Summary

In December 2015, the “Fixing America’s Surface Transportation Act” or FAST Act became law.  The FAST Act:

• Reaffirmed a continuing federal role in surface transportation, but continued a growing dependence on general funds and generally flat funding through the life of the law;

• As the first multi-year surface transportation authorization bill enacted in many years, FAST Act provided greater certainty and predictability of federal funding for state and local agencies, but Congress failed, again, to develop sustainable revenue streams to support federal surface transportation programs over the long-term.

• Did not advance the challenge of accountable and performance-based federal transportation programs that invest scarce federal resources in the most beneficial projects;

• Advanced multi-modalism in surface transportation by including for the first time a title on intercity passenger rail; and

• Improved national policy on freight and goods movement.

While the passage of FAST Act is to be applauded for many reasons, major challenges to the nation’s transportation infrastructure remain unaddressed.

Introduction

In early December 2015 Congress passed, and the President signed, a five-year surface transportation reauthorization bill, “Fixing America’s Surface Transportation Act” or FAST Act.  This was the first time in ten years that Congress had been able to enact such a multi-year bill.  Although the enactment of multi-year surface transportation bills by large bipartisan majorities had been almost automatic for a half-century, this achievement, in an otherwise deeply divided and partisan Congress and political environment, was notable.

In passing a five-year bill, Congress provided some greater degree of certainty to the state transportation departments, transit agencies, and other state and local grantees, which still receive significant assistance from the federal government for their transportation capital programs.  The over $60 billion a year from both the Highway Trust Fund (HTF) and general funds insures that the surface transportation program remains one of the largest in the federal domestic budget, and for many states and transit agencies these grants still represent very significant portions of their own capital programs.  So, the continuity and certainty of federal funding, established by FAST Act, will be important to these grantees and will allow them to continue with their own investment programs.

Finally, and perhaps most important, leading members of both Houses were determined to stop the pattern of repeated, short-term extensions of surface transportation programs and continual transfers of general funds to HTF (or “patches”) that had occurred over the last seven or eight years.  While a caucus of members in the House of Representatives generally prefer devolution of the federal transportation programs to the States and the elimination of the federal gasoline tax, FAST Act still passed in the House by 359 to 65.  In the Senate objections were raised about the “pay-fors,” on which the transfer of about $70 billion of general funds to HTF was based.  Without those pay-fors, program levels for the highway and transit programs could not have been maintained.

Overview of Impact on Transportation Policy

While FAST Act seemed to break a pattern of Congressional stalemate on major domestic grant programs and provided the certainty in funding that was welcomed by transportation stakeholders and state and local grantees, in other regards, the bill continued many significant transportation trends that had characterized national transportation policy for many years.  The bill was a combination of good and bad and a mixture of continuing policy trends and some new features. 

First, FAST Act re-affirmed that the federal government will continue to play a significant role in surface transportation investments, despite the discomfort of many conservatives in both Houses.  Spending on highways and transit programs has strong support not only from Democrats in both Houses, but a substantial majority of the Republican caucuses.  Indeed, many argue that the federal government has a defined role in transportation funding, as prescribed in the Constitution. 

Second, while FAST Act reaffirmed the national interest in surface transportation, these are not the “salad days” of the last half of the 20th century, when federal funding increased dramatically with each new surface transportation reauthorization bill (growing by 30 to 35 percent, for example, with the acts of 1991 and 1997). 

While there is a slight bump in federal funding for highways and transit in the first year of FAST Act, funding under this statute will remain essentially flat or stagnant throughout its five-year life (except for modest inflation adjustments in the later years).  Since investment needs and total spending on transportation at all levels of government keep growing, the federal share or proportion of the total investment pie will keep shrinking.

Thus, the relative federal role, while re-affirmed, continues to shrink proportionately, while those of states and localities continue to grow. Public funds at all levels are increasingly used to leverage private investment in public transportation infrastructure.

Third, the slight increases and the continuity of federal funding that were important features of FAST Act were dependent upon the transfer of abut $70 billion of general funds to HTF, thus continuing a pattern that has been established over the past decade.  The reliance on these transfers or “patches,” in order to assure HTF’s viability and to maintain current levels of federal spending for highways and transit means that Congress has failed, again, to develop sustainable revenue streams to support federal surface transportation programs. 

While the enactment of FAST Act may delay the time that another “patch” or general fund transfer to HTF will be necessary, Congress will still have to face in the future the need to identify the “pay-fors” to enable such transfers and to justify their appropriateness.  This is hardly a recipe for assuring the adequacy of federal transportation funding or for sustaining the federal role in surface transportation over the long term.

Fourth, for all the continuing debate about whether funding for, and investment in, transportation facilities should be user-based, Congress has effectively decided that HTF should be a mixed fund,that is supported by both user fees and general taxpayer funds.  Already, over 25 percent of HTF will derive from transfers from the general fund; when surface transportation programs that are entirely funded with general funds are included, the proportion of general funds grows to over 30 percent.

As a result of these circumstances, the battle over allocations of federal surface transportation funds between the states has been much muted during the consideration of more recent surface transportation authorization legislation, including FAST Act.  For many years, those states that received less in federal surface transportation grants than they raised from the federal gasoline taxes collected within their borders sought to have their relative shares of federal funding increased at the expense of those states (whose allocations of grants exceeded their contributions, in terms of federal motor fuels taxes).

With HTF increasingly dependent upon general funds, all states now receive more in federal aid than they contribute in federal transportation-related user charges, and the battle over relative allocations of grants between states seems less central to the consideration of federal surface transportation legislation.

The reality of a mixed fund for surface transportation should, also, minimize the battle between modes for funds. If general taxpayers are funding such a significant portion of federal transportation grants then certainly a stronger argument can be made that the funds should be used across modes and for broader intermodal and multi-modal solutions to mobility and accessibility problems.

Fifth, FAST Act made no real progress in introducing performance measures and real accountability in the federal surface transportation programs, which remain at the important, initial, steps, enacted in the Moving Ahead for Progress in the 21st Century Act (MAP-21).  For now, the articulation of performance metrics will have to rely on the regulatory and guidance processes of the Federal Highway Administration (FHWA) and the Federal Transit Administration (FTA).  Nor did FAST Act introduce any significant reforms in the deeply fragmented, often politicized, and frequently ineffective transportation planning and capital programming processes of metropolitan and state agencies.

Until such planning and programming reforms and requirements for performance that carry consequences are enacted at the federal level there are few assurances that the investment decisions made at the state and local levels will reflect national purposes, strategic goals, and the most benefits (economic, social, and environmental).  At a time of scarce resources, there should be a premium on “wise” investment decisions.  While FAST Act did not reverse provisions on performance and accountability contained in MAP-21, it did little to advance them.

Sixth, FAST Act did, however, advance the cause of multi-modalism in surface transportation.  For the first time a surface transportation authorization bill contained a title on intercity passenger rail (although Amtrak and other passenger rail programs will continue to be funded strictly from general funds, rather than from HTF, and through regular annual appropriation processes).

Like the highway, transit, and highway safety programs in FAST Act, passenger rail programs are authorized for five years, and FAST Act directed that net income from Amtrak’s Northeast Corridor (NEC) operations are to be used to invest in NEC capital improvements, rather than to subsidize money-losing long-distance trains (although, under the Fiscal Year 2016 Omnibus Appropriations bill, the effect of this provision will be delayed, in order to give Amtrak and the Federal Railroad Administration the opportunity to implement the new accounting structure).

Finally, a positive feature of FAST Act was its advancing national policy on freight and goods movement.  For the last 15 or 20 years there has been a growing awareness at the federal level that there are clear national interests and purposes, in addressing bottlenecks in the national freight network and in enhancing intermodal movements of goods. 

In 2005 Congress had established a grant program for projects of national and regional significance in the surface transportation re-authorization legislation adopted that year.  The program was intended to be largely directed to freight projects, but was fully earmarked by Congress. 

In the 2009 economic recovery legislation Congress established the Transportation Investment Generating Economic Recovery (TIGER) program, a multi-modal competitive grant program for state and local projects, designed to have significant economic impacts.  Administered by the U.S. Department of Transportation, the TIGER program has been extended repeatedly by Congress in annual appropriations measures.   Over its six-year life, TIGER has been the source of funding for several freight projects.

Building on these earlier freight initiatives, Congress included in FAST Act a discretionary freight projects program of $4.5 billion over the five-year life of the bill for so-called “nationally significant freight and highway projects.”  With its emphasis on highway bottlenecks, this program does not meet all of the needs of a nationally significant freight program, but it is an important step, as is the five-year $6.2 billion formula program for states to carry out freight projects.

These can be important initiatives, if these programs are actually used to improve the national and regional mobility of freight shipments and the accessibility of major market for freight movements and deliveries.

Conclusion

In its re-affirmation of a continuing federal role in surface transportation, FAST Act is an important statute.  However, this legislation continues a trend toward a growing dependence on general funds for these programs and stagnation in the general level of federal funding for surface transportation.  The inevitable result is a growing burden on states and localities to address the needs of an aging, deteriorating, and often-congested national surface transportation network.

FAST Act also fails, as did its most recent predecessors, to address the issue of establishing sustainable revenue streams to support the federal role in transportation.  Moreover, the challenge remains to assure accountable and performance-based federal transportation programs, in order to increase the probability that scarce federal resources will be invested in the most beneficial projects.

While the passage of FAST Act is to be applauded for many reasons, major challenges to the nation’s transportation infrastructure remain unaddressed.
 ___________________________________________________________ 
EMIL H. FRANKEL served as Assistant Secretary for Transportation Policy, U.S. Department of Transportation, under President George W. Bush, 2002 to 2005, and Commissioner of the Connecticut Department of Transportation, 1991 to 1995

Earlier this year, the president announced a $60 billion initiative to provide two years of “free” postsecondary training to students at community colleges. Schools receiving funding would be required to adopt evidence-based reforms to improve student outcomes as well as create programs that provide occupational training or fulfill transfer requirements to 4-year colleges and universities. However, these proposals do not fully cover the cost of attendance, just tuition and fees. Additionally, the proposals do nothing to actually lower the high cost of college, but merely subsidize today’s expensive tuition rates. 

In this paper, the American Action Forum (AAF) examines the impact of “free college” proposals on students, finding:

·      Only one out of three college students in the United States would be eligible to participate. 

·      Over a quarter of college students would be prevented from accessing a public federal relief program because the program doesn’t allow students at private institutions or 4-year public universities to participate.

·      Only $32 billion of the $80 billion combined federal and state investment for President Obama’s proposed free college program would result in a degree or credential. The majority - $48 billion- would be a loss.

Not All Students Win

Current estimates place the total college-going population in the United States at around 21 million, with about 15 million or some 72 percent of those students enrolled in public institutions.^[1] Free college proposals would not extend cost relief to students enrolled in private higher education programs, as a result more than one in four college-going students would automatically be prevented from accessing any form of tuition relief. Of those enrolled at public institutions, about 6.8 million, or roughly 45 percent are enrolled in two-year community colleges. This means that the president’s proposal would only benefit approximately one out of every three college students in the United States.

The institutions where students would be ineligible to benefit from the president’s proposal serve a large percentage of the nations’ financially needy students. Just over 30 percent of Pell recipients attend public four-year colleges and universities while another 33 percent of Pell recipients are enrolled at private non-profit and for-profit institutions.^[2] These students demonstrate one example of how the president’s proposal would prevent those in need of financial assistance from accessing it.

Why Push Students Towards Community Colleges?

From an outcomes standpoint, the president’s proposal allocates taxpayer dollars toward one of the worst-performing sectors of American higher education: community college. It also seeks to drive enrollment into what is already American higher education’s largest sub-sector. 

Community college participants not only represent the group with the highest unemployment rates amongst individuals with a college education, but their earnings are lowest amongst all individuals having a post-high school credential. They also represent the largest group of students who begin college but never complete a degree. Some estimate that 48.8 percent of all students at two-year public community colleges have not completed a degree and are not enrolled at another institution 6 years later.^[3] As a sector, these students also have the highest tendency to default on their federal student loan debt.^[4]

Chart 1: Earnings and Unemployment Rates by Educational Attainment

Source: Bureau of Labor and Statistics

The Taxpayer Cost of Funding Access Versus Degree Completion

In general, federal investments in higher education are made in an effort to increase the degrees and credentials needed to ensure a productive workforce with lower unemployment rates, higher wages, and economic growth. By definition, free college proposals increase access to higher education. However, more students does not equate to more degrees granted, which yield the kinds of wage gains consumers seek or the tax and productivity gains that governments expect to realize.^[5] The lack of evidence or support for improving completion has been a common refrain amongst both economists and pundits.^[6]

In fact, in November of 2014, just two months before the president unveiled his plan, the National Student Clearinghouse released an in-depth study that students at the public 2-year colleges at the heart of the president’s plan are highly unlikely to earn a degree or credential that is capable of generating economic benefits.

Chart 4. Six-Year Outcomes by Starting Institution Type

Source: National Student Clearinghouse Research Center

As the chart shows, after six years only about 39 percent of public community college students end up completing a degree.^[7] This means that the president’s free college proposal would effectively be spending $36 billion of a $60 billion investment on up to 5.4 million students who will likely never receive any type of college credential. Add the share of the $20 billion that states would be required to invest on top of the federal match and the total potential loss on an $80 billion federal and state investment could be close to $48 billion.

A full accounting of the economic costs is difficult to produce, especially given the migration of students from 2- to 4-year institutions. Still, research from the American Institutes of Research found that between 2004 and 2008 almost $4 billion in federal and state taxpayer monies through grants and appropriations went to community college students who dropped out after just one year of study.^[8]

It’s important to note that the federal taxpayer’s $60 billion investment provides grant dollars to students regardless of whether they actually receive a degree. It provides aid whether the student is part of a family of four that lives below the poverty line or part of a family that makes $200,000 per year.

Addressing the Real Challenge

The easiest way to understand the challenges with the president’s free community college proposal is to compare it to additional investment in, for example, existing federal programs designed to foster affordability, such as the Pell grant. 

If the focus of the program is to guarantee that individuals will complete and reap the benefits of advanced education, it is unclear why a policy should specify which entity provides the service.

As many critics have already noted, free community college ends up subsidizing individuals for whom higher education may already be affordable. On the other hand, this proposal works against efforts to redress problems associated with under-matching: the idea that qualified students from less-affluent households do not end up pursuing degrees at competitive colleges and instead enroll in less-selective or two-year community colleges. In either of these circumstances proponents need to be able to articulate why a free community college run via state-based incentive grant programs would be more efficient or effective at improving affordability, access, or completion than Pell grants or other programs already in place.

The under-matching issue is particularly salient because it suggests free college is highly inconsistent with existing federal education policy. For years, Congress and the Department of Education have pursued efforts to facilitate and improve consumer choice by stressing the need for cost calculators and ratings systems so that potential students can make decisions that best fit their unique circumstances. They have encouraged shopping for schools and programs, mindfulness towards things like graduation and attrition rates, and to make academic major and career decisions based on wages and future employment prospects. They have brought under- and over-matching into the national policy debate and actively sought ways to help students avoid making college-going decisions based on what costs the least and instead focus on where their skills and talents fit best.

The current free community college proposals effectively throw nearly all of this by the wayside with incentives for students to enroll in institutions with the highest attrition rates, some of the lowest wage curves and the greatest unemployment variability of any post-high school training institution. They make cost the main driver of the educational investment decision, without placing any downward pressure on the cost of college. They effectively place a financial penalty on millions of poor students for choosing a 4-year public or private option that may be a better fit for their capabilities or increase their chance of graduation. 

As policy goes, promoting choice while financially encouraging participation in a single sub-sector sends consumers mixed signals. It also creates economic inefficiencies by giving free college to students at one institution type without regard for what they’re studying.  This risks an overflow of students in programs that labor markets neither need nor want.

Conclusion

In the current proposals, free does not attempt to drive down the costs of providing students with an education, but instead shifts the burden of paying for higher education away from students and onto taxpayers. A balanced policy discussion on these proposals must consider the costs and tradeoffs. Not every student benefits from this proposal and not every state will be willing or able to participate. Where public resources are scarce, the fact that so many students start college but never finish raises important questions about whether existing funding could be more efficiently allocated to achieve more favorable outcomes.

There is strong public agreement that policymakers need to develop new ways of promoting higher education completion and build new tools that will help students, institutions, and government manage college affordability. Every option should be weighed through a balanced assessment of the extent to which it meets the needs of all parties that it may affect, the costs on taxpayers, and its impact on affordability.

• Download

Summary

• The Arctic has received very little attention from the U.S. policymakers and the administration despite growing in economic and geostrategic importance as a result of receding sea ice levels.
• The federal government must uphold its constitutional responsibility to provide for the nation’s common defense by asserting American leadership in the Arctic.
• The United States should develop regional defenses and promote democratic norms in the wake of increased Russian and Chinese activity in the Arctic.

Introduction

With over 1,000 miles of Alaska’s Arctic coastline, the United States stands to benefit from the region’s increasingly accessible natural resource deposits and commercial maritime transit routes. However, many Americans do not understand the significance of this relatively unexplored frontier.

With the exception of the Cold War, the United States has invested very few resources in order to secure its position as a global leader in Arctic affairs. By failing to properly develop the region’s defense and commercial infrastructure, America will lose the ability to assert itself in this increasingly important area of geopolitical concern. Doing so will cede leadership to other nations – namely Russia – and continue a dangerous trend of reactive, rather than proactive engagement on the part of the American political and military establishment.

Source: Wikimedia

A Changing Arctic Climate

The Arctic is warming at an unprecedented rate. Early climate models indicated that the Arctic would experience ice-free summers starting at the end of the 21^st century, but current estimates now project that this phenomenon will begin as early as mid-century, if not sooner.

The eight Arctic nations – the United States, Russia, Canada, Norway, Sweden, Finland, Denmark (via Greenland), and Iceland – as well as China, South Korea, Japan, and other countries stand to benefit from these monumental climatic shifts as new opportunities for resource extraction and trans-oceanic shipping emerge. Many in the international community have therefore called for development of the region’s defense and commercial infrastructure.

The Emerging Arctic Energy Market

According to a 2008 United States Geological Survey report, “The extensive Arctic continental shelves may constitute the geographically largest unexplored prospective area for petroleum remaining on Earth.” This high volume of untapped reserves – 84 percent of which may lie in offshore deposits – could amount to 13 percent of global undiscovered oil and 30 percent of global undiscovered natural gas.

Although Russia holds a large claim over these resources as a result of its vast Arctic coastline, the U.S. Department of the Interior’s Bureau of Ocean Energy Management estimates that Alaska’s outer continental shelf contains 26.21 billion barrels of undiscovered oil and 131.45 trillion cubic feet of undiscovered natural gas.

A recent report issued by the National Petroleum Council – a body of individuals appointed by the Secretary of Energy to advise the federal government on matters concerning energy, the environment, security, and the economy – states that exploration of and eventual output from these untapped reserves “may provide a material impact to U.S. oil production in the future, potentially averting decline, improving U.S. energy security, and benefitting the local and overall U.S. economy.”

Developing Arctic Infrastructure

Potential for near-term escalation of conventional warfare in the Arctic region remains low. A number of unique security challenges have nevertheless emerged as countries respond to the region’s burgeoning economic and geostrategic opportunities with the development of commercial and defense infrastructure.

From an immediate threat outlook, additional infrastructure is required to provide for safe transit through Arctic sea lanes. Unprecedented opportunity for natural resource extraction also warrants additional infrastructure to support potential oil spill and other environmental disaster cleanup initiatives.  Ship casualties in Arctic Circle waters ranging from structural damage to complete destruction have skyrocketed from just three in 2005 to 55 in 2014. Until countries invest in Arctic infrastructure to deal with increased transit and commercial development, these ship casualties will only become worse.

United States

America’s political and military leadership appears underprepared to lead in the Arctic. The U.S. Coast Guard, which is responsible for conducting maritime patrols such as search and rescue operations, faces a number of increasingly severe challenges as sequestration and inconsistent congressional budgeting have undermined its efforts to modernize and expand its fleet of polar icebreakers and surveillance aircraft. Together, these assets are the key tools for any nation seeking to project power and operate in the Arctic.

A frequently cited Coast Guard report states that adequate U.S. presence and capability in the polar regions requires at least three heavy and three medium polar icebreakers. Currently, the Coast Guard only operates one of each – the Polar Star heavy icebreaker, which can break through thick Arctic ice year-round, and the Healy medium icebreaker, which is used in support of Arctic scientific research. The Coast Guard also has one inactive heavy icebreaker, the Polar Sea, which is need of repair. Navy officials indicate that icebreakers “are the only means of providing assured surface access in support of Arctic maritime security and sea control missions.”

Source: Wikimedia

Facing a major challenge in maintaining the 39-year-old Polar Star, the Coast Guard is now attempting to acquire a new heavy icebreaker. With a price tag of approximately $1 billion, one polar icebreaker would consume the entire Coast Guard annual acquisition budget. Any new procurement would require significant support from both the federal and legislative branches. However, the Obama Administration has cut five-year funding for icebreaker acquisition by 81 percent since Fiscal Year (FY) 2013. Rather than receiving the previously budgeted $508 million through FY 2015, the program has only been allocated $9.6 million. The president’s FY 2016 budget request did not help the situation, requesting a mere $4 million for acquisition of an additional heavy polar icebreaker. Such delay in funding creates an uncertain and dire operating climate for the Coast Guard.

In an attempt to remedy the glaring capability gap, the Obama Administration announced in September 2015 that it would jumpstart production of an additional polar icebreaker by two years to FY 2020. If procured that year, the additional polar icebreaker would not come into service until 2024 or 2025. Given that the Polar Star’s service life is expected to end between 2019 and 2022, current projections of Coast Guard assets and capabilities indicate that “there will be a period of perhaps two to six years during which the United States will have no operational heavy polar icebreakers” if the Polar Star is not further extended or the Polar Sea repaired.

Even if the Coast Guard were to acquire an adequate fleet of icebreakers to escort ships through polar sea ice, the Navy may still be unable to operate in certain Arctic conditions. According to a war game published by the United States Naval War College (USNWC), “Strategic and operational planners will simply need to accept that certain areas in the Arctic remain off-limits to U.S. warships unless the commander is willing to accept risks, the ice recedes away from the area of interest, or ships are produced with additional ice strengthening.”

The USNWC war game further notes that U.S. Arctic capability gaps also concern the lack of air and ground support in Alaska’s Arctic, “to include military hangars and fuel storage, as well as roads from Fairbanks to [northern] airfields and supply nodes.” Coupled with the fact that there are no naval installations in Alaska’s Arctic that can accommodate icebreakers, such lack of regional infrastructure – including poor communications and satellite capabilities – significantly delays the response time of emergency responders and defense entities. The state of Alaska and Army Corps of Engineers do have a plan to make one port just outside of the Arctic Circle more accessible to icebreakers.

Outdated maps of America’s Arctic waters also exacerbate these capability gaps, forcing ships to rely on half-century old estimates of the region’s operating environment when in transit. Updated mapping of Alaska’s Arctic shoreline will not be completed until at least 2035.

Canada

The Canadian Rangers provide the bulk of the Army’s ground presence in the Arctic. This reserve force is comprised of approximately 5,000 soldiers operating over 170 patrols throughout Canada’s northern provinces and territories. Because of the Rangers’ small size and lack of resources, Canada’s Arctic Response Company Groups and Army Initial Reaction Unit assist it in providing northern ground support. Collectively, these forces provide a focused, yet limited presence in the Arctic.

Like the United States, Canada has inadequate Arctic maritime forces. Although it maintains six icebreakers (two medium and four light), none of them are equipped to operate year-round in the Arctic environment. Their warships are also not properly hardened for Arctic transit and the Navy’s fleet of Arctic Offshore Patrol Ships is still years from reaching operational status.

The Canadian Air Force’s Arctic presence is strengthened by its aerial situational awareness capabilities. It will procure three additional RADARSAT II satellites by 2018. These systems enhance marine surveillance, ice monitoring, disaster management, environmental monitoring, resource management, and mapping around the world. Canada’s fleet of 18 CP-140 Aurora surveillance aircraft also regularly operates in the Arctic and are currently receiving extensive upgrades. The Air Force relies on a small fleet of CF-18 aircraft that operate out of four northern forward operating bases to deter Russian incursion into its sovereign airspace.

Russia

Of all the Arctic nations, Russia undoubtedly maintains the largest Arctic military infrastructure. It established an Arctic regional command in 2014 and has begun building 10 airfields and 13 air-defense radar stations throughout the region. Russia currently operates 17 icebreakers – four heavy, six medium (with three more under construction), and seven light.

Russia has installed a variety of anti-aircraft missile regiments, missile system batteries, and coastal defense missile battalions in the Arctic. It is also developing extensive Arctic airlift capabilities, with an expected 2017 completion date for the Tiksi airport in Northeast/Central Russia. This facility will reportedly be outfitted with MiG-31 aircraft, one of the planes Russia has used to violate sovereign airspace of Arctic nations. Russia has also moved forward in constructing runways on numerous Arctic islands to increase its intelligence, surveillance, and reconnaissance capabilities in the region.

This buildup is meant to assist Russia in securing its claims to the Arctic’s vast natural resources and potentially lucrative shipping routes. A stronger Arctic presence will also act as force multiplier for the Russian Navy, allowing it to reassign resources through what is becoming an increasingly accessible trans-oceanic route.

China

As a rising geopolitical power and the world’s largest energy consumer, China sees the Arctic as a critical source for gaining access to natural resources and economical shipping routes. Having previously taken a more aggressive tone against notions of Arctic sovereignty, the country now advocates for peaceful cooperation with Arctic nations. China is currently conducting a five-year assessment of polar resources and governance in order to advance its cooperative agenda. Through its Arctic Research Center in Shanghai, China collaborates with Nordic nations on scientific research. It also works with international partners at the Chinese Arctic Yellow River Station in Norway.

China further cooperates with Arctic nations by participating in both bilateral and multilateral scientific missions. It currently operates the Xuelong (Snow Dragon) light research icebreaker, which conducted a trans-Arctic voyage from Shanghai to Iceland in August 2012, and is building a second, more advanced research icebreaker that is expected to enter into service in 2016. China claims these assets will primarily be used for Arctic scientific research.

China is also a member or observer of numerous Arctic-related regional associations and multilateral organizations. Most recently, it was admitted to the Arctic Council (which the United States currently chairs) as an observer nation. The Chinese are expected to continue advocating for their status as a self-described “near-Arctic” nation in these multilateral forums, despite being nearly 1,000 miles away from the Arctic Circle at its closest border.

Multilateral Cooperation in the Arctic

The international community has demonstrated an interest in maintaining stability in the Arctic, placing an emphasis on establishing multilateral lines of communication and cooperation. In particular, the Arctic Council and the International Maritime Organization have proven to be most effective in terms of fostering cooperative engagement between the Arctic nations. These organizations have produced a variety of international agreements, namely: the Agreement on Cooperation on Aeronautical and Maritime Search and Rescue in the Arctic; and the Agreement on Cooperation on Marine Oil Pollution Preparedness and Response in the Arctic; and the Polar Code, which creates standards for ships operating in the region. Although significant in terms of fostering ties between the Arctic nations, these agreements largely outline existing responsibilities that any country ought to follow under international maritime laws and norms.

Recently, the Arctic nations announced the formation of the Arctic Coast Guard Forum, “an operationally-focused, consensus-based organization with the purpose of leveraging collective resources to foster safe, secure and environmentally responsible maritime activity in the Arctic.” The United States currently chairs this forum in tandem with its chairmanship of the Arctic Council.

Some believe current economic and environmental conditions in the Arctic do not provide sufficient opportunity for cooperation. According a report published by The Arctic Institute, “Sharing responsibility or handing over tasks to other countries is often not even an option, as states are struggling to even provide capabilities in their own Arctic maritime domains.” Diplomatic complications may also emerge as countries must identify how they seek to engage with Russia in Arctic affairs following the country’s annexation of Crimea and repeated incursions into the sovereign airspace of Arctic nations.

Leading from Behind in the Arctic

At current standing, the United States lacks both the capabilities and political will to lead in the Arctic operating environment. The American political and military establishment must therefore reevaluate its northern posture so as to foster greater stability and economic prosperity in the region. Failing to do so could ultimately threaten U.S. national security objectives, prompting reactive and potentially hostile engagement in the future.

Both Congress and the administration must uphold their responsibility to provide for the common defense by ensuring that the nation is prepared to lead wherever American interests are at stake. Doing so will require an intensive and immediate development of America’s Arctic defense infrastructure coupled with sustained promotion of democratic norms through both multilateral organizations and bilateral agreements.

Summary

Puerto Rico’s declining economy has driven their country into $71 billion in debt. As a whole, Puerto Rico has poor growth, an oversized public sector, and a tax system inadequate to the commonwealth’s needs It is vital for Puerto Rico to prioritize the funding of essential programs while reducing other spending in order to regain economic stability. In this paper AAF outlines key reforms to improve Puerto Rico’s budgetary and economic outlook. Highlights of these reforms follow:

• Tax reform in order to more efficiently collect due revenue and chip away at the debt without inhibiting the economy.

• Congress should regularize Puerto Rico’s federal reimbursement policies of rum excise taxes to improve predictability and stability.

• Streamlining, consolidation e and where appropriate, privatization, of the large public corporations responsible for $48 billion of debt. Addressing these debts is essential to stabilize the Commonwealth’s finances as a whole and to finance large unfunded pension liabilities.

• Education reforms that address the Commonwealth’s needs while more efficiently stewarding public resources.

• Governance reform that could allow the Commonwealth to save between $20 to $60 million from consolidation of municipalities.

• Consolidation Puerto Rico’s 130 governmental agencies could save between $60 and $300 million annually.

• Modernization of Puerto Rico’s labor market that addresses high employment costs and unfavorable business environment.

• Exemption from the federal minimum wage, which is very high, in order to lower employment costs and incentivize employers to hire more workers.

• Improvements to Puerto Rico’s budget process to enhance the transparency of its finances in order to pinpoint and resolve problems in the budget within a reasonable time.

Introduction

The Puerto Rico economy has been on a steady decline for over a decade, and it is now in the process of grappling with a budget crisis that has endangered its capacity to finance governmental obligations; the current outstanding stock of which totals about $71 billion. Any strategy to remedy the dual challenge of weak growth and heavy indebtedness must engage policymakers at the local, Commonwealth-wide, and federal levels. This study outlines the growth and fiscal context in which potential policy reform must take place, and examines potential elements of a pro-growth fiscal consolidation for the Commonwealth of Puerto Rico.

Growth and Fiscal Context

The ultimate goal of economic and budgetary reforms in Puerto Rico should be a growing economy, and the ability of the Commonwealth to regain access to capital markets in a sustainable fashion. The growth challenge is longstanding and real. As shown below, Moody’s anticipates that, at best, Puerto Rico will have very modest growth and not until over five years from now.

Figure 1: Historical and Projected Output

imilarly, the Commonwealth has a deep structural budget challenge that has led to large and persistent budget deficits. Note that these reflect not only general fund deficits, but commonwealth-wide activities.

Figure 2: Structural Deficits

Rectifying these deficiencies will necessarily involve some difficult choices. However, it is important to put in context the Commonwealth’s capacity to absorb consolidation, its ability to restrain spending, its levels of taxation, and the size of its indebtedness. Indeed, it is a common assumption that Puerto Rico has engaged in protracted austerity and, despite this, is over-taxed and over-leveraged.  The data do not support these assumptions.

Consider first the amount of government spending in Puerto Rico. Despite attempts to scale back spending, Puerto Rico has failed to significantly reduce budgetary expenses. Looking at Figure 3, it is evident that Puerto Rico has ample room to reduce government spending when compared to the U.S. states.[1] Puerto Rico is well above the mean in this calculation which supports a capability of reduction. It is essential for the economic stability of Puerto Rico to prioritize the funding of essential programs and debt while at the same time reducing other spending. Previous research indicates that it is valuable for Puerto Rico to preserve core functions of government while focusing spending reductions on transfer programs.

Figure 3: Government Spending

On the other side of the ledger, tax collections as a percent of GDP for Puerto Rico are actually lower, at 11 percent, than any other U.S. states.[2]

Figure 4: Tax Collections

Similarly, the idea that Puerto Rico is dramatically over-leveraged is another misconception surrounding the Commonwealth.  Looking at total debt as a percent of GDP (with and without unfunded pension and other post-employment benefits [OPEB]) Puerto Rico is well below the mean and median when compared to the mainland. Looking at the total debt including unfunded pension and OPEB, Puerto Rico has debt equal to 98 percent of its GDP; substantially below the states’ median of 120 percent.[3]

Figure 5: Total Debt[4]

 Taken as a whole, the data depict a Puerto Rico that has poor growth, or actual decline; an overly large public sector; and a revenue base that simply is too small to finance it. This mismatch will ultimately lead to irreversibly large debt burdens, but that has not yet occurred. There remains time to address the growth and budget challenges.

Commonwealth Reforms

Puerto Rico must take the lead in charting a course beyond its current economic and budgetary torpor, which means any fiscal consolidation must first involve self-initiated reforms. Puerto Rico’s tax measures need modernization which can ultimately generate funds to chip away at the debt without inhibiting the economy. The debt portfolios of Puerto Rico’s large public corporations – with on the order of $48 billion in debt outstanding – must be addressed as part of a growth plan for Puerto Rico as well as pension liabilities.[5]  Additional savings can be achieved in other areas of the Commonwealth’s budget. One example is in education spending, where savings can be achieved without harming educational attainment.  The Commonwealth must also make structural changes at the municipal level by streamlining local administration. These measures should be paired with reforms that enhance labor markets, an essential element of a pro-growth consolidation. As part of this reform effort, and to gauge budgetary and economic progress, the Commonwealth must undertake meaningful budgetary process reform that enhance the transparency of its finances.

Tax Reforms

Tax reform and administration can play an important role in improving Puerto Rico’s budgetary trajectory. This should include a needed modernization of its current tax compliance tax system, including the property tax system, to more efficiently collect due revenue.

Puerto Rico has a substantial problem with the collection rate of its Sales and Use Tax (SUT) that is noted in both Krueger’s and the governor’s plan. At 56 percent[6], Puerto Rico’s collection rate is significantly lower than the U.S. states’ total tax collection rate of 83 percent[7]. By improving the Commonwealth’s SUT tax collections to 65 percent-70 percent (still well below the U.S. states’ total tax collection rate), Puerto Rico could yield approximately $100 to $150 million in annual receipts[8]. The collection rate can be improved by using a quasi-independent government agency (similar to the federal IRS) or a Public Private Partnership to assign taxes.

Another outdated system needing modernization addressed in the Krueger and Governor’s report alike is the property tax. The Krueger Report estimates that revamping the property tax can yield $100 million for 2016 increasing to $350 million by the year 2018[9] by reevaluating the assessed real-estate values (currently from 1954), updating the registry and adjusting down tax rates. A transition period using discounts—based on income-level and change in tax due—could be used to ease the burden of the increased valuation. Since Municipalities normally retain the property taxes, the central government revenue could cut its transfers to municipalities by the amount of additional property tax receipts.

Public Corporations Reforms

Puerto Rico has the opportunity to benefit from the consolidation and privatization of the public entities responsible for issuing $48 billion in debt. The funds generated and debt removed in the process can be repurposed to finance a growing unfunded pension.

Public Private Partnership (P3) initiatives can benefit the Commonwealth by alleviating debt, unfunded pension, and OPEB liabilities while reducing the outsourcing of government expenditures. Another side benefit of P3 initiatives is a lower cost and better service for customers; the result of private corporation structures improving the current inefficient system. Examples of P3 statute success include the Luis Munoz Marin International Airport, PR-22 & PR-5 toll roads, and 21^st Century Schools project. Public Private Partnership is also present in the Governor’s plan.

PREPA can benefit by pursuing P3 projects for its power generation facilities or for developing the infrastructure intended for the transportation of Liquid Natural Gas. Doing this lowers power generation costs, generates new jobs for its residents, while eliminating debt financing costs. PRASA can improve profitability by refining water collections, reducing operation costs, and increasing usage rates. The major government-owned insurance corporations could be consolidated and privatized. Consolidating and partially privatizing the EDB, MFA, and GDB could reduce overhead expenses and finance IRR projects for the development of Puerto Rico. The potential developments include infrastructure projects, microfinance institutions, and other economic development initiatives as well as providing capital to small and medium sized enterprises to reduce volatility and having programs available to encourage foreign direct investment. The estimated effects of consolidating and privatizing specific public corporations can be found in Table 1.

Table 1: Potential Proceeds[10]

Pension costs continue to be a rising problem for Puerto Rico where the current unfunded pension nears $36 billion[11]. In order to resolve this issue, Puerto Rico can change all pensions to defined contribution plans, increase the minimum retirement age, and increase employee contributions.

Education Reforms

Education reform is another necessary measure to conserve funds unnecessarily subsidizing the University of Puerto Rico (UPR) as well consolidate the excess spending on schools and teachers.

The University of Puerto Rico is heavily subsidized by the commonwealth as all UPR students benefit from a blanket subsidy.  The flat rate that the UPR students pay is far lower than many schools on the mainland. Educational support could be shifted to a need-based scholarship approach to reduce the General Fund’s spending without hindering those reliant on the subsidy. Krueger estimates that a program such as this can reduce spending $500 million by FY2020[12].

Turning to the K-12 system, in the past decade the number of students in Puerto Rico dropped 40 percent while teachers increased 10 percent. This created an average student-teacher ratio of 12 which is considerably lower than the median for the mainland of 16. Krueger finds that a gradual cut in the number of teachers saves approximately $400 million per year by FY 2020[13]. Another cost reducing reform incorporated in both the Krueger and Governor’s report is a consolidation of schools. By reducing the number of schools, the remaining will be able to invest more in modern facilities and better technology. The governor’s report estimates that consolidating schools and reducing Puerto Rico Department of Education payroll is expected to save $50 million in 2017 growing to approximately $225 million in 2020[14].

Government Reforms

With 50 of the 78 municipalities operating at a deficit[15], the Commonwealth could benefit from regionalizing its less financially stable municipalities. Smaller municipalities spend 98 percent[16] of their budgets on payroll, according to Bayamón Mayor Ramón Luis Rivera Cruz, which can be consolidated to reduce operational costs and improve tax revenues while providing more services for the people. The Commonwealth could save between $20 to $60 million from consolidation of municipalities. The Commonwealth could also save between $60 and $300 million annually by consolidating Puerto Rico’s 130 governmental agencies.

Labor Market Reforms

Another measure that needs to be made at the Commonwealth level is the modernization of the labor market. The employment costs are too high and other laws make doing business burdensome in Puerto Rico.

The local labor laws of the Commonwealth are problematic to the employee base by supporting an unfavorable business environment. In order to overcome this problem, Puerto Rico needs to alter the laws that magnify employment costs: defining overtime as any time spent working in excess of an 8-hour day, 30 paid vacation days compared to the norm of 15, mandatory end-of-year bonus, and Sunday premium pay.  Puerto Rico also needs to change the laws that disrupt the ease of business: onerous requirements for laying off employees compared to the mainland and a probationary period of only 3 months. The reform of these local labor laws is supported in both the Krueger and Governor’s report.  Additionally, vocational training could be subsidized to advance the skills of Puerto Rico workers while reducing the employment costs related to business hiring.

Another problem with investing in Puerto Rico is a local permitting process that it is disorderly and not centralized. One reform backed by the Governor’s is to issue Executive Orders designating strategic emergency areas for expedited permitting processes in select-essential infrastructure and service projects. This is the quickest and easiest means of attaining improvement without the need for legislation. Another reform that would streamline the process is to modify the language in the Permit Process Reform Act. Current reviews and approvals take too much time to conclude and, due to the multiple agencies involvement, are redundant. By consolidating all the infrastructure and environmental related permits in the State Office of Permits Management, reviews and approvals will be much faster and effective. Another suggestion is to integrate all of the municipal permitting processes Commonwealth’s existing permit filing and processing platform centralizing the process.

Budget Reforms

The Commonwealth needs to improve the budget process, especially the budget’s transparency, in order to pinpoint and resolve problems in the budget within a reasonable time. The Commonwealth did not pass a balanced budget for the general fund in 5 of the past 6 years. Furthermore, the government has failed to report annual financial statements within 305 days of each fiscal year in 9 of the past 13 years. Meanwhile, there remains considerable skepticism over the the accuracy of relevant budget data. The lack of urgency to get the budgets passed in a timely manner is detrimental to Puerto Rico’s ability to turn around its fiscal situation which is a concern evident in both the Krueger and Governor’s report.

Federal Reforms

There are a number of policy reforms Puerto Rico must undertake to improve its economic and budgetary outlook, but those alone will not adequately address Puerto Rico’s challenges. There are also a number of reforms needed at the Federal level to increase the labor force and reduce expenses on the island.

Tax Reforms

Puerto Rico is in part dependent on the vicissitudes of federal tax writers owing to its reliance on federal reimbursements of rum excise taxes. These reimbursements are currently based on an enhanced, “temporary” basis, that is routinely extended in Congress’s annual tax “extenders” package. If this package is not renewed, Puerto Rico and the U.S. Virgin Islands would receive lower reimbursements – a combined $168 million in subsidies.[17] As part of broader tax reform measures, Congress should regularize its tax policies to improve predictability and stability.

Labor and Welfare Reforms

Puerto Rico has a significant problem with a labor force that continues to decline. This problem stems from Federal regulation hindering the Island’s ability to set compensation for its workers and is agreed with in the Krueger and Governor’s report. The Governor’s report finds that only 40 percent of the adult population are employed or looking for work compared to 63 percent on the U.S. mainland.[18]

Minimum wage laws contribute to the abysmal labor force participation rates by inflating employment costs. The U.S. federal minimum wage is very high relative to local average where minimum wage is equivalent to 77 percent of per capita income compared to 28 percent in U.S. states.[19] Puerto Rico needs to be exempt from the federal minimum wage in order to lower employment costs and incentivize employers to hire more workers. The Krueger report estimates the income taxes raised from higher worker participation to be $50 million in the early years and quickly growing. Another reason for the low labor participation rate is the welfare system providing generous benefits often exceeding that of minimum wage employment. One estimate from the Krueger Report shows that a household of three eligible for food stamps, AFDC, Medicaid and utilities subsidies can receive $1,743 per month compared to the minimum wage earners take home earnings of $1,159.

As of now, there is no incentive to join the work force because of this discrepancy between benefits of workers and non-workers. In order to combat this problem, the Federal government could give the Common wealth the ability to adjust welfare requirements and benefits accordingly; such as, continuing food stamps for a while even after a person returns to work, providing lower housing benefits to more people rather than higher benefits to a few, and cutting back on Medicaid benefits paid out over and above the Federal minimum standard. These suggestions for lowering the welfare system’s generous benefits are supported in the Krueger report.

Regulation Reforms

The Jones Act is another legislation that needlessly increases the costs on the island. The Jones Act does this by requiring the use of American crew and U.S. manufactured vessels on all shipping routes where the origin and destination are within the United States and its territories. Puerto Rico’s inclusion under the Jones Act results in Puerto Rico import costs twice as high as neighboring Caribbean Islands. This issue of high transport costs is present in both the Krueger report and the Governor’s report where it is suggested that Puerto Rico be exempt from the Jones Act. Despite there being exemptions in other U.S. territories (the U.S. Virgin Islands and Guam), Puerto Rico continues to be overruled with a rejection as current as earlier this year. A more lawmaker friendly proposal is to exempt Puerto Rico from the Jones act for specific imports, such as energy/fuel and food which would drive down the costs of living on the island. The lowered costs from this exemption could feasibly invite more tourists as well as retain the fleeing workers.

Conclusion

To improve its budgetary and economic outlook, Puerto Rico must embrace a multi-layered and multipronged fiscal consolidation that also enhances labor and investment opportunities in the Commonwealth. This approach should recognize the Commonwealth’s wherewithal to absorb difficult choices, while improving long-term economic growth. These reforms should focus on reforms to the Commonwealth’s tax system, and directing savings towards Puerto Rico’s public sectors. Additional reforms should improve Puerto Rico’s investment climate and spur domestic economic activity.

Executive Summary

In his 2015 State of the Union address, President Obama announced a $60 billion  initiative to provide students with two years of free community college. The program would require states to opt in and commit 25 percent of the necessary funding.  In return, the federal government would pick up the remaining 75 percent of funding for tuition and fees.

In this paper, the American Action Forum (AAF) examines the impact of “free college” proposals on states, finding:

• State spending would need to increase by approximately 5 to 13 percent in order to meet the contribution to be eligible to receive federal matching funds.
• All states would spend at least $3.7 billion to $4.1 billion annually.
• Only $32 billion of the $80 billion combined federal and state investment for President Obama’s proposed free college program would result in a degree or credential. The majority - $48 billion- would be a loss.

Introduction

Recent proposals to provide the first two years of community college to students for free have become popular. These proposals do not fully cover the cost of attendance, just tuition and fees. The proposals also rely on states to provide funding commitments and meet “maintenance of effort” conditions.  In addition, the proposals do nothing to actually lower the cost of college, but merely subsidize today’s expensive tuition rates. Free in this context does not attempt to drive down the cost of providing education, but instead shifts the burden of paying for higher education away from students and onto taxpayers.

The Cost of State (Lack of) Participation^[1]

The president’s proposal is structured as a matched investment, including an additional commitment to making and sustaining appropriate funding increases. The annual cost of expected state investment varies substantially, from as low as $7.4 million in Vermont to more than $349 million in New York (see Chart 2). At a minimum, the incremental one-year additional cost for all states would be approximately $3.7 billion.^[2] This does not take into account tuition increases nor does it address the potential for students not currently in the higher education market to enroll in community colleges or for students at 4-year institutions to shift to community colleges.  Under very simple assumptions[3] the collective annual cost to states could easily reach $4.1 billion.

Pairing these minimum additional commitments with existing annual state appropriations for higher education^[iii] suggests that the average state would need to increase its annual appropriations by 5.1 percent to 5.7 percent to cover states’ shares of the proposal’s match requirement. Across individual states those percentages vary from as low as 2.1 percent in Montana to 12.4 percent in New Hampshire.

Chart 2: Baseline Estimate of State Costs & Percentage of Budget Increase Needed to Cover Costs

Source: AAF Analysis (See Appendix 1)

States must not only be fiscally capable of such investments but also be willing to participate in the program. While the program currently has yet-to-be-defined performance standards, these standards and the necessity of reporting on them will impose further regulatory costs and paperwork burdens. In either case the incentive grant structure means that different levels of state participation will impact the eligible college-going population differently. Just four states enroll one-third of all public college students (California, New York, Texas and Florida) and half of all public community college students come from just seven states. California alone enrolls approximately one out of every five community college students in the United States.

Given that states, not students, initially determine participation levels, they substantially determine which students are eligible. For one or more of the “Big Four” states above to either be unwilling or unable to meet the incentive grant requirements, the result would result in anywhere from 200,000 to 1.6 million students to be ineligible for the proposed public subsidy.

The ability to both inject and sustain additional resource commitments into public higher education will depend in part on individual states’ public priorities, notably health care and pensions commitments. Despite the last recession ending just five years ago and a number of states having experienced budget surpluses in recent years, declines in state higher education investment continues to persist. Between 2008 and 2015, 31 states cut per-student funding by more than 20 percent, and 6 states cut funding by more than one-third.^[5]

Proposals for free college are expected to offset such declines by creating matching federal investments, however, it is unclear to what extent capacity for such investments exists. Up to 16 states are expected to run deficits in the next year or two years. In states with sizeable public higher education systems like Illinois, Maryland, Wisconsin and Pennsylvania, deficits are expected to run between $750 million and $6 billion.^[6]

Chart 3 plots state 2- and 4-year enrollments against George Mason University’s Mercatus Center index of states’ fiscal health.^[7] The bottom five states in particular are classified by the Center as being in “financial peril” based on high deficits and unfunded debt obligations. If one assumes that the bottom quintile was financially unable to meet the matching fund requirements, more than 1.6 million community college students in these states would find themselves ineligible for relief provided by the president’s proposal.

Chart 3: Ranking of States’ Fiscal Health Compared to Public Full-Time Equivalent Enrollment in Degree-Granting Institutions, by State, Fall 2012

Note: Based on FTE fall enrollments.

Source: NCES, Digest of Education Statistics 2013, Table 307.20.; Mercatus Center at George Mason University

State willingness to participate may also be affected by college graduates’ migration patterns. Unlike other infrastructure investments, such as highways, human capital does not have to remain within the state where it was created. As Figure 1 shows, hundreds of thousands of students with either some college or associates degrees annually move to different states.

Figure 1:

Source: U.S. Census Bureau, 2014 American Community Survey 1-Year Estimates

Finally, evidence supporting the success of higher education programs of this type is thin and is more apt to suggest states generally struggle to meet federal eligibility conditions.

The U.S. Government Accountability Office published a report in December 2014 on state and federal programs to improve college affordability that was only able to identify three federal/state incentive-style programs. Of these, the one closest in structure to free college proposals, the College Access Challenge Grant Program, appropriated $142 million in 2013 (the last year of the program) but only ended up spending half of the money because not all of the states were able to meet the maintenance of effort conditions required for the funding.^[8]

Addressing the Real Challenge

A simpler solution would be to investment additional funds into existing federal programs designed to foster affordability, such as the Pell grants. In comparison to Pell, state-based grants or allocations to institution types by design must curtail access since individuals do not dictate initial eligibility.

At the state level, the obvious obstacle is that states most capable of meeting the conditions necessary to receive matching federal funds will be those who are likely to need it the least. As seen in Figure 2, the financially healthiest states almost exclusively service small community college populations and states with high existing public subsidies at these schools. The states that need the dollars the most, the ones where economic investment can – over the long run – foster a larger, more stable revenue base, will find it most difficult to secure the investment dollars and maintain those levels over time.

In this regard, a state-based incentive grant is a policy that actually promotes gaps between the haves and the have-nots. If the federal government announced a program allowing families that agreed to make sustained investments in their children’s higher education to have the balance of their tuition and fees paid for, it would immediately be met with derision as a tool that assists the wealthy at the expense of the poor. Enacting similar policy at the state level is also biased toward some.^[9]

The idea behind free community college is that states and institutions are supposed to make investments that will eventually lead to structural changes to higher education that students, families and policymakers all seek: lower-cost education programs and higher completion rates. Unfortunately, there is no evidence or logical support for the idea that free community college “bends the cost curve” or provides states making substantial financial commitments with assurance or incentives that completion rates will rise as a result.

Conclusion

While proposals to make community college free continue to attract attention, these proposals are expensive and fail to actually help students in need. Additionally, these proposals require states to make investments that they’ve shown a great deal of reluctance committing to while being offered no promise or guarantee of a return on the money spent.

A balanced policy discussion on these proposals must consider the costs and tradeoffs. Not every student would benefit from these proposals and not every state will be willing or able to participate. Where public resources are scarce, the fact that so many students start college but never finish raises important questions about whether existing funding could be more efficiently allocated to achieve more favorable outcomes.

Appendix 1: Calculating Costs and Budgets

• Download

Introduction:

At one point in Alice's Adventures in Wonderland, the eponymous character is lost in the woods, leading to her famous exchange with Cheshire Cat. After the Cat asks Alice where she wants to go, she explains, “I don’t much care where,” and finally she resigns saying, “So long as I get somewhere.” The Cat retorts, “Oh, you're sure to do that, if only you walk long enough.” The Federal Communication Commission’s (FCC) Lifeline subsidy is also wandering without care for goals or a destination.  Now that changes are being considered by the FCC, a more robust review of the program is needed.

Internet access is a key component in our rapidly advancing economy. While incredibly valuable, 15 percent of Americans remain disconnected, mostly by choice.[1] To expand broadband adoption, the FCC plans to extend the Lifeline program to subsidize its purchase by low-income households. Given current eligibility and a lack of targeting mechanisms, we project the program to cost as much as $4.6 billion per year.

The agency should look towards broader and much needed reforms for the program before implementing this expansion. In particular, the FCC should:

• Define the problem that Lifeline aims to solve;
• Cap the budget;
• Reform eligibility requirements;
• Reconsider the current contribution method, which is harmful to the poorest families; and
• Implement an economically rigorous evaluation.

Overview Of The Lifeline Program

New broadband subsidies surely will extend the cost of the program. Although the Commission has committed to expanding Lifeline to include broadband, this expansion should have faced much broader skepticism. As a recent Government Accountability Office report noted, “the Lifeline program, as currently structured, may be a rather inefficient and costly mechanism to increase telephone subscribership among low-income households.”[2]

Lifeline is very much a program rooted in another competitive landscape. Shortly after the AT&T system was broken up, the Lifeline subsidy program was put into place to help low income families afford telephone services. In the years before the breakup, universal service became synonymous with service for all, even though no such provision was written into the original 1934 Communications Act. A decades-long back and forth beginning in the 1930s between federal regulators and state regulators culminated in local costs being shifted to long distance service. The FCC pulled apart some of these geographic cross subsides in the 1980s with the imposition of new charges.[3] Simultaneously, a reconceptualization of government’s role in society and communication policy argued that ubiquitous communications infrastructure would contribute to the national unity and opportunity. Both provided reasons for the creation of Lifeline.

Yet, the FCC has never tried to evaluate just how successful the program is. Even after countless GAO reports suggested this big change, the agency has held that the structure of the program “makes it difficult to determine a causal connection between the program and the penetration.”[4] A growing literature since Garbacz & Thompson (1997) contradicts the official agency view.[5] Independent studies have been conducted of the program for nearly as long as its existence, and almost universally it is found to be economically inefficient and ineffectual in achieving its stated goals.[6] Moreover, previous broadband expansion projects have an unfortunate history of waste, fraud, and abuse. The experience of the Rural Utilities Service serves as an example, where billions of dollars were mishandled and half of the projects they outlined were never finished.[7]

The logic of the subsidy is simple, as the price decreases, consumption goes up. For broadband, the offer of lower prices via the Lifeline program will be fighting forces within the market, which have pushed down prices because of competition. As the market for Internet has developed and prices have dropped, consumers naturally find themselves more willing to buy Internet access. One method to understanding this change are so called price elasticity of demand studies, which estimate how a 10 percent increase in price will affect a change in demand. Below is a chart that shows the trend in these projections over the last 15 years. In the 1990s, a 10 percent change in price would alter demand by nearly 3.5 percent. By 2014, the demand decreased by around .5 percent for a 10 percent increase in price. In all, consumers are becoming less sensitive to price changes.

Various Sources[8]

As the market has developed to become less elastic, the effectiveness of broadband subsidies has greatly diminished. While these price inducements might have been effective a decade ago when the technology was new, Lifeline will increasingly become ineffective as time passes.

Lifeline Reforms

Now that the FCC has decided to reform the Lifeline Program, five major changes need to be pursued. First, the problem that Lifeline aims to solve needs to be defined. Second, the budget needs to be capped. Third, the eligibility requirements require reforms. Fourth, the current contribution method needs to be debated further and reconsidered, as it is harmful to the poorest families. Fifth and finally, Lifeline needs an economically rigorous evaluation process.

Defining the Purpose of Lifeline

As explained in the recent Notice of Proposed Rulemaking,

“The purpose of the Lifeline program is to provide a hand up, not a hand out, to those low-income consumers who truly need assistance connecting to and remaining connected to  telecommunications and information services. The program’s real success will be evident by the stories of Lifeline beneficiaries who move off of Lifeline because they have used the program as a stepping stone to improve their economic stability.”[9]   

Even though the ostensible aim of the program is to ensure that those who cannot pay for the service are able to do so, the FCC has never formally backed up these aims with targets. Only when these two work together can a program effectively operate. Will the program be aimed at reducing the digital divide? How many people are being targeted with the program? Which demographic characteristics would be best suited for this program at making it effective? By which metrics will we know that Lifeline has been a success? How will we separate these program effects out from market shifts?

In various proposals from the White House and the FCC, the problem has mainly been portrayed as a divide between the rich and the poor. However, AAF analysis of Pew data suggests that the real digital divide is between the old and young, which demands a rethinking of the proposed programs and of Lifeline.

In a post arguing for the expansion of Lifeline, FCC Chairman Tom Wheeler explained:

“While more than 95 percent of households with incomes over $150,000 have broadband, only 48 percent of those making less than $25,000 have service at home. A world of broadband ‘haves’ and ‘have-nots’ is a world where none of us will have the opportunity to enjoy the full fruits of what broadband has to offer.”[10]

Similarly President Obama’s proposal, known as ConnectHome, aims to connect more low-income communities to high-speed Internet.[11] While it is undeniable that Internet adoption is higher in households with more income, it is worth investigating what the most significant factors behind non-adoption are. In other words, is the digital divide really a chasm of income?

The Pew Research Center recently released its 2000-2015 study on Internet adoption that provides some fascinating data for answering this particular question. Below is a chart created using Pew data that examines the advances made in Internet adoption by income bracket over the last 15 years. While incredible strides have been made, gaps remain between the rich and poor.

However, this graph alone doesn’t tell the whole story. As has been widely documented, age also correlates strongly with Internet usage and may be a confounding variable in the analysis.[12]

AAF analyzed Pew data to further break down the relationship between age, income, and Internet adoption. As we restricted the age range of the analysis, we found that the income brackets were slowly converging. When looking only at the age range of 18-29, Internet usage is near universal regardless of income bracket.

These findings would seem to indicate that age is a far more significant factor than income when determining whether or not an individual uses the Internet.  Income is, and will continue to be, a contributing factor. But among the demographic that has grown up with and seen the enormous benefits of the Internet, few are being priced out of the broadband market.

This age-based digital divide also puts the Pew “Who’s Not Online and Why” study into perspective.

As their report shows, only 19 percent of offline adults listed price as the primary reason for non-adoption.[13] Meanwhile relevance and usability make up 66 percent,[14] which are the two areas we would expect to see increased considering who the offline population consists of.

These findings have major implications for public policy. In a world of scarce resources and limited government funding, it is important that we fund the programs that would have the greatest impact on the problems we intend to solve. If our goal is to bridge the digital divide and help more Americans access the Internet, it’s hard to see how offering a subsidy without a solid basis in outcomes is the best use of funds.

The “problem” of broadband is often played up as an issue of affordability, but when considering the underlying factors of age, relevance, usability, and computer ownership, it seems unlikely that a subsidy would have a strong effect on closing the digital divide. The lesson of studies performed on wired and wireless telephony support this general conclusion, as does the FCC’s recent study of broadband adoption. Indeed, that study exemplified the problems endemic at the agency at solving the Internet broadband gap, enrolling just over 7,000 people in the pilot program, which 74,000 were expected. Comcast has shown itself to be effective in getting low income families online, as nearly 500,000 families are connected via their Internet Essentials program. [15] Partnering with private companies could help alleviate these issues and go a long way in explaining non-adopters don’t get onto the Internet. Most importantly, the FCC needs explicitly to name those constituencies intended for Lifeline and outline those metrics that will serve to explain when the program has been effective.

Cap the Budget

Capping the budget of the Lifeline program is a vital. Regardless of how targeted the program ends up being, it would be fiscally irresponsible to leave Lifeline uncapped. Lifeline remains the only USF program without a budget cap and consequently the one with the most historic abuse.[16]  While common objections to doing so usually revolve around funding levels, our analysis estimates four levels of cost of the program for both expected usage and the high-end potential, as well as targeted or non-targeted programs. Given current eligibility and no targeting mechanisms, the program has the potential to cost as much as $4.6 billion. If the program reduced eligibility and was targeted it could cost as little as $522 million. 

By outlining eight different scenarios, we can see a range of potential budget options for the program with the three primary factors in the analysis being current or reduced eligibility, targeted or non-targeted, and full or expected Usage. Current eligibility refers to the existing eligibility standards as calculated by the GAO.[17] Reduced eligibility estimates those households under 135% of the poverty line as eligible for subsidies. If no additional methods are used to narrow down the population receiving the subsidy, they are marked as non-targeted, while targeted was determined as only those currently without Internet access among the eligible population receiving the subsidy. Full usage was defined as 100 percent of eligible persons using the subsidies. Expected usage places 33 percent of eligible persons as receiving the subsidies, as reports have suggested that 66 percent of non-adopting households would not subscribe to broadband at any price.[18]

If the Commission is concerned about fluctuations in enrollment due to changing economic conditions, then set a cap that is higher than the expected operating budget. During the 2008 economic collapse, Internet adoption dropped 4 percentage points among households making $30,000 or less. This, however, occurred when the adoption rate was close to 50%. Setting a cap that is 10 percent higher than the expected operating budget would ensure that Lifeline is able to meet the increased demand for broadband subsidies in the event of another recession, but would only work if the agency is diligent in weeding out those who are abusing the system.

Reform Eligibility Requirements,

Currently, individuals can be eligible for Lifeline if they sit below 135% of the poverty line. However, Lifeline might be available if that person qualifies for Medicaid; Supplemental Nutrition Assistance Program (Food Stamps or SNAP); Supplemental Security Income (SSI); Federal Public Housing Assistance (Section 8); Low-Income Home Energy Assistance Program (LIHEAP); Temporary Assistance to Needy Families (TANF); National School Lunch Program's Free Lunch Program; Bureau of Indian Affairs General Assistance; Tribally-Administered Temporary Assistance for Needy Families (TTANF); Food Distribution Program on Indian Reservations (FDPIR); Head Start; or any number of State assistance programs. Because many of these programs use different criteria for determining eligibility, the culmination of all these program leads to an eligible population of 42 million households.[19]

This is roughly a third of the U.S. population that would be eligible. Subsidizing such a large population is not only expensive but is also contrary to the basic goal of Lifeline, which is to get those Americans connected that currently are not. One possible cost cutting measure would be to simplify the eligibility system to only include those families who are below 135% of the poverty line. Our analysis indicates this would cut the eligible population down from 42 million to 28 million households. This was labeled in our budget projections as “Reduced Eligibility”.

One of the most important reforms for Lifeline is ensuring that eligibility is not determined by those providing the communication service, also known as the eligible telecommunications carriers (ETC). Fortunately, it appears that the commission is agreed on this point. We applaud the commission for recognizing the perverse incentives this creates since the ETC’s have little reason to limit their Lifeline subscribers and thus potential income from the program. However, as detailed above, the overall spending for Lifeline can be significantly reduced if the eligible population can be meaningfully targeted. 

Another way to ensure we are targeting the population that most needs this assistance would be to work with existing ETC's to determine who already subscribes to Internet (or has within the last 3 months to prevent gaming the subsidy) and disqualify them from receiving subsidies. This would ensure that we are targeting those who need help crossing the digital divide, not subsidizing those who have already made the leap. We labeled this group as “targeted.” As part of good governance, the FCC should align eligibility with the goals as defined previously. 

Reform the Current Contribution Method

The Universal Service Fund (USF) is structured such that the program subsidizes internally those who cannot afford telecommunications services. Each month, these rates change, and last year they averaged to a 5.82 percent tax on wireless carriers.[20] Paradoxically, like many flat taxes, the burden falls disproportionately onto poor families and young people, who are more likely to use wireless as their primary means of Internet access. According to one study, 80 percent of poor households pay into the fund through taxes only to get nothing back in return.[21] In fact, MIT economics professor Jerry Hausman estimates that the each dollar raised via wireless taxes costs the economy between $0.72 and $1.12, making the USF program a wash.[22]

As a result of the reclassification of broadband, Internet broadband might soon be considered a USF taxable service meaning that the contribution base could expand dramatically. If this does happen, voices calling for a change in the contribution method of all USF programs might subside. Yet, the mechanism is out of line with prevailing tax logic. The USF is akin to the Supplemental Nutrition Assistance Program, and should be treated as one. Even though the USF is funded via a fee, the money collected doesn’t benefit the group from which the levy is collected, like a toll booth helps to pay for a road. The USF benefits those who are not connected to communication services and thus should be understood as general program, and paid for through the normal tax structure.[23] Although it is not the purview of the FCC to make this kind of change, Lifeline reforms should make the program more explicit and more closely tied to Congressional approval, as it is legally mandated with the power of taxing. 

Implement Meaningful Evaluations of Lifeline

The FCC has a responsibility to ensure the programs they institute are having a net positive effect. Due to the inefficiencies of government bureaucracy and the weak effect subsidies will have in the current market, it is vital that the FCC perform a cost-benefit analysis to ensure they are doing more harm than good.

This problem is further complicated when examining the essentially meaningless goal the FCC has set for itself. The GAO report on Lifeline strongly urged the commission to set measurable goals as a concrete way to evaluate the performance of the program. In the 2012 NPRM on Lifeline reform, the FCC claimed to have fixed the problem by setting the goal of increased broadband penetration of low-income consumers. The problem came when the commission decided how to measure this goal:

“As with our first goal, as an outcome measure of the availability of broadband service to low-income consumers, we adopt the broadband penetration rate of low-income consumers, i.e. the extent to which low-income consumers are subscribing to broadband. Progress towards our goal of ensuring the availability of broadband service to low-income consumers will be indicated by a narrowing of the difference between this outcome measure and the broadband service penetration levels of non-low-income consumers in the “next highest income” bracket”. [24]

Without choosing a specific benchmark or threshold of broadband penetration, there is no way the FCC can’t meet the goal. Natural market forces will continue to push up adoption rates at all income levels, and seeing as most income brackets are already at saturation levels, the gap between the low-income bracket and the next highest income bracket will naturally become smaller over time.

FCC should conduct a basic analytical projection of the proposed broadband program based on current Lifeline adoption levels for wired and wireless telephone service before taking any further steps with the program. In particular, there are number of regressions that could help to pinpoint these targets.[25] From here, a specific level of broadband penetration above the 3-year average growth of adoption for low-income consumers can be established, thus creating an actionable goal.

Conclusion

The FCC has not done enough to set goals and establish mechanisms for evaluation of the Lifeline program. In particular, the FCC needs to

• Define the problem that Lifeline aims to solve;
• Cap the budget;
• Reform eligibility requirements
• Reconsider the current contribution method, which is harmful to the poorest families; and
• Implement an economically rigorous evaluation.

Only when these changes are implemented will the program truly begin to serve those whom should be receiving Lifeline assistance.

• Download

Test

Introduction

The Centers for Medicare and Medicaid Services (CMS) have finalized a rule regarding the Medicare reimbursement methodology for biosimilar products. Biosimilars are prescription medications which have been approved by the Food and Drug Administration (FDA) as being “highly similar” to a specific biologic medication (known as the reference product).[1] Thus, it is easiest for most people to think of biosimilars as the equivalent of generics for small molecule brand name medications, though this is not scientifically accurate. While small molecule generics are chemically identical (save for potentially any inactive ingredients) to their respective brand name drugs—and can be because they are chemically manufactured—exact copies of biological products, by their nature of being developed from living organisms, cannot be produced,[2] and patients may respond differently to the reference product and the biosimilars.

In recognition that biosimilars are not simply generic versions of biologicals, the Biologics Price Competition and Innovation Act was included in the Affordable Care Act (ACA) to establish a different pathway for approval through the FDA and a different methodology for reimbursement from CMS. The law directs that biosimilars be paid based on 100 percent of their volume-weighted Average Sales Price (ASP) plus 6 percent of their reference product’s ASP. (For comparison, small molecule drugs are paid 106 percent of the volume-weighted ASP of all brand name and generic versions of a particular drug.) Biosimilars were given this more favorable add-on to the reimbursement rate in recognition that they are more expensive to produce than small molecule generics. However, there is now a debate as to whether or not biosimilars should be paid with a single billing code based on the ASP of all biosimilars for a single reference product, as has been finalized by CMS, or the ASP for each individual biosimilar, separately from any other biosimilar of the same reference product. Economic arguments and patient safety concerns may support the latter, though the statutory text regarding this matter is somewhat ambiguous.

Biosimilars are Not Generics

There are several issues to consider when weighing these options. Biosimilars are not exactly the same as their reference biologic by definition. Biosimilars are also not required to treat all the same indications and conditions as the reference product. Therefore, one biosimilar may treat all or most of the same conditions as the reference product, while another biosimilar for that same reference product may only treat one indication. Further, even if two biosimilars of a single reference product treat the same indications, they are not necessarily biosimilar to each other (this would require a separate determination from the FDA). Thus, the biosimilar products are not necessarily equally effective or valuable and therefore should not be reimbursed the same amount. Reimbursing these drugs equally does not allow for fair compensation of higher quality products or the fact that one biosimilar may treat more indications than another, which presumably increases its cost of development. If price is the only factor on which companies are competing, a downward spiral may result, pushing prices to an unsustainably low level, eventually forcing high-value manufacturers to exit the market. As competition is reduced, prices could eventually be allowed to rise for the few companies left standing, even if those products are not those most preferred by physicians and patients.  The sterile injectable generic drug experience indicates that drug shortages can occur once the price has bottomed out.[3] The net effect of this policy, therefore, could be that the thriving biosimilars market that policy makers were expecting in the US might not materialize, and patients would not get the benefit of more treatment options and lower prices. 

Further, because biosimilars will not be exactly the same, they cannot be used interchangeably (a drug must specifically receive “interchangeable” status by the FDA to qualify as an acceptable substitute without physician’s orders). However, some are worried that physicians may inappropriately prescribe such drugs interchangeably if the reimbursement amount is the same. This could happen either because the equal reimbursement inaccurately signals to the physician that the drugs are interchangeable, or because the price of a drug changes whereby a physician could lose money by prescribing one biosimilar which in turn encourages the physician to change the prescription. Conversely, physicians have the responsibility of ensuring they understand the differences between the drugs they prescribe, and drug manufacturers should take care to ensure this information is readily available.

Difficulty Tracking Adverse Events

Another concern expressed by some is that using a single billing code will not allow for adequate pharmacovigilance to track and monitor adverse events among patients. If this is true, there is certainly reason to adjust this policy. Patients deserve to know that the biosimilars they are taking can be appropriately tracked to their respective manufacturer should there be a problem with a specific biosimilar.

Despite some claiming that the small molecule drug industry, which uses a single billing code (HCPCS), does not have problems tracking adverse events using the National Drug Code (NDC) assigned to each product along with the manufacturer lot number and company name for each batch of drugs produced, research by the Food and Drug Law Institute has found that adverse events for small molecule drugs are likely often misattributed to the innovator drug.[4] This happens because reports may only contain a nonproprietary name which results in drugs with the same nonproprietary name being grouped together.[5] Further, the FDA’s Adverse Event Reporting System (FAERS) database does not have a data field for an NDC number and the manufacturer lot number was only reported 10 percent of the time which would arguably make it difficult to rely on these pieces of information for tracking adverse events.[6]

Additionally, because biosimilars are not exactly the same as their reference product, as generics are for their brand-name drug, biosimilars may have a different clinical profile from the reference product and therefore potentially greater differences in the side effects, making the consequence of misattribution for biosimilars potentially much more severe. Finally, while biosimilars in the European Union, Canada, Australia, and Japan are all identified by the International Nonproprietary Name (INN) given by the World Health Organization (WHO), and those countries do not report problems tracking adverse events, WHO has stated that the current approach may not suffice as more biosimilars enter the market, at which point more distinct identifiers may be necessary.[7] In a report published by the National Center for Biotechnology Information, the authors state “the effectiveness of [current] surveillance methods may be compromised when there are multiple manufacturers of products that share common drug nomenclature or coding.”[8]

Some argue that the FDA naming practice, in which a unique suffix will be provided for each drug manufactured and added to the product’s proper name, will allow for sufficient pharmacovigilance. Others contend the FDA naming policy will not suffice, as a drug’s various naming and identifying components are not always used in billing and the FDA’s new Sentinel Initiative, to more effectively and accurately track adverse events electronically, will rely on multiple sources of information, including claims data.[9] The bottom line is the more differentiating factors, the better. The Secretary of Health and Human Services does not want to be in the position of finding out that the practice is insufficient after an adverse event occurs and the effected patients cannot be identified.

The Statutory Text is Unclear

Finally, there is disagreement over interpretation of the statutory text which prescribes how biosimilars should be reimbursed by CMS. The reimbursement formula contains two parts: 100 percent of the ASP of the biosimilar product (the base payment) plus 6 percent of the ASP of the reference product (an add-on payment, typically provided, much like a dispensing and administrative fee). There is no disagreement over the part of the statute that states the 6 percent add-on should be calculated solely based on the volume-weighted ASP of the reference biological product, and that the ASP of the reference product should not be included in the base payment for biosimilar products, which is clearly and explicitly provided in §3139(a)(1)(B) of the ACA:[10][11]

Disagreement arises over CMS’ reading of how to calculate the base payment. The text states that payment is the sum of:

                    ``(A) the average sales price as determined using

                the methodology described under paragraph (6) applied to

                a biosimilar biological product for all National Drug

                Codes assigned to such product in the same manner as

                such paragraph is applied to drugs described in such

                paragraph; and

                    ``(B) 6 percent of the amount determined under

                paragraph (4) for the reference biological product (as

                defined in subsection (c)(6)(I)).'';”

CMS has interpreted this language to mean that all NDCs of all biosimilars with the same reference product should be combined to generate one shared ASP, just as all NDCs of small molecules (generic and brand name) are combined to create a single shared ASP. This position is supported by the explicit exemption of the reference product from the calculation of the base ASP, which is only necessitated if CMS’s interpretation is correct, and by the ACA’s reference to paragraph (6), which sets payment policy for multiple source drugs, as opposed to paragraph (4), which prescribes the payment amount for single source drugs or biologicals.

However, this same language is interpreted by others to mean that all NDCs of a single biosimilar product (each biosimilar may have multiple NDCs based on formulary variations, dosing, or other minor differences) should be combined to generate that product’s unique ASP. They rely on the textual reference to NDCs of a single “product”, and the financial differences between biosimilar and small molecule manufacturing (discussed supra) to support this position. If the latter interpretation is correct, biosimilar reimbursement would more closely resemble reimbursements for biologics than those for small molecule generics.

While Congress’ intent in using this particular language is unknown, in cases of statutory interpretation such as this, the administrative—in this case, CMS’—interpretation is accepted as correct.

Conclusion

Ultimately, the decision over how biosimilars, or any medication, should be reimbursed by CMS should be determined by economic principles based on the value of the medication to patients and without putting patient safety and access to such products at risk. If the statutory text does not clearly provide for the favorable regulatory outcome of these factors, it should be amended. Because the rule finalized by CMS would provide equal reimbursement for all biosimilars of a single reference product, without regard for the relative value of such products based on factors such as the number of indications they are approved to treat or the side effects they cause, higher-value products may get squeezed out of the market. Even if CMS has appropriately interpreted the statute in development of this rule and in a consistent manner with how reimbursement policies have been determined for other drugs based on the same section of the law, the underlying policy appears to be flawed. This reimbursement methodology will likely lead to undesirable economic impacts, could stifle a fledgling biosimilars marketplace, and also lead to potential issues related to patient safety and access to appropriate treatment options, and therefore Congress should proactively fix this flawed approach.

• Download

The American Action Forum (AAF) studied the surge in Bakken oil production, its impact on the environment, and finds air quality in the region unchanged. Generally, as U.S. and Bakken oil production has increased, air quality has actually improved. Oil production has increased in the region by 1,450 percent, while more than 90 percent of the air quality days in the region remain “good,” a trend unchanged since 2008. In addition, these economic boons are in concert with falling greenhouse gas emissions in the U.S.

Oil Production Boom

 For the past several years, new techniques to extract oil and natural gas have led to surging production across the U.S., including the Bakken shale formation in North Dakota and parts of eastern Montana. The graph below charts the incredible growth in the region’s production, measured in millions of barrels per day.

As the data reveal, the Bakken region has increased output from 0.08 million barrels per day in January 2008, to 1.16 million by August 2015, a 14.5-fold uptick. During this time, the nation as a whole has also witnessed incredible growth in output. The graph below details U.S. oil production from 2008 to 2014, measured in thousands of barrels a day.

Although the Bakken gains have been more pronounced, as shown above, the U.S. has still increased production by 74 percent, from just over five million barrels a day to more than 8.7 million in 2014. This expansion has helped to drive down gas prices for consumers at the pump and contributed to the realistic possibility of U.S. energy independence for the first time in more than a generation. However, as noted, this growth has increased cries that the fracking and oil boom has led to declining environmental quality. The data reveal these claims as mostly hollow.

Air Quality in Bakken

EPA labels air quality on a given day by the following: “good,” “moderate,” “unhealthy for sensitive groups,” “unhealthy,” “very unhealthy,” and “hazardous.” As AAF examined earlier this year, air quality is generally improving across the U.S. and the Bakken has enjoyed cleaner air as well. EPA generally defines “good” air days as the following: “air pollution poses little or no risk.” In the Bakken region, despite the oil production boom, more than 90 percent of year is labeled “good” by EPA. See the graph below:

Throughout the history of the oil boom in the Bakken, the percentage of the year with “good” air quality days has consistently stayed above 90 percent, with more than 94 percent considered “good” in 2014, a slight improvement above the 2013 figure (91 percent). The rate of “unhealthy” days in the Bakken remains low, with less than three last year, down from nine in 2012. There is actually a positive correlation (albeit, the correlation coefficient is just 0.36) between rising Bakken oil production and the overall number of “good” air quality days in the region.

Compare the Bakken’s results to upstate New York’s Marcellus region, which has banned fracking. In upstate New York, the average jurisdiction has never experienced more than 90 percent “good” air days in a year. These jurisdictions have averaged roughly 83 percent, compared to 93 percent for the Bakken. Increased oil and natural gas operations certainly don’t directly result in cleaner air, but there is little evidence in the Bakken area that production is causing poor air quality days for local residents.

Although “good” air quality days are a general indicator that the levels of environmental pollution are well within acceptable limits, AAF also examined smog in the Bakken region. EPA data tracks days throughout the year when ground-level ozone (smog) is the dominant pollutant. Even by that metric, the Bakken region has made slight improvements. The chart below tracks days in which ozone is the dominant pollutant for the average jurisdiction in the Bakken (with linear trendline).

As the data reveal, smog has gradually declined as the dominant pollutant in the area. In 2008, it was the main pollutant for 294 days; that figure dropped to 270 days by 2014, a decline of eight percent. This improvement is hardly earth-shattering, but it helps to dispel any notions that oil production in the Bakken is associated with dangerous air quality.

Greenhouse Gas Emissions

In addition to the positive news of energy independence and strong air quality, the amount of greenhouse gas (GHG) released into atmosphere continues to decline. This, despite the renaissance in domestic energy production. The graph below track carbon dioxide emissions from 2005 to 2013.

As noted above, U.S. GHG emissions have declined from six billion tons in 2005 to roughly 5.5 billion tons by 2013, a drop of nine percent. Even Montana, located partially in the Bakken region, has managed to reduce emissions nine percent during the same period. Likewise, Pennsylvania, which has drastically increased its natural gas production, has decreased carbon dioxide emissions by eleven percent from 2005 to 2013.

Conclusion

Energy booms are not incompatible with clean air and a strong environmental record. The Bakken region increased oil production 1,450 percent, all while maintaining “good” air quality for more than 90 percent of the year. The U.S. economy may not have to choose between energy production and environmental stewardship after all.

In recent years, policymakers and labor advocates have proposed raising the minimum wage at federal, state, and local levels. On the local level, these efforts have been somewhat successful as a number of major cities have enacted minimum wage increases over the last few years. These cities include San Francisco and Seattle, which both enacted laws to implement a “living wage” of $15 per hour. Beginning this year, several major cities took steps to phase in these large minimum wage increases. Now that we are almost to the end of 2015, these cities provide a new source of evidence to evaluate the labor market implications of raising the minimum wage. American Enterprise Institute’s Mark J. Perry has been tracking employment patterns both in Seattle and in the rest of the state of Washington, finding that restaurant employment in the city has been falling since the beginning of the year while it has been rising in the rest of the state. But, how have the rest of the major metropolitan areas fared? The facts show that this year restaurant employment in the major metropolitan areas with minimum wage increases has consistently experienced slower growth than restaurant employment in the rest of their states.

2015 Major Cities with Minimum Wage Increases

This year, the seven major U.S. cities that implemented minimum wage increases are Chicago, Louisville, Seattle, Washington, Oakland, San Francisco, and San Jose. These minimum wage increases are detailed in Table 1.

Table 1: Major City Minimum Wage Increases in 2015

Most of these minimum wage increases are the beginning of much larger minimum wage hikes. For instance, San Francisco’s and Seattle’s minimum wage increases were the first steps to increasing the cities’ minimum wages to $15. In other cities, the minimum wage implementation is much further along. San Jose raised its minimum wage from $8 to $10 in 2013. Since then, the minimum wage in San Jose continues to rise with inflation. This year it increased by only 15 cents from $10.15 to $10.30. Overall, the San Francisco Bay Area has been full of cities implementing substantial minimum wage increases. Besides San Francisco and San Jose, the cities in the Bay Area that are phasing in minimum wage hikes are Berkeley, Emeryville, Mountain View, Oakland, Sunnyvale, and Richmond.

How to Evaluate Employment Trends

As noted in a previous American Action Forum (AAF) paper, when evaluating the impact of the minimum wage it is necessary to ask the right question. For instance, evaluating the effect of the minimum wage on total employment trends would generally yield irrelevant results because only about 2 percent of all wage and salary workers earn at or below the federal minimum wage. It is very unlikely that a minimum wage increase would affect the majority of workers who earn significantly above the minimum wage. As a result, any analysis that examines the entire workforce is likely to understate the labor market consequences of raising the minimum wage because the vast majority of workers do not earn low wages.

Thus, in order to fairly examine the labor market consequences of increasing the minimum wage, it is essential to examine low-wage workers who actually earn at or slightly above the current minimum wage. This includes young, low-skilled workers, the exact population policymakers are trying to help by increasing the minimum wage. One way to zero in on the low-wage population is to examine recent job growth trends in an industry that actually employs low-wage workers. This paper examines 2015 job growth in restaurants, specifically food services and drinking places, which in 2013 employed 48.78 percent of all workers who earned at or below the federal minimum wage. Food services and drinking places continue to employ low-wage workers today, as the average hourly pay rate of production and nonsupervisory workers in that industry was only $11.51 in September 2015. Evaluating job growth in food services and drinking places provides a valid idea of how recent minimum wage hikes are impacting job creation for low-wage workers.

Methodology

To analyze trends in city restaurant employment relative to the rest of the state, we use the most localized data for each metropolitan area available in the Federal Reserve Bank of St. Louis’s Federal Reserve Economic Data (FRED). For each city, we use seasonally adjusted food services and drinking places employment data in the metropolitan statistical area (MSA) or metropolitan divisions (MD), illustrated below.

Table 2: Metropolitan Areas Analyzed

Although we do not directly analyze the cities that raised the minimum wages (except in the case of Washington, DC), analyzing employment trends in the entire metropolitan areas may still provide useful information on the impact of the minimum wage on low-wage employment. In addition, by evaluating the metropolitan areas, we are also able to take into account trends in smaller localities that are following in the footsteps of the major cities by raising their own minimum wages. This is particularly true in the San Francisco Bay Area where a large number of smaller towns surrounding the major cities have enacted their own minimum wage increases. For instance, the Oakland-Hayward-Berkley MD includes Oakland and Berkeley, both of which began phasing in minimum wage increases. In addition, the San Jose-Sunnyvale-Santa Clara MSA includes San Jose, Sunnyvale, and Mountain View, which all increased their minimum wages to $10.30 this year.

For each metropolitan area, we measure the growth in food services and drinking places employment in 2015. Consequently, we measure the percent change in food services and drinking places employment from December 2014 to September 2015, the most recent month with available data. We do the same for the state that the metropolitan area is located in and exclude the metropolitan area itself. Since Washington, DC is not located in a state, we compare restaurant employment growth the District to the growth in Virginia and Maryland combined.

We then combine food services and drinking places employment in all of the metropolitan areas and combine employment in all of the metropolitan areas’ surrounding states. This allows us to compare restaurant employment growth in all the metropolitan areas that raised the minimum wage together to restaurant employment growth in all of the rest of the states together.

Results

The chart below shows that restaurant employment in metropolitan areas with cities that raised the minimum wage in 2015 has lagged behind employment in the rest of the states.

Clearly, restaurant employment has grown far slower this year in cities that raised the minimum wage than in the rest of the states in which those cities are located. This year restaurant employment in the metropolitan areas with major cities that raised the minimum wage only grew 1.1 percent through September. In the surrounding state areas, however, restaurant employment grew 2.8 percent.

Table 3 shows the percent growth in restaurant employment in each metropolitan area and in the rest of the area’s state.

Table 3: 2015 Growth in Restaurant Employment: Metropolitan Area vs. Rest of the State

Restaurant employment in the major cities that raised the minimum wage this year has consistently been slower than the rest of the states in which the cities are located. Restaurant employment in Louisville grew 2.6 percent, 1.1 percentage points lower than restaurant employment growth in the rest of Kentucky. In Seattle, where the minimum wage is increasing to $15 per hour, restaurant employment has only grown 0.6 percent this year while restaurant employment in the rest of the state of Washington grew 6 percent.[1] In the other city that is phasing in a $15 minimum wage, San Francisco, restaurant employment only grew 1.4 percent and in the rest of California it grew 3.2 percent.

There are also a few exceptions in the major cities we examine. In particular, restaurant employment in Chicago grew at about the same rate as in the rest of Illinois (0.5 percent versus 0.4 percent). Also, restaurant employment in Oakland grew faster than the rest of California (4.1 percent versus 3.2 percent). Despite the faster growth in Oakland, however, restaurant employment in the three major cities of the San Francisco Bay Area only grew 1.9 percent. That is 1.7 percentage points slower than the 3.6 percent growth in restaurant employment experienced in the rest of California.

In Washington, DC, meanwhile, restaurant employment fell 0.6 percent. In Maryland and Virginia, it increased 0.9 percent. Interestingly, Washington, DC, is the only city that we are able to directly measure the change in restaurant employment in the District itself (rather than the metropolitan area) and it by far has the most negative employment trends. In all other cities we measure the change in restaurant employment in the metropolitan regions where these cities are located, which include towns that did not increase the minimum wage. As a result, it is possible that the employment trends in the smaller towns also included in metropolitan areas measured could be making the employment growth in the metropolitan regions more positive than what is actually being experienced in the cities that raised their minimum wage this year.

Conclusion

While cities continue to phase in large minimum wage hikes, the verdict is still out on how these policies will impact local employment. The early data, however, do indicate a slowing of restaurant employment growth. As these steep minimum wage increases continue to be implemented, it is possible that these trends may become more worrisome over time.

Nearly two years ago, the Affordable Care Act (ACA) implemented a new individual insurance marketplace, along with the promise of stability and affordability. The new individual marketplace was to rival that of the employer sponsored insurance (ESI) marketplace in stability and predictability, while premiums were to rise at rates much lower than the historical average. This paper seeks to evaluate the degree to which these promises have been fulfilled heading into year three. In this study we find that the cost of both the benchmark Silver plan and the lowest cost Bronze plan will increase by 10 percent in 2016. 

Data and Methodology

The primary sources of data for this report are the 2014, 2015, and 2016 individual market medical landscape files that are available through the Centers for Medicare and Medicaid Services (CMS). These landscape files contain data on the health insurances plans that were offered through the federally-facilitated or federal-state partnership exchanges, which include 36 states in 2014, 37 states in 2015, and 38 states in 2016—only 35 states are included in all three datasets.

Breakaway Policy and the Robert Wood Johnson Foundation maintain a dataset on Silver plans offered in all states in 2014, which we use to calculate premiums for 2014 benchmark Silver plans in non-federal exchange states. In order to obtain information on 2015 and 2016 silver plans that were not included in the data available through CMS, we use plan compare tools on state-based exchanges where they are available. Overall, we analyze data on 2014, 2015, and 2016 Silver plans for 43 states, which account for 461 of the 501 rating areas nationwide. Our estimates on the number of health insurance issuers in a given rating area are also based on Silver plan data.  

We use plan compare tools on state-based exchanges to obtain information on 2015 and 2016 Bronze plans not included in the federal data. However, we were unable to include data on 2014 Bronze plans offered outside of federally facilitated or federal-partner exchanges. As a result, our analysis of Bronze plans is restricted to the 36 states that were included in the federal exchange data in 2014, which account for 407 of the 501 rating areas nationwide. In computing average effects across rating areas, we calculate a weighted average of the effects in each rating area using the potentially eligible population. The potentially eligible population is defined as the number of individuals who are either uninsured or insured through the individual market, ineligible for Medicaid or the Children’s Health Insurance Program, and determined to be a legal resident. We estimate this population using the 2010-2012 American Community Survey. All of the premium estimates in this report are based on those offered to a 27-year-old non-smoker. Unless otherwise noted, the premium changes do not include any subsidies that are available for middle-to-low income households. In our analysis of subsidized 2014 benchmark plans, we consider the post-subsidy premium for a 27-year-old non-smoker that earns $30,000 per year. Where we consider the implications for families, we calculate the premium for a married couple of 27 year olds with two children. We also use both the 2014 income contribution scale and 2014 federal poverty guidelines for calculating the premium subsidy for all three years. This is a simplification that ignores a slight increase in both the federal poverty guidelines and the income contribution scale that roughly offset each other in terms of the overall effect on the premium subsidy for an individual whose income does not change. More importantly, we assume that household income remains the same. An estimate that included some increase in income would lead to a reduction in premium subsidy.

Determining Premium Growth

Each state market operates under different conditions and costs of health insurance vary between rating areas of different states. There are also a variety of plans within each rating area that consumers can purchase, and the enrollment numbers for each of those plans will fluctuate according to several factors. Additionally, risk protections that were guaranteed by the ACA (a guarantee that has proven empty), the growing pool of insured, and changes in the plans offered can all affect premiums in more indirect ways. Averages mask a lot of this variation; therefore, it is necessary to look at the national insurance marketplace from various viewpoints if any legitimate analysis of premiums is to be made. 

In the discussion of health insurance costs, the premiums for the second-lowest cost Silver plan, referred to as the benchmark Silver plan, is often the focus because of its high enrollment numbers—roughly 11 percent of the individual market. This year, in 461 observed rating areas, the average premium for the benchmark Silver plan increased by 10 percent. However, many of those plans have changed. Roughly 71 percent of the 2015 benchmark plans will not remain benchmark plans in 2016, with many of those plans leaving the market all together. Of the 2015 benchmark plans that are still offered, but are not necessarily still the benchmark plan, in 2016, there has been an average premium increase of 10 percent. Since 2014, the average cost of the benchmark has grown by 13.5 percent. 

While the average cost of benchmark Silver plans serves as a quick way to gauge the cost of health insurance, averages mask variation.  Non-benchmark Silver plans enrolled roughly 59 percent of the marketplace in 2015, while Bronze plans—the cheapest metal level on the market—enrolled about 20 percent. Premiums for the lowest cost Bronze plan have increased by 10 percent between 2015 and 2016, and nearly 14 percent since 2014. Using these lower-cost plans as a judge may give us a better understanding of how people are affected throughout the market because lower cost plans are more likely to enroll those ineligible for subsidies and cost sharing. 

Average rate increases disguise substantial variation across the country, which is divided into 501 different rating areas. For example, we estimate that the lowest cost Bronze plans increased by about 10 percent on average while 20 of the 461 rating areas observed saw an average 10 percent decrease. On the other hand, 85 of 461 observed rating areas saw increases of over 25 percent. There is similar variation in the premiums of 2016 benchmark plans as premiums will increase by more than 30 percent in some areas while others will see decreases of to 20 percent or more.

Though averages only tell part of the story, it is noteworthy that the increases in both benchmark Silver plans and lowest cost Bronze plans are already approaching the 8 to 12 percent increases that afflicted the individual market before 2014. Increases of this size were promised to stop under the ACA.

Plan Turnover and its Impacts

Last year, 53 percent of consumers enrolled in a 2014 plan shopped around for a different plan in 2015. This year, the administration is encouraging people to do the same. Those people who do not shop around for a new plan will be auto-enrolled in the plan for which they are currently enrolled or a similar cost plan if their current plan is not available. For example, of those who are currently enrolled in a 2015 benchmark Silver plan many will have to enroll in a new plan in 2016. This turnover in benchmark plans—and all the other plans on the market—is problematic because it affects both premiums and overall cost of insurance as subsidies change accordingly. Another unfortunate outcome of this turnover is the possibility of people being forced to switch health care providers as networks change with plans.

In 321 of the 461 rating areas, the benchmark plan has changed, and many are no longer offered. As mentioned earlier, the average increase in premiums of 2015 benchmark plans was 10 percent, and when subsidies are considered, premiums for those eligible increased 6.7 percent. In many rating areas, a rise in competition—introduced by new entrants and the process of bidding for the benchmark plan—can lead to a decrease in the benchmark premium even as premiums rise for most of the other offered plans. As this happens, many people in those rating areas will see premiums grow while subsidies decrease. In areas where there are only two insurers, average benchmark premiums increased by 13 percent while premiums increase by 8.3 percent on average in those areas where there were more than two insurers.

In the employer sponsored insurance market, consumers can expect a consistent premium growth average of 6 percent without needing to worry about switching plans year to year. The change in benchmark plans from year to year also indicates the high level of flux in the market.  This too illustrates the amount of uncertainty consumers face with an unstable market.

Conclusion

It is important to remember that the current marketplace structure provided by the ACA is relatively new. In many ways, the health insurance marketplace is just as unpredictable as it was a year ago. In regards to cost, it is marked by disparities between states and rating areas. There is still a significant amount of movement among insurers as they enter and exit markets, introduce new plans, and remove old plans. Because of these factors, it will still take time to pinpoint the underlying cost growth in the individual market.

• Download

In previous work, AAF focused on the online and traditional gig economy by exploring the changes in this labor force since the economic downturn.[1] But within this broad class, a narrower category of networked technologies has taken the spotlight to become a constant focus for policymakers, critics, and presidential hopefuls. These technologies meld peer-to-peer networks on both the production and consumption sides to sell goods and services.[2] While labor practices have been a constant topic of conversation, many have lost sight of the primary benefit of new network-based service technologies, and the chance that it affords policymakers. Disruptive technologies bring new and expanded services to consumers. As consumers switch and efforts are made to place the new services into the old regulatory buckets, the costs of the old system are highlighted and policymakers have the opportunity to rebalance asymmetric regulations. The burdens of the regulatory system are extensive but often go unseen. While there has been a lot of concern about how to regulate disruptive technologies, consumers and governments stand to benefit from these innovative changes and a hands off approach that incentivize their creation. This paper highlights both the state of the modern gig economy, its consumer benefits, and the cost and affiliated burdens imposed by regulators.

How Networked Technologies Are Bringing Benefits to Consumers

Disruptive technologies and innovative business models benefit consumers because they internalize the power of markets, thus bringing to bear their power to help transform a service. Markets are characterized by three broad features: they must attract a sufficient proportion of potential market participants both as buyers and sellers to come together; they must mitigate the problem of congestion by allowing participants alternative possibilities or making the markets clear quickly; and finally, they must make it safe to participate in the market as simply as possible.^[3] Disruptive network technologies employ advanced communication methods to help solve the first and third issues, while also bringing into the market new participants and protection mechanisms.

Broadly speaking, what unites these new technologies are four broad features:[4]

1.        Enabled by advanced communication networks, new technologies reduce transaction costs and allow new trades to occur.

2.      These trades often occur with assets that otherwise will not be used, thus bringing “dead capital” back to work for consumers.

3.      These new organizational forms push services from isolated exchange to market exchange.

4.      Review systems and other dynamic policing mechanisms instill trust into the both side of the bargain.

The reduction of transaction costs

For some time, the existence of companies posed a problem for economists. When consumers want a good or service, why is it that they cannot just go to the market and contract for it? In other words, why are there so many companies? As Nobel Prize winning economists Ronald Coase and Oliver Williamson explained, there are huge costs that one has to incur to participate in the market, to find out who has the good or service, to make a deal, and to ensure that the provider follows through on their agreement. Thus, companies capitalize on what economists have come to call transactions costs. The nature of firms is intimately tied with the transactions costs within the market.   

Advanced networked technologies such as Internet purchasing, mobile app commerce, etc., have dramatically reduced the cost of interacting, bargaining, and monitoring, allowing for the creation of new kinds of organizations. Thus, with the internalization of market processes, trade in services and goods is expanded. In turn, consumers have flocked to these networked services because of the added convenience, lower prices, and higher quality services. This new organizational form first took hold with the Linux operating system and has become especially well known via the Wikipedia project. Decentralized exchanges and cheap communication over the Internet allowed for programmers to connect to collectively edit and advance Linux, while those same tools have allowed for Wikipedia editors to collect and publish over 5 million English language pages.^[5] The biggest names in this space, including Uber, Lyft, and Airbnb operate in industries with large barriers to entry, which also might be understood as having high initial setup costs, a kind of transaction cost.

While today these platforms are transforming specific services, there are countless smaller players that have created their own niche market that could not have been possible before ubiquitous communication. Two of the more well-known organizations include Kiva, which allows people to lend money via the Internet to low-income and underserved entrepreneurs and students across 82 countries, and Kickstater, which crowdfunds creative projects. These patronage platforms evince how powerful networks can be if they utilize a vast network of suppliers, who have committed relatively small funds, to affect a targeted domain of consumers. The case of financing is, however, just a small part of a much larger change in how assets are being deployed in the marketplace. 

Resurrecting dead capital

At any given time, there are countless assets not being used. In the US, about 1.5 rooms exist for every person, cars will often sit idle in parking lots during business hours, and countless other durable goods will go unused for long swaths of time.^[6] Because these assets cannot easily be bought, sold, valued or used as an investment, they are often called dead capital.^[7] Reducing the cost to transact means that these assets can be brought back into the market to be bought and sold, creating beneficial exchanges for both consumers and producers. While much of the focus has centered around Uber, Lyft, and Airbnb, it is likely that niche markets for specific goods and services like tools, boats, musical instruments, and extra factory space will see the biggest growth in upcoming years.^[8]  

Consider Airbnb. Renting out spare rooms and couches resembles the once common practice of "boarding out." Boarding was especially popular in the early 1900s with lower-income and recently immigrated families to supplement income and help assimilate newcomers, who were often within that family’s social network.^[9] Boarding out was prevalent, but fell out of favor as regulation of boarding homes increased. Airbnb has brought back these kinds of services to consumers, reemploying extra rooms into the economy.

There are a number of clear benefits from expanded services. For one, consumers and producers have greater choices. One of the biggest advantages of more choice could come in the form of narrowing of prices. Thus, while prices will not always decrease, it could be that the absolute range of prices will be narrowed around an average.^[10]

As methodological techniques advance and data is collected, it is likely that the real benefits will be conferred to the owners of the asset. Early data from Airbnb and other online marketplaces suggests that the introduction of these peer-to-peer production firms has the effect of substituting rentals for ownership, thus conferring onto these new owners the financial benefits that go along with ownership. Moreover, used-good prices do tend to see drops that come hand in hand with consumer surplus increases.^[11] As a recent paper on this topic explained,

Consumption shifts are significantly more pronounced for below-median income users, who also provide a majority of rental supply. Our results also suggest that these below-median income consumers will enjoy a disproportionate fraction of eventual welfare gains from this kind of ’sharing economy’ through broader inclusion, higher quality rental-based consumption, and new ownership facilitated by rental supply revenues.[12]

While still early, these results are promising and suggest that both sides of the bargain actually win.

The transition from isolated to market exchange

By reducing transaction costs and reintroducing dead capital back into the market, new technologies are able to transform isolated exchanges, a one-to-one interaction, into market exchanges, that brings both customer and producer knowledge to a wider audience. In the parlance of economists, search costs are reduced and the market is expanded.

The concept of isolated exchange is one that dates back to Aristotle’s work on economics.[13] In this kind of exchange, the deal is limited to those in a close network, often among friends, and is not standardized, nor does it take into consideration other players in the market. The knowledge imbued in prices that comes from extensive and liquid markets is simply not the same when those interactions are just one offs. Transitioning from these one off trades to a more fluid market exchange has serious benefits, especially for consumers. As experiments in this space have shown, in some cases only four buyers and four sellers are needed to bring about competitive outcomes, but only if the structure of those exchanges is a market.^[14]

Limited knowledge about demand for a service has long been a limiting factor in the expansion of some services. Airbnb has built a massive back catalogue of rooms in areas that have extra spaces but aren’t typically in the same places that hotels aggregate. Indeed, this is a driving factor behind the company. Instead of a typical hotel room, many flock to the site because it offers a unique and comfortable space to connect with new people and rest.^[15]

Taxicabs have long been criticized for not providing service in some neighborhoods and to some economic groups. In part, this is due to knowledge problems in that the demand for a product or service exists, but isn’t known to exist. According to one analysis of actual Uber rides, the expansion of the company has largely benefited low-income New Yorkers, who live in the bottom half of New York City median household income.^[16] While there has always been a demand for these services in underserved neighborhoods, knowledge of that demand can only now be quantified, pushing service to expand into these new areas. Indeed, in a survey of consumers of these products, the third most cited reason for use is that the product or service cannot be elsewhere.^[17]

Personal reputation and trust enabled by reviews

Like other kinds of trade, economic exchange enabled by new tech is built on a system of trust, which is supported via rating systems and electronic verification techniques. The sophistication of these rating systems enables the ecosystem and is clearly changing the nature of consumer regulation of bad actors. Paradoxically, while consumers have lost trust in large companies, peer-to-peer networks and social selling might be adopted in part because they are more psychologically appealing as they provide an experience that is distant from bureaucratic organizations.[18]

It is a mistake to think that online reputational systems are simple affairs. As should be expected, buyers consider reviews and use them for a variety of purposes when purchasing goods. Moreover, “negative reviews do hurt sellers, but that a negative reputation hurts more than a positive one helps on some dimensions but not on others.”^[19] Also, online sellers are able to distinguish themselves by replicating unique selling points highlighted by successful brands.

Combined, these forces act as constraints on businesses as evidenced in research on Yelp, a service rating web site. For one, it has been found that a one star increase in Yelp’s five star rating system leads to between a 5 and 9 percent increase in revenue for independent restaurants. For chain restaurants, however, ratings seem to not have an appreciable effect on revenue. However, as Yelp penetration has expanded, the market share of chain restaurants has diminished.^[20] In other words, online consumer reviews are standing in for more traditional forms of reputation.

Yet those traditional forms of reputation are also undergoing changes, mirroring one of the oldest forms of selling, direct selling. Direct selling is built on relationships, distinct from physical and online retail outlets. As such, the performance of the service or good is being continually monitored because there is a close connection between the buyer and seller. Now, those kinds of connections and the resulting consumer insights are being extended online into online social channels via technology, fostering closer experiences. New platforms create the spaces for communication and personal brand management that had largely gone unnoticed apart from direct sellers. Thus, the constant two-way communication imbued in networks has expanded the scope of this important constraint into other methods of selling.   

This more robust knowledge ecosystem has had direct effects on taxicabs much like Yelp reviews. Because there is extensive data provided by Uber for use, there are some interesting relationships to be found in their rise and effects on the quality of local taxis. After controlling for a number of other factors, Uber’s expansion is connected with a decline in consumer complaints per trip about taxis in New York. In Chicago, there is a similar story. The networked service’s growth is associated with a decline in broken credit card machines, air conditioning and heating, rudeness, and talking on cell phones.^[21]

But for taxicabs, this is a two way street. As a profession, taxi drivers are subject to some of the most significant levels of workplace violence, up to 33 times higher than average. Nonfatal assaults, homicides, robberies, and verbal abuse are among the highest for taxicab drivers, while the perpetrators of the less serious actions typically face few legal consequences. With Uber, personal reputation is quantified. Moreover, because there is no cash exchange, the possibility for robbery is severely diminished, and if there are problems, identifying information on the rider is kept which allows for ex post enforcement. All combined, the safety of the driver is better assured with Uber.

The Unseen Cost of Regulation

While innovation in business models for independent businesses has enhanced economic opportunity for sellers and quality of service options for consumers, changes in the marketplace have also given rise to new efforts to regulate these firms. The laws some want to apply to these upstarts are meant to ensure safety, quality, and transparency. Yet, as the above section highlights, desirable outcomes for both consumers and producers are occurring with these networked technologies. As has been detailed for some time, the laws meant to ensure consumer protections act as hurdles for new entrants, imposing costs on the economy, producers and consumers.

However, it is difficult to calculate just how extensive this drag is. Two problems plagued past studies of regulation. For one, specific and narrow areas of regulation were often studied, like environmental or telecommunication regulation, and thus limited in scope. On the other hand, more generalized methods, like the World Bank’s Ease of Doing Business Index, quantified the regulatory burden in a comparative fashion among regions.^[22] While this method has proven useful to researchers in making comparisons and connecting this data with economic growth figures, the downside is that the absolute cost to consumers is difficult to ascertain. Efforts to calculate the total burden of government mandates have been taken up in recent years, and are taking a number of different forms.

One way to get at the cost of regulation can be found in AAF’s Regulation Rodeo, which compiles the cost of final rules as published in the Federal Register. The innovation of this program lies in data collection and ease of access, as it compiles official figures. According to the total implemented rules so far this year, consumers will ultimately have to bear $90 billion and 48 million paperwork hours.^[23] However, this method doesn’t include all of the proposed rules, which when summed for the whole of 2014 came to a grand total of $181.5 billion.^[24]

Another method of articulating the cost of regulation uses textual analysis to count all those instances in the Code of Federal Regulations (CFR) where individuals or companies are constrained in their actions.^[25] Using this method, researchers found that the total number of restrictions in the entire CFR rose from about 835,000 in 1997 to over 1 million in 2012.^[26]

Both of these projects help to illuminate the extensive unseen costs built into markets that occur when the federal government either requires or restricts some kind of action on the part of companies or individuals, which in turn gets placed as a burden on consumers. Though it takes many forms, what unites disruptive technologies and business models is that they are disrupting regulatory regimes and lowering the barriers to entry.

The reduction of taxicab medallion prices across the country provides a visceral example. Before Uber and Lyft, driving a taxicab in Chicago required that you first pay the city $357,000 for a taxicab medallion.^[27] For New York City, the cost to get into taxis was much higher, at right around $1 million.^[28] In Boston, the price of a medallion is $625,000, while in San Francisco, medallions were selling for $300,000 with the city taking a $100,000 cut.^[29] Those prices are falling dramatically. In Chicago, the price fell to about $270,000 and the number of medallions being transferred also dropped dramatically. In New York city, the cost of these medallions were down about 25 percent at the beginning of 2015, and the relatively liquid market has come to a standstill.^[30]

So why did the medallions cost so much and why are they now falling? Only a certain number of medallions were ever available at one time, so a medallion’s cost represents in part the going rate of exclusion also known as an entry barrier. With those entry barriers coming down, so too has the cost of the medallion. The story is the same for nearly every city. At first, there was free entry and exit into the taxicab market. Then, as a result of efforts to regulate the industry, the medallion system was put into place. Only those with a medallion could drive a taxi. When the first set of medallions were stamped in the 1930s, those grandfathered into the system got a windfall. As economists say, the asset (medallion) is capitalized and the costs to acquire it are baked in. But from then on, anyone who wants to enter into the market has to pay for access to the market by purchasing or leasing a medallion. Once competition enters into the market, the asset holders are left with real asset decreases, leading to what Buchanan called into a transitional gains trap.

And yet, there isn’t a huge difference between the medallion system and other regulations meant to affect quality, and quantity of service. Occupational licenses, which aim to ensure these consumer benefits, are proving to be another sad story of regulatory overreach, affecting about 1 out of every 3 jobs, compared to less than 5 percent in the early 1950s.^[31] Both the medallion system and occupational licensing schemes act as a fixed cost and deter entry, thus increasing prices for consumers and keeping people out of certain professions. Only those that can jump through the regulatory hoops are able to secure their benefits, which has real and immediate impact. While the goals of these regulations are often laudable, the implementation imposes inefficiencies in the market, including limiting consumer choice, raising consumer costs, increasing practitioner income, limiting practitioner mobility, and depriving the poor of adequate services.^[32] The White House has even chastised the broad nature of these regulations, noting that:

Too often, policymakers do not carefully weigh these costs and benefits when making decisions about whether or how to regulate a profession through licensing. In some cases, alternative forms of occupational regulation, such as State certification, may offer a better balance between consumer protections and flexibility for workers.^[33]

It is important to note that the real benefit of these regulations – improved quality of service – is often hard to find in the data.^[34] So not only do these laws impart costs, they do so in a way that does not improve the underlying services. By bringing together consumers and producers of services and goods, networked technologies and independent business models not only place pressure on this clunky regulatory system, but also benefit consumers via the introduction of new services.

Conclusion

New technologies are upending the typical role of regulation. In a previous generation, it was regulation that protected consumers. Now reciprocal rating systems and transparency provide consumers critical buying information, giving all of us a chance to rethink why we regulate. Importantly, review and rating systems come as a result of expanding networks, which have brought a bevy of benefits to consumers and producers. Legislators need to be acutely aware of the benefits of change as they craft policy and work to ratchet down laws already on the books. Moreover, many services are growing and are being funded because there are opportunities separate from the traditionally regulated economy. In their zeal to protect and tax, governments all too often get in the business of picking winners and losers. New waves of products are forcing cities to justify their policies. Here’s hoping the innovators win.     

In previous work, AAF focused on the online and traditional gig economy by exploring the changes in this labor force since the economic downturn.[1] But within this broad class, a narrower category of networked technologies has taken the spotlight to become a constant focus for policymakers, critics, and presidential hopefuls. These technologies meld peer-to-peer networks on both the production and consumption sides to sell goods and services.[2] While labor practices have been a constant topic of conversation, many have lost sight of the primary benefit of new network-based service technologies, and the chance that it affords policymakers. Disruptive technologies bring new and expanded services to consumers. As consumers switch and efforts are made to place the new services into the old regulatory buckets, the costs of the old system are highlighted and policymakers have the opportunity to rebalance asymmetric regulations. The burdens of the regulatory system are extensive but often go unseen. While there has been a lot of concern about how to regulate disruptive technologies, consumers and governments stand to benefit from these innovative changes and a hands off approach that incentivize their creation. This paper highlights both the state of the modern gig economy, its consumer benefits, and the cost and affiliated burdens imposed by regulators.

How Networked Technologies Are Bringing Benefits to Consumers

Disruptive technologies and innovative business models benefit consumers because they internalize the power of markets, thus bringing to bear their power to help transform a service. Markets are characterized by three broad features: they must attract a sufficient proportion of potential market participants both as buyers and sellers to come together; they must mitigate the problem of congestion by allowing participants alternative possibilities or making the markets clear quickly; and finally, they must make it safe to participate in the market as simply as possible.^[3] Disruptive network technologies employ advanced communication methods to help solve the first and third issues, while also bringing into the market new participants and protection mechanisms.

Broadly speaking, what unites these new technologies are four broad features:[4]

1. Enabled by advanced communication networks, new technologies reduce transaction costs and allow new trades to occur.

2. These trades often occur with assets that otherwise will not be used, thus bringing “dead capital” back to work for consumers.

3. These new organizational forms push services from isolated exchange to market exchange.

4. Review systems and other dynamic policing mechanisms instill trust into the both side of the bargain.    

The reduction of transaction costs

For some time, the existence of companies posed a problem for economists. When consumers want a good or service, why is it that they cannot just go to the market and contract for it? In other words, why are there so many companies? As Nobel Prize winning economists Ronald Coase and Oliver Williamson explained, there are huge costs that one has to incur to participate in the market, to find out who has the good or service, to make a deal, and to ensure that the provider follows through on their agreement. Thus, companies capitalize on what economists have come to call transactions costs. The nature of firms is intimately tied with the transactions costs within the market.   

Advanced networked technologies such as Internet purchasing, mobile app commerce, etc., have dramatically reduced the cost of interacting, bargaining, and monitoring, allowing for the creation of new kinds of organizations. Thus, with the internalization of market processes, trade in services and goods is expanded. In turn, consumers have flocked to these networked services because of the added convenience, lower prices, and higher quality services. This new organizational form first took hold with the Linux operating system and has become especially well known via the Wikipedia project. Decentralized exchanges and cheap communication over the Internet allowed for programmers to connect to collectively edit and advance Linux, while those same tools have allowed for Wikipedia editors to collect and publish over 5 million English language pages.^[5] The biggest names in this space, including Uber, Lyft, and Airbnb operate in industries with large barriers to entry, which also might be understood as having high initial setup costs, a kind of transaction cost.

While today these platforms are transforming specific services, there are countless smaller players that have created their own niche market that could not have been possible before ubiquitous communication. Two of the more well-known organizations include Kiva, which allows people to lend money via the Internet to low-income and underserved entrepreneurs and students across 82 countries, and Kickstater, which crowdfunds creative projects. These patronage platforms evince how powerful networks can be if they utilize a vast network of suppliers, who have committed relatively small funds, to affect a targeted domain of consumers. The case of financing is, however, just a small part of a much larger change in how assets are being deployed in the marketplace. 

Resurrecting dead capital

At any given time, there are countless assets not being used. In the US, about 1.5 rooms exist for every person, cars will often sit idle in parking lots during business hours, and countless other durable goods will go unused for long swaths of time.^[6] Because these assets cannot easily be bought, sold, valued or used as an investment, they are often called dead capital.^[7] Reducing the cost to transact means that these assets can be brought back into the market to be bought and sold, creating beneficial exchanges for both consumers and producers. While much of the focus has centered around Uber, Lyft, and Airbnb, it is likely that niche markets for specific goods and services like tools, boats, musical instruments, and extra factory space will see the biggest growth in upcoming years.^[8]

Consider Airbnb. Renting out spare rooms and couches resembles the once common practice of "boarding out." Boarding was especially popular in the early 1900s with lower-income and recently immigrated families to supplement income and help assimilate newcomers, who were often within that family’s social network.^[9] Boarding out was prevalent, but fell out of favor as regulation of boarding homes increased. Airbnb has brought back these kinds of services to consumers, reemploying extra rooms into the economy.

There are a number of clear benefits from expanded services. For one, consumers and producers have greater choices. One of the biggest advantages of more choice could come in the form of narrowing of prices. Thus, while prices will not always decrease, it could be that the absolute range of prices will be narrowed around an average.^[10]

As methodological techniques advance and data is collected, it is likely that the real benefits will be conferred to the owners of the asset. Early data from Airbnb and other online marketplaces suggests that the introduction of these peer-to-peer production firms has the effect of substituting rentals for ownership, thus conferring onto these new owners the financial benefits that go along with ownership. Moreover, used-good prices do tend to see drops that come hand in hand with consumer surplus increases.^[11] As a recent paper on this topic explained,

Consumption shifts are significantly more pronounced for below-median income users, who also provide a majority of rental supply. Our results also suggest that these below-median income consumers will enjoy a disproportionate fraction of eventual welfare gains from this kind of ’sharing economy’ through broader inclusion, higher quality rental-based consumption, and new ownership facilitated by rental supply revenues.[12]

While still early, these results are promising and suggest that both sides of the bargain actually win.

The transition from isolated to market exchange

By reducing transaction costs and reintroducing dead capital back into the market, new technologies are able to transform isolated exchanges, a one-to-one interaction, into market exchanges, that brings both customer and producer knowledge to a wider audience. In the parlance of economists, search costs are reduced and the market is expanded.

The concept of isolated exchange is one that dates back to Aristotle’s work on economics.[13] In this kind of exchange, the deal is limited to those in a close network, often among friends, and is not standardized, nor does it take into consideration other players in the market. The knowledge imbued in prices that comes from extensive and liquid markets is simply not the same when those interactions are just one offs. Transitioning from these one off trades to a more fluid market exchange has serious benefits, especially for consumers. As experiments in this space have shown, in some cases only four buyers and four sellers are needed to bring about competitive outcomes, but only if the structure of those exchanges is a market.^[14]

Limited knowledge about demand for a service has long been a limiting factor in the expansion of some services. Airbnb has built a massive back catalogue of rooms in areas that have extra spaces but aren’t typically in the same places that hotels aggregate. Indeed, this is a driving factor behind the company. Instead of a typical hotel room, many flock to the site because it offers a unique and comfortable space to connect with new people and rest.^[15]

Taxicabs have long been criticized for not providing service in some neighborhoods and to some economic groups. In part, this is due to knowledge problems in that the demand for a product or service exists, but isn’t known to exist. According to one analysis of actual Uber rides, the expansion of the company has largely benefited low-income New Yorkers, who live in the bottom half of New York City median household income.^[16] While there has always been a demand for these services in underserved neighborhoods, knowledge of that demand can only now be quantified, pushing service to expand into these new areas. Indeed, in a survey of consumers of these products, the third most cited reason for use is that the product or service cannot be elsewhere.^[17]

Personal reputation and trust enabled by reviews

Like other kinds of trade, economic exchange enabled by new tech is built on a system of trust, which is supported via rating systems and electronic verification techniques. The sophistication of these rating systems enables the ecosystem and is clearly changing the nature of consumer regulation of bad actors. Paradoxically, while consumers have lost trust in large companies, peer-to-peer networks and social selling might be adopted in part because they are more psychologically appealing as they provide an experience that is distant from bureaucratic organizations.[18]

It is a mistake to think that online reputational systems are simple affairs. As should be expected, buyers consider reviews and use them for a variety of purposes when purchasing goods. Moreover, “negative reviews do hurt sellers, but that a negative reputation hurts more than a positive one helps on some dimensions but not on others.”^[19] Also, online sellers are able to distinguish themselves by replicating unique selling points highlighted by successful brands.

Combined, these forces act as constraints on businesses as evidenced in research on Yelp, a service rating web site. For one, it has been found that a one star increase in Yelp’s five star rating system leads to between a 5 and 9 percent increase in revenue for independent restaurants. For chain restaurants, however, ratings seem to not have an appreciable effect on revenue. However, as Yelp penetration has expanded, the market share of chain restaurants has diminished.^[20] In other words, online consumer reviews are standing in for more traditional forms of reputation.

Yet those traditional forms of reputation are also undergoing changes, mirroring one of the oldest forms of selling, direct selling. Direct selling is built on relationships, distinct from physical and online retail outlets. As such, the performance of the service or good is being continually monitored because there is a close connection between the buyer and seller. Now, those kinds of connections and the resulting consumer insights are being extended online into online social channels via technology, fostering closer experiences. New platforms create the spaces for communication and personal brand management that had largely gone unnoticed apart from direct sellers. Thus, the constant two-way communication imbued in networks has expanded the scope of this important constraint into other methods of selling.   

This more robust knowledge ecosystem has had direct effects on taxicabs much like Yelp reviews. Because there is extensive data provided by Uber for use, there are some interesting relationships to be found in their rise and effects on the quality of local taxis. After controlling for a number of other factors, Uber’s expansion is connected with a decline in consumer complaints per trip about taxis in New York. In Chicago, there is a similar story. The networked service’s growth is associated with a decline in broken credit card machines, air conditioning and heating, rudeness, and talking on cell phones.^[21]

But for taxicabs, this is a two way street. As a profession, taxi drivers are subject to some of the most significant levels of workplace violence, up to 33 times higher than average. Nonfatal assaults, homicides, robberies, and verbal abuse are among the highest for taxicab drivers, while the perpetrators of the less serious actions typically face few legal consequences. With Uber, personal reputation is quantified. Moreover, because there is no cash exchange, the possibility for robbery is severely diminished, and if there are problems, identifying information on the rider is kept which allows for ex post enforcement. All combined, the safety of the driver is better assured with Uber.

The Unseen Cost of Regulation

While innovation in business models for independent businesses has enhanced economic opportunity for sellers and quality of service options for consumers, changes in the marketplace have also given rise to new efforts to regulate these firms. The laws some want to apply to these upstarts are meant to ensure safety, quality, and transparency. Yet, as the above section highlights, desirable outcomes for both consumers and producers are occurring with these networked technologies. As has been detailed for some time, the laws meant to ensure consumer protections act as hurdles for new entrants, imposing costs on the economy, producers and consumers.

However, it is difficult to calculate just how extensive this drag is. Two problems plagued past studies of regulation. For one, specific and narrow areas of regulation were often studied, like environmental or telecommunication regulation, and thus limited in scope. On the other hand, more generalized methods, like the World Bank’s Ease of Doing Business Index, quantified the regulatory burden in a comparative fashion among regions.^[22] While this method has proven useful to researchers in making comparisons and connecting this data with economic growth figures, the downside is that the absolute cost to consumers is difficult to ascertain. Efforts to calculate the total burden of government mandates have been taken up in recent years, and are taking a number of different forms.

One way to get at the cost of regulation can be found in AAF’s Regulation Rodeo, which compiles the cost of final rules as published in the Federal Register. The innovation of this program lies in data collection and ease of access, as it compiles official figures. According to the total implemented rules so far this year, consumers will ultimately have to bear $90 billion and 48 million paperwork hours.^[23] However, this method doesn’t include all of the proposed rules, which when summed for the whole of 2014 came to a grand total of $181.5 billion.^[24]

Another method of articulating the cost of regulation uses textual analysis to count all those instances in the Code of Federal Regulations (CFR) where individuals or companies are constrained in their actions.^[25] Using this method, researchers found that the total number of restrictions in the entire CFR rose from about 835,000 in 1997 to over 1 million in 2012.^[26]

Both of these projects help to illuminate the extensive unseen costs built into markets that occur when the federal government either requires or restricts some kind of action on the part of companies or individuals, which in turn gets placed as a burden on consumers. Though it takes many forms, what unites disruptive technologies and business models is that they are disrupting regulatory regimes and lowering the barriers to entry.

The reduction of taxicab medallion prices across the country provides a visceral example. Before Uber and Lyft, driving a taxicab in Chicago required that you first pay the city $357,000 for a taxicab medallion.^[27] For New York City, the cost to get into taxis was much higher, at right around $1 million.^[28] In Boston, the price of a medallion is $625,000, while in San Francisco, medallions were selling for $300,000 with the city taking a $100,000 cut.^[29] Those prices are falling dramatically. In Chicago, the price fell to about $270,000 and the number of medallions being transferred also dropped dramatically. In New York city, the cost of these medallions were down about 25 percent at the beginning of 2015, and the relatively liquid market has come to a standstill.^[30]

So why did the medallions cost so much and why are they now falling? Only a certain number of medallions were ever available at one time, so a medallion’s cost represents in part the going rate of exclusion also known as an entry barrier. With those entry barriers coming down, so too has the cost of the medallion. The story is the same for nearly every city. At first, there was free entry and exit into the taxicab market. Then, as a result of efforts to regulate the industry, the medallion system was put into place. Only those with a medallion could drive a taxi. When the first set of medallions were stamped in the 1930s, those grandfathered into the system got a windfall. As economists say, the asset (medallion) is capitalized and the costs to acquire it are baked in. But from then on, anyone who wants to enter into the market has to pay for access to the market by purchasing or leasing a medallion. Once competition enters into the market, the asset holders are left with real asset decreases, leading to what Buchanan called into a transitional gains trap.

And yet, there isn’t a huge difference between the medallion system and other regulations meant to affect quality, and quantity of service. Occupational licenses, which aim to ensure these consumer benefits, are proving to be another sad story of regulatory overreach, affecting about 1 out of every 3 jobs, compared to less than 5 percent in the early 1950s.^[31] Both the medallion system and occupational licensing schemes act as a fixed cost and deter entry, thus increasing prices for consumers and keeping people out of certain professions. Only those that can jump through the regulatory hoops are able to secure their benefits, which has real and immediate impact. While the goals of these regulations are often laudable, the implementation imposes inefficiencies in the market, including limiting consumer choice, raising consumer costs, increasing practitioner income, limiting practitioner mobility, and depriving the poor of adequate services.^[32] The White House has even chastised the broad nature of these regulations, noting that:

Too often, policymakers do not carefully weigh these costs and benefits when making decisions about whether or how to regulate a profession through licensing. In some cases, alternative forms of occupational regulation, such as State certification, may offer a better balance between consumer protections and flexibility for workers.^[33]

It is important to note that the real benefit of these regulations – improved quality of service – is often hard to find in the data.^[34] So not only do these laws impart costs, they do so in a way that does not improve the underlying services. By bringing together consumers and producers of services and goods, networked technologies and independent business models not only place pressure on this clunky regulatory system, but also benefit consumers via the introduction of new services.

Conclusion

New technologies are upending the typical role of regulation. In a previous generation, it was regulation that protected consumers. Now reciprocal rating systems and transparency provide consumers critical buying information, giving all of us a chance to rethink why we regulate. Importantly, review and rating systems come as a result of expanding networks, which have brought a bevy of benefits to consumers and producers. Legislators need to be acutely aware of the benefits of change as they craft policy and work to ratchet down laws already on the books. Moreover, many services are growing and are being funded because there are opportunities separate from the traditionally regulated economy. In their zeal to protect and tax, governments all too often get in the business of picking winners and losers. New waves of products are forcing cities to justify their policies. Here’s hoping the innovators win.     

Introduction

In 2011, two million Americans were infected with—and 23,000 killed by—drug-resistant bacteria.[1] Over the past 20 years incidents of drug-resistant infections acquired in hospitals have increased from a low of 10 percent of all hospital acquired infections to 60 percent today.[2] Despite the evident need for new, novel antibiotics, almost no major pharmaceutical companies are investing in antibiotic development. Only 11 new antibiotics were approved between 1998 and 2014, and no new classes of antibiotics have been approved since 1987.[3]

What’s Going Wrong?

Antibiotics developers, like all pharmaceutical manufacturers, are market actors sensitive to incentives. Currently there are very few incentives to invest in antibiotic research and development (R&D) and many reasons to direct those funds elsewhere.

Antibiotics are Expensive to Produce

Because we rarely face entirely new bacteria, antibiotic development is often intended to better combat a well-known bacterium that has adapted defenses against currently available treatments through repeated exposures. Bacteria’s ability to evolve quickly renders once therapeutic drugs useless against later generations.[4] Scientists must meet this challenge by developing either a more powerful generation of an existing drug, or a new drug with a unique method of attacking for which the bacteria has no adaptation.[5] Unfortunately, both these options—especially the latter—are expensive and scientifically complex, and therefore do not create much incentive for investment in R&D relative to other drugs that can more readily be improved upon.

Antibiotics Provide Low Return on Investment

Antibiotics offer drug manufacturers a particularly low return on their investment. They are intended to be single use drugs that cure the condition they treat.[6] Even an expensive antibiotic is unlikely to earn profits equal to a cheap treatment for a chronic disease; which must be purchased regularly throughout the patient’s lifetime.

Unlike other classes of drugs, antibiotics are often their own worst enemy financially. The more often an antibiotic is prescribed and used, the more opportunities the bacterium has to evolve and develop a resistance to the drug.

FDA Approval Process is Long and Expensive

Pathways to Food and Drug Administration (FDA) approval are slow and difficult. By the time a new antibiotic comes to market, the strain of bacteria it treats may have had over a decade to develop its resistance; particularly if the new drug is an update of a similar drug, rather than a novel form of attack.

Another important market influence acting against antibiotic R&D is the seeming irrationality of consumers. Bacterial threats to individual and population health are not as immediate as they once were, and deaths caused by uncomplicated bacterial infections are still relatively rare.[7] Public consciousness is currently more focused on “scarier” threats like cancer and heart disease, and most people do not take the threat of untreatable, drug-resistant bacteria seriously. This lack of public understanding of the seriousness of the situation causes consumers in situations that are not immediately life-threatening to undervalue antibiotics – they are unwilling to pay more for new treatments for old bugs, despite the substantial R&D costs that go into producing dugs that the bacteria are not resistant to. In fact, at the time of discovery, the average net value of a new antibiotic for the developer today is a loss of $50 million.[8]

What’s Being Done About It?

The slowdown in production of new and effective antibiotics is a clear public health threat. The FDA and Congress have both taken steps to address the problem, but most promising responses have unsurprisingly come from the free market.

Government Response

The FDA has released guidance directing farmers raising animals for consumption to alter their use of antibiotics that are approved for human use. The FDA is asking these farmers to voluntarily limit their use of antibiotics to therapeutic purposes under the oversight of a licensed veterinarian.[9] Though this guidance is voluntary, it appears that most livestock operations are willing to comply.[10] This is an important step in minimizing the opportunities bacteria have to develop resistances to drugs that are important to human health.

In 2012, Congress passed the Food and Drug Administration Safety and Innovation Act (S. 3187), which among other provisions, included the Generating Antibiotic Incentives Now (GAIN) Act.[11] The GAIN Act allows for more antibiotics that treat serious and life threatening infections to have an extended five-year exclusivity period beyond that of other drugs.[12] This law attempts to counterbalance the disincentives of a long FDA approval process and low returns on R&D investment, but may also drive up the price of these drugs by preventing competition from generic manufacturers. Nonetheless, the scientific reality is that antibiotics are still liable to lose their effectiveness (and therefore value) over time.

Congress is also currently considering the Promise for Antibiotics and Therapeutics for Health (PATH) Act to minimize the negative effects of the long FDA approval proves.[13] If enacted, this bill would allow for an expedited approval pathway for antibiotics for use in narrowly defined patient populations with serious, specific illnesses and an unmet need for treatment. These developers would later be able to apply for an expanded approval and the drug could be used off label if appropriate.

Private Sector Response

The response to the vacuum in antibiotic production has been most impressive in the private sector. The rise in new tests and specialized firms can both prevent and meet the need for new drugs.

The development of rapid diagnostics and changes in technology of Laboratory Developed Tests can provide health care providers with accurate diagnoses quicker.[14] By providing a diagnosis in hours rather than days, better diagnostic tests can enable the use of antibiotics before an infection has time to spread. It will also help reduce misuse of antibiotics.[15] These benefits will help slow the development of drug-resistant strains of infectious bacteria.

Specialized drug development firms and academic institutions with outside grant funding that focus exclusively on the development of antibiotics have largely replaced major pharmaceutical companies in the field.[16] With no other drugs on the market, small firms can focus exclusively on R&D. Once a drug has been developed, firms may apply for FDA fast-track approval for small, targeted populations of the sickest patients.[17] If successful, this business model encourages the sale of the patent to large drug companies with the resources to manufacture the drug on a large scale, and follow through on the full FDA approval process. By eliminating manufacturing and trial costs, small developers can generate a modest profit developing otherwise profitless antibiotics.[18]

• Download

Two new Federal Reserve rulemakings in the last month have pushed Dodd-Frank’s aggregate financial costs past $35 billion, with $29.3 billion in final rules. With more than five years of implementation, it’s clear that regulators still have dozens of new measures left to propose and finalize. The most expensive portions of the law could ultimately lie ahead for consumers.

Federal Reserve’s Actions

Broadly, the Federal Reserve occupies a somewhat discrete place in the regulatory world. It doesn’t consume the oxygen of the Environmental Protection Agency or even the Securities and Exchange Commission. On its own authority, it has imposed roughly $300 million in costs pursuant to Dodd-Frank, in addition to 4.8 million paperwork burden hours. These might seem like pedestrian figures, but it’s important to recognize that it would take more than 2,400 employees working full-time (2,000 hours a year) to complete the Federal Reserve’s new paperwork. Now, two rulemakings, one jointly authored by the Federal Reserve and one issued independently, have only added to the agency’s Dodd-Frank portfolio.

Recently, a quintet of financial regulators issued a final rule implementing a Dodd-Frank measure on “Margin and Capital Requirements for Covered Swap Entities.” Despite initial projections of costs ranging up to $46 billion, both the proposed “Regulatory Impact Analysis” (RIA), and final RIA centered on a cost figure of $5.2 billion; it now represents the most expensive rule under the landmark financial reform bill in its five-year history.

Given that there were five agencies (Comptroller of the Currency, Federal Reserve, FDIC, Farm Credit Administration, and Federal Housing Finance Agency) involved in the formation of this rulemaking, the proposed rule stretches to 281 pages, but the analysis from the proposed to final version is almost identical. Both proposed and final rules estimated $644 billion of initial margin would be required.  As with many of the estimates in the analysis, the range of figures is questionable. For example, the margin required could be as low as $280 billion or as high as $3.6 trillion, but the agencies chose $644 billion because it “falls roughly in the middle.” This is only true, if you have a creative definition of “roughly” and “middle.” The mathematical midpoint of the above range is $1.94 trillion or three times the central figure the agencies selected.

The midpoint figure of margin dictates a baseline for calculating compliance costs. As the rule notes, “Using our initial margin estimate of $644 billion, every basis point of lower return equals $64 million. If we assume a 100 basis point difference in return, the opportunity cost of the fully phased-in initial margin requirement would be $6.4 billion.” This 100 basis point is the high-end range of opportunity costs; at 25 basis points, the costs total only $2.9 billion. A discounted figure of $5.2 billion is the primary benchmark for the rule’s cost, in addition to more than 52,000 paperwork burden hours.

The baseline for cost arises from three different global studies conducted in 2013. In order to approximate the U.S. share, the regulators merely reduced the amounts in the study by 65 percent, since the U.S. share of the swaps market is roughly 35 percent. In other words, the main benefit-cost estimates rest on assumptions piled on assumptions to yield an incredibly uncertain final calculation.

Strangely, there is no direct discussion of high-end costs in the RIA, just the preamble to the final rule. This could be due to the OCC’s sole drafting of the RIA. In the actual text of the rule, the cost scenarios are more explicit: “The estimated annual costs of the initial margin requirements range from $672 million to roughly $46 billion depending on the specific initial margin estimate and incremental funding cost that is used to compute the estimate.” A 68-fold difference in cost is hardly the level of precision desired from the nation’s premier financial regulators implementing one of the most important provisions of Dodd-Frank. It is even more peculiar that the $46 billion figure was omitted from both the proposed and final RIA.

What does this mean for the nation? The map below attempts to approximate the regional impact of the margin rule based on the affected establishments and the assumptions contained in the cost estimate.

Given the rule’s wide disparities in costs, it’s safe to assume the benefit figures will be equally lacking. The $644 billion margin is considered the prime quantified benefit, just as it contributes to the costs. In the words of regulators, it “creates a new buffer against losses for covered swap entities and other swap market participants.” In addition to this quantified benefit, regulators included four non-quantified benefits, including 1) reduced uncertainty, 2) limiting participants from taking excessive risks, 3) added protection against insolvency of counterparties, and 4) decreased risk associated with swaps.

Finally, on October 30, the Federal Reserve issued additional requirements for “Systemically Important U.S. Bank Holding Companies,” or “Too Big to Fail” institutions. Banks with this systemically important tag will be required to maintain a minimum amount of unsecured long-term debt. Under section 165 of Dodd-Frank, this measure generally applies to all institutions with more than $50 billion in assets.

The cost of this “Too Big to Fail” oversight measure, if finalized, would make it the seventh costliest Dodd-Frank measure to date, at $1.5 billion. To derive this figure, regulators found the covered banks had a capital shortfall of roughly $120 billion, approximately “1.7 percent of aggregate risk-weighted assets.” To replace this shortfall, the Federal Reserve estimated it would cost between 20 and 200 basis points, depending on the type of debt. This translates into a cost range of $680 million to $1.5 billion annually. For perspective, if finalized in FY 2014, it would have been the most expensive rule, according to the Office of Information and Regulatory Affairs.

In terms of benefits, the “Too Big to Fail” proposal projects that it would, “materially reduce the risk that the failure of a covered [bank] would pose to the financial stability of the United States.” The text does not quantify a benefit, but notes that prior to Dodd-Frank, the odds of major financial crisis were between 3.5 percent and 5.2 percent. Presumably, this rule alone won’t significantly lower those odds, only in concert with other rules addressing risk will Dodd-Frank reduce the chances of another crisis.

Dodd-Frank’s Mounting Burdens

Thanks in part to the two new rules mentioned, Dodd-Frank’s aggregate costs have climbed substantially since AAF’s last update on the law’s five-year anniversary. Since July, the law’s costs have increased to $29.3 billion, from $24 billion. Paperwork has also accelerated to 72.8 million, up from just 61 million in July. Those are only the final rule costs; combined with proposals that will likely be final, and the Dodd-Frank’s burdens climb to $35 billion in costs and 75 million paperwork burden hours.

What do these huge figures mean for the economy? Someone must comply with this paperwork and pay for the costs. Whether it’s employees at affected entities, shareholders, or eventually consumers, everyone generally pays for the cost of federal regulation. On the compliance side, there is strong evidence these costs now translate into more compliance officers, solely responsible for filling out forms and ensuring companies follow the new law. For example, in the “Finance and Insurance” industry, the number of compliance officers had increased 13 percent since 2010 and 20 percent since 2005. Meanwhile, wages paid to those compliance officers have risen 24 percent since 2005 (unadjusted for inflation). The graph below tracks the growth of compliance officers in the “Finance and Insurance” industry.

The effects of Dodd-Frank extend beyond just finance, however. In addition to the anecdotal evidence that the law has driven many industries from certain profit-making fields to compliance-driven, there is strong quantitative evidence that federal regulation has contributed to a surge in compliance officers nationwide. Since 2010, the growth in compliance officers nationwide has topped 21 percent. For comparison, from July 2010 to July 2015, total nonfarm payroll in the U.S. increased by just 8.9 percent. In other words, the nation is generating new compliance officer jobs more than 2.3 times faster than the average nonfarm job.

This is hardly the end of the law’s impact. Even five years after passage, the latest Unified Agenda of federal regulation from the administration logs at least 80 scheduled Dodd-Frank regulations pending. According to the data, the Securities and Exchange Commission has the largest slate, with 22 rulemakings. The Federal Deposit Insurance Commission and the Consumer Financial Protection Bureau are second and third with 11 and 10 pending regulations. It’s clear that after litigation, implementation of Dodd-Frank could take another five years, with economic impacts for decades in the future.

Consumer Impact

Far beyond the walls of compliance offices and law firms, Dodd-Frank is reaching into family bank accounts and small businesses with its burdens and costs. The purpose of Dodd-Frank was partly to stem abuses and fix systemic weaknesses in the financial sector that came about when the housing bubble burst and helped cause the financial crisis. But, in its efforts to make the financial system safer, Dodd-Frank has restricted the availability of many financial products and credit, particularly for low-income borrowers, young people, and minorities.

Lending, whether commercial, real estate, or consumer, has not recovered in this economic recovery as quickly as the average post-WWII economic recovery due to the severity of the financial crisis and environment of tightened credit encourage by Dodd-Frank. Real estate lending has been the slowest to recover, taking six years to reach its pre-recession levels.

Specifically, the swaps rule will duplicate the burdens from another Dodd-Frank product: the Volcker Rule. Volcker was only adopted by four agencies (CFTC, SEC, OCC, and the Federal Reserve), unlike the five for swaps, but it took up 1,000 pages to tell banks to be less risky. In its aftermath, banks have been forced to shed their market-making operations, thereby diminishing liquidity and preventing consumers from being able to trade quickly and at a steady price. Combine that with the higher capital requirements and banks are deciding not to take up excess space on their balance sheets holding inventories of assets awaiting a buyer.

For example, JPMorgan carried $2.7 trillion in corporate bonds in 2007; this year that number is down to $1.7 trillion and falling. That lack of liquidity hurts consumers directly as their options for financial products and services are limited, and indirectly, as less liquid U.S. banks lose their competitiveness – Europe and much of the rest of the world are not restricted by bans on certain activities. It also should come as no surprise to regulators that the banning of certain bank activities, like proprietary trading, will only drive those activities into less regulated or even unregulated areas of the financial system, thereby undermining the entire purpose for these rules.

Further, it is well documented that as liquidity decreases, the cost of capital for businesses increases. There is a positive relationship between stocks’ excess monthly returns and bid-ask spreads for any given level of systematic risk. Thus, average returns are higher for stocks with higher bid-ask spreads, and the increase in the bid-ask spread as a result of rules like Volcker and the swaps rule, will result in higher capital costs for those businesses. The resulting increased capital costs mean slower economic growth and reduced job creation.

More generally, Dodd-Frank’s new rules, oversight agencies, and capital requirements – specifically, the requirements arising from the Too Big To Fail rule, at the current rate, are slated to reduce our GDP over the 2016-2025 period by $895 billion, or $4,346 per working-age person. That is money out of the pockets of consumers as a result of a law that was supposed to be helping consumers.

Conclusion

These two actions by the Federal Reserve have added $6.7 billion in costs, hardly a trivial figure for the nation’s economy, and more specifically, the financial system. This has contributed to Dodd-Frank’s $29.3 billion financial imposition, including more than 72.8 million paperwork burden hours. To put the paperwork in perspective, it would take 36,437 employees working full-time to complete a year of the law’s figurative “red tape.” The margin rule is evidence that, even after five years, the slow crawl of Dodd-Frank regulation will continue to pile up significant economic burdens for the financial system and consumers.

Research published by the American Action Forum and the Empire Center

FOREWORD

“Unfortunately, the real minimum wage is always zero, regardless of the laws, and that is the wage that many workers receive in the wake of the creation or escalation of a government-mandated minimum wage, because they lose their jobs or fail to find jobs when they enter the labor force.”

— Thomas Sowell

Under a proposal championed by Governor Andrew Cuomo, New York would become the first state in the nation to mandate a minimum wage of $15 an hour. That’s more than double the federal minimum wage, and 67 percent higher than the $9 statewide minimum already scheduled to take effect in New York at the end of 2015.

Advocates of such a policy believe that low-income workers will be its primary beneficiaries. This paper, however, suggests that the poorest New Yorkers would have the most to lose from a sharp rise in the government-mandated wage floor.

The authors, economists Douglas Holtz-Eakin and Ben Gitis of the American Action Forum, draw on three credible research models to estimate low, medium and high impacts from raising the statewide minimum wage to $12 or $15.

The key finding: a $15 minimum wage ultimately would cost the state at least 200,000 jobs, with proportionately larger employment decreases in upstate regions. That’s the authors’ “low-impact” scenario, based on a model developed by the Congressional Budget Office, of which Holtz-Eakin is a former director.

The other two models point to even bigger losses, indicating that a $15 an hour minimum wage would lead to 432,200 and 588,000 fewer jobs under the “medium impact” and “high impact” scenarios, respectively.

Job losses would be smaller, but still more than New Yorkers should be willing to tolerate, if the state was to set the minimum at $12 an hour, according to Holtz-Eakin and Gitis. 

Based on national labor force data, the authors of this paper estimate less than 7 percent of the wages generated by a $15 wage, and less than 6 percent of the wages generated by a $12 wage, would actually go to households in poverty. 

As Holtz-Eakin and Gitis note, these findings are consistent with the preponderance of economic research, which has long indicated that higher minimum wages are associated with a decline in employment. To be sure, economists differ on the strength of the effect, which is why this paper draws from different approaches—with the CBO model at the lower bound—to illustrate a range of possible impacts.

The “Fight for $15” is rooted in well-founded and understandable concern about the challenges faced by low-income households, especially those struggling to get by in hyper-expensive New York City. But a $15 minimum wage is only likely to make those challenges worse.

As shown in this paper and as strongly suggested by other research on the general issue, enacting the biggest increase ever in New York’s minimum wage would benefit some low-income workers at the expense of others. The losers would be stuck with the ultimate minimum wage: zero.

In a prolonged period of slow economic growth, the potential loss of at least 200,000 jobs would be an extreme and unacceptable tradeoff.

—E.J. McMahon

President of the Empire Center for Public Policy

EXECUTIVE SUMMARY 

We applied the methodology of three different independent research models to examine the employment and earnings effects of raising New York’s statewide minimum wage to $12 per hour and to $15 per hour, respectively, by 2018 in New York City and 2021 in the rest of the state.

We focus on how raising the state minimum wage would affect the very workers whom the policy is intended to help. Overall, we find significant tradeoffs to raising New York’s proposed minimum wage.

While a minimum wage hike would benefit some workers by increasing their earnings, it would also hurt hundreds of thousands of others whose earnings would sink because they could no longer find or keep a job.

Our medium-impact estimates (below) show that raising the state’s minimum wage to $12 per hour would affect 2.3 million low-wage workers while costing the state 216,200 jobs. On net, total wage earnings among low-wage workers would rise by $1.1 billion.

Similarly, we find that increasing the state’s minimum wage to $15 per hour would affect 3.1 million workers and cost 432,500 jobs. Total wage earnings among low-wage workers would rise by $4.6 billion, after accounting for earnings declines from job losses.

Taken at face value, the net total wage earnings gains for both the $12 and $15 per hour minimum wage may seem large. But, it is important to point out that only a very small amount of the total wage earnings gains would benefit workers in poverty.

In previous American Action Forum (AAF) research, we found that only 5.8 percent of net total wage earnings gained from a $12 federal minimum wage and only 6.7 percent of the net total wage earnings gained from a $15 federal minimum wage would actually go to households in poverty. In New York, this means that of the $1.1 billion gained from the $12 minimum wage, only $61.9 million would go to workers in poverty. Likewise, only $308.0 million of the $4.6 billion gained from the $15 minimum wage would help workers in poverty.

Because the exact effect of the minimum wage on employment remains unsettled, we check the robustness of our results by employing a range of estimates from literature that imply low, medium and high employment impacts.

INTRODUCTION

The federal minimum wage has been set at $7.25 an hour since July 2009. In recent years, some American policymakers and labor advocates have argued for further increases in the wage at the federal, state, and local levels. On the federal level, the Obama administration and top congressional Democrats have rallied behind a proposal to raise the federal minimum to $12 per hour by 2020. Under another proposal championed by, among others, Sen. Bernie Sanders, the federal minimum would rise to $15 per hour, a level now in the process of being implemented in Seattle and a handful of other localities.

AAF, along with the Manhattan Institute, previously analyzed the employment and earnings effects of these two policies on a national level, and found that they would induce substantial job losses with little benefit for those in poverty.

New York’s minimum wage currently stands at $8.75 and is scheduled to reach $9 an hour on December 31, 2015.  However, based on recommendations of a Wage Board empaneled to study the issue at the request of Governor Andrew Cuomo, the state labor commissioner has issued an order raising the minimum wage to $15 an hour for employees of fast-food chain restaurants in New York.

Soon after the Wage Board recommendation for fast-food workers, the governor announced he would propose legislation raising New York’s statewide minimum wage to $15 per hour for all workers. The governor’s proposal would parallel the fast-food schedule, raising the state’s minimum wage to $15 per hour by 2018 in New York City and by 2021 for the rest of the state.

To test the impact of such policies, this paper estimates the employment and earnings effects of increasing New York’s minimum wage to two alternative higher levels— $12 and to $15 per hour—in each region of the state, focusing on the low-wage workers whom such raises would be intended to assist. In doing so, we project a range of job losses if lawmakers were to raise the state’s minimum wage to $12 or to $15 per hour and the net change in total wage earnings for all low-wage workers in the state.

PREVIOUS RESEARCH

We utilize research by the CBO (2014)[1]  Meer & West (2015),[2] and Clemens & Wither (2014)[3] to provide a range of estimates for the impact of a $12 and a $15 minimum wage on New York state employment and total wage earnings. These studies examined different labor-market aspects of the minimum wage, resulting in different conclusions regarding the policy’s impact on employment and wage earnings. Using these three studies, we consider the effects of the minimum wage under three scenarios — low, medium and high employment impacts.

In our previous paper, we used these same studies to examine the impact of raising the federal minimum wage on national employment and total wage earnings. Our medium-impact scenario was based on the model developed by Meer & West (2015). With that model, we found that raising the federal minimum wage to $12 per hour by 2020 would cost 3.8 million jobs and, on net, increase total wage earnings by $14.2 billion. Increasing the federal minimum wage to $15 per hour by 2020 would cost 6.6 million jobs and, on net, increase total wage earnings by $105.4 billion.[4]

CBO

In 2014, the Congressional Budget Office (CBO) examined the impact of raising the federal minimum wage to $9.00 or $10.10 per hour, two of the most popular proposals at that time. For the $10.10 proposal, the CBO found that the policy would result in employment falling by 500,000 jobs relative to their projected 2016 baseline. The CBO assumed that, in addition to those earning between $7.25 and $10.10 getting a raise, those earning just above $10.10 would also see their wages increase. Specifically, those who earn up to 50 percent more than the minimum-wage hike would see their hourly earnings rise. As a result, people earning below $11.50 (who stay employed) would benefit from a wage increase of some sort.

The CBO concluded that net total wage earnings for low-wage workers would increase by $31 billion: 19 percent of those additional earnings would go to families below the poverty threshold; 52 percent to families with incomes one to three times the poverty threshold; and 29 percent to families with incomes more than three times the poverty threshold. We employ these findings when developing our projection of a low-impact employment scenario resulting from an increase in New York’s statewide minimum wage.

Meer & West

While there is an ongoing debate regarding the impact of the minimum wage on the level of employment, Texas A&M economists Jonathan Meer and Jeremy West suggested in their recent research that the negative impact of the minimum wage is best isolated by focusing on employment dynamics—that is, the change in job growth once the higher wage is implemented. Specifically, they found that a 10 percent increase in the real minimum wage is associated with a 0.30 to 0.53 percentage-point decrease in the net job-growth rate. They found that, as a result, the 10 percent minimum wage increase reduces future employment by 0.7 percent.

Previously, the AAF applied Meer & West’s work to California’s recent law that raises the state’s minimum wage to $10 per hour (effective 2016). Using Meer & West’s result, the AAF found that this wage increase in California would mean a loss of 191,000 jobs that would never be created.[5] In addition, the AAF found that if every state followed suit, more than 2.3 million new jobs would be lost across the United States. We employ the estimates found in Meer & West’s study to characterize medium-impact employment consequences of raising the minimum wage in New York.

Clemens & Wither

In late 2014, Jeffrey Clemens and Michael Wither of the University of California at San Diego released research examining what happened to low-wage workers the last time that the federal government raised its minimum wage—rising in three steps, during 2007–09, from $5.15 to $7.25 per hour. Using data from the Survey of Income and Program Participation (SIPP), they focused on how the minimum-wage hike affected employment and earnings among those whom the minimum-wage hike affected most: low-wage workers earning below $7.50 per hour.

Clemens and Wither found significant, negative consequences for low-wage workers. From 2006 to 2012, employment in this group fell by 8 percent, translating to about 1.7 million jobs.[6] The job loss in this low-wage group accounted for 14 percent of the national decline in employment during this period.[7] The minimum-wage hike also increased the probability of working without pay (e.g., unpaid internships) by 2 percentage points. Workers with at least some college education were 20 percent more likely to work without pay than before the minimum wage rose.

As a result of the reduction in employment and paid work, net average monthly earnings for low-wage workers fell by $100 during the first year after the minimum wage increased and fell by an additional $50 in the following two years. We use the Clemens & Wither estimates to estimate our high-impact employment scenario.

METHODOLOGY

In identifying the number of workers whom the minimum wage hike would affect in New York, we assume that those most directly affected by the minimum wage increase are the workers who, we project, would earn between $9 per hour (New York’s statewide minimum wage effective December 31, 2015) and the new minimum wage level in 2021 (2018 for the New York City region) under current law.[8] For the $12 minimum wage, this includes all workers who would earn between $9 and $12 per hour; for the $15 minimum wage, it includes everyone who would earn between $9 and $15 per hour. This is the group of workers that we assume would both suffer from all the job losses and benefit from any wage earnings gain.

To estimate the actual number of workers who would be impacted by the minimum wage and thus either lose their jobs or see their wages rise, we first project total regional employment levels by 2021 (2018 for New York City). We accomplish this by using the state Labor Department’s Employment Projections[9] to calculate the projected compounded annual growth rate for total employment in each region. Starting with 2014 total employment reported in each region by the New York Quarterly Census of Employment and Wages,[10] we project employment levels to 2021 (2018 for New York City).

After projecting total future employment levels in each region, we estimate how many of those workers would earn between $9 and $12 per hour and between $9 and $15 per hour under current law. With regional American Community Survey (ACS) data, we obtain the percent of workers who would earn at least $9 per hour but less than $12 and $15 per hour by 2018 in New York City and 2021 in all other regions.[11]^{, [12]} We multiply that percentage by the projected total employment levels in each region to estimate the number of workers who would earn between $9 per hour and the new minimum wage levels in 2021 (2018 for New York City) under current law.

WORKERS IMPACTED[13]

Table 1 on page 5 shows the projected number of workers statewide and in each region who would be impacted if lawmakers raised New York’s minimum wage to $12 or to $15 per hour.

Table 1: Workers Impacted by Minimum Wage Hike (in thousands)

 We estimate that raising New York’s minimum wage to $12 per hour would impact about 2.3 million workers statewide, about 24.4 percent of the employed population. Raising it to $15 per hour would impact about 3.1 million workers or 33.2 percent of the workforce.

The percent of the population affected in each region depends entirely on the frequency of low-wage employment. Long Island, for instance, appears to have the fewest low-wage workers as a $12 minimum wage would impact 18.1 percent of workers and a $15 minimum wage would impact 25.1 percent. In the North Country, however, those minimum wage hikes would affect 30.5 percent and 41.6 percent of workers, respectively.

EMPLOYMENT

Supporters of a minimum wage generally echo Governor Cuomo’s view that such a policy is needed to lift workers out of poverty, “improve the standard of living for workers, encourage fair and more efficient business practices, and ensure that the most vulnerable members of the workforce can contribute to the economy.”[14]  

However, the potential employment consequences of mandating a minimum wage hike call the merits of this policy into question. When a state government increases the minimum hourly pay for private-sector workers, it effectively increases the per-hour cost of low-wage labor. Employers have three main ways to pay for the additional labor cost: lower profits, higher prices, and fewer workers.

While many labor advocates assume that employers will absorb a state-mandated increase in labor costs through a reduction in profits, the evidence suggests that the vast majority of low-wage workers are in industries with thin profit margins, such as retailers and restaurants.[15] In these industries, businesses tend to pay for minimum wage hikes by increasing prices, reducing current and future employment, or both. While the exact impact of a minimum wage hike on employment is debated, extensive economic literature, from the 1950s to today,[16] concludes that raising the minimum wage has a negative impact on employment levels and job creation.[17] Moreover, the literature shows that the workers who tend to become jobless are the low-skilled, low-wage workers whom the policy intends to help.

The CBO, Meer & West, and Clemens & Wither research on the issue demonstrate negative labor market consequences with varying degrees of severity. In this section, we apply their findings to proposals to raise New York’s statewide minimum wage. In particular, for each region we will estimate the labor market consequences of raising the state’s minimum wage to $12 and to $15 by 2021 (2018 for New York City). We follow our previous study’s methodology by assuming that all job losses occur within the group of workers who would earn between $9 and the new minimum wage level under current law. As a result, we assume no job losses among workers earning more than the new minimum-wage level.

To preview our estimates, we find that increasing New York’s minimum wage to $12 per hour would reduce employment by 76,600 to 290,400 jobs. Raising the minimum wage to $15 per hour would cost the state 200,000 to 588,800 jobs.

$12 Minimum Wage

Overall, we estimate that if the minimum wage rose to $12 per hour, low-wage employment in New York would be 76,600 to 290,400 lower than under current law.

Table 2: Job Losses from $12 per hour Minimum Wage

Study Models - Low, Medium and High Impacts

Using the CBO report, our low-impact employment scenario, we find that raising the minimum wage to $12 per hour by 2018 for New York City and 2021 for the rest of the state would cost 76,600 jobs statewide. This means that total employment would decline by 0.8 percent. The job losses would range from about 1,600 in North Country to 37,400 in New York City. Although the rural areas of New York would lose fewer workers than the urban areas, the job losses would be proportionally more significant. 

In the Hudson Valley and on Long Island, for instance, employment would only decline by 0.7 percent and 0.6 percent, respectively. But in the North Country and Southern Tier, employment would decline by 1 percent.

In our medium-impact employment scenario, derived from Meer & West, we find that this minimum wage increase would reduce employment significantly, costing 216,200 low-wage jobs. Meer & West’s research models the impact of the minimum wage on total employment levels, not just low-wage workers. So while the other models are sensitive to the prevalence of low-wage work in each region, our medium-impact employment model is not. As a result, the estimated percent reduction in employment is the same across the state.

Finally, with the Clemens & Wither model, our high-impact employment scenario, we estimate that raising the minimum wage to $12 per hour would cost 290,400 jobs. This model does depend on the prevalence of low-wage employment, which allows us to examine how raising the minimum wage in New York would affect each region differently. We find that the decline in employment in this scenario ranges from 2.6 percent on Long Island to 3.4 percent in the North Country.

$15 Minimum Wage

We estimate that 200,000 to 588,800 fewer low-wage jobs would exist in 2021 (2018 in New York City) if New York policymakers increased the state minimum wage to $15 per hour.

Table 3: Job Losses from $15 per hour Minimum Wage

Study Models - Low, Medium and High Impacts

Using the CBO estimate, we find that increasing the state’s minimum wage to $15 per hour would cost 200,000 jobs. Among the state’s regions, the reduction in employment would range from 1.6 percent on Long Island to 2.7 percent in the North Country. The reduction in job creation captured by the Meer & West estimate reveals that New York would have 432,500 fewer jobs than under current law, reducing projected employment by 4.7 percent.

Using the Clemens & Wither estimate points to a loss of 588,800 jobs, a 6.4 percent decline in employment. Under this scenario, we find that the job losses would range from 5.2 percent on Long Island to 7.3 percent in the North Country.

TOTAL WAGE EARNINGS

In this section, we project how increasing New York’s statewide minimum wage to $12 and to $15 per hour by 2021 (2018 in New York City) would affect total wage earnings of low-wage workers. This involves calculating the total wage earnings increase for those employed and the total wage earnings loss for those who can no longer find jobs. After subtracting total wage earnings lost from total wage earnings gained, we derive the net change in total wage earnings for all low-wage workers.

Assumptions and Methods

For all workers who would keep their jobs and would earn between $9 and the new minimum wage level under current law, we assume that increasing the minimum wage would raise their hourly pay rate to the new minimum wage level.

In the $12 minimum wage scenario, for all workers who earn between $9 and $12 per hour in 2021 (2018 in New York City), we assume that wages would rise to $12 per hour—if they stay employed.

Likewise, in the $15 minimum wage scenario, we assume that all works earning between $9 and $15 per hour would see their wages rise to $15 per hour—if they stay employed.

Under both the $12 and $15 minimum wage scenarios, we assume that the minimum wage increase itself would have no impact on hours worked per week and weeks worked per year—for those who keep their jobs. In addition, we assume that all who would be jobless as a result of the minimum wage increase would see their annual wage earnings fall to $0.

To estimate the change in total wage earnings, we utilize ACS data to project average hourly pay, annual pay, and hours per year in each region for workers who would earn between $9 and $12 per hour and between $9 and $15 per hour under current law.[18] With that information, we estimate the total wage earnings lost among the low-wage workers who lose their jobs and the total wage earnings gained among those who remain employed. Overall, we calculate the net change in total wage earnings for low-wage workers by subtracting total wage earnings lost from total wage earnings gained.

$12 Minimum Wage

As shown in Table 4, the impact of raising New York’s minimum wage to $12 per hour on the total wage earnings of low-wage workers depends largely on how many become jobless.

Table 4: Net Wage Earnings Change from $12 per hour Minimum Wage (dollars in millions)

Study Models - Low, Medium and High Impacts

In the low-impact CBO and medium-impact Meer & West employment scenarios, raising the New York’s minimum wage to $12 per hour would, on net, increase total wage earnings among low-wage workers. In the CBO model, a $12 minimum wage would, on net, raise low-wage worker total wage earnings by $3.8 billion or 0.7 percent. In the Meer & West model, it would raise low-wage worker total wage earnings by $1.1 billion or 0.2 percent. However, in the high-impact Clemens & Wither employment model, the earnings lost stemming from the reduction in jobs actually outweighs earnings gained among those who remained employed. In particular, low-wage worker total wage earnings would, on net, decline by $370.2 million or 0.1 percent.

In this case, the regions with the most low-wage workers could potentially see the largest total wage earnings gains. In the Meer & West scenario, for instance, net total wage earnings would rise by 0.5 percent in the North Country and decline 0.1 percent on Long Island.

$15 Minimum Wage

Table 5 illustrates how raising the New York minimum wage to $15 per hour would affect total wage earnings for low-wage workers.

Table 5: Net Wage Earnings Change from $15 per hour Minimum Wage (dollars in millions)

Study Models - Low, Medium and High Impacts

In total, we estimate that raising New York’s minimum wage to $15 per hour would, on net, increase total wage earnings for low-wage workers by between $1.3 billion and $10.6 billion. The net total wage earnings gains in the $15 minimum wage scenario would be much larger than in the $12 minimum wage scenario because the $15 minimum wage would both impact more workers and raise their average wages by a larger margin. 

In the medium-impact Meer & West employment scenario, total wage earnings would, on net, rise by $4.6 billion. The CBO model again results in the largest net total wage earnings gain ($10.6 billion) because it projects the fewest job losses. In this case the Clemens & Wither model projects the smallest net total wage earnings gain ($1.3 billion) because it projects the most job losses.

Once again, the regions with the most low-wage workers tend to have the largest total wage earnings gains. In the medium-impact Meer & West employment scenario, total wage earnings in the high-wage Long Island region would, on net, only increase by 0.1 percent. But in the low-wage North Country region, total wage earnings would, on net, increase by 1.6 percent.

POVERTY AND PRICE IMPACTS

It is important to understand that very little wage earnings gained from raising the minimum wage actually benefits workers who live in households below the poverty line. Due to poverty data limitations, we are unable to estimate what portion of the projected total wage earnings gained in New York from each minimum wage hike would go to workers in poverty.

In our previous, nationally focused report, however, we found that under the Meer & West model only 5.8 percent of net total wage earnings gained from raising the federal minimum wage to $12 per hour would actually go to workers in poverty. Applying this estimate to New York would mean that only $61.9 million of the $1.1 billion net total wage earnings gained would go to workers in poverty. Likewise, we found that only 6.7 percent of the increase in net total wage earnings from a $15 per hour federal minimum wage would benefit workers in poverty. In New York, that means that of the $4.6 billion gained from a $15 minimum wage, only $308.0 million would benefit workers who are in poverty.

 It should also be noted that these projections of employment impacts and net changes in total wage earnings do not take into account the economic impact of any price increases resulting from higher minimum wages. There is a substantial body of research concluding that raising the minimum wage would also raise prices, and that the higher prices would disproportionately burden low-income households and families. MaCurdy (2014) analyzed how the 90 cent federal minimum wage increase from $4.25 in 1996 to $5.15 in 1997 impacted prices. The restaurant industry, which has the highest concentration of low-wage workers, increased average prices by 1.85 percent. Families spent $136 (in 2010 dollars) more per year on average due to this modest minimum wage hike. Moreover, the price increases facing families in extreme poverty imposed a slightly larger percentage increase in their annual expenditures than it did for families who were not in poverty. In particular, the minimum wage hike increased annual expenditures for families with incomes less than half the poverty threshold by 0.63 percent. For families with incomes more than three times the poverty threshold, it increased their expenditures by 0.56 percent.[19]

 A $12 and $15 per hour minimum wage in New York may indeed increase total wage earnings after accounting for job losses. But, only a very small portion of the total wage earnings gained from the policy would actually benefit those it intends to help: households in poverty. Meanwhile, evidence suggests it would disproportionately burden those in poverty with higher prices.

CONCLUSION

Across the nation, lawmakers continue to debate the merits of raising the minimum wage to $12 per hour and to $15 per hour. In this paper, we estimate the employment and wage earnings effects of raising the state minimum wage in New York to $12 and to $15 per hour. We find that any potential benefits from raising the minimum wage would be greatly offset by the negative labor market consequences of the policy.

In particular, the wage earnings gains from raising the minimum wage would come at a significant cost to the large number of workers who would become jobless. In effect, raising the minimum wage transfers wage earnings from the low-wage workers who are unfortunate enough to become jobless to the low-wage workers who remain employed.

• Download PDF

Introduction

The Environmental Protection Agency (EPA) recently released a new round of air quality standards for ground-level ozone. The measure lowers the threshold for “nonattainment” status from 75 parts per billion (ppb) to 70 ppb and imposes $1.4 billion in nationwide annual costs. Generally, “nonattainment” counties exceed EPA’s threshold for ozone (“smog”) pollution. Counties deemed nonattainment must work with their states to devise a State Implementation Plan (SIP) to meet EPA goals. According to EPA, this must, “show how the nonattainment area will attain the primary [ozone] standard ‘as expeditiously as practicable,’ but no later than within the relevant time frame.”

But what are the deeper economic effects? Looking back at the previous standard of 75 ppb, the American Action Forum (AAF) found that even this less stringent threshold reduced total wage earnings, average annual worker pay, and employment. Observed nonattainment counties experienced losses of $56.5 billion in total wage earnings, $690 in pay per worker, and 242,000 jobs between 2008 and 2013.

How EPA Regulates Ozone

Under the Clean Air Act, EPA must set air quality standards for six types of criteria pollutants, including ozone. Ozone regulation can be difficult for states and localities because unlike other pollutants, there is no factory or machine that directly emits ozone; it generally forms as a product of a number of chemical reactions between other pollutants and sunlight. When a particular county has ozone concentrations above EPA’s threshold, 75 ppb as of 2008, it is considered a nonattainment area. Affected states and localities must develop a plan to both reduce the amount of ozone in their nonattainment counties as well as manage activities that could contribute to higher ozone levels. There is no one plan or pollution strategy that will allow a state or locality to comply with ozone regulation. In fact, EPA acknowledges there are some “unknown” technologies or strategies that will be required to comply with the most recent standards. Typically, compliance involves a combination of proposals from local governments, review by EPA, and constant monitoring by all jurisdictions to ensure a compliance pathway. Ozone regulation affects virtually every activity that burns fossil fuels.

Since there are so many factors involved in the formation of ozone, a nonattainment designation means that any economic activity involving emissions – including such industries as manufacturing and construction – could face certain operational restrictions. Since nonattainment is generally consistent throughout a county, it is possible that some of these industries could choose to slow hiring and limit raises or move their operations to other nearby counties that do not face such restrictions in order to be more productive and profitable. This study assesses these consequences by examining wage and employment patterns in nonattainment counties, relative to their neighboring attainment counties.

Methodology

In this paper, we study the 2008 Ozone rule, and specifically, its impact on county level labor markets — total wage earnings, average annual worker pay, and total employment. To accomplish this, we compare the 208[1] counties that the EPA deemed in nonattainment (and therefore subject to larger restrictions) to the 234 attainment counties that border nonattainment areas before and after the regulation took effect. As a result, we estimate the impact of the regulation on nonattainment county labor markets relative to labor markets in neighboring attainment counties, which are presumably legally, economically, and socially similar, but were not directly impacted by the rule. In total, we examine 442 counties from 28 states.

Empirical Model

In order to find the relationship between the ozone regulation and total wage earnings, average annual worker pay, and total employment growth in nonattainment counties, AAF performs regression analysis to test the change in the annual growth rates relative to neighboring attainment counties after the regulation was published and went into effect in 2008. For the pre-regulation period, we calculate the compound annual total wage earnings, average annual worker pay, and total employment growth rates from 2004 to 2007. The same is done for the regulation period, which is 2008 to 2013. For each variable, we pool the data from both time periods and use a binary variable to indicate if the regulation is in effect.

Our empirical model is captured in equation 1:

The dependent variable, Labor Market Growth Rate_i, represents the county level average annual growth rate of total wage earnings, average annual worker pay, or total employment; these data are based on county level wage and employment data found in the Bureau of Labor Statistics’ Occupational Employment Statistics.^[2] We use three binary variables to indicate whether a county is ever affected by the regulation, whether the regulation is in effect, and the interaction between them. The first binary variable, Nonattainment_i, equals 0 in both periods if the county is attainment or 1 if it is nonattainment. This variable by itself measures labor market growth in nonattainment counties relative to bordering attainment counties before the regulation took effect. The second binary, Regulation_i, equals 1 for every county in the period after the regulation took effect, 2008 to 2013, or 0 in the period before the regulation took effect, 2004 to 2007. This variable by itself captures the change in labor market growth in the attainment counties after the regulation took effect. Finally, the interaction term only equals 1 if both conditions, being a nonattainment county and being in the post-regulation period, are satisfied. If it is an attainment county, the pre-regulation period, or both, then the variable equals 0 and it drops from the model.

In this paper, we are interested in the combination of Nonattainment_i and the interaction term. The sum of those two variables’ estimated coefficients captures the total change in labor market growth rates in nonattainment counties when the regulation took effect relative to the change in bordering attainment counties, which were not impacted by the regulation.

In testing the impact of the regulation on total wage earnings, average annual worker pay, and total employment growth rates, there are plenty of other factors that influence the labor market and need to be held constant. We include state Freddie Mac House Price Index^[3] as a variable to control for the negative macroeconomic labor market trends during the Great Recession. Additionally, we control for the percent of county employment in the services industry,^[4] population of each county,^[5] and percent of the population with at least a Bachelor’s degree^[6] to reflect the quality and size of the labor force. Lastly, we include state-level tax rates^[7] to control for the effect of state fiscal policy on the labor market. Similar to the employment measurements, we take the compound annual growth rates for each of these variables for before and after the regulation took effect. Finally, we cluster our standard errors to control for any analysis errors that may be correlated with the counties over time as well as potential heteroscedasticity present in our data.

Results

We find that nonattainment status had significant negative effects on total wage earnings, average annual worker pay, and total employment for a county. Table 1 illustrates the results for the impact of the ozone regulation on the labor market in nonattainment counties.

The results consistently show that the ozone regulation had statistically significant adverse effects on nonattainment county labor markets. In particular, we find that in nonattainment counties, the ozone regulation reduced the annual average growth rates of total wage earnings, average annual worker pay, and total employment by 0.4 percentage point, 0.3 percentage point, and 0.1 percentage point respectively. To put this in perspective, the average annual growth rate of total wage earnings in nonattainment counties was 1.7 percent when the regulation took effect. Our results indicate that the average annual growth rate would have been 2.1 percent had the regulation not taken effect. Moreover, the annual growth rate of average annual worker pay in nonattainment counties would have increased on average from 1.6 percent to 1.9 percent. The average annual total employment growth rate would have increased from 0.1 percent to 0.2 percent.

Implications

We find consistent and statistically significant evidence that the ozone regulation reduced the growth rate of total wage earnings, average annual worker pay, and total employment in nonattainment counties. Table 2 illustrates the implications of our results on pay and employment.

We find that in nonattainment counties, the ozone regulation was associated with a 0.4 percentage point decline in the annual growth rate of total wage earnings. This translates to a loss of $54 million per year in total wage earnings on average in each county. By 2013, nonattainment counties on average lost $272 million in total wage earnings due to the ozone regulation. This means that the ozone regulation reduced total wage earnings by $56.5 billion across all the observed nonattainment counties.

In addition, we find that between 2008 and 2013 the ozone regulation reduced the annual growth rate of average annual worker pay by 0.3 percentage point. This means the regulation cost each worker on average $140 per year. So by 2013, each worker in nonattainment counties lost a total of $690.

Finally, we find evidence that the ozone regulation reduced job growth in nonattainment counties by 0.1 percentage point. This indicates that the regulation cost each nonattainment county an average of 233 workers each year. By 2013, each nonattainment county lost 1,200 jobs on average. Across all the observed ozone regulated counties, this adds up to a loss of 242,000 jobs total. To put the employment figure in perspective, the U.S. economy created 223,000 jobs in July 2015

Discussion

These findings should not come as a huge surprise given the existing research. The results of AAF’s study resemble those of past studies of ozone regulation’s effects on wages and employment. For instance, in 2002 Professor Michael Greenstone found that earlier iterations of ozone regulation under the Clean Air Act resulted in losses of “590,000 jobs, $37 billion in capital stock, and $75 billion (1987$) of output” in nonattainment areas.

Our study’s findings are not as dramatic, but that is likely due to differences in scope and magnitude of regulation. Professor Greenstone’s study primarily examined “pollution intensive industries” whereas this study looks at all industries and all employees. Many industries are only tangentially related to the standards, whereas higher polluting industries will likely experience the most acute effects. Finding that this rule has negative effects on an economy-wide scale in spite of dilution from industries that do not directly pollute further implicates the effects of ozone regulation. In addition, Professor Greenstone examined earlier standards that may have been more difficult to comply with at that time due to unfamiliarity with mitigation approaches and new technology.

The other curious aspect of our findings is why there is such an outsized imbalance between lost jobs and lost wages. The basic phenomenon this study sought to examine was the change in employment and pay in nonattainment counties relative to neighboring attainment counties. We find that the regulation reduced total wage earnings in each county by $54 million each year and cost 233 jobs annually. But clearly, the job lost in the average nonattainment county did not pay roughly $232,000 annually. Rather it seems that the total wage earnings fell mostly because the average annual pay per worker who remained employed fell substantially as well.

Conclusion

EPA’s ozone standards affect a broad array of industries, and due to the nature of how ozone actually forms, nonattainment areas can have difficulty meeting standards quickly. California, for example, will have nearly a generation to reach the new standards. This could result in certain businesses making economic decisions to reduce hiring, limit raises, and move or expand operations elsewhere. The result? $56.5 billion in lost wages and 240,000 fewer jobs in nonattainment counties struggling to meet standards.

Since the 1980s, the Lifeline program has provided a subsidy for telephone service for low income households. However, changes in the broader communication ecosystem has prompted the Federal Communications Commission (FCC) to implement reforms to the program. Among the changes that the Federal Communication Commission (FCC) is currently considering for the wired and wireless Lifeline subsidy, the most important should be the implementation of an evaluation program. According to our model, which incorporates the most recent data, the program costs $2,078 to add an additional wireless subscriber to the already existing telephone system for one year. In total, just under 6 percent of all the recipients of Lifeline added wireless service because of the program.    

There are other ways to induce the adoption of new technologies. In 2012, U.S. consumers paid on average over 17 percent in wireless taxes and fees, which ranks these taxes among the highest for goods.[1] More than just good government and tax policy, lowering the wireless tax rate could go a long way to meet the goals of a majority of Americans having access to telephone service.

Time and again, the FCC has resisted implementing a program evaluation of Lifeline, even after reports from Government Accountability Office (GAO) suggested these changes, because, as the agency proclaims, the structure of the program “makes it difficult to determine a causal connection between the program and the penetration.”[2] Well structured, independent studies have been conducted of the program for nearly as long as its existence, and almost universally it is found to be economically inefficient and ineffectual in achieving its stated goals.[3]

By updating a much cited model with variables for wireless, we were able to test Lifeline’s usefulness for expanding wireless use.[4] While there is a statistically significant correlation between wireless penetration rates for a state and their Lifeline program, the overall effect is small. An increase in the size of the program by 10 percent would only increase wireless penetration from .08 percent to .09 percent in our three models.[5] Assuming the higher number, the cost of adding a marginal user was $2,078. In 2012, 1.05 million marginal households got wireless services because of the program, even though 18.1 million households received the subsidy for both wired and wireless services. In total, just under 6 percent of the Lifeline households used their subsidy to get onto wireless services.  The appendix contains the methodology for these findings.

 Our outcomes are in line with other studies, and suggest that the program is rather ineffective in getting people into wireless services.

Appendix

The chart below displays the results of three models. The first model includes just a poverty variable, the second includes just an income variable and the third model includes both. Data from 2012 was used because it is the most recent date for the dependent variable, total wireless penetration by state, which comes from the Centers for Disease Control.[6] Because Montana, South Dakota, and Wyoming have such small populations, the CDC could not reliably estimate their telephone use, leading to 48 total observations. The tax variable is the total combined state and local taxes percentages.[7] The poverty variable is an estimate from the March Supplement of households living in poverty as a percentage of total households by state.[8] The income variable is a Census estimate of median household income from 2012.[9] The Lifeline variable represents the expenditures on Lifeline divided by the number of poor households.[10] The density variable is a percentage of population living in urban areas.[11] Because the previous model tries to track the effect of previous household consumption of telephone service, the Wireless 2010 variable is a two year lag variable, capturing those households that consumed some kind of wireless telephone service in 2010.[12]

Introduction

As the world meets in Paris at the United Nations (UN) Conference on Climate Change during the first two weeks in December, it is important to take note how the U.S. has already regulated greenhouse gases (GHGs). According to American Action Forum (AAF) research, regulators have already imposed $26 billion in annual costs to limit GHGs and have proposed an additional $1.7 billion. However, to meet President Obama’s climate goals the nation will have to spend up to $45 billion more each year by 2025.

What are the benefits of these investments? According to the Environmental Protection Agency (EPA) estimates, previous actions will avert a combined 0.0573 degrees Celsius of warming. Meeting the president’s 2025 goals could add reductions up to 0.125 degree Celsius. In other words, full achievement of the president’s climate goals will cost more than $73 billion in annual burdens to alleviate less than two-tenths of one degree of warming.

Past GHG Regulation

President Obama and EPA have aggressively regulated greenhouse gases from a variety of sources. AAF research counted at least 15 final rules designed to ensure proper reporting of GHGs, increasing fuel economy for cars and heavy-duty trucks, methane reductions from fracking, and limiting GHG emissions from new and existing power plants. These rules have consumed more than 3,000 pages in the Federal Register. The total cost from these 15 measures is $230 billion in net present value costs, more than $26 billion in annual burdens, and 2.9 million paperwork burden hours. These totals omit Department of Energy rules that also limit GHG emissions (and impose substantial burdens) and EPA’s Mercury Air Toxics and Cross-State Air Pollution measures. The recent regulatory burdens fighting climate change act essentially like taxes, raising the cost of energy for consumers and manufacturers.

Below is a list of the five most significant GHG regulations since President Obama took office:

Combined, all 15 major rules aim to eliminate more than six billion tons of carbon dioxide (CO₂) during the life of the rulemakings or roughly 860 million annually. For comparison, the U.S. emitted 6.1 billion tons in 2005. Yet, despite the heavy costs of regulation now and the questionable future benefits, regulators will continue to issue new rules to address climate change.

President’s Climate Goals

At the beginning of his administration, President Obama set a goal to reduce U.S. GHG emissions by 17 percent, below 2005 levels. Based on the latest Energy Information Administration (EIA) projections, the U.S. will emit 5.49 billion tons of GHG in 2020. (This baseline includes all regulatory and legislative actions as of October 2014.) A 17 percent reduction from 2005 levels means the U.S. needs to emit just slightly more than five billion tons in 2020, so the U.S. is still 303 million tons away from the goal.

That might not appear to be difficult, given the nation has already cut more than 900 million tons annually, but consider that time is hardly on the administration’s side. What’s more, relaxing the initial deadline for power plant compliance means the “Clean Power Plan” will only save 81 million tons of CO₂ by 2020. Any similar regulation affecting stationary sources of GHG will likely reap only marginal cuts in the remaining four to five years.

The 2025 goals are even more daunting. According to EIA, the nation will emit 5.5 billion tons, barely an increase above 2020, but President Obama has called for a reduction of up to 28 percent below 2005 levels. This puts the president’s goal at 3.9 billion tons or a reduction of more than 1.2 billion tons by 2025. This includes EIA’s baseline, in additional to the Clean Power Plan, and pending proposals for a second round of heavy-duty truck efficiency and a new fracking measure. With all of these reductions assumed, the nation must still double its GHG cuts to date and then find additional reductions of more than 400 million tons of CO₂. Even though the Clean Power Plan is scheduled to cut 265 million tons of GHGs by 2025, it is just a fraction of the remaining goal.

The graph below displays the heavy lifting required to meet the president’s targets:

Cost of Achieving Goals

What will it cost to cut an additional 1.2 billion tons by 2025? If history is any guide, up to $45 billion in annual regulatory costs. To arrive at this figure, AAF examined the cost of reducing past GHG emissions from the Obama Administration. For example, examining the full slate of the administration’s regulatory actions on climate yields $261 billion in total costs to reduce more than seven billion tons of GHG. This yields a cost of $37.04 to prevent one ton of CO₂ from entering the atmosphere. Comparing annual costs ($28 billion) to annual GHG reductions (902 million tons) results in a slightly lower figure: $31.11 to prevent the release of an additional ton of CO₂.

With these figures in mind, we know in 2020 the nation will need to eliminate 303 million tons above EIA’s baseline and what is currently in proposed form. By 2025, that figure grows to 1.2 billion tons. Assuming the cost of eliminating CO₂ remains constant for the next decade nets the following figures: $11 billion in additional annual costs by 2020 and $45.5 billion by 2025. Using the lower figure of $31.11 to reduce emissions could impose costs of $9.4 billion (greater than the burden of the Clean Power Plan) and $38.2 billion, respectively. The 2025 goals are the equivalent of two to three years’ worth of government-wide regulation.

With those costs pending, what’s left to regulate?

The administration has already targeted oil and natural gas wells twice, existing and new power plants, and the transportation sector. Yet, there are obviously more sources of GHG left for regulators.

EPA has already put out a proposed “Endangerment Finding” on aircraft emissions, although eminent regulation is unlikely until international regulators conclude their work. The fracking boom has generated plenty of sources of methane, beyond the well sites, that could be attractive for the Department of Interior or EPA.

In addition, the regulation of refineries or large manufacturing plants could be an attractive target in the future, although controlling GHG emissions from those sources will hardly net significant cuts. For example, the largest refinery (by emissions) in the U.S. emits 10.5 million tons of equivalent CO₂ and there are only 145 refineries that report to EPA. Compare that to the power plant sector, where the largest facility emits 22.2 million tons and there are 1,572 facilities that report to EPA. In total, power plants emitted 2.1 billion metric tons of equivalent CO₂ in 2013_, compared to just 177 million tons for refineries. Suppose the nation eliminated 100 percent of refinery emissions by 2025; that would reduce the amount needed to reach the president’s 2025 goal by just 14 percent. 

Finally, according to EPA data, electricity generation and transportation generated 58.5 percent of U.S. GHG emissions in 2013. Industrial emissions, which regulators will doubtless address during the next decade, consumed 21 percent. The next closest contributor of GHG emissions is the agricultural sector, at 8.8 percent. Hardly a popular villain, regulating farms, large or small, will be politically difficult for any administration, expensive, and likely to yield few returns. Regardless of which industry is next for regulators, the price of reducing an additional 1.2 billion tons will remain high.

Benefits

There will be benefits from continuing to reduce GHGs. Some industries will flourish and others will shrink. There will be indeterminate health gains and some climate benefits. EPA admits climate benefits will remain small because climate change is a global problem and there is only so much a nation with four percent of the world’s population can do to avert the problem.

EPA has produced estimates from its past efforts to alleviate climate change. In five previous rules, EPA projected temperature declines ranging from 0.0042 Celsius to 0.0176 Celsius by the year 2100. Combined, previous EPA action to limit climate change could avert just 0.0573 degrees by 2100 (including the second round of heavy-duty trucks in this baseline). Those small reductions come at the total cost of more than $208 billion.

There are also GHGs reductions associated with these five measures. Combined, EPA projects they will eliminate 546 million annual tons of CO₂. Using these metrics as a baseline, AAF assumed the same ratio of GHG reduction to temperature reduction for the president’s 2025 goal. Assuming the rate remains constant for the next decade, reducing emissions by an additional 1.2 billion tons could avert 0.125 degrees of warming. In other words, unless there is a drastic change in EPA’s modeling, meeting President Obama’s 2025 climate goal will reduce global temperature by 0.1823 degrees Celsius; a number that is difficult to measure, much less halt drastic climate change. This is also a high-end figure, as agencies typically give a range of temperature declines and AAF always assumed the most beneficial climate scenario.

What about monetizing the benefits? According to the White House, the “Social Cost of Carbon” (SCC) represents “an estimate of the economic damages associated with a small increase in carbon dioxide (CO₂) emissions, conventionally one metric ton, in a given year.” It also represents a per ton monetary equivalent of the value of avoided climate change. For example, if a regulation reduced emissions by 265 million tons in 2025, it could yield $13.5 billion in climate benefits, assuming a discount rate of 3 percent.

The latter is important when comparing costs and benefits because costs are likely incurred initially and benefits are typically years, decades, or in the case of climate change, generations away. White House guidance dictates a discount rate of three and seven percent. “As a default position … a real discount rate of 7 percent should be used as a base-case for regulatory analysis.” However, for climate change, the administration has ignored its own guidance. It SCC analysis uses 5 percent, three percent, 2.5 percent, and an extreme that represents the most catastrophic effects of climate change. EPA completely omits a seven percent rate and instead concocts a 2.5 percent rate.

Assuming a discount rate of five percent, (the average of three and seven) as EPA does in one analysis, yields a $16 benefit per ton of CO₂ eliminated in 2025. Given AAF’s analysis that the cost of eliminating a ton of CO₂ ranges between $31 and $37, it appears the costs far exceed the benefits, at least at the five percent rate.

Conclusion

After $28 billion in annual regulatory costs to address climate change, the U.S. is hardly finished if it wants to meet the president’s goals. As much as $45 billion per year in regulatory burdens will likely be required to meet the 2025 benchmark, with just a 0.125-degree temperature reduction as a reward. As the U.S. approaches another round of climate negotiations, these figures demonstrate the American people have already shouldered a heavy burden to reduce emissions. 

Introduction 

On October 6, 2015, the Centers for Medicare and Medicaid (CMS) published its final rule on Stage 3 of the Meaningful Use Requirements for the Electronic Health Record Incentive Program.[1] This program adjusts payments to Medicare and Medicaid providers for implementing and “meaningfully using” (or not) interoperable electronic health records (EHR) systems. These standards are being implemented in three stages to gradually move providers toward the desired end point: to “provide efficiencies in administrative processes which support clinical effectiveness, leveraging automated patient safety checks, supporting clinical decision making, enabling wider access to health information for patients, and allowing for dynamic communication between providers.”

At each stage, CMS has chosen various metrics—of both quality and health information technology (HIT) functionality—to evaluate a provider’s performance in caring for their patients. In order to attest to each stage of the incentive program, providers must track and report their performance on such metrics. The first meaningful use requirements (Stage 1) were outlined in 2010 and focused on capturing patient data, such as demographic information and family medical history. Stage 2 began in 2014 and focused on the exchange of information between providers and patients as well as providers within a given practice in order to improve treatment adherence and care coordination. Beginning in 2016, penalties will be issued to providers that fail to meet the program’s requirements in the preceding year.[2] Stage 3 objectives are focused on improving the interoperability of EHR systems in different practices.

Modified Stage 2 and the 2015-2017 Transition Period

The new rule combines stages 1 and 2 into a “Modified Stage 2,” which will allow providers to meet a single set of objectives for up to three years (2015-2017), rather than having to meet the objectives of both Stage 1 and 2 separately. This is intended to reduce the reporting burdens and allow providers to move more quickly through the stages. Providers will be required to at least attest to the Modified Stage 2 beginning in 2015, with limited exceptions. This should help ensure that all providers will be ready and able to transition to Stage 3, and thus possess the same capabilities, in 2018—a necessary achievement in order for interoperability to be successful.

Calendar year reporting will begin in 2015, as opposed to fiscal year reporting, as has been the case. Providers still only need to attest to meaningful use for any continuous 90-day period through 2015, but beginning in 2016 and in 2017, they must attest for the entire year.[3] Full year reporting will be required of all providers beyond that—this is critical to meeting the goals of the program: the more data collected, the more useful it is for comprehensive analysis and care improvement.

Beginning in 2015, measures that are redundant, duplicative, or considered to be “topped out” will be removed; this will be an ongoing process to reduce the burdensome reporting requirements and to allow for the most appropriate use of funds.[4] Topped out measures do not provide an opportunity to differentiate performance among providers and presumably no longer need to be incentivized. Certain other measures will be selected as “high-priority,” particularly for use in the Clinical Decision Support objective.

Objectives and Corresponding Measures for Modified Stage 2: 2015-2017

The following objectives and corresponding measures will be used for evaluating whether or not a provider has met the necessary requirements for attestation to Modified Stage 2 in 2015, 2016, and 2017:

Stage 3

Stage 3 is intended to bring about advancements in care delivery by requiring more advanced EHR functionality and standards for structuring data, increasing thresholds compared to Stage 1 and 2 measures, and requiring more coordinated care and patient engagement. All providers will be required to meet the Stage 3 objectives in 2018 for the entire calendar year, but providers will be encouraged and able to begin attesting to Stage 3 in 2017.

Objectives for Stage 3: 2017 and beyond

The following objectives and corresponding measures will be used for evaluating whether or not a provider has met the necessary requires for attestation to Stage 3 in 2017 and subsequent years. All CEHRT must be able to perform the necessary functions to meet these objectives, including recording and reporting all necessary data electronically.  

Medicaid Providers

The Medicaid EHR Incentive Program is voluntary for the states, and each state may choose whether or not to administer such a program. Providers treating both Medicare and Medicaid patients can avoid Medicare penalties by successfully demonstrating meaningful use to their state Medicaid agency, even if it occurs after the Medicare attestation period closes, if the state chooses to participate. States must determine whether and how electronic reporting of CQMs would occur, or whether they wish to allow reporting through attestation. The federal government will fund 90 percent of the state’s administrative costs for implementing the technology and systems necessary for compliance.

Medicare Advantage Incentive Payments

Medicare Advantage Organizations (MAOs) are unique in how they are treated under this incentive program because providers treating MA patients are not paid directly by CMS, as providers treating traditional fee-for-service (FFS) providers are. Instead, CMS pays MAOs a flat monthly fee for each enrollee at the beginning of each month. Through these incentive programs, MAOs may receive payment adjustments according to how well their patients are treated, similarly to how payments are adjusted for providers. In Stage 3, there will not be any changes to the existing policies and regulations for MAOs.

Conclusion

CMS is “statutorily required to require more stringent measures of meaningful use over time.” Since 2009, requirements have gradually increased and required more advanced technology. Each progression is designed to move all providers in the same direction, create standards of practice and for capturing data, and allow for improved patient outcomes. Progress is being made: in 2014, 57 percent of office-based physicians electronically shared health information with their patients and 42 percent electronically shared patient information with other providers, including 26 percent who electronically shared patient information with providers outside their office or group.[5] CMS is also seeking comments on adopting meaningful use attestation into the new MIPS payment scheme which will be implemented beginning in 2019.

• Download

Executive Summary

By 2020 – a mere five years away – we project the United States and Silicon Valley economies will face crucial labor force skill shortages. Our analysis and policy recommendations make the case for a multi-pronged approach to meeting the projected skills gaps: increase the growth rate in immigrant workers across the economy; adjusting visas to better reflect skills mismatches; improving training and education opportunities for the existing immigrant workforce; and allowing training and education earned abroad to be applied to degree attainment and U.S. licensure processes here. Regardless of the policy mix chosen, failing to address these challenges will undermine the United States’ global competitiveness.

United States

United States Private Sector Projected Worker Shortages

In 2020 the United States will be short roughly 7.5 million private sector workers, with substantial shortfalls occurring across all skill levels. While improving training for the existing workforce, increasing the growth rate in the immigration workforce at all skill levels would substantially reduce overall shortages. We find that increasing the growth rate of immigrant workers by 25 percent across all skill levels would decrease the United States’ labor shortage to 6.4 million, mostly by eliminating middle- and high-skill labor shortages.

Silicon Valley

Silicon Valley Private Sector Projected Worker Shortages

Silicon Valley meanwhile will be short 72.5 thousand private sector workers, with the shortfalls occurring at lower- and middle-skill jobs. Increasing the growth rate of immigrant workers by 25 percent across all skill levels would decrease the region’s shortage from 72.5 thousand to 46.4 thousand.

Policy Options

Changes in immigration policies and patterns can help fill the middle-skill gaps in the Silicon Valley and meet our nation’s economic needs. Policy options that could increase the immigrant workforce to fill these jobs include:

1. Broad Immigration Reform that Ensures Employers’ Labor Needs Are Met

Congress could enact broad immigration reform policies that ensure that future labor needs are met through the legal immigration system. The employment-based visa program needs to be reformed to reflect the different workforce needs of employers and our country and to include flexibility for the visa quotas to adjust to economic realities. 

2. Adjust the Mix of Visas to Better Match Skill Needs of Employers

The proportion of working age immigrants entering the United States with at least a Bachelor’s degree has increased substantially. In particular, in 2013 over 50 percent of working age immigrants entering the United States and over 60 percent entering Silicon Valley had at least a Bachelor’s degree. One option would be to expand visas that target immigrants with an Associate’s level education or to create new temporary and permanent work visas that are specifically for workers at this skill level.

3. Increase Educational and Skills of Existing Immigrant Workforce

State educational policies could encourage and enable immigrant workers already living in the United States and Silicon Valley to obtain an Associate’s degree or comparable industry-recognized credential to meet the middle-skills gaps. Federal, state, and local governments could promote effective integration of education (adult education, community college, four-year institutions) and workforce support services. Integrating these systems can assist immigrant workers in accelerating language acquisition while concurrently developing knowledge and job skills needed to obtain occupational credentials. Government offices could share successful preparation models. State and local models, such as career pathways, can be applied to preparing immigrant workers for middle-skill jobs. 

Employers could also provide development opportunities for native and immigrant staff, such as in-house training, partnering with community colleges and postsecondary institutions, and joining with community organizations to provide contextualized English language classes and supportive services. 

4. Obtain Credit for Work, Experience, and Credentials from Abroad

Employers and the nation’s economy benefit from having skilled immigrant professionals that are readily able to contribute their expertise and credentials gained from abroad. State educational policies could enable immigrant workers to receive credit for work, experience, and credentials attained from abroad. This could increase the value of education and training for immigrant workers because it would limit redundancy in their personal education, accelerate their degree completion, and save them time and money. Additionally, states could consider how immigrants can gain credit for training and work experience gained abroad when applying for licensure in the United States, while maintaining public health and safety standards. 

Introduction

Silicon Valley’s regional economy is among the fastest growing in the United States. From 2003 to 2013 real Gross Domestic Product (GDP) in the San Jose-Sunnyvale-Santa Clara metropolitan area grew at an average annual rate of 4.5 percent, significantly higher than the 1.8 percent average growth rate nationwide.[1] During that time period, employment in San Mateo and Santa Clara counties grew 9.8 percent, more than twice as quickly as the 4.8 percent job growth in the entire United States. Silicon Valley is certainly growing fast and employers will continue to demand a multitude of workers at all skill levels. But, with a national labor force participation rate at a historic low, will there be enough workers to fill all those job openings?

In this paper, we project the difference between the skills employers will demand in 2020 and the skills workers will supply to estimate potential skill gaps in the economy. In turn, we project that by 2020 the United States and Silicon Valley economies could be confronted with significant worker shortages at multiple skill levels. There are different options to developing a pipeline of skilled workers, and this paper focuses on how adjusting the growth of immigrant workers could substantially close the impending skill gaps in the United States and Silicon Valley.  

Although we find that it would be helpful to use immigration to address future skill gaps, doing so would require substantial shifts from recent immigrant trends. In particular, from 2003 to 2013 the portion of immigrants coming to the United States and Silicon Valley with Bachelor’s degrees has increased substantially and the portion with all other skill levels has decreased. As a result, absent a change in policy or immigration patterns, increasing the immigrant workforce would bring in additional high-skilled workers, but not a large number of low- or middle-skilled workers. Our analysis leads to policy recommendations to help fill middle-skill gaps by ensuring that immigrant workers can gain skills in the United States and apply their work experiences and credentials attained abroad.

Education and Immigration Characteristics in the United States and Silicon Valley

When examining the impact of immigration on the supply of skilled labor, it is important to analyze regions that actually rely on foreign born workers and a highly skilled labor force. This is why we compare labor market trends in the entire United States to trends in Silicon Valley, where there is a highly educated workforce and a large concentration of immigrants.

In this paper, we analyze growth of workers at four distinct education levels. These four categories of workers include those who (1) have not completed high school, (2) obtained a high school degree but did not complete any higher level of education, (3) completed some college or obtained an Associate’s degree but did not obtain a Bachelor’s degree, and (4) obtained a Bachelor’s, Master’s, Professional, or Doctorate degree.

Table 1 compares educational attainment of workers ages 25 to 64 nationwide to Silicon Valley.[2]

Table 1: Worker Education Characteristics

Overall, Silicon Valley has a much higher prevalence of highly skilled workers and a lower concentration of middle-skilled workers than the entire United States. In particular, 33.7 percent of workers in the United States have Bachelor’s degrees. In Silicon Valley, those workers represent 50 percent of the workforce. However, the United States has a greater percentage of workers with high school and Associate’s degrees than Silicon Valley.

Silicon Valley also has a much higher concentration of foreign residents than the entire United States. Table 2 compares the immigrant population in the United States to Silicon Valley.[4]

Table 2: Immigrant Residents

In the entire United States, immigrants represented 13 percent of the population in 2013, but, in Silicon Valley they were 36.2 percent of residents. In other words, a person in Silicon Valley is almost three times more likely to be an immigrant than a person in the United States as a whole.

Going forward, Silicon Valley will continue to demand skilled workers at a higher rate than the entire United States. The California Employment Development Department projects that between 2012 and 2022, 8 of the 30 occupations with the most job openings in San Mateo County and 13 of the top 30 in Santa Clara County will require a Bachelor’s degree or more.[6] The Bureau of Labor Statistics meanwhile projects that only 3 of the 30 occupations with the most job openings in the entire United States will require a Bachelor’s degree or more.[7]

In the following, we utilize recent employment trends to estimate if there will be enough workers (both native and foreign) to meet employment needs in 2020 at all skill levels in the United States and Silicon Valley.

Methodology for Estimating Workforce Skill Shortages

United States

There are three steps to projecting education shortfalls in the labor market by 2020. First, we project the number of workers the labor market will have at each of four education levels. Second, we project the number of workers at each of those four education levels that employers will demand. And finally, to yield the shortage (or surplus) of workers by skill level we calculate the difference between the projected number workers at each education level employers will demand and the number of workers at each education level who will actually exist.

To project the number of workers who will exist across all skills, we use Current Population Survey (CPS) March Supplements to estimate the compounded annual growth rate of workers at each of the four education levels in each industry from 2003 to 2013.[8] We perform this exercise for all workers, immigrant workers, and non-immigrant workers with the national population weights available in the CPS data files. Assuming those long-term average growth rates will remain constant going forward, we project the growth in workforce levels in each industry from 2013 to 2020 for all, native, and foreign workers at each education level.

When estimating the number of jobs at each skill level employers will seek, we use data reported by the Georgetown Center on Education and Workforce (CEW) to calculate the compounded annual growth rates of workers demanded at each skill level in every industry.[9] Then starting with actual employment levels in 2013, we use these compounded annual growth rates to project the growth in the number of jobs available in 2020 that will require less than a high school education, a high school degree, some college or an Associate’s degree, and a Bachelor’s, Master’s, Professional, or Doctorate degree.

Finally, by comparing the projected number of workers who will be in the labor force at every education level in 2020 to those employers will require, we estimate the future shortages or surpluses in workers by skill level in each industry.

Silicon Valley

To our knowledge, a detailed data set similar to the CPS is not available for Silicon Valley. To project the potential worker shortages or surpluses specifically in Silicon Valley, we use the exact same data and methods as we did for the United States with one key difference. Instead of using the national population weight in the CPS, we use regional Silicon Valley demographic characteristics to weigh the national data set so it more closely resembles Silicon Valley’s population. The demographic information we use to weigh the national CPS data came from the Census Bureau and includes four main categories: race, Hispanic ethnicity, immigrant status, and age.[10]

By combining the four main demographic categories we derive 112 unique demographic subcategories. We then calculate relative probabilities and apply them to the entire national sample as weights for the 112 different subcategories of people. In essence the relative probabilities indicate the number of people each observation in the national sample would represent in Silicon Valley. For instance, one relative probability indicates that for every person in the national sample who in 2013 was foreign, Asian, not Hispanic, and between the ages of 25 to 34, there were 2.3 people who met those characteristics in Silicon Valley.

Limitations to the Silicon Valley Analysis

There are three central limitations to our Silicon Valley analysis that are important to understand.

First, to derive the relative probabilities, we use the four main categories to estimate the percent of people who fall into each of the 112 subcategories in the national sample and in Silicon Valley. We then divide the percentage of people in each subcategory in Silicon Valley by the percentage in national sample. With the CPS, we could directly find the percent of the sample that falls under each of the 112 demographic subcategories in the national sample. For Silicon Valley, we only have direct information for the distribution of people in the region for each of the 4 major demographic categories. Thus, to find the percent of people in each of the 112 unique categories we multiply together the percentages of the four major categories. For instance, we find the percent of people in Silicon Valley who are foreign, Asian, not Hispanic, and between the ages of 25 to 34 by multiplying the percent of people in Silicon Valley who are foreign by the percent who are Asian, the percent who are not Hispanic, and the percent who are ages 25 to 34. The major limitation with this part of our analysis is that we assume that each subgroup has the same demographic characteristics as the entire region. For instance, 32.2 percent of the entire region of Silicon Valley is Asian and our methods result in us assuming that 32.2 percent of people who are foreign, not Hispanic, and ages 25 to 34 are also Asian.

Second, the Silicon Valley population weights help the national data resemble that specific region, which allows us to track the growth in workers at all education levels within the region. This in turn allows us to project the growth in skills supplied by Silicon Valley workers to 2020. But, it does not allow us to make Silicon Valley specific projections for the growth in skills required by employers. As a result, for our Silicon Valley analysis, we assume that the projected growth rates in workers demanded in each industry matches the national growth rates and that the distributions of workers demanded by skill level in 2020 will match the national distributions.

Third and finally, although we weigh the national sample from the CPS to reflect the Silicon Valley population, we do not actually analyze a sample of people from that region of the country. As a result, if there are trends unique to Silicon Valley that are not captured in our weights, we are not able to account for them.

Despite these limitations, we are confident that our analysis provides informative projections of skills needs across industries within the Silicon Valley by 2020.

Projected Labor Shortages

Our projected skills shortfalls and surpluses in 2020 for the private sector of the entire United States and just Silicon Valley are shown in Table 3.[11] Estimated shortfalls and surpluses for each industry are available in the Appendix.

Table 3: Private Sector Projected Worker Shortages: Nation vs. Silicon Valley

Overall, we find that both the entire nation and Silicon Valley will face substantial and similar worker shortages in 2020. In the United States, there will be about 7.5 million or 4.9 percent fewer workers than demanded by employers. Silicon Valley employers meanwhile will be short 72.5 thousand workers, which translates to a 5.0 percent shortfall.

In the United States, a worker shortage will occur at every skill level. In 2020, the nation will be short 5.3 million workers (29.3 percent) who did not complete high school, 1.2 million (3 percent) with a high school degree, 123 thousand (0.3 percent) with some college or an Associate’s degree, and 876.9 thousand (1.8 percent) with a Bachelor’s degree or higher.

Silicon Valley, however, will face a greater shortage for middle-skilled jobs. In particular, the region will be short 66.9 thousand workers with a high school degree. This means that the gap between employer demands and worker supply for high school degree holders will be 17.3 percent, over five times larger than the 3.0 percent gap nationwide. Silicon Valley will also have a 7.4 percent shortage of workers with some college or an Associate’s degree, which is much larger than the national shortfall of 0.3 percent.

Finally, the table shows that while the entire nation will be short highly skilled workers, Silicon Valley will have surplus of workers with at least a Bachelor’s degree. In particular, we project that the region will have 78.4 thousand or 16.3 percent more workers with a Bachelor’s degree or above than employers will demand.

It is important to note that the projected surplus of workers with at least a Bachelor’s degree in Silicon Valley could be due to the limitations of this analysis previously highlighted. In particular, due to lack of data we assume that the projected growth rates in workers demanded in each industry in Silicon Valley matches the national growth rates and that the distributions of workers demanded by skill level in 2020 will match the national distributions. In reality, however, it is likely that employer demand for workers with at least a Bachelor’s degree will grow faster in Silicon Valley than the rest of the nation. As a result, we suspect that the surplus of worker’s with at least a Bachelor’s degree shown in Table 3 may be overstated or may not actually exist.

Immigrant Workers’ Impact on the Shortfall in Skill Levels

Adjusting the growth rate of immigrant workers at all skill levels reveals how important immigrants can be to the future of the labor force. In particular, keeping the number of immigrant workers from growing would cause the skill gaps to widen substantially. Increasing the growth of immigrant workers, however, would reduce the projected skill shortfalls. In general, we find that the higher the education level, the more responsive the worker shortfalls and surpluses are to increasing or decreasing the growth of immigrant workers.

Stopping Growth of Immigrant Workers                         

If instead of growing at the same rate as the previous ten years, immigrant workers in each skill level did not change and stayed at their 2013 levels, then the private sector labor shortages in 2020 would increase substantially both nationally and in Silicon Valley. This is illustrated in Table 4.

Table 4: Private Sector Worker Shortage if Immigrants Increase 0%: Nation vs Silicon Valley

If the number of immigrants at each education level did not change after 2013, then the overall worker shortage in 2020 would grow from 7.5 million to 11.2 million nationally (4.9 percent to 7.3 percent) and from 72.5 thousand to 138.8 thousand in Silicon Valley (5.0 percent to 9.5 percent). In general, the higher the education level, the quicker the worker shortage would rise.

Nationally shortages for workers with high school degrees and greater would grow substantially. The shortfall would rise from 1.2 million to 2.2 million (3.0 percent to 5.4 percent) for workers with a high school degree, 123 thousand to 931.3 thousand (0.3 percent to 2.1 percent) for workers with some college or an Associate’s degree, and 876.9 thousand to 2.8 million (1.8 percent to 5.6 percent) for workers a Bachelor’s degree or above. Generally, the higher the education level the more the shortage would grow if immigrant workers were to simply stay at their 2013 levels.

In Silicon Valley, keeping immigrant workers at 2013 levels would cause the shortages of workers with high school degrees and some college or Associate’s degrees in 2020 to increase and the surplus of workers with Bachelor’s degrees or above to decrease. In particular, the 2020 shortage of workers with high school degrees would increase from 66.9 thousand to 71.2 thousand (17.3 percent to 18.5 percent). The shortage of workers with some college or Associate’s degrees would rise from 31.2 thousand to 50.4 thousand (7.4 percent to 11.9 percent). Meanwhile, the surplus of workers with Bachelor’s degrees or above would fall from 78.4 thousand to 28.7 thousand (16.3 percent to 5.9 percent).

Increasing Growth of Immigrant Workers

Increasing the annual growth rate of immigrant workers at each skill level would reduce the projected 2020 private sector skill shortfalls. Table 5 shows the shortages and surpluses of workers if the annual growth rate of immigrant workers at each skill level increased by 25 percent.

Table 5: Private Sector Worker Shortage if Immigrant Growth Rate Increase 25%: Nation vs. Silicon Valley

The nationwide private sector total labor shortage would fall from 7.5 million to 6.4 million (4.9 percent to 4.2 percent) and the Silicon Valley private sector labor shortage would decrease from 72.5 thousand to 46.4 thousand (5.0 percent to 3.2 percent). Like above, worker shortages and surpluses are more responsive to increasing immigration growth at the higher end of the skill distribution. Specifically, increasing the growth rate of immigrant workers would reduce the middle- and high-skilled worker shortages the most.

Nationally, the shortages of workers with some college or Associate’s degrees would disappear entirely and the shortages of workers with Bachelor’s degrees or greater would decline substantially. The shortage of workers with some college or an Associate’s degree would become a 0.2 percent surplus and the shortage of workers with a Bachelor’s degree or greater would fall to 0.7 percent.

In Silicon Valley, the shortage of workers with some college or Associate’s degrees in the private sector would fall from 31.2 thousand to 25.1 thousand (7.4 percent to 5.9 percent). Meanwhile, increasing the growth rate of immigrant workers with a Bachelor’s degree or greater would increase the surplus of workers in that category from 78.4 thousand to 91.6 thousand (16.3 percent to 19.1 percent).

How Immigrants Influence Labor Force Skill Levels

Immigrant workers expand the talent pool that employers can utilize to meet their workforce needs. Overall, we find that in recent history a greater proportion of working-age immigrants coming to the United States and Silicon Valley have been entering with at least a Bachelor’s degree, while the proportion with all other skill levels has been falling. Meanwhile, once in the United States and Silicon Valley very few immigrants upgrade their skills by attaining higher education.

Immigrants are more frequently coming to the United States and Silicon Valley with High Skills Levels

The evidence suggests that the proportion of new immigrants coming to the United States and Silicon Valley with at least a Bachelor’s degree has increased substantially over time. Table 6 compares the skills of prime working age immigrants (25 to 54) who came to the United States from 2002 to 2004 to the skills of those who entered from 2012 to 2014.[12]

Table 6: Prime Working Age Immigrant Entrant Skill Levels in United States

Over the ten-year period, the proportion of prime working age immigrants who came to the United States with at least a Bachelor’s degree increased by a substantial 14.1 percentage points from 38.3 percent in the 2002 to 2004 period to 52.4 percent in the 2012 to 2014 period. Meanwhile, the percentage of entrants at all other skill levels fell over the ten-year time frame.

Table 7 shows how skills of immigrant entrants who moved to Silicon Valley changed over the same time period.

Table 7: Prime Working Age Immigrant Entrant Skill Levels in Silicon Valley

The proportion of immigrants entering Silicon Valley with at least a Bachelor’s degree increased at an even faster rate than it did nationwide. Over the ten-year time period, the proportion of immigrants entering with at least a Bachelor’s degree increased by 14.9 percentage points from 46.6 percent to 61.4 percent. Just like for the entire United States, the proportion of prime working age immigrants entering with low- and middle-skill levels decreased.

Tables 6 and 7 indicate that immigrant workers entering the country are already well situated to increase the high skilled labor force. However, they are not well positioned to fill low- and middle-skilled jobs. This is particularly important in Silicon Valley, where we found that in 2020 there would be a surplus of workers with Bachelor’s degrees and a shortage of workers with some college or an Associate’s degree.

Once in the Country, Few Immigrants Upgrade their Skills

Recent data suggest that once immigrants arrive in the United States and Silicon Valley, they do not upgrade their skills as frequently as native residents. This is particularly true for middle-skilled education, as immigrants obtain some college or an Associate’s degree at a slower rate than any other education level. Using data from the Survey of Income and Program Participation (SIPP), we analyze how immigrants and non-immigrants enhanced their education levels from 2008 to 2013.[13]

Table 8 reveals what happened among all immigrants in the United States.[14]

Table 8: Immigrant Learning in United States

From 2008 to 2013, the proportion of immigrants who completed some college or an Associate’s degree was virtually unchanged, only increasing by 0.2 percentage point. The proportion with at least a Bachelor’s degree only increased 1 percentage point.

Table 9 reveals that meanwhile non-immigrants upgraded their skills more rapidly.

Table 9: Non-Immigrant Learning in United States

From 2008 to 2013, the proportion of non-immigrants in the United States with some college or an Associate’s degree increased by 1.4 percentage points and the proportion with at least a Bachelor’s degree increased by 2.6 percentage points.

Table 10 reveals that immigrants in Silicon Valley are not upgrading their skills much more rapidly than immigrants in the entire United States.

Table 10: Immigrant Learning in Silicon Valley

Over the five-year time period, the proportion of immigrants in Silicon Valley with some college or an Associate’s degree only increased by 0.2 percentage point. The proportion of immigrants with at least a Bachelor’s degree grew by 2 percentage points, which was quicker than the increase that occurred nationally.

Table 11 indicates that non-immigrants in Silicon Valley upgraded their skills at rates similar to the entire nation.

Table 11: Non-Immigrant Learning in Silicon Valley

By 2013 the proportion of non-immigrants in Silicon Valley with some college or an Associate’s degree increased by 1.3 percentage points. In addition, the proportion with at least a Bachelor’s degree increased by 2 percentage points. This means that in Silicon Valley, while immigrants obtain a Bachelor’s degree at the same rate as non-immigrants, they still lag behind in the some college or Associate’s degree category.

Policy Implications

While trends in the last ten years show that immigrant workers are well situated to fill jobs that require a Bachelor’s degree, our projections suggest that the United States and particularly Silicon Valley will face substantial shortfalls in workers with low- to middle-education levels. As a result, in order to use immigration as a tool to reduce worker shortages, particularly at the middle-skill level, changes in immigration policies and patterns would be useful. Four ways policy could increase middle-skilled workers in the United States and Silicon Valley are: (1) broad immigration reform that ensures employers’ labor needs are met; (2) adjust the U.S. visa mix so that a larger portion of immigrant workers entering the United States have middle education levels; (3) increase the educational and skills attainment of immigrant workers that are already in the United States; and (4) enable immigrants in the United States to obtain credit for their work, experience, and credentials gained from abroad.

Broad Immigration Reform

Employers in the United States and Silicon Valley need a reliable supply of workers to maximize their productivity. Congress should enact broad immigration reform policies that ensure that future labor needs are met through the legal immigration system. This includes solutions that allow immigrant workers to upgrade their skills through efficient and effective education and workforce programs and for employers in the United States and Silicon Valley to access these workers. Further, the employment-based visa program needs to be reformed to reflect the different workforce needs of employers and our country and to include flexibility for the visa quotas to adjust to economic realities. 

Adjust the U.S. Visa Mix

Over a ten-year period, the proportion of working age immigrants entering the United States with at least a Bachelor’s degree increased substantially. In particular, in 2013 over 50 percent of working age immigrants entering the United States and over 60 percent entering Silicon Valley had at least a Bachelor’s degree. These trends are likely largely explained by the fact that the U.S. visa system is designed to primarily grant legal residence to immigrant workers with very high skills and has given more and more preference to those with at least a Bachelor’s degree over time.

Examining worker admissions under permanent worker visas (aka green cards) in Table 12, it is easy to see how the U.S. visa system over time has given more preference to high skilled workers.

Table 12: Permanent Admissions for Workers

Between 2004 and 2013, the total number of workers admitted to the United States on a permanent worker visa has remained virtually unchanged at about 72 thousand. However, during that time period, the percentage of workers admitted with visas that require at least a Bachelor’s degree, EB-1 and EB-2, increased substantially. In particular, the percent of permanent workers admitted with an EB-1 visa rose 4.1 percentage points from 18.4 percent to 22.5 percent and the proportion admitted with an EB-2 visa rose a dramatic 21.2 percentage points from 21.9 percent to 43.1 percent. As a result, visas that require a high level of educational attainment now account for two-thirds of all permanent worker admissions in the United States.

The consequence of this, however, is that fewer visas are being made available for workers with middle skills. In fact, the proportion of permanent workers admitted to the United States with the visa that allows for middle skills, EB-3, has declined significantly. The proportion of permanent workers admitted with an EB-3 visa dropped 29.3 percentage points from 57 percent to 27.7 percent. Within EB-3, the visa under the “skilled” category requires at least two years of training and most closely aligns with an Associate’s level education.[17] However, the proportion of permanent workers admitted with a “skilled” EB-3 visa declined 15.2 percentage points from 27.1 percent to 11.9 percent.

Table 13 shows the distribution of workers admitted under temporary employment visas, such as the H-1B.

Table 13: Temporary Admissions for Workers[18]

Overall, the number of workers admitted with temporary worker visas has spiked over the last ten years from 675,647 in 2004 to 1,613,868 in 2013. Despite the number of temporary worker admissions increasing under most visa categories (particularly for NAFTA workers), employment admissions with a nonagricultural visa (H-2B), which is intended for low-and middle-skill workers, remained flat at about 100,000 per year. As the table shows, this resulted in nonagricultural worker visa admissions falling from 12.9 percent of temporary worker admissions in 2004 to only 6.5 percent in 2013.

Since we project that Silicon Valley will have a substantial shortage of workers with some college or an Associate’s degree in 2020, the region could benefit from expanding visas that target immigrants with an Associate’s level education or creating new temporary and permanent work visas that are specifically for workers at this skill level.

Increase Educational and Skills Attainment of Immigrants Already in the United States

Federal and state educational policies could encourage and enable immigrant workers already living in the United States and Silicon Valley to obtain an Associate’s degree or comparable industry-recognized credential to fill middle-skills gaps. As indicated above, once in the United States, immigrants do not upgrade their education as frequently as non-immigrants. This is particularly true when it comes to obtaining some college or an Associate’s degree, as the proportion of immigrants at those education levels remained virtually unchanged in both the United States and Silicon Valley from 2008 to 2013.Encouraging immigrants to obtain an Associate’s degree or comparable industry-recognized credential in the United States not only helps meet employers’ demands for middle-skill workers, but also can benefit immigrants themselves. A study of 4,000 immigrant professionals by the World Education Services, IMPRINT, and George Mason University’s Institute of Immigration Research found that immigrants “who had invested in additional U.S. education were more likely to be employed and successful than those who had only received education abroad.”[19]

The Workforce Innovation and Opportunity Act, enacted on July 22, 2014, presents an opportunity to improve education and workforce services for all job seekers, including immigrants. With the help of this bill, federal, state, and local officials can promote effective integration of education (adult education, community college, four-year institutions) and workforce support services. Integrating these services can assist immigrant workers accelerate language acquisition and concurrently attain educational or other industry-recognized credentials needed for the growing industries and occupations in their local or regional economies. 

Coordinating education and workforce support services is particularly important for immigrants because they frequently face cultural, linguistic, and socioeconomic barriers to successfully advancing their education and achieving their employment goals. In particular, immigrants may face unique barriers such as lack of professional networks and limited English language abilities.[20] In Silicon Valley, more than half of working adults speak a language other than English in the home, of which 58 percent are fluent in English and 42 percent are English Language Learners (ELL).[21] Of those ELLs in the Silicon Valley, 57 percent have a high school diploma or less.[22] Coordination and alignment of education and workforce support programs may help ensure that immigrants who seek to improve their employability can do so efficiently.

One potentially successful strategy to prepare immigrant and non-immigrant job seekers for employment is career pathways.[23] The federal government, states, philanthropic organizations, and local services are investing in career pathways initiatives, which align adult basic education, occupational training, postsecondary education, and supportive services to provide comprehensive and flexible education and training programs.[24] Career pathways programs aim to meet the needs of working students, including immigrants and English language learners.[25] Further research is needed on 1) how career pathways can help immigrant workers, including ELLs, enroll in and successfully complete educational and training programs, including those leading to an Associate’s degree or other industry-recognized credentials, and 2) how career pathways can build an immigrant workforce for growing, non-traditional industries and occupations to meet employer demand.

Employers can also play an important role in improving immigrant worker skills to fill middle-skill job openings. In a survey of 340 organizations of different sizes, industries, and locations, the Association for Talent Development noted that in 2013, organizations, on average, spent $1,208 per employee on training and development.[26] For example, many employers offer several types of development opportunities to staff such as a combination of mentoring, rotations, and on-the-job training.[27] Employers also partner with community colleges and postsecondary institutions to develop and offer customized virtual or classroom training for staff.[28] Immigrant workers may benefit from contextualized English literacy classes or supportive services, and benefit from employers that partner with organizations such as adult education and community-based organizations to provide these programs.[29]

Further, employers can incentivize staff to upgrade their skills by offering monetary bonuses, additional time off, and other rewards upon successful completion of training. Many workers participate in tuition assistance programs, where the employer contributes to part or all of the employees’ cost to earn postsecondary degrees and credentials.[30]

Credit for Work, Experience, and Credentials from Abroad

Immigrant workers bring to the United States diverse expertise and skills developed through employment and other experiences from abroad. To reduce educational redundancies and improve education and training efficiency for immigrant workers, state educational policies could enable immigrants to more frequently receive credit for work, experience, and credentials attained from abroad.

Prior learning assessments (PLA) is one model that postsecondary institutions are using to allow students to gain academic credit for knowledge developed outside of the classroom, such as through employment, professional training, volunteer activities, or military training. PLAs have helped adult learners accelerate their degree completion. In a study of 62,475 students across 48 postsecondary institutions, the Council for Adult and Experiential Learning (CAEL) found that PLA students had a much higher and faster degree-earning rate and were more persistent in accumulating credit towards a degree than non-PLA students. In particular, 13 percent of PLA students earned an Associate’s degree compared to 6 percent of non-PLA students; PLA students saved an average of 1.5 to 4.5 months of time towards the Associate’s degree.[31] Also, Hispanic students who earned credit through PLAs had higher graduation rates and required less time to complete degree programs then non-PLA students.[32] Further, since PLAs can save students time and money, they can incentivize immigrant workers to upgrade their skills through educational attainment.

Additionally, states could consider how immigrant professionals can gain credit for training and work experience gained abroad when applying for licensure in the United States, while maintaining public health and safety standards. This can ensure that skilled professionals are more readily able to contribute their expertise to the benefit of employers in the United States and economy at large. 

Conclusion

We find that by 2020 both the entire United States and Silicon Valley specifically will face substantial worker shortages. Silicon Valley in particular will have a significant shortage of workers for jobs that will require middle-level skills, such as an Associate’s degree. One way to address these potential skill shortages is through immigration. If policymakers were to increase the growth rate of immigrant workers at all skill levels by 25 percent, the overall worker shortages would decline considerably. Given recent immigration trends, however, this effort would require substantial policy changes, such as reforming our immigration visa program to account for the nation’s need for workers with middle skills or enabling more immigrants in the United States to advance their education.

Appendix