Chairman Smith, Ranking Member Johnson, and Members of the Committee, thank you for the opportunity to appear today. In this testimony, I wish to highlight the following points:

• “Midnight” regulations or rules typically issued after Election Day, but before the next president takes office tend to cause a period of increased regulatory activity. Under both Democratic and Republican administrations, the midnight quarter between November and January is highly active for rulemakings, including controversial measures. OIRA Administrator Howard Shelanski has pledged to curb this practice, but Americans won’t know the success of this effort for another year.

• Both the Department of Energy (DOE) and the Environmental Protection Agency (EPA) have a busy schedule of regulations remaining in 2016. The recent history of DOE and EPA rules has already resulted in tremendous economic costs with more than $500 billion in burdens.[1] This figure has real world consequences for consumers, employment, and the bottom line of countless companies.
• Reform options to curtail the practice of excessive midnight regulation could include the use of the Congressional Review Act (CRA), legislative approaches that ensure proper notice before final publication, and commitments from both the outgoing and incoming administrations to halt any expedited rulemakings.

History of Midnight Regulation

The term “midnight” regulation likely harkens back to the debate over President John Adams’s midnight judges of 1801. After the election of Thomas Jefferson, President Adams signed the commissions of several judges, including three on his last day in office. Today, midnight regulation is generally known as any rule published after Election Day, but before the next president takes office. Scholars often study the midnight quarter, from November to January, during this presidential transition.

During any quarter from November to January, federal regulators will still issue new rules, but there is an incentive for the outgoing administration to cement as many regulatory priorities as possible before the next administration takes office. Given the reality that some midnight regulation will occur, the question is whether a particular midnight quarter experiences significantly more regulatory activity than other quarters with no presidential transition.

For 2000 and 2008, the last time political parties switched hands in the White House, there was a pronounced spike in regulatory activity during this midnight period, for both economically significant rules (those with an economic impact of $100 million or greater) and all rulemakings. The graphs below chart these spikes.

As seen above, both 2000 and 2008 saw significant surges in regulatory activity. For significant activity, regulators implemented more than 40 rules during the 2000 midnight period or more than double the average for non-midnight years. In 2008, regulators also finalized nearly 40 significant rules. For comparison, in 2007 during the midnight quarter, there were only 22 significant measures.

The graph examining all rules looks nearly identical to the significant rule graph. In 2000, cabinet-agency regulators issued more than 130 rules, which is 1.7 times greater than the non-midnight average (74.3). The story was similar in 2008, when the administration approved 137 rues, nearly twice the non-midnight average.

It should be noted that 2004 and 2012 did not experience any noticeable increase in activity; in fact, 2012 experienced some of the slowest midnight activity during the period studied. Why? A simple explanation is there was no need to rush rules through the process because at that time for both President Bush and President Obama, there was no midnight for their term in office. Both presidents knew they had four additional years in office.

There is some evidence, as reported by the Washington Post, that the Obama Administration intentionally delayed controversial rules until after Election Day.[2] One notable rule, Tier 3 standards for vehicle fuel, will slightly increase the price of gasoline. The White House signaled, at a time of high gas prices, that this rulemaking might be politically controversial. Thus, Tier 3 wasn’t officially proposed until May 21, 2013.[3]

Neither EPA nor DOE have been immune to the surge in midnight rules. According to OIRA, EPA is the most active regulator, with regard to total costs and benefits, and DOE is not far behind, in third place among cabinet agencies.[4] Any regulatory wave will likely involve these two agencies and the evidence reveals they have played notable roles in past midnight regulation. In 2000, the administration issued eight economically significant EPA and DOE rules or more than five times the non-midnight average of 1.5 rules. In 2008, regulators finalized four significant rules from these agencies; while not as high as 2009, this figure is still more than twice the non-midnight average. The graph below highlights the midnight trends at EPA and DOE.

It’s important to note that the phenomenon of midnight regulation has occurred under Democratic and Republican administrations. The problem generally lies outside of politics, although cementing political priorities motivates midnight regulation. If a rule is sped through the regulatory process, the public and OIRA likely have less time to review and scrutinize the rulemaking. If expedited too quickly, regulators could ignore public input altogether, a hallmark of the rulemaking process. OIRA’s staff, already overworked, could fail to identify legal, economic, or practical implementation problems with regulation. This does not benefit anyone affected by federal regulation, which is to say, everyone: the environment, consumers, employees, and businesses.

There is a growing body of research that suggests expedited rulemakings lead to poor economic analysis. The Mercatus Center at George Mason University has found shorter review times lead to lower-quality regulatory analysis.[5] This poor analysis could result in unintended consequences for rules after implementation or the inability of scholars and regulators to judge the success or failure of a rule. The Administrative Procedure Act established a scheduled and deliberate timeline for federal regulation and the outcome of a presidential election should not alter the rulemaking process.

Finally, a more immediate concern for Congress and the administration might not be midnight regulation, but rules rushed before the carryover provision of the CRA becomes effective. According to AAF’s calculations based on initial House and Senate calendars, any regulation issued in late May would likely be subject to a CRA resolution of disapproval in 2017.[6] There are many factors that affect the day when the carryover provision applies and the administration’s possible motivations: number of days Congress is actually in session, the outcome of 2016 elections, and procedural hurdles that might halt any rush in regulation this spring. It is too early to tell whether the administration acknowledges the day when new rules are subject to the CRA and if current rules are on an expedited schedule. However, Congress and scholars should examine activity in the coming months to determine if there is a mini-rush in federal regulation to avoid a CRA vote.

Remaining DOE and EPA Rulemakings

Given the historical problem of midnight regulation and the outsized role of DOE and EPA in federal rulemakings, it’s necessary to scrutinize their remaining regulatory agendas and their records to date.

Since 2006, DOE and EPA have imposed 298 rulemakings with a quantified cost, benefit, or paperwork burden. Since that time, the agencies have published $513 billion in regulatory costs, accompanied by more than 33 million paperwork burden hours.[7] To put these figures into perspective, it would take more than 16,500 employees (working 2,000 hours annually) to complete 33 million hours of paperwork. The $513 billion figure amounts to a regulatory burden of $1,594 per person from just these two agencies. EPA’s paperwork burden has grown 26 percent since 2006 and now sits at a historic high.

As noted above, based on AAF data (available at www.regrodeo.com), not only has regulatory activity at DOE and EPA peaked during election years, but costs have as well:

As shown above, regulators tend to publish expensive regulations during election years, but this trend is less pronounced than during midnight periods. The median cost for election year rules from EPA and DOE rules is $38.1 billion, compared to $29.3 billion for all years. It’s important to note 2016 is included in the above tally, even though there is only one month of data. Already, the two agencies have added $15.7 billion in costs, more than 2013, 2008, and 2006. January could portend a busy year for EPA or DOE or possibly fade like 2013.

Regulatory activity does not appear to be waning at DOE. There are six economically significant rulemakings at OIRA for review currently (as of this writing):[8]

• Efficiency Standards for Portable Air Conditioners
• Efficiency Standards for General Service Lamps
• Efficiency Standards for Manufactured Housing
• Efficiency Standards for Packaged Boilers
• Efficiency Standards for Water Heating Equipment
• Efficiency Coverage Determination for Computers and Backup Batteries

If there is a slowdown in regulation, it likely won’t be from DOE. There is little data available now on the possible costs and benefits of these pending proposals, but based on previous data, the average significant DOE rulemaking from the Obama Administration lists $487 million in annual costs and $1.6 billion in annual benefits; for net present value, the average costs are an astounding $9.8 billion. Thus, if the past averages are predictive, these six rulemaking could impose more than $2.9 billion in new annual costs.

For EPA, there is presently just one economically significant measure pending at OIRA, but the historical totals from the agency have had a tremendous impact on the U.S. economy. For example, on average the net present value costs for EPA final rules tops $10 billion, with an annualized burden of $1.6 billion; the average paperwork burden associated with these rulemakings eclipses 870,000 hours.

Beyond what is currently at OIRA, there is a slate of rulemakings that could be finalized later this year, into the midnight period. In June EPA plans to finalize a new round of emissions standards for fracking.[9] The proposed version of this rule projected costs near $420 million.[10] July is also slated to be a busy month for EPA and DOE. Another round of dishwasher efficiency standards is expected then, with possible costs of $7.1 billion.[11] There are plans to finalize a second round of efficiency standards for heavy-duty trucks and engines. The total cost of that rule could approach $31 billion, according to initial agency estimates.[12] It’s notable that this rulemaking was initially scheduled to arrive at OIRA for review in January 2017 during the heart of the midnight period.[13] A revision to the rulemaking timeline sped the process up to July of this year.[14] If all of these measures are finalized in 2016, it could easily challenge previous records for regulatory burdens.

What do all of these figures and rulemaking timelines mean for consumers and businesses? Typically, consumers bear the cost of regulation through higher prices, but that is not always a certainty. Someone must bear the burden when a regulation imposes transition costs: employees at a firm through layoffs or reduced wages, lower returns for the company, or costs passed on to the consumer in the form of higher prices.

For consumers, last year AAF detailed how DOE’s regulations routinely forecasted both layoffs for company employees and higher prices.[15] For example, consumers are paying $2,300 in price increases from just ten recent rules. Few consumers will be affected by every rule, but a hypothetical individual buying a new refrigerator, furnace fan, and water heater could face a “regulatory tax” of $620. In most instances, the consumer would have no knowledge that federal regulations drove up the price of the item.

EPA regulations are also not immune from driving up prices for consumers. The agency’s regulatory analysis for a recent renewable fuel measure admitted the regulation could increase food prices by “roughly $10 per person per year.”[16] Today, this equates to more than $3.2 billion in higher food prices because of the mandate. EPA also conceded that: its Mercury Air Toxics rule would increase electricity prices by 3.1 percent, its Cross-State Air Pollution rule would boost prices by 1.7 percent,[17] and the Clean Power Plan could drive up electricity costs by 3.2 percent.[18] In isolation, these prices increase appear minor. However, they amount to a combined 8 percent hike in utility bills, a significant impact for many middle-to-low income Americans.

Agencies also acknowledge that these regulations have tremendous benefits for the environment and society. Yet, some assumptions underlying these benefit figures have been questioned by academics across the ideological spectrum. Sofie Miller of the GW Regulatory Studies Center has found DOE’s reliance on low discount rates may not accurately represent average consumer preferences, especially for low-to-middle-income Americans.[19] For example, Miller notes that a new rule for furnace fans could cause prices to increase by $64 to $154 per unit. Some consumers will keep the fans long enough to realize significant energy savings and generate net benefits, despite the higher upfront purchase price. However, even DOE’s own analysis estimates 30 percent of consumers will experience net costs. This one-size-fits-all approach to regulating efficiency often results in higher burdens for millions of Americans.

One of the biggest questions with DOE rulemakings is the agency’s use of standard discount rates: three and seven percent. These lower discount rates, used by most agencies, tend to benefit higher-income households. As Miller notes, “[T]hey have more certain future streams of income; low-income households, on the other hand, do not benefit from the same certainty.”[20] Using discount rates that better represent low-to-middle-income households can quickly change DOE rulemakings with net benefits into rules with net costs to consumers. Some efficiency rules function as a regressive transfer of income from low-income individuals to high-income households.

Finally, although both DOE and EPA are active regulators, the former rarely receives the same level of media scrutiny. For example, EPA’s “Clean Power Plan” received more than 4.3 million public comments.[21] By contrast, DOE’s most expensive rule of 2015 for fluorescent lamps received just 25 comments.[22] This disparity is largely a function of EPA consuming most of the regulatory oxygen. This does not indicate DOE measures only include unsubstantial burdens or fail to adversely affect consumers. Indeed, since 2006, the agency has imposed more than $174 billion in regulatory costs or $540 in burdens per person.[23] As noted above, many of these costs impose regressive impacts for low-income households. DOE might not be in the news now, but it certainly has the regulatory agenda to warrant heightened coverage.

Possible Reforms

Addressing midnight regulation, as two administrations can attest, is not an easy task. In 2008, OIRA Administrator Susan Dudley noted the “pressures and incentives to complete regulations before the stroke of midnight.”[24] To push agencies to adhere to a moderate pace, White House Chief of Staff Joshua Bolten urged agencies to propose rules no later than June 1, 2008 and finalize them no later than November 1, 2008.[25] Despite these warnings, the 2008 midnight period was still a record time for regulation, with nearly 40 economically significant measures. As the Government Accountability Office (GAO) reports, there was a major rule issued every other day in President Bush’s last month in office.[26]

Today, Howard Shelanski has attempted to curb the practice of midnight regulation as well. In a memo to agencies, “Regulatory Review at the End of the Administration,” he urged: “agencies should strive to complete their highest priority rulemakings by the summer of 2016 to avoid an end-of-year scramble that has the potential to lower the quality of regulations that OIRA receives for review and to tax the resources available for interagency review.”[27] Despite this memo, and the past pleas of Joshua Bolten, there is little penalization for agencies rushing regulation.

Perhaps, Congress or the administration could explore more formal measures to address midnight regulation, as opposed to temporary and ad hoc memos. Congress could propose legislation that would establish enforceable guidelines for end-of-administration rulemakings, similar to the Bolten memo. This legislation wouldn’t end regulation during the midnight period, but it could ensure an orderly process and allow OIRA an opportunity to thoroughly vet new rules. An incoming president could also choose to issue an executive order outlining steps for midnight rules, but enforceable legislation would be preferable to an executive order.

Congress already has pending legislation that could address some of the problems outlined. The ALERT Act (H.R. 1759) would prohibit a rule from taking effect unless it has been posted on the Internet for at least six months. This would address some of the issues with hurried regulation. For example, an efficiency rule for heat pumps during the 2000 midnight quarter was proposed in October of 2000 and finalized by January 2001, a period of just 107 days.[28] Some comment periods last longer than the entire rulemaking history of the heat pumps measure.

If reform fails, the next administration is left with a few drastic options, including a regulatory freeze. President Obama’s Chief of Staff, Rahm Emanuel, implemented a temporary freeze in 2009 in response to midnight rules from the Bush Administration. In a memo, he requested all agencies to refrain from publishing new rules, pullback any rules that had been submitted to the Federal Register, and consider extending the effective date of rules already formally published.[29] This acted as a brief de facto moratorium on regulation. For example, the administration withdrew eleven significant regulations at OIRA in its first week and didn’t approve its first significant rule for more than a month. Again, Congress could choose to formalize these temporary freezes into legislation, which would discourage the practice of rushing regulation.

Finally, if no pending reforms are signed into law, Congress can always address midnight rules through the CRA. The carryover provision allows legislators in 2017 to review rules issued 60 session or legislative days before the end of Congress. If there are any particularly egregious midnight regulations, the CRA allows Congress to address many of them at the start of the next session. However, the CRA has only been used successfully once and it is a blunt policy instrument. If a CRA resolution is signed into law and a rule is rescinded, it may not be reissued in “substantially the same form.”[30] Although there will be dozens of rules eligible for CRA votes next year, it’s unlikely that more than a handful will receive votes.

Conclusion

The nation won’t know if midnight regulation is a pervasive problem yet again until 2017. If too many rulemakings are rushed, there could be legal, economic, and other unintended consequences for the nation. Congress can address the problem of midnight regulation legislatively, either through use of the CRA or a more direct legislative approach that targets expedited rulemakings.

Thank you for your time. I look forward to answering your questions.