The year 2017 has seen the emergence of the Congressional Review Act (CRA) as a powerful political tool. The law allows Congress to fast-track legislation to supersede and cancel rules recently enacted by federal agencies.

Until now, the CRA has largely been an empty threat because legislation under the act requires the approval of both houses of Congress and the White House. Obviously, presidents are unlikely to approve the repeal of a rule that their administration has crafted. But this dynamic changes when control of the White House shifts between parties. Before this year, the CRA had been used just once, in 2001 by a Republican Congress and new George W. Bush administration to dispense with a workplace “ergonomics” rule adopted by the departed Bill Clinton administration. This year, the CRA has become a potent weapon under the new Trump administration and a Republican Congress intent on reversing regulatory initiatives undertaken by the departed Barack Obama administration.

Now that Congress and President Trump are making extensive use of the act, the ramifications of these repeal resolutions become more significant. A CRA resolution does not just nullify an existing rule; it categorically and permanently prohibits the agency from reissuing the rule “in substantially the same form” unless Congress someday passes new legislation authorizing the agency to reissue the regulation. The key question, then, is what it means for a new rule to be “substantially similar” to a CRA-canceled one.

Courts have never authoritatively interpreted this phrase precisely because the CRA has virtually never been used before this year. But with the act’s newfound life, it’s almost certain the federal judiciary will have to do this in the future. When such cases arise, how will courts interpret the act’s terms? Will they simply look to the text of the CRA, the ambiguous “substantially the same form” language, or will they broaden their analysis to the legislative history of the CRA itself? How will current statements from members of Congress factor into the decisionmaking process? Congressmen and senators have already predicted that some of this year’s overturned rules will return in some form in the future.

In this article, we argue that future courts should place predominant weight upon the legislative history surrounding the disapproval of a rule under the CRA. For instance, when Congress recently used the CRA to cancel the Securities and Exchange Commission’s Resource Extraction Rule (discussed below), lawmakers repeatedly suggested that the SEC “go back to the drawing board” and craft a new rule. In contrast, when Congress used the CRA to cancel the Interior Department’s Stream Protection Rule (also discussed below), no supporters of the resolution expressed interest in the agency rewriting the regulation, suggesting that Congress’s purpose in that instance was to cancel the rule based on its general subject matter.

This interpretive approach won’t be dispositive for all rules, especially when legislative debate is lacking (which it has been for several votes). But it does mark the best approach, given the ambiguous text of the CRA.

Brief History of the CRA

The CRA was adopted as part of a package of major regulatory reforms after the “Republican Revolution” following the 1994 elections. In a sign of the relatively limited reach of the CRA and the somewhat more bipartisan atmosphere of the 1990s, the legislation passed 100–0 in the Senate.

The CRA legislation was a response to the Supreme Court’s 1983 decision in INS v. Chadha in which Congress lost its “legislative veto,” the ability to undo executive actions through congressional action alone. In contrast to the procedure struck down in Chadha, however, every CRA resolution must be approved by both the House and Senate, and then signed by the president. In other words, a CRA resolution is just the passage of a law, only with expedited procedures in the House and Senate. According to the primary sponsors, Sens. Harry Reid (D–Nev.), Ted Stevens (R–Alaska), and Don Nickles (R–Okla.), it was designed to address “burdensome, excessive, inappropriate, or duplicative” federal rules.

The scope of the CRA, unlike many other regulatory reform initiatives, applies to all federal agencies, including independent commissions. For instance, actions from the Consumer Financial Protection Bureau and the SEC would be covered under the CRA. The legislation’s reach was intended to be as broad as possible. The drafters of the CRA were explicit on this, saying it covers:

  • formal rulemaking, the relatively rare, trial-type process for formulating some regulations and for rate-setting,
  • informal rulemaking, the much more common notice-and-comment rules found in the Federal Register,
  • other public information from agencies, usually published in the Federal Register, and
  • guidances and other interpretive documents that qualify as “rules” under the Administrative Procedure Act, but that are usually exempt from review under ordinary procedural requirements.

Once an agency finalizes a rule and reports it to Congress, Congress has 60 session or legislative days to consider the rule for repeal under the CRA. Members simply introduce a resolution of disapproval and it cannot be filibustered in the Senate. Although CRA resolutions take up floor time, the current House and Senate have managed to pass more than a dozen measures repealing Obama-era regulations. The question now is how many of those rules could someday return from the grave, in some form or another?

New Life for Old Rules?

As previously noted, the CRA states that a disapproved rule “may not be reissued in substantially the same form.” This language is ambiguous; the term “substantially similar” does not lend itself to an obvious single interpretation.

Fortunately, the lawmakers who originally drafted and enacted the CRA offered some useful guidance, preserved in the legislative history. Whether a subsequent regulation is “substantially similar” to a nullified regulation depends first and foremost on the original statute that authorized the regulation. The drafters of the CRA established a hierarchy of discretion, from broad to most restrictive, relying on the text and grant of authority of the underlying statute. This was set forth in detail by Reid, Stevens, and Nickles in a joint explanatory statement in the Congressional Record.

Three important points emerge from that statement. First, if the law authorizing the disapproved rule provided broad discretion to the agency, then regulators would likely have broad authority to issue a substantially different rule. That information is helpful, but it still does not clarify precisely where to draw the line between substantial differences and minor differences.

Second, if the original law that authorized the initial agency action did not mandate a particular rule, then regulators have discretion “not to issue any new rule.” While this does not go directly to the question of how to interpret “substantially similar,” it does illustrate how the drafters of the CRA place great importance on the intent of Congress when it enacted the statute authorizing the regulation with respect to whether a given agency could return to the regulatory drawing board. If Congress were to strike down the Clean Power Plan under the CRA, for example, then the Environmental Protection Agency could choose not to issue another rule at all because no statute specifically requires the EPA to do so.

That second category is important because a substantial fraction of federal rules do not have specific mandates and their underlying organic statutes are often silent on specifics. For example, the ergonomics rule that was struck down in 2001 was not explicitly authorized in statute. The Department of Labor initiated the rule on its own discretion; 16 years after it was struck down, the Labor Department has not issued another rule on the same subject matter. Apparently, Labor officials believe that any new ergonomics rule would be “substantially the same form” as the CRA-stricken rule and thus likely to be struck down by courts.

A third, and perhaps most important, point from the explanatory statement is that if Congress had been explicit in its authorizing statute and the grant of power to a federal agency was “narrowly circumscribed,” then “the enactment of a resolution of disapproval for that rule may work to prohibit the reissuance of any rule.” This can be interpreted to mean that if, say, Congress says the level of particulate matter in the atmosphere should be limited to 12 micrograms per cubic meter and a CRA measure strikes that down, the agency is prohibited from issuing that standard again.

The senators’ explanation of this third category has direct relevance for the measures Congress has struck down recently. For instance, Section 1504 of the 2010 Dodd–Frank finance legislation mandated that the SEC require resource extraction issuers to disclose payments to foreign governments. That section was explicit about the information that companies had to report. Yet Congress struck down the rule. For the attorneys at the SEC, how do they craft a new rule that follows section 1504 yet is somehow substantially dissimilar from the rule Congress struck down? It’s an unenviable position and one that might be answered by the legislative history when Congress repealed the rule.

In sum, where there is little agency discretion and Congress has expressly delegated certain tasks to an agency, even if the final rule comports with original intent, Congress can change its mind and strike down the rule. How can an agency issue a substantially dissimilar rule while concurrently following the original intent of the statute? The agency likely cannot thread that narrow needle absent new congressional authorization.

Qualitative vs. Quantative Approach

To examine the probability of agency action after a CRA resolution of disapproval, we suggest a “qualitative” versus “quantitative” comparison. By “qualitative” we mean broad grants of authority with no finite or numerical guidelines. This extends to original regulations as well. Such authority would seem to allow agencies to craft new rules, albeit after some time at the “drawing board.”

On the other hand, if agencies are implementing a statute by which Congress originally set quantitative standards, but Congress repeals the rule under the CRA, then it would seem almost impossible for an agency to promulgate a substantially different rule. If the EPA sets ground-level ozone concentrations at 70 parts per million and Congress strikes down those standards, what is a substantially different rule? Is 65? Is 75? Quantitative or numerical prescriptions in statutes or in the text of a regulation would seem to make crafting a new rule after a CRA action nearly impossible; Congress’s decision to cancel the agency’s rule would seem to effectively rescind the original statutory delegation of regulatory power altogether.

Below, we offer a few contemporaneous and hypothetical examples to illustrate this quantitative and qualitative dichotomy.

Resource Extraction Rule / This rule generally required specific disclosures from companies dealing with foreign governments. Dodd–Frank set no numerical standard, but it did include specific instructions for the information to be disclosed. For instance, section 1504 of Dodd-Frank mandated that an annual report contain the type and total amount of payments to foreign governments or corporations. This appears straightforward, but given the rulemaking history it is not. Generally, the resource extraction rule is more of a qualitative example with some caveats.

A federal court struck down the first iteration of the SEC’s resource extraction rule, so the agency issued a new rule in 2016, within the window for disapproval under the CRA. Congress acted and struck down the measure. After losing in the courts and failing with Congress and the president, many might assume this regulation is finished. However, the legislative history surrounding the CRA resolution reveals that even Republicans voting for the CRA resolution presumed that their act would not prevent the SEC from promulgating a different resource extraction rule someday. House Financial Services Committee Chairman Jeb Hensarling (R–Texas) remarked: “Let’s also remember that this joint resolution does not repeal section 1504 of Dodd–Frank. I wish it did, but it doesn’t…. It simply tells the SEC to go back to the drawing board.” Likewise, Sen. Mike Crapo (R–Idaho) explicitly noted that he expected yet another rule: “What the SEC will need to do is to go back to the drawing board and come up with a better rule that complies with the law of the land.”

Whether the SEC does actually draw up a new rule will be instructive, but Congress was hardly specific on the changes or the form of the third iteration of the resource extraction rule that lawmakers want to see. Ultimately, the discretion still lies with the regulators at the SEC to determine if and how a new rule will develop.

For those hoping it never does, the SEC has technically already fulfilled its statutory obligation. It did issue a final rule and then a court struck it down. It then issued another regulation and the elected leaders of the nation rescinded it. The SEC might defer to the other branches of government and acknowledge that creating a substantially different rule would be difficult and would almost assuredly lead to another lawsuit. If regulators do proceed, we’ll at least have one interpretation of “substantially the same form” in the next two to three years.

Stream Protection Rule / In contrast to Congress’s view of the resource extraction rule, congressional majorities clearly had no interest in preserving the Department of Interior’s authority to promulgate a new “Stream Protection Rule.” Interior’s regulation was ostensibly designed to prevent dumping coal deposits into waterways, a largely qualitative and discretionary rulemaking.

The rule united environmentalists while also acknowledging it would adversely affect coal industry employment and have a social cost of more than $1 billion. Given the political forces involved, as soon as the regulation was published, Senate Majority Leader Mitch McConnell (R–Ky.) declared he would use the CRA to undo the regulation. He noted the rule was “part of the [Obama] administration’s plan to demolish these coal communities right now and long after the president has left office.”

Presented with an opportunity to use the CRA to nullify this rule, majorities in both houses—including some Democrats—overturned the rule. And the commentary accompanying their votes contrasted sharply with the aforementioned commentary accompanying the Resource Extraction Rule’s CRA resolution. As to the Stream Protection Rule, Rep. Paul Gosar (R–Ariz.) remarked: “But the fact is, you can’t put lipstick on this pig. Whether you call it the Stream Buffer Zone Rule or the Stream Protection Rule, the rule still stinks.” Rep. Rob Bishop (R–Utah) was even more frank, recognizing that Congress should go forward and set specific standards for stream protection, not let federal agencies write new rules. He noted, “What we are doing here today with this effort is to reestablish the Article I authority that we have in the Constitution by saying we are responsible for the policy, not some agency of the executive branch.”

Democrats were also aware of the power of the CRA, predicting the Stream Protection Rule would not return. Sen. Chris Van Hollen (D–Md.) charged: “The Congressional Review Act doesn’t make sense here. If you want to trim a tree, you don’t chop it down and bury it under cement so it will never grow again.” He also predicted the permanent end of the rule, remarking: “[A] resolution of disapproval under the Congressional Review Act does not just send a rule back to the drawing board. Instead, the resolution repeals the rule and prohibits the agency from ever proposing anything like it again.”

In contrast to the resource extraction measure, it is clear Republicans don’t want the Stream Protection Rule returning and Democrats don’t expect it to return. Given the congressional debate, a redo of the rule, at minimum, would likely have to demonstrate insignificant costs and neutral economic effects to the coal industry.

Clean Power Plan / The Clean Power Plan (CPP) mandates greenhouse gas standards for existing power plants, but the deadline for a CRA resolution repealing the CPP had already expired when President Trump took office. Still, the CPP offers an interesting interpretive hypothetical.

The CPP mandates a 32% reduction in the emission of greenhouse gases by 2030, brought about by using three “building blocks”: heat-rate improvements at coal plants, a shift in electricity generation from coal to natural gas, and a build-out of zero-carbon energy. If Congress could use the CRA to nullify the CPP, then what would a “substantially” different CPP look like? Is a 28% cut sufficiently dissimilar? What about a 35% reduction? Suppose the next president sets a goal of reducing emissions by 80%; would that count as substantially different? Here, one might be able to reasonably argue that a 3% or 80% reduction would not be substantially similar to the original CPP, and thus not prohibited by the CRA.

Or suppose that current EPA Administrator Scott Pruitt reissues the CPP, but includes only one building block: limited heat rate improvements for some aging coal plants, reducing emissions by 2–3% but costing a mere fraction of the CPP’s original $8.4 billion estimated burden. Could the EPA reasonably argue that this new rule is substantially dissimilar from the prior regulation?

In this context, the EPA’s argument would be bolstered by the broad discretion that the Clean Air Act vested in the agency in the first place. Congress never spoke explicitly about greenhouse gases in the Clean Air Act, so the EPA could be just as creative in reissuing a new rule. Even if a new CPP is terrible policy, it might be a terrible policy that the CRA does not prohibit.

Overtime Rule / In 2017, the Labor Department greatly expanded the overtime threshold for non-exempt employees, from those earning up to $23,000 annually to those earning up to $47,000. This rule fell just days outside of the CRA review window, but it offers a more restrictive quantitative example compared to the CPP.

Assume that Congress could have used the CRA to repeal the Overtime Rule. Indeed, before Congress adjourned in 2016, members and staffers openly discussed repealing the rule; many staffers suggested that they wanted new legislation establishing overtime thresholds, not a new rule. To them, a CRA vote would have meant no new expansions from the agency.

Like the other quantitative examples above, imagine a new administration attempts to resurrect the Overtime Rule after a CRA repeal. Is a $40,000 employee earnings limit substantially similar, or $55,000? Suppose it’s two decades until a new rule is considered and it sets the threshold at $75,000. Perhaps the agency could argue that rule would be dissimilar. However, adjusting for inflation the threshold would still be around $47,000 in 2017 dollars.

There is hardly a knowable legal path to regulating again after Congress strikes down a rule with specific, quantifiable standards, whether in the regulation or from the authorizing statute. In practice, such a resolution would act as an absolute bar to a new rule, save new delegation from Congress.

Endangerment Finding for Aircraft / The final example involves a rule or determination from the EPA that aircraft emissions contribute to global climate change and endanger public health. Many legal experts consider the endangerment finding a necessary condition for regulating specific greenhouse gas emissions from aircraft under the Clean Air Act. There was a reason, after all, that the Obama administration chose to begin with similar endangerment findings and then followed with emissions regulations for vehicles and existing power plants.

What would have happened if Congress had used the CRA to disapprove and nullify the Aircraft Endangerment Finding? If it is a legal prerequisite to regulation, how could the EPA have conceived of a substantially different regulation? In cases like these, it is binary: either the EPA views these emissions as a danger to public health or it does not. With a CRA vote, Congress would have said the agency can’t reissue a rule that determines the emissions endanger public health. Without a new finding, there is no subsequent regulation on aircraft. In the future, these sorts of binary, authority-or-no-authority rules might prove attractive to members of Congress who want to preclude specific rules from ever returning.

What Did Congress Reveal?

As the debates display, there was mixed opinion on whether all of the repealed rules would return in some form or if new versions were precluded by the CRA vote. Given that public remarks might not always mirror personal feelings, we reached out to several House and Senate staffers on the condition of anonymity to gauge their intentions and instincts on whether various disapproved rules would return.

One Senate aide remarked, “A majority voted for [repealing] each regulation, so it’s fair to say they don’t want to see the rule.” However, he acknowledged, “In the next five to 10 years, the Federal Communications Commission rule [on consumer privacy] seems to have generated a lot of opposition, so an activist FCC head could be pretty clever coming back to that in a way that is plausibly not the ‘same form.’ ”

There seems to be universal recognition that the FCC’s privacy rules could return, although all agreed that would have to be decided by the courts. Another aide noted, “I think Congress will attempt to act on one this year or next through legislation that makes it clear [internet service providers] are subject to privacy rules, but that the [Federal Trade Commission] (not the FCC) is the privacy regulator and it generally treats ISPs and edge providers like Google and Facebook the same.” Another aide noted that his boss didn’t want to see any new rules regardless of the form. If there were to be new regulation in these fields, Congress would act with new legislation.

Finally, the press has already begun speculating on the motives for repealing each rule and which ones might return. The Huffington Post focused on the rescission of drug testing requirements for unemployment applicants. The conventional thinking is that the Trump administration will write even tougher standards for more expansive testing than the Obama-era rule. That would mark the first chance to determine a substantially dissimilar rule. Rep. Kevin Brady (R–Texas), who sponsored the resolution to repeal the drug testing rule, acknowledged, “My understanding is that they will promulgate a new rule.” It will then fall to the Trump administration and perhaps the courts to determine what “substantially the same form” means in the context of reissued rules.

From Congress to the Courts

Of course, the CRA will not interpret itself. In the end, the federal courts will determine the act’s capacious terms, once an appropriate case presents itself. Such cases might take a variety of forms.

First, pro-regulatory activists might try to bring a constitutional lawsuit challenging the CRA itself. Specifically, they might seek out sympathetic persons, companies, or other entities who would have benefited from a rule that Congress nullified with a CRA resolution and attempt to convince judges that the CRA itself violates the Constitution. (One progressive organization, the Center for Biological Diversity, already has filed a federal lawsuit in Alaska arguing that Congress and the president are constitutionally prohibited from using the CRA to impede an Interior Department regulation.) While a direct constitutional challenge would aim primarily to nullify the CRA, not interpret it, such litigation could theoretically urge the court to adopt a “narrowing construction” of the CRA’s terms to limit the case’s constitutional stakes.

In any event, this approach faces several jurisdictional and substantive impediments. First, plaintiffs attempting to bring such a constitutional challenge would need to show that they have standing to bring the suit, including a showing that a signed CRA resolution injures them in a way that can be redressed by the courts. Moreover, courts generally decline to hear cases in which plaintiffs desire simply to increase regulatory burdens on third parties. Second, even if the plaintiffs manage to leap these significant jurisdictional hurdles, they still would need to formulate a compelling substantive argument that the CRA is not just bad policy but unconstitutional, and so far the CRA’s critics have failed to find such a silver bullet. (The Center for Biological Diversity, for example, argues that the CRA should be struck down because of INS v. Chadha, but the CRA contains precisely what Chadha’s “one-house veto” lacked: a resolution approved by both houses of Congress and signed by the president.)

Instead of a direct constitutional lawsuit, the more likely scenario is one in which parties challenge an agency’s action or inaction in the aftermath of a CRA resolution. That is, if Congress passes a CRA resolution nullifying a regulation and the relevant agency acquiescently declines to resurrect the rule, a pro-regulatory party might sue to try to force the agency to act. Or, if an agency attempts to resurrect a CRA-stricken regulation, perhaps after a change in presidential administrations, then a party might file a lawsuit to block the agency from implementing the new regulation.

A challenge to agency inaction would, like a direct constitutional challenge, face significant jurisdictional and substantive hurdles. Agencies generally retain immense discretion not to undertake a rulemaking process. The courts are generally loath to force agencies to regulate, just as courts are generally loath to force prosecutors to prosecute; thus, agencies possess something akin to “prosecutorial discretion.”

But there are two important exceptions that the CRA’s critics might seize upon. First, if a substantive statute like the Clean Air Act appears to require the agency to promulgate a regulation, then courts are more likely to entertain a lawsuit forcing the agency’s hand. Second, if an agency announces that it cannot undertake a new rulemaking solely because a prior CRA resolution has stripped the agency of its authority on the subject, then a party could file a lawsuit challenging the agency’s position that it lacks regulatory authority. In such a case, a court could conclude that the agency has not lost its regulatory authority—because new rules requested by the challengers would not be “substantially” similar to the CRA-repealed rule—and direct the agency to consider the merits of a new regulation. This would be similar to what the Supreme Court required of the EPA in Massachusetts v. EPA, where the EPA disclaimed authority to regulate greenhouse gas emissions as air “pollutants.”

To avoid that latter option, an agency would be wise to cite two grounds for declining to undertake a post-CRA regulation: namely, that the CRA resolution stripped the agency of its authority, and that the agency would exercise discretion not to promulgate the new regulation even if the CRA did not prevent it. This might prevent challengers from suing over agency inaction, at least so long as the relevant underlying substantive statute does not require the agency to regulate.

If a lawsuit challenging agency inaction is an unlikely route to adjudicating the meaning of the CRA’s “substantially the same form” provision, then a lawsuit challenging agency action is the most likely path. That is, if an agency—perhaps after another presidential election—attempts to promulgate a regulation relating to a matter that had been the subject of a previous CRA-nullified regulation, then parties affected by the new regulation will almost certainly file a lawsuit against the agency, arguing that the new regulation is unlawful under the CRA. In such a case, the court would squarely confront the question of whether the new regulation is of “substantially the same form” as the old CRA-nullified regulation.

No matter how the issue reaches the court—be it a lawsuit challenging agency action or agency inaction—the court’s analysis would not involve application of the increasingly controversial doctrine of “Chevron deference.” Normally a court defers to an agency’s interpretation of a statute’s ambiguous terms—and the CRA’s “substantially the same form” provision certainly seems ambiguous. But as the courts have stressed, Chevron deference is appropriate only when the statute at issue is one that has been committed to that particular agency’s exclusive administration—e.g., provisions of the Clean Air Act, which are administered exclusively by the EPA. Thus, when an agency is administered by multiple agencies—e.g., the Freedom of Information Act or the Administrative Procedure Act—then no agency receives Chevron deference in interpreting it. The CRA falls neatly into this latter category, so the courts would likely interpret the CRA’s terms undeferentially (or, in legal terms, “de novo”).

But to be clear, the court’s specific task would be to interpret the meaning of the CRA’s terms and not the meaning of a CRA resolution, though the two are deeply intertwined. To interpret the CRA’s “substantially the same form” provision, the court would apply the traditional tools of statutory construction: the meaning of the CRA’s words, both as those words are normally understood and as they might be understood in this particular legal or legislative context; the CRA’s broader structure; the CRA’s legislative history; the broader set of presumptions (or “canons”) of statutory construction that impute congressional intent in one interpretive direction or another; and so on. Taken together, the court would be attempting to ascertain Congress’s “intent,” though particular judges vary widely in the extent to which they are willing to look beyond the statutory text’s plain meaning for indications of congressional intent.

Conclusion

As we have sought to make clear in this article, the CRA is best understood as embodying Congress’s understanding that proper application of the “substantially the same form” provision will turn, in each case, on both the substance of the statute that Congress originally enacted authorizing the regulation, and Congress’s subsequent purpose in passing the CRA resolution. In other words, the Congress that enacted the CRA intended for courts and agencies to pay attention to the intentions of the subsequent Congresses that use the CRA.

But this is a difficult standard for any court to apply, especially for judges who are wary of ascribing such flexible meaning to any statute. Ironically, the judges who are ordinarily the most skeptical of agencies might also be the most skeptical of interpreting the CRA flexibly to limit future agencies. And the judges who ordinarily treat statutes more malleably might be the ones most likely to construe the CRA strictly—but in service of preserving the agencies’ discretion and power in the future.