Say, for instance, EPA takes up a statutory duty to create a rule to reduce a certain air pollutant, but also must take into account the varying costs of doing so.  After taking in and responding to the legally required public comments, EPA staff meet with some of the President’s staff to talk about the different options for the rule.  Anything wrong if those conversations are fully off the record?  What if the President’s staff tells EPA just how it needs to analyze the options, and that, whatever the final rule, its benefits must outweigh its costs?

By now, we’re used to the White House’s Office of Management and Budget/OIRA review, and arguments against cost benefit analysis or centralized review tend to be huffed-out, hyperbolic, normative statements against politically watered down regulations.  It is far more interesting, in my mind, to read arguments proffered shortly after Reagan’s Executive Order 12291 (adopted by Clinton as the more frequently cited 12866) that set up the now-routine notion of OMB as regulatory gatekeeper.

The DC Circuit in 1981 ruled that folks from OMB could indeed talk with folks at EPA off the record after the comment period closed.  The case, Sierra Club  v.  Costle, largely stands for that proposition today; but, it is fun to read the commentary it inspired in its day.  So, I commend the article in Cato/AEI’s Regulation magazine, “Regulatory Oversight Wins in Court.”

The supporters seem to think the rulemaking process is a part of process of making the statute in the first place; so it makes perfect sense that Congress should sign off on the rules promulgated pursuant to its own initiatives. Supporters also assume the elected representatives’ quick votes on the rules provide a measure of democratic accountability. They are generally skeptical of the competence and accountability of bureaucrats.

Objectors seem to think rulemaking is a function by which Presidents impose policy and assert power as a useful opposing branch to Congress. Objectors assume voters can hold agencies accountable every four years during the Presidential elections. They are generally skeptical of the political motivations and monetary capture of Congresspeople.

I haven’t seen as much discussion on what REINS means for the process of governing, and the values we attach to the various actors within government. So, below are a few questions.

Should a current Congress be able to prevent the promulgation of statutes passed by a prior Congress, without actually repealing the statute?

Say the GOP won both houses in 2010 with a veto proof majority and promptly passed laws requiring the Occupational Safety & Health Administration to revise its regulations to prevent only the workplace hazards causing “severe or frequent injuries.” OSHA works on the new rule for a few years, researching the severity and frequency of each occupation’s injuries, and finally produces the rule to Congress in 2013. Meanwhile, Democrats swept back into control of the House in 2012. The Democrats don’t have the votes to repeal the 2010 “NOSHA Act,” but when presented with OSHA’s rule, reject it by resolution. And they do so on every revised rule.

Is it desirable to allow a representative to vote in favor of a popular bill, but against its implementation?

Obstruction by resolution might not be by a later Congress against its predecessor. As I mentioned in a prior post, a representative might vote for the “Everyone Likes it in Theory” Act, but against the “Actually Putting it into Practice” regulation. REINS, then, might afford our elected officials another tool in the trickery of campaign ads.

Certainly it is possible that a representative will sincerely believe an agency got something wrong in its rule, and want to send it back for revision. That presents its own danger–the sometimes endlessness of noodling in minutia. Until now, we’ve left it for agencies to do the fine tuning, which takes years. REINS invites politicians into that process.

What does it mean to interpret a vague piece of legislation; when agencies add the necessarily tremendously detailed rules to statutes, are they in fact legislating or implementing existing legislation?

On one end of the spectrum, if a court believes that an agency actually changed a statute through rulemaking, the rule will be overturned. On the other end, a rule carrying out a specific statutory directive will stand.

In between are those rules that inspire the most written about doctrine in administrative law, Chevron, in which the statute was a little fuzzy and the agency decided on a particular interpretation.

Or rules that apply expertise where Congress asked for such expertise: like, Congress instructing the EPA Administrator to prescribe emission standards for air pollutants “which in his judgment cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare.” 42 U.S.C. 7521(a)(1).

Is a rule an executive or legislative function?
Writers commonly describe agencies as performing quasi-legislative (rulemaking) and quasi-judicial (enforcing, adjudicating) tasks. I’ve wondered whether it’s appropriate to allow the “quasi-legislative” description to place agencies within the legislative branch in a separation of powers argument. Indeed, in my mind, rulemaking is neither an executive nor legislative function.

Rulemaking is simply an agency’s placing into executable rules already existing legislation. If an agency changes the legislation in the process of making a rule, the rule is invalid.

Execution, I think, is better left to those activities that enforce rules in effect. Note, of course, that a great deal of interpretation (guidance, decisions on when to enforce, allocation of resources) goes on in the act of enforcing.

Rulemaking, though, involves an effort to take a law and apply a framework with which it will apply to the real world. The idea has long been that Congress is institutionally unable to prescribe every detailed rule, so it delegates to experts that last step, teeing up one more question for now:

What is the best structure for, and by what process can we assure, an appropriate balance of expertise and accountability in the final rules governing our day-today lives?

The Constitution failed to provide a framework for the administrative state, even though (thanks to Professor Mashaw we know that) the framers should have seen it coming. Thus, statutory law (the APA) provides our structure, and that is what REINS aims to alter.

Far more than the canards of jobs, red tape, or the benefit of having regulations generally, the discussion REINS should be inspiring is of the basic processes of lawmaking and rulemaking. Will better (whether your opinion of “better” means fewer, more, or more effective) regulations result from providing Congress an up or down vote on every promulgated rule?

Will that process add significantly to the time it takes to put any given rule into effect, and is that good or bad? Will that process push agency rulemaking staff to work with Congressional staff and lobbyists far more while drafting rules, and is that good? Will they pay more attention to politics and less to economists and scientists? Will the need to pass Congressional approval become a response to public comments?

I tend to think REINS allows for political cover and massive regualtory delay. I doubt it will ever make the President’s desk, and if it does it’ll be vetoed. However, in another time, if such a change indeed comes along, I will dream of a world in which voters pay attention to how their representatives vote (on both the bill and the rule); in which every representative has the philosophical capacity to vote for or against general principles and the technocratic capacity to vote for or against the subsequent rule; and in which every representative can speed read sufficiently to fully understand and give a fair assessment of a rule within 15 to 30 days.

Finally, a correction: when I first posted on REINS I’d only read Section 1. REINS requires both houses of Congress to approve by joint resolution any new major rule. I thought a vote wasn’t required, which would have problematically allowed Congress to kill rules through inaction. That’s not the case, but the actual provisions, which do require an up or down vote, pose some problems.

The process is roughly this, for major rules: an agency submits to Congress its rule; the majority leaders of the House and Senate introduce resolutions accepting the rule, and then pass it on to the relevant committee; that committee then has 15 days to allow the joint resolution to stand, or propose amendments to the underlying statute; the joint resolution then goes tot he calendar for an up or down vote that must happen within 15 session days, with debate limited to 2 hours.

Jonathan Adler praises REINS’ expedited review and mandatory vote, but it is a strange comfort. Agencies take several years to develop major rules, and Congress is to vote on the thing in about 30 days. If one house plays more safe than sorry, and rejects the rule, it is back to the perdurable drawing board.

Several headlines this past week provide glowing examples of the latter lesson.  The CBO issued its latest statutorily required report giving an ongoing and often changing economic analysis of the American Recovery and Reinvestment Act of 2009′s affect on outputs and employment.

According to the Washington Times and Investor’s Business Daily, the news is this, respectively: “CBO: Stimulus hurts economy in the long run” and “The CBO Quietly Downgrades Obama’s $825 Bil Stimulus.”

According to Jay Bookman at the Atlanta Journal Constitution, and Daily Kos blogger, the news is this, respectively: “CBO reports stimulus package was a major economic success” and “CBO Gives Thanks to the Stimulus.”

The Washington Times article offers the best example of lesson #1:

The Congressional Budget Office on Tuesday downgraded its estimate of the benefits of President Obama’s 2009 stimulus package, saying it may have sustained as few as 700,000 jobs at its peak last year and that over the long run it will actually be a net drag on the economy.

I suppose the lesson for the thoughtful audience of political speak is to raise a yellow card at any reference to “as few as” or any other variant suggesting a range.  Here, the Washington Times used the low end of the range of employment created, on average, during 2010 – the range was between 700,000 and 3,300,000 people.    (Another lesson for the political rhetorician facing a thoughtful crowd: aim to use the ‘opposing end’ of an estimate, and use it as a concession proving the point – like, ‘even if so and so’s best numbers come true, we’d still have such and such problem.’)

Here, anyway, is how the CBO director summarized the 2011 third quarter:

Using such analysis, CBO estimates that ARRA’s policies had the following effects in the third quarter of calendar year 2011 compared with what would have occurred otherwise:

• They raised real (inflation-adjusted) gross domestic product (GDP) by between 0.3 percent and 1.9 percent,
• They lowered the unemployment rate by between 0.2 percentage points and 1.3 percentage  points,
• They increased the number of people employed by between 0.4 million and 2.4 million, and
• They increased the number of full-time-equivalent (FTE) jobs by 0.5 million to 3.3 million. (Increases in FTE jobs include shifts from part-time to full-time work or overtime and are thus generally larger than increases in the number of employed workers.)

And here’s the full report – it’s not long.

Finally, I need to dispose of a lingering aftertaste left from my local paper, which today wrote:

 In many cases, that is, the stimulus may not have “created” jobs so much as shuffled them around.

This is backed up by research done by the Mercatus Center at George Mason University in Northern Virginia. Researchers there found that in three out of five cases, businesses receiving ARRA money hired people away from other jobs, rather than giving jobs to the unemployed.

When a person leaves a job, in many instances the left-behind job becomes vacant and refilled.  Thus, an employed person leaving a job for another job still creates a job opening.  That’s common sense, but Jon Chait nicely spelled it out a few months back.  The editors at the Times Dispatch, predictably not readers of Chait, are sadly also not patrons of common sense.

Most commentary I’ve read argues that the commerce clause does/does not provide authority to Congress to implement the individual mandate.  The tax and spending clause tends to get mentioned as the mandate being a type of tax.

I haven’t seen much debate on what would seem an easier route to Congressional authority: attaching the mandate a conditional string to health-related funding to the states.  Is that out there and I’ve just missed it?

A policy in which we choose not to have fully private health care and insurance–in other words, one in which we have chosen not to simply let those die that cannot pay for coverage–requires some government spending.  States and the federal government do most of that spending.  I suppose it could be possible for the federal government to get out of the game, and allow that basically moral decision to occur within each state–do we, as a state, wish to create programs like mandatory emergency room coverage, social security, medicare, and medicaid to help prevent death by empty pocket? For now, though, the federal government has a major role in that prevention.  Why can’t it attach conditions to its health-related spending just as the conditions of the drinking age attach to highway funds?

The condition couldn’t be the individual mandate – but could require states accepting funds to implement something that achieves the same goal, which in my mind is getting rid of the incentive to not getting insurance until sick.

(1) In a nutshell, the net neutrality rules forbid internet providers from favoring some content providers over others.  So, for instance, Verizon can’t take a bunch of money from Disney and send along blazing fast ABC programming to your monitor while non-Verizon-paying video producers get throttled.  Rather, internet providers must blindly provide equal access to the highway without special lanes for the paying preferred.

(2) It’s been widely discussed lately that resistance to “control” is at the heart of both the Tea Party and Occupy Wall Street movements, the former concerned with government control and the latter concerned with corporate control.

(3) Hutchison asserts that prohibiting companies from making deals with each other (resulting in faster or slower websites for those of us downstream from the deal-making) is harmful government control.  It is doubtlessly correct that the FCC is exercising some control.  But her comment is in a contextually, and horribly incorrect tunnel of vision looking only at the sphere of companies making deals.

(4) The rules, pretty clearly, do much more to keep control from happening than to allow control.  Net neutrality is basically a mandatory Autobahn.

(5) Thus, Hutchison very ably demonstrated the problem of rhetoric in the dearth of context.  In this case, not many folks were fooled.

Meanwhile, jobs go begging: in Alabama, which passed the nation’s harshest anti-immigrant law; in Georgia, where the governor suggested using convicts to work in the fields after 11,000 jobs went unfilled; and in the orchards of Washington, where the flow to the far north has diminished mainly because of the recession.

Well then, why not hire only people with full citizenship? One farmer in Colorado, John Harold, tried doing just that, hoping to fill harvest positions with jobless locals looking for extra cash. But as my colleague Kirk Johnson reported, many of those locals did not last even a full day; they complained of the hard work in the onion fields of Colorado.

The problem, through good times and bad, is that there are millions of jobs that Americans will not do. The solution, some combination of path to citizenship with guest worker programs, should be within the grasp of the better political minds.

A reasonable conclusion from this premise is that we should allow immigrant to perform work that, if not by law than by some social or physical prohibition, consists of conditions too treacherous for American citizens. Rather than addressing those conditions, we should let these folks seeking citizenship do it. So, before laws improved the conditions of factory jobs from, say, what we read about in Sinclaire’s The Jungle, it would have been appropriate to allow Russian folks to die in those factories that were too treacherous for Americans.

I cannot accept that conclusion.  Rather, there are only two morally appropriate conclusions that, in turn, allow fairly straightforward responses:

1) The work is too treacherous for anyone to perform.

Thus: The farm work should be better regulated.  If work is not safe enough for Americans it is not morally appropriate to benefit from such work from non-Americans, and it is especially inappropriate to allow Americans to facilitate such work.

2) American citizens are too weak or snobbish to perform perfectly acceptable and available work.

Thus: Social norms need to adjust toward acceptance and appreciation of farm work.  If people seek unemployment benefits where farm work is available, they should be required to accept the work or not accept benefits if that work is otherwise within whatever parameters the unemployment folks set out.

As for immigration policy, it should be reasonable and be enforced.  So, regarding those not already in the country, we should debate the policy for entrance.

One last premise with which I disagree: the conflation of immigration and deportation policies.

Immigration policy is different than deportation policy.  So we should also have a clear headed deportation debate.  My proposal: folks that are are fairly established in the country should be allowed to seek legal citizenship if willing to do so.  In that respect, I agree with Timothy Egan’s conclusion while disagreeing with the premise.

The non-codified meaning of a “job creator” is any person that creates a job.  Of course, jobs are not created by people they are created by demand.  People certainly make decisions regarding who to hire and how much to pay them.  But without a duty to perform, there is no job to be done.  Owners and hiring managers can have as money available to them as they want; they will not hire unless there is unmet demand.  The real creation, then, occurs when anything happens that creates the need for paid work.   Restaurant patrons, book buyers, students, litigants, and patients are job creators.

To be sure, people making over $250,000 create jobs because they, too, usually have to pay for goods and services.  But, then, so is everyone else a job creator.   The person making $0 is a job creator when he turns up in an emergency room and gets treated (there’s the job) by a physician that is then paid with tax money (there’re some more jobs – tax collectors, fund administrators, and hospital administrators).

In any event, I searched the news for some quotes to back up the first paragraph’s claim.  Turns out there’re quite a few folks out there irked with the “job creator” talking point.  Some just want to point out that lower tax rates for top income-earners doesn’t create jobs.  Fair enough on the economic point.

But, more to the accuracy of “job creators,”  Robert Friedman wrote a column in The Hill touting the job creation of new businesses:

“Tax cuts for job creators!” It is a rallying cry echoing these days from both ends of Pennsylvania Avenue. For Republicans in Congress it means never raising taxes on the wealthiest 2 percent of the population. The White House, meanwhile, is considering a general reduction in payroll taxes for all.

Both scenarios, however, miss the real job creators: new businesses under one year old and typically unincorporated, which have added an average of 3 million net new jobs a year to the American economy. That’s more than all other categories of business combined, according to recent studies by the National Bureau of Economic Research and the Ewing Marion Kauffman Foundation.

And Mary Sanchez at the Kansas City Star argued that real job creators are those in the middle class:

This middle class is a vast middle tier of those who work to live, and strive to work a little harder to get a little more in life. Middle class people may save, but they don’t accumulate enough wealth to live off. Almost every buck they get, they spend.

That point matters: Spending creates jobs. In our economy, middle class consumers are the real job creators. Depress their income, and you depress employment.

We’ll never get around to holding politicians truly accountable unless this fuzzy middle demographic — a massive one as a potential voting bloc — gets wise about where it came from in the first place, and how it foundered.

The great prosperity of the American middle class in the late 20th century didn’t just magically transpire. The important groundwork was laid by the federal government via investment. Consider what the creation of the federal highway system did for developers and builders who created our suburban communities and all of the businesses that followed. Or the impact of the GI Bill on so many people who returned to the workforce after World War II.

…Much today is made of the massive federal deficit. I have a way we can solve that: more jobs. More jobs mean more growth, more tax revenue. But America’s job creators — middle class consumers — are tapped out. Business owners can’t hire until they have consumers to sell to. That leaves the job of stimulating demand to the government. Time for government to lay the groundwork for our future by investing in our middle class.

These arguments suffer from the same error in the GOP talking point–identifying one segment of the economy as the real job creators.  Friedman’s new businesses are obviously the part of business that will create a bunch of jobs, being that they are new and in need of staff.  And, just a guess here–new business start, fail, start up again, and fail more often than old business; so more new businesses exist each year creating jobs that may or may not last.  But while businesses, whatever the newness, literally do the hiring, the jobs are not there without demand.

Ms. Sanchez gets to the point that demand within the economic system is the trigger to job creation.  But it is a disservice to clarity to begin the analysis of how to stimulate jobs with a favored segment of society already in mind.  The actions and decisions of rich people, middle income people, or low income people may have varying and more or less important roles in the sprouting up of a new job to be done; starting the discussion with one segment of income in mind, though, distorts one’s holistic economic thinking.  Demand often stems from folks in the middle class buying things.  But, as mentioned above, it also stems from rich and poor people getting sick, and the people at the end of the causal chain of demand might not do the paying.

The acronym for the Regulations from the Executive in Need of Scrutiny Act highlights, as conservative  sources celebrate, the bill’s intent to rein in federal regulations.  In a nutshell, it allows Congress to do through inaction what the Congressional Review Act–part of the mid-90s Contract with America–already allows Congress to do with legislative action: to disqualify major, new regulations.  Under the REINS Act, if an agency proposed a rule that had an estimated economic impact of more than $100 million (this happens about 50-100 times each year), both chambers in Congress would be required to approve of the final rule.  If Congress didn’t pass a joint resolution within 70 days in the legislative session, the proposed rule would die.   The Congressional Review Act already allows Congress to disqualify an agency’s rule, but both chambers have to vote affirmatively to do so.  Thus, under current law, Congress must pass a joint resolution to kill a regulation; under the proposed law, it must pass a joint resolution to affirm a regulation.

The politics of the bill are less interesting than the more fundamental concerns with administrative law, so we’ll quickly dispel the former.  The bill would lower the effective rate of (REIN in) major federal rules.  Some people like that and others don’t.

A more sophisticated issue, segue-ing us from politics to administrative theory, is accountability.  A blessing and curse of our fleet of civil servants is that they are purportedly experts, insulated from the fickle tides of public opinion; at the same time, is is difficult to ascribe accountability to the insulated rulemakers.  The conundrum of accountability versus expertise has existed probably as long as representation; but has at least been well pondered since Woodrow Wilson zeroed in on it in 1886.

Supporting insulated expertise, here is Noah Sachs:

Since the Progressive era, U.S. administrative law has operated from the premise that agency action should be somewhat insulated from political pressure and horse trading. The REINS Act would mark a radical abandonment of that goal, an attempt to correct an oversight problem that doesn’t even exist. It would deliver a body blow to the already-sluggish agency rulemaking process by politicizing it and entangling it in the congressional morass. And, over the long term, it would do serious damage to American health and prosperity—stopping agencies from promulgating important rules that, among other things, would help prevent bank failures, ensure the safety of the food we eat, and control toxic pollution in the air we breathe.

The results would likely be devastating. In the near term, the REINS Act could be a back-door means of gutting health care reform. The GOP lacked the votes in the Senate to repeal the Patient Protection and Affordable Care Act, but, under the REINS Act, it could do serious damage to the statute. The law has more than 40 different provisions that call on the Department of Health and Human Services (HHS) to enact implementing regulations. These forthcoming rules, most of which will be considered “major,” will cover issues such as prevention of Medicare fraud and extending dependent coverage to people as old as 26. With the REINS Act in effect, they could be quashed if the House objects to them, or if Republicans simply stall a floor vote on them beyond 70 days.

And supporting accountability, here is Jonathan Adler:

The primary purpose of the Act is to ensure greater political accountability for major regulatory initiatives.  Federal regulatory agencies only have that power delegated them by Congress, but regulatory agencies are not always particularly responsive to Congressional concerns.  Nor are members of Congress always willing to take responsibility for how the power they have delegated gets exercised.  Requiring a straight up-or-down vote on new major regulations is a way to address both problems and the expedited procedures ensure that traditional legislative logjams and special interest obstruction won’t prevent consideration of significant regulatory initiatives.  This is why I believe the REINS Act is more about transparency and political accountability than anything else.

Adler’s support of the bill, trumpeting transparency and accountability, might sound more coherent if the Congressional Review Act did not already exist.  But, it is difficult to see how a bill that allows Congress to quietly overturn proposed rules by not voting is more transparent or accountable than existing law that allows Congress to overturn a rule by voting.  The argument can only be on this question: which scheme promotes greater transparency and accountability–one in which X must reject Y; or one in which X must affirm Y, and is deemed to reject Y if X does not act.  I don’t think it’s impossible to argue that requiring affirmation is as accountability-securing as requiring rejection, but administrative law makes it a stretch.  If agencies were actually government bodies with free reign to create rules, willy-nilly, a required Congressional affirmative would make exceeding sense.  As it is, though, agencies cannot pass rules that statutes do not authorize.  So Congress has already provided an affirmative by passing legislation authorizing rulemakings.  A second vote to pass the actual rule resulting from passed legislation seems to me to decrease accountability.  A representative might vote yes to the “Everyone Like It in Theory” Act, but vote against (or not vote at all) for the “Actual Details of Putting It into Practice” regulations.

The question that brings us into a truly philosophical examination into our administrative state is: how does the REINS bill strike our notions of the separation of powers?  Arguments relying on the separation of powers principle rely on neatly demarcated branches of government, and folks arguing for and against the bill tend to either (1) place agencies in the executive branch or (2) emphasize that they are substantively controlled by legislation; thus, at least rulemaking should be regarded as a legislative branch activity.

Sally Katzen argues that the REINS Act would be unconstitutional, relying on an agency-as-executive approach:

Over twenty years ago, Chief Justice Rehnquist set forth several tests for whether a statute violates the Constitution’s separation of powers. One is that a statute is suspect if it “involve[s] an attempt by Congress to increase its own powers at the expense of the executive branch.” Much of the discussion surrounding the REINS Act suggests that that may be an apt characterization of the bill’s sponsors’ intent. Another of Rehnquist’s tests is whether an act of Congress “impermissibly interfere[s] with the President’s exercise of his constitutionally appointed functions,” which clearly includes the obligation to “take care that the laws be faithfully executed.” For over a century, the executive branch has taken care to faithfully execute the laws by, among other things, developing and issuing regulations implementing legislation. Justice Scalia, who of all the Justices most aggressively guards the President’s authority, has relied in key separation of powers cases such as Morrison v. Olson and Mistretta v. United States on the fact that the activities at issue in those cases were ones in which the executive had traditionally engaged.  That characterization is clearly applicable to agency rulemaking as well.

Jonathan Adler wants to distinguish “execution” from rulemaking:

Several members of the subcommittee suggested the REINS Act imposed unconstitutional constraints on executive power, particularly the executive’s responsibility to faithfully execute and enforce federal laws.  Therefore, they suggested, the REINS Act could conflict with Article II, Section 1 of the Constitution.  Set aside the curiosity of House Democrats, including Rep. Conyers, defending executive power.  This objection is based on a fundamental confusion about the nature of executive power. The power to “enforce” the laws – that is, the power to take action to see that legal rules are complied with – is distinct from the power to make the rules pursuant to a delegation of authority from Congress. So, for instance, the EPA’s power to impose fines or other sanctions on companies that violate emission limitations is distinct from the EPA’s power to set the emission limits. A requirement that federal regulatory agencies obtain Congressional approval before major rules may take effect requires Congressional assent for the latter, but has not effect on the former.

Sally Katzen raised a more nuanced separation of powers concern, but one that I also find unconvincing, and for largely the same reasons. She noted that under Morrison v. Olson, “a statute is suspect if it ‘involves an attempt by Congress to increase its own powers at the expense of the executive branch,’” and it is reasonable to see the REINS Act as an effort to constrain the executive. Just look at the bill’s full title and findings. The problem with her argument is that it ignores the distinction between executive and legislative functions.

The powers to investigate and prosecute are core executive functions. Any effort by Congress to limit such powers and aggrandize its own is problematic.

The executive power is distinct from the power to adopt legislative-type rules, however.  The latter is not a core executive function. Rather it is a quasi-legislative power that must be delegated by Congress. As the Supreme Court has stressed time and again (and as I noted in my testimony), federal agencies have no authority to promulgate regulations beyond that which has been given by Congress. And what Congress has given, it may take back. Restraining the exercise of such authority, whether by adopting rules for the exercise of regulatory authority (as under the Administrative Procedure Act or the Congressional Review Act) or limiting the scope of such authority is perfectly acceptable, so long as other Constitutional requirements (such as bicameralism and presentment) are satisfied. As the REINS Act satisfies such requirements, there is no problem. The REINS Act does not curtail executive power so much as it places limits on the legislative-like power delegated by Congress.

Adler is right to prevent over-simplistically placing agencies into one branch– whenever people talk about separation of powers, I recommend a seasoning of sufficient salt to add complexity to the dish.   I appreciate recent scholarship, such as that from Professor Mashaw, that identifies things we would see as administrative functions before the solidification of an administrative law field.  But, it is nonetheless clear that the framers in 1787 did not write up a structure of federal government that foresaw the contemporary administrative state.  So, framer-centric arguments about separation of powers elude the post-framing constitutional problems that arose as delegations of quasi-legislative and quasi-judicial powers arose.

Adler, though, seems to think that agencies, at least when making major rules, should be fully and solely creatures of Congress.  I’m not sure that is the proper understanding of the nature of agencies’ regulatory action.  An agency’s rule is not a new law; it is the carrying out of a Congressional statute.  If a rule goes beyond what an authorizing statute allows, it will be overturned.  And so, it is not a stretch to pull rulemaking away from Adler’s description of legislative activity and toward his notion of executive enforcement: a rule might be understood as analogous to a police office’s enforcement of a criminal ordinance with a policies to identify manifestations of that crime.  Likewise, Congress might declare that X is prohibited, and agencies then enforce against X by identifying X in X-1, X-2, and X-3 manifestations.  Rules, in other words, are an agency’s specified enforcement strategies of a broader Congressional policy.

I’m not interested in placing agencies definitively within wither the executive, legislative, or judicial branch.  The point, in fact, is that they do not belong, and should not be conceptualized, as being in either.  Sadly for the textualist, there is no appropriate constitutional language that provides direct guidance to the Court on administrative law.  That is not to say the language is irrelevant; but it is inadequate.

And speaking of language, it is interpretation that makes all of this most interesting.  The argument that agencies execute legislation would make tremendous sense if statutes were never vague.  But they are, which undoubtedly emphasizes the latter portion of the hyphenated quasi-legislation.

That is not to say, though, that Congress ought to take control of the rulemaking process when agencies work in the world of Chevron.  The appropriate response to the mysteries of our administrative state is not to force agencies into an existing branch.  The appropriate method for Congress to affect policy is by passing statutes.  It would be interesting to read the Court decision in a case deciding whether a latter Congress can prevent a former Congress’ legislation from taking effect by preventing associated regulations from becoming final.  My hunch is that the practice would fail.   Once a bill become law Congress cannot direct the interpretation of that law except by passing a new bill.  Chevron doesn’t require an agency, in the face of vague statutory language, to go get a Congressional interpretation; the Court has created a space for agencies to reasonably interpret statutes, and thus create policy, in a sphere outside that of our generally recognized governmental branches.