Atlas Roofing Shrugged

Another Burger Court precedent is about to be “abandoned” or “overruled” like Roe, Lemon, Chevron, Abood, and Bakke.

I've finally had a chance to finish Jarkesy v. SEC. There is no reason for that opinion to have been nearly 100 pages. The Justices need to grant more cases, reduce the time of oral argument, and spend less time writing encyclopedias. I could read Atlas Shrugged three times, and it would still take less time than reading through all the Court's decisions from this term. Some of the majority opinions are longer than the John Galt speech. Speaking of Rand…

From the 1930s through 2010, the Securities and Exchange Commission (SEC) would seek civil penalties in federal court. But that practice changed with the Dodd Frank Act, which gave the SEC the choice to bring suit in federal court, or through "in-house" adjudications before an ALJ. Unsurprisingly, the SEC prefers Door #2. The government is more likely to win before an administrative law judge. Moreover, the rules favor the government, as the Federal Rules of Civil Procedure and Evidence do not apply to these internal proceedings. And if the defendant appeals the agency's ruling to federal court, the ALJ's findings are nearly dispositive. Justice Gorsuch's concurrence well explains how this process is stacked against the accused.

The crux of Chief Justice Roberts's majority opinion is that the SEC must bring suit for civil penalties in a federal court, where a jury trial is available under the Seventh Amendment. To be frank, I'm not sure that Jarkesy really cares about having a jury. Rather, he would prefer to be in a federal court with a neutral Article III judges, and all of the protections of the Federal Rules of Civil Procedure and Evidence. I question how many SEC suits brought in federal court wind up before a jury trial. The Seventh Amendment is a means to an end.

Jarksey split 6-3. Justice Sotomayor wrote a vigorous dissent. In her view, Jarkesy was controlled by Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n (1977). This case upheld the use of in-house adjudication for the Occupational Safety and Health Administration. Chief Justice Roberts did not mention Atlas Roofing until a footnote on page 18 of his majority opinion. And he buries his analysis of that precedent at the tail-end of his decision.

Justice Sotomayor charges:

It should be obvious by now how this case should have been resolved under a faithful and straightforward application of Atlas Roofing and a long line of this Court's precedents. The constitutional question is indistinguishable. The majority instead wishes away Atlas Roofing by burying it at the end of its opinion and minimizing the unbroken line of cases on which Atlas Roofing relied.

The Court distinguishes Atlas Roofing, and finds it "does not extend to these civil penalty suits for fraud" by the SEC so "that case does not control." Justice Sotomayor finds this distinction is non-existent. I won't get into that debate here. Rather, it is far more important that Atlas Roofing is being set up to be overruled. Here, Atlas Roofing may join other Burger Court chestnuts that have roasted over an open fire, including Roe, Abood, Chevron, and Bakke.

Roberts explains that the reasoning of Atlas Roofing does not withstand scrutiny, and is inconsistent with later decisions--all factors that cut against stare decisis:

The reasoning of Atlas Roofing cannot support any broader rule. The dissent chants "Atlas Roofing" like a mantra, but no matter how many times it repeats those words, it cannot give Atlas Roofing substance that it lacks

Roberts, who usually scoffs at law professors, cites his preferred stable of scholars to show that Atlas Roofing is an outlier:

Reading the dissent, one might also think that Atlas Roofing is among this Court's most celebrated cases. As the concurrence shows, Atlas Roofing represents a departure from our legal traditions. See post, at 12–20 (opinion of GORSUCH, J.). This view is also reflected in the scholarship. Commentators writing comprehensively on Article III and agency adjudication have often simply ignored the case. See, e.g., R. Fallon, Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915 (1988) (no citation to Atlas Roofing); J. Harrison, Public Rights, Private Privileges, and Article III, 54 Ga. L. Rev. 143 (2019) (same); W. Baude, Adjudication Outside Article III, 133 Harv. L. Rev. 1511 (2020) (same).

The Court stops short of overruling Atlas Roofing. The Court did "not reach" Jarkesy's argument that Granfinanciera, S. A. v. Nordberg (1989) "overruled Atlas Roofing." Though Chief Justice Roberts does leave open the possibility that Atlas Roofing was already "abandoned" like Lemon:

The dissent chides us for "leav[ing] open the possibility that Granfinanciera might have overruled Atlas Roofing." Post, at 25, n. 8 (opinion of SOTOMAYOR, J.). But the author of Atlas Roofing certainly thought that Granfinanciera may have done so. See Granfinanciera, 492 U. S., at 79 (White, J., dissenting) ("Perhaps . . . Atlas Roofing is no longer good law after today's decision."); see also id., at 71, n. 1 (Granfinanciera "can be read as overruling or severely limiting" Atlas Roofing).

Justice Gorsuch's concurrence goes further, and explains that Atlas Roofing cannot be squared with original meaning:

The high-water mark of the movement toward agency adjudication may have come in 1977 in Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n. Some have read that decision to suggest the category of public rights might encompass pretty much any case arising under any "'new statutory obligations,'" Brief for Petitioner 22 (quoting Atlas Roofing). It is a view the government essentially espouses in this case. But without reference to any constitutional text or history to guide what does or does not qualify as a public right, that view has (unsurprisingly) proven wholly unworkable.

Michael Showalter explains how the Court tried to reconcile Atlas Roofing, as precedent, and original meaning. But this decision is merely a temporary way-station on the road to overruling. One last chance, as Richard Re might call it.

Justice Sotomayor sees the writing on the wall. If Jarkesy is extended beyond the SEC, many other agency adjudications would have to be brought in federal court. Indeed, some agencies do not even have the authority to seek civil penalties in federal court, so Congress would have to grant that power:

Some agencies, like the Consumer Financial ProtectionBureau, the Environmental Protection Agency, and the SEC, can pursue civil penalties in both administrative proceedings and federal court. Others do not have that choice. As the above-cited statutes confirm, the Occupational Safety and Health Review Commission, the Federal Energy Regulatory Commission, the Federal Mine Safety and Health Review Commission, the Department of Agriculture, and many others, can pursue civil penalties only in agency enforcement proceedings. For those and countless other agencies, all the majority can say is tough luck; get a new statute from Congress.

I think it unlikely that a divided Congress would ever enact such a statute, so these agencies would be severely weakened. Justice Sotomayor describes this change as a "seismic shift" that would unleash "chaos."

I think Atlas Roofing is on borrowed time. And there is a vehicle on the horizon that presents this issue. 

The SpaceX case against the NLRB pending in the Fifth Circuit has expressly preserved the issue of whether Atlas Roofing should be shrugged: 

The NLRB may argue that the Supreme Court has already upheld its proceedings against a Seventh Amendment challenge. See NLRB v. Jones & Laughlin Steel Corp. (1937); Atlas Roofing Co. v. OSHRC (1977). SpaceX expressly preserves its argument that both these cases should be overruled as they were wrongly decided. Notably both these cases are missing from the Supreme Court's litany of historic public rights cases in Jarkesy at 2131-34. And the Court leaves open that Atlas Roofing has been overruled in footnotes 3 and 4. Moreover, the Court notes Atlas Roofing adopts a problematic "circular" definition of public rights. Id. at 2138-39. And the dissent expressly recognizes that "it is unclear how OSHA, or the National Labor Relations Act at issue in Jones & Laughlin, would fit the majority's view of the public-rights doctrine." Id. at 2165 (Sotomayor, J., dissenting).

The Fifth Circuit reinvigorated the Seventh Amendment in Jarksey. And it was affirmed. The SpaceX case could continue that trend.

At a recent conference, I was asked if I could overrule any one Supreme Court precedent, what would it be? My answer, without hesitation, was NLRB v. Jones & Laughlin Steel. People often point to Wickard v. Filburn as the focal point of the Supreme Court's expansion of federal power. But really that explosion began with Jones & Laughlin Steel. That case adopted the "substantial effects" test, which eliminated the distinction between what is local and what is national. Jones & Laughlin Steel, in addition to obliterating limits on Congress's commerce powers, also wiped out protections of the Seventh Amendment. As a result, the National Labor Relations Authority can bring suits in-house, and not before federal courts. Atlas Roofing may be on the chopping block, so let's see if the Roberts Court would reach back to the New Deal. 

Comments

[ReaderY]

Atlas Roofing is straightforwardly distinguishable. Under Jarkesy, the 7th Amendment applies when Congress federalizes causes of action analogous to traditional civil torts, in which specific individuals are damaged.

OSHA safety regulations in general are not like that. They, like many regulations, are (in general) prophylactic in their character, designed to prevent harm, not to remedy harm that has already occurred. Prophylactic regulations are not like civil torts. I think it’s pretty clear that under Jarkesy, the 7th Amendment wouldn’t apply.

Nonetheless, Jarkesy might apply to some OSHA enforcement actions. For example, when OSHA seeks damages after an accident has occurred, the situation would be more analogous to a traditional civil tort, as it would be remedial rather than prophylactic in character, and there would be identified accident victims who would ordinarily sue for civil torts if Congress hadn’t preempted the common law.

[ReaderY]

The dissent may be right that the distinction is inconnsistent with past precedents. But I think Justice Sotomeyor is wrong to suggest that this distinction is meaningless, arbitrary, or unworkable. When the federal government files a civil action that effectively steps into the shoes of private plaintiffs who would previously have sued on their own behalf in a civil tort, the 7th Amendment applies. This means it’s not limited to the common law as it was defined in 1789. But it’s not completely unbounded.

[Dilan Esper]

The problem with overturning Jones & Laughlin Steel is that the pre-New Deal commerce clause test was absolutely stupid, unworkable, immoral, inconsistent with constitutional text, and would destroy the country and turn us into a Banana Republic if reinstated. Hammer v. Dagenhart should make lists of the worst SCOTUS cases of all time-- not only did it uphold a result that is profoundly immoral, but it did so based on a brain-dead holding, that running a factory that manufactures things for sale all over the country is not interstate commerce.

I don't like Wickard and Raich very much, so I get the basic idea of pushing back against things that go ridiculously in the other direction (I'm sorry, but growing marijuana on your windowsill for your own use is in no way connected with interstate commerce), but the basic reason the Commerce Clause expanded is because we are now an interconnected country with massive interconnected commercial relationships. That moved a ton of stuff that was once the province of the states into the commerce clause, and that shift of power was basically necessary for the United States to be as prosperous as we are. The option of uniformity of commercial regulation is extremely important. We'd be a crappy, poor country without it.

[ReaderY]

Why immoral?

If one accepts that having 2 levels of government and sovereignty isn’t immoral, so that there isn’t a single set of rules the entire world is subject to, why should having 3 levels be any less moral than 2? There may be practical reasons not to do it, at least in particular cases and circumstances. But immoral? It seems to me that once one accepts that something other than a monopolis is moral, does it really matter so far as morality is concerned how many different levels there are? Why not 4?

Moreover, there are a lot of reasons to think that a single economy may not be ideal. Suppose all the country’s hamburger was processed in a single huge vat. Sure, it would be more efficient when it was working than lots of little local hamburger plants scattered around the country. But efficiency is not the only criterion. Variation, redundancy and compartmentalization are both often critical to resistance to stress and hence survival. With a centralized food supply, a single outbreak of disease would wipe the entire food supply out. Look at the Irish potato famine which follwed a whole society’s dependence on a single variety of plant for sustenance.

The idea that centralality, uniformity, and standardization are always good, prevalent in the 1930s, is a distinctly outdated idea today. We now know that variation, redundancy, and compartmentalization are essential to species survival and evolutionary success, because they provide robustness and survivability against predators, disease, disasters, and other stresses.

Economic and political systems are no different. Efficiency is only advantageous up to a point. The societies that survive the periodic changes and disasters that befall human history, like the species that survive the periodic changes and disasters that befall evolutionary history, are the inefficient ones. The efficient ones become great for a time, then can’t adapt to changed environments and die off.

[Dilan Esper]

Striking down a child labor law is immoral.

Moreover, there are a lot of reasons to think that a single economy may not be ideal.

It's not that a we have to have a single economy in all respects. It's that without the OPTION- if a single economy is UNCONSTITUTIONAL- it would destroy our ability to have regulated capitalism. We'd become a tiny basketcase with a ungovernable mess of an economy while other countries that have national systems would pass us. We'd lose our position the same way Argentina lost their's.

[SMP0328]

I agree with Dilan Esper. Jones & Laughlin Steel is a middle ground between the rigidity of the pre-New Deal reading of the Commerce Clause and the all-bets-are-off reading under Wickard and Raich. Over the last 30 years, the Court has pushed back against Wickard via the anticommandeering doctrine under the 10th Amendment and requiring the activity being regulated be economic in nature. I don't think this Court is willing to go further. The Court might overrule Wickard and/or Raich, but I think it is highly unlikely it would overrule Jones & Laughlin Steel.

Incidentally, the Child Labor Amendment would likely be adopted in short order if Hammer v. Dagenhart were ever reinstated. That proposed amendment is still technically before the States (no ratification time limit).

[Corve5]

That we would be a poor, crappy country without the bullshit interpretations of the Commerce Clause is an argument for amending the Constitution, not ignoring it.

[Dilan Esper]

No it isn't. Following conservatives' dumb and minority positions about what the Constitution means is one millionth less important than having a working government that delivers prosperity.

[Corve5]

Prosperity for whom? The left's idea of prosperity is taking from white men and giving to everyone else.

[circleglider]

Most directly, it’s the prosperity of an otherwise unemployable and unelectable million or two over-educated busybodies who are convinced the world would far apart without their righteous moral guidance.

[Dilan Esper]

It's the prosperity that actually is predicted by conservative economists, who very much espouse consistent regulation and uniformity.

[Squirrelloid]

I don't think the founders ever envisioned the federal government regulating the operation of a business, much less one operating in a single location (most businesses in the founding era). They would have been appalled by the idea. That's a matter for state or local government.

The whole point of the interstate commerce clause was to *stop the states from disfavoring commerce from other states*. (It was written to address a problem that emerged under the articles of confederation, where states imposed tariffs on goods from other states). It was not intended to allow the federal government to regulate businesses.

So immoral? The expansion of the commerce clause beyond what it was intended to do without a constitutional amendment is what was immoral.

[ReaderY]

Disagree. If the Framers had only intended the Negative Commerce Clause and never intended to authorize positive regulation, why would they have adapted the language they did, which explicitly authorizes Congress to regulate interstate commerce and says nothing about prohibiting states from discriminating against other states? Shouldn’t we assume they meant what they clearly said?

I think there’s no question the Commerce Clause permits regulating business when they are engaged in interstate commerce. The only difference is, what is this “interstate commerce”?

In 1900, interstate commerce was basicall ythe movement of goods across state limes. Businesses which sold and moved goods across state likes could be regulated, directly, for those sales and movements, a much narrower conception of “interstate commerce.” But it still involved the direct regulation of businesses.

As mentioned in my previous comment, I don’t think there’s any basis for the negative commerce clause. We don’t necessarily want to make the movement of hazardous materials, diseased plants or animals, or toxic waste across states lines, for example, as easy as possible; not all movement is an unvarnished positive. As I see it, under the Commerce Clause it’s for Congress and only Congress to decide what’s good for the country as a whole and what isn’t. It’s not for courts to incorporate their own dubious economic theories into the text based on things the Framers supposedly thought but never said.

If Congress likes the policy behind the negative commerce clause, it can simply pass a statute that incorporates it. Then it will be Congress’ policy, as it should be.

[Dilan Esper]

The framers intended that any problem of national scope be within Congress' power. We know that because that was the instruction to the Committee of Detail that produced the Commerce Clause.

[Brett Bellmore]

It wasn't what they wrote into the Constitution, though.

[circleglider]

It wasn’t what they wrote. It wasn’t what they intended. And they certainly wouldn’t have seen the operation of free markets on a continental scale as anything approaching a “problem” in need of fixing.

Maybe there’s some hope for the Roberts Court after all.

[Dilan Esper]

The point isn't that it is "what they wrote"; it's that if you have competing frames of the Commerce Clause, and one effectuates that intent, you should favor that one over the one that doesn't (especially when the latter one arbitrarily excludes various types of commercial regulations from the clause).

[Squirrelloid]

" I don’t think there’s any basis for the negative commerce clause. We don’t necessarily want to make the movement of hazardous materials, diseased plants or animals, or toxic waste across states lines, for example, as easy as possible; not all movement is an unvarnished positive."

Under my reading, it's the state responsibility to stop the movement of improper materials across state lines, and congress can simply not impede them from doing so. ie, the commerce clause gives congress the power to undo state prohibitions or disincentives on interstate movement of goods *when it wants to*, it doesn't *require* it to do so in all cases. It doesn't forbid state regulation on commerce (dual sovereigns), just gives the federal government a superior say in the matter. (Handling diseased animals, for example, would be an exercise of the police power - which the federal government is never given and does not have. This is properly the domain of the states.)

1900 is pretty late for looking for evidence of original understanding.

https://www.senate.gov/artandhistory/history/minute/Interstate_Commerce_Act_Is_Passed.htm

"Before 1887, Congress had applied the Commerce Clause only on a limited basis, usually to remove barriers that the states tried to impose on interstate trade."

So for the first ~100 years, Congress understood the commerce clause in pretty much the way I claimed.

Now, I also think the Supreme Court decided Wabash v. Illinois wrongly. The commerce clause does not forbid state regulation of commerce, even interstate commerce. The federal government can, if it chooses, overrule state regulations on interstate commerce, but it can also leave them in place by not acting. So the court was wrong to strike down the Illinois law without an act of Congress to the contrary.

[John F. Carr]

"I question how many SEC suits brought in federal court wind up before a jury trial. The Seventh Amendment is a means to an end."

In Massachusetts a defendant may get out of small claims court by insisting on the right to a jury trial. Later consent to a bench trial could result in the case going back to small claims.

[ReaderY]

I suspect this would be the case here as well. It may just not be possible to get an Article III judge without a jury. You either request a jury, which gets you to federal court, or you don’t, which keeps you in administrative proceedings.

[JoeFromtheBronx]

What changed the nature of what is "local" and "national" is modern day realities.

[Dilan Esper]

Indeed. And a lot of self-styled originalist conservatives do not understand that.

[Soronel Haetir]

Sorry, I simply fail to see how regulating a business that does zero interstate commerce, such as my family's liquor store, should come within the ambit of federal control at all. The distributors we buy from, sure. But we are legally *forbidden* (by the state) from performing either interstate sale or purchase.

We *must* buy all our stock from a list of in-state distributors, and all our sales *must* be on-site

I see no problem with limiting "commerce" to the founding era's buying and selling of goods. (I might be convinced to extend this to services, but even that's a tough question). I believe the Roosevelt-era courts did the nation a great disservice simply allowing Congress to do as it would rather than requiring constitutional amendments (amendments which I suspect would have been readily adopted at the time).

[JoeFromtheBronx]

"zero interstate commerce, such as my family’s liquor store, should come within the ambit of federal control at all. The distributors we buy from, sure"

Where does this alcohol come from? Is it all produced in the state? Is it a niche business? Do you not sell to anyone who comes from out of state or use the alcohol for some sort of interstate purpose?

How about the people who work at all the distributors and all the goods required to distribute including vehicles, machinery, etc? Which is interconnected with businesses like your own. It is all an interconnected web of commerce.

Are they all purely intrastate? What about the federally backed money, bankruptcy protection, unemployment insurance, health care, etc. involved?

As to needing to buy stock from in-state distributors, alcohol is a specialized good given the 21st Amendment, but the Supreme Court still has limited the discretion of state favoritism.

Congress for goods generally can also limit such favoritism for goods involved in interstate commerce.

Why "services" is not part of interstate commerce is unclear too. It's an artificial rule not found in text & services are a standard part of "commerce" and were then too.

There was no need for an amendment here. Interstate commerce is involved.

[Squirrelloid]

The commerce clause allows the federal government to regulate commerce among the states, not anything remotely connected to commerce among the states.

The local retailer who buys from an in-state distributor and sells only at a physical location is *not* interstate commerce in any sense. Even if someone comes from out-of-state and purchases an item at their establishment, the transaction happens entirely within the state. It doesn't matter where the purchaser is from.

The distributor they buy from (who purchases from out of state) - sure, those out-of-state purchases are commerce between the states, but then the sales in-state are not interstate. (The whole business is not 'interstate commerce', only the specific transactions that cross state lines).

Commerce is specifically transactions, not the whole business. A business is not commerce. Whether the commerce clause should reach a transaction is fact-specific to that transaction, and requires that specific transaction involve the movement of goods or provision of services across state lines.

[Dilan Esper]

That's silly. I will concede the conservative argument on Wickard and Raich: if someone is distilling their own liquor, from their own grain, and selling it on premises, the connections to interstate commerce are pretty slim.

But why should the sale of Jack Daniels in a California liquor store be outside the commerce clause. Why can't the federal government, for instance, regulate what is on that Jack Daniels label or the alcohol content within the bottle? And why can't the federal government regulate who it is sold to? That's all commerce among the several states. It's totally artificial to cut it off at the liquor distributor or store.

[Squirrelloid]

How is the local store engaged in interstate commerce? Which *transaction* they engage in crosses state lines?

Also, they shouldn’t be able to regulate the bottle label at all, even if the bottle does cross state lines. That’s not a transaction.

(Alcohol content, sure, that's a regulation of what can be transacted across state lines, because 35% is a different good than 40%, and you could ban one transaction and allow the other. I don't see a legitimate policy justification for the government to care about % of alcohol in interstate commerce, but that's not a constitutional bar).

[ReaderY]

And? It’s hardly the last time something came out of committee looking rather different from what went in. And we generally follow what the committee actually and the assembly approved after lots of discussion and compromise, not whatever the original idea was when it first went into committee. It’s therefore not a LEGITIMATE interpretation of the phrase.

It’s like throwing out the language of a negotiated contract and instead saying that ones prefered side’s original proposal was what was originally meant, and the negotiated language just obscured the wording. The whole point of legitimate republican government is that social rules are negotiated, and negotiations are respected. Agreements have to be honored.

I’m a legitimist, not an originalist. “Commerce among the several states” doesn’t have to mean exactly what it meant in 1789. But it must nonetheless mean some reasonable, plausible interpretation of “commerce among the several states.” And “any problem of national importance” just isn’t a reasonable or plausible interpretation of “commerce among the several states.” It means something entirely different.

[JoeFromtheBronx]

"any problem of national importance” is not the test

That was generally speaking what the framers were aiming for in drawing lines, as was noted at one point, but there are specific enumerated powers.

The average federal commercial regulation does have a "plausible" connection to interstate commerce.

[Brett Bellmore]

And still fails, because the commerce clause doesn't say Congress can regulate anything that has a connection to interstate commerce, it says it can regulate that interstate commerce itself.