Monday saw President Trump force through another executive order - “Reducing Regulation and Controlling Regulatory Costs”. The headline was the introduction of a new “one-in, two-out” rule for new regulations:

for every one new regulation issued, at least two prior regulations be identified for elimination, and that the cost of planned regulations be prudently managed and controlled through a budgeting process.

Anything that can be done to focus regulators’ minds on the costs imposed on private businesses and groups of new regulation is probably, on net, positive. But the UK has had a policy like this since 2005, first adopting a “‘one-in, one-out” rule, then a “one-in, two-out” rule and now a “one-in, three-out” variant. The results are widely acknowledged to be mixed. Here are 4 lessons from the UK the Trump administration should bear in mind.

1. Focus on costs, not counting regulations

What really matters is not the number of regulations but the costs imposed on private businesses and civil society organizations. A “numbers” approach could be gamed: a department could introduce a new regulation, and remove a defunct one, while imposing new business costs. Thankfully, both the UK government and Trump’s executive order now recognize this. Section 2, part c) of the order says:

any new incremental costs associated with new regulations shall, to the extent permitted by law, be offset by the elimination of existing costs associated with at least two prior regulations

In the UK though, “one-in, one-out” eventually meant that for every new regulation introduced with a net cost to business, regulations up to an equivalent net cost would be eliminated. It would be better named a “pound-for-pound” rule. When upgraded to “one-in, two-out” every new regulation with net costs to business had to be compensated for by regulatory removal or revision at double the monetary cost of the new regulation. And so on. Whether badly drafted or otherwise, Trump’s version reads more like the “one-in, one-out” rule on cost, albeit having to find the cost compensation across two regulations. If implemented in this way, it could become messy to implement for many agencies. Judging regulation by pure cost rather than numbers, as the UK has done, would be a stronger constraint.

2.  Judge by net costs rather than gross costs

Any new measure, whether regulatory or deregulatory, will generate some costs to private businesses and civil society. If Trump is serious about deregulation, it therefore makes much more sense to assess “net” costs, rather than “gross” costs as a target for the new rule. This was recognized in Britain which now carries out the net cost methodology. Otherwise perverse incentives are created: departments or agencies will be cautious about ever proposing deregulatory measures where benefits to business exceed new costs, because they would still have to find gross cost savings elsewhere. As Stuart Benjamin outlines, steps taken to make pipeline construction easier, for example, otherwise might end up delayed as the agency scrambles around finding existing regulations with gross costs to remove to compensate for the very small costs of a deregulating measure. This might seem an obvious point, but at the moment the order is ambiguous – simply stating that the Director of the OMB will provide guidance “for standardizing the measurement and estimation of regulatory costs.”

Read the rest of this post »

President Trump issued an executive order on Friday that includes a ban on the entry of virtually all nationals from several countries. The same day, the New York Times published my argument that the portion of the ban that bars immigrants or legal permanent residents violates the law, which bans discrimination against immigrants based on national origin.

Andrew McCarthy of National Review Online was kind enough to take the time to publish a response (“Trump’s Exclusion of Aliens from Specific Countries Is Legal”). Because Mr. McCarthy’s article demonstrates significant confusion over my argument, the facts, and the laws at issue, it surprised me to see National Review editor Rich Lowry also cite it favorably. Despite the weakness of its analysis, the piece provides me an opportunity to clarify and reinforce some aspects of my argument that brevity required me to excise from the Times.

1. The Constitution gives the power to make immigration laws to Congress. Mr. McCarthy writes:

Under the Constitution, as Thomas Jefferson wrote shortly after its adoption, “the transaction of business with foreign nations is Executive altogether.” … In the international arena, then, if there is arguable conflict between a presidential policy and a congressional statute, the president’s policy will take precedence in the absence of some clear constitutional commitment of the subject matter to legislative resolution.

In other words, the president can ignore congressional limits in this area. He cites case law in which courts describe the president’s foreign affairs powers with respect to relations with foreign governments as expansive, but cites no case that concludes the president can ignore Congress to exclude immigrants. It is reminiscent of President Nixon’s famous argument that “when the president does it, that means it is not illegal.” It is Congress, not the president, that makes immigration law. “[O]ver no conceivable subject is the legislative power of Congress more complete than it is over… the admission of aliens,” ruled the Supreme Court in Oceanic Steam Navigation Co. v. Stranahan.

Mr. McCarthy had no problem defending this view when the actions at issue were President Obama’s, which were also justified based on “security,” but now adopts it to defend President Trump’s. As my Cato colleagues wrote at the time, “it is not for the president alone to make foundational changes to immigration law—in conflict with the laws passed by Congress and in ways that go beyond constitutionally authorized executive power.”

2. President Trump cannot use the supposed “purpose” of a statute to override its plain meaning. Mr. McCarthy quotes the relevant portion of the Immigration Act of 1965 (8 U.S.C. 1152(a)) that amended the Immigration and Nationality Act of 1952, which clearly prohibits discrimination in the issuance of an immigrant visa based on national origin. But Mr. McCarthy states:

…the purpose of the anti-discrimination provision (signed by President Lyndon Johnson in 1965) was to end the racially and ethnically discriminatory “national origins” immigration practice that was skewed in favor of Western Europe. Trump’s executive order, to the contrary, is in no way an effort to affect the racial or ethnic composition of the nation or its incoming immigrants.

Mr. McCarthy gives no citation for this claim—which contradicts everything the president and his advisors have been saying about the intent being to ban Muslims—but regardless of Mr. Trump’s intention, the result of his actions does affect the ethnic composition of the country, which was indeed one of the actions that Congress in 1965 thought it was banning.

But Mr. McCarthy is again claiming that the president can ignore the plain meaning of the laws of Congress, this time based on its supposed “purpose.” But as my colleagues at the Cato Institute put it, “Unenacted legislative intentions are not law under the Constitution.” It is the text on the page that makes law. Mr. McCarthy condemned this type of legal reasoning as a “post-law” argument when President Obama reasoned this same way in the Obamacare case, King v. Burwell, yet he eagerly adopts it now to defend President Trump.

Read the rest of this post »

In a committee vote the tightness of which surprised no one, this morning President Trump’s nominee for education secretary, Betsy DeVos, was approved on a purely partisan basis by the Senate Health, Education, Labor and Pensions committee. DeVos’s nomination now moves to the full Senate.

While the rhetoric surrounding DeVos has been heavily targeted at her competence, the main issue seems to be that Democrats generally oppose private school choice programs while Republicans generally do not. Even questions about the Individuals with Disabilities Education Act (IDEA) at DeVos’s confirmation hearing—would she support attaching IDEA rules to public funding that disabled students could take to a chosen school?—were primarily about choice.

Choice is fundamentally different from public schooling. With choice, families have real power—the power to leave a school not serving them and take their education dollars elsewhere. This is why Florida’s McKay scholarship program for children with disabilities—which DeVos tried to defend before being cut off in questioning at her nomination hearing—has very high satisfaction levels among parents using it. Public schools, in contrast, get taxpayer money no matter what, and require seemingly endless political, bureaucratic, and legal combat to hopefully—just hopefully—get improvements made.

Of course, choice needs freedom from stultifying rules and regulations to be meaningful. Specialization, competition, innovation—none can meaningfully exist without educators having the freedom to engage in new and different ways of delivering education.

The powerful inclination to wrap programs in incapacitating layers of red tape…er, “accountability”…is a major reason that the federal government should not try to deliver school choice, or govern education at all. (The Constitution is the other big one.) It is simply too dangerous to have one government—the federal government—supply choice nationwide. But there is good reason to fear that the Trump administration will try to do it nonetheless, based on Trump’s promise to make a  $20 billion choice “investment.”

Empowering parents with choice is the right way to deliver education. But the clear and present danger of freedom-smothering rules and regulations, as we’ve seen brightly illustrated by the debate over DeVos, accompanies any government funds. Which is why choice must not be delivered by Washington.