Cato Op-Eds

Individual Liberty, Free Markets, and Peace
Subscribe to Cato Op-Eds feed

A restriction of free speech by any other name is still unconstitutional. No matter how much the Missouri government wants to regulate alcohol it may not do so by restricting the freedom of speech. Cato joins the ACLU and the Freedom Center of Missouri on an amicus brief supporting a challenge to a Missouri law prohibiting alcohol producers from advertising alcohol prices unless the prices are displayed inside a retailer—and they may not advertise who their retail partners are unless they list more than one. This both limits and compels speech.

The government argues that advertising is “commercial speech” and therefore not afforded the same protections under the First Amendment as every other type of speech. The “commercial speech” doctrine traces back to Valentine v. Chrestensen, an infamous case from the 1940s in which the Court arbitrarily decided commercial advertising wasn’t constitutionally protected. The Court has slowly been hacking away at this arbitrary rule, eventually creating an intermediate protection for commercial speech in under the Central Hudson Gas & Electric Corp. v. Public Service Commission (1980), where courts must balance four factors before deciding whether to nullify a restriction on commercial speech.

The Central Hudson test requires courts to determine: (1) whether the speech is false or misleading, (2) whether the government has a “substantial interest” in regulating the speech at issue, (3) whether the censorship directly advances the government’s interest in regulating the desired object, and (4) whether the government’s speech-regulation is no more restrictive than necessary. As reasonable as the words may seem in the abstract, the test has sprouted a thicket of arbitrary rulings contrary to the original meaning of the First Amendment.

What exactly qualifies as a “substantial interest” for a government has never been particularly clear. Nor has the Court been able to demonstrate where in the First Amendment some types of speech are afforded greater protections than others. Justices from John Paul Stevens to Clarence Thomas have criticized the Central Hudson test for its arbitrary factors and lack of grounding in the text of the First Amendment.

Our brief argues that alcohol advertising is just as much a form of speech as literature and political speech, deserving of protection from arbitrary government restrictions. The Central Hudson test should be eliminated in favor of the standard First Amendment protection for any kind of speech—which of course isn’t absolute, but is also not subject to government whim.

Despite the “commercial speech” doctrine’s emergence seemingly over night, and that the Central Hudson test still stands after nearly 30 years, Cato once more opposes this impediment to both the freedom of speech and economic liberty in the hope that, as Shakespeare put it in Henry VI, Part III, “many strokes, though with a little ax, hew down and fell the hardest timbered oak.”

Last night, the Senate passed the Formerly Incarcerated Reenter Society Transformed Safely Transitioning Every Person (FIRST STEP) Act, a bill which could provide relief to thousands of federal inmates and an untold number of future defendants for federal drug crimes. The Senate was widely considered to be the biggest hurdle to the bill’s success, and so activists and other supporters are celebrating in a way seldom seen in Washington criminal justice circles. With the president’s vocal support of the bill, it is likely to pass the House and be signed into law.

Although most news stories about the bill feature the “unlikely allies” from across the political and ideological spectrum that supported this bill, such language downplays the many years of tireless effort organizations on the left and right have put into bipartisan reform. Previous efforts have come up short, in one instance painfully close to the finish line, so the relief and elation many advocates are expressing are well-earned. While Cato policy precludes its scholars endorsing legislation, my most sincere congratulations to the many organizations and individuals who fought so hard to make yesterday’s passage happen.

All that said, the FIRST STEP Act is aptly named because it only serves a small segment of the nation’s incarcerated population and thus cannot be the endpoint in the ongoing struggle for criminal justice reform. The United States holds over two million people in jails and prisons, mostly at the state and local levels. To the eligible prisoners who will be affected by FIRST STEP, the law may be a godsend, but for the vast majority of inmates, nothing will change when the president signs the bill into law. Taking nothing away from the achievements of FIRST STEP’s proponents, so much more needs to be done at every level of our criminal and carceral system.

Law enforcement officers continue to harass and over-police neighborhoods, bringing more people into the correctional system than need to be there in the first place. Prosecutors overcharge and over-sentence offenders, particularly those that exercise their constitutional rights to jury trials. Judges are still hamstrung by mandatory minimum sentencing laws that remove discretion and mercy in favor of punishment that may do more damage than good to the community.

Again, congratulations to our friends on the left, the right, and those in between who worked so hard to pass FIRST STEP. Cato is already working with some of these organizations on the next steps to making our criminal system fairer and less punitive.

President Trump has repeatedly cited drug smuggling as a reason to build a wall along the Southern border. But my new Cato policy analysis shows that, if stopping drug smuggling is the goal, a border wall is about the worst possible investment. Here are a few of the main findings:

  • Hundreds of miles of border fences built from 2003 to 2009 had no effect on marijuana smuggling.
  • Marijuana legalization starting in 2014 has cut marijuana smuggling between ports of entry (i.e. where a wall would go) 78 percent from 114 pounds per agent in 2013 to just 25 pounds per agent in 2018.  
  • Since marijuana is the primary drug smuggled between ports of entry, the total value of all drugs seized by the average Border Patrol agent fell 70 percent from 2013 to 2018.
  • As a result, the average inspector at ports of entry made drug seizures that were three times more valuable than those made by Border Patrol in 2018. In 2013—prior to legalization—the average Border Patrol agent made more valuable seizures.
  • By weight, the average port inspector seized 8 times more cocaine, 17 times more fentanyl, 23 times more methamphetamine, and 36 times more heroin than the average Border Patrol agent seized at the physical border in early 2018.

The best proxy measure for changes in drug smuggling is the amount of drugs seized by Border Patrol. To control for enforcement levels, the paper looks at seizures per agent. From 2003 to 2013, the rate of seizures per agent was virtually constant, even as the number of agents doubled and the government constructed hundreds of miles of border fences. But enforcement didn’t stop the flow—only when states, starting in 2014, started to fully legalize marijuana did marijuana smuggling decline (Figure 1).

Figure 1: Marijuana seizures and length of border fences

[HTML] Figure 2
!function(e,t,s,i){var n="InfogramEmbeds",o=e.getElementsByTagName("script")[0],d=/^http:/.test(e.location)?"http:":"https:";if(/^\/{2}/.test(i)&&(i=d+i),window[n]&&window[n].initialized)window[n].process&&window[n].process();else if(!e.getElementById(s)){var r=e.createElement("script");r.async=1,,r.src=i,o.parentNode.insertBefore(r,o)}}(document,0,"infogram-async","");

Because marijuana is primarily smuggled between ports of entry, the most valuable drug smuggling now occurs at ports of entry. As Figure 2 shows, Border Patrol was seizing more drugs—by value—in 2013, but by 2018, value of drugs seized at ports of entry was three times the value of drugs seized between ports of entry by Border Patrol. In other words, a border wall would not target the most valuable drugs crossing the border.

Figure 2: Value of drug seizures per agent by location of seizure [HTML] Figure 6
!function(e,t,s,i){var n="InfogramEmbeds",o=e.getElementsByTagName("script")[0],d=/^http:/.test(e.location)?"http:":"https:";if(/^\/{2}/.test(i)&&(i=d+i),window[n]&&window[n].initialized)window[n].process&&window[n].process();else if(!e.getElementById(s)){var r=e.createElement("script");r.async=1,,r.src=i,o.parentNode.insertBefore(r,o)}}(document,0,"infogram-async","");

All of this should inform policymakers on how to address illegal immigration as well. Policies that make legal immigration easier undermine the incentivizes for black market activity. As Figure 3 shows, when lesser-skilled guest worker admissions increase, the number of apprehensions per Border Patrol agent declines. Since 1949, a 10 percent increase in guest workers was associated with an 8.8 percent decrease in apprehensions per agent. The current H-2A and H-2B guest worker programs for seasonal workers are already helping the situation greatly, but they could be improved and expanded on to cut illegal entries further.

Figure 3: Lesser-skilled guest worker admissions and apprehensions per Border Patrol agent [HTML] Figure 8
!function(e,t,s,i){var n="InfogramEmbeds",o=e.getElementsByTagName("script")[0],d=/^http:/.test(e.location)?"http:":"https:";if(/^\/{2}/.test(i)&&(i=d+i),window[n]&&window[n].initialized)window[n].process&&window[n].process();else if(!e.getElementById(s)){var r=e.createElement("script");r.async=1,,r.src=i,o.parentNode.insertBefore(r,o)}}(document,0,"infogram-async","");

If Congress wants to address drug smuggling, it should legalize marijuana nationwide and invest in better ports of entry, not build a wall. The same lessons that drug smuggling have provided should apply to illegal immigration, cutting demand for illegal immigration by making more legal options available to immigrants.

This morning, Acting Attorney General Matthew Whitaker signed the Trump Administration’s new regulation banning bump stocks. The final rule is largely unchanged from the one put up for notice and comment several months ago, but contains over 100 pages of responses to the tens of thousands of comments submitted both in favor of and opposition to the rule (the majority of which opposed the rule, including a comment submitted by us and Josh Blackman).

The Administration’s main contention continues to be that language of the National Firearms Act and Gun Control Act—which has not needed clarification in some 80 years—is ambiguous in regards to bump firing, a contention we dispatched back in June. Yet the government is attempting to use this supposed ambiguity to shoehorn bump stocks into a statute that regulates possession of machine guns.

Regardless of what public opinion is at this moment, the law means what it says. The executive branch has the power to interpret existing law, not write new ones. The administration argues, essentially, that because the statute did not provide a separate definition of the terms “automatically” and “function,” that it gets to insert their own meaning. That simply isn’t the case. Administrative interpretations are supposed to do just that—interpret existing law—not give new meaning to an old one.

In this case, the existing law specifically defines “machine gun”; several administrations have reviewed bump stocks and repeatedly determined that they don’t fall in that category. It’s been clear for decades that Gatling guns and bump stocks were not machine guns. This regulation is not an attempt to clarify a vague law, but to seize political expediency to expand the power of the executive.

If the government really wants to regulate bump stocks, it needs to do so by passing a new law, not by assigning new meaning to an old one. The Founders weren’t short-sighted; there is a reason laws that affect the entire nation have to come through Congress, not reimagined by bureaucrats.

What’s worse is the fact that the administration’s attempt to skirt the Constitution is for something as inconsequential as bump stocks. We are talking about seriously damaging the integrity of our legal system over a novelty item. In a country as divided as ours, this seems like a squandering of political capital.

The powers of the federal government are few and defined, not open to unlimited interpretation by unelected bureaucrats. In light of this, our regulatory comment potentially gives Cato a right to intervene in the coming litigation. Stay tuned.

In a series of retrospectives on 1968, NPR reports on a sort of jury nullification that took place in Long Island and created a pre-Stonewall victory for gay rights and sexual freedom. In the summers, gay men would come out from New York City and elsewhere for sunbathing, house parties, and in some cases anonymous sex in a wooded area between the towns of Cherry Grove and Fire Island Pines. And some were arrested, fined, or even imprisoned. Then in 1968 some decided to fight back and hired a lawyer willing to fight for them:

In the autumn of 1968, close to two dozen gay men were acquitted of consensual sodomy charges in a series of criminal trials on Long Island. The trials and acquittals marked a pivotal moment in what eventually became the gay rights movement. They demonstrated to the larger gay community — then mainly closeted — that gay people could band together to resist police harassment….

In late August, 1968, police arrested 27 men in Cherry Grove. A few pleaded guilty to consensual sodomy and payed a fine of $250. But 22 men fought the charges in court.

Benedict Vuturo, a prominent Long Island criminal defense lawyer, was retained by the Mattachine Society. In the fall of 1968 Benny Vuturo, as he was known, demanded jury trials for all of the gay men he was defending.

“Benny said there’s terrible crimes on the mainland of Long Island, murders and rapes, and here the cops go and they beat the bushes and try to find these gay fellas who are not harming anyone,” said reporter Karl Grossman, who covered some of the trials for the Long Island Press.

“The juries, one after another, concurred, and they found the defendants not guilty, not guilty, not guilty. And that was the end of the police raids on Fire Island. To me, it really was a testament to the common sense of eastern Long Island residents who served on those juries, and to the jury system.”

Vuturo was hoping to lose one of the trials so he could challenge New York’s sodomy law but he won every case.

The state’s sodomy law was overturned in 1980, 12 years after the Fire Island trials. 

There wasn’t much doubt that the men had been doing what the law prohibited. Yet Long Island juries found them not guilty. That’s a phenomenon often called jury nullification, defined by the Legal Information Institute as “A jury’s knowing and deliberate rejection of the evidence or refusal to apply the law either because the jury wants to send a message about some social issue that is larger than the case itself, or because the result dictated by law is contrary to the jury’s sense of justice, morality, or fairness.” Read more about jury nullification in this Cato Institute book and in a Wall Street Journal article discussed here. It won’t surprise you to hear that judges and prosecutors don’t want juries to know their rights. I wrote about Stonewall here.

It turns out that NBC News covered the 1968 Fire Island story four months ago.

Senator Elizabeth Warren says yes, because patent protection gives drug companies monopoly power that they exploit. Her suggestion raises several issues.

First, Warren is right that while patents might incentivize innovation, they also keep prices elevated while new drugs are under patent, thereby reducing utilization. Patent policy should seek to balance these two effects, and current policy might not be the right balance.

Second, Warren’s suggested policy - more government, rather than reduction or elimination of existing patent protection - fits the standard progressive approach: assume the fix to imperfect government is more government, rather than less. 

Third, libertarians are divided over government patent protection. On the one hand, libertarians endorse a government role in defining and enforcing property rights generally, so why should intellectual property be any different? (And as a bonus, intellectual property protection is an enumerated power). Further, standard economics suggests that private investment in new ideas might be insufficient without patents.

On the other hand, the current patent system generates a non-trivial frictions: patent trolls try to “hold up” firms that might use patents for new products. This causes no ineffiiciency if Coasian bargaining costs are zero, but that seems unlikely. 

Existing research, moreover, provides little evidence that patent protection spurs innovation (although pharma may be an exception).

So, Warren raises a valid concern over patents. But rather than having government manufacture generic drugs (what could possibly go wrong?), why not just scale back existing patent protections?

By the end of 2019, Facebook promises to establish an independent body to handle appeals of its content moderation decisions. That intention follows an earlier suggestion by Mark Zuckerberg that Facebook might establish a “Supreme Court” of content moderation. Like the real Supreme Court, Facebook’s board will presumably review the meaning and application of its Community Standards, which might be considered the basic law of the platform.

There are many questions about this new institution. This post looks at how its members might be selected.

To fix ideas, let’s begin with how members of the U.S. Supreme Court are selected. Appointments to the highest court are procedurally simple and normatively complex. Article II of the Constitution says the president “shall nominate, and by and with the Advice and Consent of the Senate, shall appoint…Judges of the Supreme Court…” Advice and Consent can mean a simple majority or a supermajority of senators. Whatever the rule, senators vote only once on a nominee. Thereafter justices “hold their Offices during good Behavior.” (Article III) In practice that means justices continue serving however unpopular their decisions. Of course, justices may be impeached and removed from the Supreme Court. However, Congress has not removed justices or judges because of their decisions. It really does require bad behavior.

Clearly the framers of the U.S. Constitution valued judicial independence, especially a certain distance from the unfiltered will of the majority. A candidate for the presidency may promise voters to nominate a favored judge to the Court; he lacks power to seat anyone. Candidates for the Senate may promise to support or oppose a nominee to the Court, but no senator or group of senators can decide whom to nominate. Once seated, the “good Behavior” standard means a justice serves until retirement, death, or impeachment and removal. The first two are by far the most likely means of departing the Court. The justices need not fear their peers in the executive or the legislative branches or indeed, the people themselves, since they may not be recalled by an angry electorate.

But justices are not free to exercise the judicial power of the United States as they wish. Presumably they are obligated to interpret and apply the words of the Constitution which both empowers and limits the government. The courts are not independent of “We, the People” understood over time as the will of majorities and supermajorities expressed as the text of and amendments to the Constitution. The presidents and senators who nominate and appoint justices also depend directly or indirectly on voters. In these ways, the selection of Supreme Court justices balances independence and representation, thereby fostering the legitimacy of the Court and its decisions.

Following legal and constitutional values might enhance the legitimacy of Facebook’s “appeals court.” Facebook clearly values independence in this new review board: Zuckerberg describes the board as “an independent body, whose decisions would be transparent and binding.” But he also says, the board would “uphold the principle of giving people a voice.”

You might think Facebook’s Community Standards cannot support its review board the way the Constitution supports the Supreme Court. The Constitution gained consent through a deliberative process that led to approval of the document eventually in all states. Nothing like that happened at Facebook. Yet every user consents to abide by Facebook’s Community Standards when joining the platform. We might question the quality of that consent, but it seems similar to the consent given to the Constitution by those of us who joined the “platform” after 1789.

What about applying the Community Standards? Zuckerberg says the board should be independent of Facebook for three reasons:

First, it will prevent the concentration of too much decision-making within our teams. Second, it will create accountability and oversight. Third, it will provide assurance that these decisions are made in the best interests of our community and not for commercial reasons.

Members of the board might serve during “good behavior” like justices. This would create space for its members to interpret and apply Facebook’s Community Standards as they saw fit.  Perhaps extended terms for members of the board could attain the same end. But remember: even Supreme Court justices can be removed for bad behavior.

Facebook’s board also needs to be independent in the sense of being free of both politics and commerce. Many people fear that Facebook’s content governance reflects the political commitments of its managers and employees. The interpretation of Facebook’s Community Standards also could become the plaything national political forces.  In both instances, concerns about independence reflect worries about misrepresentation. What should be for all turns out to reflect the will of a few. How might the selection of review board members better represent Facebook’s users?

Let’s begin with a straightforward idea of representation. Imagine Mark Zuckerberg appoints the members of the board. Zuckerberg is accountable to Facebook’s users because they can exit the platform and thereby harm or destroy his business. That constraint would mean his appointments represent the concerns of users, along with other matters important to the business. Zuckerberg is a faithful agent of his customers not because he wishes to be so but because he must be.

Facebook wants its content moderation to be accepted as legitimate by its users (and by others). Would users accept this market theory of representation for the board? Many people doubt that markets constrain business managers. Others will think of representation as direct voting rather than indirect responses to consumer desires. Partial acceptance of the market theory may not be adequate to legitimize the new body.

So instead of one person, maybe every adult should elect the members of the board. But direct election seems impossible. The institutions to make that happen do not exist and would take a long time to create. If created, the elections would likely have low turnout with dire implications for the legitimacy of the board.

Any decisionmaker faced with a similar situation tends to act on what might be called the stakeholder theory of representation. Facebook could determine which groups have a strong interest in content moderation by the company. They could then consult with these organized interests about who should serve on the board and then appoint them. The stakeholders would nominate while Facebook managers appoint the “justices” of the review board. Facebook might well see these appointments as representative of its users. The appointees would not work for Facebook and hence be independent in a sense. More realistically, if these selections were done correctly, Facebook would give its critics (and supporters) a seat at the appeals court.  Its critics might become more constructive or even supporters of Facebook’s content moderation.

But who would these appointed stakeholders represent? They would be suggested by groups with intense interests in Facebook’s content moderation. For them, the benefits of organizing to influence content moderation would outweigh its costs as noted in a famous book.  For most Facebook users, the opposite would be true; the costs of organizing would outweigh its benefits. As a result, the appointees would likely have atypical views about the meaning and application of Facebook’s Community Standards. In other words, representatives of stakeholders are unlikely to be representative of Facebook’s users.

Turning to stakeholders to help with a political challenge is natural. Above all, they are there, and you know them. Stakeholders do indeed offer a measure of representation, and perhaps also some independence from forces outside any organization. Indeed conflict among stakeholding members might enhance the board’s independence.  But the representation they offer is flawed. And perhaps, at this stage of institutional design, Facebook might look for alternatives that offer more in the way of both independence and representation.

Return Mail, Inc. is a small technological company that developed a and patented a system for processing returned mail after a failed delivery attempt, using optical scanners, computer databases, and other mechanisms. When it sought to enforce its patent against the United States Post Service (USPS), it knew that in the wake of the 2011 America Invents Act (AIA), the U.S. Patent and Trademark Office (PTO) could change its mind and conclude that the patent was granted in error and should have no further force. It also knew, however—or so it thought—that once the government made a decision regarding a patent, the government would be expected to speak with one voice. Instead, two different governmental agencies came to different conclusions and attempted to argue amongst themselves over Return Mail’s rights.

Article II of the Constitution vests the executive power in the president alone because the president is uniquely accountable to the entire American public.  Yet the USPS, although part of the government, operates independently of direct presidential control and is able to take legal positions that conflict with presidential directives and priorities. The Supreme Court has permitted the creation of such agencies, but it has never sanctioned these agencies to directly contradict presidential decisions and to seek the resolution of such disputes in the judiciary branch.

Such a system creates significant problems for the public in general, because it can never know who is actually speaking for the government and which directives it must comply with. It’s particularly problematic in the world of patents—which Congress from very early on determined must have uniform application throughout the country and charging a single agency with reviewing applications and a single court with hearing all appeals. Allowing myriad government agencies to reach their own conclusions on the meaning and scope of patents would undermine the system that Congress has taken pains to construct over more than 200 years (regardless of the proper scope of patents and other legislative reforms that may be worth pursuing).

Further, allowing the government to disperse the executive power between the president and independent agencies has a risk of undermining the due process protections that must be afforded to patentees before their patent rights are taken away. The Supreme Court has always maintained that patentees’ rights are protected by the essential guarantees of the Due Process Clause. Under the current system for administrative patent cancellation proceedings, the director of the PTO (who is responsible to the president and terminable at will) has the power to select administrative patent judges and assign them to particular cases. In cases where an executive agency seeks review of a patent issued by the PTO, the government—through that agency—has an interest in the outcome of the proceedings, but it is also the one sitting in judgment of the case. Our system of government has rejected procedures where one can be a judge in his own case. Allowing one government agency to challenge a patent while another government agency sits in judgment of that challenge would undermine more than 400 years of Anglo-American jurisprudence.

The Cato Institute, joined by Professor Gregory Dolin of the University of Baltimore, have filed an amicus brief urging the Supreme Court to hold that fidelity to our constitutional structure requires construing the AIA to apply solely to the resolution of disputes between private parties where the PTO remains a neutral and disinterested adjudicator and where politically accountable branches remain responsible to the citizens for the decisions that they have reached.

The case of Return Mail, Inc. v. U.S. Postal Service will be argued at the Supreme Court in the new year.

This blog post is part of a larger series on stock-market “short-termism”. See also my entries on share buybacks and progressive corporate governance reforms.

I. Introduction

To recapitulate the “myopia thesis”: managers of publicly traded firms are hostage to diversified shareholders who forego careful study of the firm’s fundamentals and instead respond to the latest, easily digestible quarterly earnings report. Rather than undertaking investments that might have a substantial return down the road, managers mimic the priorities of transient shareholders uninterested in a firm’s long-term strategy. Future-oriented firms that resist this temptation will find it more difficult to raise capital. This will then jeopardize their ability to survive long enough to reap the returns from long-term investments. 

The myopia hypothesis predicts that: 1) stock markets undervalue firms that sacrifice short-term profitability for longer-term growth 2) firms will therefore rationally forego long-term investments such as research and development (R&D) and capital expenditures (CAPEX). In this post, I will critically examine the evidence for such claims.

II. Profits, P/E Ratios and IPOs

Several economic indicators challenge the first pillar of short-termism thesis. In a recent NBER working paper, Steve Kaplan contrasts the early predictions of the myopia theorists in the 1980s with subsequent trends in corporate profits. If the short-termists of yesterday had been correct, the earnings posted by publicly held firms throughout the 80s and 90s would have evaporated over the medium-to-long term. Instead, we’ve since witnessed a steady upward march of corporate profits into now unprecedented territory.[1]

Moreover, while earnings have been on the rise over the past several decades, the price-to-earnings ratio (P/E) of the median firm listed on the S&P has risen even faster, currently at 25 compared to a historical median of 15.[2] A higher P/E ratio indicates that shareholders are valuing future earnings very highly. The very names that come to mind when one thinks of dynamic, future-oriented firms: Google, Apple, Amazon, Microsoft, Facebook all have exceptionally P/E ratios (~35), in many cases despite long periods of losses (Amazon netted more income in the final quarter of 2017 than in all 54 post-IPO quarters cumulatively). Investors act as if today’s unexceptional profits are going to grow substantially over the coming years and are pricing this optimism into current share prices. For these five firms, total profits of $101 billion in 2017 translated into a valuation worth 15% of the entire S&P 500.[3] The latest “Special Report” of the Economist magazine, while describing the potential monopoly threat that these tech giants pose, nonetheless concedes that their currently stratospheric valuations are in anticipation of future profits:

To justify its valuation, Facebook’s rate of “monetisation” will have to surge, suggesting that it extracts a bigger fee from other firms who want to reach consumers. To justify its $820bn market value, Amazon will have to increase its share of American retail to 12% (Walmart’s share today is 7%). Likewise Netflix will have to roughly double its nominal fee per user over the next ten years. Though tech firms’ profits as a share of gdp today are not extraordinarily large, Wall Street is predicting they will be in a decade’s time, with the median ratio for the five firms rising to 0.28%. That is above the 0.24% median level of Standard Oil, us Steel, at&t and ibm when they were each clobbered by antitrust regulators. The tech firms are expected to have higher returns on capital than the oligopolies of old, suggesting that they are better at extracting income per dollar of assets.

The Economist goes on to elaborate the ways in which prominent firms are leaving money on the table in the short-term to build future marketshare:

For Amazon and Netflix the rents flow in the other direction because their prices are low today: in total they subsidise their combined 240m paying subscribers to the tune of about $50 per person per year, based on the amount of additional free cashflow they would have needed to cover their cost of capital in 2017

Further corroboration can be found in the timing of IPOs. Notable examples such as Uber notwithstanding, the overall trend in IPOs has been toward more and more “premature” births of firms into public markets. The average profitability of a firm at the date of its IPO has been declining over the past several decades, as investors are willing to purchase shares in firms earlier in their life cycle.[4] Kaplan, citing IPO statistics by Jay Ritter, notes that just 4% of biotech firms that had an IPO between 2013-2016 were posting profits at the time. If publicly traded companies were indeed hobbled by myopic shareholders, this would present a massive arbitrage opportunity for investment vehicles operating on a longer time horizon. Venture capital and private equity funds are equipped for this purpose yet have not seen the abnormal profits or increased marketshare that one would expect if publicly traded firms were consistently failing to anticipate trillion dollar bills a few steps ahead on the sidewalk.[5]

III. R&D Up, Not Down, In Public Firms

The second key piece of evidence militating against the myopia thesis is that publicly traded U.S. firms, particularly those with the bubbliest valuations, are conducting more R&D than ever before, not less. Let’s begin by noting the strong upward trend in the overall private sector’s spending on R&D, which I’ve divided by total GDP (known as “R&D Intensity”):

Next, let’s look at R&D spending by the five aforementioned firms that shareholders are tripping over themselves to invest in:

The only way to reconcile the myopia hypothesis with the foregoing data is to maintain that these firms’ valuations would be even higher if they were spending less on R&D. For some reason, that counterfactual rings false. Post-recessionary declines in capital expenditures, on the other hand, seem superficially sympatico with short-termism, but can be adequately explained by reference to lower capacity utilization, and is in fact a worldwide phenomenon not restricted to the “overly financialized” Anglosphere.[6]

Let’s recall the mechanism by which short-termism is said to operate: diversified, rationally ignorant shareholders with short time-horizons turnover at high rates and punish future-oriented investments that incur immediate costs, such as R&D. We would expect, therefore, that publicly traded firms, whose shareholder profiles check those boxes, would suffer more acutely from myopia than privately held firms. Indeed, studies have found that publicly traded companies that go private register more patents post-transition, and, conversely, that privately held firms suffer a decline in patent quality after an IPO.[7][8] More directly to the point, a 2015 paper found that compared to privately held firms matched on a battery of relevant characteristics, public corporations engage in less net investment, and are less likely to capitalize on new investment opportunities as they emerge.[9]

A 2018 Federal Reserve Board working paper that uses corporate tax return data, comes to the opposite conclusion. Because private and public firms are subject to the same IRS filing requirements, the researchers were able to use the exact same measure of R&D investment when comparing the two groups, where previous studies had to impute measures of R&D for private firms. Moreover, whereas the Asker et al paper only had access to an unrepresentative sample of private firms and had to combine their measurement of capital expenditures with merger and acquisition activity as a rough proxy for net investment, the FRB study’s IRS data allows it to directly measure long-term investment, and to divide this measure into its physical and intangible (R&D) components. Beyond being far more granular, this dataset also captures the full universe of private U.S. corporations, allowing the researchers to compare a representative sample of private firms against their publicly traded counterparts, mitigating the selection effect that has plagued past comparisons. Not only do the Fed researchers find statistically significant differences between private and public firms, the magnitude of these effects is substantial:

…public firms invest…46.1 percentage points more in long-term assets than their private firm counterparts. It is not simply that public firms invest more relative to their asset base and thus out-invest private firms, they also direct a greater share of their investment portfolios to long-term assets. Public firms allocate 9 percentage points more of their total investment dollars to long-term assets than comparable private firms. The long-term investment advantage of pubic firms over private firms largely stems from their outsized investments in R&D. Public firms invest 39.2 percentage points more in R&D expenditures relative to physical assets, and dedicate 11 percentage points more of their investment budgets towards R&D than private firms[10].

Moreover, the researchers directly attribute this result to the difference in the shareholder profiles facing private vs. public firms:

The access to capital investment and the ability to spread risks among many small shareholders appears to facilitate heavier investments in R&D, arguably the riskiest of asset classes.

This result does not appear to stem from an omitted variable confounding the comparison between private and public firms. By exploiting the variation in R&D spending pre and post-IPO within the same firm, the researchers similarly note:

We find that public firms do not alter their short-term investment relative to physical assets following an IPO. These firms do, however, increase their long-term investments, and particularly investments in R&D: firms increase their R&D-to-physical asset ratios by 34.5 percentage points, and their R&D-to-total investment shares by 17.1 percentage points. Using an event study framework, we show that this increase in R&D expenditures occurs immediately upon IPO and persists for at least 10 years. We also examine changes in investment behavior following stock market delistings: results are less precise, but generally point to a reduction in R&D investments upon going private.

[1] Kaplan (2017)

[2] ibid

[3] Roe (2018)

[4] Fama and French, 2004; Ritter, 2016; cited in Kaplan, 2017

[5] Supra note 4

[6] Supra note 4

[7] Lerner, Sorensen and Stromberg (2011)

[8] Bernstein (2015)

[9] Asker, Ferre-Mensa, and Ljungqvist (2015)

[10] Feldman et al 2018

Late Friday afternoon, a federal judge in Fort Worth ruled that, because the individual mandate could no longer be upheld as a tax (because Congress in 2017 eliminated the monetary assessment for noncompliance), it was unconstitutional – and that it couldn’t be severed from the rest of the Affordable Care Act, so all of Obamacare is invalid. Fantastic, right? This is what I and many others have been working for since the law was signed in March 2010 and, while it took a while, we finally reached to the mountaintop – a second bite at the apple to undue John Roberts’s betrayal, right?

Well, not quite. Much as Judge Reed O’Connor’s ruling seemed to parallel the ruling by Judge Roger Vinson nearly eight years ago, in the litigation that culminated NFIB v. Sebelius in 2012 – Josh Blackman even evoked that early decision in a clever allusion to Groundhog Day – this time around there are different statutory facts being evaluated and so a different legal posture. 

Mind you, it’s absolutely correct that a “shared responsibility payment” that is $0 can no longer be justified as a tax, even under Chief Justice Roberts’s twistification. That is, a bare command to buy insurance is unconstitutional because it goes beyond federal power under the Commerce Clause and Necessary and Proper Clause (so ruled a majority of the Supreme Court, including Roberts).

But that’s not the end of the ball game because the question of whether the individual mandate can be severed from some or all of the rest of the ACA is a different one than whether the mandate itself is constitutionally kosher. Judicial doctrines of severability are somewhat complicated and call for judgment rather than bright lines, but they boil down to two questions: (1) Is the remainder of the statute “fully operative as a law”? and (2) Would Congress have passed the remainder? In Cato’s severability brief in NFIB, we argued that (1) “The individual mandate was essential to the Act’s scheme for achieving near-universal health care coverage at an acceptable cost” and (2) “Severing the individual mandate from its related provisions in Titles I and II will produce new comprehensive health care legislation that Congress did not enact and would never have enacted.”

But this time around, the Tax Cuts and Jobs Act of 2017 reduced the tax-penalty to $0 without eliminating so much as the guaranteed-issue and community-rating provisions (the parts most closely tied to the mandate), so (1) either the rest of the law would seem to be working (or not) irrespective of the individual mandate, and (2) we now have the scheme that Congress actually passed. In other words, Congress had the opportunity to sever as much of Obamacare as it wanted – legally speaking; there was only so much Republicans could do practically through “reconciliation” given the Democrats’ ability to filibuster more substantive legislation – and it effectively ratified the entirety of Obamacare with a $0 mandate.

So I’m quite skeptical that the severability ruling will be upheld on appeal, even by the conservative-friendly Fifth Circuit. There are also potential issues of standing, given that it’s based on compulsion to follow a law that has a $0 enforcement mechanism and no other legal consequences.

The case might not even get to the Supreme Court. And if it does, remember that the five justices who ultimately upheld the ACA are still on the Court. Plus Justice Brett Kavanaugh twice rejected challenges to the law when he was on the D.C. Circuit – albeit on technical grounds, and without engaging in severability analysis – which is why attacks on him from the left for wanting to gut Obamacare were so misguided.

In short, Friday’s ruling gave me a wistful thought about what might have been, but this case just isn’t the silver bullet that will finally kill the monster that has so damaged our health care system, economy, and rule of law. Indeed, I imagine that at a certain point we’ll stop talking about Obamacare – that point may have come when Senator John McCain voted against its “skinny repeal” last year – and just debate how best to reform “the health care system.”

For more analysis of Friday’s ruling in Texas v. United States, see Volokh Conspirators Jon AdlerIlya Somin, and special guest star Josh Blackman in two posts (with more to follow). It’s telling that I haven’t yet seen Randy Barnett, the intellectual godfather of the original individual-mandate challenge, opine on the ruling. I think he knows, as well as the rest of us, that this case isn’t the blockbuster some have made it out to be.

President Trump claimed last week that “people are pouring into our country, including terrorists.” This came after his unsubstantiated claim that Middle Easterners are traveling in the caravan. Center for Immigration Studies (CIS) fellow Todd Bensman has repeatedly defended these types of claims by equating immigrants from “countries of interest” with “terrorists.” This conflation is common and rarely challenged as Homeland Security officials and members of Congress frequently describe immigrants from these countries as a terrorist threat. Despite Border Patrol apprehending tens of thousands of foreign nationals from these countries of interest and many thousands more who have undoubtedly entered illegally, not a single person has been killed by a terrorist who entered as an illegal border crosser from any of the countries of interest.

Special Interest Alien Apprehensions

The terminology used to describe these immigrants varies considerably between sources. In 2011, the Department of Homeland Security (DHS) Inspector General defined the term “specially designated countries” to mean countries “that have shown a tendency to promote, produce, or protect terrorist organizations or their members.” Border Patrol Chief David Aguilar described “Special interest countries” as “basically countries designated by our intelligence community as countries that could export individuals that could bring harm to our country in the way of terrorism.”

These definitions could apply to nearly every country in the world, as just about every major country has “produced” or “exported” at least one terrorist. With several exceptions, the lists have consisted primarily of countries with large Muslim populations. The designated countries have changed repeatedly over the years:

In 2003, DHS released a list of 52 countries. In 2004, the list included 35 countries, two of which were new. In 2007, the list was referenced in a news article, and though the full list was not quoted, it included another country (Tanzania) that was not on either of the prior two lists. In 2018, DHS released yet another list of countries where CBP “had Enforcement Actions against aliens from the following ‘Special Interest Aliens’ countries for FY18.” This partial list included yet five more countries that were not on the prior lists. Altogether, these lists have contained 63 countries. Only 14 have shown up on all the lists.

From 2007 to 2017, Border Patrol apprehended 45,006 immigrants from any of the countries ever designated as a “country of interest” (See Table 1). During the same period, it apprehended 4,109 from countries that made it onto all three lists. Given the inconsistency in these lists and for sake of completeness, Figure 1 shows the annual number of special interest aliens apprehended by Border Patrol separated by the different lists and those apprehended from countries that appeared at least once on a single list.  Fiscal Year 2007 is the earliest year that Border Patrol has made the number of apprehensions by citizenship publicly available.

Figure 2 shows that only 0.85 percent of all Border Patrol apprehensions from 2007-2017 were special interest aliens.  This includes aliens from any country that ever appeared on any special interest countries list.  The other 99.15 percent, 5.25 million apprehensions, were for illegal immigrants from countries other than those that have ever appeared on a special interest countries list.

With 16,979 apprehensions, Indians were the most common “special interest aliens” apprehended from 2007 to 2017, followed by Brazilians with 12,925 apprehensions. Both countries were on the 2003 list, but not listed in 2004 or 2018. Bangladeshis were the most commonly apprehended “special interest aliens” who were on all three lists with 2,469 apprehensions.

No Terror Attacks on U.S. Soil

Zero people were murdered or injured in terror attacks committed on U.S. soil by special interest aliens who entered illegally from 1975 through the end of 2017.  However, seven special interest aliens who initially entered illegally have been convicted of planning a terrorist attack on U.S. soil.  They all entered illegally from Canada or jumped ship in American ports before the list of special interest countries even existed.  None of them successfully carried out their attacks and none illegally crossed the Mexican border.

Five of those seven illegal border crossers resided as illegal immigrants in the United States.  Walid Kabbani, a native of Lebanon, walked across the Canadian border illegally in 1987 to deliver a bomb to his co-conspirators in the United States.  He was discovered by a local police chief and arrested before he could carry out his attack.  Algerian-born Ahmed Ressam attempted to enter with false documents in 1999 on his way to attack Los Angeles International Airport as part of the so-called Millennium Plot.  U.S. border inspectors apprehended him, discovered his bomb in the spare tire well of his car, and then arrested him.  In the scuffle to detain Ressam, he broke free of U.S. law enforcement officers and ran into the United States before being apprehended a short time later.  Since Ressam technically entered the country unlawfully when crossing the Canadian border, we included him on this list. 

Algerian-born Abdelghani Meskini aided Ressam in his plot after he entered the United States illegally as a stowaway on a ship.  Palestinian Gazi Ibrahim Abu Mezer, who was born in Israel and traveled on Israeli papers, was apprehended at a bus stop after illegally entering the United States in 1997.  Somali-born Nuradin M. Abdi originally entered the U.S. unlawfully in 1995 on a fake passport.  While he did not cross the border unlawfully, he did so on a false passport and would have been blocked like Ressam was if his subterfuge was discovered by Customs agents.  In order to include the maximum number of possible terrorists so that we bias the results against ourselves, we included Abdi.    

From 1975 through 2017, a total of nine terrorists entered the United States illegally and only three did so along the Mexican border: Shain Duka, Britan Duka, and Eljvir Duka.*  They crossed as children with their parents in 1984 and were arrested as part of the planned Fort Dix terror attack that the FBI foiled in 2007.  The Dukas are ethnic Albanians from Macedonia – neither country has appeared on any special interest countries list.  The only terrorists who crossed the border with Mexico illegally did so as children, decades before becoming terrorists, and were not even from the special interest countries.

In addition to the five illegal immigrant border crossers from the special interest countries, two people from those countries entered illegally and applied for asylum.  We typically count these people as asylum seekers, but Bensman and others might include them as illegal border crossers.  As a result, we decided to include them here.  Although data is a little sketchy on these instances, Pakistan-born Majid Shoukat Khan and Shahawar Matin Siraj are two of the eleven asylum-seekers who probably initially entered illegally before asking for asylum.  They entered in 1996 and 1998, respectively.  They are both from a special interest country, neither of them committed an attack, neither injured or murdered anyone, and they both crossed the Canadian border.

Although there have been zero attacks committed by illegal border crossers from any of the special interest countries, foreign-born people who entered legally from those countries were responsible for 99.5 percent of all murders and 94.7 percent of all injuries committed by foreign-born terrorists on U.S. soil from 1975 through the end of 2017.  This isn’t surprising as the 9/11 terrorists are responsible for over 98 percent of all the murders and 87 percent of all of the injuries committed by foreign-born terrorists over this time.  Of the small number of foreign-born terrorists who committed attacks or were convicted of planning to do so on U.S. soil from 1975 through the end of 2017, the most successful strategy was to first enter legally.  There is no evidence that that pattern of activity has changed.     


So far, there have been zero people murdered or injured in terror attacks committed by illegal border crossers on U.S. soil.  This includes those who entered as illegal immigrants and those who entered illegally and then applied for asylum.  Only seven terrorists from special interest countries, all of whom entered prior to the government putting those countries on a list, even entered the U.S. illegally by crossing a land border.  Two of them were arrested within hours of doing so, two other received asylum, and none of them crossed the Mexican border. 

Our above evidence is based on past events.  The future could be different, but those who think that special interest aliens from these countries will enter illegally across the Mexican border and commit terrorist attacks here should present some compelling evidence before policymakers take them seriously.


*Numbers based on a forthcoming updated Cato Institute policy analysis.




That’s what the second author said about a new paper on Greenland’s ice, which arrived just in time for the annual meeting of the signatories of the UN’s 1992 treaty on climate change, this time in Katowice, Poland. Appearing in Nature, Rowan University Geologist Luke Trusel and several coauthors claimed ice-core data from Central-Western Greenland revealed melting in the recent two decades that has been “exceptional over at least the last 350 years.” The paper appeared in the December 6 issue of Nature.

How exceptional?

“Our results show a pronounced 250% to 575% increase in melt intensity over the last 20 years” as measured in four ice cores in west-central Greenland. Three of the cores were in the Jakobshavn Glacier, the largest-discharging glacier in the entire Northern Hemisphere. The Ilulissat icefjord, created by the glacier, some 25 miles in length, has historically calved nearly 50 cubic kilometers of ice per year into Disko Bay, near the town of Ilulissat. 

They then correlated their ice-core data with a model for ice behavior in all of Greenland. The correlations, while significant, were modest, with the explained variance of the island-wide melting maxing at around 36%. The melt reached its maximum in the very strange summer of 2012, where the amount at the Summit site, near Greenland’s highest elevation, was the largest since the summer of 1889—worth noting because that was well over 100 years ago.

There’s a long-standing quality weather station at Ilulissat, and it certainly shows summer warming of about 2⁰C from its beginning around 1850 to the 1920s.

For a broader comparison, we looked at the summer temperature anomalies for the 5 X 5 degree gridcell that includes Disko Bay and the icefjord. Because it is relatively hospitable and settled, there are a number of stations within the cell so the data is quite reliable. The data we show is from the Climate Research Unit at the University of East Anglia, version HadCRUT4.

There’s very little to see in this temperature record. The authors are well-aware of this and offer a rather unsatisfactory explanation:

The non-linear melt-temperature sensitivity also helps explain why episodes of mid-twentieth-century warmth resulted in less intense and less sustained melting compared to the last two decades, despite being only marginally cooler…Additional factors, such as recent sea-ice losses, as well as regional and teleconnected general circulation changes may also play a part in amplifying the melt response.

“Teleconnected” means things that happen in different locations but occur at the same time, without the reason for the connection being necessarily known.

What is known is that the correlation between summer temperature and melt in the cores on the glacier is relatively low (but statistically significant) with local summer temperatures only explaining 11% of interannual variance in melt. The melt data is noisy, but shows a spectacular increase in the last two decades for the area around Disko Bay.

Figure 1.  Regional temperature history for Disko Bay, which includes the Ilulissat icefjord. 

The “250% to 525%” increase actually refers to the two somewhat separated locations, one on the Jakobshavn Glacier and the other on the peninsula that forms the northern boundary of Disko Bay

The actual data show the baseline summer melt is very close to zero (and in some cases is below zero, i.e. a gain), for the vast majority of years back into the 18th century. Five times a number that is very close to zero is still a small number. It’s “off the chart” if the chart stops at very near zero and ends at number five times very near zero, as it does in the paper.

The sudden change in melt is nonetheless impressive. And it was certainly published at a time that would have maximum impact on the partygoers in Katowice.

It has some competition, though, in the other direction. In 2013, the Danish “NEEM” team successfully drilled a core through the previous (penultimate) interglacial and found that summer temperatures were 6-8⁰C warmer than the 20th century average for 6000 years. They estimated Greenland only lost 30% of its ice. That anomaly is known as the Eemian period and was around 118,000 years ago.

Humans can’t induce a summer warming of that magnitude and length simply because there’s not enough fossil fuel handy. And the math there is somewhat reassuring.  If all the ice came off Greenland, sea level would ultimately rise 23 feet. So 30% of that is 6.9 feet, spread out over 6000 years. That works out to a Greenland-induced sea level rise of 1.1 feet per millennium when it was much warmer.

Thanks to Ryan Maue for the Disko Bay analysis

The escalating tariffs imposed by the United States and China in recent months are making people and markets nervous, and just about everyone would like to see a deal of some sort. Over this period, prospects for a deal have gone up and down with the latest Trump tweet, but some reports suggest that the prospects now look promising. The current negotiating timetable is based on a 90-day period that Presidents Trump and Xi agreed to at their dinner in Buenos Aires on December 1 during the G20 talks. But what exactly would a deal look like? The Trump administration talks a lot about what it wants China to do, and the media mostly focuses on that. But what, if anything, will the administration give as part of this deal? Mike Santoli of CNBC Closing Bell asked  White House trade adviser Peter Navarro this question recently:

MIKE SANTOLI: Peter, you started out by saying how President Xi in 45 minutes gave a presentation showing a willingness to address all the United States’ concerns and then you say but everything else they have done essentially means that you’re very skeptical, I assume, of the likelihood of them delivering that. In exchange for what was the President suggesting that they would perhaps promise these reforms? In exchange for what from the United States?

PETER NAVARRO: …In terms of what we give back, I mean therein lies the rub. President Trump has been eloquent in stating that we are the piggy bank of the world. We’re in all these sorts of one sided bad trade deals and the problem that we have whenever we negotiate with whoever we negotiate is they are getting such a great deal they really don’t want to give us anything. So we are not prepared to give them anything in terms of a deal quid pro quo because we are so much behind the eight ball. We are not stealing their technology. We are not forcing the technology transfer. We are not manipulating our currency. We are not counterfeiting and pirating Chinese goods and flooding their markets. We are not having state owned enterprises run rampant around the world, basically exploiting the rest of the world. So what is there for us to give? What we have to give is access to our markets, period. The largest market in the world. Access to our financial markets, our capital markets. This is a great gift that we give to other nations. But we’re not going to do it anymore – President Trump’s made it clear – we’re not just going to give that away and be exploited.

In short, Navarro says that the administration will not be giving anything. Of course, even if he knew otherwise, he probably feels like he has to say this as part of the administration’s negotiating strategy. But regardless of what he or the administration says about this, the reality is likely to be different. Generally speaking, trade deals involve concessions by both sides. And with China in particular, given its “century long humiliation” of being pushed around by foreign powers, one-sided deals are going to be very difficult politically. And that seems to be how China is thinking about these negotiations, as a spokesperson for China’s Ministry of Commerce recently suggested: “In the next 90 days, China and the U.S. will negotiate on major issues of bilateral concern based on the principle of ‘mutual respect, equality, mutual benefit, and being mindful of each other’s concerns,’ aiming at the ultimate goal of removing all additional tariffs, and striving to reach consensus.”

So what might China want or expect to get out of a deal? In a Free Trade Bulletin we wrote last year, we suggested some possibilities:

… if this is to be a negotiation, China will have demands, too. One of these is likely to be that the United States begins treating China as a market economy in its antidumping calculations. This will be difficult for the United States to accept. However, if the United States can get enough concessions in other areas (such as SOEs and overcapacity in steel production) and can impose some additional disciplines on China’s own abuses of antidumping measures, perhaps it could go along with this demand.

Similarly, there has been talk of revising the review of foreign investment carried out by the Committee on Foreign Investment in the United States (CFIUS) so as to broaden its scope. Chinese companies have previously challenged CFIUS decisions blocking their investment, and expanding the scope of this review is likely to be of concern. As part of a negotiation on trade and investment, the United States could commit to maintaining the existing focus for reviewing Chinese investments and keeping the review process from being abused.

Since we wrote that piece, the enactment of the Foreign Investment Risk Review Modernization Act (FIRRMA) has provided a legal framework to expand, rather than limit, the scope of CFIUS review. However, some of the details are still to be determined, and the U.S. could apply the law in a narrow way so that the process will not impede investment unreasonably. Relatedly, the U.S. could relax some of its restrictions on selling high-tech products to Chinese companies, in areas where national security is not at issue. These are just a couple ideas, and there are many other possibilities.

We are more skeptical of a significant deal than some commentators seem to be, but we do think there is a chance, although it will almost certainly take longer than 90 days. But the Trump administration needs to offer something in exchange for any concessions by China. If the administration is not anticipating giving something to make a U.S.-China deal look balanced, the prospects for a deal do not look very good at all. 

The national Cato 2018 Paid Leave Survey of 1,700 adults finds widespread support for creating a federal paid leave program, with 74% in favor. However, 69% of Americans would oppose establishing a federal paid leave program if it meant that fewer women would get promoted and become managers. But would establishing a federal program actually do this? Research suggests that it could and that’s why we asked about it on the survey:

Read about the full survey results and methodology here.

First, let’s consider the different career outcomes between women in the United States and women in Western Europe and Scandinavia. Academic research finds that American women are more likely to rise up the corporate ladder than their European counterparts who have access to generous family social welfare programs. 

An analysis of OECD countries reveals that American women are about 3 to 14 times as likely as Scandinavian women to be employed as managers with 14.6% of American women who are managers compared to 4.6% of Norwegian, 4.2% of Swedish, and 1% of Danish women. American women are also more likely than women in France (5.1%), the United Kingdom (7.8%), Germany (2.7%), and the Netherlands (3.6%) to hold managerial positions. 

Another measure of corporate success is the share of women who serve on company executive committees—these include the CEO and those who directly report to the CEO. The 20-First’s 2018 Global Gender Balance Scorecard finds that a majority—53%—of American companies have three or more women on executive committees.(1) In contrast, only 14% of European companies have three or more women serving on company executive committees. Some argue that once a company starts to have three or more women on executive committees, the company culture changes to see women in leadership roles as the norm rather than the exception.

But correlation is not causation, as every good statistics student knows. Just because Western European and Scandinavian countries have generous paid leave programs and family policies doesn’t mean it necessarily causes fewer women managers in Europe.  However, academic studies suggests that it might.

Studies have found that government-provided paid leave may lead to fewer women getting promoted and becoming managers because they take longer leaves than they otherwise would. Other studies have noted that employers, particularly smaller companies that have difficulty accommodating workers taking leave, may be less willing to hire female employees to begin with. Some argue that American women’s corporate success may be due to the fact that the United States does not provide generous family leave policies like Europe.

Because of this research, the Cato 2018 Paid Leave Survey examined what Americans think about establishing a federal paid leave program if it had harmed women’s careers.

The survey finds that voters would not be willing to sacrifice the progress women have made in the workplace if that were the cost of implementing a federal paid leave program: 69% would oppose while 29% would support making this trade-off.

A majority (57%) of mothers of children under three also oppose the idea if federal paid leave were to harm women’s careers. Majorities of Democrats (63%), independents (68%), and Republicans (77%), men (67%), and women (71%) all would oppose establishing a federal paid leave program if it meant fewer women would get promoted.

These data indicate that policymakers have good reason to evaluate the academic research on the effects of government-supported paid leave programs on the corporate leadership gender gap before inaugurating this new program. 

For public opinion analysis sign up here to receive Cato’s upcoming digest of Public Opinion Insights and public opinion studies.


Read about the full survey results and methodology here.

The Cato Institute 2018 Paid Leave Survey was designed and conducted by the Cato Institute in collaboration with YouGov. YouGov collected responses online October 1-4, 2018 from a national sample of 1,700 Americans 18 years of age and older. Restrictions are put in place to ensure that only the people selected and contacted by YouGov are allowed to participate. The margin of error for the survey is +/- 2.4 percentage points at the 95% level of confidence.


(1)The number for American companies is based on 37 out of 38 being companies based in the United States and 1 from Brazil.

This story from WAPO sent a chill down my spine:

Until quite recently, Zamira Hajiyeva was living the high life, according to British authorities. She had a $15 million townhouse in London’s tony Knightsbridge neighborhood, a golf club in the English countryside and a gold-plated shopping habit at Harrods.

That was before a British court this year asked the 55-year-old from Azerbaijan an impertinent question: How did she afford those purchases?

That query is at the heart of a bold British push to try to reverse what the government believes is a flood of foreign investment stemming from overseas corruption and criminality.

In other words, the U.K. is shifting the burden of proof onto foreign investors; they must now prove their wealth is legitimate.

No doubt, some foreign investestments into the U.K., U.S., and elsewhere are attempts to launder ill-gotten gains.

But the slippery slope implications of the U.K.’s new approach are terrifying.

To start, governments can expand this “presumption of guilt” to its own citizens, not just foreign investors.  Can you prove all your wealth is from legitimate sources?  Unless you have kept scrupulous records your entire life, probably not.  And even if you can, doing so is likely an accounting and legal nightmare.

And where would this presumption of guilt stop? Will we someday have to prove we do not use outlawed drugs?

The presumption of innocence is fundamental to a free society.  The U.K. is playing with fire.


As the Cato Institute continues to press the case for Jones Act reform, defenders of this flawed and failed law have repeatedly made clear that they’ve taken notice. Fresh evidence of this was seen earlier this month with the publication of an op-ed on the leading maritime website Entitled “CATO’s Continued Attempt to Skin the Jones Act,” the piece was an obvious preemptive salvo launched a day prior to Cato’s recent conference on the law’s shortcomings. A close reading, however, reveals it to be another instance of Jones Act defenders missing the mark.

Examining the law’s history, author Sal Mercogliano—a professor at Campbell University—claims that prior to the outbreak of World War II that “the Jones Act, reinforced by the Merchant Marine Act of 1936 ensured that not only was there a domestic shipbuilding industry, but it could be ramped up to support the building of over 5,000 merchant ships…” This is, at best, incomplete. As the book Global Reach points out, during this time U.S. merchant shipbuilding was almost non-existent and the fleet itself in obvious decline:

By the mid-1930s American merchant shipbuilding had come to almost a complete halt. In the nearly twenty years following the end of World War I, America’s merchant fleet, including its cargo and passenger ships, was becoming obsolete and declining in numbers; nearly 90 percent of the merchant fleet was more than twenty years old, and few ships could do more than ten or eleven knots. Although the Maritime Commission established by the Merchant Marine Act of 1936 had planned to build 50 ships a year under its CDS provisions, by 1939 the United States had only about 1,340 cargo ships and tankers, fewer than the total built by U.S. shipyards in 1914-17, even accounting for wartime losses. In no respect was the U.S. commercial industry capable of meeting the demand for sealift posed by the looming conflicts in Europe and the Pacific.

It’s also worth pondering why, if the Jones Act should be regarded as a success, the Merchant Marine Act of 1936 was even needed.

Dr. Mercogliano’s explanation of stupefying U.S. shipbuilding costs—commonly estimated to be three to five times greater than those of Asian shipyards—similarly leaves much to be desired:

CATO contends that the Jones Act is a burden that American can no longer bear. Specifically, they cite the higher cost to build ships in America as opposed to overseas. The largest builders of commercial ships in the world today are the Republic of Korea, the People’s Republic of China, and Japan; nine out of every ten ships afloat are built in East Asia. The question that needs to be raised is why? It is the exact reason that the CATO Institute rails about with the United States – government subsidies. The South Korean government announced the injection of over $700 million dollars into Hyundai Merchant Marine to stabilize the largest Korean shipping line. It was announced that the South Korean government would be infusing over $1 trillion into shipbuilding, in violation of the World Trade Organization.

While it is true that Asian shipbuilders receive government largesse, both the amount and its alleged explanatory power are vastly overstated. The $1 trillion figure cited—a stunning two-thirds of South Korea’s GDP—has no basis in reality. In fact, a recent WTO case brought by Japan against South Korea over its shipbuilding subsidies uses the figure of $11 billion in financial assistance.

The main reason why U.S. shipyards are so wildly expensive is not subsidies but scale. Asian shipyards have it, while U.S. shipyards—where annual output of tankers and cargo ships is typically in the single digits—do not. As this 2015 National Defense University (NDU) report on the U.S. shipbuilding sector explains:

Many of the successful foreign shipbuilders enjoy significant economies of scale from commercial market share dominance. The Jones Act does not deliver the market share necessary for the United States to achieve competitive economies of scale and, therefore, the United States will likely never be competitive in the global commercial market.

A 2009 NDU report similarly held the Jones Act at least partly responsible for U.S. shipyards’ lack of competitiveness:

Regarding the international market for commercial ships, U.S. shipbuilders face steep competition from shipbuilders in Asia who offer lower prices, are more efficient, and have higher industry best practice ratings. This is particularly true for the construction of vessels over 1,000 tons. This can be partially attributed to protectionist policies, such as the Jones Act, that have shielded domestic shipbuilders from the pressures of global competition. Thus while U.S. shipbuilders have remained competitive within the U.S. market, they were less so compared to foreign shipyards. None of the shipyards that the Industry Study Team visited expressed confidence that U.S. shipyards, as they are currently configured, could compete effectively in the global shipbuilding market.

Furthermore, the United States is hardly innocent in this regard. As yet another NDU report points out, the Philly Shipyard alone has received “approximately $500 million in direct subsidies and leases its facility from the city for only $1 per year” while the Austal shipyard in Alabama and VT Halter shipyard in Mississippi have received millions more.

There’s also the small matter that the Jones Act itself, which mandates the purchase of U.S.-built ships, functions as a massive de facto subsidy scheme.

And for all of the finger pointing at foreign subsidies, it’s instructive that when the Clinton administration reached an international deal for an end to shipbuilding subsidies—one that included Japan and South Korea—U.S. shipbuilders performed an about face and turned it down. One would expect a rather different response if subsidies were indeed the decisive factor preventing these shipbuilders from competing with their foreign counterparts.

Mercogliano later goes on to discuss the Jones Act’s national security justification and the heavy U.S. reliance on foreign-flagged shipping during Operations Desert Shield and Desert Storm. While conceding that, despite having the Jones Act in place, the conflict “highlighted a shortfall in American sealift capacity,” Mercogliano seems to argue that this had largely been remedied by the time of the 2003 Iraq War with “all the cargo shipped to that conflict [going] on ships crewed by American merchant mariners.”

Other sources, however, paint a rather different picture. According to a 2009 Naval War College paper the shipping situation was in fact arguably even more dire than during Desert Shield/Storm:

With 77.6 percent of cargo being moved by United States government owned vessels, TRANSCOM still had to turn to the United States and foreign flagged merchant fleets during the deployment of OIF. In fact, since the United States flagged merchant fleet continued to decline between Desert Shield/Desert Storm and OIF, United States flagged merchant vessels delivered only 1.3 million square feet or a mere 6.3 percent of OIF deployment cargo. Therefore, even with TRANSCOM’s high priority of funding after Desert Shield/Desert Storm they were still required to use foreign flagged merchant vessels to move 3.3 million square feet or 16.0 percent of OIF deployment cargo.

While this looks like an 11 percent improvement from Desert Shield/Desert Storm, OIF required 12.1 million square feet less of cargo. Therefore, if the required cargo to be moved was equal to that of Desert Shield/Desert Storm then foreign flag vessels would certainly have been used to carry the majority of the additional cargo. In fact, using a conservative estimate of foreign flag vessels picking up 50 percent of the difference in cargo between the two conflicts would have brought the percentage of cargo carried by foreign flag vessels to 28.6 percent, one percent higher then (sic) during Desert Shield/Desert Storm.

Since this time the Jones Act fleet has only further deteriorated. When Operation Iraqi Freedom began in 2003 the number of Jones Act ships stood at 151. Today it is a mere 98.

None of this is to suggest that the piece is entirely devoid of useful facts or information. Mercogliano, for example, notes that the Jones Act’s namesake, Sen. Wesley Jones of Washington state, was “particularly interested in the cabotage provision between his home state and the Alaska territory.” This does not surprise, with the Jones Act resulting in two Canadian shipping companies that served Southeast Alaska driven from that market—a development that almost surely benefited competitors in Jones’s home state.

Refreshingly, Mercogliano also acknowledges that, should the Jones Act be repealed, “Americans could see a reduction in their freight rates and the opening of new water routes between American ports.”

Perhaps most importantly, while he uses his conclusion to oppose repeal of the Jones Act, Mercogliano concedes that reform is needed stating, “Does the Jones Act need reform? Yes, what 98-year-old law does not need updating.” We may differ on the exact reforms required, but acknowledging that not all is well with the Jones Act is an important first step. 

Today, the Department of Homeland Security (DHS) released a report detailing deportations (henceforth “removals”) conducted by Immigration and Customs Enforcement (ICE) during fiscal year 2018.  This post present data on removals in historical context – combined with information from Pew and the Center for Migration Studies on the number of illegal immigrants present in the United States.

ICE deported 95,360 illegal immigrants from the interior of the United States in 2018, up from 81,603 in 2017.  Removals from the interior peaked during the Obama administration in 2011 at 237,941 (Figure 1).  The Trump administration would have to increase the pace of interior removals dramatically to reach Obama’s previous peak.  Unless something dramatic changes, that won’t happen as local law enforcement agencies are much less likely to cooperate with President Trump’s ICE than they were with President Obama’s ICE.   ICE also removes large numbers of illegal immigrants apprehended at the border.  Since 2012, border removals have outnumbered those from the interior of the United States.


Figure 1

Interior and Border Removals by ICE, 2008-2018

Source: Immigration and Customs Enforcement.


The Obama administration removed 1,242,486 from the interior of the United States during its full eight years, averaging 155,311 removals per year.  Data from the earlier Bush administration are more speculative, but they show an increase in deportations during the last half of Bush’s administration that continued during Obama’s first term before flattening and, finally, dropping rapidly in his second term.    

The percentage of all illegal immigrants removed from the United States is a better measure of the intensity of interior enforcement than the total numbers removed.  Based on estimates of the total size of the illegal immigrant population from Pew, ICE removed about 0.89 percent of the illegal immigrant resident population from the interior of the United States in 2018, up from 0.76 percent in 2017.  Interior removals as a percent of the illegal immigrant population peaked at 2.11 percent in 2009. 


Figure 2 

Removals as a Percent of the Illegal Immigrant Population


Sources: Immigration and Customs Enforcement, Pew, and Author’s Estimates.


ICE under President Obama’s administration removed an average of 1.38 percent of the interior illegal immigrant population per year of his presidency.  The Obama administration’s interior removal statistics show a downward trend beginning in 2011 and continued until the end of his administration.  So far, ICE under President Trump has only managed to deport an average of 0.83 percent of the illegal immigrant population each year.   

The Obama administration also focused immigration enforcement on criminal offenders (not all illegal immigrants are criminals).  During the Obama administration, 52.6 percent of all illegal immigrants removed were convicted criminals, including those convicted of immigration crimes.  About 57 percent of those deported during the Trump were convicted criminals. 


Figure 3

Criminal Removals as a Percent of All Removals


Source: Immigration and Customs Enforcement.


The Trump administration is continuing to ramp up interior immigration enforcement.  The 2018 figures show a reversal of the declining interior immigration enforcement efforts under the Obama administration, but they have not reached Obama’s previous deportation records.  Trump is unlikely to come close, but he will continue to try.


President Trump’s administration has increased immigration enforcement at worksites, just as he promised.  Immigration and Customs Enforcement (ICE) arrested 2,304 people at worksites in the fiscal year 2018, a more than 7-fold increase from the previous fiscal year and about 6.7-fold more than the last full year of the Obama administration. 

ICE’s worksite arrests fall into two categories: criminal or administrative.  Any citizen or noncitizen whom ICE suspects of having committed a criminal violation, such as identity fraud, can be arrested.  Administrative arrests are for civil violators of the Immigration and Nationality Act (INA), which means that only non-citizens can be arrested for such violations.  An administrative arrest is usually the first step toward deportation for an immigrant.  ICE doesn’t make most of its initial administrative arrests as it relies on other law enforcement agencies to arrest illegal immigrants, but it does directly arrest those at worksites in these cases.

Figure 1 shows the rapid and recent increase in all ICE criminal and administrative arrests.  The number of arrests peaked in 2008 at 6,287 after several steady years of increases, but then declined during Obama’s presidency with a single blip in 2011.

Figure 1

All ICE Arrests at Worksites

Sources: Immigration and Customs Enforcement and Congressional Research Service.

President Trump’s ICE is making more administrative arrests and a greater percentage of those come from worksite enforcement (Figure 2).  In 2016, only 0.35 percent of all of ICE’s administrative arrests (those that ICE makes itself) were at worksites.  In 2018, 3.7 percent of all of ICE’s non-custodial arrests were at worksites.  Since 2017, administrative arrests are up 8.9-fold compared to a 5.6-fold increase in criminal arrests over the same time.  As a share of all ICE administrative arrests, those conducted at worksites are up about 10.7-fold over 2016, the last year of the Obama administration, and 8.8-fold since 2017.      

Figure 2

Worksite Administrative Arrests as a Percent of All ICE Administrative Arrests

Source: Transactional Records Access Clearinghouse, Immigration and Customs Enforcement, and Congressional Research Service.

The I-9 worksite audit was a major innovation in immigration enforcement unrolled during Obama’s administration, which Trump’s administration is using with alacrity (Figure 3).  Although the number of worksite audits increased about 4.4-fold from 2017 to 2018, from 1,360 to 5,981, only a small fraction of all establishments were actually audited.  The Business Dynamics Statistics (BDS) at the U.S. Census reports that there were more than 6.8 million business establishments in 2016, the last year for which data is available.  Not all of those 6.8 million establishments have employees besides the founder and, since he doesn’t have to fill out an I-9 form, not all of them would be subject to audits.  With the aid of a back of the envelope estimate, I assume that about 3.9 million of those establishments had more than one employee in 2016.  Thus, it’s unlikely that the 5,981 ICE I-9 audits in FY2018 covered more than 0.15 percent of all business worksites. 

Figure 3

I-9 Audits

Sources: Immigration and Customs Enforcement and Congressional Research Service.

ICE clustered its I-9 audits by time in 2018.  It sent out 5,278 (88 percent) of the 5,981 notices of inspection for the audits during the two periods of January 29 through March 30 and July 16 through July 20.  Those big operations were meant to frighten illegal immigrants and their employers.  Since ICE only audited about 0.15 percent of establishments, it has to rely on fear to increase compliance with immigration laws at worksites.

The Trump administration was supposed to reduce expensive economic regulation.  In many other sectors of the economy, Trump has followed through on this promise.  Immigration is the exception. 


Nearly 8 in 10 Americans (78%) support creating family and medical leave savings accounts. The national Cato 2018 Paid Leave Survey of 1,700 adults finds the public favors allowing workers to set aside money in tax-advantaged savings accounts that could be used if they need to take family or medical leave. A fifth (20%) would oppose the creation of family and medical leave accounts.

Read about the full survey results and methodology here.

Establishing family leave savings accounts enjoys rare bipartisan support: 82% of Democrats, 80% of Republicans and 69% of independents support offering tax-advantages to people who set aside money for parental, family, or medical leave. 

Women (80%) and men (76%) also overwhelmingly agree about establishing these types of accounts. These numbers include 85% of Democratic women, 82% of Republican women, and 78% of Democratic and Republican men. 

Support for parental and family leave savings accounts aren’t reserved for the wealthy. Nearly three-fourths (73%) of those earning less than $25,000 a year also support such savings accounts. Support heads upwards from there with 86% in favor among those earning $80,000 a year or more.

Finding an image of a piggy bank or savings jar with “family leave” or “parental leave” written on the side to compliment this post was difficult. Instead, one can easily find images of piggy banks and savings jars with labels to save for retirement, to buy a house, for college as well as future travel and stocks. For these needs and wants, we’ve cultivated a culture of saving.

Over time our society has fostered a set of social norms that we instill in our young people starting at an early age. We were reminded when we were young, just as we remind the next generation, to start saving early for an education, a house, big purchases, emergencies, investments in the future, etc. We were reminded to delay consumption today so we have more for later. Doing so can provide peace and security to feel like we’ve done our part to prepare for the future. People usually don’t enter the world with a set of expectations about the value of saving and what they should save for. Thus it’s useful to establish social norms that encourage saving early on.

However, a gap exists in our present cultural norms when it comes to saving for parental and family leave. The fact that I could not easily find an image of a piggy bank or savings jar with “parental leave” or “family leave” as a savings goal is indicative of our culture not yet establishing a norm for these savings goals. While family leave savings accounts need not crowd out other innovations to help support working people who take time to care for new children, family members, or themselves, it can be part of the discussion. These poll results suggest society may be ripe for establishing new social norms of saving for parental and family leave.



As the $867 billion farm bill goes to the president for his signature, the Congressional Budget Office reminds us why farm subsidies don’t make much sense.

CBO just released a new “options” report that includes more than 300 pages of ideas for reducing budget deficits. The report suggests ways that Congress could cut farm subsidies, and it describes how the world has changed since these programs were put in place in the 1930s:

During the Great Depression of the 1930s, the 25 percent of the population that lived on farms had less than half the average household income of urban households; federal commodity programs came about to alleviate that income disparity.

One argument for eliminating Title I commodity support programs is that the structure of U.S. farms has changed dramatically since then: The significant income disparity between farm and urban populations no longer exists. In 2014, about 97 percent of all farm households (which now constitute about 2 percent of the U.S. population) were wealthier than the median U.S. household. Farm income, excluding federal program payments, was 52 percent higher than median U.S. household income.

Moreover, payments made through programs that support commodity prices and incomes are concentrated among a relatively small portion of farms. Three-quarters of all farms received no farm-related government payments in 2014; most program payments, in total, went to mid- to large-scale farms (those with annual sales above $350,000).

Title 1 refers to ARC, PLC, and other programs that shovel billions of taxpayer dollars to the growers of corn, soybeans, wheat, and other crops.

Let me reiterate:

  • About 97 percent of all farm households are wealthier than the median U.S. household.
  • Farm income was 52 percent higher than median U.S. household income.
  • Subsidies are slanted toward the largest farms.

Further arguments against farm subsidies are here, here, and here.