Cato Op-Eds

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

On Monday, July 24, the House will consider the Fiscal Year 2018 Intelligence Authorization Act under suspension of the rules in an attempt to fast-track the legislation, which contains some significant “Russiagate”-related provisions. 

Section 501 calls for a new Intelligence Community assessment “of the most significant Russian influence campaigns, if any, conducted during the 3-year period preceding the date of the enactment of this Act, as well as the most significant current or planned such Russian influence campaigns, if any.” Significantly, the classified report, which is due 60 days after enactment, is to also have an unclassified summary, meaning the public may learn still more about exactly when alleged Russian efforts to influence the 2016 presidential election began.

What may be lacking, as was the case with the IC assessment published in January 2016, is any new or meaningful, specific declassified intelligence that actually validates IC claims that the Russians were, in fact, responsible for the interference. Ironically, it has been yet another IC leaker—Reality Winner—who has provided us with the most interesting technical assessment of alleged Russian election-related activities. 

Section 502 of the bill mandates interagency reports on potential future threats:

(1) IN GENERAL.—As provided in paragraph (2), for each Federal election, the Director of National Intelligence, in coordination with the Under Secretary of Homeland Security for Intelligence and Analysis and the Director of the Federal Bureau of Investigation, shall make publicly available on an internet website an advisory report on foreign counterintelligence and cybersecurity threats to election campaigns for Federal offices. Each such report shall include, consistent with the protection of sources and methods, each of the following:

(A) A description of foreign counterintelligence and cybersecurity threats to election campaigns for Federal offices.

(B) A summary of best practices that election campaigns for Federal offices can employ, in seeking to counter such threats.

(C) An identification of any publicly available resources, including United States Government resources, for countering such threats.

To be truly effective, such an approach would almost certainly require the declassification of some fairly specific intelligence on alleged or actual Russian practices. It will very interesting to see exactly how much push-back this provision gets from the ODNI, DHS, and FBI.

Section 503 calls for a classified report on “containing an assessment of the financing of threat activity by the Russian Federation” which is also due 60 days after enactment.

Given how legislatively dysfunctional Congress has been this year overall (the annual National Defense Authorization Act was nearly two months behind its usual House floor schedule this year), it’s unclear whether the House and Senate will be able to agree on these provisions and actually get the Intel Auth bill to Trump for his signature before the year is out—or whether Trump will threaten a veto over any Russia-related provisions in any final bill. One thing is clear: that this bill is on the suspension calendar means that GOP House Intelligence Committee “Russiagate” investigative lead Mike Conaway (R-TX) and his Democratic counterpart, Adam Schiff (D-CA) are apparently on the same page about these issues. And for a president desperate to move past the “Russiagate” quagmire, that’s bad news indeed.

“Are Shoe Brands Charging You More Based On Your Gender?” asks Footwear News. The website cites data compiled by Datafiniti showing that high-end footwear brands tend to price women’s shoes higher than men’s. As examples, the median price of a Gianvito Rossi pair of women’s shoes was $750 versus $469 for the gent’s.

Such data has become increasingly of interest following the New York City study entitled “From Cradle to Cane: The Cost of Being a Female Consumer,” which was then replicated by The Times newspaper in London, both purporting to show that women were discriminated against in pricing terms for a range of everyday products. Particular examples in the surveys included children’s scooters, women’s hair products and razors. This comes alongside previous studies from the University of Central Florida and Consumer Reports, which found that women tended to pay significantly more for deodorant, haircuts, and dry cleaning.

Does all this show “sexist” pricing and discrimination against women?

Here’s some economics to bear in mind which might explain why the prices of goods might differ in these types of surveys:

  1. There can be cost of production differences in some cases: often men’s clothes and shoes tend to be more standardized than those targeted at, and bought by, women. This explains why both women’s clothing and dry cleaning of that clothing can be more expensive too, given the clothes cannot be handled in the same way. It does not seem a stretch of the imagination to suggest that women’s hair cuts tend to be more expensive as a function of length, time, and the variety of tastes as well.
  2. Though things like different colored razors have the same functional uses, the branding and designs are themselves part of the product. It might be that many women place inherent value on some products being branded as “for women” or else enjoy the scent, colors etc. Indeed, there is nothing to stop women buying male or unisex branded products (some women I know do), suggesting the continued existence of gendered variants is evidence of demand. Prices that maximize profits will be higher for women than men if women are less responsive to price (i.e. have more inelastic demand). This might be because of greater attachment to brand, or because women perceive products tailored to women as being of higher quality.
  3. For some products targeted at women, women may buy fewer units of any given brand relative to men (on products where women prefer variety), which would (other things given) lead to higher prices too.
  4. The surveys themselves may be biased towards standardized products with male and female versions. Tyler Cowen speculates, for example, that women might (on average) have a relative taste for variety and quality on these type of low value products, whereas men tend to spend more cheaply on these but more on products not differentiated by gender (such as automobiles, electronic sound systems etc).
  5. The example of nightclub cover charges, where women frequently obtain “free” entry, shows that for some goods and services men tend to pay more than women. Even in the shoe study, the data shows that the median price for moderately priced sneakers tends to be higher for men than women’s shoes. A pair of Nike women’s shoes was $80 vs. $85 for men. Some of the above economics might explain this (particularly the elasticity explanation), but with things such as nightclub nights the charge may help cover for probability of externalities – i.e. drunken violent behavior. Another example of this is vehicle insurance, where men tend to pay more than women for near identical products due to higher costs associated with a tendency towards more risky behaviors for men.

As seen then, there are many different rational explanations for why different prices might be charged for products targeted at different genders. This is certainly not an area where there are obvious “market failures” which justify government intervention.

The implicit criticism by some of this practice seems to be opposition to the mere existence of “gendered” products at all. Indeed, in a debate I had on the subject in the UK, my female opponent claimed that women did not want products to be “gendered.” If this is true, it suggests there is an exceptional entrepreneurial opportunity for companies to develop more unisex products. 

MacRumors has a piece out today noting that Apple has raised its lobbying game in Washington over the last six months, spending $3.6 million on a team of lobbyists who’ve visited House and Senate offices on issues ranging from “general patent reform” to “green technology” to “issues related to implementation of Section 1502 of the Dodd-Frank Act.” What’s missing from the lobbying disclosure form is any mention of federal government surveillance practices, whether it be Section 702 of the FISA Amendments Act or that nasty encryption-related battle Apple had with the FBI in the wake of the San Bernardino shooting in 2015. 

As Reuters noted earlier this month, the tech industry generally has been rather quiet about FISA reform, though members of the Reform Government Surveillance consortium (of which Apple is a member) like to point to a letter they sent to key Congressional committees earlier this year as evidence of their committment to getting NSA and the FBI to clean up their acts on domestic surveillance. But as the old saying goes, talk is cheap.

Apple, as the richest and most successful tech company in human history, certainly has the resources to make it’s lobbying campaign–or even a surveillance reform-focused PAC–far more robust and politically threatening to pro-Surveillance State House and Senate members. That it has declined to do so to date is telling. Until Apple and the other members of the RSG make it clear to House and Senate members that there will be a steep political price to pay for failing to rein in NSA and the FBI, don’t expect significant domestic surveillance reforms to make it into law.

The federal government funds an array of aid programs aimed at promoting growth in less-developed countries. Funding goes to federal agencies, such as the U.S. Agency for International Development, and it also goes to international aid groups, such as the World Bank.

The federal government spends more than $30 billion annually on foreign aid. Aid spending has more than doubled in the past two decades. The Trump administration has called for substantial cuts to foreign aid spending.

Ian Vásquez has published a new study on foreign aid at He argues that despite the political enthusiasm for aid, there is little evidence that it is effective. Indeed, it is often counterproductive. There is no correlation between foreign aid and economic growth, and efforts to condition the receipt of aid on market reforms have failed. As such, Ian argues that the federal government should end its development aid programs.

However, Ian also discusses the good news on economic development. Decades of experience show a strong relationship between economic growth and market-oriented policies. Countries eager to improve living standards can do so themselves with domestic reforms, such as securing property rights, reducing arbitrary and bureaucratic regulations, and encouraging entrepreneurship and investment.

Ian’s new study on foreign aid is here.


As I argue in my recently published policy analysis here at Cato, the American-led war on terror has clearly failed. Unfortunately, rather than accept the obvious fact that the campaign was badly misguided and focusing homeland security efforts in more fruitful areas, the Trump administration appears ready to embrace, and perhaps even to escalate, the American commitment in the Middle East. Though President Trump himself has frequently voiced concerns about nation building in Iraq and the mission in Afghanistan, few of his senior advisers appear to share his worries. And sadly, few voices from the foreign policy establishment have questioned the need for continued American intervention.

The near total lack of debate begs a simple question: Why do so many smart people support the continuation of a strategy despite its abject failure over sixteen years and in the absence of anything even remotely approaching a new theory of victory?

Though there are undoubtedly many different contributing factors, one important cause is the influence of several mutually reinforcing fallacies about terrorism and the use of force.

The first of these is the “political will” fallacy. This is the misguided idea that the United States can outlast the Taliban, Al Qaeda, ISIS, and other local actors simply by illustrating sufficient political resolve. Once the terrorists and insurgents understand that the United States is truly  “in it to win it” they will admit defeat. The reality, however, is that resolve is not something the White House can create. Resolve is a force that stems from how meaningful the objective is to a nation and how much its people are willing to pay to achieve it.

Given this, America’s adversaries clearly enjoy a decided advantage. Local actors like the Taliban have a tremendous stake in the outcome in Afghanistan – it is their home, after all. Americans, on the other hand, are rightly dubious of the value of slugging it out for a country of little significance to their security. Thus, much as happened during the Vietnam War, no matter how much firepower the United States brings to the fight local adversaries like the Taliban will always have greater resolve to keep fighting.

Buttressing this problem is the “organizational fallacy.” The war on terror has been focused on defeating terrorist groups, primarily by killing the people linked to them. But as the war on terror has illustrated, a group-centric, terrorist-killing approach is the wrong way to think about the problem. Terrorist organizations are not really organizations in a classical sense, nor are they static entities. As the United States broke up Al Qaeda Central, eventually killing Osama bin Laden, Al Qaeda simply morphed and other groups emerged elsewhere to carry the banner, including the Islamic State. Moreover, the Islamic State’s success showed that even killing tens of thousands of fighters does not eliminate the problem of terrorism, even if it degrades the physical organization itself. As should be clear by now, killing individual terrorists does nothing to address the root causes of political conflict in the Middle East. Drone strikes cannot kill the ideologies, ideas, grievances, and emotions that motivate violence in the first place. In fact, many people, including American military commanders, believe that the group-centric, terrorist-killing approach has caused more problems than it has solved.

The final fallacy is what the economist Friedrich Hayek called the “fatal conceit.” The fatal conceit is the mistaken belief that a small group of central planners can manage economic markets or other complex social functions. Sadly, this conceit runs rampant in foreign policy circles, where government officials routinely make plans to reshape nations and influence political outcomes around the world. Even worse, not even sixteen years of utter failure appear to have dampened the power of this fallacy. 

As intelligent people, presidents and their foreign policy teams naturally believe that they can learn from history and apply those lessons to doing better in the future. And so they ignore the evidence that clearly indicates that the complex tasks of political and social engineering – which military intervention and nation building certainly are – are well beyond the ability of any group of people to manage, no matter how smart or well-intentioned they are. The result is that political leaders continue to throw money and military forces at Afghanistan and the Middle East, calling it nation building or a war on terror, believing they are directing events and making progress toward victory.

But these are blunt instruments, not precision tools, and as history has shown their impact has little to do with the plans made by the American government. The forces at play are so many and so complex that no one could have predicted the outcomes of the past sixteen years. Who would have predicted that the Taliban in 2017 would be at their strongest since the 2001 invasion, or the emergence and success of the Islamic State? And yet think tanks, military leaders, and other foreign policy experts continue to argue that the United States should persist until “we get it right.”

At this point, these three fallacies are so deeply embedded in American strategic culture that it is impossible to imagine how the war on terror will ever end. Though Trump may yet confound his advisers and restrain them from their preferred escalation of American effort, a renewed American commitment to the current, failed strategy in Afghanistan and elsewhere seems almost certain.

Yesterday, Attorney General Jeff Sessions reanimated the suspended Department of Justice program that allows local police to circumvent their own state laws to profit from seizing property from individuals who have not been charged, much less convicted, of a crime. Over at Democracy: A Journal of Ideas, I wrote that this may have the unintended consequence of increasing racial profiling on American roads:

Virtually everywhere police stops are counted and measured demographically, black and/or Hispanic drivers are over-represented in those pulled over and subsequently searched for contraband. The vast majority of searches of drivers across ethnicities come up empty, and statistics show that black and Hispanic drivers who are searched are less likely to be carrying contraband than whites who are similarly searched.

Stopping drivers to search for drugs and drug proceeds is much cheaper than developing leads and building cases against large drug organizations through buy-and-bust operations or long-term stings, making interdiction through traffic stops all the more appealing. For that reason, while the disparity in stops almost certainly exists independent of asset forfeiture laws, increasing the use of forfeiture will likely result in an increase of racial profiling.  

These traffic stops that officers initiate to search a car aren’t the typical traffic stops many Americans have experienced. These stops are intrusive, probing, invasive interrogations that are designed to get presumptively innocent people to give up their right to not be searched. These stops are degrading, and the people who experience them–disproportionately black and Hispanic drivers–resent being treated like criminals with something to hide. As I wrote in the Case Western Reserve Law Review, this can diminish police legitimacy in minority communities and thus can erode public safety.

Attorney General Sessions paid lip service to new protections for innocent property owners who might have their cars or cash confiscated by police officers under this program. Even if those very limited protections reduce the number of innocent people whose property is taken by police, there are no protections against the invasive methods police use to find the property in the first place. These pretextual investigatory stops are themselves harmful to innocent drivers and this program provides cash incentives for police departments to employ them.

Congress should act to rein in the DOJ’s forfeiture powers and respect state limits on civil forfeiture. Likewise, state governments should remove the financial incentive police departments have to shake down innocent drivers. 

You can read the Democracy piece here. My CWRLR article is here.


There’s been some buzz this week about a new poll that, according to its creators, shows overwhelming support for Dodd-Frank and the Consumer Financial Protection Bureau (CFPB). A stunning 74 percent of respondents, the creators claim, support Dodd-Frank while 77 percent support the CFPB.

But has not been as widely reported is how skewed the questions were. Given the questions, it’s surprising there was not more support for Dodd-Frank and the CFPB. 

Consider the question that resulted in the 74 percent support Dodd-Frank figure:

Now please listen to this description of the Wall Street Reform law that was passed after the financial crisis. In addition to requiring federal oversight of a larger range of financial companies, this law also prohibits banks from certain risky practices, and created the Consumer Financial Protection Bureau to fight against abusive financial practices that hurt consumers. It also bans taxpayer-funded bailouts of large banks and financial companies and, instead, sets up a system where investors rather than taxpayers bear the losses of bank failures. Please tell me whether, overall, you favor or oppose this law.

No more taxpayer-funded bailouts? No more socialized losses and privatized gains? Sign me up! I would love a law that does this. Except the very reason that many people, including me, oppose Dodd-Frank is because of the belief that it does just the opposite, that it entrenches the too big to fail concept and makes bailouts more likely down the road. The debate over Dodd-Frank is not pro-bailouts versus anti-bailouts, or pro-accountability for bad business choices versus anti-accountability. No one is advocating for bailouts or taxpayer-funded losses. The debate is over whether the particular provisions in Dodd-Frank are likely to lead to more bailouts or fewer, whether the law increases the chances of another crisis or decreases it. 

This question also lobs a number of very nuanced and deeply controversial terms at the respondent – words like “risky” and “abusive” – with almost no context.  What is the optimal level of “risk” for financial institutions, who should assess “riskiness” and how risk can be adequately hedged are questions at the core of ongoing debates over the cause of the 2009 financial crisis and future policy considerations. Labeling a set of practices as “risky” is conclusory and deliberately ignores any existing controversy.

The term “abusive” as used by the CFPB is even more controversial than the term “risk.” Fraud has been illegal in America since the colonial era. The Federal Trade Commission has had the task of policing “unfair” and “deceptive” practices since 1914. These terms are well understood in American law. But Dodd-Frank, in creating the CFPB, charged the new agency with a new mission: protect consumers from practices that are not only “unfair” or “deceptive,” but also “abusive.”  What is an “abusive” practice? What is an example of a practice that is neither “unfair” nor “deceptive” (i.e., where the consumer is not misled or deceived) but is nonetheless “abusive.” The question that resulted in a 77 percent approval rating for the CFPB is no better.  It reads:

The Consumer Financial Protection Bureau, or CFPB, is the first federal agency whose focus is protecting consumers when they use mortgages, credit cards, bank accounts, and other financial products and services. Its mission includes preventing deceptive, unfair and abusive lending and collection practices by banks and other companies. From what you know about the Consumer Financial Protection Bureau, or CFPB, would you say you favor or oppose the CFPB?    

Once again, sounds great. Fighting deception and abuse? Love it. But of course, methods matter. Suggesting that the debate over the CFPB is between those who are for abusive practices and those who are against them is simply disingenuous. I am profoundly concerned about the CFPB, its structure, and its approach to its mission. I believe the term “abusive” is vague and gives the CFPB unnecessary and unprecedented latitude, which has led the CFPB to itself behave in a manner that is abusive to Americans. But I am not in favor of abuse. Quite the opposite. 

To the extent that the results of this poll are held up as proof that the American people support Dodd-Frank, the CFPB, or any of the policies the poll claims to examine, this proof is weak stuff. The questions gloss over the true debate, and therefore fail to accomplish the poll’s purported goal of uncovering Americans’ opinions on these issues. In court, I would claim the questions are leading the witness. Here I will simply claim that the questions are misleading.

With President Donald Trump frustrated all over again with congressional Republicans’ inability to coalesce around a bill repealing and replacing ObamaCare, it seems like a good time to dust off this National Review Online column where I offer 14 ways Trump can pressure Congress and build public support for legislation:

1. End Congress’s illegal ObamaCare exemption.

2. End ObamaCare’s unconstitutional cost-sharing subsidies.

3. End ObamaCare’s illegal “reinsurance” payments.

4. Block Big Insurance’s “risk-corridor” raid on the Treasury.

5. Investigate the Obama administration’s illegal spending.

6. Allow freedom of conscience and choice in contraceptives coverage.

7. Illustrate how Americans can avoid ObamaCare penalties.

8. Illustrate how ObamaCare makes it easier than ever for people to wait until they are sick to purchase coverage.

9. Publish ObamaCare’s vital signs.

10. Release the documents.

11. Praise states that refused to implement ObamaCare.

12. Direct states to prepare for ObamaCare repeal.

13. Renounce IPAB.

14. Let seniors opt out of Medicare without losing Social Security benefits.

To its credit, the Trump administration has been doing some of #9. But not enough. Read here for more.

The landmark Supreme Court case District of Columbia v. Heller established that the right to keep and bear arms under the Second Amendment is an individual right. McDonald v. City of Chicago, which applied the right against state laws, clarified that states can’t “single out” the Second Amendment and treat it with less respect than any other fundamental right.

Heller and McDonald are based on the premise that if states are allowed to decide the scope of our constitutional rights, citizens will receive inferior rights in certain states. Maryland and the U.S. Court of Appeals for the Fourth Circuit are encouraging this system of inferior rights by ignoring the key principle of these cases and allowing state legislatures to define how the Second Amendment should function.

Over a decade ago, James Hamilton accepted a laptop purchased on a stolen credit card and was subsequently convicted of three felonies in Virginia. He completed probation and paid restitution. In 2013, the governor of Virginia restored his Second Amendment rights. Since then, Hamilton has been an exemplary law-abiding citizen; he became an armed security officer with the Virginia Department of Criminal Justice Services and is now a protective security officer with the Department of Homeland Security. A family man and junior wrestling coach, Hamilton is far from a danger to society.

Maryland, however, treats Hamilton as a dangerous felon. When he applied for a handgun permit, he was denied due to his past conviction. A Maryland statute bars felons from owning firearms, with no exception for someone whose rights had been restored. Worst of all, when Hamilton brought his case to federal district court, it was dismissed for failure to state a claim. He did not even argue that the statute was unconstitutional; he argued that the statute, as applied to him, wrongly infringed on his constitutional rights. Yet the Fourth Circuit affirmed the district court, so Hamilton is now petitioning to the Supreme Court for his final chance to reclaim his Second Amendment rights in Maryland.

Cato is supporting his request with an amicus brief. We argue that the Second Amendment has not been adequately protected in lower courts—especially the Fourth Circuit—and that judicial deference has granted states an unacceptable amount of power over civil liberties. With lower courts split on the issue, the Supreme Court has a duty to clarify exactly how states must apply Heller and McDonald. Although Heller established a presumption that state restrictions on felon firearm ownership are valid, this presumption is rebuttable and not immune from as-applied challenges. A decade-old, non-violent felony should not act as a scarlet letter, preventing a person from challenging a statute in court.

When it considers the petition in Hamilton v. Pallozzi this fall, the Supreme Court should take this case and reaffirm a strong commitment to safeguarding our Second Amendment rights against state infringement.

PDF here

Statement for the Record
of David Bier of the Cato Institute[1]
Submitted to Subcommittee on Immigration and Border Security,
House Committee on the Judiciary 
Hearing on
“Agricultural Guestworkers: Meeting the Growing Needs of American Agriculture”
July 18, 2017

Foreign agricultural workers allow farms to expand production, lower prices, and raise incomes for most workers in the United States. Government intervention in the labor market inhibits the ability of farmers to plan the planting and harvesting of crops appropriately, leading to a reduction in production at the start of the season or crops rotting at the end. This government-created uncertainty also makes it more difficult for U.S. companies that rely on U.S. agricultural products to expand. At the same time, unnecessary regulations on agricultural guest workers limit their availability, incentivizing illegal immigration.

To fix these problems and end the regulatory uncertainty, Congress should grant a lawful status to the existing unauthorized immigrant workforce—a disproportionate share of which works in agriculture—and it should reform the current H-2A temporary worker program for future agricultural workers. Excessive regulatory costs and arbitrary limitations on the occupations that H-2A workers may perform limit the use of the program. Any revised H-2A program should allow guest workers to change employers without ex ante government permission and to freely negotiate wages or other benefits without fear of losing their status.

Foreign Agricultural Workers Increase Farm Production

Labor costs account for 17 percent of the variable costs for production on U.S. farms.[1] For fruits, vegetables, and nursery products, the price of labor can account for almost half of all variable costs.[2] All else equal, increases or decreases in the supply of labor result in increases or decreases in production. Higher production has many salutary consequences for the United States, including increased returns on investment, lower food prices, and increased employment.[3]

In 2013, a team of economists employed by the U.S. Department of Agriculture (USDA) estimated that a guest worker program that boosted the number of total temporary workers by 156,000 over 15 years would raise production by up to 2 percent annually in certain labor-intensive subsectors.[4] By contrast, a removal of 5.8 million unauthorized immigrants would decrease agricultural output in certain subsectors by up to 5.4 percent. A similar study found that a 50 percent reduction in foreign workers in the dairy industry would reduce output by 7.9 percent.[5]

Farmers reduce planted acreage when they cannot hire as many workers as they would like to. As the immigration attorney for one farm put it in March 2017, “You’re either reducing your acreage or you’re going for H-2A.”[6] Farmers who miscalculate end up having crops rot in the field. One study found that California fruit and vegetable farmers had to plow under $13 million worth of produce in 2016, a threefold increase since 2011, due to unexpected labor scarcity.[7]

Machines Cannot Fully Replace Human Laborers

Over the last several decades, the price of machinery has fallen relative to the wages of farm laborers.[8] This has led many farms to hire fewer workers and shift to machines. Although opponents of opening the international labor market sometimes argue that the shift to machines proves that foreign labor is unnecessary, this is incorrect for several reasons.

Many industries simply cannot mechanize production at this time. Many fruits and vegetables need to be harvested by hand. The USDA has concluded that some of these sectors, such as apples, oranges, and asparagus, that compete internationally would lose market share if labor costs rise.[9] Even in sectors that can adopt machinery, however, USDA concluded that “hand-harvested produce is usually of better quality, since it is hard to replicate the skill and care of hand harvesters.”[10] This means that a scarcity-inducing labor market policy also harms consumers by diminishing quality. 

Even if increased labor scarcity induced farmers to replace labor with machines, this would defeat the stated purpose of the tighter labor policy: better wages for U.S.-born farm workers. A major justification for the 1965 termination of the Bracero guest worker program for Mexican farm workers was that it would raise wages for American farmworkers. An excellent recent study, however, found that farm wages in Bracero-heavy areas actually did not rise relative to those in other areas after its cancellation.[11] The authors concluded that farmers responded to the increased labor scarcity with mechanization, not with higher wages. 

Foreign Farmworkers Don’t Harm Americans

Guest workers rarely displace domestic farmworkers. Another recent study that analyzed data from the North Carolina Growers’ Association (NCGA) proves this point. NCGA advertised 6,500 farm jobs and accepted 90 percent of all applications from U.S. residents, which amounted to just 245 U.S. residents accepting a job in the field out of a total of 500,000 unemployed North Carolinians.[12] Of these, only seven actually finished the growing season. This is compared to 90 percent of the more than 6,000 H-2A workers who finished.[13] This coincides with a study of California vegetable producers that found at most one U.S. resident is displaced from a farm job for every 81 foreign farm workers.[14]

Even though foreign farm workers may displace a tiny percentage of U.S. workers from farms, this does not necessarily do permanent long-run economic damage. Immigrant farm workers enable U.S. workers to move into related industries where higher pay is available. As foreign farm workers have entered, U.S. workers have moved to other industries, which explains their lack of interest in agricultural jobs. Indeed, U.S. workers make up the vast majority of workers in occupations that depend on manual agricultural labor, such agricultural managers and supervisors, agricultural inspectors, purchasing agents for farm products, and others.[15]

For this reason, increases in the availability of farm labor ultimately benefit U.S. workers elsewhere in the economy. USDA used a general equilibrium model to estimate the effect of removing 5.6 million unauthorized immigrant workers and concluded that about 1 percent of income accruing to natives is dependent on those workers, a disproportionate share of whom work in agriculture.[16]

The Supply and Demand for Farmworkers

Employment figures indicate that agricultural workers are currently in high demand. Farm unemployment is usually higher than the national average because the work is often seasonal or temporary, but unemployment in agriculture over the last 12 months is as low as it has been during any period in a decade.[17] August 2016 saw the lowest unemployment rate for that month since before the year 2000, as did February and March 2017. Average unemployment for the last 12 months was the lowest of any 12-month period since 2007.[18]

Few economic sectors in the United States rely more heavily on foreign workers than agriculture, meaning that immigrants are responsible for a large portion of its production. The USDA has found that more than 70 percent of hired crop farmworkers migrated to the United States from other countries.[19] It also found that half of the total lacked proper authorization to live or work in the United States.[20] In addition, another 135,000 foreign workers received H-2A visas in 2016 to come and work as temporary farm workers in the United States.[21] The use of this program has increased substantially over the last two decades, doubling in the last five years alone (Figure 1).[22]

Figure 1
H-2A Agricultural Visas Issued, FY 1997-FY 2016
Source: U.S. Department of State

This increase in the supply of guest workers is beneficial, not only due to the positive economic factors described above, but also due to the positive impact on border security. Over the last 60 years, guest worker entries correlate negatively with apprehensions of illegal aliens at the border. Logically this relationship makes sense. Government has artificially restricted the supply of a service below the level of market demand. The excess demand seeks an outlet in the underground economy. The United States experienced a similar process during alcohol prohibition with bootleggers replacing the legitimate liquor market. When Congress first enacted immigration quotas in the 1920s, many commenters at the time noted the relationship between these two progressive policies, calling illegal immigration “human bootlegging.”[23]

Figure 2 highlights this relationship for the post-World War II period. The red line is the number of guest workers entering each year. The blue bars represent the number of apprehensions that each border agent made during the year, which is the best measure of total illegal immigration available. Typically, more apprehensions per agent mean more attempted crossings. As it shows, when illegal immigration first started in the late 1940s, Congress responded to the market, increased the supply of visas, and continued to do so until the supply met demand. Combined with a stricter enforcement policy in the 1950s, the problem nearly evaporated, but it returned with a vengeance after Congress eliminated the Bracero program in 1965.[24]

Figure 2
Guest Worker Entries and Apprehensions of Illegal Aliens per Border Patrol Agent 1946-2015
Sources: Border Patrol; Immigration and Naturalization Service

In recent years, the number of guest workers entering each year has risen sharply, which again has coincided with a drop in illegal immigration, but the H-2A program still needs reforms. From 2007 to 2011, H-2As represented just 10 percent of farm employment.[25] Although it has doubled in size since then, it is still a small share of the total. The biggest problem is that H-2A workers may not work in any permanent or non-seasonal jobs. This eliminates roughly half of all farm jobs, including in the dairy and livestock industries. H-2A visas are also unavailable for meat and poultry processing. The H-2A program is unworkable for certain seasonal industries that cannot guarantee a certain length of employment as the program requires. USDA also notes that the inflated wage requirements may discourage some employers from using it.[26]

An expanded agriculture guest worker program would benefit the U.S. economy and improve border security. Foreign farm workers are in demand and Congress should allow the free market, rather than the black market, to meet that demand.


[1] The Cato Institute is a libertarian 501(c)(3) nonprofit think tank founded in 1977 and located in Washington D.C.


[1] Patrick O’Brien; John Kruse; and Darlene Kruse, “Gauging the Farm Sector’s Sensitivity to Immigration Reform via Changes in Labor Costs and Availability,” WAEES, February 2014,

[2] Steven Zahniser, Tom Hertz, Peter Dixon, and Maureen Rimmer, “The Potential Impact of Changes in Immigration Policy on U.S. Agriculture and the Market for Hired Farm Labor: A Simulation Analysis,” U.S. Department of Agriculture, Economic Research Service, Economic Research Report Number 135, May 2012.

[3] See: Zahniser, et al.

[4] Ibid.

[5] Rosson, Parr, Flynn Adcock, Dwi Susanto, and David Anderson, “The Economic Impacts of Immigration on U.S. Dairy Farms,” National Milk Producers Federation: Arlington, VA, June 2009, fi les/fi le/NMPF%20Immigration%20Survey%20Web.pdf.

[6] Geoffrey Mohan, “To keep crops from rotting in the field, farmers say they need Trump to let in more temporary workers,” Los Angeles Times, May 25, 2017,

[7] Kelsey Brugger, “Labor Shortage Leaves $13 Million in Crops to Rot in Fields,” Independent, June 22, 2017,….

[8] Sun Ling Wang, Paul Heisey, David Schimmelpfennig, and Eldon Ball, “Agricultural Productivity Growth in the United States: Measurement, Trends, and Drivers,” United States Department of Agriculture Economic Research Service: Economic Research Report 189, July 2015,….

[9] Linda Calvin and Philip Martin, “The U.S. Produce Industry and Labor: Facing the Future in a Global Economy,” United States Department of Agriculture Economic Research Service: Economic Research Report 106, November 2010,….

[10] Ibid.

[11] Michael A. Clemens, Ethan G. Lewis, and Hannah M. Postel, “Immigration Restrictions as Active Labor Market Policy: Evidence from the Mexican Bracero Exclusion,” NBER Working Paper No. 23125, February 2017,

[12] Michael Clemens, “International Harvest: A Case Study of How Foreign Workers Help American Farms Grow Crops – and the Economy,” Partnership for a New American Economy and the Center for Global Development, May 2013,….

[13] Ibid.

[14] Devadoss, Stephen, and Jeff Luckstead. “Contributions of Immigrant Farmworkers to California Vegetable Production,” Journal of Agricultural and Applied Economics 40(3): pp. 879-94, 2008,

[15] American Community Survey, 2011-2015

[16] Zahniser, et al.

[17] Bureau of Labor Statistics, “Unemployment Rate - Agricultural and Related Private Wage and Salary Workers,”

[18] Bureau of Labor Statistics, “Unemployment Rate - Agricultural and Related Private Wage and Salary Workers,”

[19] U.S. Department of Agriculture, “Farm Labor - Background,” Tuesday, September 27, 2016,

[20] Ibid. See also: Jeffrey Passel and D’Vera Cohn, “Size of U.S. Unauthorized Immigrant Workforce Stable After the Great Recession,” Pew Research Center, November 3, 2016,….

[21] U.S. Department of State, “Worldwide NIV Workload by Visa Category FY 2016,”….

[22] U.S. Department of State, “Nonimmigrant Visa Statistics - Nonimmigrant Visa Issuances by Visa Class and by Nationality,”….

[23] David Bier, “Tougher Restrictions on Immigrants Won’t Work,” Newsweek, November 16, 2016,

[24] Alex Nowrasteh, “Guest Worker Visas Can Halt Illegal Immigration,” Cato Institute, May 5, 2014,

[25] Zahniser, et al.

[26] Zahniser, et al.

It’s been two years since the “Blurred Lines” verdict, but the daze has just begun. According to a BBC report last week, recording artists are now being instructed not to talk publicly about their musical influences for fear of exposure to copyright infringement claims.

“Blurred Lines” was a chart-topping 2013 pop song by Pharrell Williams, Robin Thicke, and Calvin “T.I.” Harris. Marvin Gaye’s family successfully claimed that the track infringed on Gaye’s 1977 song, “Got To Give It Up,” winning $5.3 million in damages and 50 percent running royalties. The case is now on appeal at the Ninth Circuit.

If copyright law was focused on actual, you know, copying, this case would have never gone to the jury. The Gaye family holds the copyright to the sheet music, not the actual recording, so its claim should stand or fall based on the notes on the page. Which is to say, it should fall: the two songs are set in different keys and use different sets of chords (see good analyses here and here). Accordingly, “they sound similar” shouldn’t even be a relevant argument, much less a winning one.  

The Gaye family based its case on shared “elements” in both songs, as well as Pharrell Williams’ admissions that he was inspired by Gaye and that the song captured the “feel” of Gaye’s earlier tune. The trial court allowed the case to go to the jury and included an instruction that “substantial similarities” between elements of the two songs was evidence of infringement.

This way lies mayhem. If the liability standard set in this case were generally followed, the normal processes of musical creation—and artistic creation more generally—would be illegal. Artistic creation is inherently a social endeavor, each new work part of an ongoing conversation with other artists and the audience. Artists naturally pick up on what has been said previously in that conversation, borrowing elements from the past and rearranging them and adding to them to create something new. The existence of distinctive styles, genres, and movements in the arts is possible precisely because of the ubiquity of artists’ drawing on what has come before.

This unfortunate case demonstrates how copyright’s expansive coverage of “derivative works” is antithetical to the stated purposes of the law. Copyright is supposed to incentivize artistic expression, but now we have come to the point where artists are being urged to muzzle themselves to keep themselves away from the law’s reach. A decision by the Ninth Circuit to toss out the verdict would be welcome news—but only a small, first step toward reining in a law run amok.

This post was coauthored with Rachel Chiu.

Momentum is building for air traffic control (ATC) reform. With health care reform prospects dashed for now, and tax reform facing a difficult path, ATC reform could be an area for legislative progress in coming months. The Trump administration and House leadership are on-board with an ATC privatization plan passed through the lower chamber’s transportation committee. And while the Senate is always a hurdle for fiscally conservative reforms, privatization supporters have leverage because current funding for the ATC system runs out at the end of September.

Why do we need major ATC reform? This is a high-tech industry that is rapidly evolving, yet our system is trapped inside of the hopelessly sluggish Federal Aviation Administration (FAA). In other countries, independent ATC systems are moving ahead with an array of innovations. We are falling behind in a very real way, which has important ramifications for airport congestion, flight delays, and aviation safety.

Consider one cool new ATC technology: “remote” or “virtual” control towers. The iconic airport towers that have the big windows for controllers to see runways are likely on the way out. They will be replaced by visual and infrared cameras on runways able to pan and zoom, with the electronic feed going to control centers either nearby or hundreds of miles away. The feed will be displayed on wall-sized high-definition monitors that will be overlaid with electronic flight and sensor information.

The United States is behind on remote towers, as we are on many ATC technologies. The first remote tower was built by Saab and put in operation in Sweden in 2015, as shown in the photo. The company describes some of the advantages of remote towers here, including superior performance at nighttime and during bad weather. 

Sweden has implemented the system at three airports and is planning more. Its neighbor Norway “is going further and consolidating control of 15 small northern airports into one virtual tower. Another 17 airports may be added to that centre later,” according to The Economist.

Britain’s privatized ATC company is installing a remote tower at London City Airport. The feed from 14 cameras will go to a control room 80 miles away. The new system—built by Saab—will be better and cheaper than a traditional tower, and it will free up land space.

Ireland has successfully tested remote tower systems at two of its airports, and Germany is installing a remote tower at Saarbrücken Airport with the controllers sitting “350 miles away in Leipzig.” Wow.

Canadian firm Searidge Technologies is installing remote tower systems in Budapest and other cities. The privatized British and Canadian ATC companies are co-owners of Searidge.

What about the United States? The FAA has been slow to explore remote tower systems, as it has been slow to adopt satellite-based navigation, electronic flight strips, and other ATC advances. Remote tower systems are being demonstrated at airports in Virginia and Colorado, but the FAA is not part of those experiments.

Instead, Bob Poole notes that the FAA is building new towers in Charlotte ($112 million), Philadelphia ($200 million), and other cities. All are expensive traditional towers. Remote towers are the future, but the FAA is stuck in the past.

Here is a crucial political point: ATC reforms face opposition in Congress by members who fear that a privatized system will hurt small, rural airports. But remote towers could be the savior of small airports because they offer a much cheaper way to provide ATC at low-traffic facilities. The Economist says the remote towers in Norway “could bring the cost of air-traffic services down by as much as 40%. The savings could be used to help to maintain services at small airports which might not have control towers of their own, or are not busy enough to maintain a full-time tower service.”

Remote towers are the exact type of innovation that a privatized U.S. ATC service would pursue to improve quality and cut costs. But for a host of bureaucratic reasons, it is one of many opportunities that our lethargic government system is slow-walking or ignoring.

For more on remote towers, see a new Reason Foundation study here and a study by the company that installed the German system here. For more on ATC reform, see here.

On Monday, the Trump administration once again officially certified that Iran is in compliance with the Joint Comprehensive Plan of Action (JCPOA), the agreement that rolled back Iran’s nuclear program and subjected it to unprecedented levels of inspections and monitoring in exchange for sanctions relief. But, according to multiple reports, Trump was very close to refusing to do so.

Apparently, there is a split in the administration. Some of Trump’s national security advisors, along with some hawks on Capitol Hill, are intent on torpedoing the Iran nuclear deal. And Trump was set to officially claim, contrary to the facts, that Iran was not living up to the agreement. At the last minute, another camp in the administration’s national security team, including Secretary of State Rex Tillerson, Secretary of Defense James Mattis, and national security advisor H.R. McMaster, prevailed upon the president to tell the truth and certify that Iran is indeed complying with its obligations under the JCPOA.

These are very ominous signs about the intentions of President Trump. Recall that the president has repeatedly denounced the Iran nuclear deal as a pathetic capitulation that must be reversed. That is not only wrong, but it puts the administration in a difficult spot. The other parties to the agreement – including Britain, France, Germany, China, and Russia, as well as the International Atomic Energy Agency (IAEA) – all know that Iran made major concessions and they all concede Iran is so far complying with the deal’s stringent limitations on its nuclear program.

If the United States is the one to pull out of the deal, falsely claiming Iran is acting in violation of the agreement, those other major countries won’t play along with us. International sanctions won’t be re-imposed, and the United States will look like the rogue actor. Up until now, hardliners in the administration have advocated trying to “provoke Iran into being the one to scrap the nuclear deal, [thereby] leav[ing] the United States in a stronger position.” Worryingly, it looks like they’ve just about lost patience with that approach, and intend to effectively pull out of the deal at the next 90-day interval requiring U.S. certification.

Other pressure tactics continue. For example, Trump’s State Department yesterday announced a new set of economic sanctions on Iran for behavior unrelated to the nuclear agreement. And this isn’t the first time. These actions needlessly ratchet up tensions, bolster the hardline voices inside Iran who believe the JCPOA capitulated too much, and contribute to widespread sentiment among the Iranian population that they are not seeing the promised economic benefits of complying with the deal.

Here is the ultimate irony of President Trump’s reported intention to refuse to re-certify Iranian compliance: it is actually the United States, not Iran, that appears to be in material breach of the JCPOA. At the G-20 summit earlier this month, President Trump repeatedly urged fellow world leaders to stop doing business with Iran. This pretty clearly violates not only the spirit but also the letter of the deal, which says that the United States “will refrain from any policy specifically intended to directly and adversely affect the normalization of trade and economic relations with Iran inconsistent with their commitments not to undermine the successful implementation of this JCPOA.”

Add all of this to the bewilderingly chaotic and unstable regional situation in the Middle East. U.S. allies Saudi Arabia, Egypt, Bahrain, and the United Arab Emirates are in a standoff with another U.S. ally, Qatar, in part over the latter’s unacceptably cordial relations with Iran. In addition, a flock of external powers, including the United States and Iran, are clustering inside Syria’s civil war, mainly to battle ISIS, and have repeatedly clashed militarily. In that environment, a belligerent and impressionable President Trump with top national security advisors pushing to nullify the JCPOA is a dangerous mix that could portend not just the end of a successful diplomatic agreement that has taken Iranian nuclear weapons development off the table for the foreseeable future, but possibly even war.

Despite the perilous instability in the Middle East, when it comes to the JCPOA and U.S.-Iranian relations, America is not a victim of circumstance. Nothing is forcing the United States to increase pressure on Iran. Nothing is forcing the Trump administration to abrogate the deal. This is a choice. If it all falls apart and Iran is consequently freed from its obligations to limit its nuclear development, the United States will be right back on the path to war with Iran, exactly where it was before July 2015 when the JCPOA was signed. If the deal, the sanctions relief, and the opportunity for peace unravel, it will have been elective.

Americans should now be well aware of the dire consequences of permitting hawkish presidential administrations to rush headlong into war and regime change in the Middle East. Iraq cost us, and the region, dearly. A confrontation with Iran would be an order of magnitude worse. All Trump needs to do to avoid that is admit the truth and certify that Iran is complying with the JCPOA. 

Today through Monday, July 24th, Cato and the Urban Institute are hosting a joint online debate contemplating costs, benefits, and possible reforms to zoning regulation. Participant opinions will run the gamut, from anti-zoning to pro-zoning, which should make the conversation lively. Participants will include the following:

  • Emily Talen, professor of urbanism, University of Chicago
  • Robert Dietz, chief economist and senior vice president, National Association of Homebuilders
  • Dana Berliner, senior vice president and litigation director, Institute for Justice
  • Lance Freeman, professor of urban planning, Columbia University
  • Richard Rothstein, research associate, Economic Policy Institute; fellow of the Thurgood Marshall Institute, the NAACP Legal Defense Fund, and the Haas Institute at the University of California (Berkeley)
  • Craig Anthony Arnold, professor of law and affiliated professor of urban planning in the Department of Urban and Public Affairs, and chair of the interdisciplinary Center for Land Use and Environmental Responsibility, at the University of Louisville
  • Derek Hyra, associate professor in the School of Public Affairs, American University

I will moderate the debate alongside Rolf Pendall, co-director of the Metropolitan Housing and Communities Policy Center at the Urban Institute.

On the anti-zoning side, participants are expected to argue that zoning increases housing costs and segregation, while reducing property rights, individual liberty, and economic growth. On the pro-zoning side, participants are expected to contend that zoning is a boon to environmental justice, growth management, and community preservation.

The debate will conclude by contemplating possible reforms to zoning regulation. This portion is particularly important because there is often substantial local pressure to zone restrictively. Overcoming it often requires ingenuity.

Join us for the full conversation here.

Amtrak’s co-CEO Wick Moorman has announced that the passenger railroad is thinking of offering a new service to compete with the airlines: economy seating that is crammed together as tightly as airline seats. This was immediately blasted by Senator Charles Schumer (D-NY), saying, “Amtrak should not throw out one of the best things about Amtrak and train travel — that is, you at least get a seat you can sit in and be comfortable.”

In fact, this idea makes no sense not because heavily subsidized train travelers somehow deserve more comfortable seats but because it would cost Amtrak more in lost revenues than it will save. Airlines fill 85 percent of their seats and on lots of flights they fill 100 percent. Amtrak fills only 51 percent of its seats, so cramming more seats into a railcar will simply mean more empty seats.

According to USA Today, Amtrak seat pitches–the distance from the back of one row of seats to the back of the next–are 39 inches for day trains and 50 inches for overnight trains. Airline seat pitches are 30 to 33 inches while buses are 28 to 31 inches. That means Amtrak could squeeze in four rows of seats where it now has three on day trains and five rows where it now has three on overnight trains.

Amtrak’s overnight trains rarely have more than four coaches. Substituting one economy coach for two regular coaches would save a little bit on fuel and maintenance and results in an overall loss of seating capacity. Many coach riders on the overnight trains are price sensitive, so most of the people attracted to the economy coaches would have otherwise taken the regular train. Thus, Amtrak is likely to lose more revenue than it gains by attracting few people away from buses or planes.

Outside the Northeast Corridor, most of Amtrak’s day trains also tend to have about four coaches, so the same logic applies. Northeast Corridor trains may have eight coaches or more, so it is possible Amtrak could substitute three economy coaches for four regular coaches, saving the cost of one car.

That wouldn’t be enough to make Amtrak more competitive, as Amtrak’s real competition in the Northeast Corridor is not the airlines but the bus companies, especially Megabus and Bolt. Right now, the cheapest Amtrak fare from Washington to New York is $49, and if you want to go tomorrow the price is as high as $180. Megabus fares start at $1.50 and top out at $27.50, though when I looked even many buses tomorrow have fares as low as $15.50.

That means, to be competitive, Amtrak would have to reduce the fares for its economy class by at least two thirds and probably a lot more while its costs would go down by, at most, one quarter. If it only reduces the fares by a quarter it will end up drawing most of its economy customers away from its own regularly priced trains, not from the bus companies. Thus, economy seating would minimally lower costs but significantly reduce revenues.

Moorman, who was formerly CEO of the Norfolk Southern Railway, may not have been serious about this suggestion, but it is disappointing that he made it at all because I can’t see any way for Amtrak to win from it. Moorman, who was made CEO last September, has already been replaced by former Delta Airlines CEO Richard Anderson. To insure a smooth transition, the two are acting as co-CEOs until the end of this year. While Anderson is obviously familiar with the passenger industry, his only tenuous connection to the railroads is that his father was an office worker for the Santa Fe Railway.

Amtrak’s real problem, which neither Moorman nor Anderson can afford to admit, is that it has no reason to exist. Too slow to compete with the airlines and too expensive to compete with buses, passenger trains are simply not competitive in any market without huge subsidies, and even then either buses or airlines attract far more passengers. In 2016, federal and state Amtrak subsidies  per passenger mile were more than ten times subsidies to airlines or buses. For those subsidies, Amtrak carried about 1 percent as many passenger miles as domestic airlines and about a third as many as scheduled intercity buses.

The Northeast Corridor, whose trains supposedly make money, in fact has, by latest count, a $38 billion maintenance backlog, which wouldn’t exist if the trains were actually profitable. To pay that off in 20 years would require more than a doubling of fares (with no loss of riders) on Amtrak trains as well as all commuter trains that use the Northeast Corridor, and by the end of that time there would be more things that need to be replaced.

Neither Moorman or Anderson will be able to solve these problems. Instead, they are going to have to depend on the political winds blowing in their direction, which is never a safe place to be.

The Federation for American Immigration Reform (FAIR) released a report claiming that E-Verify lowered unemployment rates in states that implemented it.  FAIR’s report is deeply flawed.  The first section of this blog will catalog FAIR’s errors and show that states with mandatory universal E-Verify typically had higher unemployment.  The second portion of this blog will use the synthetic control method to look at E-Verify’s effect on unemployment in Arizona after the E-Verify mandate.  The flaws in FAIR’s report are important to highlight as more states are considering a universal E-Verify mandate.  There is little evidence that E-Verify mandates lower unemployment but much evidence that they raise it.   

Criticisms of FAIR’s Report

E-Verify is a taxpayer funded federal government run system that is supposed to exclude illegal immigrants from the workforce.  The system would be used at the point of hire to verify that any new worker is actually authorized to work in the United States.  FAIR attempted to show that states with E-Verify have higher employment growth relative to other states.  This is likely an attempt to overcome one of the stronger criticisms of E-Verify: It is an expensive labor market regulation that will increase unemployment by raising the cost of hiring new workers among other problems.  However, FAIR excluded the first state to mandate E-Verify and made numerous other silly methodological choices that make their results unreliable. 

First, the FAIR authors excluded Arizona from their report.  Arizona was the first state to mandate E-Verify for all new hires.  Unemployment rates as measured by U3 were lower in Arizona than in the rest of the United States prior to the implementation of E-Verify and they shot up afterward, remaining consistently above the rest of the United States (Figure 1).  The result is even more extreme for the U6 unemployment rate that the FAIR report insisted on using (Figure 2).  Narrowing the comparison to the southwestern states of California, Colorado, Nevada, New Mexico, Oklahoma, Texas, and Utah shows similar results whereby Arizona had relatively lower unemployment prior to mandating E-Verify and higher unemployment afterward (Figures 3-4).  Utah mandated E-Verify for some employers during this time but excluding that state does not affect the results.  Mandatory E-Verify did not appear to improve employment in Arizona. 

Figure 1

Arizona Unemployment Rate (U3) vs. United States Unemployment Rate (U3)

Source: Bureau of Labor Statistics.

Figure 2

Arizona Unemployment Rate (U6) vs. United States Unemployment Rate (U6)


Source: Bureau of Labor Statistics.

Figure 3

Arizona unemployment Rate (U3) vs. Southwest Unemployment Rate (U3)


Source: Bureau of Labor Statistics.

Figure 4

Arizona Unemployment Rate (U6) vs. Southwest Unemployment Rate (U6)


Source: Bureau of Labor Statistics.

Second, FAIR did not distinguish between different types of E-Verify mandates.  They treated West Virginia’s law that the “West Virginia Legislature requires any employer that has employees working on the grounds or in buildings of the Capitol Complex, or who have access to sensitive or critical information, to use E-Verify” with being as strict as Alabama’s statewide mandatory E-Verify requirement for all new private and public sector hires.  Those different laws are going to have different impacts so it makes no sense for FAIR to treat them as the same.  This blog only looks at states with universal E-Verify mandates because that is FAIR’s policy goal.  

Third, FAIR did not show an average effect across states with mandatory universal E-Verify.  I constructed a simple average of U6 unemployment in all states (Arizona, Alabama, Mississippi, and South Carolina) for each year they had a universal E-Verify mandate and compare them to states that did not (Figure 5).  States are added to the average when they passed new mandates.  The vertical lines show when each state passed its E-Verify mandate.  States with an E-Verify mandate had consistently higher unemployment than states that did not.  This does not mean that E-Verify caused the persistently higher unemployment rates in these states as these state governments might have passed E-Verify because of high unemployment rates.  However, this simple analysis does provide evidence that states with E-Verify mandates did not then have lower unemployment rates than other states.

Figure 5

E-Verify Mandate States Unemployment Rate (U6) vs. United States Unemployment Rate (U3)


Source: Bureau of Labor Statistics.

Fourth, there were an estimated 350,000 unauthorized immigrants living in Arizona in 2009.  In the same year, there were only 210,000 illegal immigrants living in all of Alabama, Mississippi, and South Carolina.  If E-Verify were to have an effect on unemployment then it would be most obvious is Arizona.  However, it is impossible to tell when each state enacted mandatory E-Verify by eyeballing the U6 rate (Figure 6).  The state of the economy drives the unemployment trends more than this labor market regulation.  If E-Verify has an effect, it is small.

Figure 6

Unemployment Rates (U6) by State and Nationwide


Source: Bureau of Labor Statistics.

There are at least two reasons why the E-Verify mandate did not much affect the unemployment rates in these states.  First, illegal immigrant workers were small fractions of the workforce in three-quarters of them.  Second, E-Verify’s effect is attenuated because many employers in these states do not follow the E-Verify mandates and it is ineffective at actually excluding illegal immigrant workers even when it is used properly.

Fifth, the FAIR report analyzed a period of time that coincided with recovery from the Great Recession.  In other words, FAIR chose a start date for its report after which nationwide unemployment rates were guaranteed to fall.  The authors of the FAIR report made no effort to correct for that huge confounding factor. 

Synthetic Controls Method

This section uses the synthetic controls method (SCM) to see how E-Verify affected relative unemployment in states that mandated the program.  SCM “allows for effect estimation in settings where a single unit (a state, country, firm, etc.) is exposed to an event or intervention.  It provides a data-driven procedure to construct synthetic control units based on a convex combination of comparison units that approximate the characteristics of the unit that is exposed to the intervention.  A combination of comparison units often provides a better comparison for the unit exposed to the intervention than any comparison unit alone.”  SCM creates an alternative version of the unit that was not affected by the intervention based on numerous indicator variables for other units prior to the intervention.  Researchers use the SCM widely

The comparison units are states that must be similar to Arizona prior to the passage of mandatory E-Verify which I track back to 2004 (other papers have gone back to 1998).  Furthermore, the indicator variables must also be related to unemployment.  I chose states in the American southwest and indicator variables of the state level minimum wage laws, population, the percent of the population that are illegal immigrants, the state unionization rate, and per capita GDP.  SCM shows that Arizona’s unemployment rate rises above what it would have been in Arizona without E-Verify (Figure 7).  This means that E-Verify is likely the cause of the increase in relative unemployment in Arizona since its mandate.

The Root Mean Square Percentage Error (RMSPE) measures how close the pre-E-Verify Real Arizona is to the pre-E-Verify Synthetic Arizona Without E-Verify and the closer it is to zero (but not too close), the better it is.  The RMSPE is an excellent 0.0504892.  The Synthetic Arizona without E-Verify is constructed from 33.9 percent of Nevada, 29.1 percent New Mexico, 20.9 percent Texas, and 16.1 percent Utah.  The indicator variables are 16.7 percent the Arizona  U3 unemployment rate in 2007, 28.7 percent the population of the states, 14.4 percent the state’s minimum wage, 9.2 percent the state’s population who are illegal immigrants, a fraction of a percent from the unionization rate, and 31 percent from the state’s GDP per capita.  These are good weights distributed over many indicator variables.  When I excluded states with a partial E-Verify mandate on firms by certain sizes (Georgia, North Carolina, Tennessee, and Utah), the results were very similar. 

Figure 7

Synthetic Control Method Arizona, U3 Unemployment Rate


Sources: Bureau of Labor Statistics, Bureau of Economic Analysis, Eren and Ozbeklik, American Community Survey, and Pew.  

I then ran two robustness checks to see how much confidence we should have in these findings.  The first is called an in-time placebo and it changes the date of E-Verify’s enactment to see if there was a similar effect.  If there was a similar effect in other years then E-Verify is less likely to be the culprit. However, there was not a similar effect for other years which should increase confidence in the original SCM finding that E-Verify increases unemployment. 

The second robustness check is called an in-place placebo and it runs the SCM for every single state to see if Arizona stands out (Figure 8).  Arizona is in the bolded green line and it does not stand out relative to other states in the southwest.  This robustness check should decrease our confidence that E-Verify had much of an impact on the unemployment rate in either direction because Arizona straddles the horizontal 0 line.  A quick use of the SCM over these years with these indicator variables cannot show that E-Verify increased unemployment in Arizona but it does show that E-Verify did not lower unemployment. 

Figure 8

In-Place Placebo, U3 Unemployment Rate


Sources: Bureau of Labor Statistics, Bureau of Economic Analysis, Eren and Ozbeklik, American Community Survey, and Pew.  

This wonderful paper by Magnus Lofstrom, Sarah Bohn, and Steven Raphael at the PPIC uses different indicator variables over a different period and finds a negative employment effect from E-Verify.  I trust their more thorough review of the evidence than my own but I needed to run my own simpler SCM here in order to look at the same years analyzed by FAIR.        


FAIR’s report made several questionable methodological choices in analyzing how E-Verify affects unemployment rates on the state level.  Most egregiously, FAIR excluded Arizona even though it was the first state to mandate universal E-Verify.  Any thorough and reputable study on the employment effects of E-Verify must include all of the states that actually mandated it.  Furthermore, unemployment rates in states with universal mandatory E-Verify were mostly higher than in other states.  Lastly, I used the synthetic control method to show that Arizona’s unemployment rate did not drop as a result of E-Verify and there is some evidence that E-Verify even increased unemployment in Arizona.  Whatever one can say about E-Verify, it does not lower unemployment rates.

Many states are considering an E-Verify mandate for all employers and employees in their states. E-Verify is a taxpayer funded federal government run system that is supposed to exclude illegal immigrants from the workforce. In practice, E-Verify imposes high economic costs and does little to dampen the wage magnet that attracts illegal immigrants to the United States.

State attempts to mandate E-Verify have run into serious roadblocks in at least three states this year and are likely dead there. For instance, Illinois’ HB 3415 is still stuck in committee and is very unlikely to become law. In fact, Illinois is so averse to E-Verify that it even tried to prohibit its use of that system by any employer in the state. New Jersey’s A 3249 is extremely unlikely to pass in a state government dominated by Democrats. Maine HP 904 is effectively dead as that legislator has more pressing matters to attend to.

However, a watered-down form of mandatory E-Verify has an excellent chance of becoming law in North Carolina. The Protect North Carolina Workers Act (HB 35) passed the House and will likely be taken up by the state Senate when it returns in late August. The Protect North Carolina Workers Act would require mandatory E-Verify for every new employee hired by a firm with 15 or more employees. Legislators have exempted domestic and farm workers from E-Verify, likely because they believe those sectors of the economy would be devastated if they also had to comply with the mandate. 

E-Verify’s cheerleaders sell the system as a silver bullet that will resolve illegal immigration with little help from other government programs. E-Verify takes personal identity information that workers must provide on I-9 tax forms and then checks the validity of their Social Security Number against information held in Social Security Administration (SSA) and immigration status against Department of Homeland Security (DHS) databases. If both databases decide that the identity information is valid then E-Verify approves the employee for work. If E-Verify encounters an error then it returns a tentative non-confirmation that requires the employee and employer to sort out the error. In 2012, 46.5 percent of contested E-Verify cases took longer than eight working days to resolve. If the error is resolved then the employer can permanently hire the worker. If the error is not resolved then E-Verify issues a final non-confirmation whereby the employer must fire the worker.

That whole system can only work well if E-Verify and the information in government databases is reliable and accurate. Unfortunately for workers and employers, that information is wrought with errors. The last major survey of E-Verify’s accuracy rates in 2012 found that it incorrectly approved 54 percent of illegal immigrants to work legally in the United States. E-Verify approved them because it relies on documents presented by the workers themselves. The biggest hole in E-Verify is that it checks the validity of documents but does little to check whether they belong to the actual person who presents them. For example, E-Verify flags a blatantly false Social Security number but it will not flag a valid one when it is paired up with the right name even if neither belong to the correct worker.  Any problems created by a nationwide E-Verify mandate will lead to calls for a national biometric identity card to fill the gaps.

E-Verify also forces some employers to fire some legal workers. Around 0.15 percent of E-Verify queries result in a false final non-confirmation, locking out otherwise legal citizens and permanent residents from employment. Even more absurd, about half of all new hires in states that require mandatory E-Verify for all new hires are not run through the system. The goal of E-Verify is to disincentivize illegal immigration by denying workers employment but mandatory E-Verify in Arizona only lowered wages of Mexican illegal immigrants by about five percent. That hardly counts as demagnetizing the wage magnet. Enforcing E-Verify is about as difficult as enforcing the current I-9 system except that it adds another layer of bureaucracy for employers and employees to overcome. 

A North Carolina E-Verify mandate for firms with 15 or more employees will have little effect on unlawful immigration but will increase the cost of hiring new workers. North Carolina has grown rapidly over the last few decades. Its relatively low taxes, tolerable administration of justice, and manageable regulation environment have made it a magnet for workers and firms from around the world. Imposing a new, costly, and ineffective labor market regulation like E-Verify is a bold strike against the other wise policies that have made North Caroline an economic success.  

Congress should privatize federally-owned businesses such as Amtrak, the postal service, and the air traffic control system. Privatization would cut costs and improve customer service, as I discuss here. It would also boost U.S. innovation and exports.

Typically, federal government businesses do not export their goods, services, or technologies. They have no incentive to do so. They are content to live the quiet life, and they have little reason to innovate or seek foreign markets. As such, keeping business activities trapped inside of bureaucracies restricts growth opportunities for the economy.

In a new article on air traffic control (ATC), Rui Neiva of the Eno Center illustrates the theme. He discusses how the privatized Canadian ATC company developed and implemented “electronic flight strips” years ago, while our government ATC bureau is still struggling to adopt this advance.

By 2009, all NAV CANADA facilities, including those serving general aviation, were equipped with electronic flight strips. Given that NAV CANADA is both an operator and a manufacturer of equipment for ATC, the technology that they developed has been sold to multiple countries worldwide, including Australia, Denmark, Dubai, Italy, and United Kingdom.

Canada privatized its ATC system in 1996 as Nav Canada. The company is a leader in ATC innovation and has developed numerous technologies, such as the flight strips, that it exports abroad. Bob Poole noted, “The technical expertise at Nav Canada has led to a thriving business marketing innovative ATC hardware and software and advising other air navigation service providers on modernization.”

Nav Canada earns income from foreign contracts and royalties, which in turn helps fund its ongoing research. Another interesting byproduct of Canadian privatization is Searidge Technologies, which is a Canadian company that is a leader in selling “remote tower” services to ATC systems worldwide. Nav Canada was an early investor in Searidge and today is co-owner.

Now consider another industry run by our government: currency printing. Our money is printed by the U.S. Bureau of Engraving and Printing. But Canada has long contracted out the printing of its money to the Canadian Bank Note Company and other firms.

Canadian Bank Note Company has used its domestic expertise as a base to go global, and today it prints money, stamps, and high-end security products for clients in 60 counties. Meanwhile, the Bureau of Engraving and Printing supplies only the U.S. market and does not export.

Government agencies fall behind because they are cut off from global markets and the need to compete. Look at the U.S. government-made bills in your wallet. Boring! Now look at this cool Canadian fiver or this New Zealand fiver, both manufactured from high-tech plastics by the Canadian Bank Note Company. The NZ bill won “international bank note of the year” in 2015.

The lesson is that we deny opportunities to American entrepreneurs when we keep businesses entombed in the government. Moving ATC, currency printing, and other activities to the private sector would allow American workers to capitalize on their skills and sell their innovations worldwide.

More on air traffic control here.

More on privatization here.

Join me, Michael Sargent, and Rep. Tom Massie at the Heritage Foundation tomorrow to discuss air traffic control reform.


Liu Xiaobo passed away on July 13th, 2017, at age 61. This piece was originally posted at on September 8, 2016.

When I think of examples of successful self-control and dignity under the most difficult circumstances, one person comes to mind before all others: 刘晓波, Liu Xiaobo, the Chinese writer and human rights activist, who was sentenced on Dec. 25, 2009, to 11 years of imprisonment in China on the charge of “inciting subversion of state power.” Liu’s body is in prison and he is being made to suffer deprivation of liberty, health, companionship, and more by state authorities, but he will not allow himself to be consumed by the hatred that would destroy a person with less self-control. Before being subjected to years of imprisonment and abuse, he had tried to lead a life of freedom and responsibility.

“What I demanded of myself was this: whether as a person or as a writer, I would lead a life of honesty, responsibility, and dignity,” Liu said.

Liu read his final statement after his sentencing, but was cut off after 14 minutes by the “judge” on grounds that the prosecution had spent only 14 minutes making the case against him. In it, he said, “I have no enemies and no hatred.” Liu exemplifies great self-control and mindfulness. He is focused on his goal: to foster a society of free and responsible persons who live together in equal freedom and mutual respect. He refuses to be consumed by hatred, because hatred and bitterness foster violence, cruelty, revenge, and oppression.

“Hatred can rot away at a person’s intelligence and conscience,” Liu said. “Enemy mentality will poison the spirit of a nation, incite cruel mortal struggles, destroy a society’s tolerance and humanity, and hinder a nation’s progress toward freedom and democracy. That is why I hope to be able to transcend my personal experiences as I look upon our nation’s development and social change, to counter the regime’s hostility with utmost goodwill, and to dispel hatred with love.”

I blink away tears every time I read Liu Xiaobo’s final statement. He not only refuses to be consumed by hatred, he takes the unjust occasion of his imprisonment to express his love for his wife, Liu Xia, who was forbidden by the state authorities to see him.

“I am serving my sentence in a tangible prison, while you wait in the intangible prison of the heart,” he said. “Your love is the sunlight that leaps over high walls and penetrates the iron bars of my prison window, stroking every inch of my skin, warming every cell of my body, allowing me to always keep peace, openness, and brightness in my heart, and filling every minute of my time in prison with meaning. My love for you, on the other hand, is so full of remorse and regret that it at times makes me stagger under its weight. I am an insensate stone in the wilderness, whipped by fierce wind and torrential rain, so cold that no one dares touch me. But my love is solid and sharp, capable of piercing through any obstacle. Even if I were crushed into powder, I would still use my ashes to embrace you.”

Has there ever been a more poignant expression of love? “I would still use my ashes to embrace you.” As another translation phrased it, “I will embrace you with my ashes.”

Liu had spent time in America teaching, but in 1989, as the Tiananmen Square protests gathered adherents, he left the safety of Columbia University to return to China. For his efforts to avert violence and save lives, he was arrested and brutalized for two years without charge. Although he had come to believe that some elements of the political cultures of “the West” could be helpful to China, he was not blinded to the flaws of other countries, cultures, and political systems. In a 1989 essay written in New York (“Chinese Politics and China’s Modern Intellectuals”), he resolved to be true to himself.

If I, as a person who has lived under China’s autocratic system for more than thirty years, want to reflect on the fate of humanity or on how to be an authentic person, I have no choice but to carry out two critiques simultaneously. I must:

1. Use Western civilization as a tool to critique China.

2. Use my own creativity to critique the West.

Liu Xiaobo exemplifies, to perhaps the highest degree, the self-control of a human being. The Chinese state has imprisoned him, but he refuses to be their willing slave. He refuses to be controlled by the state. They cannot even make him hate them.

Many of Liu’s papers are collected in his book No Enemies, No Hatred: Selected Essays and Poems, by Liu Xiaobo, ed. by Perry Link, Tienchi Martin-Liao, and Liu Xia (Cambridge, MA: Harvard University Press, 2012).


The death of Liu Xiaobo from liver cancer on July 13, under guard at a hospital in Shenyang, marks the passing of a great defender of freedom—a man who was willing to speak truth to power. As the lead signatory to Charter 08, which called for the rule of law and constitutional government, Liu was sentenced to 11 years in prison for “inciting the subversion of state power.” Before his sentencing in 2009, Liu stood before the court and declared, “To block freedom of speech is to trample on human rights, to strangle humanity, and to suppress the truth.” With proper treatment and freedom, Liu would have lived on to voice his support for a free society. 

While Liu’s advocacy of limited government, democracy, and a free market for ideas won him the Nobel Peace Prize in 2010, China’s leadership viewed him as a criminal and refused to allow him to travel to Oslo to receive the award. Instead, the prize was placed on an empty chair at the ceremony, a lasting symbol of Liu’s courage in the face of state suppression. Beijing also prevented liberal Mao Yushi, cofounder of the Unirule Institute, from attending the ceremony to honor Liu. 


The mistreatment of Liu, and other human rights’ proponents, is a stark reminder that while the Middle Kingdom has made significant progress in liberalizing its economy, it has yet to liberate the minds of the Chinese people or its own political institutions. 

The tension between freedom and state power threatens China’s future. As former premier Wen Jiabao warned in a speech in August 2010, “Without the safeguard of political reform, the fruits of economic reform would be lost.” Later, in an interview with CNN in October, he held that “freedom of speech is indispensable for any country.”

Article 33, Section 3, of the PRC’s Constitution holds that “the State respects and protects human rights.” Such language, added by the National People’s Congress in 2004, encouraged liberals to test the waters, only to find that the reality did not match the rhetoric.

The Chinese Communist Party pays lip service to a free market in ideas, noting: “There can never be an end to the need for the emancipation of individual thought” (China Daily, November 16, 2013). However, Party doctrine strictly regulates that market. Consequently, under “market socialism with Chinese characteristics,” there is bound to be an ever-present tension between the individual and the state.

In an interview with the Wall Street Journal (September 22, 2015), President Xi argued that “freedom is the purpose of order, and order the guarantee of freedom.” The real meaning of that statement is that China’s ruling elite will not tolerate dissent: individuals will be free to communicate ideas, but only those consistent with the state’s current interpretation of “socialist principles.” 

This socialist vision contrasts sharply with that of market liberalism, which holds that freedom is not the purpose of order; it is the essential means to an emergent or spontaneous order. In the terms of traditional Chinese Taoism, freedom is the source of order. Simply put, voluntary exchange based on the principle of freedom or nonintervention, which Lao Tzu called “wu wei,” expands the range of choices open to individuals. 

Denying China’s 1.4 billion people a free market in ideas has led to one of the lowest rankings in the World Press Freedom Index, compiled by Reporters without Borders. In the 2016 report, China ranked 176 out of 180 countries, only a few notches above North Korea—and the situation appears to be getting worse. Under President Xi Jinping’s consolidation of power in preparation for this year’s Party Congress, the websites of liberal think tanks, such as the Unirule Institute, have been shut down, and virtual private networks (VPNs) are being closed, preventing internet users from circumventing the Great Firewall.

Liu’s death is a tragic reminder that China is still an authoritarian regime whose leaders seek to hold onto power at the cost of the lives of those like Liu who seek only peace and harmony through limiting the power of government and safeguarding individual rights.