Cato Op-Eds

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

In 1993, Jony Jarjiss entered the United States on a temporary visa for fiancés of U.S. citizens. The relationship fell apart, and in 1994, an immigration judge ordered his removal for overstaying his visa. Iraq refused to accept him back, and so for 23 years, Jarjiss has checked in with immigration authorities. Then, due to a new deal with Iraq, the Trump administration arrested him in July 2017 and is now attempting to deport him. He is “terrified” of the persecution that he may face as a Christian upon his return and is attempting to reopen his immigration case. The government is trying to remove him before he has that opportunity.

During the campaign, President Trump promised Iraqi and Syrian Christians protection in the United States, and they rewarded him with their votes. Yet in March, the United States struck a bargain with the government of Iraq: President Trump would leave Iraq off his travel ban executive order in exchange for Iraq accepting 1,400 Iraqis subject to deportation orders.

The deal was strange considering the president argued letting in any Iraqis could let “dangerous people” enter. But in any case, Immigration and Customs Enforcement (ICE) began rounding up Iraqis in June and has arrested 279 so far, about half of whom were ordered removed from the United States at least a decade ago. Most are Chaldean Christians like Jarjiss.

As their lawyers scrambled to find them legal representation to challenge their removals in immigration courts, the United States government began shipping them from Michigan across the country to 26 different states, obfuscating their efforts to obtain lawyers, and pressuring them to sign away their rights to challenge. ICE prison guards harassed Iraqis in what the ACLU alleges was a concerted effort to intimidate them. Jarjiss was shipped to Ohio.

Finally, in July, a federal district court judge temporarily blocked the removals in the case of Hamama v. Adducci. The judge’s order explains that the Iraqis’ efforts have “been significantly impeded by the Government’s successive transfers of many detainees across the country, separating them from their lawyers and the families and communities who can assist in those legal efforts,” and that these people are “confronting the grisly fate Petitioners face if deported to Iraq.” The order allowed Iraqis to file motions to reopen their removal cases in immigration courts.

Since July, immigration courts have granted 87 percent of all of these motions. In the 10 cases where judges have ruled on the merits, all received the right to remain the United States. Yet ICE continues to detain all of the others. In more difficult cases, it could take years for judges to reach the merits of the case, but ICE refuses to consider applications for release from Iraqis.

The ACLU is now challenging their continued detention. Under Supreme Court precedent, ICE cannot detain immigrants indefinitely. Their removal must be “reasonably foreseeable,” and the ACLU is demanding evidence that the removal would meet that standard. They also assert that prolonged detention requires an “individualized hearing before an impartial adjudicator.”

Most of these Iraqis were legal permanent residents, and they have already gone through immigration court proceedings at some point. Half received orders of removal at least decade ago. They lost their cases at that time. Some lost because they could not credibly claim a fear of persecution at that time; others because they lacked attorneys to represent them in court; others because they failed to apply for asylum in the time limit; and still others because they committed crimes that bar them from relief.

Until this year, the Iraqi government would only accept deportees with unexpired Iraqi passports, and it refused to issue new passports to those who the United States attempted to deport. The government has portrayed all of these people as serious criminal threats who need to be removed, but about 75 percent committed only nonviolent crimes years or decades ago. Even those barred from asylum still may have claims under the Convention against Torture. Some simply overstayed a temporary visa.

Moayad Jalal Barash who came to the United States when he was a young child served time for a drug conviction when he was 17. He is now 47 and has grandchildren in the United States. Another man served 2 years for a drug charge in 1987. Najah Konja has lived in the United States for 40 years and served time for a drug convictions decades ago. Jihan Asker paid $150 fine for a misdemeanor in 2003 and has not been arrested in 14 years since. Some like Jarjiss have no criminal history at all.

Regardless of the specifics, the U.S. government should not be attempting to thwart the legal rights of people. In 2017, the U.S. State Department described a “genocide against Yezidis, Christians, and Shia Muslims” in areas of Iraq controlled by ISIS. Other credible reports of persecution by Shia militia against Muslims have surfaced. The United States should not deport nonviolent people to places where they could be killed.

Reporters and pundits are claiming that Republicans are pursuing a tax giveaway to the rich. In fact, the proposed tax changes, especially in the early years, heavily favor middle-income households.

The skewed media coverage of the GOP tax plan is partly based on a slanted presentation of results by the Tax Policy Center (TPC). As I have noted, TPC has expert analysts and does a lot of great work, but they tilt their reports to make it appear that high-earners are favored by recent Republican proposals.

The summary of TPC’s new report on the House GOP tax plan says, “The largest cuts, in dollars and as a percentage of after-tax income, would accrue to higher-income households.” The largest cuts in dollars go to high-earners—no kidding! TPC’s own data (current law data is here) reveal that the top quintile of households will pay an average of about $65,000 in income and estate taxes in 2018, compared to about $3,200 for the middle quintile. Thus, it is very difficult to cut taxes without the top group receiving the largest dollar cuts.

The TPC report presents the data shown in columns 1 and 2 below, which suggest that GOP cuts favor high earners. However, based on TPC data, I calculated columns 4 and 6. Column 4 reveals that the lower and middle groups would receive the largest percentage cuts when total federal taxes are the denominator.  

But, as I discuss here, column 6 presents the best data showing the relative size of GOP cuts. It shows income and estate tax cuts as a percentage of current income and estate taxes paid. Under the GOP plan, the middle quintile gets a big 26 percent tax cut, on average, while the top two quintiles would receive cuts of 16.9 percent and 7.4 percent. Looked at this way, middle earners would get the largest tax cuts under the GOP plan.

President Trump’s nomination of Jerome Powell as the next chairman of the Federal Reserve System is a bet that he will continue Janet Yellen’s policies and not rock financial markets. The expectation is that Powell will follow the Fed’s already-announced normalization schedule, which calls for slowly reducing the Fed’s $4.2 trillion balance sheet, by rolling off maturing mortgage-backed securities (MBS) and longer-term Treasuries, and gradually increasing the target range for the fed funds rate.[1]

The presidential vote of confidence for Powell reflects the White House and Treasury’s desire for low interest rates to fund the public debt and support high asset prices, as well as the probability that the Senate will quickly confirm the nominee.

Mr. Powell also appears open to revisiting financial regulations, such as the Volcker rule and regulations that discriminate against smaller banks. In testimony, before the Senate Committee on Banking, Housing, and Urban Affairs, on June 22, 2017, Governor Powell set forth “Guiding Principles to Simplify and Reduce Regulatory Burden.” One of the key principles is that Fed policymakers “should assess whether we can adjust regulation in common-sense ways that will simplify rules and reduce unnecessary regulatory burden without compromising safety and soundness.” He emphasized that the Fed’s goal should be “to establish a regulatory framework that helps ensure the resiliency of our financial system, the availability of credit, economic growth, and financial market efficiency.” However, during his tenure on the Fed’s Board of Governors since May 2012, he has consistently voted in favor of tightening the Fed’s grip on financial regulation.[2] Thus, one must remain skeptical about whether he would embrace market-friendly deregulation.

Unconventional Monetary Policy and Uncertainty

Unconventional monetary policy — in the form of quantitative easing (i.e., large-scale asset purchases) and ultra-low interest rates — has misallocated credit, distorted interest rates, encouraged risk taking, inflated asset prices, fueled government deficit spending, and done little to promote long-run private investment.[3]

By purchasing massive amounts of high-risk MBS and long-term government bonds, the Fed helped lower longer-term interest rates but steered credit away from private investment, which was also impeded by stricter macro-prudential regulations. Moreover, by keeping short-run interest rates near zero for more than seven years, paying interest on excess reserves (IOER) above the effective fed funds rate, and convincing markets that rates would stay low for a long time (forward guidance), the Fed has increased the reach for yield and appears more interested in priming Wall Street than in letting markets set interest rates and allocate credit.

The Fed’s current operating procedure is to administer the target range for the fed funds rate using IOER and reverse repos, in contrast to the pre-crisis arrangement whereby the Fed’s open market desk bought or sold short-term Treasuries to increase or decrease bank reserves, and then let market participants determine the effective funds rate.[4] As a result, changes in the monetary base are no longer “high powered” — the money multiplier has collapsed and the monetary transmission mechanism that prevailed prior to 2008 is broken.

Unconventional monetary policy, macro-prudential regulation, and the lack of any monetary rule have increased uncertainty about the future of monetary policy and, thus, have had a negative effect on private investment. To his credit, Mr. Powell has recognized some, but not all, of these problems. In January this year, he told members of the American Finance Association in Chicago that “the current extended period of very low nominal rates calls for a high degree of vigilance against the buildup of risks to the stability of the financial system.” However, he downplayed that risk by saying that “the bottom line is that there has not been an excessive buildup of leverage, maturity transformation, or broadly unsustainable asset prices.”[5]

The Limits of Monetary Policy

The purpose of the Bernanke-Yellen monetary policy has been to lower longer-term rates and pump up asset prices creating a wealth effect to spur spending and real economic growth. But there has been a differential impact favoring the housing and government sectors, while private investment has been sluggish due to regime uncertainty, regulatory costs, and a fall in the private saving rate. From a long-run perspective, the Fed cannot permanently increase wealth by monetary policy. Powell recognizes the limits of monetary policy when he notes that “ultimately, the only way to get sustainably higher interest rates is to improve the broader environment for growth, by adopting policies designed to increase productivity and potential output over the long term — policies that are mainly outside the scope of our work at the Federal Reserve.”[6] Recognition of the limits of monetary policy is an important first step toward sound monetary policy.

The Schizophrenic Nature of Fed Policy

In his January speech, however, Powell is silent on the schizophrenic nature of Fed policy — a policy designed to increase risk taking, allocate credit to favored sectors, and stimulate economic activity but that plugs up the monetary transmission mechanism by paying IOER, and discourages lending to productive ventures by an onerous system of macro-prudential regulation. As market analyst Brian Barnier notes, “Low interest rates don’t help if companies face high risk and uncertainty. Central banks that cause volatility and uncertainty have been defeating their own interest rate actions.”

Normalizing Monetary Policy and Instituting a Rules-Based Regime

Regime uncertainty could be reduced by first normalizing monetary policy by reducing the size of the Fed’s balance sheet and ultimately eliminating IOER and restoring a market-driven fed funds rate. A rule-based monetary regime could then be instituted to guide monetary policy. In the present unconventional regime — with the absence of a competitively determined fed funds rate and a weak link between base money (i.e., currency in circulation plus bank reserves), broad monetary aggregates, and nominal GDP — the implementation of monetary rules such as the Taylor rule and a final demand rule would fail.

Under unconventional monetary policy, we therefore are stuck in a fully discretionary fiat money regime — and, thus, in a fog of uncertainty. Maintaining such a system ignores the institutional uncertainty brought about by not having a credible monetary rule. Such a rule would help depoliticize monetary policy and incentivize the Fed to take a long-run perspective, thereby reducing uncertainty.[7] As Karl Brunner has pointed out regarding the knowledge problem facing monetary policymakers,

We suffer neither under total ignorance nor do we enjoy full knowledge.  Our life moves in a grey zone of partial knowledge and partial ignorance.  More particularly, the products emerging from our professional work reveal a wide range of diffuse uncertainty about the detailed response structure of the economy… . A nonactivist [rules-based] regime emerges under the circumstances … as the safest strategy. It does not assure us that economic fluctuations will be avoided.  But it will assure us that monetary policymaking does not impose additional uncertainties … on the market place.[8]

In a speech at the Forecasters Club of New York in February, Powell argued that monetary rules can help guide policy, but he sees those rules as too simple to take account of the complexity confronting policymakers. He has in mind the Taylor rule, which would set the nominal fed funds rate based on a single equation:

(1) R = r* + π + a (ππLR) + b (gap)


R = nominal federal funds rate

r* = neutral real federal funds rate

π = the inflation rate

πLR = 2 percent

gap = percentage deviation of output from its potential level or unemployment from its natural rate.

Taylor, in his 1993 article, set a = 0.5 and b = 0.5.

According to Powell,

I think it’s fair to say that simple policy rules are widely thought to be both interesting and useful, but to represent only a small part of the analysis needed to assess the appropriate path for policy. I am unable to think of any critical, complex human activity that could be safely reduced to a simple summary equation. In particular, no major central bank uses policy rules in a prescriptive way, and it is hard to predict the consequences of requiring the FOMC to do so, as some have proposed. Policy should be systematic, but not automatic.

In fact, complexity is what makes a rules-based regime desirable. No one on the Federal Reserve Board or the Federal Open Market Committee predicted the 2008 financial crisis. The purpose of a systematic, rules-based monetary regime is to keep the economy on track and prevent a sharp decline in final demand. No rule can be perfect; there is always a learning process.  But as Brunner noted, a nonactivist rule would reduce uncertainty inherent in a period-by-period discretionary monetary regime. A long-run strategy based on achieving a stable growth path of nominal final demand would avoid the type of errors associated with a purely discretionary regime.

The Great Moderation and the Case for a Final Demand Rule

During the “Great Moderation” (1987–2006), under Fed chairman Alan Greenspan, the trend rate of growth of final demand, as measured by nominal final sales to domestic purchasers (FSDP), was 5.4 percent per year — split into real growth of 3 percent and inflation of 2.4 percent.[9] Cato’s former chairman Bill Niskanen found that variation around that trend “had significant effects on asset prices and the real economy, and most of this variation was a consequence of the Fed’s response to financial crises.” Figure 1 from his 2006 Cato Journal article is reproduced below.

Figure 1: Nominal Final Sales to Domestic Purchasers

Based on his research, Niskanen concluded that the Greenspan Fed implicitly followed a final demand rule but that it overreacted in increasing demand when faced with financial crises. Niskanen sees the primary duty of the Fed as maintaining “a steady increase in aggregate demand consistent with a low target rate of inflation.”

The Great Moderation was also in line with the Taylor rule but that rule depends on knowledge of r* and the output gap, both of which are difficult to estimate. As Powell noted in his February speech, “The neutral rate changes significantly over time, and estimates of its level entail substantial uncertainty.”  Moreover, “there is particularly high uncertainty about measuring the deviation of output from its potential,” and the values of the coefficients, a and b, in Taylor’s rule need to be specified.

Niskanen prefers a final demand rule over an interest rate rule, in part, because it does not require imputing values to r*, or estimating the output gap.[10]  In the 2009 edition of Cato’s Handbook for Policymakers, he recommended that “Congress should amend the Full Employment and Balanced Growth Act of 1978 to clarify the congressional guidance on the conduct of monetary policy.” In particular, he argued that

Congress is best advised (1) to specify a target rate of increase of final sales and (2) to instruct the Federal Reserve to minimize the variance around this target rate. The target rate of increase of final sales may best be about 5 percent a year, sufficient to finance a realistic rate of economic growth of 3 percent and an acceptable rate of inflation of about 2 percent.[11]

Niskanen saw nominal FSDP as “a feasible target” because it is “almost completely determined by U.S. monetary policy, whereas the rate of economic growth and the inflation rate are separately affected by a variety of domestic and foreign conditions.” The problem is that, under the Fed’s current operating procedure, the link between base money creation and final demand has been severed. Moving to a rules-based regime thus requires normalizing monetary policy and restoring the monetary transmission mechanism as discussed earlier.

Some congressional leaders think it’s time to create a rules-based monetary regime. The Financial CHOICE Act of 2017 (H.R 10), which recently passed the House, would make the Fed responsible for specifying a monetary rule and justifying to Congress any deviations from it.[12]  Whether the CHOICE Act passes or not, it is important to consider alternative monetary rules and to be prepared to make the case for rules over discretion when the opportunity for reform arises.

The Phillips Curve Is a Poor Guide for Monetary Policy

The Phillips Curve model of the economy, which posits an inverse relationship between unemployment and inflation, has been a poor guide for monetary policy, yet the Fed still incorporates that relationship into its thinking.[13] With the rate of unemployment now at 4.1 percent, policymakers are puzzled why inflation hasn’t increased to the Fed’s target of 2 percent. They could look to their own operating procedures used since October 2008. Without IOER and Dodd-Frank type regulations, banks would be lending more, and base money would have a stronger impact on overall money growth and the price level.

Powell’s Challenge

Mr. Powell, no doubt, will be under pressure from the White House and Treasury to keep rates low — even if markets are pushing them upward. Intervening to postpone necessary adjustments, however, would only complicate future policy changes and increase the costs of adjustment.

It is essential that Powell understand the risks involved in the post-2008 operating techniques and the underpricing of risk that unconventional monetary policy has occasioned. His challenge will be making the transformation to a new policy regime that gets the Fed out of the business of allocating credit and pegging interest rates at artificial levels.


Congress has ultimate authority for monetary policy. During the confirmation process, there needs to be a discussion of the limits of monetary policy and how Mr. Powell sees the future of monetary policy, and the steps he would take in a crisis situation. Finally, Congress needs to make the Fed accountable for its mistakes and ensure it abides by the rule of law.


[1] See George Selgin, “Operation SNAIL,” Alt-M, September 26, 2017.

[2] See Binyamin Appelbaum, “In Choice of Fed Chairman, Trump Downgrades Deregulation,” New York Times, October 29, 2017.

[3] There is no doubt that nonmonetary forces have contributed to historically low interest rates. Real rates have been declining for some time, due to slower productivity growth, demographics, and other factors (see Powell’s January 7, 2017 speech). It is hard to deny, however, that Fed policy has not contributed to the low-interest environment and helped fuel selected asset prices. Indeed, Powell, in his January address to the American Finance Association, argued that, with regard to the impact of “highly accommodative monetary policies, … studies generally show that they lowered rates across the curve and moved other asset prices as well.” At the same time, he admitted that “isolating the effects of these policies is challenging.” If Mr. Powell is correct that rates are mostly reflecting nonmonetary factors, then asset prices may be sustainable, but if he is wrong, then there is a strong possibility that as the Fed exits its unconventional polices, there will be a significant market correction.

[4] For a detailed discussion of the Fed’s pre- and post-crisis operating procedures, see George Selgin, “Interest on Reserves and the Fed’s Balance Sheet,” testimony before the House Subcommittee on Monetary Policy and Trade, May 17, 2016. See also Norbert Michel and Selgin, “Fed Must Stop Rewarding Banks for Not Lending,” American Banker, May 30, 2017.

[5] The IMF is less sanguine. In its latest Global Financial Stability Report (October 2017), the IMF raises “concerns about a continuing buildup in debt loads and overstretched asset valuations [that] could have global economic repercussions” (p. 42).

[6] Jerome Powell, “Low Interest Rates and the Financial System,” Speech at the 77th Annual Meeting of the American Finance Association, Chicago, January 7, 2017.

[7] Thomas Hoenig, vice chairman of the Federal Deposit Insurance Corporation and former president of the Federal Reserve Bank of Kansas, has argued “that monetary and regulatory policies have for some time been overly focused on short-run effects at the expense of long-run goals, which has unintentionally served to increase uncertainty and economic fragility.”  See Hoenig, “The Long-Run Imperatives of Monetary Policy and Macroprudential Supervision,” Cato Journal (Spring/Summer 2017), p. 195.

[8] Karl Brunner, “The Control of Monetary Aggregates,” in Controlling Monetary Aggregates III, p. 61. Boston: Federal Reserve Bank of Boston, 1980.

[9] Final sales to domestic producers (FSDP) is defined as “the sum of nominal gross domestic product  plus imports minus exports minus the change in private inventories.” See Niskanen, “Monetary Policy and Financial Regulation,” in Cato Handbook for Policymakers (2009), 7th ed., p. 377.

[10] In his 1992 Cato Journal article, “Political Guidance on Monetary Policy,” Niskanen examined three viable monetary rules: (1) targeting the price of gold or a broad price index, (2) targeting a monetary aggregate, and (3) targeting nominal GDP or domestic final sales. He argued that “any one of these rules would be better than guidance based on interest rates or exchange rates, or on any real variable such as the growth of output or the level of the unemployment rate” (p. 281). His preferred rule, however, is to minimize “the variance around an approved target path of nominal domestic final sales” — an objective that “is probably the most that can be expected of monetary policy” (p. 285).

[11] Market monetarists, such as Scott Sumner and David Beckworth, prefer to target nominal GDP (NGDP) rather than final sales. Their arguments for a final demand rule, however, are similar to Niskanen’s. For example, Beckworth argues that “a NGDP target aims to stabilize total dollar spending. It is one target that has embedded in it both the supply of and the demand for money (i.e. total dollar spending = money supply x velocity of money). The beauty of a NGDP target is that the Fed does not need to know what is exactly happening to the money supply or money demand. All the Fed only needs to worry about is the product of the two components. There is no need to track the money supply or estimate money demand. By focusing on total dollar spending, the Fed will be fostering a stable monetary environment where movements in money supply and money demand are offsetting each other.” Bill Woolsey, in comparing a monetary rule targeting NGDP versus one targeting FSDP, finds no significant difference.

[12] See Title X of H.R. 10: “Fed Oversight Reform and Modernization.” H.R. 10 also calls for a Centennial Monetary Commission to examine the Fed’s history and to recommend reforms.

[13] See, e.g., J. A. Dorn, “It’s Time to Bury the Phillips Curve,” Investor’s Business Daily, September 26, 2017.

[Cross-posted from]

As Republicans work on details of their tax legislation, many reporters and pundits are painting the effort as slanted in favor of the rich. But that is not what a fair reading of the official data shows. The House bill proposed last week provides the largest percentage cuts to middle-income taxpayers, at least in the early years.

The Joint Committee on Taxation (JCT) has published estimates on the House bill. In the table below, the first three columns show the results for 2019, which is the first year in the JCT analysis.

Column 3 shows that middle-income groups would enjoy somewhat larger percentage tax cuts than higher-income groups under the GOP plan. Those are the raw JCT results.

However, the JCT calculates its results using a denominator that not only includes income taxes, but also payroll and excise taxes, even though Republicans are not changing those taxes. JCT’s inclusion of payroll and excise taxes slants its presentation to exaggerate the benefits to higher groups compared to lower groups.

Columns 4 and 5 include just individual and corporate income taxes since those are the taxes that Republicans are cutting. Column 4 shows my rough estimate (details below) of the 2019 totals of those taxes since the JCT did not provide those figures. In aggregate, households earning less than $40,000 do not pay any income taxes, so they are n/a.

Column 5 shows the GOP tax cuts as a percent of estimated income taxes paid under current law. You can see that using this more sensible denominator, the results show huge percentage tax cuts for middle earners. Households in the $40,000 to $50,000 group would get their income taxes cut almost in half, in aggregate. The House bill would make the tax code substantially more “progressive,” which is not a good idea since the code is already far too progressive or unequal.

Two caveats. The JCT does not include estate tax changes. Also, I focused on 2019, but over time the tax cuts bend more toward the top end. However, my main point is that distribution tables can show the same model results in very different ways, suggesting different messages. To me, the adjusted JCT data reveal that the GOP has moved too far left in designing its tax package.

The most important goal of tax reform is to raise economic output and incomes over the longer term, and the GOP’s business tax reforms will do that. However, because the GOP tilted left on its individual changes, those changes will not spur any economic growth, according to the Tax Foundation. Sadly, that looks like it could be a missed opportunity for tax reform this year.

Data Note: I could not find an estimated breakout of federal taxes between income, payroll, and excise for 2019 on JCT’s website. So I used a recent TPC estimate for 2019 (T17-0045) of the current law breakdown in the same income categories as JCT. Since TPC’s tax and income estimates are somewhat different than JCT’s, I adjusted the TPC data to match the JCT figures.

Since the 1986 Immigration Reform and Control Act (IRCA) came into force, it has been against the law for illegal immigrants to work in the United States. Prior to IRCA, they could be deported if discovered by the Immigration and Naturalization Service (INS), that era’s federal immigration agency, but illegal immigrants were not barred from working. For decades prior to 1986, immigration restrictionists and labor unions had tried to make illegal immigrant employment illegal. Prior to IRCA, the INS even instituted the Texas Proviso whereby they wouldn’t enforce harboring or other statutes against employers of illegal immigrants. IRCA ended this arrangement in 1986.

Since November 6, 1986, everybody seeking a job in the United States must fill out an I-9 form at the point of hire to supposedly prove that they are legally allowed to work in the United States. The new hire must show the employer some documents that corroborate the information on the I-9 form, copies of which the employer must keep on site. This requirement has had multiple negative effects. 

The first is that it has decreased the wages of illegal immigrants relative to similarly skilled native-born Americans and legal immigrants by diminishing employer demand for their labor. While this is the point of the I-9 check, it is disingenuous of immigration restrictionists to complain about slow immigrant wage convergence when the I-9 regulations that they support are slowing down that convergence for lower-skilled immigrants.

The second negative effect of IRCA is its creation of a large-scale black market for legal documents in the United States. The value of document fraud increased after ICRA because false documents became necessary for illegal immigrants to fill out an I-9 form to work. Since working is the main reason immigrants, especially illegal immigrants, come to the United States, a growing cottage industry of black market identity documents rose to serve them.

This is where identity theft enters the immigration debate. Many illegal immigrants in the United States work and fill out I-9 forms which many voters think means that they must have stolen somebody’s identity and committed a felony. That’s not necessarily true. Aggravated identity theft is a felony that requires the user to knowingly use another person’s identity, but falsely using another person’s identity is a misdemeanorMerely purchasing an identity from another person and using that for employment purposes doesn’t count as a felony because the purchaser does not know if it belonged to somebody else because it could be fake. 

In popular terminology, as opposed to legal terminology, anybody who uses somebody else’s identification without the owner’s permission commits identity theft. This makes it complicated for supporters of legalization to argue that illegal immigrants who are otherwise law-abiding should have a path to citizenship because, if they have worked on an I-9, then there is a good chance that they’ve worked using somebody else’s identity.

The Internal Revenue Service (IRS) further complicated this issue by granting Individual Taxpayer Identification Numbers (ITIN) to foreigners ineligible for a Social Security Number (SSN). ITINs are available for foreign investors, spouses of foreign workers, and others so they can pay taxes. Illegal immigrants understandably take advantage of the ITIN. A recent government report suggests that about 1.2 million ITIN filers are also using a name or SSN that does not match their ITIN.

Most illegal immigrant use of other people’s identities is likely for employment purposes. Rarely does it result in truly awful crimes. Virtually all employment-related identity theft would disappear if illegal immigrants were legalized and future immigrant flows were liberalized because immigrants wouldn’t have to steal identities in order to work. The best solution would be to remove the I-9 requirement entirely and shrink, or separate, the SSN or other government-issued identification requirements for employment authorization. If there is no reason to steal SSNs for employment then nobody will steal them for employment. But those arguments only get legalization proponents so far against people who are disgusted by rampant employment-related identity theft.

Many of the so-called instances of employment-related identity theft committed by illegal immigrants are actually identity loans where the owner of a government-issued identity allows the illegal immigrant worker to use that legal identity for work purposes. There are few statistics available to estimate the frequency of identity loans but they appear to be commonplace. One survey-based paper found that immigrant farmworkers estimated that 50 to 70 percent of their coworkers used loaned documents whose procurement is usually arranged by a supervisor, friend, colleague, or a fourth party.

Other family members frequently supply the loaned identities. For example, many unlawful immigrants use the SSNs of parents, children, or other family members who either voluntary lend the information or are not using it – a form of family wealth. Oftentimes, the actual owner of the identity is outside of the country or too young to work. A good example of this is the story of Manuel:

An unauthorized immigrant from Zacatecas, Manuel had first settled in the San Francisco Bay Area and sought work in construction. However, because the construction industry in San Francisco is unionized, Manuel had discovered that potential employers would check his I-9 form with E-Verify, a federal database that matches Social Security cards and names. So Manuel contacted an uncle in Zacatecas who had obtained a Social Security card in the early 1970s when he had first migrated to California—to ask about using his card. His uncle, who had returned to Mexico permanently with no plans to reenter the United States, agreed.

When next applying for work, Manuel presented his uncle’s Social Security card and a fake legal permanent resident card (green card) bearing his uncle’s name—which he had purchased from a local vendor—to the employer. The employer, in turn, was able to verify that the SSN was on file with the Social Security Administration (SSA) and that it matched the uncle’s name. This ploy allowed Manuel the luxury of finding work in a sector of the economy normally closed to the unauthorized. In entering a unionized construction job, Manuel benefited from more comfortable work conditions and employer-provided benefits. Moreover, in contrast to what he had earned in agriculture—typically about $25,000 a year—Manuel was able to earn $60,000 to $80,000 a year in construction.

Some researchers claim that the lending of identity is a form of community-wide mutual assistance or insurance but the family property explanation is better.  

Sometimes, the fourth party identity-loaner will loan his or her own identity so that the taxes paid by the illegal immigrant worker will contribute to welfare fraud. In this scenario, taxes paid by the illegal immigrant identity-borrower count toward government benefits like unemployment or Social Security for the identity owner. Sometimes, the American identity owner even pays the illegal immigrant to work with their identity in order to commit welfare fraud. This brief description explains how the scam works:

Elisabeta remembers that she had just arrived in the United States when her neighbor, Amparo, approached her to ask whether Elisabeta would “work” her papers (trabajar sus papeles). At the time, Elisabeta was barely 15 and was a single girl seeking work in agriculture, an industry dominated by males. She had come to the United States to earn money to send to her recently widowed mother, who lived with Elisabeta’s older sisters on the family farm in rural Jalisco, Mexico. Yet before she could get to work, she needed to procure the identity documents—a fake Social Security card and a “green card” (or resident alien visa)—that would allow her to find a job. She lacked the cash necessary to purchase a set of fake papers at a flea market or from a fly-by-night document mill. Thus, Elisabeta faced the kind of dilemma that plagues many unauthorized new arrivals: Without “papers,” she could not find employment. And yet, without employment, how could she possibly purchase “papers”?

Amparo drew on her knowledge of Elisabeta’s situation when she made the case for the loan. “Why don’t you just work my papers?” she had proposed. “Because being recién llegada (recently arrived) and all, where will you possibly get them otherwise?”

Amparo herself was in her midfifties. She walked slowly and stiffly. As she grew older, she told Elisabeta, working in the fields made her back hurt. So Amparo offered Elisabeta what seemed to the latter like a bargain. If Elisabeta used Amparo’s mica (green card) and seguro (Social Security number) to find work, Amparo would pay her an additional $100 for every $1,000 that she earned.

The “tip” that Amparo offered to entice Elisabeta to “work” her documents is the kind of financial incentive that often accompanies document exchange in California’s Central Valley. While the exchange appeared advantageous to Elisabeta, Amparo herself also stood to benefit. A person’s unemployment payments are based on the calendar quarter within the previous 12-month period during which he or she earned the highest amount. Thus, while Elisabeta would gain identity documents that would allow her to find work, her work history would fatten Amparo’s unemployment checks at the end of the season.

A preliminary analysis of law enforcement press releases, far from a precise measure, reveals a few patterns that should be the basis for future research. For instance, when non-citizen immigrants commit identity theft, they mostly do so for employment purposes. When American citizens commit identity theft, they usually do so in order to commit welfare fraud, financial fraud, or in order to sell identities to illegal immigrants either on their own or as part of a conspiracy that involves the immigrants themselves. For example, some of the instances of identity theft committed by U.S. citizens were committed to steal wages, create fake schools to supply “student” visas, marriage fraud, H1-B visa fraud, visa fraud and harboring, manufacturing and selling fraudulent identities, EB-5 visa fraud, defrauding migrant clients, bribery, and illegal issuing of driver’s licenses, among others. In many of these cases, immigrants would purchase the identities but U.S. citizen intermediaries initially stole them.

Borrowing another person’s identity with the owner’s permission for employment purposes can still be a crime but it is a lot less egregious than taking somebody’s identity without their permission. Working in the United States is extremely valuable but government immigration laws make it very difficult for foreigners to do so lawfully. Most illegal immigrants have to work on a legal or false identity in order to earn a wage, which incentivizes identity loans and identity theft. Although researchers need to conduct more detailed quantitative and qualitative work on identity loans borrowed and identity theft committed by illegal immigrants, it is a more nuanced issue with fewer and different victims than many commentators realize.

Special thanks to Jen Sidorova for her help on this blog post. 

An overwhelming majority (82%) of Americans agree that “it would be hard to ban hate speech because people can’t agree what speech is hateful,” the Cato 2017 Free Speech and Tolerance Survey finds. Seventeen percent (17%) disagree. Majorities across partisan and demographic groups alike agree that hate speech is hard to define and thus may be hard to regulate.

Full survey results and report found here.

How Do Americans Define Hate Speech?

When presented with specific statements and ideas, Americans can’t agree on what speech is hateful, offensive, or simply a political opinion

Besides slurs and biological racism, Americans are strikingly at odds over what speech and ideas constitute hate.[1] For instance, a majority of Democrats (52%) believe saying that transgender people have a mental disorder is hate speech. Only 17% of Republicans agree. On the other hand, 42% of Republicans believe it’s hateful to say that the police are racist, while only 19% of Democrats agree.

Among all Americans, majorities agree that calling a racial minority a racial slur (61%), saying one race is genetically superior to another (57%), or calling gays and lesbians vulgar names (56%) is not just offensive, but is hate speech. Interestingly a majority do not think calling a woman a vulgar name is hateful (43%), but most would say it’s offensive (51%). Less than half believe it’s hateful to say that all white people are racist (40%), transgender people have a mental disorder (35%), America is an evil country (34%), homosexuality is a sin (28%), the police are racist (27%), or illegal immigrants should be deported (24%). Less than a fifth believe it’s hateful to say Islam is taking over Europe (18%) or that women should not fight in military combat roles (15%).

Liberals and Conservatives Define Hate Speech Differently

Liberals and conservatives significantly diverge over what speech they define as hateful, offensive, or simply an opinion.

Majorities of liberals say that slurs against racial minorities (81%) and LGBT people (73%), saying that one race is genetically superior to others (75%), or saying transgender people have a mental disorder (59%) are hateful. Strikingly, majorities of conservatives don’t think any of these ideas are “hateful” although most consider them “offensive” or hateful.[2] In fact, conservatives are about 40 points less likely than liberals to think that saying transgender people have a mental disorder (17% vs. 59%) or saying racial slurs (43% vs. 81%) are hateful. While strong majorities of conservatives agree these are at least offensive or hateful, they are less likely to equate these phrases and ideas with hate specifically.

Although majorities of conservatives did not find any of the statements included on the survey hateful, they were more likely than liberals to find several statements hateful. First, conservatives are about twice as likely as liberals to think it’s hateful to say the police are racist (39% vs. 17%). Second, conservatives are somewhat more likely to believe it’s hateful to say that America is an evil country (39% vs. 29%). Third, conservatives are somewhat more likely than liberals to think it’s hateful to say that all white people are racist (44% vs. 35%).

Liberals are more likely than conservatives to view a variety of political opinions and speech as either offensive or hateful.

Liberals are more than 40 points more likely than conservatives to think it is offensive or hateful for a person to say that homosexuality is a sin (90% vs. 47%), women shouldn’t fight in military combat roles (87% vs. 47%), illegal immigrants should be deported (80% vs. 36%), or Islam is taking over Europe (79% vs. 33%). Not even a majority of conservatives find these statements to be offensive or hateful.

Notice that two of these, women fighting in combat roles and deporting illegal immigrants, are policy positions that a substantial number of Americans hold. For instance, a Quinnipiac survey found 22% of Americans believe unauthorized immigrants should be “required to leave” the United States. A Fox News poll found 26% don’t think women should fight in military combat roles. Furthermore, the perception that the police allow racial bias to impact their jobs is also a view held by about a quarter (26%) of Americans, according to a Quinnipiac survey.

Yet, to merely express these as political positions or perceptions would also be viewed as highly offensive to a large share of the population.


These data demonstrate why it would be difficult to regulate hateful or offensive speech—Americans can’t agree what speech is hateful or offensive. Across a variety of statements, liberals and conservatives varied dramatically in the speech they found hateful, offensive, or neither hateful nor offensive.

Furthermore, what may be a policy preference or an assessment of public policy for one person, such as deporting unauthorized immigrants or perceiving a racial bias in policing, may be highly offensive and hateful to another. Thus, attempts to regulate offensive speech would have the effect of shutting down dialogue and erect barriers to public policy debate.

Full survey results and report found here.

Sign up here to receive forthcoming Cato Institute survey reports.

The Cato Institute 2017 Free Speech and Tolerance Survey was designed and conducted by the Cato Institute in collaboration with YouGov. YouGov collected responses online August 15–23, 2017 from a national sample of 2,300 Americans 18 years of age and older. The margin of error for the survey is +/- 3.00 percentage points at the 95% level of confidence.

[1] In this report, biological racism refers to a belief that some races are genetically superior or inferior to one another.


The Obama administration ramped-up and sustained interior immigration enforcement operations through the end of FY2013, which was the longest such period of sustained enforcement in U.S. history. President Obama inherited an expanding immigration enforcement apparatus and built on it further by making “Secure Communities” mandatory in near every county of the United States, appointing immigration enforcer and former-Arizona governor Janet Napolitano as the head of DHS, and treating Central American asylum-seekers in a heartless fashion.     

Immigration restrictionist groups and some of President Obama’s supporters hid or excused the fact that President Obama’s interior enforcement operations were so extensive. The restrictionists argued that President Obama’s deportation numbers were puffed up to include those captured at the border. Their argument contained just enough truth to pass a 10-second investigation but ignored the fact that ICE removals from the interior of the United States were higher for a longer period under Obama than for any other President (Figure 1). 

Figure 1

ICE Removals from the Interior of the United States


Source: ICE FOIA Library.

Although the Obama administration did sustain a high number of removals, it also focused more on criminals (Figure 2). The latter category included those who committed victimless crimes, like immigration violations, as well as real crimes like violent and property offenses. In January 2016, 53 percent of those removed by ICE were criminals compared to just 33 percent when Obama first took office in January 2009 (Figure 3). In March 2017, the last month for which I have full data, criminals were 59 percent of all removed immigrants. Trump’s deportation priorities haven’t affected that yet.

Figure 2

ICE Removals from Interior of the United States by Criminality


Source: ICE FOIA Library.

Figure 3

Criminal ICE Removals from the Interior of the United States


Source: ICE FOIA Library, author’s calculations.

It is difficult to judge the Trump administration’s removal record after less than a year in office.  After all, reorienting the enforcement priorities of a bureaucracy as large as the Department of Homeland Security takes time.  However, a few facts do stand out. 

First, ICE arrests are up dramatically during the Trump administration. 

Second, ICE removals from the interior of the United States are down since February 2017. Obama’s ICE deported 21,705 people per month from the interior of the United States from October 2016 through the end of January 2017, before Trump took office. However, President Trump has managed, so far, to deport an average of only 17,142 people per month – 21 percent below Obama’s record for the first four months of the fiscal year. The February-September 2017 removal number is 14.6 percent lower than for the same period in 2016 (Figure 4). 

Figure 4

ICE Removals from the Interior of the United States, Monthly Average from April-October 2017.


Sources: ICE FOIAs, Washington Post, and Author’s Calculations.

The number in Figure 4 is estimated from the number of deportations by September 9, 2017, extended for an additional three weeks until the end of the fiscal year. That simple exercise estimates that ICE will deport about 223,953 people by the end of the fiscal year. Interior removal numbers are publicly known for the months of October 2016 through March 2017 thanks to FOIAs, so estimating the average number of monthly deportations after March was simply based on the total number of deportations that year.  

Third, the immigration court backlog is getting worse. The extension of due process rights to immigrants, an insufficient number of judges, and executive actions have all certainly contributed to the slowdown. Regardless, the number of new arrests and Trump administration efforts to clear the backlog will lead to a substantial rise in deportations from the interior of the United States. Indeed, it may already be occurring but the numbers are hidden by the monthly averages from April through October 2017.    

The Cato 2017 Free Speech and Tolerance Survey finds only 20% of current college and graduate students believe their college or university faculty has a balanced mix of political views. A plurality (39%) say most college and university professors are liberal, 27% believe most are politically moderate, and 12% believe most are conservative.

College Democrats Less Likely Than Republicans to Think Faculty Is Liberal

Democratic and Republican students see their college campuses very differently. A majority (59%) of Republican college students believe that most faculty members are liberal. In contrast, only 35% of Democratic college students agree most professors are liberal. Democratic students are also about twice as likely as Republican students to think their professors are moderate (32% vs. 16%) or conservative (14% vs. 9%).

Full survey results and report found here.

College Students Agree Student Body is Liberal

Current students believe that most of their campus’ student body is liberal. Fifty-percent (50%) believe that most students at their college or university are liberal, 21% believe most are moderate, 8% believe most are conservative, and 19% believe there is a balanced mix of political views.

Democratic and Republican students largely agree on the ideological composition of their campus student body.

Consequences of Campus Political Climate

These perceptions of ideological homogeneity on college campuses may explain why 72% of Republican college students say the political climate prevents them from saying things they believe because others might find them offensive. About a quarter (26%) of Republican college students feel they can share their political views.

Far fewer Democratic college students feel they can’t share their beliefs. College Democrats (51%) are 21 points less likely than Republican students (72%) to feel they can’t share their views. Nevertheless over half of Democratic students also feel the need to self-censor. Interestingly, college students who identify as independents (70%) are about as likely as Republicans (72%) to say they feel uncomfortable sharing their opinions as well.

Only current students who identify as “very liberal” do not feel the need to self-censor (74%). However, liberal college students (55%) instead feel that they have to hide some of their political views, as well as 69% of moderates, 71% of conservatives, and 83% of strongly conservative students.

In sum, there is a widespread perception that most faculty and students at American colleges and universities are liberal. But it’s not just perception. Empirical studies of faculty ideology confirm what most students observe. Surveys conducted by sociologist Neil Gross and others find that, indeed, most of the professoriate is liberal. Sam Abrams at summarizes some of this data to show that as of 2014 about 12% of professors are conservative while about 60% are liberal. Further, the Cato Free Speech and Tolerance Survey finds that nearly half (46%) of current college and graduate students identify as Democrats or independents who lean Democratic. About a quarter identify as Republican (27%).

These results matter because if universities become political echo chambers, it could lead to the exclusion of non-conforming political views, self-censorship, and less rigorous academic inquiry. Without a free exchange of ideas, there may be less thorough checking of academic work and the quality of research may decline. By extension, the public may lose confidence in the process of academic inquiry and become skeptical of its results.

Full survey results and report found here.

Sign up here to receive forthcoming Cato Institute survey reports.

The Cato Institute 2017 Free Speech and Tolerance Survey was designed and conducted by the Cato Institute in collaboration with YouGov. YouGov collected responses online August 15-23, 2017 from a national sample of 2,300 Americans 18 years of age and older. The margin of error for the survey is +/- 3.00 percentage points at the 95% level of confidence.

A main goal of the House Republican tax plan is to reduce marginal tax rates, which are the rates that affect economic decisionmaking. Cutting marginal rates reduces the harm caused by many of the special-interest breaks in the tax code, while reducing the value of those breaks to recipients. That is why even some supporters of breaks that the House plan would not repeal are not pleased with the rate-cutting reforms.

As an example, the House tax plan does not repeal the Low-Income Housing Tax Credit (LIHTC), but supporters of the subsidy program are not happy. The credit is taken by big banks on their corporate tax returns, and it has spawned an industry of real estate, financial, and services firms, as well as state and local housing bureaucracies.

In an upcoming Cato study, Vanessa Brown Calder and I describe the flaws in the LIHTC and recommend its repeal. We discuss some of the flaws in this op-ed.

The House tax plan would affect the LIHTC in three ways:

  • Cutting the corporate tax rate from 35 to 20 percent would reduce the value to corporations of investing in LIHTC properties because the tax-loss benefits would be smaller.
  • Eliminating tax-free private activity bonds would end subsidies that are often paired with LIHTC subsidies on housing projects.
  • Changing the LIHTC inflation factor would slow the growth of the $9 billion credit over time.

Those would be good changes, but as the tax debate moves to the Senate, it would be nice to see some members push for LIHTC elimination. That would free up $9 billion a year to offset a further reduction in the corporate tax rate.

President Trump said, “tax reform must dramatically simplify the tax code [and] eliminate special-interest loopholes.” And yesterday, House Speaker Paul Ryan said he wanted to “clean out the special interest loopholes in the tax code.”

Great! You folks should add the LIHTC to the clean-out list. It is a complex corporate giveaway that provides limited benefits and suffers critical flaws.

Look for our forthcoming study, “Low-Income Housing Tax Credit: Costly, Complex, and Corruption-Prone.”

A year ago, voters made Massachusetts one of eight states (plus D.C.) to legalize recreational marijuana. For all its predecessors, the prospect of legal marijuana for both medical and recreational use has meant increased tax revenue, new jobs, and a reduction in black market sales. Technical issues inevitably arise—how to tax different products and concentrations? what constitutes impaired driving?—but with another nine months to go before the initiative takes full effect, there should be plenty of time to implement this reform.

The Bay State does face one unique issue: how to transport marijuana from the mainland to surrounding islands like Nantucket and Martha’s Vineyard. The problem is that transporting marijuana by sea or air—across waters under federal jurisdiction—is illegal.

If distributors were to ferry marijuana across Nantucket Sound, the Coast Guard would have full authority to seize and enforce the federal prohibition. Flying cannabis to the island is just as risky a proposition because federal law prohibits pilots from knowingly flying planes that contain illegal substances on board. Curiously, however, the applicable regulation says that it “does not apply to any carriage of narcotic drugs, marihuana, and depressant or stimulant drugs or substances authorized by or under any Federal or State statute or by any Federal or State agency.” This raises the obvious question of whether state-legalized marijuana is actually allowed to be transported by air.

The language of this regulation is ambiguous regarding what I’ve taken to calling “Schroedinger’s Weed”—both legal and illegal at the same time —and the FAA hasn’t provided any guidance on the matter. Even without such guidance, marijuana businesses in Alaska have been taking advantage of the regulatory language in a different way, specifically the part about “knowing” transport. Alaskan businesses simply inform airport police that they are transporting marijuana, while also providing their proper (state) Marijuana Control Board documentation. None of the parties inform the airplane owner or pilots, keeping them unaware of the marijuana.

This type of plausible deniability is likely not viable long-term or on a large scale because the room for error is high and, after all, it remains a legal grey area. A possible alternative would be to narrowly tailor state laws that expressly makes legal the transportation of marijuana within and between any state territory—whether in Alaska, Massachusetts, or anywhere else. That would create a bona fide “authorized” use by state statute and explicitly invoke the regulatory safe harbor without ambiguity—and it’s hard to imagine congressional will to close the alleged loophole. 

If the Massachusetts legislature declines to enact such a law, then the only completely secure legal option would be to grow and sell all marijuana on the islands themselves rather than transporting it in. Even there, however, there would have to be an initial import of marijuana seeds, which may trigger the same federal air and maritime laws (but may be easier to conceal).

Moreover, under regulations being developed for the Massachusetts market, all marijuana must first be tested in a certified lab before being introduced for consumer consumption. The establishment of such testing facilities on the individual islands would make Nantucket or Martha’s Vineyard weed considerably more expensive. These islands are known as being playgrounds for the rich, to be sure, but such impediments may still make marijuana commercially unfeasible, particularly in competition with the black market.

Still, the Cannabis Control Commission, a Massachusetts state agency, is determining ways to ensure that Nantucket and other state islands have access to marijuana—and there have been moves for statutory reform in the state legislature. Meanwhile, Mass Medi-Spa Inc., a marijuana dispensary, is planning on opening in Nantucket within the next year regardless of the state of the laws.

Let’s hope that their vision for pro-federalism, pro-liberty policy reform doesn’t go up in smoke.

Thanks to Cato legal intern Addison Hosner for his research assistance on this post.