Border Patrol made 405,036 apprehensions of approximately 324,029 unique individuals in the 2020 fiscal year, down from 859,501 apprehensions of approximately 799,336 unique individuals in 2019. In 2020, Border Patrol arrested 2,438 criminal aliens convicted of 3,150 crimes. Of those 3,150 convictions, 40 percent were for immigration offenses while the remainder were for more serious crimes. Thus, approximately 1,463 of the 2,438 criminal aliens arrested for Border Patrol had committed non‐immigration crimes. Of the 324,029 unique individuals apprehended by Border Patrol, the 1,463 criminal aliens with non‐immigration crimes accounted for about 0.45 percent of all aliens apprehended by Border Patrol in 2020. In other words, less than one‐half of one percent of the illegal immigrants apprehended by Border Patrol had committed non‐immigration crimes and about 0.75 percent had committed a crime including immigration crimes (Figure 1).
Criminal aliens apprehended as a percent of all apprehended illegal immigrants are up slightly in 2020 because the number of all people apprehended has roughly halved in the last year, but the total number of criminals apprehended is the lowest recorded. The number of criminal aliens apprehended is 43 percent lower than in 2019 and 87 percent lower than in 2015. In 2015, each Border Patrol agent apprehended one criminal alien on average. In 2020, only 1 in 8 Border Patrol agents apprehended a criminal alien on average. Either criminal aliens found a way to cross the border undetected, which is unlikely, or many fewer are coming.
By comparison, about 8 percent of the U.S. adult population had been convicted of a felony. Although it’s not an apples‐to‐apples comparison as the U.S. adult felony conviction rate includes immigrants who have a lower criminal incarceration and conviction rate, we can confidently estimate that native‐born Americans have a rate of felony conviction about 10 times higher than that of illegal immigrants apprehended by Border Patrol in 2020.
In 2020, convictions for driving under the influence accounted for about 12 percent of the convictions , drug crimes accounted for 12 percent, 7 percent for assault, battery, or domestic violence, 5 percent for property crimes, 5 percent for sexual offenses, 2 percent for weapons charges, 18 percent for other crimes, and 0.1 percent for homicide and manslaughter (Table 1).
The difference above between the number of apprehensions and individuals apprehended is due to the recidivism rate. Many illegal immigrants get caught, are returned or removed from the United States, and try again. That recidivism rate has trended downward over time because the government imposed harsher penalties on illegal border crossers. However, the government has been quickly returning illegal immigrants apprehended along the border without consequences since March 2020 in response to the COVID-19 pandemic. As a result, the recidivism rate has shot up in 2020 for the reasons I stated here.
How much has the recidivism rate shot up? It was 7 percent in 2019 and 20 percent for the entire 2020 fiscal year. In September 2020, the recidivism rate was a shocking 37 percent.
Nick Miroff of the Washington Post wrote an excellent piece about the skyrocketing recidivism rate wherein he quotes Border Patrol chief Rodney Scott as saying: “We’re returning people very, very quickly, but our ability and willingness, if you will, to prosecute people, to have a consequence to the illegal activity of crossing the border, has been reduced.” Most of those being apprehended are single Mexican adults, a dramatic turnaround from 2019 when only 20 percent were Mexicans.
Returning illegal border crossers immediately combined with other restrictions on asylum and Mexican enforcement policy may have dissuaded many Central Americans from trying to enter unlawfully or to seek asylum, but it has helped shift it back toward Mexicans. The Trump administration’s cancellation of H-2B visas for seasonal non‐agricultural work, primarily used by Mexicans, has likely also contributed to the surge of Mexicans.
The Department of Labor (DOL) released a rule last week that raised the “prevailing wage”—the minimum wage that employers must pay to H-1B and other foreign workers. In justifying the rule, DOL claimed that most employers were paying more than the current prevailing wage, so raising it shouldn’t affect them. Indeed, DOL said that the prevailing wage should approximate the wages that many H-1B employers were already paying to their workers. But it then went ahead and imposed prevailing wage rates that are far higher than the wages that H-1B workers are now receiving.
DOL summarizes its logic for raising the prevailing wage as follows:
the Department’s data show that many of the largest users of the H–1B program pay in many cases wages well over 20 percent in excess of the prevailing wage rate set by the Department for the workers in question.… Employers must pay the higher of the actual wage they pay to similarly employed workers or the prevailing wage rate set by the Department. Both possible wage rates generally should approximate the going wage for workers with similar qualifications and performing the same types of job duties in a given labor market as H–1B workers. It is therefore a reasonable assumption that … the wage rates they produce would, at least in many cases, be similar.
Where the Department’s otherwise applicable wage rate is significantly below the rates actually being paid by employers in a given labor market, it gives rise to an inference that the Department’s current wage rates … are not reflective of the types of wages that workers similarly employed to H–1B workers can and likely do command in a given labor market.… Put another way, when many of the heaviest users of the H–1B program pay wages well above the prevailing wage, it suggests that the prevailing wages are too low, and thus can be abused by other firms. (85 FR 63872, 63886).
To hear DOL tell it, then, most H-1B companies are already complying with both the letter and intent of the law and should have no problem with this new rule because the higher wages will only affect a few low‐paying employers. It concludes that the prevailing wage rate “should approximate” the actual wage being paid for the “largest users of the H-1B program” and that the actual wage and prevailing wage should “at least in many cases, be similar.” But the prevailing wage rates that the rule actually adopted completely contradict these agency findings.
DOL produced wage rates that are almost entirely dissimilar from the actual wages offered to H-1B workers overall as well as among the top users in 2020. Table 1 shows the new hourly prevailing wage rates compared to the actual hourly wage offers in 2020. Overall, 94 percent of H-1B job offers were below the prevailing wage rates under the IFR. The new IFR prevailing wage rate is 20 percent more or higher than the actual wage offers for 88 percent of H-1B jobs in 2020. Overall, the average H-1B employer will have to increase actual wage offers by more than 30 percent. Among the top H-1B employers that DOL specifically indicates its findings should apply to, the new prevailing wages average 31 percent above, and again, the new prevailing wages are 20 percent above for 89 percent of H-1B jobs among the top users.
Notice that DOL claimed that a 20 percent difference was “significant” and proved the the prevailing wage rates were wrong, but now it has moved the wages more than 20 percent higher than actual wages, despite its clear statements. Far from costing employers nothing, the 30 percent wage hike will cost employers tens of billions of dollars in additional wages.
This is the second way in which DOL has fundamentally misrepresented its rule to the public. I previously noted that DOL made a massive error in its assumptions about where wages would fall under its new methodology. It’s possible that the one bad assumption explains why the prevailing wages are so far from the actual wage offers. Either way, DOL’s rule directly contradicts the agency’s own findings. Hopefully, a court will find that any agency action that contradicts its own findings is arbitrary and capricious and stop the rule.
In assessing the New Deal's contribution to economic recovery, I've naturally tended to draw on fairly recent research. That keeps me from being accused of being out of date. But it makes me vulnerable to the charge of overlooking the testimony of experts who studied the New Deal's consequences at first hand.
To that charge, I plead an emphatic Not Guilty! Those who know me will back me up when I say that I'm actually an antiquarian at heart, who'd much rather read a musty old report than any recent journal article. So I've read plenty of contemporary writings on the course of the depression and recovery, and the New Deal's contribution to them, including those of several of FDR's own advisors. These works often support the critical assessment of subsequent economic historians. If anyone is guilty of exaggerating the New Deal's contribution to the recovery, it's those popular historians who gloss over its failures while declaring that anyone who points to them must be a Hoover Republican!*
Of those failures, none was more glaring than that of the National Recovery Administration, the subject of my previous post in this series. And that failure was no less evident to those who witnessed its consequences as it has been to most economic historians since. I might cite numerous contemporary works to make the point—no other product of New Deal legislation met with more caustic or widespread criticism. But none makes it more assiduously than the 1935 Brookings Institution publication, The National Recovery Administration: An Analysis and Appraisal.Read the rest of this post »
Andrew Forrester, Michelangelo Landgrave, and I published a new working paper on illegal immigration and crime in Texas. Our paper is slated to appear as a chapter in a volume published by Oxford University Press in 2021. Like our other research on illegal immigration and crime in Texas, this working paper uses data collected by the Texas Department of Public Safety (DPS) that records and keeps the immigration statuses of those arrested and convicted of crimes in Texas. As far as we’ve been able to tell, and we’ve filed more than 50 state FOIA requests to confirm, Texas is the only state that records and keeps the immigration statuses of those entering the criminal justice system. Texas gathers this information because its runs arrestee biometric information through Department of Homeland Security (DHS) databases that identify illegal immigrants. Unlike other states, Texas DPS keeps the results of these DHS checks that then allows a more direct look at immigrant criminality by immigration status.
The results are similar to our other work on illegal immigration and crime in Texas. In 2018, the illegal immigrant criminal conviction rate was 782 per 100,000 illegal immigrants, 535 per 100,000 legal immigrants, and 1,422 per 100,000 native‐born Americans. The illegal immigrant criminal conviction rate was 45 percent below that of native‐born Americans in Texas. The general pattern of native‐born Americans having the highest criminal conviction rates followed by illegal immigrants and then with legal immigrants having the lowest holds for all of other specific types of crimes such as violent crimes, property crimes, homicide, and sex crimes.
Since Texas is the only state that records and keeps the immigration statuses of those arrested, we can’t make a direct apples‐to‐apples comparison between Texas and other states (every state should record and keep this information so we can answer this important question). It could be that illegal immigrants in Texas are the most law‐abiding illegal immigrant population in the country – or the least law‐abiding. Until other states start recording and keeping the data, we won’t know for sure. But there is much suggestive evidence that the illegal immigrant criminal conviction rate in Texas is comparable to their crime rates across the country.
For instance, the ratio of the nationwide estimated illegal immigrant incarceration rate to the native and legal immigrant incarceration rates is very similar to the same ratios for the criminal conviction rate in Texas. The similarity is evidence that the pattern in Texas holds nationwide, at least to the extent that convictions and incarcerations are correlated. The only way that illegal immigrants could have a higher incarceration rate is if there is something seriously wrong with our method of estimating their total population in the United States and the actual number is much smaller or we are seriously undercounting illegal immigrants who are incarcerated. Neither is very likely, but it’s important to mention the possibility.
We go a bit further in this working paper by looking at how local variation in the illegal immigrant population is correlated with crime rates on the country level in Texas for the years 2012–2018. The relationship between changes in the illegal immigrant population and crime is known as an elasticity. The elasticity between two variables estimates how one variable, the illegal immigrant population here, affects another variable like the number of illegal immigrant convictions or the total crime rate. We control for the number of law enforcement officers per capita. We basically find no relationship. The only statistically significant relationship worth reporting is a negative association between total violent crime convictions and the illegal immigrant share with a point estimate of -0.104 that is significant at the 5 percent level. This exception suggests that a 10 percent increase in the illegal immigrants share of the population is associated with a 1 percent decline in violent crime convictions in our sample of Texas counties.
Our working paper isn’t the only new research on illegal immigration and crime. Christian Gunadi, an economist who recently graduated from the University of California Riverside, examined how the DACA program affected crime rates. Gunadi tested the theory, based on Gary Becker’s crime research, that issuing work permits to young illegal immigrants increases the opportunity cost of committing crime by making it easier for them to be legally employed. Gunadi found, when he analyzed the individual‐level incarceration data, that there was no evidence that DACA statistically significantly affected the incarceration rate of young illegal immigrants. Gunadi also looked at crime on the state level and found that the implementation of DACA is associated with a reduction in property crime rates such that an additional DACA application approved per 1,000 population is associated with a 1.6 percent decline in the overall property crime rate. That second finding is consistent with the Beckerian crime model.
Other recent research into immigration and crime similarly find no relationship between immigration and crime or a slightly negative relationship, but their methods are not as robust so I don’t place as much weight on them. However, a recent working paper written by Conor Norris and published at the Center for Growth and Opportunity used difference‐in‐differences and the synthetic control method to see how the passage of SB-1070 in Arizona in 2010, which was an immigration enforcement law, affected crime there relative to other states. It found that violent crime in Arizona increased by about 20 percent under both methods.
Norris’ paper is interesting and worth developing further. For instance, most of the research on the economics of crime focuses on how higher opportunity costs lowers crime rates. In that way, increasing legal employment opportunities can lower crime while making it more difficult for illegal immigrants to work can push some of them toward committing crimes because they’d have less to lose. In 2007, the Arizona state legislature passed the Legal Arizona Workers Act (LAWA) that mandated E‐Verify on January 1, 2008. E‐Verify is intended to prevent the hiring of illegal immigrants. Forrester and I wrote a short blog post showing that the passage of LAWA may have increased the monthly flow of non‐citizens into Arizona state prisons, but the effect was short‐lived as many illegal immigrants either left the state or figured out how to get around E‐Verify.
The above new research and the vast quantity of papers on how immigration doesn’t increase crime and frequently lowers it leads to an interesting question: Why do so many people think that immigration increases crime? The Christian Science Monitor had an interview segment recently where they asked criminologists why so many Americans think immigrants increase crime even though the weight of evidence says that they are less likely to commit crimes than native‐born Americans. According to a recent Gallup poll, 42 percent of respondents thought that immigrants increase crime, 7 percent thought that immigrants decrease crime, and 50 percent said immigrants didn’t affect crime.
Much of the effect could be that people who don’t like immigration could just ascribe all types of negative behavior to them in order to justify their dislike. This probably explains a lot of it, but it would be a disservice to stop there. We must examine the possible other reasons. Another potential reason is that many people think that immigrant criminals could have been prevented from coming in the first place, so there’s more of a focus on their crimes (availability bias) because many people think that they are more preventable than crimes committed by native‐born Americans. In that way, many people could think that allowing any crime by immigrants is a choice and that crime could go away at the stroke of a pen. That’s not how the world works and that doesn’t explain why so many people think that crime rates go up with immigration, but if that form of control bias is combined with a conflation between the number of crimes and the crime rate then the mistake is understandable if not based on an accurate understanding of the variables.
Another reason could be that native‐born Americans who have the same ethnicity as recent immigrants might have a much higher incarceration rate, so the respondents to these surveys lump them in together and conclude that immigrants boost the crime rate. Among native‐born Americans, Hispanics do have a higher incarceration rate but Asians have a much lower rate. This is further complicated by the fact that Puerto Ricans, who are not immigrants, likely have the highest incarceration rate of any Hispanic sub‐group in the United States (see Table 1) and it would be quite silly for someone to blame immigrants for the higher Puerto Rican incarceration rate.
There is more and more evidence that immigrants, regardless of legal status, are less likely to commit crimes than native‐born Americans. However, a substantial number of Americans still think that immigration increases crime. As more evidence builds over time, we can only hope than Americans respond by updating their opinions so that they fit the facts.
The United States has welcomed more than 85 million legal immigrants to the United States since its founding. But at no time since it has maintained records has the country witnessed as fast a decline in legal immigration as it has seen in the second half of fiscal year 2020 (which finished September 30). Overall, the second half of FY 2020 saw 92 percent fewer immigrants from abroad than the first half, which was larger than any annual decline in the history of the United States.
Figure 1 shows the monthly immigrant visa issuances under the Trump administration since March 2017. As it shows, legal immigration almost wholly stopped in April and May 2020—after the State Department closed its consulates and President Trump issued a proclamation suspending new visa issuances to most immigrant categories. It has recovered slightly since then, but it remains 84 percent below last year (which was also a down year).
Figure 2 shows the number of new arrivals of legal permanent residents or immigrant visas approved by year from 1820 to 2020, with the third and fourth quarter of FY 2020 added. The United States witnessed a more than 90 percent falloff in new immigration from abroad during the second half of FY 2020. This brings the annualized legal immigration rate from abroad to 0.03 percent of the U.S. population. This is the lowest rate of immigration except for three years during World War II and one year during the Great Depression.
The 92 percent drop in the second half of FY 2020 is larger than the drop during any single year in American history—larger than the 73 percent decline in 1915 coinciding with the start of World War I, larger than the 70 percent decline in 1925 coinciding with Congress closing legal immigration from Europe, larger than the 63 percent declines in 1931, 1942, and 1918 following the onset of the Great Depression and U.S. entries into each world war. Table 1 shows the data for all available years and the change for the second half of 2020 from the first half. While it’s only half a year, Figure 1 indicates how slow the immigration recovery has been. It is unlikely that the 2021 will be much different if President Trump is reelected.
Before 1924, immigrants were never required to receive immigrant visas abroad to enter and become legal permanent residents, and from 1924 to 1952, nearly all immigrants had to receive immigrant visas abroad to become legal permanent residents. In recent years, about half of all new legal permanent residents have adjusted their status to permanent residence from temporary statuses, such the H-1B visa, refugee status, or illegal status. Generally, the number of new “immigrants” include both the number of new arrivals from abroad and those adjusting in the United States, but it’s also important to see who is entering from abroad because that reflects real changes in the U.S. population. The number of work visas, of course, have also declined just as dramatically.
This historic slowdown is important for both the short‐term and long‐term economic growth of the United States. Fewer workers mean that jobs will take longer to fill and slow the economic recovery, and in coming years, fewer workers will support more retirees. If the United States remains closed long enough, it could push worldwide patterns of immigration away toward other countries with more welcoming policies.
Last week, the Wall Street Journal examined government efforts to secure early access to doses of the most advanced COVID-19 vaccines, and how this access could prove to be a game‐changer for these economies in 2021. As shown in the following WSJ chart, many governments have contracted with multiple pharmaceutical companies in order to ensure that they have access to at least one vaccine that successfully completes “Phase III” trials, which are now underway for most of the listed drugs. Among these governments is the United States, which has thus far secured vaccine commitments from Oxford/AstraZeneca (0.91 doses per capita); Novavax (0.3); Sanofi/GSK (0.3); BioNTech/Pfizer (1.83); J&J/Janssen (0.3); and Moderna/NIAID (1.52). This puts the United States second, behind only the United Kingdom, in contracting for early vaccine doses and thereby potentially saving thousands of American lives and restarting the struggling domestic economy if one or more of those vaccines pans out.
Beyond the sheer size and scope of the U.S. effort, what’s perhaps most striking here is the extent to which the Trump administration jettisoned its economic nationalism in pursuit of a game‐changing COVID-19 vaccine. Indeed, as shown in the chart below (based our own independent research), each of the vaccines that the United States has secured appears to be heavily reliant on globalization — of investment, manufacturing, testing, and research personnel — to produce the final doses at the absolute maximum speed and scale.
This summary, moreover, is only the tip of the iceberg when it comes to the truly‐global effort to beat COVID-19. Unmentioned in the chart above, for example, are all of the other people from around the world — investors, researchers, production workers, etc. — that make the listed companies, facilities and processes hum, as well as the previous global collaborations that have driven pharmaceutical innovation for decades.
Also unmentioned are the numerous other vaccine candidates that are in earlier stages of development. For example, a recent survey by the Coalition for Epidemic Preparedness Innovations (CEPI), which is working with the World Health Organization to ensure global access to a COVID-19 vaccine, showed over a hundred potential manufacturers of both vaccine‐related drug substances (inputs) and drug products (doses) located in dozens of countries around the world (including, of course, the United States):
Since the unfortunate onset of COVID-19, American politicians of all stripes — but especially in the Trump administration — have blamed “globalization” for the pandemic and promised to re‐shore U.S. manufacturing to bolster American “resiliency” and national security. Yet these very same officials have quietly embraced that very same globalization when it matters the most.
The Department of Labor’s (DOL) new rule changes how it calculates the mandatory minimum wage—called the prevailing wage—for employers of H-1B and permanent foreign workers. DOL adopts a fundamentally flawed methodology as its basis to inflate the prevailing wage. But a bigger issue is that DOL itself failed to understand how much its methodological changes would artificially raise the required wages. DOL estimated the wage effects of its rule using completely erroneous assumptions, and so it understates to the public the wage increases by, in many cases, as much as 26 percent.
The prevailing wage is supposed to approximate the wages of similarly skilled U.S. workers. DOL currently uses the Bureau of Labor Statistics’ (BLS) Occupational Employment Statistics (OES) survey to create a prevailing wage for four skill levels within each occupation in every area of the country. The creation of the skill levels—which is the focus of this rule—is contentious because the OES doesn’t directly record skills. Instead, BLS creates these skill levels mathematically based on the reasonable assumption that higher wages within an occupational category within a specific area generally reflect higher skills.
Table 1 compares the new and old prevailing wage methodologies. Previously, DOL had assumed that the bottom third of the wage distribution represented entry level wages, while the rest (the top two thirds) were not entry level. After averaging the wages in the bottom third and top two thirds to create the Level 1 and Level 4 wages, it placed the two other wage levels equally distant between them. The new rule, however, ignores all wages below the 45th percentile—making that the starting wage—and rather than averaging all other wages to produce Level 4, it only averages the top decile of wages (the 90th to 100th percentile).
The big error that DOL makes is that it assumes that averaging the top decile will equal the 95th percentile. Since the top decile (the top 10 percent of wage earners) includes some extreme outliers and a very small sample size, those outliers skew the level 4 wage far higher than the 95th percentile. Because the Level 2 and Level 3 wages are dependent on the Level 4 wage calculation, it also mistakes where these wage levels will fall on the wage distribution. This becomes incredibly important when DOL then tries to estimate the wage impact of the new rules:
To estimate the wage impacts of new percentiles contained in this [rule], the Department used publicly available BLS OES data that reports the 10th, 25th, 50th, 75th, and 90th percentile wages by SOC code and metropolitan or non‐metropolitan area. In order to estimate wages for the new IFR levels of 45th, 62nd, 78th, and 95th percentiles, the Department linearly interpolated between relevant percentiles for reported wages at each SOC [job] code and geographic area combination. [Endnote: For example, if OES reports a wage of $30 per hour at the 25th percentile and $40 per hour at the 50th percentile then the 45th percentile is interpolated as $30+($40-$30)*((45–25)/(50–25)) = $38 per hour.] For the 95th percentile, the Department used OES wages reported for the 90th percentile at each SOC code and geographic area combination. (p. 126)
In other words, DOL did not even try to estimate what the actual 95th percentile would be. For reasons that are only explicable as laziness or hastiness, it did not ask BLS to calculate the new prevailing wages until after it issued the rule. Instead, it just settled on using the 90th percentile as a stand‐in for the level 4 wage to estimate the rule’s effects because it thought that it would be similar. It then tried to calculate the level 3 and level 2 wages assuming that they would be around the 78th and 62nd percentiles, respectively. But this was wrong.
Table 2 is an example of DOL’s assumed wage levels for San Francisco architects and New York City management analysts under the new rule and what the actual wage levels now are. For wage levels 2–4, DOL’s estimate is wildly inaccurate—underestimating the actual amount by 15–18 percent for level 2, 23–25 percent for Level 3, or 24–26 percent for level 4. These huge differences appear in every single occupational category. Notice that in the examples below, the actual level 2 wage is higher than what DOL assumed would approximate the level 3 wage, and the actual level 3 wage is higher than what DOL assumed would approximate the level 4 wage.
DOL repeatedly refers to the top three wage levels as “percentiles,” misstating its own methodology which is based on averaging the top decile and then imputing the middle two wage levels equally distant from the other levels. Because the level 4 wage is so inflated, the other two middle wage levels are also widely off to the point that what DOL refers to as the “78th percentile” (level 3) is actually above the 90th percentile in the two examples above.
Because DOL decided not to ask BLS for the actual OES data that it would use to create the new wage levels, it missed how egregious this mistake was. Moreover, because DOL refused to let the Office of Management and Budget review the rule, this massive error slipped through unnoticed. Finally, because DOL made the rule take effect almost immediately and refused to accept public comments on the rule, it has no time to correct it before the rule begins to effect tens of thousands of American businesses. DOL should immediately rescind the rule and start the process over.