Category: Economics

Another try to pass a farm worker bill

The farming industry regularly comes up with a bill to nomalize the status of unauthorized farm workers.  Close to 600 farm organizations are backing a House bill aimed at ensuring workforce supply. It includes a provision to normalize legal status – i.e. provide for amnesty. HR 9535, Securing Agriculture’s Workforce Act of 2026 was introduced by Glenn Thompson, a Republican congressman from Pennsylvania. It overhauls the H-2A program for temporary (i.e. seasonal) workers, focused on farm work.

The farm industry says, “While exact numbers are difficult to determine, it is estimated that roughly 70 percent of hired crop workers are foreign-born, with about 40 percent lacking legal work authorization. Unauthorized labor peaked at 55 percent around 2000 but remains a key part of the nation’s farm labor pool.”

I have written about how the H-2A program use has surged in the past ten or so years. (Also here.) It went from 150,000 workers in 2015 to 400,000 workers in 2025. Enrollments now appear to be at the 500,000 level.

What is happening: unauthorized workers in farming have either flattened out or declined, and very few non-immigrant American citizens are going in.  Estimates of the size of the paid (i.e. non-family) farm workforce are all over the map – perhaps 1.5 million, 2.2 million on the very high side. This is a heavily labor contractor-run workforce, concentrated in five states. The H-2A program is dominated by labor contractors. (Go here for 2026 California analysis.)

Among the provisions of the bill are 350-day contracts (vs seasonal) and arrangements that could run to three years.

As for normalization of legal status, Section 2 of the bill provides for not a path to citizenship but only legal status in the H-2A program, with no access to public benefits. “Applicants generally would have to demonstrate either that they voluntarily departed the United States or that they were unlawfully present on May 31, 2026, and performed at least 180 days of agricultural work during the previous two years. The waiver would apply only to specified immigration violations that occurred before the worker obtains H-2A status and would not create a new immigration status or a pathway to permanent residence.” (Go here.)

 

Tourism to U.S. down 20% from 2019

The decline in overseas tourism to the U.S. cannot be explained away as a Canadian boycott. While global tourism is 4% up this year (1.5 billion visits), the most recent figures (August, 2026) show a large decline in U.S. visitors.

The U.S. has been experiencing a decline in overseas arrivals (thus excluding the Mexican and Canadian borders).  Overseas arrivals, January–August 2026, were 21.41 million, down 5.8%. August alone was down 11.8%.

But worldwide tourism is up. In the first quarter of 2026, arrivals increased 2%, with Europe up 4% and Asia-Pacific up 3%.

According to the U.S. Department of Commerce’s International Trade Administration (ITA), international arrivals to the United States in 2025 decreased 5.5% from 2024. With the exception of a blip due to the World Cup, the downward trend has worsened in 2026.

Per this report, U.S. Travel Association president Geoff Freeman said this month that the U.S. was on pace to receive approximately two million fewer overseas visitors this year than last, leaving visitation roughly 20% below its 2019 peak,

Variouse sources: here, here, here and here.

Why anti immigration has stopped metastasizing in the U.S. 

Anti-immigration political movements arise out of perceptions of lost control, cultural change and unequal treatment.  Anti-immigration politics is surging in Germany, is strong in the UK – and is peaking in the U.S. despite Trump’s efforts.

A snapshot of Germany, the United Kingdom and the U.S.:

Germany: Alternative für Deutschland (AfD): 201% of seats in the Bundestag (Feb 2025 elections),  success in Saxony-Anhalt 9/6/26 – about 44%, in Mecklenburg–Western Pomerania 9/20/26 – 38% of vote vs. 17% in 2021. In Berlin, 16%.

United Kingdom: the political party situation is less clear because of the controversial dynamics of the Reform Party’s leadership. In the past two weeks landings of boats conveying potential asylum applicants have been interrupted in highly visible forms of protest.  Per YouGov in August, 69% think immigration over the last decade has been too high, and the share saying migration has been “mostly bad for the UK” went from 10 points since 2016 to 43%.

United States: Over the past 25 or so years, the Dem and Rep parties gradually and then quickly homogenized their leadership into inclusive (Dem) and restrictionist (Rep) positions.  However, at the electoral / voting level, American politics has remained internally confused, captured in a mindset which I have described as “We love immigrants, just not as many.”

And, as in the case in Germany and the UK, the nursing beds of anti-immigration politics are in regions with little past history of immigration, with a surge of immigration occurring recently, bringing in persons who are overly of difference cultures i.s. emerging world.

People not like us

The emerging country imprint of immigration is more visibly pronounced in Europe, which has absorbed migration not just within the EU but from Eastern European countries, and also migration from the Middle East and Africa.

On contrast, the major sources of migration into the U.S. since 1990 has been, first, from Latin America, which for generations shared a transnational workforce mainly in farming, and in the past 15 years from Asia, which exhibits an approved profile of hard work and commitment to education. The U.S. has been absorbing immigrants for centuries.  This Asian wave, as well, has a visible layer of elites that overtly contribute to society – Asian doctors and IT professionals. This is unlike migrating elites in the U.K., especially Russians, who remain removed from overt contribution to society (the Mayfair set).

Proxies for citizen power

Trump’s efforts to create an illusion of non-citizen voting is based on a clever observation.Immigrant formal voting power is very limited by law. This does not take into account that government policies which support migrant populations may appear to, in effect, grant immigrants rights associated with being voting citizens – such as housing.  A core issue for anti-immigrants, not often well articulated in an elegant way a la The Atlantic or NY Times, is that citizen sovereignty is eroding – appears to be in concrete ways. (Go here for a summary of a book on this.)

Why anti-immigration politics in the U.S. has peaked

In contrast to Germany and the UK, we are seeing the end of the rise of anti-immigration politics in the U.S. This is due primarily to the much greater experience in successful integration of immigrants into an already multi-racial, multi-cultural society. There is poorly known success of greater economic mobility among immigrants. Also, Americans have far less emotional attachments to historical icons from the past than do many Europeans.

The November elections may show the political failure of extreme politicization of anti-immigration. Congresswoman Salazar’s problem is a canary in the mine.  Proposition 187, curtailing access to public services by unauthorized persons, passed overwhelming in 1994 in California  — part of Republican Pete Wilson’s gubernatorial campaign.  The state supreme court declared it unconstitutional. California had supported Republican presidential candidates in most elections 1952-1988. Republicans remained competitive in the early 1990s. After Proposition 187, however, Democrats dominated presidential elections, the legislature and statewide offices. Arnold Schwarzenegger’s victories in the unusual 2003 recall and in 2006 were the principal exceptions.

 

 

 

 

 

More on Hispanic home buying

For some years Hispanic households have led the first time purchase of homes. I posted on this in 2021. Here are some current facts, quoting housing industry analysis:

The broader national context highlights this point clearly. While total U.S. owner-households increased by 316,000, this figure masks contrasting trends among different groups. Non-Hispanic white households experienced a decline of 441,000 owner-households, while non-Hispanic Black households saw a reduction of 103,000. These declines were fully offset by the gains among Hispanic households.

In effect, Hispanic buyers were not just contributors—they were the sole drivers of growth. Without their participation, the U.S. would have recorded a net loss of approximately 125,000 homeowners.

The Hispanic population is the youngest major ethnic and racial group in the United States, with a median age of 31.

By adding 441,000 owner-households and driving 92.6% of new household formation, Hispanic buyers have become the backbone of the housing market. Their influence extends beyond numbers, shaping trends, demand patterns, and future expectations.

Source here.

The AI workforce: still very U.S. dominant

MacroPolo’s Global AI Talent Tracker collects information on the global AI workforce, in particular what it defines as AI researchers.   The trend has been and remains towards migration to the United States, both for education and work. It’s headline: “47%of the world’s top AI researchers were trained in China. 72% of them now work in the United States.”

The composition of US-based AI researchers shifted since 2010. In 2010, 40% of AI researchers at top US institutions were foreign-born. In 2024, it exceeded 65%. The shift is concentrated in machine learning and deep learning.

China is the largest source of AI professionals. Among China-educated AI researchers, 72% now work in the United States. The next largest sources are India (80% US-bound), South Korea (45% US-bound), and Europe (40% US-bound).

Among Chinese researchers,  the share of them who received their undergraduate education in China has risen to 38%.  But only 11% of China-educated researchers now work in China—72% work at institutions in the United States.

American institutions employ 59% of the researchers studied, and much of this workforce was educated abroad. About 80% of US-educated researchers remain in America, compared with 65% in Britain and 20% in India.

Go here for a 2021 posting on the AI workforce.

Adverse economic impact of ICE enforcement

A Brookings study of May 2026 finds that the surge in ICE enforcement reduced the number of expected filled jobs by 688,000.  Its is not a absolute reduction in employment, rather a decline in normally expected growth. Bottom line (see below): the U.S. lost the equivalent of one year in expected growth.

The period studied was the first half of 2025.  This a shortage from expected for all would-be job occupants – not distinguishing between citizens, authorized and unauthorized foreign born.  The analysis indicates the relatively lower employment was due in part to arrests and in part due to fear of arrest, which the “shock and awe” approach was designed to do.

(The total is from employment data, not surveys.)

Since the number applies only to the first half of 2025, it is certain that the total losses have been much higher. Because of the pandemic and the Biden-caused surge in employment during his administration, I think it is safer to use the increase of total employment in 2019 as a reference. In that year it was about 1.8 million. Assuming that all-2025 effect was a relative loss was a high as 1.5 million-plus, that means that the loss approximated one year’s normal gain in employment. Using another assumption that 75% of the loss is for unauthorized workers, which stood at about 8 million in 2019, that implies that, in order of magnitude, one tenth of unauthorized workers left employment.  This is a conjecture, and only for 2025.

“We find that employment trajectories in surge and non-surge cities were closely aligned before enforcement began. They diverged precisely when ICE arrests intensified. The gap widened over time, and the knock-on effects extended far beyond those directly targeted. In surge cities, employment fell most in immigrant-intensive sectors, but job losses spread further.”

“We isolate the surge from other forces: tariffs, inflation, and AI, operating simultaneously on the broader economy. The findings are stark: Job losses in surge cities far exceeded the number of people arrested, and the damage extended well beyond the workers who were directly targeted. The pattern of excess job loss is consistent with fear-driven labor withdrawal, business disruption, and weaker local demand.”

How does education affect the proclivity to emigrate to the U.S? The case of Mexico

The desire of Mexicans wanting to emigrate stands at about 20% of the population. That share is not greatly different from many other countries. But the share of Mexican workforce age working residing outside Mexica is high – 10%, of which almost all in the United States. Mexico is part of the American basin of workers, mostly with little formal education.

It’s unclear whether improvement in education status translates into lower propensity to emigrate.  Costa Rica’s high school graduation rate (60%) is higher than Mexico’s (40%) but Costa Ricans are more inclined to emigrate.  That said, a look at progress in education in Mexico is worth a look.

In 2000, only 20% of Mexican adults had completed what is called high school in the United States. Now, 40%. A lower education threshold also saw substantial gains, from under 70% to around 90%.

Higher education is almost entirely at state colleges. Attendence grew about 50% between 2011 and 2022.   About a half million graduate a year. (Putting that in perspective, that 500K is equivalent to somewhat less than 1% of the total workforce of Mexicans, in the U.S the annual number of graduates is about 1.5% of the work force. (But the HS graduation rate in the U.S. is above 90%.) This flows a lot more advanced educated persons into the Mexican workforce than in the past, but the absorption rate is low – to be sure, the wages are higher than for high school graduates by the quality of work is the lowest among all OECD countries. Half of college graduates are overqualified for their jobs.

If access to migration to the U.S. were easier, would college educated Mexicans wish to migrate? Would they have a better chance of getting work for which they are qualified, considering both wage and desired social status? Compared to less educated, who take many jobs Americans don’t want? The median wage of Mexican in the U.S. is $19 vs the national median wage of $27.

Where Haitian TPS workers work

Note that on June 25 the Supreme Court allowed the administration to proceed with terminating Haiti’s TPS designation.

Haitian TPS workers are heavily concentrated in a small number of occupations. Of the estimated 200,000 Haitian TPS beneficiaries who are employed in the United States, about 11% work as cooks or restaurant servers, compared with 2.4% of the total U.S. workforce. Another 11% work as stockers, warehouse workers or packers, compared with 2.7% of all U.S. workers.

Agricultural workers account for 7.5% of employed Haitian TPS beneficiaries, while about 0.6% percent of total U.S. employment is in agriculture.  Retail employees represent 7% of Haitian TPS workers, compared with 3.2% of the overall workforce. Nursing assistants account for 6.5% of employed Haitian TPS beneficiaries, compared with 0.9% of all U.S. workers.

Together, these five occupational groups employ roughly 43% of working Haitian TPS beneficiaries, compared to 10% of the total workforce.

The median hourly wages of these jobs is slightly under $20 an hour. The median wage of all high school graduate worker is about $24.50.

Much of this data from the American Immigration Council.

What unauthorized farm workers said in a survey

Farm labor particularly in the California produce industry is heavily foreign born and unauthorized. The USDA’s national crop workforce estimated as of January, 2025: “In 2020–22, 32% of crop farmworkers were U.S. born, 7% were immigrants who had obtained U.S. citizenship, 19% were other authorized immigrants (primarily permanent residents or green-card holders), and the remaining 42% held no work authorization. (Go here.)

The United Farmworker Foundation issued a report in July based on a survey of 2,250 farm workers, most of them in California. The survey was conducted in mid 2025. The surveyors did not ask the legal status of respondents, but it appears that virtually all were unauthorized to be in the country

Some highlights:

Tenure in the U.S.: 80% had been in the U.S. for at least ten years;40% over 20 years. 40% had been in farm work in the U.S. for over 20 years. 65% work at least ten months a year in agriculture.

Their undocumented status’ most painful aspect is inability to visit family members in their country of origin. At the same time, 69% said that if they were deported, they would lose family attachments in the U.S. If their status were made legal, 87% would visit family members in their country of origin.

92% said that ICE was affecting their employment. 88% reported fearing detention or deportation. 61% had reduced their shopping, 29% had avoided medical attention.

42% did not know their rights if approached by ICE. Only 20% said they knew their rights and how to get legal representation.

 

 

More insights into the H-1B workforce.

Classify this posting as yet another view of a 30 years plus phenomenon of the transnational Indian IT workforce.

Despite a formal cap of 85,000 on new H-1B workers, the entirely actively working H-1B workers today appears to be about 700,000. A new paper provides some insights into this workforce.

The size of the active workforce compared to annual entrants is due to these workers not only being able to extent their three year stay to six years, but also to extend beyond that

This is an overwhelmingly Indian workforce. India provides about 70% of H-1B workers; China, about 10%.

From a quarter to a half of these workers in a single sector, Computer Systems design. The size of this workforce engaged directly in computer and math (the core technical section) is about 1.3 million. Of these, about 1/10th are H-1B workers.  But they are concentrated among the more highly educated ranks of the total 1.3 million workers, according to this study.

However, note this: Congressional Research Service data show that in about the past 10 years, 69% to 92% became green card holders. As most H-1B workers thus gain green card status, the presence of foreign born computer systems high end workers is very likely to be much more than 1/10th of this workforce.

The employment impact on U.S. born workers is selective. It appears that  U.S. born workers in this core technical segment has not grown, which to me is a sign that H-1B workers are partially blocking the entry of Americans into this workforce.  However, when H-1B workers are present, the non-STEM trained workers in the computer sector increase in size.

the article, focused on the United States, does no address how former H-1B workers populate American and India IT firms.