A new research model from the Penn Wharton Budget Model (PWBM) has brought clarity to the immigration debate in the U.S. by analyzing the macroeconomic implications of different policy scenarios. The model is at the core of a paper titled “Immigration and the Macroeconomy” authored by PWBM experts – Efraim Berkovich, director of computational dynamics; Daniela Costa, economist; and Austin Herrick, senior analyst.
“We find that, after an initial period, increasing legal immigration improves both the government’s fiscal balance and the economy on a per-capita basis,” the paper stated. “Legalization policies [or regularizing undocumented immigrants], on the other hand, worsen the government’s fiscal balance due to increased spending, while having modest effects on the economy broadly.” Lawful immigrants receive government transfers over their lifetime such as Social Security benefits, Medicaid, and the Supplemental Nutrition Assistance Program (SNAP); if they are not sufficiently productive, they create a “retirement benefits imbalance,” the paper pointed out.
The legalization plan the paper modeled is similar to the legalization provisions in the Biden immigration plan. That plan was akin to “a one-shot legalization for people who are already in the U.S.,” said Herrick.
The main finding of the study is that temporary increases in legal immigration rates lead to long-term fiscal benefits, in aggregate and on a per-capita basis. Those benefits would be available partly because legal immigration reduces the burden on younger workers to provide for older retirees — as measured by the old-age dependency ratio — the study added. Policies that make way for legal immigration produce “long lasting, multi-generational effects” as the children of new immigrants enter the workforce, the authors noted.
The model studied the likely effects of a six-year, 25% increase in legal immigration between 2022 and 2027. Such a policy could lead to a 0.08% increase in per capita GDP by 2032 and 0.30% by 2052, it predicted. Under the same policy, government debt would be 0.41% lower by 2032 and 1.34% lower by 2052.
Policies that legalize unauthorized immigrants have dramatically different outcomes for per capita GDP and government debt. “[They] imply a trade-off between higher wages for newly legalized workers and increased government debt through additional spending on social programs for those same immigrants,” the paper stated.
The model predicted that a policy of “full legalization” — where undocumented immigrants are given legal status — leads to a 0.02% increase in government debt by 2032 and a 1.26% increase by 2052. Per capita GDP, meanwhile, is 0.01% lower than baseline in 2032 and 0.37% lower in 2052.
The paper used Congressional Budget Office estimates that about 70% of the undocumented immigrants offered legalization would opt for it. Others may be reluctant because of the costs involved and penalties such as taxes or fines they may have to pay for having stayed illegally in the country, Herrick said.
Cyclical Old-age Dependency Ratios
The overarching question in the immigration debate is on the old-age dependency ratio, or the burden retirees bring on younger workers. The paper estimated that the U.S. population will increase to 356.5 million by 2040, with a foreign born share of 14.5%; the foreign born share in 2020 is estimated at 13.99%. Those estimates projected that by 2060, the population increases to 375.8 million, with the foreign born share rising slightly to 15.6%.
Over that same period of time, the average age of adults (anyone aged 18+) in the U.S. population increases from 48.6 in 2020 to 51.7 in 2040 and 53.9 in 2060. While among adults, immigrants are not younger than natives (the average immigrant adult in 2020 was 49 years old, versus 48.6 years old among natives), the average adult immigrant arriving into the country is 36.1 years old, the paper stated.
“This is the direct mechanism through which increased immigration reduces the old-age dependency ratio,” the paper pointed out. Increased immigration also increases the number of newborns because these immigrants are younger, and therefore closer to prime childbearing age upon arrival, the authors noted. “After two to three decades, these immigration waves contribute to reducing old-age dependency a second time, as the children of arriving immigrants become old enough to contribute to the labor market themselves.”
“You see a cyclical generational pattern in the old-age dependency ratio,” said Herrick of the principal finding of their study. “Immigrants tend to arrive within a fairly consistent age band so you can see their lifecycle throughout the projection as the dependency increases and decreases.”
“Immigrants tend to arrive within a fairly consistent age band so you can see their lifecycle throughout the projection as the dependency increases and decreases.” –Austin Herrick
“If you allow more immigrants to come into the country, your [dependency rate] will go down because immigrants are of working age and you just have literally more people to offset those retirees,” Herrick continued. “But after some time, as they begin to retire, that pushes your dependency ratio back up, because now all those retirees need support from the country (through programs such as Social Security and Medicare). But then later on, as their children begin to enter the workforce in greater numbers, that drives that ratio back down.”
Significantly, the paper quantified the expected productivity boost that legalizing undocumented immigrants provides. The paper explains that legalizing undocumented immigrants does not drive down the old-age dependency ratio. “Policies which increase immigration flow will relieve some of the burden on current younger workers in having to provide for older retirees only if immigrants provide a net growth to the economy,” the paper noted.
“There are a number of forces going in opposite directions, and the magnitudes of these things matter very much for the final effect,” said Berkovich. “I expected the legalization policies to be more growth-inducing in the long run than they turned out to be [in the paper’s model], which was surprising to me. That’s the contribution of this paper — to try to get a good estimate of these effects and seeing which way to go [with immigration policy].”
“In the case of legalization [of undocumented immigrants], you can see the opposite effect,” said Herrick. “When unauthorized immigrants are offered legal status, they are more likely to remain in the country, either because they don’t have to worry about deportation or because they are more able to become rooted in the community. But as a consequence, as they get older, they too will require support from the same social programs that citizens and legal immigrants are entitled to. And that has negative impacts in the government’s finances because these people are remaining in the country longer than they otherwise would.”
How Labor Productivity Shapes the Value of Immigration
Individual labor productivity determines the macroeconomic value of immigration, the paper noted. Higher labor productivity generates higher income for those more productive individuals (and higher revenue for the government), but it also generates an indirect positive spillover effect, the authors wrote.
“When effective labor supply rises due to rising average labor productivity, the return to capital also rises,” they explained. “Higher returns to capital incentivize more capital investment and thereby raise the average wage level for all workers.” But capital responds with a time lag to the increase in labor supply, reducing wages and income in the interim. That means “additional immigrants begin to weigh on the fiscal balance and the economy in the near-term until capital investment catches up.”
“I expected the legalization policies to be more growth-inducing in the long run than they turned out to be [in the paper’s model], which was surprising to me.” –Efraim Berkovich
“We see from the data that when an unauthorized person becomes legalized, they become more productive,” said Berkovich. That may have to do with “institutional factors” such as access to a wider range of jobs, he explained.
But immigration doesn’t always provide a productivity boost as is commonly understood, Berkovich said. “The assumption was that if you legalize the unauthorized population, you get this productivity boost. But as it turns out, from our model estimates, the size of that productivity boost is insufficient to overcome the demographic aging bomb, meaning that the [undocumented immigrants] who would have left the country before they collect Social Security now are legalized to stay, they collect Social Security, and that creates problems for the long-term fiscal health of the U.S. government budget.”
The paper cited data from the Congressional Budget Office that showed that between 1998 and 2018, unauthorized immigrants earned, on average, 48% less than citizens and lawful immigrants. The lower earnings of unauthorized immigrants are, in part, explained by their lower educational attainment — 42% of them lacked a high school diploma, compared with 15% of authorized immigrants and 6% of citizens, it added.
But education alone does not explain the earnings differences because “even at each level of education, unauthorized immigrants had lower earnings than the other two groups,” the researchers pointed out. “This disparity may be due to institutional factors which prevent unauthorized immigrants from participating in the job market to the extent of their underlying ability. Under this assumption, legalization leads to higher labor productivity.” That advantage to the broader economy may be negated when legalization increases federal spending on Social Security and other welfare programs, though the budget effect is partially offset by higher rates of tax compliance by the newly lawful immigrants, they added.
Experiments in the Study
In designing a framework to parse through multiple scenarios, the researchers considered two different types of immigration policies — policies that alter the inflow of legal immigrants and those that legalize unauthorized immigrants. Next, they conducted experiments with five scenarios:
- Almost zero legal immigration: This policy reduces the expected baseline inflows by 99% for five years.
- Half legal immigration: A reduction of 50% in the baseline inflow for five years.
- A quarter more legal immigration: A 25% increase in the baseline inflow for five years.
- Double legal immigration: A 100% larger immigration inflow for five years.
- Quadruple legal immigration: A four times larger than baseline immigration inflow for five years.
The researchers identified several outcomes from their experiments, notably these:
- Labor productivity increases slightly right after the legalization policy takes place because of the assumption that legal immigrants have access to formal labor markets and better work opportunities than unauthorized immigrants.
- After the discrete increase in productivity in the first 3-4 years after a legalization immigration policy, there is a decreasing trend with respect to baseline. Newly legalized immigrants are more likely to remain and, because they have a lower education achievement on average, the long-run available labor productivity per capita declines.
- When unauthorized immigrants gain legal status and their productivity rises due to the status change, output per capita goes up. Since legalized immigrants are less productive than the population average and are more likely to stay in the country, output per capita drops in the long-run.
- Unauthorized workers earn less than natives (mostly due to demographic characteristics, especially education). Legalization increases their propensity to remain in the country, thus decreasing per capita wage.
- Unauthorized immigrants are older than authorized immigrants or older than native born Americans. That goes against popular perceptions that legalization of unauthorized immigrants has significantly more fertility effects.
According to Berkovich, their framework will help policymakers to estimate the potential impacts of various immigration scenarios with more precision than existing models. He also hoped the paper triggers a discussion on the insights revealed by their model. “We created a model that we think can handle some of the subtle issues surrounding immigration,” he said.