Immigration to the United States is a complex and thorny topic. Setting aside the impact of immigration on American culture and politics, as well as the balance of taxes and spending, this AIER explainer looks at the economics of immigration, both legal and illegal.
To understand the economic effects of immigration, we need to understand the motivations for immigrants to come to the US, particularly the economic incentives. Various economic models can be used to analyze which types of immigrants countries like the US tend to attract and what impacts they have. While no single model provides a complete picture, together they offer a more comprehensive understanding of the economic effects of immigration.
Labor as a Factor of Production
Factor Complementarity
The simplest way to think about immigration is as an increase in the labor supply. Labor is a factor of production, that is, a general type of input into the production process. Other factors of production may include land, capital, natural resources, and entrepreneurship. Technology is usually not considered a factor of production. Instead, technology is thought of as those ideas, processes, and techniques that make all factors of production more productive.
Entrepreneurs hire labor to make and do things. Without labor, entrepreneurship, capital, land, and the like are not productive. By the same token, without capital, land, and labor are not productive. This feature of factors of production is called factor complementarity: the more other factors a given factor can interact with and supplement, the more productive it is. Each hour of labor is more productive when it can use more capital (machines, entrepreneurship).
What entrepreneurs pay owners of labor (workers) in exchange for their services is called a wage. Wages are the price of labor (and human capital). The price of capital is usually called the interest rate, the price of entrepreneurship is economic profit, and the price of land is land rent.
Workers want to go where the wage is highest. If they can make more money in another industry or another location, they have a clear incentive to switch industries or locations. By balancing labor supply and demand, this constant switching in the economy drives long-run wage differences between industries and locations toward zero.
Thinking of immigrants and other workers as mere quantities of labor risks oversimplification, however. Like other workers, immigrants bring their distinct talents and interests into the labor market, learn on the job, and create new ways of doing things. Even unskilled workers have different levels and types of specialized knowledge that keep them from being perfectly interchangeable. Workers can also be capitalists: investors or entrepreneurs. By one estimate, immigrants in the US start businesses at 1.8 times the rate of natives. Illegal and legal immigrants have different outside options and therefore different levels of bargaining power with their employers.
In all these ways immigrants and workers are much more complex than mere quantities of labor, but to understand an economic phenomenon, we often have to start with a simple model before adding complexities.
Mobility and Specific Factors
Labor is mobile across industries and, if regulations allow it, across locations in the long run.
But how long is the long run? If we exclude human capital from the definition of “labor” and consider “labor” to refer only to raw, manual labor power, then labor moves quickly and readily across industries and, if allowed, locations. The skills required for picking crops, cleaning rooms, preparing fast food, mowing lawns, and such can be easily learned, and so workers can move easily among these jobs. We shouldn’t expect wages for these jobs to be very different. If demand for cleaning rooms rises, causing cleaners’ wages to rise, workers should move from fast food into cleaning, causing wages to fall back for cleaners and rise for fast food workers, until they are roughly equal and there is no longer any incentive to switch jobs.
But human capital is much more occupation- and industry-specific. It is difficult to apply the skills of a physician to the job of computer programming. The wage return to human capital should equilibrate only slowly across industries. Nevertheless, the same processes still do work, even if over a generation or two. If the demand for computer programming falls because of AI, for example, we should expect programmers’ wages to fall, while everyone else benefits from cheaper, better software. Programmers themselves will look for work in careers where the skills they’ve developed will be of some use, but the effect is even stronger on the upcoming generation. Fewer students will strive to learn programming skills; instead, more are likely to apply themselves to medicine, law, or finance, assuming wages in those fields remain high. As more new workers flood these fields instead of programming, the wages for computer programmers will gradually recover to attract the people needed to perform tasks that cannot be automated, while the wages for physicians, attorneys, and finance workers will fall, until the wages across these higher-skilled industries are roughly equal and there is little financial incentive for students to develop one set of skills rather than another.
In a dynamic economy, equilibrium is never actually reached. But the theoretical equilibrium does work like a kind of gravitational force, drawing workers away from where they are least needed and toward where they are most needed.
Labor mobility across locations is very observable where it’s allowed. In the United States there are ordinarily no restrictions on moving from place to place. As a result, we should expect wages for the same skill level in different places to be roughly equal unless some places are nicer than others to live in, that is, have different amenities.
US evidence suggests that total real compensation (price-adjusted wages plus amenities) does tend to equalize across locations, for a given skill level, when fully open immigration is allowed.[1] In 2000, the places with the lowest real wages in the US were places like Santa Cruz, Honolulu, San Francisco, Santa Barbara, San Jose, rural Hawaii, and Jacksonville, North Carolina (Chen and Rosenthal, 2008). San Diego, Cape Cod, rural Montana, and rural Vermont were also near the bottom. Presumably, these are the nicest places in the country to live in: workers were willing to accept lower wages to be there. Among the highest-earning geographies were Kokomo, Indiana; Wilmington, Delaware; Flint, Michigan; Waterbury, Connecticut; Detroit; Trenton; Philadelphia; and Hartford, Connecticut. Houston, Newark, and Baltimore were also high on this list. These are, apparently, the least nice places in the country to live in: workers had to have high wages (relative to rents) to live there.[2]
Still, we must remember that wages, like other prices, coordinate countless bits of information in the heads of employers and workers, relating to tastes, opportunities, and other circumstances. These circumstances and consumer tastes are constantly in flux, and therefore wages never reach a static equilibrium.
The Heckscher-Ohlin and Stolper-Samuelson Theorems
The Heckscher-Ohlin and Stolper-Samuelson Theorems help explain how immigration affects the United States economy, which is an advanced country rich in physical and human capital. These theorems are usually used to explain the effects of foreign trade, but their logic applies equally well to immigration.
The Heckscher-Ohlin Theorem (1933) posits that different places start out with different factor endowments and then infers that places will specialize in producing the goods that intensively require their relatively abundant factors and import goods that intensively require their relatively scarce factors.
For example, the United States is capital- and land-abundant, but labor-scarce by contrast. This might also be phrased as the US having a high ratio of skilled labor to unskilled labor, and a low ratio of labor to land, compared to most other countries.
The Heckscher-Ohlin Theorem then predicts that the US will be a low-cost producer of goods that use a lot of capital, skilled labor, and land relative to unskilled labor, because those inputs are relatively cheap, and a high-cost producer of goods that use a lot of unskilled labor relative to capital, skilled labor, and land, because that input is relatively expensive. So the US will be able to sell capital- and land-intensive goods abroad while importing labor-intensive goods.[3]
The Stolper-Samuelson Theorem says that a change in the relative prices of goods more than proportionally changes the prices of the factors used intensively in their production (Samuelson, 1948). For example, let’s say the US starts out with no trade with the rest of the world. Capital-intensive and land-intensive goods are cheap, and labor-intensive goods are expensive because of the US factor endowments. Once the US opens to trade, capital-intensive and land-intensive goods rise in relative price domestically, and labor-intensive goods fall in relative price. The overall effect is to increase US productivity, as David Ricardo discovered, and as Ludwig von Mises and Donald Boudreaux have elaborated. But holders of capital and land benefit from trade, while holders of labor lose from trade, even taking into account lower prices on imported and import-competing goods.
Note that by “holders of capital” we include skilled workers, who hold a lot of human capital. “Holders of labor” are unskilled workers, who have only their raw manual power to offer. They are relatively scarce in the US, so with international trade, companies will try to economize on their use, and consumers will buy goods made with a lot of unskilled labor from abroad, where these goods can be made more cheaply.
The phenomenon of factor price equalization(FPE), predicted by these two theorems, has been much debated in economics. Everyone agrees that FPE happens to some extent, but the question is how much of the growing inequality between the wages of skilled and unskilled workers in the US and other advanced industrialized economies is the result of globalization (trade, capital mobility, and immigration) versus technology.
The counterintuitive truth to keep in mind is that the United States’ aggregate economic gains from globalization are larger the more that globalization changes factor prices—that is, suppresses the wages of the unskilled and boosts the wages of the skilled. The reason for this lies in the theory of comparative advantage, cited above. Economies benefit from trade the more their domestic prices without trade diverge from world prices. That’s the primary reason small countries benefit more from trade than large countries (Alesina, Spolaore, and Wacziarg, 2000). If trade really lets Americans economize on unskilled labor that is relatively expensive to us, that’s a huge benefit to the US economy, but it also drives up inequality. By contrast, in poor countries trade reduces inequality because skilled labor is scarce and unskilled labor is abundant. As a result, trade, capital mobility, and immigration should reduce absolute poverty, since unskilled workers in poor countries are the poorest people on earth.
Comparing Trade and Immigration
In the real world, opening trade has not led to anything near complete factor price equalization worldwide. (Remember that we do observe substantial FPE across locations in the US, however!) Even unskilled workers in the US make a much higher wage than unskilled workers in, say, Bangladesh or Bolivia. The reason for this is that total factor productivity (TFP) is higher in the US because our level of technology is higher, our laws and political system are more stable, and the government interferes less with the free-market system. As a result, an American unskilled worker is far more productive than a Bangladeshi unskilled worker. An American skilled worker is far more productive than a Bangladeshi skilled worker. Capital and land are more productive here, too.
Herein lies the economic difference between trade and factor mobility. When capital and labor can move across borders, they can access the TFP of the most productive economies. It’s clear that unskilled workers have an incentive to move from unskilled-labor-abundant economies, where wages are low, to unskilled-labor-scarce economies like the US, where wages are high. That follows just from factor endowments. But there’s even a case for skilled workers to move from developing countries to the US because TFP is higher here, allowing them to earn higher wages even though they are leaving a place where their assets are relatively scarce for one where their assets are relatively abundant.
For this reason, the economic gains from immigration are potentially much higher than those for trade. At the current margin, just about every worker in a low-TFP economy could become more productive by moving to a high-TFP economy (Clemens, 2011). The world as a whole could be much more productive and richer if workers could seek the highest wage their labor commands.
Now, there are two important things to note about this prediction. The first is that this conclusion follows only if mass immigration doesn’t have massive negative effects on TFP. Some critics of immigration have posited that this is precisely what will happen because of immigrants’ use of the welfare state, voting patterns, or cultural values and intelligence (Jones, 2022). Any full assessment of immigration policy must assess these claims and take those with strong evidence into account.
Second, as with trade, the economic gains from immigration are larger the greater the pre-immigration wage differences between economies. If healthy, English-speaking Bangladeshi unskilled workers moved to the US en masse, they’d be expected to drive down the wages of native unskilled workers, and this process is part of what generates the economic gains from Bangladeshi immigration.
Some pro-immigration commentators deny this effect. They say that immigration not only increases the supply of labor, it also increases the demand for goods and services, which in turn offsets the wage effect of immigration.
This claim is incorrect. First of all, the mechanism by which they claim immigration raises wages here implies that it increases prices. That’s the whole reason producers would bid up labor to produce more goods. But if immigration drives up prices, then it still drives down real, inflation-adjusted wages. Second, immigration can’t affect aggregate demand much in the long run. Aggregate demand affects choices most in the short run, until expectations change and prices adjust. Furthermore, the Federal Reserve should be expected to offset the alleged demand effects of immigration with tighter monetary policy, in order to reach its inflation target.
As supply and demand would predict, an exogenous increase in the supply of a factor of production will reduce its price. Immigrants will bid down the wages of native workers with precisely the same skill sets and industrial and occupational concentrations. But does this theoretical result happen in practice?
The Evidence on Immigration’s Distributional Effects
Does unskilled immigration reduce the wages of unskilled workers already here? Economists haven’t come to a consensus on this point. A key point of dispute concerns whether most unskilled immigrants to the US constitute a substitute for unskilled native labor, or not. Most unskilled immigrants to the US do not natively speak English. As a result, they may be less productive than native unskilled workers in service industries.
An influential early review of the literature suggested that the wage effects of immigration are much stronger than those of trade (Borjas, Freeman and Katz, 1997). Since then, economists have tried to identify “natural experiments” that let them draw causal conclusions more confidently.
One such natural experiment may be the Mariel Boatlift of 1980, which brought to South Florida an estimated 125,000 Cuban refugees, most of whom hadn’t finished high school and did not speak English. The original Boatlift study showed no wage effects (Card, 1990). An influential reassessment then showed a negative effect on the wages of high school dropouts (Borjas, 2017). However, a subsequent critique showed that the data Borjas used are unreliable because the sample sizes are too small to infer wages for the broader populations under investigation (Clemens and Hunt, 2019).
If immigrants are more complementary to native workers than substitutes for them, then they don’t compete much with native workers. Effectively, they offer different factors of production. That’s just what a study of “occupational task-intensity” data finds (Peri and Sparber, 2009). Along the same lines, Ottaviano and Peri (2012) find that in the long run, immigrants increase the wages of natives at all skill levels.
A more recent, high-quality study finds that Mexican immigrants impelled to the US by the peso crisis reduced native low-skilled wages and raised rents in the short run, but in the long run only reduced the wages of those low-skilled natives who entered the labor market in high-immigration years and reduced rents in locations where immigrants disproportionately entered the construction sector (Monras, 2020). Further work taking immigrants’ location decisions into account finds that immigrants concentrate in high-productivity cities so that they can earn money to send home. As a result, they improve the productivity and wages of natives more than they would otherwise (Albert and Monras, 2022).
The bulk of the evidence is therefore consistent with economic theory. Immigrants reduce the wages of natives with whom they directly compete, but they raise the wages of most native workers in the long run. Skilled immigration to the US is widely regarded as having positive effects on most workers, with the possible exception, in the short run, of native workers in identical occupations and industries.
What Economics Can Teach Us About Immigration
Economic theory tells us that immigration directly benefits economies for the same reason and in the same manner that trade does: it allows labor to work where it is most productive, making all other factors of production more productive. The government is no better at centrally planning labor markets than it is other markets: by incorporating dispersed, tacit knowledge, wages coordinate the choices of immigrants and their employers in myriad ways that no one could foresee. Moreover, immigration has a big positive effect on productivity that trade does not have: it gives workers everywhere access to the most advanced technology.
Popular discourse on immigration does not track the insights of economics. Immigration should have the biggest benefits for America as a whole precisely when it drives down the wages of our scarcest factor of production, unskilled labor. The same is true of trade. For these reasons, economists have often proposed policies to “compensate” low-skilled natives for the fact that their wages under globalization will be lower than under autarky.
Our biggest mistake, however, might be to think of people as if they were factors of production, rather than to understand that factors of production are things that people own. All of us at one time had only unskilled labor to offer the labor market. Many immigrants end up starting businesses and becoming owners of capital, raising the productivity of native Americans. Moreover, if American policies better supported skill development and productive work, fewer workers would have only unskilled labor to offer the market.
As Adam Smith discovered, economic progress depends on deepening the division of labor, which in turn is made possible by ever-larger networks of exchange. As a form of voluntary exchange for mutual benefit, immigration deepens labor markets and facilitates a more finely-grained division of labor.
This explainer does not address the indirect effects of immigration on the economy, which may work through culture, politics, or the welfare state. Those indirect effects are the ones most difficult to measure, and controversy about them is likely to continue. But before we can have a sensible conversation about immigration policy, we need to understand the economic basics.
References
Albert, C. and Monras, J. (2022) “Immigration and Spatial Equilibrium: The Role of Expenditures in the Country of Origin,” American Economic Review, 112(11), pp. 3763–3802. Available at: https://doi.org/10.1257/aer.20211241.
Albouy, D. (2016) “What Are Cities Worth? Land Rents, Local Productivity, and the Total Value of Amenities,” The Review of Economics and Statistics, 98(3), pp. 477–487.
Alesina, A., Spolaore, E. and Wacziarg, R. (2000) “Economic integration and political disintegration,” American Economic Review, 90(5), pp. 1276–1296.
Bernard, A.B., Redding, S.J. and Schott, P.K. (2013) “Testing for Factor Price Equality with Unobserved Differences in Factor Quality or Productivity,” American Economic Journal: Microeconomics, 5(2), pp. 135–63. Available at: https:// doi.org/10.1257/mic.5.2.135.
Borjas, G.J. (2017) “The Wage Impact of the Marielitos: A Reappraisal,” ILR Review, 70(5), pp. 1077–1110. Available at: https://doi.org/10.1177/0019793917692945.
Borjas, G.J., Freeman, R.B. and Katz, L.F. (1997) “How much do immigration and trade affect labor market outcomes?,” Brookings Papers on Economic Activity, 1997(1), pp. 1–90.
Card, D. (1990) “The Impact of the Mariel Boatlift on the Miami Labor Market,” ILR Review, 43(2), pp. 245–257.
Chen, Y. and Rosenthal, S.S. (2008) “Local Amenities and Life-Cycle Migration: Do People Move for Jobs or Fun?,” Journal of Urban Economics, 64, pp. 519–537.
Clemens, M.A. (2011) “Economics and emigration: Trillion-dollar bills on the sidewalk?,” Journal of Economic Perspectives, 25(3), pp. 83–106.
Clemens, M.A. and Hunt, J. (2019) “The labor market effects of refugee waves: reconciling conflicting results,” ILR Review, 72(4), pp. 818–857.
Jones, G. (2022) The Culture Transplant: How Migrants Make the Economies They Move to a Lot like the Ones They Left. Redwood City, Calif.: Stanford University Press. 13
Leontief, W. (1953) “Domestic Production and Foreign Trade; The American Capital Position Re-Examined,” Proceedings of the American Philosophical Society, 97(4), pp. 332–349.
Monras, J. (2020) “Immigration and Wage Dynamics: Evidence from the Mexican Peso Crisis,” Journal of Political Economy, 128(8), pp. 3017–3089. Available at: https://doi. org/10.1086/707764.
Ottaviano, G.I.P. and Peri, G. (2012) “Rethinking the effect of immigration on wages,” Journal of the European Economic Association, 10(1), pp. 152–197.
Peri, G. and Sparber, C. (2009) “Task Specialization, Immigration, and Wages,” American Economic Journal: Applied Economics, 1(3), pp. 135–69. Available at: https://doi. org/10.1257/app.1.3.135.
Samuelson, P.A. (1948) “International Trade and the Equalisation of Factor Prices,” The Economic Journal, 58(230), pp. 163–184. Available at: https://doi.org/10.2307/2225933.
Endnotes
[1] One paper that purports to find otherwise – Bernard et al. (2013) – does not attempt to adjust wages for local prices or amenities, substantially vitiating its results.
[3] Leontief (1953) found that US exports were more labor-intensive and less capital-intensive than its imports, but the well-known “Leontief Paradox” is explicable in that the US appears to have had a durable comparative advantage in skilled labor-intensive production.
