People Over Robots

People Over Robots

We live in a technological age—or so we are told. Machines promise to transform every facet of human life: robots will staff factory floors, driverless cars will rule the road, and artificial intelligence will govern weapons systems. Politicians and analysts fret over the consequences of such advances, worrying about the damage that will be done to industries and individuals. Governments, they argue, must help manage the costs of progress. These conversations almost always treat technological change as something to be adapted to, as if it were a force of nature, barreling inexorably into the staid conventions and assumptions of modern life. The pace of change seems irrepressible; new technologies will remake societies. All people can do is figure out how best to cope.

Nowhere is this outlook more apparent than in the discussion of automation and its impact on jobs. My local grocery store in rural Utah has hung, with no apparent sense of irony, a sign proclaiming the company’s support for U.S. workers above a self-checkout machine, a device that uses technology to replace the labor of an employee with the labor of the customer. Much ink has been spilled in explaining how automation threatens some low-skilled workers and what governments should do to help: for instance, countries could support retraining initiatives, revamp education systems, or invest in redistributive schemes. At the same time, many governments hope that machines can save their economies from the consequences of demographic decline and aging. Techno-optimists argue that the United States and many other wealthy countries need automation to make up for dwindling working-age populations and looming gaps in workforces. Happily, they suggest, the advance of technology will sweep aside the troubles of demography.

But these debates and arguments miss a very simple point. As seismic as it may seem, technological change is not a natural force but the work of human beings. Of course, technology has radically improved human lives: no one wants to live without electricity, flush toilets, or (in Utah) central heating. In other cases, however, it is new policies, and not new technologies, that societies need most.

Automation is often a solution in search of a problem. It is a choice people have made, not an inevitability and certainly not a necessity. For instance, the United States faces a scarcity of truck drivers. The American Trucking Association has estimated that in 2021 there were 80,000 fewer drivers than the total needed and that, given the age of current drivers, over a million new ones will have to be recruited in the coming decade. To deal with this deficit, many tech moguls, including Amazon founder Jeff Bezos, have invested in the research and development of self-driving vehicles, technology that would reduce the demand for drivers. For Bezos, such technology makes corporate financial sense; Amazon relies on low shipping costs to keep its prices down. But it does not make wider economic sense because millions of people would be happy to drive trucks in the United States—they just need to be allowed to work in the country.

There is no global scarcity of people who would like to be long-haul truck drivers in the United States, where the median wage for such work is $23 per hour. In the developing world, truck drivers make around $4 per hour. Yet firms cannot recruit workers from abroad even at the higher wage because of restrictions on immigration, so business leaders in the United States are impelled to choose machines over people and eradicate jobs through the use of technology. But if they could recruit globally, they would have less incentive to destroy those jobs and replace people with machines. The implacable fact of national borders steers businesses toward investing in technology that does not respond to global scarcities—and that no one really needs.

What is true for truck driving is also true for many other industries in the rich industrial world that require nonprofessional workers in specific work environments. A 2021 report by the financial services company Mercer estimated that, by 2025, the United States would face a shortage of some 660,000 home health aides, lab technicians, and nursing assistants.

Barriers to migration encourage a terrible misdirection of resources. In the world’s most productive economies, the capital and energies of business leaders (not to mention the time and talents of highly educated scientists and engineers) get sucked into developing technology that will minimize the use of one of the most abundant resources on the planet: labor. Raw labor power is the most important (and often the only) asset low-income people around the world have. The drive to make machines that perform roles that could easily be fulfilled by people not only wastes money but helps keep the poorest poor.

To be sure, some social and political concerns regarding the cross-national movement of economic migrants are legitimate, including those about how such flows of people would be managed, affect current domestic workers, and create social tensions. Advocates are also right to worry about how migrant workers would be protected from exploitation. From the perspective of individual firms and industries, it is easier to figure out how to get artificial intelligence to drive a truck than it is to lower the bureaucratic barriers of immigration restrictions.

But choosing devices over people is a mistake. It leads the world to miss out on the real economic and humanitarian gains that would come from letting people move to where they are needed instead of trying to invent machines that can supplant humans. The refusal to allow people to cross national borders as economic migrants, especially to engage in jobs that require just core labor skills, massively distorts the trajectory of technological change in ways that make everybody, especially the world’s poor, worse off.


A popular view in the West holds that the governments of wealthy societies do not need to bring more workers into their countries. If anything, they can comfortably raise barriers as technological progress destroys so-called low-skill jobs. But this is just not the case.

Some technological changes are driven by progress in basic science. Often, discussions of the future of labor assume that the path and the pattern of technological change are already determined and that the effects on jobs and workers are just natural consequences of the inevitable progress of science. But economists have developed increasingly nuanced understandings of how technological change has reshaped labor markets and wages and how innovation is a consequence—not just a cause—of the costs that firms face.

For decades, economic and policy discussions about labor markets and technology had tended to focus on how shifts in wages reflected the skills of workers, where “skill” was treated as synonymous with “cognitive skill”, with a worker’s level of formal education as a rough proxy for cognitive skill. The standard analysis was that advances in information and communications technology have helped boost the incomes of highly educated workers and reduced those of all less-educated, low-skilled workers. But that is not what happened: although the wages of the highly skilled have risen in the United States since 1979 relative to those in the broad middle of earners, the wages in traditionally low-wage occupations also rose in percentage terms more than those in the middle—and over some periods by as much as those at the top.

Research by the economists David Autor, David Dorn, and others shows that the demand for various occupations in response to technology does not change in a simple linear relationship with worker skill but depends on the nature of the tasks that that worker must perform. Jobs in service occupations, such as food preparation, cleaning and janitorial work, maintenance, in-person health assistance, and security, involve manual and nonroutine tasks. Many manual nonroutine tasks are very difficult to either automate or offshore, as they require the direct physical presence of the worker, and so these jobs remain in demand, and their wages have remained robust even in the wake of technological advances.

Driverless forklifts at a warehouse in Roosendaal, Netherlands, November 2022. Piroschka van de Wouw / Reuters
Driverless forklifts at a warehouse in Roosendaal, Netherlands, November 2022. Piroschka van de Wouw / Reuters

Undoubtedly, truly revolutionary changes have occurred in how people communicate, seek information, organize and process data, and entertain themselves. But the notion that the rapid technological change in some sectors of the economy in recent decades has accelerated the transformation of the entire economy is wildly off the mark. In fact, the growth of economic productivity in industrial countries by the standard measure of growth of “total factor productivity” (which assesses productivity by dividing total production, or output, by its costs, or inputs) has been considerably slower in recent decades than it was between the early twentieth century and 1970. Nearly every developed country has experienced a substantial deceleration in productivity growth since 1980.

Meanwhile, the supply of workers for manual, nonroutine tasks has markedly decreased in rich industrialized countries, thanks to dramatically lowered fertility and rising levels of education. The number of open jobs is increasingly out of sync with the number of domestic candidates available to fill those jobs. The U.S. Bureau of Labor Statistics Occupational Outlook Handbook forecasts that, between 2021 and 2031, the occupational categories that require less than a college degree and for which existing median earnings are less than $40,000 will see a net rise of more than five million new jobs, with home health and personal care aides adding around 924,000 jobs and cooks adding 419,000. But according to UN demographic projections, the number of people between the ages of 20 and 40 in the United States (not factoring in any migration) will fall by more than three million in the same period. The medium-term demographic future of the native born in the rich industrial world is already clear: by the 2040s, there will be millions too few native-born people in developed countries available to perform all the essential, nonroutine, manual tasks in the economy.

The more than trillionfold increase in computing power over the last century has radically changed those occupations in which people did routine and repetitive tasks. The African American women who made the moon landing possible (and were made famous in the movie Hidden Figures) worked for NASA as “computers”, busily making numerous calculations that machines can now make in an instant. There are no more human computers, and machines called computers are just markedly more effective than humans at computing. But for many tasks, no increase in computing power will make them more appropriate for machines to perform. Machines are not better at personal care, machines are not better cooks, and machines will not necessarily be better than people at driving trucks.

If self-driving trucks eventually proliferate on American roads, it will not be evidence of the relentless march of scientific progress. Rather, their use will be a demonstration of something else entirely: the consequences of barriers to the movement of labor that create massive private financial incentives that, in turn, drive powerful people and corporations to undertake enormous investments of scarce human resources in technological innovation—with the wider result of replacing people with machines. Ultimately, U.S. firms opt for automation because it is far easier to solve even very hard technical problems, such as those that self-checkout machines and driverless vehicles attempt to solve, than it has been for countries to address the social and political obstacles that prevent them from allowing foreigners to do those jobs.


The arbitrary facts of birth, nationality, and citizenship profoundly affect people’s lives. Where people are born and where they can move fundamentally shape how much money they can earn over the course of their life. Restricting the movement of people across borders creates a massive price differential between equally productive workers.

In research I did with co-authors, we looked at the earnings of workers born and educated in 42 different countries. We compared the earnings of those who remained in their home countries with those who worked in the United States. We adjusted these earnings for the differences in the price of goods and services between countries to take into account discrepancies in purchasing power. The wage differential for equally productive workers between the 42 countries we studied and the United States ranged from two to ten times as much in the United States, averaging around four times as much. Such proportions are evident across occupations (including those of waiters and truck drivers) and across education and skill levels.

This gulf in the wages of equally productive people in different countries is the largest policy-induced price distortion in the world today (and probably in all of human history). Barriers to migration generate an artificial scarcity of labor. Many industries in the United States struggle to find workers at labor costs they can afford. That deficit spurs companies to search for solutions through automation and other technologies that are unnecessary and inefficient.

Experience shows that letting more workers into a country would indeed change patterns of innovation. The United States has run this experiment before, in reverse. In the middle of the twentieth century, the United States allowed the seasonal migration of agricultural guest workers from Mexico under the rubric of the Bracero Program. The government eventually slowed the program and finally stopped it entirely in 1964. Researchers compared the patterns of employment and production between those states that lost Bracero workers and those that never had them. They found that eliminating these workers did not increase the employment of native workers in the agricultural sector at all. Instead, farmers responded to the newly created scarcity of workers by relying more on machines and technological advances; for instance, they shifted to planting genetically modified products that could be harvested by machines, such as tomatoes with thicker skins, and away from crops such as asparagus and strawberries, for which options for mechanized harvesting were limited.

Necessity may be the mother of invention, but false necessity is the mother of dumb inventions. Prohibition in the early twentieth century banned the importation, production, transportation, and sales of alcoholic beverages in the United States. These restrictions were an enormous boon to illegal distillers of moonshine, who saw demand for their booze explode. But their product needed to be shipped to thirsty drinkers. Just hiding flasks in boots (the origin of the term “bootlegger”) was insufficient. To transport more moonshine, people built “moonshine runners”: vehicles that could speedily carry heavy shipments of moonshine without drawing attention. The development of moonshine runners required technological savvy and innovation, but they still represented a dumb invention. The banning of a perfectly ordinary economic transaction—buying alcohol—led not to better liquor trucks but to the innovation of making a liquor truck look like a regular vehicle, which was a pure waste of time and talent.


Proponents of barriers to migration insist that they are necessary to protect the wages of current citizens, but that is not true. There have been times in the past century when governments were concerned that their country could not provide sufficient jobs for their citizenry, but the shifting demographics of the rich industrial world have changed that logic completely. For the foreseeable future, the challenge will be finding enough workers to fill available jobs. Even countries that have traditionally not welcomed immigrants, such as Japan, are now aggressively recruiting workers from abroad. They can do so with the knowledge that immigrants do not necessarily hurt the wages of natives. A 2017 National Academy of Sciences review found that the net impact of immigration on the average wages of domestic workers in the United States was either zero or, more likely, slightly positive. In economic terms, migrants are not substitutes for the typical U.S. worker, but complements, so that more migrants actually raise the average wage of citizens. Having more assistants, for instance, does not lower but rather raises the wages of skilled workers, such as nurses, by freeing more of their time for tasks that need their unique skills. Of course, some particularly disadvantaged U.S. workers may compete directly for jobs with migrants, but limitations on migration are neither an effective nor an efficient way to help those native workers. Programs such as the Earned Income Tax Credit are a vastly more cost-effective means of buttressing the wages of domestic workers. I have done calculations to show that, even under more pessimistic assumptions about the negative impact of migrants on the wages of disadvantaged native workers, only a modest increase in the EITC is required to completely offset those losses—and it is a tiny fraction compared with the economic loss that would result from banning migrants in the first place.

The main economic losers from the restrictions rich countries place on labor mobility are the world’s poor. Decades of well-intentioned development programs and aid initiatives cannot equal the benefit of permitting a person in a poor country to work in a wealthier, more productive one. If they want to help the world’s poor, citizens of rich countries should understand that all the worthy development projects, antipoverty programs, and foreign aid to poor countries have an inconsequentially small effect compared with the benefits of just letting people move to the rich countries that need them and work for the going wage justified by their productivity.

For instance, a widely cited 2015 paper published in Science considered the effects of an antipoverty program that transferred livestock across six poor countries, with the goal of increasing the income of chronically poor households. The program spent $4,545 per household in its first two years. By the third year, annual household consumption was higher by just $344 on average across five of the six countries where the program produced positive results—Ethiopia, Ghana, India, Pakistan, and Peru. (In Honduras, nearly all the livestock died.) And, given that so many attempts through similar projects to raise the incomes of the poor had failed, this modest gain of $344 in annual household consumption by spending $4,545 was regarded by the authors as a major success.

By contrast, my research suggests that workers without a high school degree would make, on average, as much as $13,119 more per year in the United States than would their counterparts in the five countries studied. Even if ten percent of the wage differential is absorbed in the travel costs of moving back and forth for a year, allowing the same low-skilled workers to be employed in the United States, rather than consigning them to their home country, would produce a boost in income 35 times larger than that enabled by an effective, well-designed, and well-implemented antipoverty program.

The phenomenon of global poverty today is not one of “poor people” but of people trapped in “poor places”, unable to leave because of barriers limiting their movement. The derisive caricature of poverty has it that people are poor because they lack “human capital”, but the reality is that the massive expansion of education in the developing world since the 1950s means that the average adult in Haiti today has had more schooling than the average adult in France had in 1970. But Haiti is a chaotic and low-productivity place to use any type of capital, including human capital; hence, most Haitians who have escaped poverty have done so by leaving their country. Some may fret about “brain drain”, the spurious notion that a poor country will be further immiserated by losing its best and brightest overseas. The principal attraction of arguments regarding “brain drain” seems to be that the words rhyme, as there has never been any evidence that outward migration in general has harmed a country’s prospects. It is worth remembering that many of the richest countries in the world today—including Denmark, Italy, Norway, Spain, and Sweden—had some of the highest emigration rates in the late nineteenth and early twentieth centuries.

Wage differentials create a deep desire to traverse national borders. Between 2015 and 2017, Gallup asked people around the world whether they would move permanently to another country if they could and, if so, to which country they would move. From these samples, one can estimate that around 750 million people would choose to leave their home countries permanently if they could (and even more people would be willing to move temporarily). Based on the survey, 158 million additional migrants would want to come to the United States; Australia, Canada, France, Germany, and the United Kingdom would each receive around 30 million more migrants. This is not to say that these countries should or would accommodate so many migrants, but it does suggest that there is no shortage of people ready to come and work in rich countries.

The restrictions on mobility that create labor scarcity in rich countries perpetuate poverty for millions of people who are willing and able to work productively—but are prohibited from doing so. And the scarcity propels businesses to invest wastefully in technology that need not exist. Automation, in other words, is not inevitable but driven by the artificial scarcity of labor. Companies perceive a financial incentive to choose machines over people. Without such an incentive, businesses and households would make different choices. Walmart is not making you ring up your own purchases on self-checkout machines because it thinks you always had a hankering to work for the company but because it cannot find all the workers it needs at viable costs. And households, too, can make different decisions that benefit everyone involved. Research in Singapore shows that highly skilled women are much more likely to be in the labor force when care workers are available to take up tasks in the home. The availability of home aide workers also allows the elderly to stay out of institutionalized care longer and leads to greater quality of life at a much lower cost.


It would be no small thing, of course, to figure out how countries could realize the potential of available labor in the world. The major mechanisms for global cooperation—chiefly those institutions that emerged after World War II, such as the United Nations, the International Monetary Fund, and the World Bank—set in motion the globalization of markets for goods and capital. But they did not put in place any meaningful infrastructure to support and promote the movement of labor. Money and shipping containers flowed freely across borders, but people did not.

Today, every country unilaterally sets whatever restrictions it wants on the entry of foreign nationals. Belying the facile right-wing talking point that the United States maintains “open borders” or similarly dubious claims that the world is “flat”, all rich countries have transformed their borders into cliffs, with the threshold for legal entry high and often altogether unreachable. At great expense and with mixed results, countries enforce these restrictions. In 2022, the United States allotted $26 billion to border enforcement, more than it has devoted in most years to its international development agency, USAID.

Instead of funneling resources into the quixotic pursuit of job-killing technology, countries should pursue international cooperation with regard to labor mobility. The potential gains for the countries from which people emigrate, the migrants themselves, and the countries that receive migrants are enormous. Rich countries should allow more people to live and work in their countries not out of altruism but to reckon with a growing demographic necessity.

Harvesting oranges in Lake Wales, Florida, April 2020. Marco Bello / Reuters
Harvesting oranges in Lake Wales, Florida, April 2020. Marco Bello / Reuters

The solution is the creation of a global mechanism for labor mobility. It should recruit workers fairly without excessive costs and, based on reliable information and contracts, place them in jobs that suit their abilities, protect them from abuse while they are away from their home countries, and, with time-limited mobility agreements, facilitate their orderly return home. Larger flows of economic migrants will require an industry to handle the key functions of recruitment, training, placement, protection, and return. The trucking industry, the health-care industry, and the hospitality industry cannot be expected to manage the international movement of labor to meet their needs. Instead, the task of moving people must be taken up by a globally connected and networked group of organizations and individuals. Of course, such an industry should be carefully regulated and monitored, as the risks of abuse are immense. But a well-functioning, ethical global industry that moves workers can be a huge force for good by matching people who want jobs with the enterprises that need them.

Global associations already cooperate with industry and governments to produce positive outcomes in the movement of people and things. Look up from reading this article, and you will see items around you that moved about the world as part of the 11 billion tons of maritime freight shipped every year. Or if reading online, don’t look up; your device is almost certainly one of those items. Over 4.5 billion airline passengers traveled around the world in 2019, with only 283 fatalities. (My thinly populated home state of Utah alone had 320 traffic deaths in 2022.) Safe worldwide travel on airlines has been created and supported by governments, international associations, and industry groups. Similarly, in 2018, 1.4 billion tourists traveled internationally, supported by a wide variety of industries and industry associations that facilitate such a massive movement of people safely and reliably.

Political obstacles stand in the way of the orderly cross-border movement of people to work. Countries are stuck in a Catch-22. Politicians are reluctant to create the laws, policies, and regulations that would allow labor mobility programs to flourish until such programs have been demonstrated to be safe, effective, and beneficial. But existing wage differentials between rich and poor countries, as well as the undeniable demand for workers, means that the movement of people happens anyway, but without legal sanction and with the complicity of employers. Invariably, such movement is unsafe, migrants are exploited and abused, and they cannot easily return home. As a result, the very idea of labor mobility is tainted.

It may seem paradoxical, but the pitfalls of labor mobility in the present are reasons to facilitate even more movement, only through legal and well-devised channels. The benefits of allowing people to move where their labor is needed are huge for all concerned. Rich and democratic societies need to stop blindly pursuing technological advances that economize on precisely what is abundant around the world. Wealthy countries have created strong incentives for their firms and innovators to choose machines over people. It is time to make the bet on a future built by and for people.

Lant Pritchett is Research Director of Labor Mobility Partnerships, RISE Research Director at the Blavatnik School of Development at the University of Oxford, and a former World Bank economist.

Deja una respuesta

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *