Policy 1: Implement Skills‐Based Immigration
Senator Jeff Sessions posed a hypothetical during an immigration debate in 2013. Two young people are living in Honduras, and each has a strong desire to emigrate to the United States. One has learned English, was valedictorian at his high school, and is in his second year of college. The other dropped out of high school, has minimal skills, but has a brother already living in the U.S. Considering what’s in the best interest of this country, which of the two should be allowed in?
Our current immigration policy would prefer the high school dropout with a brother living in the U.S. to the English‐proficient high school valedictorian. This preference should be reversed. The United States should evaluate skills and capabilities of prospective immigrants and use this to prioritize access to residence here.
The goals of any U.S. government policy should be to serve the interests of the citizens of the United States. This seemingly bland statement should be the explicit starting point for any discussion of policy, but it is often forgotten or strategically ignored in immigration debates. An implication of this is that the interests of prospective immigrants should have zero inherent weight in determining our policy. Of course, this does not mean that we therefore must have a policy devoid of any consideration of the interests of prospective immigrants. After all, most American citizens place weight on strictly humanitarian help for many residents of foreign countries, most Americans also care about the expected state of American society long after their personal deaths, there may be prudential reasons to believe that assistance to foreign residents may rebound to the material advantage of current U.S. citizens, and so on. But it does mean that any justification for such humanitarian actions must proceed from the interests of current citizens.
The conventional economic wisdom for the effects of a shift to a skills‐based immigration system was summarized in by the Peterson Institute for International Economics as “likely to reduce the fiscal costs of immigration and to narrow the wage gap between high‐skilled and low‐skilled labor in the United States. However, these outcomes would entail the potential cost of diminishing well‐being in poor countries.” From the perspective of an American citizen, this sounds like an excellent deal.
Most conventional economic analysis is somewhat speculative, but fortunately the vagaries of the U.S. immigration system have created a natural randomized experiment to evaluate the real effects of high‐skill immigration. H-1B visas are granted to skilled, typically technology, workers. Most years there are many more applicants than available visas, and the visas are allocated by lottery. Researchers at the Partnership for a New American Economy compared metropolitan areas that were more versus less lucky in the percentage of H-1B applicants that won the lottery. Their conclusions were that luck in the lotteries corresponded with a small net increase in total employment with college degrees in computer fields, but also a large increase in employment without college degrees in computer‐related fields. That is, not only were the winners better off, but the computer sector of the whole city was better off, and those without college degrees were disproportionately helped.
We would not need to plow new ground to move in this direction. All of the major Anglophone democracies have done a far better job of this than America and have reaped the benefits. Australia, Canada, and New Zealand all have a higher foreign‐born population than the U.S., and all three plus Britain have more new immigrants each year per capita than the U.S. They have all used some kind of points system to select for immigrants with relevant skills, such as English proficiency and educational attainment, and extra points for degrees or expertise in such fields as science, technology, and medicine. They are generally moving to a two‐stage system in which foreign applicants who achieve at least some specificed target score under such a points system are put into a pool which prospective employers can browse, and are granted visas when specific employers offer them jobs. America should implement such a system.
We should also go beyond this. Because there is vastly more demand for U.S. residency than any number of spaces that could be realistically be made available, we should embed randomized experiments into the process in which we try various selection criteria and methods, and then compare results of immigrants over a number of years after entry. This would allow rapid, empirical improvement in our system. Further, we should market American society and treat our embassies and consulates as recruiting centers for high‐potential applicants rather than purely as gates that block potential criminals. We are already a magnet for talent, and we should build upon this.
Policy 2: Redeploy 5 Percent of Government Social Expenditures to Visionary Science and Technology Investments
America’s technology strategy through most of its history was to commercialize the discoveries of European science. We began investing massively in basic research in the post‐World War II era only because there was nobody else left to do it. Today, America is the global leader in basic science. Almost half of all the most cited scientific papers are produced in the U.S. But the world is changing, and in 2013 the U.S. represented only about 20 percent of world GDP and 28 percent of world R&D spending. Over roughly the past 20 years, the fraction of American scientific papers with a non‐American co‐author has grown from 12 percent to 32 percent. Science is re‐internationalizing.
To thrive in this new world, we need to adapt. We should give ground grudgingly but recognize that over time more science will be done outside the United States. We should participate aggressively in research collaborations such as international space‐exploration efforts and the European CERN particle‐physics facility, and we should fund exchanges and other vehicles to ensure that we gain maximum benefit in return for our basic‐research investments.
But we will also need to spend more money. According to the OECD, public social spending in the United States (including not only cash government expenditures but so‐called tax expenditures, private transfers that are compulsory by law, and the like) was just about 20 percent of GDP in 2012–13. That is on the order of $3 trillion per year. Total U.S. spending on R&D across all levels of government, academia, and non‐profits is on the order of $160 billion per year. We should reduce social expenditures by 5 percent, and redeploy this money to roughly double annual government R&D spending.
To spend this money most productively, we should think differently about what basic science we conduct here. We should bias basic‐research funds not toward those areas that inherently hold the greatest promise, but toward those in which the long‐run economic benefits are likely to remain in the United States because they require the build‐up of hard‐to‐transfer expertise or infrastructure that is likely to generate commercial spin‐offs. University and research laboratory rules and the patent system should recognize the long‐run desirability of researchers creating private wealth in part through the exploitation of knowledge created by these publicly supported institutions.
More fundamentally, the sweet spot for most government research funding will likely be visionary technology projects, rather than true basic research on one extreme or commercialization and scale‐up on the other. We have a long track record of doing this well and an existing civilian infrastructure that can be repurposed, including most prominently the Department of Energy’s national laboratories, the National Institutes of Health, and NASA. Each of these entities is to some extent adrift and should be given bold, audacious goals. They should be focused on solving technical problems that offer enormous social benefit, but are too long‐term, too speculative, or have benefits too diffuse to be funded by private companies.
What would this mean in practice for these leading national assets? A few examples may help point to an answer. We should, for instance, return the DOE’s labs to a more independent contractor‐led model with clearer goals but greater operational flexibility. We might set for one lab the goal of driving the true unit‐cost of energy produced by a solar cell below that of coal, and a second lab the same task for nuclear power. We could combine the 27 independent institutes and centers of the NIH into a small number of major programs and task each with achieving measurable progress against a disease. We could task NASA with leading an international manned mission to Mars.
Careful deliberation could lead to different specific targets, but in my experience great technical organizations have a characteristic spirit that starts with goals that are singular, finite, and inspiring. That is, each organization should have one goal. The goal should be sufficiently concrete that we can all know if it has been achieved or not. And it should be sufficiently impressive that people are proud to work toward it, without being so obviously outlandish that it just inspires cynicism. Beyond goals, political leadership has the responsibility for selecting extremely able senior leadership, providing adequate resources, granting operational autonomy, and measuring progress.
Skills‐based immigration and investments in these kinds of visionary projects would be synergistic. Talented potential immigrants worldwide would be inspired by these kinds of projects to try to come to America. The most talented immigrants would play key roles in these projects, and they and their descendants would become citizens.
The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.