[Linkpost] Lant Pritchett ‘The political acceptability of time-limited labor mobility: Five levers opening the Overton window’
Lant Pritchett has written an important article arguing that the political barriers to large amounts of immigration are likely to fall substantially in the coming decades due to declining fertility in rich countries and the consequent need for more workers to support an ageing population.
From a global development point of view, I think this could be one of the most important ‘mega-trends’, given the large benefits to immigrants of immigration. From a longtermist point of view, I also think it could be one of the most important mega-trends in the coming decades (below AI, pandemic risk, and Great Power conflict). The geopolitical and long-term economic implications of historically unprecedented immigration seem potentially important. E.g. China’s population is already declining and fertility is very low. They will likely have to substantially increase immigration in the coming decades. What will this mean for the Chinese political system? What does the fertility decline and immigration mean for trends in very long-run economic growth?
A key passage in Pritchett’s piece:
“Figure 1 illustrates to keep the support ratio constant at the 2020 level of 3.08, an additional 101 million elderly need 101*3.08=310 million more labor force aged in 2050. But given the demographic changes that have (mostly) already happened—after all, everyone who will be over 28 years old in 2050 is already born—there will not be 310 million more, but rather 143 million less native-born workers. The “support ratio constant” labor force aged population would have to be 454 million higher than the projected 624 million labor force aged of the Zero Migration scenario. Maintaining a constant support ratio would imply that in 2050 42.1 percent of the labor force aged population would be the result of migration between 2020 and 2050. Put another way, there would have to be about 13 million additional workers per year into these developed regions if the support ratio is to remain constant.”
i.e. to retain constant support ratios for the elderly in rich countries, 42% of the working age population would have to be immigrants.
Abstract. A substantial expansion of migration and labor mobility in the rich industrial countries currently seems outside the Overton window, the range of acceptable political discourse. If anything, the general mood seems to favor even greater restrictiveness. I argue that five trends that are underway that could, within a decade or less, bring larger flows of migrants and labor mobility—including a major expansion of time-limited labor mobility—squarely onto the global and domestic political agenda of rich industrial countries
As an example of how powerful these demographic shifts will be, this recent paper claims that ~all of Japan’s poor economic performance relative to other developed nations since the ’90s can be explained by its demographic shift (specifically the decline in the population share of working age adults). Think about how much consternation there has been about Japan’s slow growth. We’re all headed that way.
Interestingly, AFAIK Japan has not drastically liberalized its immigration much in response to its slow growth. The proportion of foreign-born residents has grown a bit, but not much. Maybe this is changing, but we’ll see if anything actually happens, and Japan has been struggling to grow for decades.
If we’re talking about the need for more workers on timelines of decades, advancements in AI and automation will be relevant. Will immigrants be necessary in decades?
Also, fertility rates are falling pretty quickly globally, so immigration might only help so much for so long on its own.
But what about the other way around? It is far better if jobs move from the rich to the poor world. Online Kenyan administratives for UK and US firms, Peruans for Spanish firms, etc…
The most consequential effect of the pandemics has been that it was a demo of how easily many service jobs can be performed online, and that allows for “online migration”. Workers can stay at home enjoying their very cheap local cost of life while getting higher foreing wages.
Labour and goods and capital migrations have similar economics effects, but goods and capital movements do not create the same political resistance.
I think both of these trends can occur simultaneously
I’m not sure it’s very helpful to think of this as “jobs moving from one country to another”. It makes it seem zero-sum, whereas it is actually a positive-sum efficiency gain
Migrants to higher-income countries benefit from public goods like better services and public safety in addition to higher incomes
As Lant has pointed out, the higher income someone gains from moving from a low- to high-income country is enormous. IIRC it can be something like a 10x increase in consumption even if they’re working the same job. So even if we imagine some fixed pool of jobs, I’m not sure it is “far better” for jobs to move from high- to low-income countries. Given the choice of working the same job in a high-income or a low-income country, I think many people would choose to move to the high-income country.
A 10x increase in consumption doesn’t pass the sniff test, and indeed migrants to the US earn on average 2x more than before they migrated, 3x if they come from very poorest countries. (source, table 2)
Interesting, thanks for checking that!
What I had in mind were the data from this Pritchett paper. He sets out a range of estimates depending on what exactly you measure. For example he shows that the US wage for construction work is 10x the median of the poorest 30 countries (p. 5). The income gains for a low skill worker moving to the US vary depending on where they’re coming from, but range from 2.4x (Thailand) to 16x (Nigeria) (p. 4).
That’s pretty different than the paper you cite. I’m not sure what accounts for that right now. Hopefully we see more work in this area!
Yeah the discrepancy comes from assuming that immigrants in a category would earn the same as natives in that category. The first problem is that there’s substantial occupational downgrading; immigrants almost always work in lower-paid occupations than their pre-migration occupation. The second problem is that even within the same occupation, immigrants tend to have lower wages than natives (although they also have faster wage growth).
The Hendricks and Schoellman paper, in contrast, focuses on getting immigrants to the US to report their own wages before and after migration—so I think it’s a better reference on the wage gains from migration than comparing average wages.
But the pool of jobs is not fixed at all! Globalization show how easily is moving jobs to poor countries and how strong is resistance to immigration. Inmigrant communities often become traditionalists or end up in ghettoes in the receiving countries, while development in poor countries beguings a chain chain reaction of social emacipation.
What is the great welfare story of the late XXth century? Export oriented development. Mass inmigration is not so brigth… New technology allow for export oriented development in a substantial part of the services sector of the West.