An archetypical Washington, D.C. swamp denizen thought he caught me with my accuracy pants down the other day. Last Sunday’s post restated a point I’ve made repeatedly – that when countries let in too many immigrants, their economies tend to suffer lasting damage because businesses lose their incentives to improve their productivity – the best recipe for raising living standards on a sustainable, and not bubble-ized basis, as well as for boosting employment on net by fostering more business for most existing industries and enabling the creation of entirely new industries.
The reason mass immigration kneecaps productivity growth? Employers never need to respond to rising wages caused by labor shortages by buying labor-saving machinery and technology or otherwise boost their efficiency. Instead, they continue the much easier and cheaper approach of hiring workers whose pay remains meager because immigrants keep swelling the workforce.
It’s a point, as I’ve noted, strongly supported by economic theory and, more important, by evidence. But Todd Schulte, who heads a Washington, D.C.-based lobby group called Forward.us, wasn’t buying it. According to Schulte, whose organization was founded by tech companies like Facebook with strong vested interests in keeping U.S. wages low, “the decade of actual [U.S.] productivity increases came directly after the 1986 legalization AND 1990 legal immigration expansion!”
He continued on Twitter, “giving people legal status and… expanding legal immigration absolutely has not harmed productivity in the last few decades in the US.”
So I decided to dive deeper into the official U.S. data, and what I found was that although there are bigger gaps in the productivity numbers than I’d like to see, there’s (1) no evidence that high immigration levels following the 1986 amnesty granted by Washington to illegal immigrants and the resulting immigration increase mentioned by Schulte improved the national productivity picture over the pre-amnesty period; and (2) there’s lots of evidence that subsequent strong inflows of illegal immigrants (who Schulte and his bosses would like to see amnestied) have dragged big-time on productivity growth.
First, let’s examine the productivity of the pre-1986 amnesty decades, which provides the crucial context that Schulte’s claim overlooks.
According to U.S. Bureau of Labor Statistics figures, during the 1950s, a very low immigration decade (as shown by the chart below), labor productivity grew by an average of 2.63 percent annually. Significantly, this timespan includes two recessions, when productivity normally falls or grows unusually slowly.
Figure 1. Size and Share of the Foreign-Born Population in the United States, 1850-2019
During the 1960s expansion (i.e., a period with no recessions), when immigration levels were also low, the rate of labor productivity growth sped up to an annual average of 3.26 percent.
The 1970s were another low immigration decade, and average labor productivity growth sank to 1.87 percent. But as I and many other readers are old enough to remember, the 1970s were a terrible economic decade, plagued overall by stagflation. So it’s tough to connect its poor productivity performance with its immigration levels.
Now we come to the 1980s. Its expansion (and as known by RealityChek regulars, comparing economic performance during like periods in a business cycle produces the most valid results), lasted from December, 1982 to July, 1990, and saw average annual labor productivity growth bounce back to 2.24 percent.
As noted by Schulte, immigration policy changed dramatically in 1986, and as the above chart makes clear, the actual immigant population took off.
But did labor productivity growth take off, too? As that used car commercial would put it, “Not exactly.” From the expansion’s start in the first quarter of 1982 to the fourth quarter of 1986 (the amnesty bill became law in November), labor productivity growth totalled 10.96 percent. But from the first quarter of 1987 to the third quarter of 1990 (the expansion’s end), the total labor productivity increase had slowed – to 5.76 percent.
The 1980s are important for two other reasons as well. Nineteen eighty-seven is when the Bureau of Labor Statistics began collecting labor productivity data for many U.S. industries, and when it began tracking productivity according to a broader measure – total factor productivity, which tries to measure efficiency gains resulting from a wide range of inputs other than hours put in by workers.
There’s no labor productivity data kept for construction (an illegal immigrant-heavy sector whose poor productivity performance is admitted by the sector itself). But these figures do exist for another broad sector heavily reliant on illegals: accommodation and food services. And from 1987 to 1990 (only annual results are available), labor productivity in these businesses increased by a total of 3.45 percent – worse than the increase for the economy as a whole.
On the total factor productivity front, between 1987 and 1990 (again, quarterly numbers aren’t available), it rose by 1.23 percent for the entire economy, for the construction industry it fell by 1.37 percent, for the accommodation sector, it fell by 2.30 percent, and for food and drinking places, it increased by 2.26 percent. So only limited evidence here that amnesty and a bigger immigrant labor pool did much for U.S. productivity.
As Schulte pointed out, the 1990s, dominated by a long expansion, were a good productivity decade for the United States, with labor productivity reaching 2.58 percent average annual growth and total factor productivity rising by 10.87 percent overall. But when it comes to labor productivity, the nineties still fell short of the 1950s (even with its two recessions) and by a wider margin of the 1960s.
But did robust immigration help? Certainly not in terms of labor productivity. In accommodation and food services, it advanced by just 0.84 percent per year on average.
Nor as measured by total factor productivity. For construction, it actually dropped overall by 4.94 percent. And although it climbed in two other big illegal immigrant-using industries, the growth was slower than for the economy as a whole (7.17 percent for accommodation and 5.17 percent for restaurants and bars).
Following an eight month recession, the economy engineered another recovery at the end of 2001 that lasted until the end of 2007. This period was marked by such high legal and illegal immigration levels that the latter felt confident enough to stage large protests (which included their supporters in the legal immigrant and immigration activist communities) demanding a series of new rights and a reduction in U.S. immigration deportation and other control policies.
Average annual labor productivity during this expansion grew somewhat faster than during its 1990s predecessor – 2.69 percent. But annual average labor productivity growth for the accommodation and food services sectors slowed to 1.19 percent, overall total factor productivity growth fell to 1.19 percent, and average annual total factor productivity changes in accommodations, restaurants, and construcion dropped as well – to 6.36 percent, 2.67 percent, and -9.08 percent, respectively.
Needless to say, productivity grows or shrinks for many different reasons. But nothing in the data show that immigration has bolstered either form of productivity, especially when.pre- and post-amnesty results are compared. In fact, since the 1990s, the greater the total immigrant population, the more both kinds of productivity growth deteriorated for industries relying heavily on illegals. And all the available figures make clear that these sectors have been serious productivity laggards to begin with.
And don’t forget the abundant indirect evidence linking productivity trends to automation – specifically, all the examples I’ve cited in last Sunday’s post and elsewhere of illegal immigrant-reliant industries automating operations ever faster — and precisely to offset the pace-setting wage increases enjoyed by the lowest income workers at least partly because former President Trump’s restrictive policies curbed immigration inflows so effectively.
In other words, in the real world, changes in supply and demand profoundly affect prices and productivity levels – whatever hokum on the subject is concocted by special interest mouthpieces who work the Swamp World like Todd Schulte.
RealityChek 