Ajay Rajadhyaksha is global chair of research at Barclays.
Jay Powell delivered a Christmas surprise last month, and it wasn’t the fun kind. However, peeved as investors might be at the Federal Reserve’s flip-flopping, the US economy faces a far bigger and imminent challenge that they mostly seem to be ignoring.
For all the new administration’s rhetoric about unshackling the US economy, the supply of workers is set to suddenly slow down, and the economic implications are . . . suboptimal.
We’re not referring to the president-elect’s plan to deport undocumented immigrants. Most estimates suggest that there are 11mn undocumented immigrants in the country. Despite campaign rhetoric, deporting all or most of them — even if the US tried to — is an unsolvable logistical challenge. ICE agents would have to go door to door, community to community, trying to find people who don’t want to be found.
State and local authorities would have to co-operate. Other countries would have to agree to take millions back. Congress would have to pass hundreds of billions in new funding for the effort. And businesses up and down the country would complain. For example, a very large percentage of the fruit-pickers in the country are undocumented migrants. Industries as diverse as food processing, construction, hospitality and others would be greatly impacted.
The new administration is therefore likely to ultimately soften its stance on deporting undocumented immigrants within the country (in practice if not in rhetoric) Deportations will rise, but likely to the same level as the Obama years, focused on those with criminal records, and should run at an annual rate of 400-500k/year.
Even so, the US still faces a looming labour supply shock, and it’s coming from south of the border.
Estimating net immigration into the country is not an easy process. The official source for native-born and immigrant workers — the BLS’ Household Survey — is (politely put) inaccurate and fluctuates wildly. So let’s go about it the long way. We defined five immigration categories — humanitarian immigrants (who don’t have a valid visa but claim asylum at the border), unlawful entries, overstays, green cards granted abroad, and temporary resident visa holders.
Then we had to collect gross immigration data on each of these categories from 14 different sources, such as the Department of State, encounters and releases from Customs and Border Protection, admissions data from DHS, etc etc.
Much time was spent trying to avoid double counting — such as for individuals who changed immigration status, or who enter and leave the country repeatedly. Then, to complicate things further, we needed to estimate emigration flows (people who permanently move out of the US) to arrive at net immigration flows.
Tl;dr from 2014 to 2019, net immigration flows every year came close to 1mn. They varied a little by year; 2018 saw only 700k net immigrants, but the numbers were always within a few hundred thousand people. In 2020 there was a net outflow of people mostly due to overseas residents going home during the pandemic.
Then came 2022. By our estimate, net immigration in 2022 was 3.3mn, rising to 4mn in 2023. As of the end of April 2024 (which is the latest data available), net immigration was shaping up to be another 3mn-plus year. So yes, the last few years have seen immigrants enter the country at several times the pace of 2014-19.
And with this massive immigration has come a giant increase in labour supply.
We estimated labour force participation rates in every category of immigrants (and yes, some guesses were made; that’s the nature of such an exercise). Immigration (legal, illegal, humanitarian — all of it) added just under half a million workers to the US labour force every year from 2014-19. But from 2022-24, we added over 2mn workers every year.
And of course, that is now set to change rapidly. In June, president Biden reinstated the “Remain in Mexico” policy by executive order; an order that he had rescinded in December 2021. In the past, this policy was successful in slowing down immigration. This time is proving no different, with new entrants already slowing down sharply in recent months.
It takes around six to nine months for new immigrants to officially hit the workforce. Which means that president Biden’s decision should show up in shrinking labour supply soon. We estimate this action alone will knock off 750k from labour supply next year. Moreover, when president-elect Trump takes over, he’s likely to put in place other measures to slow down immigration, as he has promised.
This means that US labour supply from immigration will rapidly go back to the averages of the Trump term — around half a million new workers a year. That’s a massive drop-off from the 2mn-plus of the last three years.
But so what? As long as demand for labour is strong, why does it matter where workers come from?
The problem is that the US is an ageing country. It’s not getting old as quickly as, say, Germany or Italy. But even in the US, the native-born population is growing below replacement rates. Moreover, this job expansion has gone on for so long that the US labour force participation rate (LFPR) is already very high by historical standards — especially in the critical 25-54 year cohort — and has been on the way down in recent months.
Throw in the absolute explosion in US household wealth since Covid-19 — which disincentives older people from staying in the workforce — and it’s clear that a sharp shrinkage in immigrant labour supply will bite. Strange as it sounds, the number of jobs created in an economy is a function of both demand for workers but also the supply of workers.
What would the country have done if 2022-24 immigration had averaged half a million instead of 2.2mn, and demand for workers stayed strong? It’s an interesting question. After all, the US enjoyed very healthy job growth in 2017-18 even when immigrant labour supply was just half a million a year.
But an important difference is that labour force participation had dropped in the second Obama term, and had room to rise considerably — as it did — in the first Trump term.
This time, with LFPR already high and the US continuing to age, wages would probably have had to rise very sharply, far more than actually happened. After all, the retiree sitting on the sidelines will only enter the job market is they are enticed into doing so. And even with higher wages, actual job creation would likely have been far less.
Meanwhile, high wages would have passed into inflation, pushing it even higher than it peaked this cycle. The Fed would have hiked rapidly, and kept going until companies’ demand for workers cooled off. And the business cycle would almost certainly not been as long and seemingly indestructible as it has been in the past couple of years.
Investors will have many things to focus on as the new administration takes over — including tariffs, the deregulatory agenda, tax cuts, a “strategic bitcoin reserve” and, um, the possible annexation of Greenland, Canada and the Panama Canal.
But the quickest impact on the economy might come from a shrinking supply of new workers.
Source link
