Immigration, Housing, and Politics

Immigration tends to raise the cost of housing, benefiting native property owners but hurting renters and first-time home buyers. That’s the conclusion of a robust economic literature dating back many years. Unfortunately, campaign season has a way of turning previously unremarkable statements into bitter political disputes. So when candidates Donald Trump and JD Vance argued that immigration has caused housing costs to rise, the New York Times ran a tendentious attempt at rebuttal instead of a simple review of the evidence.

The Recent Literature

A foundational paper in the study of immigration and housing appeared in a 2007 issue of the Journal of Urban Economics. By examining changes in rents and housing prices in Metropolitan Statistical Areas (MSAs) as new immigrants moved in during the period from 1983 to 1997, author Albert Saiz concluded that a 1 percent increase in an MSA’s population due to immigration is associated with a 1 percent increase in rents. Effects on housing prices were larger but less precisely estimated.

A subsequent study by Mussa, Nwaogu, and Pozo in 2017 followed most of Saiz’s methods but introduced a key innovation by examining the impact on neighboring MSAs. Examining 2002 through 2012, the authors largely replicated Saiz’s estimated rent and price impacts on the MSAs where immigrants settled, but they found that the “spillover effects” — the impacts on rents and prices in neighboring MSAs — were actually larger. Specifically, the authors found that a 1 percent population increase due to immigration in an MSA was associated with a 0.8 percent increase in rents in that MSA, along with a 1.6 percent increase in surrounding MSAs. For house prices, the increase was again 0.8 percent in the target MSA but 9.6 percent in surrounding MSAs. The large spillover effects are likely due to native flight from high-immigration areas.

International research on immigration and housing generally comes to similar conclusions. For example, looking across 21 different countries from 2007 to 2014, a recent European study found evidence that the arrival of immigrants significantly increases housing costs.

Of course, immigration is just one of many factors affecting the cost of housing. During the vice presidential debate, Vance called illegal immigration “one of the most significant drivers of home prices”, but he mentioned inflationary fiscal policy and several other contributing factors as well. Similarly, when Trump gave prepared remarks at the Economic Club of New York, he blamed inflation and regulation for high housing costs before even mentioning the role of immigration. One could argue that the two candidates are still overstating the role of immigration, or still not paying enough attention to other factors, but the basic point that immigration raises the cost of housing should not be in dispute.

Enter the New York Times

If the New York Times had stuck to emphasizing that housing is a multi-faceted issue, with immigration being one cost-driver among many, that would have been a reasonable contribution. However, three Times reporters went much further in rejecting the Trump and Vance position. Their piece first dismisses expert opinion and empirical work as unclear or inconsequential, and then implies, contrary to the evidence, that immigration actually reduces housing costs. Their dismissal begins this way:

When Mr. Vance reiterated the claim [that immigration increases the cost of housing] during last week’s vice-presidential debate, for instance, he followed up with a social media thread citing three links.

The first was to a speech by a Federal Reserve governor that asserted immigrants “could” be putting upward pressure on the housing market, without citing any research. Another was not specifically related to housing shortages.

The final link was to a study that did find that recent immigration in any given city tends to come alongside an increase in rents and home prices, based on data from 2000 to 2012. But the author of that research, Susan Pozo, a Western Michigan University economics professor, said the way in which her results were used made her “kind of cringe.”

Please note how the Times reporters frame the discussion. Instead of taking an independent look at the evidence, they focus on rebutting three specific citations from Vance. Like defense attorneys advocating for a client, they imply that if Vance’s three citations are not sufficient to prosecute his case, then he must be wrong. Their readers are left unaware of the larger literature.

In any case, the reasons the reporters give for dismissing Vance’s three citations are weak. Start with the Federal Reserve governor’s official statement, which would normally be enough for the Times to establish some fact about the economy. Not this time, however. Similarly, the evidence described by the Times as “not specifically related to housing shortages” is actually a report from the Congressional Budget Office (CBO) — another source that the Times usually considers authoritative.

The CBO report notes that although the overall impact of immigration on inflation is small, “the greatest upward pressure on prices comes from increased demand for housing”. We are supposed to discount this statement, according to the Times, because it is part of a larger report that covers many different economic impacts of immigration beyond just housing. Readers can judge the quality of that reasoning for themselves.

The third piece of evidence that the Times dismisses is the Mussa, Nwaogu, and Pozo study discussed above. It is unclear how accurately the Times reporters summarized the comments from the study’s third author, but I suspect something was lost in translation, as their critique betrays a misunderstanding of the methods:

For one thing, Dr. Pozo and her co-authors found that places with growing immigrant populations tend to have rising rents — not that the immigrants are provably the reason for that rent growth. It could be that the newcomers were attracted to booming local economies that were already prone to faster rent increases.

The reporters are suggesting a mere correlation between housing prices and immigration, but the whole point of the study by Mussa, Nwaogu, and Pozo is to show causation. It is a carefully controlled longitudinal analysis that employs an instrumental-variable technique to help discourage the alternative interpretation that the Times reporters offer. One would think a coauthor of the study would have pointed that out to them.

The reporters go on to argue that the magnitudes of the effects found in the study are small relative to overall increases in housing prices over the past 10 years. Maybe, but the effects are surely significant on a year-to-year basis in particular regions. Remember, the authors found a 9.6 percent increase in housing prices in surrounding MSAs for every 1 percent population increase due to immigration.

Finally, the Times reporters discuss how immigrant labor is important for the housing industry, implying that immigration could actually lower housing costs for that reason. It’s true that the effect of immigration on housing is theoretically ambiguous because immigrants both demand housing and supply some labor for building new units, but that’s no reason to throw up our hands. Resolving theoretical uncertainties is exactly what empirical analysis is for. If immigrant labor is lowering the cost of housing faster than immigrant demand is increasing the cost, we would see that in the studies discussed here. We don’t. Instead, the effect of immigration is clearly to raise costs.

Conclusion

As we like to emphasize, the effects of immigration are neither entirely positive nor entirely negative. Instead, the issue confronts policymakers with trade-offs. In the housing context, the primary trade-off is the impact on property owners vs. the impact on renters and first-time buyers. The former group generally benefits from immigration, due to the increased demand for housing, while the latter group generally suffers, due to the price increases that follow the demand. The New York Times could have acknowledged the basic relationship between housing costs and immigration and then pointed out the trade-offs involved. Alas, it’s campaign season.