Home buyers and real estate agents agonize over the mortgage rate forecast and whether the Federal Reserve will cut interest rates. But the bigger story is not making headlines: Population growth is ending in the United States, which means not much need for more housing.
Housing demand and supply are critically local, so national trends seem less important. But a metropolitan area’s population growth is simply the national growth rate plus or minus a local variation. With the national population growth rate about zero, even fast-growing regions will see fewer people moving in.
Last year’s population growth came to about one percent or about 3.3 million people. (The data are estimated as of July 1 of each year, so the latest numbers are one year old now.) But most of the increase came from immigration net of out-migration. The U.S. natural increase—the excess of births over deaths—was just half a million people. The other 2.8 million people moved here from other countries.
Immigration began dropping in the months preceding President Trump’s second election. Now we have deportations and probably an increase in voluntary moves out of the country. If net immigration is currently zero, then our population growth is about 15 hundredths of one percent or roughly half a million people. We live with about 2.6 people per occupied housing unit, on average, so that half a million population increase implies a need for about 200,000 new housing units. In the last 12 months, we have built about 1.4 million units.
This top-level perspective understates the need for new housing, but only by a little. Some places are shrinking in population while elsewhere people are moving in from other parts of America. Abandoned farmhouses in Vermont don’t help Miami and Houston accommodate their growth. Adding up only the counties that are growing implies an additional 50,000 housing units needed beyond what total population growth would dictate.
Second homes amount to about two percent of the total housing stock, roughly 3.5 million units. With rising incomes, we could expect more, but we’ve actually seen the number decline. The drop has probably been in dilapidated “lake houses” while new second homes are built in resort areas. Still, the number of new second homes that will be built is probably small relative to the entire housing market.
Some of our existing housing stock will be demolished or destroyed by fire or other disasters. We don’t have hard data on this, but 100,000 units a year is a ballpark estimate based on past housing inventories relative to new construction and manufactured homes.
We also have smaller categories of housing, including college dormitories, retirement homes, liveaboard boats, etc. Still, it’s hard to argue that with zero immigration we need to build more than half a million new housing units.
However, there are certainly arguments to be made for more construction. For one thing, housing is expensive in many areas because of restrictions on development. The northeast and the west coast have laws that limit construction, pushing prices above free-market levels.
We also have some people living with roommates who would prefer to live alone. More significantly, many young families would like to have more room but cannot afford to move out of apartments or from a small house to a larger home.
Because housing demand and supply is so local, this national picture of the housing market can only serve as a starting point for people considering buying or building new housing. But the demographic trend and immigration policy are very strong factors. The birth and death rates are highly accurate, and President Trump’s immigration attitude probably won’t change during his term in office.
In years past, the country needed—and built—well over a million new housing units per year. That pace will not be normal in the coming years. If we do keep building, look for home prices to soften and apartment rents to fall. With lower prices, we’ll occupy the new housing units and live with fewer people per household, on average.