Does Housing Have a Demographic Destiny?
Thank you for reading The Sunday Morning Post! I appreciate the support I received following last week’s announcements about the future of the newsletter, including the generosity of you all, the readers. If you missed that article and want to read about what’s in store for the year ahead and how you can support The Sunday Morning Post financially, you can find it here. Thank you to everyone who became a paid supporter last week. I appreciate it very much!
I have written a lot over the past 3+ years about how certain variables that are somewhat unique to the current housing market are acting as major levers on buyer demand and seller activity. High interest rates due to stubbornly high inflation, low inventory due to the lock-in effect and a period of under-building following the Great Recession, and COVID-era movements of people around the country including the entrenchment of the work-from-home era: there has been a lot at play.
But what about more systematic, long-term factors? If demographics are destiny, what do changing birth rates, immigration, and household formation trends indicate about the future of housing? There could be big changes in the years to come thanks to these long-term variables. At the very least, the housing market of the 2030s may look a lot different than the housing market of the 2020s. People in real estate including builders and investors should be mindful.
Household Formation
I wrote about household formation stats in January 2023, particularly with regard to how COVID-era factors like diverted spending, stimulus benefits, and inward-facing behavior by consumers plus other lifestyle adjustments helped lead to a large increase in the number of new households in 2021 and 2022. This, in turn, contributed to a housing crunch: more households meant more demand, which contributed to rising prices not to mention a general frenzy of homebuyers and renters looking for places to live. In fact, from 2020 to 2022, there was a net increase of 2.7 million households in just two years. This helped contribute to rapidly rising home prices and surging rents.
But I also wrote in that article that this trend in household formation would ease off, saying:
While there might not be a major drop-off in the number of households, it is also not likely that the trend over the last 2+ years during which time the U.S. has seen an additional 2.7 million households created will continue either…the same factors that created a surge in new households are largely reversing. Interest rates are much higher, which makes it harder to buy a home. The passing of COVID-19 combined with stimulus fatigue and a split Congress make any additional fiscal stimulus very unlikely. And people are starting to feel less economically secure due to inflation and the potential of a looming recession as business activity starts to slow down.
This has mostly come to pass. Consider this: 2023 saw a net increase of just 232,000 households nationwide, an increase that was the smallest on a percentage basis in the last 80+ years! (There was actually one single year since 1940 that saw a net decline in new households formed: 2020, which was a strange year, for sure).
The slowing of household formation will have a big impact on housing, and, in fact, it probably already has. Fewer new households means less demand for new and existing home purchases. It also means fewer tenants seeking out apartments or other rental units. Less housing activity also means fewer ripple effects in the overall economy from all of the housing-related expenses like repairs and renovations, materials and supplies, and all the miscellaneous one-off costs that come with moving between apartments or buying a new home.
Why the Slowdown?
No doubt the housing crunch itself is a major contributor to household contraction. With rents up as much as they have been over the past few years, people who may have moved out to live on their own during the pandemic are now perhaps re-consolidating with roommates, which results in a loss in the number of households.
Inflation is likely a key factor too. It has just been expensive to live in general, which has probably led to the same effect of people living together in order to save money.
There are broader demographic factors at play, too. Birth rates are slowing around the world, including here in the United States. I won’t dwell on the reasons too much, other than to say we have been in the midst of a long-term change from an agrarian economy of the 19th century to an industrial economy of the 20th century to now an information/technological economy of the 21st century. The changing needs of society (and of families specifically) have led to fewer births. Economic affluence tends to lead to lower birth rates and the world is much wealthier today than it was a century ago. Changing views on family structure and the ubiquitousness of birth control are perhaps the largest factors here, too.
Is there any reason to suspect that birth rates would reverse course and start increasing? Not really. Absent some unexpected cultural shift where people start having more kids, the U.S. population will really only grow in a meaningful way through net migration into the country. Yet contrary to what people might see or feel by watching the news, net migration to the United States is actually on the decline.
Consider the chart below: after peaking in the late 1990s, net migration has steadily dropped on a percentage basis for much of the last 20 years. The spike in the chart below is from 1998, where net migration to the United States peaked at just over 6 people per 1,000 of population. As of 2023, the number had dropped to 2.748 people per 1,000. In fact, in each of the last three years, net migration has dropped by almost exactly 1.3% each year. The line extending to the right of the vertical dotted red line represents predictions for net migration in the years ahead, which shows, in general, a general steady decline in net migration at least on a percentage/per capita basis in the years to come:
If not through birth rates and migration, how else might there be an increase in new households? The other primary way would be from people simply deciding not to live together, thereby splitting up the current population of people into more separate parts. But much as it may come as a surprise to people that net migration is easing off, what may be even more surprising is that divorce rates are also on the decline. In fact, divorce rates are at nearly 40-year lows! After peaking in the last 1970s at just over 22 divorces per year per 1,000 married women, by 2022 (the most recent year for which data is available) divorce rates had dropped to nearly 14 divorces per 1,000 married women. That is a fairly significant statistical decline. I think the average person out there would say that divorce rates are at all-time highs, but it is simply not the case (perhaps a topic for a future article).
The Extreme Case: Japan
Declining birth rates, less net migration, lower divorce rates, and an overall decline in new household formation: what could it all mean for the economy as a whole and, specifically, the housing market? There are examples out there where this is already happening and where the trends are perhaps 10-20 years ahead of the United States.
Consider Japan. Japanese birth rates are actually higher than in the United States: 7.0 births per 1,000 people in 2023 as compared to 2.75 births per 1,000 people here in the United States. But the rate of decline in birth rates in Japan has been more extreme than in the United States, dropping from over 25 births per 1,000 people in the 1950s to what it is today. The chart below shows what a plummet there has been in birth rates in Japan over the past few decades:
Japan is experiencing what some economists refer to as “demographic winter.” Whereas here in the United States the population is still growing (typically by about 0.5% on average per year), in Japan the population has declined every single year since 2010. The population of Japan was just over 128 million people in 2010, but the current population is approximately 122 million. That is a staggering decline for such an industrialized nation, one that is the second largest economy in Asia.
As if that weren’t bleak, the number of marriages in Japan is also on the decline: 2023 saw 489,281 new marriages in Japan, which was the first time in over 90 years that there were fewer than 500,000 marriages in the country. Fewer marriages today means fewer new births in the years to come, compounding the demographic woes. Per Reuters, “Japan's population will likely decline by about 30% to 87 million by 2070, with four out of every 10 people aged 65 or older.” The Prime Minister has called the declining birth rate, “The gravest crisis our country faces.”
An economy with its population overweighted towards its elderly is problematic for a variety of reasons. This is a relatable problem in many parts of our own country, including right here in Maine from where I write. Maine is the oldest state in the country with a median age approaching 45 years. This presents challenges including a lack of workforce, rising healthcare costs, a changing tax base, and just the logistics of caring for an aging population.
But on housing specifically, there is another interesting factor at play in Japan, traces of which could be come evident in the United States in the years ahead: vacant homes. According to one recent report, there are over 9 million vacant homes in Japan, which represents about 14% of the overall housing stock. Many of these homes are stuck in legal limbo with no heirs to pass them on to. Others are no longer livable after years of neglect and neglected maintenance. For some of these homes, there is just simply no demand for them as they are located in remote, dying communities that are cut off from services like hospitals and grocery stores.
While there are plenty of videos on YouTube documenting stories of foreigners including Americans scooping up cheap Japanese homes, officials and advisors are quick to point out that the process is actually fairly complicated and roadblocks abound. Still, though, it is wild to think that in a time here in the United States marked by such housing pressures, that there are millions of homes in Japan (and elsewhere, too) sitting vacant. Then again, drive to almost any rural area in the United States, and it won’t take long to find vacant homes in various states of disrepair that you just wonder if the right set of people came together could be a key part of easing our housing woes.
What Comes Next
With declining birth rates, less immigration, fewer divorces, and perhaps (and inevitably) an economic drop at some point in the years ahead, there could be a systematic easing up of pressures in the housing market. Factor in steady rates of construction of new homes and rental units (although construction is off a bit in 2024 due in large part to high interest rates), and it would not be surprising if the housing frenzy of 2020-2023 looks more like a historical blip or perhaps a last gasp of irrationality before demographic destiny takes its hold and the whole system becomes depressurized. Time will tell.
Ben Sprague lives and works in Bangor, Maine as a Senior V.P./Commercial Lending Officer for Damariscotta-based First National Bank. He previously worked as an investment advisor and graduated from Harvard University in 2006. Ben can be reached at ben.sprague@thefirst.com or bsprague1@gmail.com.
Weekly Round-Up
Here are a few things that caught my eye this week!
I frequently share Federal Reserve Economic Data, also known as FRED. FRED is published by the St. Louis Federal Reserve. It is a treasure trove of information and charts for data nerds like me. In fact, I would estimate that I have referenced FRED data or shared a FRED chart in at least 75% of The Sunday Morning Post articles. Anyway, I happened upon a snippet this week in which FRED made known that it was the creation of a woman, Lora Holman, and several of her female colleagues. Thank you to the woman (and men) of FRED, who have made their resources available.
According to Census data as analyzed by The Washington Post, the three most “typical” states based on race, religion, age, and other factors are Illinois, Florida, and Pennsylvania.
Bucking trends in several neighboring coastal communities here in Maine, the voters of Mount Desert (not to be confused with Mount Desert Island as a whole) rejected limits on short-term rentals this past week. Other towns like Bar Harbor have enacted strict rules and restrictions on their short-term rentals, but in Mount Desert those who attended a recent town meeting seemed swayed by arguments around private property rights and in support of the ability for long-time residents to earn some side income from their properties, especially if they spend part of the time elsewhere. Read more via Shaun Farrar in The Bangor Daily News.
Happy Mother’s Day
I would like to wish all the moms out there a very happy Mothers Day. I am very fortunate to have a wonderful mother of my own and also a loving wife who is a phenomenal mom to our kids. I love you both.
A couple years old at this point, but one of my favorite pictures of my wife and two of our three kids. © Ben Sprague Have a great week, everybody!
Speaking to the net migration impact, it would be useful to know the absolute numbers over years since by definition the denominator effect dampens the % as the population increases over time.
I read somewhere that 2023 had the largest increase in population - almost 3.3 million. How do we connect that information to your theses?
Great article. I appreciate the facts and the data, especially re rate of immigration...