It won't help me buy a house, but maybe it flips Texas?
California lost 182,083 people in 2020, though it remains the most populous state, with just under 39.5 million residents. The state will lose a congressional seat for the first time, based on slower growth reflected in the 2020 census.
State Department of Finance officials attributed the one-year loss on a declining birth rate, reductions in immigration and an increase in deaths because of the coronavirus, which killed 51,000 people last year. They also said that as pandemic-related deaths decline, and federal immigration policy changes, California is expected return to growth in 2021.
When I read this, my ridiculous first reaction was “Will this make houses cheaper?” Sadly, the answer is no — not for me, at least. Which is not to say that this is not nevertheless an important development. I’d like to think out loud about it.
As a soon-to-be Californian interested in working on state policy — especially housing — I need to get a better grip on the Golden State’s demographic trends and their political upshot. The unprecedented shrinkage of America’s largest state seems like a good occasion to do that.
First of all, isn’t it sort of ridiculous that the state doesn’t mention domestic out-migration due to high taxes and especially out-of-control housing costs? California’s net domestic in-migration has been negative for decades. It’s true that a drop in international immigration largely accounts for the population deficit. Indeed, international migration has propped up the population and concealed the fact that fewer American residents moved to California than left California every year for more than thirty years. California, in my opinion, has the best set of natural and cultural amenities of any state. People leave Buffalo because there are no jobs, and the weather is awful. But that’s not why people leave California. They leave because, as nice as it is, for most Americans it offers a pretty shitty deal on standard of living relative to other states. People like to be able to afford groceries after paying the rent.
Anyway, if not for international immigration, California’s population decline would be a hell of lot worse than 182,083 people. According to the Public Policy Institute of California, over the last ten years, 6.1 million Californians moved to other states while only 4.9 million people from other states moved to California. This 1.2 million deficit in net domestic in-migration implies a gain of about a million international immigrants over the same period.
That the foreign-born population increased by about the same amount as the loss from domestic emigration illustrates how selection in and out of California is reshaping its now-shrinking population. But it’s not just that the state’s now an even more of a multicultural polyglot immigrant gateway than it already was (despite a significant slowdown in immigration rates). There are also clear trends in the composition of the domestic populations moving in and out of the Golden State.
As PPIC shows, the domestic trend is pretty clearly a matter of affordability. Over the last decade, California continued to attract more relatively wealthy, well-educated Americans than it lost, but it lost many more poorer, less-educated Californians than it gained.
As you can see, non-college Californians are leaving California in droves, and the trend accelerated significantly through the Trump years.
People with more schooling make more money, so those coming in should be wealthier than those heading out. That’s exactly what we see:
If you read the fine print here, you’ll see that PPIC is working from numbers for a four-person household, so “lower income” here runs from zilch - $51,750, “middle income” runs from $51,750 - $138,000 and “higher income” runs from $138,000 - Silicon Valley Overlord.
Math says that average educational attainment in California has gone up. A North Dakota’s worth of non-college folks bugged out and a mere Escondido’s worth of college grads checked in.
But what about the immigrants who more or less filled the population gap left by this passel of less-lettered California expats? Well, immigrants over the past decade were more likely to have a college degree than native-born Americans. According to the Census:
An estimated 47.4% of the foreign-born population who arrived in the United States from 2010 to 2019 had a bachelor’s degree or higher, compared to 36.3% of native-born Americans and 31.5% of the foreign-born population who entered the country in or before 2009.
In California, about 35% of the population has a college degree or higher. Now, you might think that California’s proximity to Mexico and heavy reliance on Mexican labor in construction, agriculture, food service and hospitality would mean that the foreign-born population that the state gained over the past ten years is a bit less educated than the nation’s recent immigrant population taken as a whole. I thought this might be the case for about five minutes until I remembered that Mexican immigration went net negative around 2007 or 2008, thanks to the Great Recession.
I couldn’t find the exact number, but since about half of all Mexican immigrants in the U.S. live in California or Texas, the fact that more Mexicans have been going than coming nationally strongly suggests a net loss for California. Anyway, immigration to the United States from Asia overtook immigration from Latin America about ten years ago, but it happened even earlier in California. Every year for the last decade and change, the number of new Asian immigrants to California dwarfed the number of recent Latin American immigrants, as you can see here (thanks again to PPIC):
What’s more, the recent high level of educational attainment of immigrants to the U.S. reflects the extraordinarily high level of education among California’s Asian newcomers. As PPIC reports, “Half of foreign-born residents who have come to the state since 2010—and 61% of those who have come from Asia—have at least a bachelor’s degree.”
So, the picture here is pretty clear. A less educated cohort was effectively swapped out for a more educated foreign-born cohort — plus a smattering of well-off, well-credentialed Americans. And it is just a smattering relative to California’s total population. Eyeballing it, I’d guess California has averaged something in the neighborhood of 110,000-115,000 Asian immigrants each year since 2010. California netted 154,000 domestic college grads and 114,000 domestic “higher income” households over the same time. So… if 61% of Asian immigrants in that period had a college degree, then California gained more Asian immigrants with a degree every two years than it netted domestic in-migrants with a degree over the entire decade.
This suggests that California continues to look like a great deal for educated Filipinos, Vietnamese, Chinese, and Indians (why it does is an interesting question) but much less so for educated Americans. And the overall picture pretty clearly suggests that California has become an awful deal for working class Americans and Mexicans.
Suppose we extrapolate these trends into the future. Does all this differential self-selection in and out of the state have implications for California’s partisan mix? I went into this thinking it must, but now it’s not clear to me that it does!
Nationwide, Biden beat Trump among the college-educated by eight points, 55-43. But Trump beat Biden among voters who make more than $100,000 a year by 12 percentage points — 54 to 42. If we’re just looking at education, we should assume California gained more Democrats than Republicans. If we’re just looking at income, we should assume it gained more Republicans than Democrats. Voters who are both relatively wealthy and highly educated are “cross-pressured,” as the political scientists say.
That’s why it’s not obvious that California would become more rather than less Democratic if its population of well-schooled domestic in-migrants, like me, were to continue to (slowly) grow. California overall went for Biden 63-34. Even if we’re just looking at education, the population that California gained here is more Republican than the state as a whole — at least if it’s representative of college-educated Americans. Maybe it’s not.
A sizable majority of Americans who move across state lines in any given year are younger adults without kids, which would tilt the group in a more resolutely Democratic direction. And it wouldn’t surprise me at all if young college grads willing to move to California, despite the high cost, are more liberal than young college grads overall. Likewise, it wouldn’t be shocking to find that well-to-do, well-educated Californians who fled to Texas were a bit more conservative than the newcomers who offset their loss. So maybe the groups in which California had positive net domestic in-migration didn’t make the state very slightly more Republican.
But what about the much larger net loss of non-college Californians? This one’s tough, too. Trump once again dominated among white Americans, especially those without college degrees, and California’s clearly been getting less white, in both absolute and relative terms:
Trump walloped Biden among all white voters, 58-41. Insofar as California is getting less white, it is getting less Republican. Among, whites with no college degree, Trump crushed Biden, 67-32. We know that California is losing non-college population fast, but what proportion of this group is white? I’ll dig in later and try to find out, but for now I’ll just note that California’s Hispanic population (about 40%) is larger than its white population (about 37%) and that whites are somewhat more likely to have college degrees. Putting those two things together, I’d guess maybe 45% of California’s non-college émigrés were Hispanic.
One of the big stories of 2020 was that Trump picked up some steam with Hispanic voters in a bunch of places, but California’s wasn’t one of them. Biden won California’s Hispanic vote by a nasty 56-point margin — 77 to 21. That’s a six-point improvement on Clinton’s 2016 result and appreciably better than both Trump’s dominating performance among non-college whites and Biden’s overall California result.
So, even if we assume that the white portion of the non-college population that California lost is bigger than it is in the overall population — let’s say it’s 40% — the effect of this on the state’s partisan mix would be mostly if not totally offset by the net loss of non-college Hispanic population.
Again, we see that education and income influence party affiliation in opposite directions. In California, Biden won voters making less than $50,000 by 27 points — 63-36. And he won voters making between $50,000 and $100,000 by 24 points — 61-37. Those are basically PPIC’s lower- and middle-income groups. There’s a heck of a lot more Democrats than Republicans in them, but those margins are so close to the overall California margin, it’s hard to say whether it would make much of a difference for California’s partisan balance were the state to continue to lose population from these groups at whatever rate is implied by the extrapolated trendline.
I find it both a little boring and a little interesting that all these shifts in the composition of California’s population don’t seem to augur much change in the state’s current partisan balance. Even if we take into account the ongoing naturalization of California’s rapidly increasing Asian immigrant population, we still shouldn’t expect much to change. California’s Asian population favored Biden by a 32-point margin, 63-31, which is, again, remarkably close to the statewide margin.
However, the prospect of stability in California’s partisan mix shouldn’t be taken to suggest that all these changes in the composition of the population are politically insignificant. The non-college working class and the highly educated symbol manipulation class have different interests. The fact that the working class is getting priced out of the state while the (increasingly Asian) population of economically comfortable knowledge workers continues to climb is certain to change the complexion of intra-Democratic state politics (which I still don’t understand at all).
It’s worth mentioning that California’s persistently negative levels of net-domestic in-migration can make a difference to the partisan mix in states to which working- and middle-class Californians are fleeing, even if it doesn’t make a difference in California. So where are they going?
This is from 2007-2016, but I don’t think much has changed here. There are a lot of lower and middle-income folks among all these California expats settling in Texas and Arizona and no doubt plenty of them are Hispanic and Asian — all groups with a decisive Democratic lean. We ought to expect an influx of Californians from these groups to shift the partisan balance in states like Texas and Arizona in a Democratic direction. Check this out…
The biggest group moving to Texas from California has an income in the $15,000 - $30,000 range. And most California migrants to Arizona make less than $55,000. Don’t let the loudness of Republicans stomping out of the state in a scolding partisan huff confuse you. Joe Rogan and Ben Shapiro, it turns out, are not representative of California-to-Texas movers.
Interestingly, it seems that these domestic migration trends have only picked up steam during the pandemic. Compare net domestic migration (the blue line) in LA and Phoenix during the pandemic year, below…
Last year, LA racked up record net losses while Phoenix enjoyed record net gains. The Brookings demographer William Frey writes:
Increased 2019-20 out-migration has affected population shifts in several other [than New York and Los Angeles] major metros including San Francisco, Minneapolis-St. Paul, Columbus, Atlanta, San Jose and New Orleans. At the other extreme, several large metros, such as Phoenix, Austin, Houston, Nashville and Tampa, experienced increased 2019-20 growth due to their rise in in-migration. In these areas, domestic in-migration more than made up for downturns in natural increase and immigration from abroad. [emphasis added]
What do all those bolded cities have in common? They’re Democratic strongholds in states controlled by Republicans. In Phoenix, Austin, and Houston it’s pretty likely that more or their newcomers hail from California than anywhere else. Perhaps this has something to do with the fact that Republicans in Texas and Arizona are losing their shit and transparently scrambling to rig the next election by disenfranchising Democrats!
Anyway, it’s fun to see who’s leaving California for where, but that’s actually less important than the fact that so many young, educated Americans would have moved there if state housing and fiscal policy hadn’t made it so exorbitantly expensive. That’s ultimately the explanation for the rapid growth of Austin and Phoenix relative to Silicon Valley and LA, not Californians picking up and moving to Texas and Arizona. It’s sort of funny that terrible housing and tax policy in California could end up flipping Texas — a prospect I wrote about in this 2018 Times piece — but it really could. It’d be sweet if it did. At least some good would have come of it.
Unfortunately for me, however, more of my demographic continues to move into California than out, despite its godawful housing policy, which means that the state’s small drop in population won’t have any discernible effect on demand for houses that my demographic likes to buy. Democrat-majority Texas or cheaper California house: pick one.
[Correction: In the original draft, referring to the net in/out-migration heat map chart, I said, “Outgoing Californians who make over $200,000 evidently prefer New York and Chicago.” That’s incorrect. I was reading the chart wrong. Blue is net in-migration to California. Darker blue is higher net in-migration. What the chart actually shows is that more relatively wealthy people from Chicago and New York move to California than the other way around. As noted early in the post, more wealthy people continue to settle in California than leave it. The whole state is, in effect, gentrifying.]