Added clarification: This post argues that freeing up Californian housing markets after 1970 would not have caused the California population to grow enough to swing the 2016 election.
This idea comes from Matthew Kahn at USC. In 2010, he published a paper showing that more liberal cities in California reduced the number of housing permits they issued, all other things being equal. That is to say, liberal antipathy to housing came from ideological preferences; it wasn’t a secondary effect of liberals being better-educated or concentrating in more geographically-constrained areas or whatever.
Kahn’s mechanism is that urbanites vote more Democratic, but zoning has constrained the number of urbanites. Without housing constraints, California would have 80 million people with the corresponding number of electoral votes.
But is 80 million likely? Brad Delong has an amusing back-of-the-envelop calculation for what California would look like with an unconstrained housing market. It assumes that Los Angeles is reasonably unregulated and produces a rough estimate of 8 million more people in the state, all in the Bay Area. That would give you 47 million Californians.
But his supply curve also assumes that Los Angeles is unconstrained save by its population size. Consider the below graph, from Greg Morrow’s (U-Calgary) dissertation. It shows the population that existing zoning would have allowed to live in Los Angeles if completely built-out:
The thing is, zoning restrictions certainly prevented the population of the Westside from doubling or tripling; it is less certain that it kept the population of the entire Southland from growing. Fewer people in Santa Monica, more in Riverside County.
There is another back-of-the-envelop calculation that favors a 47 million counterfactual. Growth restrictions became significant around 1970 and began to bite by 1980. Well, if California had grown at the same rate as Texas since 1970, it would have 48 million people today, not too far off the Delong counterfactual. (If it had grown at the same rate since 1980, California would have 45 million people.)
Now, to really figure out if this swung the electoral college you would have to guess at where the extra nine million people came from. Some might be from states that received Californian outmigration; others might be immigrants who would go to California rather than elsewhere. But we will ignore that for now.
Well, 37 million people got California 55 electoral votes in 2010; 45 million (remember, the 2010 population would be below 2016) would have gotten it 65-66 electoral votes.
No, not really enough to swing the election.
But if you assume that New York, New Jersey, and Massachusetts would also be much bigger, then you start getting somewhere ... and there is a completely different channel by which Jane Jacobs could be blamed for the outcome ...