Thursday, January 18, 2018

Housing: Part 279 - Building homes helps households with low incomes.

One of the many arguments against building homes in Closed Access cities is that market rate homes have very high rents and prices, and they are just bought up by rich outsiders.  This leads to complaints of "trickle down" economics.

Two things are happening here.
  1. Households react differently to housing deprivation depending on their incomes.  Households with high incomes can adjust their real housing consumption until their spending is at a relatively comfortable and normal level, by living in a smaller unit with a longer commute, etc.  As a household's income declines, they have less flexibility in this regard.  So, the adjustment households with the lowest incomes make is to move out of town.  New units means higher quantity and more moderate rents.  This means that the natural result of building new units in a housing deprived city will be that households with high incomes expand their housing consumption into the new units and households with low incomes make no change in their housing consumption.  This looks like failed "trickle down" economics.  But, the status quo was tens of thousands of lower income households moving away.  This is actually a vast improvement that favors households with lower incomes.  Not the sort of evidence that will convince naysayers, though.
  2. The other issue is that the Closed Access cities are so bad that the entire stock of housing is unaffordable for a normal family, unless it is subsidized.  So, new "market rate" units are treated by opponents as if they mean nothing to normal families.  Only subsidized units actually increase the housing that is available for the typical family.  This rhetorical treatment, which is technically true while missing the point entirely, leads to a paradigm where the market literally cannot fix the housing shortage.  There is no mechanism in this paradigm for markets to lead to an outcome that includes more housing units.  The possibility has been exorcised from the set of observable outcomes.
I decided that American Community Survey (ACS) data might help us in this regard.  Here I am using the largest 30 metro areas.  The Closed Access cities will be shown in red (here, including NYC, LA, Boston, San Francisco, San Diego, and San Jose).  These are for cumulative or annual numbers from 2005-2014.

First, as I have mentioned previously, we have a perverse migration pattern in this country where households move away from prosperity.  There is a sharp negative relationship between metro area incomes and net domestic migration.  This perversity is largely due to the Closed Access cities.  The city at the upper right, which has high incomes and high in-migration is Austin, TX.  Seattle is the city with incomes above $80,000 and in-migration of more than 0.5% annually.  That is how it is supposed to work.

How about we compare the incomes of in-migrants and out-migrants in each MSA.  All of these are expressed as a % of the MSA average income.

The incomes of migrants in general tend to be lower than average because movers tend to be young renters with lower incomes.  Interestingly, there is no systematic difference between cities of in-migrant incomes.  They tend to be about 85% of the MSA average regardless of housing policy.  This is somewhat surprising, since the Closed Access cities are the source of many complaints about wealthy outsiders buying up status units.  Some of that could be because absentee owners would not show up as migrants.  But, in terms of actual residents, there isn't much to see here.

Out-migrants on the other hand, skew poorer in cities with high rates of new out-migration.  This is entirely due to the Closed Access cities.  Here, also, if we remove the Closed Access cities, there is no relationship between net migration rates and migrant income.  But, in the set of cities with peculiarly low rates of housing starts, high average incomes, and significant out-migration, the out-migrants tend to have low incomes.

And, if we look at migration rates, by city size, rates of in-migration of Closed Access cities look fairly normal while rates of out-migration are high.  This surprised me a little bit.  I had thought that there would be more reduced in-migration in the Closed Access cities.  But, according to this data, it is the existing Closed Access households with lower incomes who bear the worst burden of housing deprivation.

I will look at another pair of graphs.  Here, I have made an adjustment for city size to estimate the unusual amount of in-migration or out-migration for each MSA.

Regarding in-migration, again, the Closed Access cities are quite normal, and again, across the board, there is no systematic relationship between relative incomes of in-migrants and the rate of in-migration.

Regarding out-migration, however, there is a strong relationship between the rate of out-migration and the relative incomes of the migrating households.  The Closed Access cities dominate the sorry end of the distribution, but they aren't outliers in terms of the relationship. It appears that wherever there are an unusual number of households moving away from a city, the movers from those cities tend to have low incomes.

By the way, whereas the Closed Access cities had normal in-migration rates for their sizes and high out-migration, Austin and Seattle have normal out-migration rates for their size and high in-migration rates.  Again, that is how it is supposed to work.

Outside the Closed Access cities, the in-migration patterns are basically what you would expect.  There is unusually low in-migration into cities like Cincinnati, Cleveland, Detroit, Pittsburgh, and St. Louis.  There is unusually high in-migration into Austin, Denver, Phoenix, and Riverside.

The rust belt cities also tend to have low levels of out-migration for their size, while the Closed Access cities have unusually high levels of out-migration.

Just for good measure, these last two graphs use housing permits per resident, which I have previously collected for 22 of these cities.  In the first graph, the scatterplot regresses outmigrant incomes against the rate of housing permits from 2005 to 2014.  The second scatterplot regresses the size-adjusted rate of outmigration against the rate of housing permits from 2005 to 2014.

By the way, in both cases, there is no relationship if you remove the Closed Access cities.  The trendlines are flat without the Closed Access cities.

This is real simple.  If you don't build enough houses, poor households will be systematically displaced.  If you do build houses, they won't be.  And, there are 5 major metropolitan areas that systematically displace their poorer residents because they don't allow enough housing units to be built.  They are outliers in this regard, and don't let anyone tell you that building houses won't solve the problem.

1 comment:

  1. My poms-poms are worn to a nub, but let me one more time say this is great blogging. Yes! Build more market-validated housing (which I think requires free markets and elimination of most zoning law).

    Here is an oddity, don't know what it means: I was reading through the latest Fed Beige Book, and there seems little difference in the tightness of labor markets as reported by the Fed:

    For example, Cleveland Fed reports: "Labor markets tightened, with wage pressures coming primarily from workers in low- and middle-skills jobs."

    Dallas: "Labor shortages persisted, with several reports that difficulty hiring was impeding growth to some extent."

    LA-SF: "Conditions in the labor market continued to tighten, with wage pressures increasing as contacts noted shortages of qualified labor in various sectors."

    BTW the unemployment rate in Indiana is 3.8%, cheap housing or not.

    At first blush, I would expect "impossible" labor shortages on the West Coast (as seen by Fedsters, anyway). I do not know how a business could get people to move the the West Coast.

    But job growth is slow in CA now. Maybe a balance is being reached. The Amazon story is worth telling. So there will be less housing inflation, and less job growth and wage growth, on the West Coast going forward. Housing and wages are high, but now reaching plateau.

    If true, that means the CPI will get undercut again.