Thursday, July 23, 2015

Housing Tax Policy, A Series: Part 45 - An observation on supply constraints and home prices

I have pointed out that real housing consumption has been in a long term decline.  In a post, today,  Political Calculations, points out how the rise in home prices as a function of incomes is coming from higher rents.  In other words, the source of the high cost of owning a home is the high cost of renting a home.

This is a little tricky.  What this means is that the value of some homes, over time, is rising due to a lack of supply.  For the household living in the house, the house value appears to remain the same.  But, since the real stock of homes is declining, relative to incomes, other households, on average, are slowly reducing their housing consumption.  Owning a home in this context is kind of like being in a lineup when they ask volunteers to step forward, and everyone else steps back.  Renters in market-priced housing notice this naturally, as their rent rises each year, and they will tend to adjust their housing consumption over time account for the changing values.  But rent-controlled tenants and home owners are insulated from this information.

If housing markets were perfectly liquid, we would sell off a few square feet of our homes each year in order to account for the stagnant aggregate supply.  But, since we can't do that, buying a home includes a pre-commitment to increase our nominal housing consumption each year that we live in the house.

This is even more tricky because most of these changing values are related to the value of location, so the change in values is somewhat invisible to us.  This is basically what is happening in places like San Francisco's Mission District.  The high demand for San Francisco housing, along with the long term constraints on building, has meant that households who moved there years ago when it was not so in demand have been passively increasing their housing consumption continually over the past few decades so that now middle and low income households are occupying housing that is more fitting for very high income households.  This causes all sorts of consternation, because part of the value of those homes came from the flavor of the neighborhood itself.  This is real value, and it is value with significant sentimental and emotional character.  The dislocations caused by this change are frustrating.

So, even though location is the main source of changing home values, it is easiest to think about this in terms of size.  What has happened to those households in the Mission District is conceptually similar to having added 100 square feet to their homes every year for the past few decades.  We sympathize with the tenants, because they didn't ask to have the equivalent of 100 square feet added to their homes each year.  It just happened to them.  But the fact remains that they are over-consuming, and there is no getting around that.

As sad as those changes are, dislocations caused by changes in the world around us are just as real as changes that we instigate.  The streets of San Francisco used to be filled with horse drawn carriages.  The world changed.  If there was a group of Amish people living in a neighborhood in San Francisco who still wanted to be able to drive their horses downtown to shop, we wouldn't be able to accommodate them.  San Francisco isn't built for horses any more.  San Francisco changed.  At some point there probably was a bitter fight between the remaining horse and buggy households and the rest of the city.  And the horse advocates would have had sentiment firmly on their side - rightly so.  But, their loss was, sadly, inevitable.

The plight of the families in the Mission District is especially noticeable because city policies have pushed much of the added cost of their housing consumption onto their landlords so that the disconnect between their actual housing consumption and their perceived housing consumption has grown to outrageous proportions.  But, really this is the dilemma that homeowners face when location value is changing.  Homeowners don't have rent control laws, but they simply mentally benchmark the nominal rental value of their homes to the values they had when they purchased them.  The added rental value of a home in a supply constrained location isn't as obvious as that hypothetically added 100 square foot room, so there becomes a growing disconnect between the value of a home to the owner-occupier and its market value.

But, it gets even trickier here.  For a household in a home with a stable rental value, there is an endowment effect to owning.  The home gains value to the owner simply as a function of the time spent living there and the place that home has in a family's history.  For a household that owns a home with stable rental value, the market value of the home remains stable, but the household keeps gaining more consumer surplus over time as the endowment effect of living in the home accumulates.  We see this very clearly in high profile eminent domain cases where homeowners fight to keep their homes even when the offering price is well above the market price.

So, when a household buys a home in a market with constrained supply, the added rental value of the home, which they are consuming over time, is simply coming out of the consumer surplus that would be building up as a result of that endowment effect.  For most families, if they don't engage in a transaction with their home, there is really no way to notice the difference between living in a home where rising imputed rent is capturing the consumer surplus and one where it isn't.

The problem the families in the Mission District are having is that market values have outpaced the endowment effect.  And the emotional dissonance is amplified by the fact that the consumer surplus was being claimed by the tenants but the market gains will go to the landlords.

Anyway, back to my original point.  Where this makes housing markets difficult to intuit, and where it makes reasonable home prices look like a "bubble" is that the effect of all of these complex conceptual issues is to raise the prices of houses today in markets where supply constraints are expected to remain strong.  In order to buy a home in, say, San Francisco, a household has to pre-commit to paying a rising imputed rent to reflect ever increasing real home consumption that the family doesn't intend to actually initiate, and that they won't notice when it happens.  The household is pre-committing to taking less of the endowment effect of living in the home as consumer surplus, and when they purchase the house at today's market price, they are transferring much of that consumer surplus to the home seller as a part of the market price today.

It seems like home prices in cities with rising rents are disconnected from reality, but what we are really seeing is a transfer of expected future consumer surplus from the buyer to the seller.  Is that confusing enough for you?

To me this is an example of the amazing ability of market values to convey information.  To own a home in a supply constrained city, you have to accept that there will be dislocations in your housing consumption that will be caused by the city's inability (or unwillingness) to accommodate changing housing demand.  This will change the nature of your home even while your housing needs remain the same.  These changes cause the expected value your home will have to potential other tenants to be higher.  Since markets have such incredible power to accommodate these values even though the individual buyers and sellers have no understanding of the complexity of the issue, we end up, amazingly, bidding up the prices of homes to their correct values.

The problem arises because politics only incorporates our conscious understanding of what we are transacting, and we impose political impositions against the very transactions that we freely engage in.  Thus, we have the strange presence of people who purchased homes in the 2005-2007 period and who simultaneously hold the view that there was a housing bubble created by predatory banks and out of control speculators.  This is an understandable position for them to have.  There is absolutely no way that we could expect ourselves to understand even a remotely small part of the inputs into the market price of any given good.  The gap between the ability of the market to accommodate that information and our own understanding is cosmic.  It's miraculous.  The problem lies in the fact that we have such effective outlets for the imposition of our conscious understanding on those markets.  The problem isn't that those people are exhibiting some sort of hypocrisy.  To expect any more from any of us would be ludicrous.  The problem is that they vote and that their votes actually change things.

So we have voted ourselves a supply problem and then we voted ourselves a demand problem.  We look at these amazing networks of coordination, and where the outcomes don't always align with our sentiments (why in the world would they), we accuse, "See that!  Capitalism did that."  The Pope calls it the "dung of the devil".  And we cheer.  My point isn't that markets achieve some sort of moral and distributive perfection.  The point is that we are in no position to judge.  As Mencken said, "Explanations exist; they have existed for all time; there is always a well-known solution to every human problem — neat, plausible, and wrong." and "The fact that I have no remedy for the sorrows of the world is no reason for my accepting yours. It simply supports the strong possibility that yours is a fake."  These statements are horribly true.  And I have no solution for the problems they do not address.  What kind of monster would I be to suggest that we replace some of our non-solutions with a lack of solutions.  As a compromise, I try only to advocate against our non-solutions when they are causing clear and identifiable harm.

Scott Alexander had a great post where he pointed out the problem of diagnosis in the face of uncertainty.  If intervention is called for 10% of the time, and we have a 90% success rate of intervening correctly when we need to, we will intervene twice as much as we should.
10% need intervention x 90% correct diagnosis = 9% correct interventions.
90% intervention not needed x 10% incorrect diagnosis = 9% incorrect interventions
We see 1% of the unaddressed problems, which can be awful and are naturally the hardest problems to address.  And we naturally want to demand more interventions.  But, this is the libertarian problem.  Those Mencken statements seem crass and fatalistic.  But, we are optimistic.  We are saying, good for us!  We have an amazing 90% success rate in diagnosing the worst externalities and problems of these amazing markets that coordinate far beyond our ability to understand what we are doing.  This is much better than we could possibly expect of ourselves.  Now, given this outstanding success rate, it follows that for every new intervention that we have missed, there are 9 interventions that we have mistakenly imposed.

So, please understand, an anti-interventionist approach may be predicated on an extreme, even indefensible, optimism about our ability to intervene effectively.

PS.  How bad could those incorrect interventions be?  Just think of whatever policy the political party you don't favor is pushing in the next election.  Human sacrifice throughout history was frequently considered a good idea.  We can pretty easily get into Godwin's Law territory here.  Interventions can very easily escape moral mean reversion.  If only we had a tradition for worrying about interventionist speculative bubbles.

2 comments:

  1. I don't have anything to add; I just want to say that I always enjoy your posts that you tag as "rambling"!

    ReplyDelete
    Replies
    1. Ha! Thanks. That tag is sort of my way of saying to myself, "Watch yourself, smarty pants." Thanks for the feedback.

      Delete