27 June 2006

Garbage Statistics and Affordable Housing

Garbage Statistics

The world is full of meaningless statistics. Sometimes, even the more respected traditional media outlets produce them. The latest example, The Economist magazine, which published a cost of living survey that excludes housing from the cost of living.

It turns out that housing accounts for 29% of the most common cost of living index (Table 709), making it the single largest factor in an individual's cost of living. But, the even more important thing about the exclusion is that housing accounts for a much larger share of the variation in cost of living from place to place, particularly at the extremes.

Housing costs vary from 69% of the national average in Albany, Georgia, to 387% of the national average in Manhattan, and there is considerable variation from metropolitan area to metropolitan area.

Miscellaneous goods and services, in contrast, make up 33% of the normal cost of living index, but the variation is much smaller. Manhattan is still the most expensive by that measure, but miscellaneous goods and services cost only 139% of the national average there, while the cheaptest miscellaneous goods and services are found in Youngstown, Ohio, where they cost 91% of the national average.

This is not surprising. Goods can typically be shipped anywhere for only the cost of transportation, which are in the continental United States, at least, even with high gasoline prices, only a small part of the retail cost of most goods. The primary drivers of area specific good prices are retail store rental costs and the prevailing local retail employee wages, which are themselves, to some extent, housing cost driven. Lawyers in New York City make more money than lawyers in Cleveland, primarily because the New York City cost of living is higher, which is primarily because real estate in New York City is so expensive. This trend is true right down the line to the baristas on the first floor of those lawyer's office buildings, and the janitors who sweep the building's floors after hours.

Indeed, housing prices alone are a good proxy for cost of living generally, even though cost of living minus housing prices is almost meaningless.

Overall Denver's cost of living is about 3% above the national average, and if you really want to measure the "cost of living" the composite figure is really the only thing that matters. Knowing category specific prices may matter to anti-trust regulators wondering if a metropolitan area's gas stations are engaged in price fixing, or for people interested in identifying ways to reduce health care costs, for instance, so the outliers can be identified and studied.

But. a composite index that excludes housing is rubbish. Metropolitan Denver's own fairly high ranking on non-housing elements of the survey (it comes in 8th according to the Economist) may have as much to do with the Gallagher Amendment (which disproportionately allocates property taxes to businesses, who in turn pass that cost on, rather than residential property), as anything.

Affordable Housing

The revelation that housing prices are the main drivers of a locality's cost of living, and that these prices decidely local, does cast some important light on the issue of affordable housing.

The United States does not have a shortage of affordable housing. There are small towns on the Great Plains that will actually give away land to newcomers who agree to settle there. And, the abundance of cheap housing isn't limited to prairie ghost towns. Some friends of the family in Rochester, New York recently bought a large single family house for their growing family for less than the cost of a one bedroom condominium in Denver. What the United States has, instead, is a lot of localized affordable housing shortgages.

Cost of living statistics don't capture it, but in fact, housing prices tend to march in lock step. In San Francisco, everything from mansions to cottages have eye popping prices. In Pueblo, Colorado housing prices tend to be reasonable regardless of whether you are looking for four thousand square feet or four hundred.

Basically, housing prices shoot up because jobs in an area grow faster than the housing supply. Housing prices tend to lag in places where job growth has been stagnant relatively to the housing supply.

It happens very fast. In Grand Junction, Colorado, when the oil shale bust occured in 1983, real estate prices dropped virtually overnight. The jobs disappeared and the housing prices responded. In the same vein, right now on the Western Slope, the revival of interest in the areas high cost oil resources, as oil prices have increased, has made the area one of the most quickly appreciating real estate markets in the state.

There is some flexibility for people to get homes other than where they work, but the vast majority of people will live within a reasonable commute of where they work, with reasonable itself a function of just how expensive in terms of housing costs a few more minutes each day of commuting turns out to be for an employee. Where a long commute can save an employee a lot of money, say in Vail or San Francisco or Manhattan, people will do it. Where it won't, people try to live close to work.

This, in turn, points to one of the problems with addressing high housing prices directly. There are two ways you can do it. Stymie job growth or build housing, and the first one is a choice almost no community would willingly make. But, building lots of housing takes lots of time and money.

It also points to an important aspect of a likely solution. It isn't terribly important to make designated "low income housing" to reduce the demand for housing, which will impact the real estate market generally. Those who can afford to pay less are going to end up with the bottom of the barrel in the housing market, in any case. But, if you build lots of higher end homes, then existing housing residents will move into them leaving their existing homes for those who couldn't afford housing before, while if you build lots of low end homes, the more affluent will stay where they are and the less affluent will move into new properties. This has limits, but the basic solution to affordable housing problems, when affordable housing really is the problem, is to build more houses.

Sometimes, of course, the problem isn't really affordable housing at all. The problem is frequently underpaid workers. If employers are thinking about building housing for the purpose of letting employees pay a submarket rent or purchase price for the housing, then the employer is basically providing an off the books paycheck increase, and one has to wonder whether it wouldn't make more sense to keep employers in the pay check writing business and putting someone else in the landlord business. If teachers, firefighters and baristas can afford to live close enough to their jobs to afford to live there, maybe you just need to pay the more.

No comments: