Has there been something in the water recently to make otherwise smart folks say odd things about the state of the housing market? Paul Krugman writes:

More bad housing news this morning. The question remains, how far is down?

Below are two pictures. The first is the CBO’s estimate of the housing-rent ratio — the ratio of home prices to an estimate of the rental rate on equivalent properties. It’s comparable to the PE ratio on stocks. What you see is that it fluctuated in a relatively narrow range until this decade, then took off for the wild blue yonder.

The question you need to ask is, is there any fundamental reason why this ratio shouldn’t go back to historical norms?

Don’t tell me about how some areas with high prices are especially desirable — that was true in the 90s, too. What we need is some macro reason to believe that things have changed.

His discussion throughout the linked post suffers a bit from jumping back and forth between the price-rent ratio and plain old prices, but let’s focus on the point above–has something changed? I certainly think so, and I imagine Paul would agree with me; after all, he spends quite a bit of time writing about how this decade is different, economically speaking, than the 90s, especially where inequality is concerned.

What if the returns to certain activities have sharply increased in recent years, and what if those activities require a presence in a highly desirable location with limited supply? Or, what if cities also function as consumption goods? In particular, what if a home in a desirable urban area is a luxury good? In that case, increases in income could have strong effects on demand for and consumption of housing. Now it also seems likely that prices overshot their marks, but that doesn’t mean that the gains from recent years must be entirely erased.

Krugman goes on to write:

Second, the average is misleading. As I argued more than two years ago, America is really two countries when it comes to home prices: in “flatland”, where new housing can easily be built, real prices never rose much; the big rises have come in the “zoned zone,” where land and permits are constrained — and those rises have been much larger than the average, so much so that they make no sense even given a maximum allowance for interest rates.

This is where the distinction between price and price-rent ratio gets important. Have rents in the zoned zone also increased substantially, or can we reasonably expect that they will (given the extent to which their zoned-ness limits supply)? If we look at individual metropolitan areas, we also see that propensity to zone is not uniform within cities. Just taking Washington, housing construction in the exurbs here strongly resembled “flatland” building practices while central city and inner suburb construction was more limited by regulation. During this downturn, prices have fallen most in the exurbs, while barely budging or continuing to increase in the center. Krugman appears to be saying that zoned zone prices will have to fall a long, long way; he may have to wait a while.

Why? I believe that central city housing, particular in certain, rich markets, is without good substitutes and is, therefore, pretty price inelastic. If you need to be near the economy or consumption options of Manhattan, or Washington, or the Loop, or San Francisco, then that’s where you need to be. Exurban office parks, by contrast, lack the economic and consumption specialization of those distinct center cities. Working, living, and shopping in Herndon isn’t all that different from same in Raleigh, or Riverside, or dozens of white collar office suburbs around the nation. Residents, then, are price sensitive in those places. When the bubble ends in Prince William County, Virginia, eliminating the prospect of further appreciation, then the prices there and in similar areas look incredibly unattractive relative to reasonable substitutes, and down, down, down go home values. Krugman seems to think that home prices in supply-limited areas should converge downward to those in “flatland.” I think the trend will continue to be divergence, big time.

This, in effect, is what Brad DeLong concludes:

Second, the zoned zone is not growing–and America’s population still is. The gap between heartland and coastal values is likely to grow over time, and that anticipated capital gain should push up prices now.

The only thing I’d change about that statement is that in many places the zoned zone is growing just as fast as it can. New York City and Washington can’t keep up with demand and so prices hold steady or increase. All the same, more and more people are willing to pay–in terms of money or space–to locate within those places.

What’s more, slow or no net population gain masks lots of population turnover. High real estate prices create a sorting effect, pushing out those with elastic demand for a place and pulling in those with inelastic demand. For places like New York and Washington, that provides a strong attraction to human capital rich workers and those strongly oriented toward consumption of luxury or specialty consumption and cultural goods. This sorting effect will enhance the divergence between some cities and others, and within those cities. (Note to all those relying on Case-Shiller data, not only do those indexes ignore condos and new construction, they also don’t distinguish between neighborhoods within an urban area.)

One final point. The zoned zone is different because it’s zoned. On this, I think we can all agree. It seems, however, that propensity to regulate increases with density, with time, and with things like congestion. In recent decades, population has moved massively to boomtowns in the SE and SW, cities which have added tens of thousands of housing units every year. I could be wrong, but I suspect that those increases are not sustainable. Rather, because those urban areas are not built to handle density, they may approach regulation-inducing saturation far faster than older, denser cities did. As an increasing share of urban America becomes zoned, we should expect supply growth to slow (unless it shifts to yet another crop of boom towns). Particularly in light of the prospects for future increases in energy and transportation costs, I suspect that housing costs will begin to absorb a steadily greater share of household income.

In other words, the great Housing Bubble of the Aughts probably acted to massively increase the nation’s stock of housing, thereby delaying the convergence of our housing prices with those in other developed nations. I would be surprised if we ever see another increase in housing supply to match the past five years–or anything even close.