What will the New Year bring for housing markets? Prediction is a perilous business, but history and basic housing economics suggest that price changes will stay modest, and that construction will increase only slowly. The best that can be said about the current market is that it offers abundant affordability and that the broader economic recovery doesn’t depend on a big housing rebound.

The most recent S&P/Case-Shiller housing-price data show a second month of seasonally adjusted price declines. The overall 20-city index dropped about 3.4 percent from October 2010 to October 2011, after falling less than one percentage point during the previous year. In nominal terms, the 20-city index is at the lowest point since March 2003, and in real terms, we’re below where we were in October 2001.

When it comes to housing prices, it hasn’t just been a lost decade: Many metropolitan areas have had two lost decades. In nine out of the 19 Case-Shiller cities with data going back to 1991, real prices are lower today than they were 20 years ago. In four of those areas -- Atlanta, Cleveland, Detroit and Las Vegas -- real prices dropped more than 15 percent over that period.

The disparate nature of those four cities reminds us that it is possible to lose big on housing in rapidly growing cities, such as Atlanta and Las Vegas, and in declining cities, such as Cleveland and Detroit.

Price Gains

Prices always reflect the interaction of supply and demand, and whenever supply is high relative to demand, prices will be low. In Cleveland and Detroit, supply is high because the cities were built up during an earlier era, when the industrial Midwest seemed to have a rosier future. In Atlanta and Las Vegas, supply is abundant because there are few limits on new construction, and builders overestimated future demand. The cities that have experienced significant price gains are those that have either strong economies or pleasant amenities; either can boost demand, as can serious natural or regulatory restraints on housing supply.

The biggest winners over the past two decades are Denver, Boston, Portland and New York. The California cities would also look better if 1991 hadn’t followed their late 1980s boom.

History suggests that 2012 will see neither a big housing rebound nor a second crash. After the last housing collapse, which first bottomed out in April 1991, prices stayed almost perfectly flat for about six years. The Case-Shiller 10-city index was only 2.3 percent higher, in nominal terms, in April 1997 than it had been six years earlier, which meant that real prices had fallen by an additional 13 percent even after the first trough.

Home sellers are often loath to take nominal losses, and that means they can sit on their property for years, keeping prices from dropping further in a market that otherwise offers enough inventory to prevent any major price upswing. Looking ahead, price swings in many markets will be limited by housing supply. In places such as Atlanta, Dallas and Phoenix, there are armies of home builders who would be delighted to supply housing once prices tick upward. Elastic supply always limits price growth -- a market truth that should have deterred more home buyers in Las Vegas and Phoenix six years ago.

Yet housing supply also puts something of a lower limit on home prices, at least in growing areas. Even in today’s depressed labor market, housing can’t be built for free and much of America is still growing. The Census reports that the U.S. had more than 240,000 more housing units in the third quarter of 2011 than it did six months earlier, and about 1 million more units than 24 months earlier.

Anemic Growth

That qualifies as anemic growth by historical standards, but it is still more than nothing, and to have any growth at all, prices must stay high enough to cover construction costs.

Moreover, the next year should have slightly more construction activity than 2009 and 2010, because the number of new households has finally started growing. During the great housing boom from 2004 to 2006, builders completed about 1.9 million units annually, but the number of new households increased, on average, only by about 1.33 million each year.

The mismatch between supply and demand helps explain why the inventory of vacant homes rose by almost 3 million units from 2005 to 2008. New construction was always going to be slow, as the country worked its way through that housing glut, but during the recession, the rate of new housing formation also plummeted. As the number of young adults living with their parents increased, the number of new households fell below 400,000 per year, less than the number of housing units being built. Despite two years of building fewer than 800,000 homes a year, the overall number of vacant homes has barely fallen.

The single best fact of 2011 for the construction industry is that the rate of household formation finally began recovering. The Census reported an increase of more than 1 million households from 2010 to 2011, and if that continues, the U.S. will finally begin working through its excess housing inventory. Enough new households will ultimately create sufficient demand to bring back the construction industry, even if it won’t bring back boom-level prices.

Given that this housing decline is likely to be with us for a while, it makes sense to look for the bright sides. Most obviously, low prices mean increased affordability. Everyone enters the world “short” housing -- that is to say, everyone at some point needs a place to live -- and low home prices help consumers in the same way as low oil and food prices. For homeowners, it’s wise to think of housing as a hedge against increases in the price of a critical human need. Certainly, the value of homes dropped, but so has the cost of obtaining housing.

It is a popular mantra, even among some economists, that the economy cannot recover until the housing market does, but I can’t imagine how either economic theory or history justifies that connection. Housing prices remained at 1991-crash levels through most of the good years of the mid-1990s. There is a tendency to spend some housing wealth, but that wealth isn’t the only driver of consumer demand. The current drop in the unemployment rate wasn’t produced by rising housing prices, and a sustained recovery depends far more on the fate of banks in Frankfurt than on home values in Atlanta.

The biggest gift of the housing bust, however, should be wisdom. We should learn the folly of bribing Americans, through the home-mortgage interest deduction and the implicit subsidies offered by Freddie Mac and Fannie Mae, to borrow as much as possible to bet on the vagaries of housing prices.

We should recognize that big leveraged bets on housing carry big risks. We also should recognize the folly of believing real-estate professionals, and other interested parties, who tell us that investing in housing is a sure path to wealth.

Housing prices go down as well as up and they always have. You should never buy a house because you’ve been convinced that it is a great investment. In almost half of the Case-Shiller markets, prices didn’t even keep up with inflation over the past two decades.

You should buy a house only if you have found a home that gives you the space and neighbors you need to live a good life.

(Edward Glaeser, an economics professor at Harvard University, is a Bloomberg View columnist. He is the author of “Triumph of the City.” The opinions expressed are his own.)

To contact the writer of this article: Edward Glaeser at eglaeser@harvard.edu

To contact the editor responsible for this article: Max Berley at mberley@bloomberg.net