Housing Bubble and Bust, Thy Name Is…Midland?


The so-old-it's-new Federal Housing Finance Agency has an interesting study [pdf] out of previous depreciations in real estate markets.

Expected finding: The rapid-drop pattern of the last three years is, as I noted the other day, historically unusual. Usually the pace of decline is much slower over a long duration (obviously, we don't know when or if current markets will bottom out). Real estate values can take the better part of a decade to find the bottom, and about that long to recover to their prior peaks.

Unexpected (to me) finding: Although California, in a walk, has already taken five of the top ten total-depreciation spots for the period studied, it turns out Midland, Texas still holds the record for longest period from peak to trough:

Home prices in Midland, TX—the worst [metropolitan statistical area] by duration of house price decline—lost over 56 percent of their value from the second quarter of 1982 to the fourth quarter of 2000 and have yet to recover 8¼ years later.

Here are the ten biggest losers in total depreciation:

  1. Merced, CA (lost 61.95% from peak in 2006Q1 through 2008Q4)
  2. Midland, TX (lost 56.15% from peak in 1982Q2 until trough in 2000Q4)
  3. Stockton, CA (lost 54.20% from peak in 2006Q1 through 2008Q4)
  4. Modesto, CA (lost 52.58% from peak in 2006Q1 through 2008Q4)
  5. Lafayette, LA (lost 52.50% from peak in 1982Q3 until trough in 1988Q4)
  6. Peoria, IL (lost 48.91% from peak in 1979Q4 until trough in 1985Q4)
  7. Salinas, CA (lost 47.50% from peak in 2006Q1 through 2008Q3)
  8. Davenport-Moline-Rock Island, IA-IL (lost 47.18% from peak in 1978Q4 until trough in 1989Q2)
  9. Cape Coral-Fort Myers, FL (lost 47.02% from peak in 2006Q1 through 2008Q4)
  10. Vallejo-Fairfield, CA (lost 45.53% from peak in 2006Q1 through 2008Q3)

How long will Midland hold its place in the real estate Hole of Shame? I don't see Merced picking up anytime soon, but could it really be 2024 before that jewel of the San Joaquin Valley hits bottom? Elsewhere in the piece, FHFA flirts with measuring booms and busts in nominal rather than real dollars, which tends to make the return-to-previous-peak come faster. (That's the only anti-inflation rhetoric you're likely to hear from the government for a long time.) But even if you use that trick, "the time it takes for an area to recover still tends to be longer than the time it takes for the same area to move from peak to trough. Correspondingly, annual depreciation rates in the downturn tend to be of larger magnitude that annual appreciation rates in recovery."

Midland shared in the statewide oil boom of the 1970s and the subsequent oil collapse of the 1980s, which put nearly a quarter-million Texans out of work. The same cannot be said for the biggest clowns of this decade:

[M]ost of the larger historical downturns were caused by sharp increases in unemployment rates and shocks to personal income. Although the U.S. economy has experienced such conditions in the last year, those factors were not among the precipitants of the latest downturn, which began in 2006, well before the financial crisis erupted in the third quarter of 2007 and the recession began in the fourth quarter of 2007.

I vaguely recall a Thor comic from the 1970s in which the god of thunder is given a tour of hell by a grim reaper figure who starts off by whispering, "There is so little to see….and so much time!"