Housing Boom and Bust in the United States: 1921-1932

“Thus, while the federal response to mortgage distress during the Great Depression provides insights about how the government might respond to the current wave of defaults, the very different conditions underlying mortgage distress during the two periods warns against drawing strong conclusions from the historical experience for the current episode.” D.C. Wheelock, ‘The Federal Response to Home Mortgage Distress: Lessons fron the Great Depression’, The Federal Reserve Bank of St. Louis Review, May/June 2008

The Obama administration has mirrored much of its response to the current home mortgage crisis in the United States on the behavior of the FDR administration to mortgage distress from 1932 onwards during the  extended Great Depression.  As this column and its follow-up tomorrow  will demonstrate, the two crises are sufficiently different as to  indicate that Obama’s  policy emulation of FDR  is dangerously incorrect.

Statistics on single-family housing starts  and the value of new units over the period 1921 to 1932 indicate a boom and bust cycle similar to that of the period 2001-2010.  Housing construction had stalled during World War I at around 100,000 new units per annum with an aggregate market value of $700 million.  Between 1918 and 1920, these numbers approximately doubled, only to fall back somewhat during the 1920-21 recession.  Thereafter, construction rebounded rapidly as the economy recovered. Single-family housing peaked in 1925 at 600,000 units per annum with an aggregate market value of $5 billion.  Real estate speculation was widespread, fueled by low interest rates,  lax lending standards, and the ease with which securities could be sold to finance construction.  In all these respects the pattern does mirror that of 2001-2006, although the magnitudes involved were markedly smaller. 

In 1926, housing starts began to decline, falling to 500,000 units in 1928, and the  aggregate market value of such new construction also fell, to $4 billion.  Interest rates began to rise in 1928 as the Federal Reserve tightened monetary policy to stem speculation, especially in the stock market. Housing starts collapsed to a 1918  trough of 100,000 per annum in 1932, with aggregate market value also collapsing to $700 million, also mirroring the 1918  trough level.  Only in 1941, and the entry of the United States into World War II, would the level of new housing starts and the aggregate market value of such new construction return to the peaks  achieved in 1925. 

However, there is one enormous difference between this 1929-32 experience and that of 2006-10  in the United States. Over the period 1929-1932, the decline in house prices was roughly comparable to the decline in consumer prices and in nominal incomes.  As a consequence, inflation-adjusted house prices changed hardly at all.  This experience differs sharply from that over the period 2006-2010, where consumer prices have slowly increased, and where nominal income is  now above the level of 2006, albeit still running perhaps 10 per cent below trend (W. Woolsey). This difference implies  a significant drop in real house prices.

The implications for the burden of residential mortgage debt as a per cent of residential wealth are dramatic.  In 1926 this burden was 23 per cent. In 1932  it had climbed to 35 per cent.  With mortgage debt rising in real terms as prices and money incomes fell, the burden of mortgage repayments became intolerable for many households. Although the nominal value of mortgage debt peaked in 1930 and then declined, deflation caused the real value of outstanding mortgage debt to continue to rise until 1932.  Consistent with Irving Fisher’s (1933) classic “debt-deflation” theory, the burden of outstanding mortgage debt increased sharply during the contraction phase of the Great Depression.  Economic recovery did not begin until the real value of such outstanding debt had begun to decline. 

Such is not at all the case over the period 2006-2010, where the real burden of outstanding mortgage debt has continuously declined as a consequence of a gently rising consumer price index (CPI) , even though aggregate  nominal income is running below trend.   The implications of this startling difference for relevant policy interventions in the US housing market will be explored in tomorrow’s column.

Tags: 1921-1932, 2001-2010, boom and bust in the housing market, the burden of mortgage debt, the Fisher debt-deflation theory