What Factors Drive A House Price Boom?

Interesting working paper from the IMF – “Price Expectations and the U.S. Housing Boom”. Essentially, expectation of future prices – unrelated to any fundamentals –  has had a significant role to play.

Between 1996 and 2006 the United States has experienced an unprecedented boom in house prices. There is no agreement on the ultimate cause for the boom. Explanations include a long period of low interest rates, declining credit standards, as well as shifts in the supply of houses and the demand for housing services. Several studies have, however, pointed out that it is difficult to explain the entire size of the boom with these factors and have offered speculation or “unrealistic expectations about future prices” as an alternative explanation. The empirical argument for an important role of house price expectations is often indirect: it is a residual that cannot be explained by a model and its observed fundamentals.

Instead of treating speculation as a deviation from a benchmark, the present paper aims to identify shifts in house price expectations directly and compare their importance to other explanations. To that purpose, we estimate a structural VAR model for the United States and use sign restrictions to identify house price expectation shocks. We then compare their effect to other shocks, including shocks to mortgage rates and shocks to the demand for housing services and the supply of houses.

Results indicate that house price expectation shocks are the most important driver of the recent U.S. housing price boom between 1996 to 2006, explaining about 30 percent of the increase. Over the entire sample, their contribution to fluctuations in housing prices in the U.S. has been smaller, accounting for about 20 percent of the long run forecast error variance of house prices. This suggests that the large contribution of price expectation shocks is historically exceptional. Regarding other shocks, mortgage rate shocks are the second most important driver of the boom: their contribution amounts to about 25 percent. Over the entire sample, they are the most important driver and account for almost 30 percent of the long run forecast error variance. Shocks to the demand for housing services and supply of houses play a subordinated role for fluctuations in house prices, both for the boom period and over the entire sample. Taken together, the four shocks explain about 70 percent of the house price boom, leaving a residual of about 30 percent. This indicates that attributing the entire residual that cannot be explained by standard shocks to price expectations will lead to an overestimation of their contribution. We also find that a model-based measure of house price expectations is positively correlated with leads of a survey based measure of house price expectations. The positive correlation with leads indicates that our measure contains similar information as a survey-based measure. In addition, it tends to provide the information more timely.

We find that the contribution of price expectation shocks to the U.S. housing boom in the 2000s has been substantial. In our baseline specification, price expectation shocks explain roughly 30% of the increase. Another 30% of the increase in house prices remains, however, unaccounted for by the four identified shocks. This indicates that attributing the entire residual that cannot be explained by standard shocks to price expectations will lead to an overestimation of their contribution. We also find that a model-based measure of house price expectations is strongly positively correlated with leads of a survey based measure of house price expectations. This indicates that our measure contains similar information as a survey-based measure, but tends to provide the information more timely. Our approach to identify price expectation shocks leaves the reason why expectations change open. When using an additional constraint to distinguish realistic from unrealistic price expectation shocks, we provide evidence that the housing boom was driven to an important extent by unrealistic price expectations. The analysis has focused on exogenous changes in expectations. An interesting topic for future empirical research is how expectations respond endogenously to other shocks.

Note: IMF Working Papers describe research in progress by the author(s) and are published to elicit comments and to encourage debate. The views expressed in IMF Working Papers are those of the author(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.

Author: Martin North

Martin North is the Principal of Digital Finance Analytics

Leave a Reply