Hot and Cold Seasons in the Housing Market

Every year during the second and third quarters (the "hot season") housing markets in the UK and the US experience systematic above-trend increases in both prices and transactions. During the fourth and first quarters (the "cold season"), house prices and transactions fall below trend. CEP Macro program members Rachel Ngai and Silvana Tenreyro propose a search-and-matching framework that sheds new light on the mechanisms governing these housing market fluctuations.

Whilst the existing literature on booms and busts in the housing market concerns infrequent but dramatic housing bubbles, Rachel and Silvana are the first to systematically document the existence and quantitative importance of these seasonal booms and busts. In the UK they find a 108 per cent difference in transaction volumes between the hot and cold seasons, and that the difference between annualized growth rates for nominal prices is above 8 per cent.

The size of these numbers confirms the importance of these seasonal fluctuations, and also presents a challenge for standard models of durable goods markets. The predictability of these prices changes - they are widely discussed by estate agents, and house price indexes are even seasonally adjusted - means that arbitrage opportunities should cause people to buy houses before the start of the hot season in anticipation of future higher prices, narrowing the price differential. More precisely, a typical no-arbitrage condition states that seasonality in prices must be accompanied by seasonality in rental flows or in the cost of housing services. Rents, however, display no seasonality, implying a substantial and, as we shall argue, unrealistic degree of seasonality in service costs.

To offer answers to these questions, they develop a search-and-matching model for the housing market. The model more realistically captures the process of buying and selling houses and it can more generally shed new light on the mechanisms governing housing market fluctuations. The model starts from the realistic premise that the utility potential buyers derive from a house is match specific; two individuals visiting the same house may attach a different value to it and hence have different willingness to pay. In that context, buyers are more likely to find a higher-quality match (and thus their willingness to pay is more likely to increase) when there are more houses for sale. Hence, in a thick market (or hot season), sellers can charge higher prices. Because prices are higher, potential sellers are more willing to sell their houses, better matches are formed, willingness to pay increases, and so on. This mechanism thus leads to a higher number of transactions and prices in the hot season.

Thus a matching model for the housing market is able to explain the coexistence of higher trading volumes and higher prices. Higher trading volumes encourage higher prices; higher prices encourage even higher trading volumes. This allows the possibility that relatively small exogenous differences in the desire to move house at different times of the year can create the much larger macro outcomes of hot and cold seasons. For example, the desire to move in the summer may be exogenously driven by the school calendar: people want to move before the start of the new school year. Rachel and Silvana document that without the "thick market" effects introduced by a matching model these exogenous differences are too small to create the observed seasonality in the housing market. However, by introducing matching they are able to quantitatively account of most of the fluctuations in transactions and prices in the UK and the US.

To read more about Rachel Ngai and Silvana Tenreyro's work on hot and cold seasons in the housing market see:

• "Hot and Cold Seasons in the Housing Market" [Full document in Adobe PDF] (L. Rachel Ngai, Silvana Tenreyro), CEP Discussion Paper 0922, April 2009