How legitimate is it to use asking price information in the absence of transactions prices? And how does the gap between the two vary over the market cycle? This paper examines these two issues by comparing two large datasets from Ireland's property market over the volatile period 2001-2012. It finds that the two series are extremely closely correlated, both across space and across time, suggesting that in illiquid markets, or in the absence of transaction price datasets, asking prices offer a very good proxy. Nonetheless, a bid-ask spread exists at any given point in time. By exploiting information on the various stages of a housing transaction, it is possible to estimate the bid-ask spread in the housing market. That spread ranges from +4% at the height of the market in 2007 to -7% in 2010.