Which house price data can researchers use? Ideally, one would use sales prices: the prices at which houses were actually sold. In practice, however, this data isn’t always available, or it only becomes available with a delay. Therefore, list prices or assessed values (such as the official tax assessments like WOZ values in the Netherlands) are often used. These seem like convenient alternatives, but they do not always measure the same thing.
Sales prices provide the most reliable picture of the housing market. They show what buyers and sellers agree on and the price at which a house actually changes ownership. List prices reflect the prices sellers hope to receive. These can be higher or lower than the final sale price, especially in times of rapid market condition change. Assessed values often lag even further behind. They are usually based on past transaction prices, which means they reflect market changes with a delay.
In a recent working paper, we investigate how these different price measures relate to each other. We use detailed data from the Netherlands and New York City. We find that the various price measures often give a similar picture at first glance, but this changes once we look at price developments over time.
This difference matters. When researchers use list prices or assessed values, price increases or decreases may be measured too early, too late, or too weakly. This affects how we understand housing market fluctuations and how we evaluate the impact of policy. Therefore, whenever possible, sales prices are the best choice. If sales prices are unavailable, list prices can be a good alternative, provided the housing market is in a relatively stable period and not characterized by large fluctuations.
Tu Giang Vu
Full reference:
Ahlfeldt, Gabriel, Hans Koster and Tu Giang Vu (2025). Mind the lag: Using assessed and list prices as proxies for housing market values. Berlin School of Economics Working Paper 83.