Seasonal Adjustment in House Price Indices in the COVID-19 Period: A Study of Residual Seasonality

The COVID-19 pandemic has caused a substantial increase in house price growth and its seasonal amplitude. This change has presented challenges to traditional seasonal adjustment methods, such as the U.S. Census Bureau's X-12 method. This study tests for residual seasonality in three major national house price indices that utilize X-12 or its variant X-12 ARIMA for seasonal adjustment: the Freddie Mac House Price Index; the Federal Housing Finance Agency House Price Index; and the S&P CoreLogic Case-Shiller Home Price Index. Evidence of the under-adjustment for seasonality in these series is established, particularly since 2020. An alternative, the Unobservable Components Model, a model-based structural time series approach, provides improved seasonal adjustment performance relative to X-12. An increase in housing demand seasonality, as proxied by Google search queries, is related to the increase in house price growth seasonality.