This study investigates the interactive effect of changes in the Gross Domestic Product (GDP) and the Consumer Price Index (CPI) on US multifamily cap rates. The data from the US and 20 of its metropolitan statistical areas (MSAs) was used from 2000 to 2021. The accompanying cap rate data is sourced to Green Street. A binary logistic regression model was specified by reducing the interaction between first-differenced GDP and CPI to a single binary variable and reducing the first-differenced cap rate series to a binary variable. Cap rate changes are forecasted, and the model is evaluated using standard goodness of fit methods, a confusion matrix, and a comparison to a buy-and-hold strategy. Overall, this study provides new evidence to explain and simplify the impact of inflation and economic growth on cap rates. The results show that the method of forecasting cap rates is highly robust in locations where growth is consistent with the national average (established cities), while it is less robust in fast-growing markets. It can be inferred that, in established cities and the US as a whole, cap rate growth can be modeled as a function of only the underlying GDP growth relative to CPI growth.
Read full abstract