Best way to Forecast House Prices
Forecasting house prices is not a simple task. Simplistic methods of assuming a constant growth rate based on past trends is very simplistic and may prove quite risky in periods that are followed by a downturn.
If we look at historical data of house price movements we will see that rarely house prices change at constant annual rates. Furthermore, usually we can observe a cyclicality in house price movements,which suggests that the possibility of the reversal of a rising path is real. Many times, the timing of this reversal is not foreseeable as it is due to unpredictable demand shocks that result in severe reduction of economic activity and reduction of demand for property.
Sometimes the house price reversal is predictable when prices rise at levels that are significantly above the prices that are affordable by middle-income households, or when the supply of new housing units rises at a faster pace than demand.
In any case, the practice of assuming a constant growth rate and project it for some years ahead should be used with great caution and only as a last resort if no alternative methodology can be used.
Ideally, the best route for forecasting house prices in the local market is through structural econometric modeling, which analyses the historical behavior of the local housing demand, supply and prices and quantifies their interactions as well as the influence of exogenous economic factors on demand and supply. Subsequently, forecasts of the exogenous factors can be used in combination with the estimated calibrated model to simulate the future behavior of the market and predict how prices will move.
Forecasting house prices through econometric modeling of the local housing market though is very data demanding and many times is not feasible due to the lack of sufficient data. In such a case, the analyst needs to examine recent percentage changes in prices, and assess whether they will continue at the same or different pace based on the prediction of demand-supply conditions in the local housing market during the forecast horizon. If total housing supply is expected to be greater than total housing demand over the forecast horizon then any recent increases in prices should not be projected into the future