How to Forecast Cap Rates

Cap rate forecasts are useful in predicting the sales price of a property investment. Such forecasts can be produced by dividing the net operating income (NOI) that the property is expected to be producing at the time of sale by an exit cap rate. A sales price estimate is necessary for calculating the expected profitability of a property investment and specifically the internal rate of return (IRR).  Studies have shown that cap rates move in a countercyclical way in relation to the rent and property value cycle.

Generating Cap Rate Forecasts

The process for generating econometric cap rate forecasts is highly technical with many steps and significant data requirements. The process for forecasting cap rates using econometric techniques includes the following steps:

  • First, the analyst needs to determine the property type for which the cap rate forecast is needed. Cap rate fluctuations and levels differ considerably across property types and metropolitan markets (Sivitanides and Sivitanidou, 1996).
  • Determine the market for which the forecast is needed. Cap rate movements and levels vary also across local markets, as they depend to a significant extent on local market conditions. Usually, the data needed for producing econometric forecasts are available at the metropolitan area level. Furthermore, real estate markets are considered to be integrated at the metropolitan level, in terms of supply and demand. Although there are submarkets within a metropolitan area that are likely to have different cap rate levels, due to different locational and other characteristics, those levels are not independent of movements in cap rates in other submarkets.
  • Obtain reliable and consistent (in terms of their measurement) cap rate data from vendors for the chosen property type and market. Note that you have to have historical data for a sufficiently long period and ideally for a full cycle over which cap rates have been rising and declining as well. The historical data may be annual or quarterly but, because cap rates move slowly, annual data may provide more robust estimates and forecasting equations.
  • Obtain also historical data for the independent variables to be used in the econometric analysis. This must include data on historical movements in critical real estate local market variables such as rents, vacancy rates, and absorption for the market and property type under consideration. You will also need data on interest rate movements and returns in alternative investment vehicles, such as stocks and bonds, as movements in local cap rates are affected by conditions in the national capital market. Historical movements in the local economy, if they can be obtained, may also help explain cap rate fluctuations in the particular market. Note that the historical data for the independent variables have to refer to the same historical period to which the cap historical cap rate data refers to.
  • A key point when developing econometric models for forecasting purposes is that there is no point in including independent variables for which reliable forecasts are not available.  Without such forecasts, it will not be possible to generate cap rate forecasts based on the estimated statistical equations.
  • Estimate alternative econometric models using different combinations of potential explanatory variables, as well as different time lags. For example, the analyst may estimate alternative equations where the rent is lagged 0, 1, or 2 years in order to see which one is more strongly correlated with movements in cap rates. The same experimentation applies to all independent variables that may affect cap rates with a time lag. Historical data on appraisal-based cap rates indicate that they are moving slowly through time. Therefore, partial-adjustment models are well suited for developing statistical equations that describe the intertemporal behavior of capitalization rates and, therefore, for generating cap rate forecasts. The basic partial-adjustment specification is derived from the following equation, which is well-suited in describing the behavior of slowly adjusting variables:
                                            C(t) = C*(t-1) + r [C*(t) – C(t-1)]                                                                    In the above equation, r captures the per-period rate of partial adjustment, while C*(t) represents the long-run equilibrium cap rate given the values of the drivers of cap rates at time t. According to Sivitanidou and Sivitanides (1999) these drivers include indicators of the strength and the risk of the local market for the particular property type considered and capital market influences. The above theoretical specification translates to the following statistical equation in which C(t-1) is the dependent variable (the cap rate) lagged by one period and the X variables represent the aforementioned factors that drive intertemporal movements in cap rates:                                                                                                                                      C(t) =  b0 + b1C(t-1) + b2X1(t-n) + ……. +  bk+1Xk(t-n)                                Note that n in the above equation represents the number of periods that it takes for the particular X variable to affect market cap rates. The above statistical equation allows quantification of the short-term effects of the independent variables, which are represented by their estimated coefficients, and the long-run effects, which are derived by dividing their estimated coefficients by r (Zheng et al., 2010). As it can be derived from the above equations, r can be estimated as r=1b1, where b1 is the regression estimate of the coefficient of C(t−1). Note that given the presence of the lagged dependent variable, the Durbin-Watson statistic is no longer a reliable indicator of the presence of serial correlation in the error term. Instead, in order to confirm that the estimated results of the regression are not distorted by the presence of serial correlation, the Durbin-h test needs to be carried out.
  • Select among the alternative econometric models the one that best explains historical movements in capitalization rates in the particular market.
  • Obtain forecasts of the independent variables that are included in the best fitting model for the market under consideration.
  • Produce cap rate forecasts for the particular property type and market by using the coefficients of the best-fit model in combination with the forecasts of the independent variables in the model obtained in the previous step.
  • Examine the cap rate forecasts produced by the model for their reasonableness.
  • If the forecasts produced strike as unreasonable and extreme use second best-fitting model to produce forecasts, and repeat the last three steps until you get a forecast that seems reasonable, given the historical behavior of cap rates in that market and property type.

When using a particular historical series of cap rate data it is important to understand whether they are based on appraisals (referred to as appraisal-based) or transactions (transaction-based). As the term implies, appraisal-based cap rates are calculated using appraised property values, while transaction-based cap rates are calculated using actual transaction prices of properties that were sold during the period to which the cap rate estimates refer to. Historical appraisal-based cap rates are usually calculated using data from the National Council of Real Estate Investment Fiduciaries (NCREIF). Historical appraisal-based cap rates and forecasts for over 50 of the largest metropolitan markets in the US can be obtained for a fee from CBRE Econometric Advisors, while transaction-based cap rate data can be obtained for a fee from Real Capital Analytics.


Sivitanides, P. and R. Sivitanidou. 1996. Office Capitalization Rates: Why Do they Vary across Metropolitan Office Markets? Real Estate Issues, Vol. 21, No. 2, pp: 34-39.

Sivitanides, P., J. Southard, R. Torto, and W. Wheaton. 2003. Real Estate Market Fundamentals and Asset Pricing. Journal of Portfolio Management, Special Issue, 45-53.

Sivitanidou, R. and P. Sivitanides. 1999. Office Capitalization Rates: Real Estate and Capital Market Influences. Journal of Real Estate Finance and Economics, Vol. 18, No. 3, pp: 297-322.

Sivitanidou, R. and P. Sivitanides. 1997. Exploring Capitalization Rate Differentials Across Property Types. Real Estate Issues, Vol. 22, No. 3, 47-54.

Zheng, L., Rose, J.M. and Hensher, D.A. (2010), “Forecasting automobile petrol demand in Australia: an evaluation of empirical methods”, Transportation Research Part A: Policy & Practice, Vol. 44, No. 1, pp. 16-38.

Related Posts

The Cap Rate Cycle
Going-in Cap Rate
Property Investment Basics: Equity Capitalization Rate
What is the Capitalization Rate?
How to Calculate a Market Cap Rate
How to identify property markets for capital gains?

Leave a Reply

Proudly powered by WordPress | Theme: Baskerville 2 by Anders Noren.

Up ↑

%d bloggers like this: