Mark-to-market and house asset valuation: An initial attempt at extending the Poterba model using the term structure of real forward interest rates

Abstract

Purpose - Downward movements in house prices can exacerbate bank crises if mark-to-market methods of asset valuation are used by lenders to assess their current balance sheet exposure. There is an imperative to find methods of house price index calculation that reflect equilibrium prices rather than temporary undershoots. The purpose of this paper is to propose a new methodology in order to evaluate whether market house prices are different from their fundamental asset prices. Design/methodology/approach - This paper proposes a method for house asset valuation that incorporates expected house price appreciation as an endogenous variable. This avoids the necessity to make conjectures about expected future house price appreciation when applying Poterba's user-cost method of house asset valuation. The methodological extension to Poterba's user-cost method of house asset valuation endogenises expected house price appreciation as the no-arbitrage expected price appreciation consistent with the term structure of real interest rates. A benchmark equilibrium house valuation can be calculated because the term structure of real forward interest rates is observable in financial markets. This enables market house prices to be compared with the benchmark equilibrium valuation in order to determine if house prices are overvalued or undervalued. Findings - The paper presents the results of a worked example to illustrate how this approach could be applied in practice. Research limitations/implications - There are a number of issues associated with the measurement of user cost which we do not address here and which the authors hope will provide fruitful avenues for future research. There are also issues regarding the impact of tax frameworks on the returns to housing, particularly the taxation of mortgage interest and imputed income. More work also needs to be done in comparing the performance of the extended Poterba model against alternative approaches, such as those that use expected inflation and/or long-run average house price appreciation, or the real interest rate spread to proxy for expected capital appreciation, and how these different approaches compare in different institutional and socio-economic contexts. Practical implications - The authors' results underscore the rationale for mortgage banks to use marking to model instead of marking to market, and this in turn should reduce unnecessary macroeconomic instability when the market prices of houses undershoot fundamental value. Originality/value - The paper shows how the term structure of real forward interest rates, observable in financial markets, can be used to extend the Poterba model.