Elsevier

Economic Modelling

Volume 25, Issue 3, May 2008, Pages 377-390
Economic Modelling

Assessing the role of income and interest rates in determining house prices

https://doi.org/10.1016/j.econmod.2007.06.010Get rights and content

Abstract

Property prices across many OECD countries have witnessed remarkable increases over the past 10 years. Two factors frequently posited for this boom are higher income levels and the benign interest rate environment experienced in many of these countries. However, empirical models of house prices struggle to achieve credible results concerning the impact of interest rates with coefficients that are frequently insignificant or of the wrong sign. In this paper we propose an intuitive theoretical model of house prices where the demand for housing is driven by how much individuals can borrow from financial institutions. This level of borrowing depends on disposable income levels and current interest rates. We empirically test this model by applying it to the Irish property market. Our results support the existence of a long-run relationship between actual house prices and the amount individuals can borrow with plausible and statistically significant adjustment to this long run equilibrium.

Introduction

The persistence of the present boom in international property prices is unparalleled in recent times. Over the five year period 2000–2005, estimates by The Economist1 reveal that the value of residential property in developed countries rose by over 30 trillion dollars — an increase equivalent to 100% of those countries combined GDPs. In North America and across Europe, countries have experienced record highs in terms of house price to income ratios. Inevitably, the concern amongst policy-makers is the inherent stability and sustainability of this asset price increase — are property markets overvalued and if so, by how much? As noted by Case and Shiller (2003), the international media has, of late, been saturated with stories/analyses documenting the imminent “collapse” of property bubbles.

Reviewing studies of cross-country property markets reveals some agreement in identifying the underlying determinants of the demand and supply of housing. Two of the key drivers frequently cited in the recent run up in house prices have been rising income levels and the benign interest rate environment faced by many countries. Less agreement, however, is forthcoming on the theoretical and empirical approaches used to model these factors. For example, it is not uncommon for price levels in the same property market when analysed with two different (and popular) approaches to be deemed either “determined by fundamentals” and consequently, sound or, conversely, “dangerously overvalued”.

It is possible to separate much of the existing literature into two broad approaches. The first we call the “econometric” approach whereby a reduced form price equation is estimated based on some underlying notion of the determinants of supply and demand. Typically, house prices are regressed on a set of potential determinants. The fitted values from the regression are then interpreted as the price level justified by fundamentals within the economy and the potential stability of the asset price increase is gauged by comparing this fundamental price with the actual price level.2 One of the problems with this approach is that variables which are believed, a priori, to be important in house price determination such as interest rates often appear with the wrong sign or are found to be insignificant. For example, in models estimated for eight different US States, Case and Shiller (2003) acknowledge that the mortgage rate had an insignificant coefficient in all but one of the regression models. Mayer (2003) also notes that the results from such regression models suggest that, historically, house purchase behaviour and housing values may not have been very responsive to changes in interest rates.

An alternative, more finance-based, approach taken in the literature can be characterised by an underlying notion of arbitrage where the returns to investing in housing relative to some other asset are evaluated or the costs and benefits of renting relative to buying are compared. One standard metric used in this context is the ratio of rental income to house prices. Deviations of the current rental price ratio from its long-run average are frequently taken to be an indication of over or undervaluation.3 A more sophisticated implementation of this approach, based on the methodology of Campbell and Shiller, 1988a, Campbell and Shiller, 1988b has been recently applied to the US housing market by Campbell et al. (2006). In this type of model, a tight relationship is imposed between house prices and interest rates. This contrasts with the former, econometric approach where the interest rate variable enters in freely into the regression specification and can often be “swamped” in the estimation yielding a very small and minor semi-elasticity effect.

However, one of the potential drawbacks of many finance based approaches is that underlying supply and demand factors such as income or demographics are not modelled. Rather, these factors enter indirectly by affecting either the growth rate of rental income or in terms of a changing discount factor. Moreover, this approach has little to say regarding any adjustment path for house prices if house prices are away from their fundamental level. In recent times many of these finance-based indicators such as the rental price ratio have deviated substantially from their long-run average for a number of different housing markets. OECD (2005) illustrate this fact for 14 out of the 17 international housing markets examined.4 However, the implied overvaluation from such measures is, at times, at variance with the results from reduced form econometric models, which tend to suggest far less evidence of overvaluation.

In this paper, we propose a simple intuitive theoretical model of the housing market which captures the important role of credit, income and interest rates as drivers of housing demand but also resolves some of the difficulties of previous approaches already highlighted. More specifically, we model the demand-side determinants of house prices as a function of the average amount borrowed by households given current disposable income levels and interest rates. In reality, the amount lent by a mortgage institution to an individual is critically dependent on current disposable income and interest rates. Based on this observation, we back out how much a financial institution would lend an individual given plausible assumptions regarding the fraction of income that goes to mortgage repayments and the duration of the mortgage using a standard annuity formula. Ultimately, this value should be an important determinant of housing demand. We believe this model captures the fact that most house purchases are mortgage-financed and the amount that mortgage providers are willing to lend is ultimately a function of income and interest rates.

In contrast to the finance approach, however, we do not derive a “fundamental” price level and then compare it with the actual level. Instead, we estimate both a long-run relationship between house prices and the amount that can be borrowed and a short-run model that examines the speed of adjustment when there is a deviation from the long run equilibrium. We apply the model to the Irish property market. This market has been to the fore of the international trend of rising house prices. Over the ten year period 1995–2005, prices for new Irish houses rose by almost 260%. Given the exceptional performance of the Irish economy over the same period, the Irish housing market is a particularly interesting case study of rising house prices in the context of increasing income levels and a low and stable interest rate environment. The former is attributed to the rise of the so called Celtic Tiger while lower interest rates have coincided with Ireland's entry and membership of the European Monetary Union (EMU).

We believe our model draws upon the advantages of both the econometric and finance based models while avoiding some of their drawbacks. In combining a theoretical and an empirical model, we think our approach has a number of merits to recommend it. First, the model is intuitively appealing, familiar as it is to most people who have taken out a mortgage. In addition, it models, in a plausible fashion, how mortgage institutions decide how much to lend.

Secondly, since we impose a realistic theoretical relationship between interest rates, income and how much one can borrow, we avoid the shortcomings of having an insignificant or incorrectly signed interest rate response — something that is characteristic of much of the previous literature. Accordingly, the proposed model is particularly useful for scenario analysis aimed at capturing the effects of changes in income and interest rate movements on house prices. This is important in light of the recent monetary tightening by policymakers in both the euro area and the US. Previously mentioned models would implausibly suggest little or no impact of higher interest rates on house prices. To further illustrate this point, we conduct a counterfactual exercise in assessing what impact the lower interest rate environment experienced by the Irish economy since joining monetary union has had on house prices relative to a regime where an independent monetary policy was pursued.

Finally, in estimating our long and short-run models, we achieve plausible and robust results in terms of the relationship between the actual and predicted price levels. This contrasts with issues of fit which can arise with the more finance-based models where the price suggested by, say, rental price ratios, are often quite out of kilter with the actual observed price.

The rest of the paper is organised as follows; in the next section we introduce our theoretical model of house prices. We then discuss the Irish housing market, while the following section describes the empirical approach adopted in this paper. The results of the empirical approach are next discussed and we assess whether Irish house prices are overvalued. Finally, we conduct our counterfactual exercise and offer a brief conclusion.

Section snippets

A theoretical model of house prices

In considering a model of house prices we define the following variables

    Pt

    actual house prices.

    Bt

    amount that can be borrowed.

    St

    supply of housing.

    Yt

    disposable income per household.

    Rt

    mortgage interest rate.

    τ

    duration of mortgage.

    κ

    proportion of household income going on mortgage repayments.

In our model, we concentrate on the role played by the demand-side factors — income and interest rates. In particular, we argue that the demand for housing is mainly a function of the amount that prospective house

The Irish housing market

Over the sample period considered (1980–2005), the Irish economy has experienced profound economic change. Ireland, in the 1980's, witnessed negligible economic growth, an average unemployment rate of 14% and high levels of personal taxation. The emergence of the so-called Celtic Tiger in the mid 1990s led to a sustained period of economic growth. Between 1995 and 2005, the size of the economy doubled with the total number of people employed in the country increasing by almost 50%. This

Concluding comments

The role of interest rates as a primary determinant of house price movements is virtually uncontested. However, economic models of house prices have struggled to successfully “incorporate” the effects of interest rate movements. In this paper we propose that the house price demand schedule can be adequately represented by the average amount borrowed, which is determined on the basis of prevailing disposable income levels and interest rates.

This approach has a number of attractions. It imposes a

References (30)

  • G. Kenny

    The housing market and the macroeconomy: evidence from Ireland

    Economic Modelling

    (1999)
  • N. Mankiw et al.

    The baby boom, the baby bust and the housing market

    Regional Science and Urban Economics

    (1989)
  • M. Roche

    The rise in Dublin city house prices: bubble, fad or just fundamentals

    Economic Modelling

    (2001)
  • T. Breusch

    Testing for autocorrelation in dynamic linear models

    Australian Economic Papers

    (1978)
  • J. Campbell et al.

    The dividend price ratio and expectations of future dividends and discount factors

    Review of Financial Studies

    (1988)
  • J. Campbell et al.

    Stock prices, earnings and expected dividends

    Journal of Finance

    (1988)
  • S. Campbell et al.

    A trend and variance decomposition of the rent–price ratio in housing markets, 2006-29 (August)

    Finance and Economics Discussion Paper

    (2006)
  • K.E. Case et al.

    Is there a bubble in the housing market?

    Brookings Papers on Economic Activity

    (2003)
  • D. Conniffe et al.

    Irish house price indices — methodological issues

    The Economic and Social Review

    (1999)
  • D. Duffy et al.

    Rising house prices in an open labour market

    The Economic and Social Review

    (2005)
  • G. Elliot et al.

    Efficient tests for an autoregressive unit root

    Econometrica

    (1996)
  • R. Engle

    Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation

    Econometrica

    (1982)
  • R. Engle et al.

    Cointegration and error-correction: representation, estimation and testing

    Econometrica

    (1987)
  • T. Fitzpatrick et al.

    House prices and mortgage credit: empirical evidence for Ireland

    Manchester School

    (2007)
  • Financial Stability Report., 2005

    (2005)
  • Cited by (110)

    View all citing articles on Scopus

    The views expressed in this paper are the personal responsibility of the authors. They are not necessarily held either by the CBFSAI or the ESCB. The authors would like to thank Karl Whelan, Maurice McGuire and Maurice Roche for helpful comments. Any errors are the sole responsibility of the authors.

    View full text