European Economic Perspectives 21

All Change

Differences in housing and financial market institutions mean that interest rate moves will affect eurozone countries in different ways, according to new CEPR research. How can policy-makers ease the tensions this will generate?

Despite the pressures for convergence as a result of the build-up to monetary union, the differences in housing and financial market institutions across the 15 EU members remain large. And according to new CEPR research by Duncan Maclennan, John Muellbauer and Mark Stephens, these differences will have profound effects on the relative responsiveness of output and inflation in the different countries to changes in short-term interest rates, as well as to asset market shocks of external origin.

Though the pressures for institutions to converge will intensify as the euro becomes established - and there will be important changes, for example, in corporate bond markets - the barriers to convergence are often deeply rooted, not least in different national housing systems. The persistence of these differences is likely to cause instability in those economies at the extremes of the spectrum of relevant dimensions, not only eurozone members such as Finland, Ireland, but also non-members such as Sweden and the UK.

Such instability may lead to potentially serious tensions between eurozone members, these researchers contend. Moreover, unless the issues are recognised by policy-makers and concerted action is taken, the barriers to convergence in housing and credit markets and associated legal, regulatory and fiscal structures will cause these tensions to persist. They matter for whether the UK and other ‘outs’ join the euro as much or more than whether business cycles are roughly synchronized at the moment of entry. Indeed, they matter for the success of the whole monetary union project.

Previous attempts to examine differences in the monetary policy transmission mechanisms between European countries have tended to focus on the macroeconomic level. A 1995 study by the Bank for International Settlements, for example, suggests that of the countries studied - Austria, Belgium, France, Germany, Italy, the Netherlands, Spain and the UK - the UK had by far the greatest sensitivity of output to interest rate rises. Moreover, since variable rate mortgage costs have a large weight in household budgets, as well as in the retail prices index, the UK is unique in having a positive inflation response to a rise in interest rates both in the year of the rise and the following year before turning negative in the third year.

Maclennan and his colleagues explore the different mechanisms by which a rise in short-term interest rates affects the major components of expenditure, such as consumption and investment. They argue that economists typically place undue emphasis on minor direct effects, such as ‘intertemporal substitution’ - the lowering of current consumption in favour of more future consumption as the interest rate rises. The indirect effects, which operate through asset prices and expectations of income growth, are of far more importance in practice but receive less attention.

For example, institutional differences between countries in pension and corporate finance systems and differences in the scale of government debt, reflected in national differences in equity and bond market capitalization, generate differences in financial asset to income ratios and in the ‘spendability’ of those assets. A rise in real house prices not only leads to a positive wealth effect on non-housing consumption, in the same way as a rise in real share prices, but also a negative income and substitution effect. This implies a positive effect on the consumption of owner-occupiers and a negative effect on the consumption of market renters.

If rental housing is owned by pension funds, whose assets are regarded as relatively illiquid by consumers, a rise in house prices has bigger consumption effects in countries with a small market rental sector and a high rate of owner occupation. Moreover, countries can vary in the effective spendability of owner-occupied housing wealth. Spendability is high where transactions costs are low and where housing is effective collateral. Together, these differences can imply a tiny impact on consumption from a rise in real house prices in some countries and a substantial positive impact in others.

Further differences can arise in the transmission mechanism between interest rates and asset prices. Maclennan et al identify as many as five distinct elements, for example, in the effect of interest rates on house prices. Not only are there important non-linear elements in these effects, but differences in history as well as in the proportions of fixed and floating rate mortgage debt, can generate big differences between countries.

Indeed, there are big differences across Europe in housing tenure, in loan-to-value ratios and mortgage debt to income ratios as measures of credit availability, and in taxes and other aspects of transactions costs that affect the size of housing wealth effects on consumption. The responsiveness of mortgage and other borrowing rates to short-term interest rates varies greatly across Europe. Financial market capitalization relative to GDP and, underlying these, banking and corporate finance systems and especially pension systems differ substantially.

Available evidence on the effects of financial and housing assets on consumption is broadly in accord with the differences between countries predicted by this analysis. The UK, for example, which tends to be at one extreme in most relevant dimensions, has a more speculative housing market and larger housing wealth effects than core eurozone countries like Germany.

Maclennan and his colleagues argue that there are strong barriers to convergence between the different national housing and financial systems. They all have their roots in different banking histories, state ownership and intervention, and legal and regulatory structures. For example, Italy’s legal process governing the use of housing as collateral is effectively non-functioning. And the mortgage markets in other countries, such as Germany and France, are notoriously closed to outside entrants.

Nevertheless, there are institutional reforms and compensating policy shifts that may ease the tensions generated by national differences and reduce the risks of membership of the monetary union. The researchers conclude by suggesting six areas where Finland, Ireland, the UK and, to some degree, Sweden, whose characteristics tend to cluster at one end of the institutional spectrum, could take action to reduce the risks of eurozone membership.

• Encourage the use of fixed-rate mortgages and loans more generally.

• Place somewhat tougher prudential upper limits on loan-to-value ratios.

• Particularly in the UK, retain a more significant ‘pay-as-you-go’ element in the public pension system.

• Encourage the development of the private rented sector and encourage the move towards market rents in as much of the social housing sector as possible.

• Reduce house price volatility not by increasing frictions in the housing market - for example, by raising UK stamp duty to the higher levels seen elsewhere in Europe - since such measures reduce labour mobility, but instead by using property taxes more flexibly.

• Encourage policy-makers in the core eurozone countries to open credit markets, the legal profession and estate agents to competition, reduce legal barriers to the use of housing collateral and cut transactions costs.

The last three of these reforms would benefit Europe as a whole by encouraging labour mobility and bringing other efficiency and welfare gains. This type of harmonization is likely to be more beneficial than the harmonization of income and corporation taxes that has been so widely discussed in recent months.

This article reviews research reported in ‘Asymmetries in Housing and Financial Market Institutions and EMU’ by Duncan Maclennan, John Muellbauer and Mark Stephens, CEPR Discussion Paper No. 2062 (January 1999). Muellbauer is at Nuffield College, Oxford and a CEPR Research Fellow; Maclennan and Stephens are at Glasgow University.

Download PDF

Return to contents

 Browse Archives