Constantinos Herodotou: Recent developments in housing markets and related policy challenges

Speech by Mr Constantinos Herodotou, Governor of the Central Bank of Cyprus, at the HL conference organised by the NBB, Nicosia, 18 February 2022. 

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
18 February 2022
PDF full text
 |  3 pages

The target of macroprudential policy is to mitigate threats to financial stability, by reducing vulnerabilities and promoting the resilience of the financial system. In order to achieve its goal, the macroprudential toolbox has been designed to include two broad categories of measures.

First category, authorities can use Borrowed Based Measures, such as limits on the Loan-to-Value ratio, which strengthen the resilience of borrowers against potential vulnerabilities and limit demand for credit. The second category are the Capital Based Measures, such as higher risk-weights on specific exposures, which strengthen the resilience of credit institutions, by increasing available buffers that can be released in case of adverse developments1

One of the macroprudential risks authorities face stems from the overheating of the residential market.

The fact that not all euro area countries receive relevant warnings and recommendations by the ESRB is an indication that the Residential Market in the euro area is characterised by heterogeneity. In the upper right quartile of the chart, we find a number of countries that registered an accumulated increase in residential property prices of at least 25% in the last three years.  Other countries have recorded accumulated growth as low as 5%. 

Using as a starting point the observed heterogeneity and by analysing the experiences of several countries, we can draw certain broad conclusions on the effectiveness of the macroprudential toolbox.

The design of Borrower and Capital Based Measures is a decision of authorities. For example, based on information from the ESRB, Belgium introduced a Loan-to-Value on both owner-occupied and buy-to-let properties. Cyprus, in order to deal with a specific sectoral exposure, further to the Loan-to-Value cap, recently introduced an even stricter Loan-to-Value for luxurious properties. However, by observing the real estate cycles registered in a number of countries, it is evident that even the idiosyncratic design of these measures does not always stop real estate cycles from materialising. The Netherlands and Slovenia are examples of countries that have recorded vulnerabilities despite the implementation of such measures.

Analysis performed at the Central Bank of Cyprus verifies this observation.

Using an econometric model, we explain the growth rates of housing loans and house prices considering the implementation of macroprudential measures2.

The analysis indicates that the Loan-to-Value ratio seems to be effective in containing housing loans, for 10 out of 12 countries in our sample and effective in containing house prices in only in 3 out of the 12 cases. Income based measures (such as Debt-Service-to-Income) and Capital Based Measures were found to be effective in around half of the countries that use them.

We can therefore conclude, that there is no "one-size-fits-all" type of macroprudential tool.

One of the reasons why macroprudential tools are not always effective, could be that vulnerabilities are not necessarily driven by the credit cycles. In the cases examined, it is evident, that credit for house purchases is not always correlated with the trends in housing prices. For example, Slovenia and the Netherlands experienced a build-up in vulnerabilities in the residential real estate market without a corresponding excessive growth in housing loans.

Structural factors could explain the above observation as they affect both demand and supply of housing. More particularly,

  • Net migration and population growth are factors that have continued to put pressure on house prices in countries such as Luxembourg, which experienced net population growth of 13,6% in 2020. Countries with low vulnerabilities in the housing market, such as Greece, experienced a negative population growth.
  • Strong preference for home ownership could also be a driving factor for the observed vulnerabilities in Luxembourg and Slovakia. Homeowners represent 92,9% of the population of Slovakia whereas in France, a country with low identified real estate vulnerabilities,  homeowners represent 65,2%. 

From the above examples, we can conclude that in designing a macroprudential tool, idiosyncratic structural factors need to be identified and accounted for.

Monetary policy also plays a role. Although macroprudential policy is the first line of defence, the ECB has recognised that Financial Stability is a precondition for price stability. It has also been acknowledged that monetary policy, can, in principle, influence asset prices such as real estate.

This interconnection, if broken down in its simplest form, has two possible outcomes that relate to financial stability. 

First outcome, the transmission mechanism of conventional monetary policy includes the ways interest rates directly influence the supply for credit for house purchases and indirectly influence the real economy through house prices3. In the history of conventional monetary policy decisions, the ECB, at times, responded to the presence of excessive credit in order to safeguard its price stability objective. In its August 2006 decision to increase interest rates, the ECB stated that the "stimulative impact of the low level of interest rates in the euro area has been an important factor behind the tendency for money and credit growth to strengthen".

In the period of quantitative easing, monetary policy tools expanded to include, among others, targeted operations to stimulate bank lending to the real economy4. These are linked to a threshold of the banks' loans to non-financial corporations and households. By recognising the impact this tool would have on housing prices, housing loans have been excluded from the this threshold.

Given that monetary policy affects real estate developments, the role of macroprudential policy becomes important in addressing idiosyncratic vulnerabilities and as such the independence of macroprudential policy should be safeguarded, especially in a monetary union.

The Second outcome is that, an increase in interest rates will increase the monthly loan payments. Performing an interest rate sensitivity analysis5, we find that households, in highly indebted countries and relatively lower income levels (such as Cyprus and Portugal) could lose up to 3,5% of their net monthly income with a 1% interest rate increase. Therefore, in calibrating macroprudential tools, authorities should not use static assumptions, but should account for probable future interest rate or income changes.

There are of course other exogenous factors that affect the housing market, such as Fiscal Policy. For example, tax incentives for house purchases that are not income based could be a driver for housing price pressures. However, an efficient social policy that targets inequalities and house deprivation, could limit risk taking by financial institutions and support financial stability during downturns.  

To sum up, the euro area residential real estate market is characterised by heterogeneity. Vulnerabilities in the real estate market are not necessarily cyclical in nature but structural factors also play a role. A single policy or measure cannot be enough to tackle the materialisation of risks from the residential market, although macroprudential authorities, as the first line of defence, have a significant toolbox in their hands, that can help in containing these risks.

1 "Macroprudential analysis of residential real estate markets", Duca et al., ECB Macroprudential Bulletin, 2019.

2 Other explanatory variables included are the year-on-year growth on GDP and the 3-month Euribor rate, as well as past observations in the growth rates of housing loans and house prices. The starting date of the estimation sample of the countries examined ranges between 2003Q1 and 2011Q3, with the last observation being 2021Q3 for all countries.

3 Frederic S. Mishkin, Housing and the Monetary Transmission Mechanism, NBER working paper 13518, October 2007.

4 The period of Quantitative Easing is defined according to the speech by Vítor Constâncio, Vice-President of the ECB "Past and future of the ECB monetary policy" and starts in July 2014.

5 Data from the Household Finance and Consumer Survey, 2017.