Club convergence in European housing prices: The role of macroeconomic and housing market fundamentals
Introduction
The topic of convergence across the EU in terms of the main macroeconomic and financial factors has been extensively researched. However, to date there has been little analysis of convergence in European housing markets, despite the well-established relationship between house prices and economic factors. The Law of One Price (LOOP) hypothesis states that prices of individual traded goods should be equal when converted at the market exchange rates. The mechanism that drives the LOOP condition is frictionless goods arbitrage. Under this condition, prices of goods in different locations should move together; furthermore, in the absence of trade barriers or nominal exchange rate fluctuations, prices of goods should converge (Parsley and Wei, 1996). While housing is not a traded good, and therefore cannot be easily arbitraged, economic and financial integration should mean that housing markets are exposed to similar shocks and prices in different locations will co-move, such that convergence may occur. First of all, convergence in house prices may itself be the result of convergence in housing market fundamentals, such as interest rates and income.1 Second, the integration of financial markets can promote convergence in the cost of financing a housing purchase, easing borrowing constraints and facilitating access to credit. The convergence in the cost of financing is linked to the convergence of real long-term interest rates and, in particular, of long-term public debt, considered the opportunity cost borne by a financial institution when granting a mortgage. Finally, the reduction of market segmentation and the elimination of foreign exchange risk may cause convergence of housing risk premia associated with returns on housing as an asset. Therefore, the convergence of the determinants of house prices (in particular income and the cost of financing), as well as the reduction in market segmentation and housing risk premia, should lead to a narrowing in the disparity in house prices across EU countries.
The housing sector, on the other hand, is a key element of the macroeconomy.2 Dwellings and the land on which they lie contribute substantially to household wealth, and this wealth is in turn a key determinant of aggregate consumption. In addition, real estate is most people's primary savings channel, and residential investment promotes growth and job creation. Mortgages are the greatest element of household debt and default during economic downturns can put the banking system at risk. The economic impact of the housing sector is a major consideration in monetary and fiscal policy, and influences business cycles (Leamer, 2007, 2015). The behaviour of the housing market is even more relevant for the economic performance of economically integrated areas, such as the Eurozone, since along with the described mechanisms that operate at national level, market integration establishes new market forces and provides a new source of shocks (Bertola, 2010).
While the process of economic integration was expected to lead to greater convergence between countries (Franks et al., 2018), some authors pointed out that, on the contrary, regional specialization of production and the emergence of agglomeration economies would lead to divergence (Krugman, 1991). Moreover, as De Grauwe (2012) points out, the process of economic and monetary integration exposed areas to asymmetric endogenous shocks, thereby magnifying and mutually reinforcing the structural imbalances that pre-dated the creation of the common currency (Ordóñez et al., 2015). Given the short-term nominal interest rate set by the European Central Bank, the existence of divergences in inflation rates among countries caused lower real interest rates in inflation-prone countries, increasing cross-border capital flows, which in turn fuelled investment in non-tradable activities as well as asset and housing bubbles.3 As a consequence, inflationary countries specialized in non-tradable activities and construction, while the less inflationary countries in the monetary union relied on exports and tradable activities. Different specialization implied structural divergence (Buti and Turrini, 2015), and the rise of imbalances in public deficits and debts, and in the current account position (Ordóñez et al., 2015). In addition, activity specialization affected the distribution of income and the rise of inequality among the European countries (Monfort et al., 2018). Following the 2008 financial crisis, it was found that housing market risk spread rapidly (De Bandt et al., 2010), revealing the importance of analysing the integration and mutual convergence of housing markets where systemic risk cannot be diversified through transnational investment. Thus, the relationship between the housing market and the macroeconomy has become of increasing importance, such as in the use of macroprudential policies aimed specifically at controlling the housing market and the mutual reinforcement of economic divergences on the one hand, and differences in housing market performance on the other.
Most studies on convergence in the housing markets to date have considered national regional markets, such as regions in the UK and USA, where there tend to be a plentiful supply of data. Many of these studies rely on testing for unit roots or cointegration between the main regions; in the UK this tends to be London and the outer regions. For instance, Holmes et al. (2011) produced evidence of convergence across US states, whilst Cook (2012) similarly showed there was convergence across the UK regions. In addition Holmes et al. (2019) using UK regional data, found no evidence of overall convergence in UK house prices, but did find evidence of UK house price convergence clubs. Holly et al. (2011) find evidence of the ripple effect in UK regional house prices, with the ripple originating in London, but the effect also fed back from the regions to London house prices. They also found evidence of New York house prices affecting London prices. Funke et al. (2019) found high levels of house price heterogeneity across China's cities, where there is also evidence of house price synchronisation within clubs. Muellbauer (1992) was the first to find the effects of the differing housing markets across the EU on their respective economies, highlighting differences between Germany and the UK. Iacoviello (2000) uses a structural VAR to analyse the relationship between European house prices and the main macroeconomic factors, finding that much of the house price volatility is driven by demand shocks and monetary policy. Goodhart and Hofmann (2007, 2008) analyse a sample of housing markets across the main industrialised countries, finding evidence of complex inter-relationships between the housing markets and macroeconomic factors, especially monetary policy. They also find that these relationships have become stronger over the more recent decades. Meen et al. (2012) analyse the relationship between the macroeconomy and housing market from a number of angles, including an assessment of how housing interacts with aggregate demand and supply, as well as the effects of credit and policy on the housing market, particularly with regard to the credit crunch after the financial crisis. Henley and Morley (1999) also found similar effects when analysing relationships between the housing markets, interest rates and consumption, and also found little evidence of convergence across housing markets. However, the early studies of EU housing markets tended to be limited by data availability on house prices, especially for Germany.
There have been a number of more recent studies on the EU housing market, such as Ferrara and Koopman (2010) and Alvarez et al. (2010) who discovered little evidence of increased co-movement between German, French, Italian and Spanish house prices since the creation of the Euro. Van Steenkiste and Hiebert (2011) using a global VAR approach found that interest rates have an increased effect on house prices in later years. Gupta et al. (2015) used cointegration to analyse house prices across eight EU countries, finding that German house prices move in the opposite direction to other EU countries. Miles (2019) has recently employed a similar dataset to the one we use in our study to determine if the creation of the Euro has increased synchronisation across the EU. Using endogenous break tests and dynamic correlations, the results show little evidence of any increase in co-movement across the EU. Tsai (2018) analyses convergence patterns in house prices among euro-zone and non-euro countries and concludes that after the introduction of the euro, house prices in the EU converged. Miles (2020), using a pairwise approach to test for convergence, finds only marginal evidence of European housing market convergence. The underlying phenomena driving the inequalities and lack of convergence across the EU stem from the common monetary policy adopted by the EU when the Euro was created in 1999, and which required a common interest rate across these countries. However due to the heterogeneity of the housing markets across the EU, the single interest rate may be appropriate for some countries, but not for many others, leading to excessive house price rises in some countries, whereas prices stagnate in others.4 A further fundamental driver of the inequalities is fiscal policy, as there tends to be differing levels of taxation and subsidies across these countries. Chang (2020) has tested for common cycles among the housing markets of Canada, the UK and USA, and finds evidence of a related driving force between Canada and the USA, but no relationship between the UK and the other two housing markets.
The aim of this study is to assess the extent of the convergence in house prices across the main European nations, particularly members of the Eurozone, and the determinants of house price convergence. As explained before, house price convergence is expected to occur in integrated economic areas as a consequence of the convergence of the housing market fundamentals within these areas. The existence and extent of such convergence is not only relevant because the housing market is an important component of the national economic activity, but also because it can condition macroprudential policies aimed to alleviate the systemic risk in the housing sector. Furthermore, real convergence in monetary unions is a key element to ensure the sustainability and resilience of the common currency area.
The contribution of our paper is fourfold. First, we test for convergence using the Phillips and Sul (2007, 2009) methodology. Previous literature has used the unit root test, either time series or panel, to test for housing price convergence. The Phillips and Sul (PS) methodology is clearly superior to these approaches for several reasons. Unit root tests for convergence choose a base or reference country for convergence. The choice of the reference country is usually rather arbitrary and is likely to condition the results on convergence. To address this problem, some authors have applied a pairwise approach, I but this can deliver inconsistent results. PS is a test for relative convergence, as it measures convergence to some cross-sectional average, in contrast to the concept of level convergence, thereby avoiding the problems linked to the choice of a reference country or the inconsistency of the pairwise approach. Furthermore, PS allows the researcher to test for club convergence, which is overlooked when using a traditional unit root test. Indeed, one of the characteristics of the European economic integration process is the existence of convergence clubs, consistent with the idea of a two- (or multi-) speed Europe. This phenomenon, which is crucial to understanding the divergences that have arisen between, for example, surplus and deficit countries, cannot be captured by unit root tests. Additionally, the PS approach is agnostic regarding the stationary or non-stationary nature of the series to be tested. Thus, in contrast to unit root tests, the PS approach does not conclude in favour of divergence even if the difference between two series may still have some nonstationary features even though the convergence condition holds. Also, unit root tests tend to conclude in favour of rejecting the null of overall convergence if there is a mixture of stationary and non-stationary series in the panel.5 The PS approach allows for converging or diverging countries within the same panel. Finally, traditional unit root tests tend to classify the difference between gradually converging series as non-stationary. The PS approach, in contrast, accounts for transitional heterogeneity and time-varying speed of convergence, thus overcoming this problem.
Second, a Bayesian dynamic panel model is used to analyse the determinants of house price convergence. Previous studies have been limited to identifying the theoretical mechanisms that could explain the existence of convergence, and then testing for said convergence, but without going on to provide any evidence as to which of the mechanisms indicated by the theoretical literature are actually driving the estimated convergence. Thus, for the first time in the same analysis, we combine two strands of the literature, on house price convergence and on the determinants of house prices, which gives us a clearer understanding of the drivers of price dynamics in an integrated area. Furthermore, since there are several drivers of convergence, we perform an analysis to determine the relative weight or importance of each one. Our analysis thus makes it possible to determine not only the fundamental drivers of the reduction in disparities in house prices between EU countries, but also the relative importance of each of them.
Third, some of the most recently published studies on house price convergence in the Eurozone only include in their analysis countries in this area, which can create selection bias. The reason is simple: from the empirical literature on both real and nominal convergence in Europe, it can be concluded that the formation of convergence clubs does not reveal a pattern of Eurozone countries versus non-Eurozone countries. That is, the clubs identified in variables such as income are comprised of a mixture of Eurozone and non-Eurozone countries, and even countries that are not members of the EU. Therefore, a convergence analysis covering only Eurozone countries could wrongly conclude that the euro has facilitated convergence (or divergence) when in fact the same degree of convergence (or divergence) is observed not only among non-Eurozone countries, but also between Eurozone members and non-members. To overcome this problem, our analysis comprises three groups of countries: Eurozone countries (Belgium, Finland, France, Germany, Ireland, Italy, the Netherlands, and Spain), EU but non-Eurozone countries (Denmark, Sweden, and the United Kingdom6), and a non-EU country (Norway).7
Fourth, despite the abundant literature on spatial spillovers in prices, and particularly house prices, the previous studies that have analysed house price convergence in Europe do not consider this type of effect, even though the location of the housing is an important factor that explains price convergence. Along with the determinants of price convergence, our Bayesian dynamic panel model also considers spatial effects.
While our results do not indicate overall convergence, they do point to club convergence among the analysed countries. Price dynamics in Europe have been driven by demand factors, including household debt and financing costs, along with unemployment and inflation, which may capture economic cycle conditions. Although the EU has no direct mandate on housing policies, the European Commission could use recommendations, guidelines and communications to influence housing market policies. The European Semester procedure could introduce country-specific recommendations to address divergences in housing market performances, preventing spillover effects of these markets on overall macroeconomic stability.
The rest of the paper is organized as follows. Section 2 discusses previous studies on the EU housing market and the potential channels for divergence. Section 3 introduces the methodologies and the data used to test for convergence. Section 4 provides the empirical results and Section 5 concludes.
Section snippets
The EU housing market and policies
The housing markets within the EU tend to be heterogenous with respect to the structural features of the housing markets and the predominant financing forms used in each country, as explained by Maclennan et al. (1998). Countries mainly vary regarding levels of home ownership, with other major differences being the types of housing finance, regulatory frameworks in the respective mortgage industries and the subsidies and taxes used by individual countries.
These differences which have evolved in
Materials and methods
In this section, we discuss the data we employ before describing in detail the equations we estimate.
In this study, two different methods are used, but we employ them in a complementary way to analyse the existence of convergence in house prices in Europe. First, we use the Phillips and Sul (2007, 2009) method of testing for convergence to test the null hypothesis of convergence against the alternative of divergence (or club convergence). Second, we use a Bayesian dynamic panel model to analyse
Descriptive
Table 1 presents the descriptive statistics of the variables in our database. These are the variables accounting for 12 European countries from 2004Q2 to 2016Q3. Fig. 1 shows the evolution of the real HP index for each country.
Club convergence
The real house price convergence across Europe is tested using the Phillips and Sul (2007, 2009) method, which as discussed shows that eliminating the cyclical components of the data improves the finite sample power and size of the club convergence test. The cyclical
Conclusion
The study tests for convergence in house prices across European countries. The results show that there are five clubs: i) Norway and Sweden, ii) Belgium, Finland, France, and Germany, iii) Denmark and the UK, iv) the Netherlands and Italy, and, v) Ireland and Spain. These clubs reflect similarities across markets, in terms of geographical proximity and market structures. There is no evidence of a Eurozone club against a non-Eurozone club, showing that despite the attempts to converge the real
Authors roles
Conceptualization: Javier Ordóñez, Bruce Morley. Data curation: Bruce Morley, Mercedes Monfort. Formal analysis: Mercedes Monfort, Laia Maynou. Funding acquisition: Javier Ordóñez. Investigation: Bruce Morley, Javier Ordóñez. Methodology: Mercedes Monfort, Laia Maynou. Project administration: Javier Ordóñez, Bruce Morley. Resources: Javier Ordóñez, Bruce Morley. Software: Mercedes Monfort, Laia Maynou. Supervision: Javier Ordóñez, Bruce Morley. Validation: Mercedes Monfort, Laia Maynou.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors would like to thank Mack and Martinez-Garcia (2011) for the use of their dataset. We would also like to thank the Editor and two anonymous referees for their constructive and helpful comments, the usual disclaimer applies. Javier Ordóñez dedicates this article to his father, in memoriam. We are grateful for the financial support received from the Generalitat de Catalunya grant 2017SGR656 (Maynou), Generalitat Valenciana grant PROMETEO/2018/102 (Ordóñez) and the Universitat Jaume I
Mercedes Monfort is Lecturer of Economics at the Jaume I University in Castellón. She holds a degree in Business Administration and Management and a MSc in Local Development Management. She received his doctorate in Economics from the Jaume I University with the university's Award for Excellence. She has published papers in Economic Modelling, Energy, Open Economies Review, Empirica, Emerging Markets, Finance and Trade, Plos One and in the International Journal of Economics and Finance among
References (81)
- et al.
Formulation and estimation of dynamic models using panel data
J. Econom.
(1982) - et al.
Spatial variability in mortality inequalities, socioeconomic deprivation, and air pollution in small areas of the Barcelona Metropolitan Region, Spain
Sci. Total Environ.
(2009) - et al.
Interpreting tests of the convergence hypothesis
J. Econom.
(1996) - et al.
Initial conditions and moment restrictions in dynamic panel data models
J. Econom.
(1998) - et al.
Are U.S. Regional incomes converging A time series analysis
J. Monetary Econ.
(1993) Are cyclical patterns of international housing markets interdependent?
Econ. Modell.
(2020)- et al.
The spatial and temporal diffusion of house prices in the UK
J. Urban Econ.
(2011) - et al.
Investigating regional house prices in the United States: evidence from a pair-wise approach
Econ. Modell.
(2011) - et al.
Property heterogeneity and convergence club formation among local house prices
J. Hous. Econ.
(2019) The time series behaviour of house prices: a transatlantic divide?
Journal of Housing Studies
(2002)
Anglo-German differences in housing market dynamics
Eur. Econ. Rev.
Regional cohesion: evidence and theories of regional growth and convergence
Eur. Econ. Rev.
House price convergence in euro zone and non-euro zone countries
Econ. Syst.
House price determinants: fundamentals and underlying factors
Comp. Econ. Stud.
Housing Cycles in the Major Euro Area Countries
Estimation of dynamic models with error components
Journal of the American Statistical Society
House prices and monetary policy in the EURO area
Euro Area Policies: Selected Issues, IMF Country Report No. 05/266
Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations
Rev. Econ. Stud.
Economic Growth
Convergence
J. Polit. Econ.
Productivity growth, convergence, and welfare: what the long run data show
Am. Econ. Rev.
Convergence in international output
J. Appl. Econom.
Inequality, integration, and policy: issues and evidence from EMU
J. Econ. Inequal.
Three Waves of Convergence. Can Eurozone Countries Start Growing Together?
β-convergence and the cyclical dynamics of UK regional house prices
Urban Stud.
Housing finance and monetary policy
J. Eur. Econ. Assoc.
The international transmission of house price shocks. Housing Markets in Europe. A Macroeconomic Perspective
The governance of a fragile eurozone
Aust. Econ. Rev.
Economics of Monetary Union
Model uncertainty in cross-country growth regressions
J. Appl. Econom.
Common Business and Housing Market Cycles in the Euro Area from a Multivariate Decomposition
Economic Convergence in the Euro Area: Coming Together or Drifting Apart?
Mapping China's time-varying house price landscape
Reg. Sci. Urban Econ.
The role of house prices in the monetary transmission mechanism across European countries
Scot. J. Polit. Econ.
House Prices and the Macroeconomy: Implications for Banking and Price Stability
House prices, money, credit, and the macroeconomy
Oxf. Rev. Econ. Pol.
Co-movement in Euro area housing prices: a fractional cointegration approach
Urban Stud.
Macroprudential Measures for Addressing Housing Sector Risks
European House Price Volatility and the Macroeconomy: the Implications for European Monetary Union
Assessing high house prices: bubbles, fundamentals and misperceptions
J. Econ. Perspect.
Cited by (21)
Revisiting international house price convergence using house price level data
2024, Economic SystemsInnovation convergence clubs and their driving factors within urban agglomeration
2023, Economic ModellingThe SPDE approach for Gaussian and non-Gaussian fields: 10 years and still running
2022, Spatial StatisticsCitation Excerpt :As an initial illustration of the method’s practical relevance, we here present an incomplete list of recent applications. In the time-period of May–Sep 2021, we find applications of the SPDE-approach to Gaussian fields in astronomy (Levis et al., 2021), health (Mannseth et al., 2021; Scott, 2021; Moses et al., 2021; Bertozzi-Villa et al., 2021; Moraga et al., 2021; Asri and Benamirouche, 2021), engineering (Zhang et al., 2021), theory (Ghattas and Willcox, 2021; Sanz-Alonso and Yang, 2021a; Lang and Pereira, 2021; Bolin and Wallin, 2021), environmetrics (Roksvåg et al., 2021; Roksvåg et al., 2021; Beloconi et al., 2021; Vandeskog et al., 2021a; Wang and Zuo, 2021; Wright et al., 2021; Gómez-Catasús et al., 2021; Valente and Laurini, 2021b; Bleuel et al., 2021; Florêncio et al., 2021; Valente and Laurini, 2021a; Hough et al., 2021), econometrics (Morales and Laurini, 2021; Maynou et al., 2021), agronomy (Borges da Silva et al., 2021), ecology (Martino et al., 2021; Sicacha-Parada et al., 2021; Williamson et al., 2021; Bell et al., 2021; Humphreys et al., 2022; Xi et al., 2021; Fecchio et al., 2022), urban planning (Li, 2021), imaging (Aquino et al., 2021), modelling of forest fires (Taylor et al., 2021; Lindenmayer et al., 2022), fisheries (Babyn et al., 2021; van Woesik and Cacciapaglia, 2021; Jarvis et al., 2021; Cavieres et al., 2021; Monnahan et al., 2021; Berg et al., 2021; Breivik et al., 2021; Lee et al., 2021; Thorson et al., 2021; Griffiths and Lezama-Ochoa, 2021), dealing with barriers (Boman et al., 2021; Babyn et al., 2021; Martino et al., 2021; Vogel et al., 2021; Cendoya et al., 2021), and so on. The plan for the rest of this paper is as follows.
Role of tourism and hotel accommodation in house prices
2022, Annals of Tourism Research Empirical InsightsCitation Excerpt :Our data set consists of annual data from 2005 to 2018 for the 27 EU member countries - Austria, Belgium, Bulgaria, Cyprus, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain and Sweden. Our research follows numerus research into EU housing markets (Gupta, André, & Gil-Alana, 2015; Maynou, Monfort, Morley, & Ordóñez, 2021; Merikas, Merika, Laopodis, & Triantafyllou, 2012; Miles, 2020; Philiponnet & Turrini, 2017). Increasing linkages in trade, financial markets, economic policies and general economic conditions exist between these countries, which directly influence housing markets (Vansteenkiste & Hieber, 2009).
Housing price convergence: new evidence from Indian cities
2024, Journal of Economic StudiesExchange rate and housing affordability in OECD countries
2024, International Journal of Housing Markets and Analysis
Mercedes Monfort is Lecturer of Economics at the Jaume I University in Castellón. She holds a degree in Business Administration and Management and a MSc in Local Development Management. She received his doctorate in Economics from the Jaume I University with the university's Award for Excellence. She has published papers in Economic Modelling, Energy, Open Economies Review, Empirica, Emerging Markets, Finance and Trade, Plos One and in the International Journal of Economics and Finance among others. She has visited the University of Adelaide and the University of Bath.
Javier Ordóñez is Professor of Economics at the Jaume I University in Castellón. He is president of the Spanish Association of International Economics and Finance (AEEFI) and Director of the International Economics Institute (IEI) of the Jaume I University. He holds a degree in Business Administration and Management with an Award for Excellence, and did his MSc in Economics and Finance at the University of Warwick. He received his doctorate in Economics from the Jaume I University with the university's Award for Excellence and European Doctorate accreditation.
Dr Laia Maynou-Pujolras is a LSE Fellow in the Department of Health Policy and is a member of the Executive Board of the Spanish Health Economics Association. She joined the London School of Economics and Political Science (LSE) as a Research Officer in December 2016 working on a Health Foundation project with Professor Alistair McGuire and Dr Victoria Serra-Sastre. She is also an Associate Researcher at the Centre for Research in Health and Economics (CRES-UPF) at Pompeu Fabra University. Dr Maynou's research focuses on health economics, in particular technology diffusion and economics, hospital efficiency, Health Technology Assessment (HTA) and public policy evaluation. All these fields of research are analysed from an empirical perspective.
Dr Bruce Morley is an economics lecturer at the University of Bath, having previously been at Aberystwyth University. His research is mainly empirical in the areas of macroeconomics, finance and the environment. He has also conducted research into sports economics, especially cricket. He received his doctorate in economics from the University of Loughborough, which was on exchange rate economics. His teaching is mainly in the areas of macroeconomics and econometrics and he is a Fellow of the Higher Education Academy.