The detrimental effects of homeownership on the Belgian labour market Daan Isebaert With: Freddy Heylen, Carine Smolders
Homeownership rate in Belgium 80% 70% 60% 50% 40% 30% 20% 10% 0% 1947 1954 1961 1968 1975 1982 1989 1996 2003 Source: volkstellingen1947/70/81/91, SEE2001, EU-SILC2009
IRL ESP FIN BEL UK PRT DNK CHE AUT FRA NLD DEU Homeownership rate across European countries in the period 1994-2001 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Source: de Graaff, T., van Leuvensteijn, M., van Ewijk, C., 2009. Homeownership, Social Renting, en Labour Mobility across Europe. In van Ewijk, C., van Leuvensteijn, M. (Eds.). Homeownership & the Labour Market in Europe (53-81). New York, Oxford University Press.
De Oswald hypothesis Sp. B Fin. Irl. BRD CH Ndl. Source: Oswald A.J., 1999, The housing market and Europe s unemployment: A non-technical paper, mimeo, University of Warwick.
BEL GRI FRA ESP ITA DEU AUT NLD IRL FIN SWE UK DNK Transaction costs on buyer, 2009 16% 14% 12% 10% 8% 6% 4% 2% 0% Source: Caldera Sánchez A. and D. Andrews, 2011, To move or not to move: What drives residential mobility rates in the OECD?, OECD Economics Department Working Papers, n 846.
The Oswald hypothesis Higher moving costs Higher Reservation wage + Lower search intensity Lower labour supply and lower employment Three theorems: A higher rate of homeownership causes lower employment rates Central mechanism = restricted mobility of homeowners Homeowners have longer unemployment spells
Houses and/or jobs: Ownership and the labour market in Belgian districts Isebaert, D., Heylen, F., Smolders, C., 2013. Houses and/or jobs: ownership and the labour market in Belgian districts. Regional Studies, forthcoming. Research question: Does the rate of homeownership have a negative effect on the employment rate in Belgian districts? i = 42 districts t = 1970, 1977, 1981, 1991, 2001, 2005 Empl it = g 1 OWN it + g 2 Schooling it + g 3 log(wage it /q it ) + others (demography, fixed effects, time dummies) Homeownership rate is endogenous, as are schooling and wage gap. Supplementary estimations analyse whether other factors determine the size of the Oswald effect (g 1 ).
Houses and/or jobs: Ownership and the labour market in Belgian districts Conclusions: We confirm the Oswald hypothesis: An increase of the homeownership rate with 1 percentage point, involves a significant drop in the employment rate of 0.35 percentage points. The effect is stronger when the fraction of high-skilled in a district is lower. (cf. Dohmen, 2005, Regional Science and Urban Economics)
Housing tenure and geographical mobility in Belgium Isebaert, D., 2013. Housing tenure and geographical mobility in Belgium. Faculty of Economics and Business Administration Working Paper, Ghent University, n 2013/855. Research question: Are homeowners less mobile? Microeconomic data with T = 1994-2009 PSBH(n=4556) en EU-SILC(n=7209) We control for various family and environment characteristics. We make a double subdivision in housing tenure choice: Private market tenants and social tenants; Owners with and without a mortgage. Probit estimation controlling for unobserved heterogeneity (Mundlak approach) en state-dependence (Wooldridge approach)
Housing tenure and geographical mobility in Belgium Conclusions: Tenants are more mobile than homeowners: the underlying assumption of Oswald s hypothesis is verified. Social tenants are less mobile than private market tenants. Homeowners with a mortgage are less mobile than outright homeowners.
Does homeownership lead to longer unemployment spells? The role of mortgage payments Baert, S., Heylen, F., Isebaert, D., 2014. Does homeownership lead to longer unemployment spells? The role of mortgage payments. De Economist, 162 (3), 263-286. Research question: Do homeowners have longer unemployment spells? EU-SILC dataset 1013 unemployment spells We follow the theoretical model in Rouwendal & Nijkamp (2010) with a supplementary subdivision between homeowners with a mortgage and outright homeowners Theory: Not only mobility costs affect search intensity. Housing costs do as well. Empirical ambiguity
Survival probability Does homeownership lead to longer unemployment spells? The role of mortgage payments 1 Kaplan-Meier estimates 0,75 0,50 0,25 0 0 10 20 30 40 Homeowners Tenants Months
Survival probability Does homeownership lead to longer unemployment spells? The role of mortgage payments 1 0,75 0,50 0,25 Outright owners Tenants Mortgage 0 0 10 20 30 40 Months
Does homeownership lead to longer unemployment spells? The role of mortgage payments Research question: Do homeowners have longer unemployment spells? Simultaneous estimation: Mixed proportional hazard model (unemployment duration) Mixed multinomial logit model (Housing tenure choice) We use instrumental variables to econometrically identify the housing tenure effect
Does homeownership lead to longer unemployment spells? The role of mortgage payments Conclusions: There is no significant difference in unemployment duration between homeowners and tenants The Oswald effect does not appear at the micro level Unemployment durations differs significantly: Outright owners > Tenants > Owners with a mortgage Housing costs are a crucial factor and might explain the different results in the existing literature.
Homeownership and rental rates EU, 2012 Source: European Commission, 2014. Tax Reforms in EU Member States 2014 - Tax policy challenges for economic growth and fiscal sustainability. Taxation Papers, 48.
Conclusions Higher moving costs Higher Reservation wage + Lower search intensity Lower labour supply and lower employment Three theorems: A higher rate of homeownership causes lower employment rates Central mechanism = restricted mobility of homeowners Homeowners have longer unemployment spells Micro- and macroeconomic research on one and the same country reveal an apparent contradiction
Conclusions This puzzle can be disentangled by looking closer into negative external effects of homeownership: Homeowners commute over longer distances. Promotion of homeownership might undermine the development of a well-functioning rental market NIMBY-effects Displacement of other workers in the same region, caused by high search intensity and low reservation wages of homeowners for local jobs
Policy Implications Gov. budget for housing Supporting homeownership 15% Supporting tenants 85% Source: Heylen, K., Winters, S., 2012. De verdeling van de subsidies op vlak van wonen in Vlaanderen. Steunpunt Ruimte en Wonen, 95p. Property acquisition is being intensely stimulated. Housing policy counteracts tenure neutrality.
Policy Implications Homeownership promotion per income quintile 0-20% 21-40% 41-60% 61-80% 81-100% Source: Heylen, K., Winters, S., 2012. De verdeling van de subsidies op vlak van wonen in Vlaanderen. Steunpunt Ruimte en Wonen, 95p.
Policy Implications How to avoid the Oswald effect: 1. Less homeowners 2. More mobile homeowners
Policy Implications How to avoid the Oswald effect: 1. Less homeowners Facilitate supply in qualitative rental housing Restore tenure neutrality by reducing excessive stimuli Cf. tax deductability of mortgage payments
Policy Implications How to avoid the Oswald effect: 2. More mobile homeowners Create relocation incentives as a complementary tool in the context of activation programs Counter the negative external effects caused by traffic congestion (more efficient public transport, Pigovian tax for commuter traffic) Directly decrease the cost of residential mobility: Lower transaction costs or expand its portability Shift to recurrent immovable property taxes or phase out the system of tax deduction on mortgage payments
Daan.Isebaert@ugent.be Isebaert, D., Heylen, F., Smolders, C., 2013. Houses and/or jobs: ownership and the labour market in Belgian districts. Regional Studies, forthcoming. Isebaert, D., 2013. Housing tenure and geographical mobility in Belgium. Faculty of Economics and Business Administration Working Paper, Ghent University, n 2013/855. Baert, S., Heylen, F., Isebaert, D., 2014. Does homeownership lead to longer unemployment spells? The role of mortgage payments. De Economist, 162 (3), 263-286. Policy recommendations: Isebaert, D., 2013. Housing and labour market performance. Manuscript submitted to the Faculty of Economics and Business Administration of Ghent University in fulfilment of the Requirements for the Degree of Doctor in Economics. Isebaert, D., 2014. The Belgian housing policy, its adverse effect on labour mobility and the negative externalities of homeownership. Documentatieblad - Bulletin de Documentation, Federal Public Service Finance, forthcoming.