劳动力市场变化对家庭抵押贷款的异质性影响

The heterogeneous impact of labormarket shifts onhousehold mortgage-taking Michele Cantarella,Ilja Kristian Kavonius Disclaimer:Thispaper should not be reported as representing the views of theEuropean Central Bank(ECB). The views expressed are those of the authors and do not necessarily reflect those of the ECB. Abstract This paper examines how structural change in labor markets affects householdcredit outcomes.Using a Shift-Share instrumental variable approach, we find thatoccupational shifts negatively influence mortgage holding for households facing fa-vorable job market conditions, such as stable employment and income growth.Ourresults, robust to alternative specifications, suggest that when both individual andeconomy-wide career prospects are favorable, the opportunity costs of settling downgrow accordingly. JEL codes: G51, J24, D84, O15Keywords:Structural change, Job security, Household credit Non-technical summary This paper investigates how structural changes in the labor market influence households’mortgage-taking. By analyzing data from the European Union Labour Force Survey (EU-LFS) and the European Union Statistics on Income and Living Conditions (EU-SILC), weaim to understand the relationship between job market conditions and mortgage-taking.The EU-LFS provides detailed information on occupational growth, while the EU-SILCoffers a rich set of variables at both the individual and household levels. The research reveals that households holding favorable positions in the job market areless likely to hold mortgages if their occupation has also been growing at an economy-widelevel. This effect is particularly noticeable over a five-year period compared to a one-yearperiod.The study highlights that while permanent job contracts and longer job tenuregenerally increase the likelihood of having a mortgage (about 2.7% and 0.4% more likely),the positive impact of job permanency on mortgage-taking diminishes if the occupation isgrowing (about -3.5%). Additionally, past and future income changes influence mortgage-taking behavior (about 1.5% and 2.3%).However, households experiencing and, mostimportantly, expecting positive income changes are less likely to take on mortgages if theiroccupation is also growing (about -2.2% and -4.3%). We propose a general model to measure mortgage-taking for individuals in variousoccupations across different years and countries.This model includes variables such asjob tenure, job permanency, and past and future income changes.By interacting thesevariables with occupational growth, the study aims to capture the heterogeneous effectsof labor market shifts on mortgage-taking.The model also controls for individual andhousehold characteristics, including gender, age, education, income, household size, anddegree of urbanization. To address endogeneity concerns, we use a Shift-Share instrumental variable strategy.This involves instrumenting occupation-level growth with the inner product of industry-occupation shares and industry growth. The underlying assumption is that technologicalshocks are exogenous and that industries differentially exposed to these shocks will shiftlabor demand away from medium-skilled jobs.To further ensure robustness, we employadditional approaches, such as controlling for equalized housing costs, using a leave-one-outapproach to purge local demand housing shocks, and measuring variations in labor supplyat the intensive margin through overtime hours. Our findings suggest that while job security and income growth typically facilitatemortgage-taking, the broader economic context and occupational growth can alter thisrelationship.Policymakers should consider these dynamics when designing interventionsto support household credit access. 1Introduction The past few decades have witnessed profound structural changes in labor markets, markedby the polarization of employment opportunities (Autor and Dorn, 2013). While existingresearch has explored the causes and consequences of these changes, much less attentionhas been paid to how these changes affect a household’s long-term decisions and, in moredetail, mortgage-taking. This omission is surprising because job polarization and access to housing have, re-spectively, been both identified as key drivers of income (Goos et al., 2009) and wealthinequality (Kuhn et al., 2020). Understanding whether (and for whom) shifts in the demandfor skills impede or facilitate mortgage access is therefore essential for both distributionaland macro-prudential policy. This impact might in fact be highly non-uniform. Job polarization is usually studied asan economy-wide phenomenon, but in fact, it also has a personal dimension: an individual’sjob security does not always move in step with the overall fortunes of their occupation. Asa result, we cannot assume that mortgage-taking responds in the same way when macro-level occupational growth and micro-level job prospects diverge. One could imagine thatstrong aggregate growt