Essays on household behavior in the mortgage market and house price dynamics.
The housing market depends critically on credit market conditions and the link between housing and credit has recently attracted considerable interest in empirical papers. The strong interlinkage between both markets raises numerous question: How do borrowers choose between different mortgage products? How do households make reinvestment decision for their largest single component of household wealth? How are housing and mortgage markets interlinked? What are long-run fundamentals in house prices? In this dissertation we study these household choices on the housing and mortgage market and the resulting market outcomes. We now briefly summarize the research in the different chapters.
Even after the crisis, house prices are overvalued according to price-to-income ratios in many countries. As unsustainable house prices may raise concerns about financial stability, it is important to understand the factors behind the evolution of house prices. In the first chapter, we introduce the borrower’s ability to pay (ATP) as a new long-run house price fundamental. ATP represents the borrowing capacity of a household, given a fixed fraction of income that goes to mortgage payments. An important contribution is that ATP also incorporates specific mortgage characteristics and the mortgage interest deduction (MID). Therefore, ATP takes into account the effect of changes in interest rates, fiscal rules and mortgage product innovation.
We apply the model to the United States of America, United Kingdom, Belgium, the Netherlands, Sweden, Norway, Finland and Denmark. Conventional and panel cointegration tests indicate that house prices and our measure of ATP adjusted for the MID and mortgage characteristics are cointegrated in a statistically significant fraction of the countries. When the mortgage product innovations and the MID are excluded from the measure of ATP, the null hypothesis of no cointegration in all countries is not rejected. This confirms our hypothesis that it is necessary to model the changes in mortgage characteristics and tax rules. Other evidence indicates that the elasticity of house prices with respect to changes in ATP is close to one and Granger-causality tests indicate that the measure of ATP Granger causes house prices. Evidence in the other direction is much weaker. Finally, we test the ATP model against more conventional user cost models and find evidence that the speed-of-adjustment from disequilibrium is faster in the ATP model. Therefore, the ATP framework provides an intuitive alternative to standard house price models.
As the first chapter indicates that mortgage characteristics have important implications for house price evolutions, we study the mortgage choices that individuals make in the subsequent chapters. In the second chapter, we study the choice between fixed rate mortgages (FRM) and adjustable rate mortgages (ARM) using data from Belgium and its Flemish region. For many households, the choice between fixed and adjustable rate mortgages is probably the most important choice with respect to the mortgage contract. In the aggregate, the individual mortgage choices have important policy implications for the transmission mechanism of monetary policy through the effect of mortgage payments on household budgets. The dominant share of adjustable rate mortgages in the UK was mentioned as an important reason by the United Kingdom’s economics and finance ministry (HM Treasury 2003a, 2003b) for opting out in the third stage of the Economic Monetary Union to introduce the euro.
Notwithstanding the different ways monetary policy shocks are propagated, there still remains large cross-country variation in mortgage shares within the Eurozone. Whereas the variation in market share over time is small in most countries, the structure of the Belgian mortgage market varied significantly during the last decade. In this paper we seek to understand this variation in the mortgage structure over time. The results indicate that households base their decision on initial mortgage payments and do not consider expected changes in ARM rates. This results in a remarkable large variation in mortgage structure over time in comparison to other countries. Borrowers who expect to move in the near future are more likely to choose the FRM. As all mortgages are portable by law, mobile borrowers prefer to reduce variability in mortgage payments, which was not observed in previous empirical studies.
Survey evidence suggests that many American and European consumers do not spend a lot of time comparing mortgage products. Indeed, many American consumers spend the same amount of time choosing where they want to go on vacation and more time researching the next car they want to buy (Finkelstein, 2010). In Europe, mortgage shopping does not appear to be fun either. Survey evidence from the Eurobarometer suggests that 21 percent of respondents in the 27 member states of the European Union did not compare mortgage products and took the first mortgage that they looked at.
In the third chapter we show, however, that mortgage shopping is associated with a substantial monetary payoff. Therefore, we bring a unique dataset in the discussion from a website where borrowers (not the lenders) can post their complete set of received mortgage rate offers. A borrower who shops for 5 mortgage offers, is able to save 7000 euro in net present value on average. The results indicate that the decision to continue shopping is not much affected by the savings from mortgage shopping, but rather by the time spent comparing prices. Indeed, the time spent between the first and fifth offer is only 13 days for the median borrower. The potential savings suggest sub-optimal mortgage shopping as the opportunity cost of time to renegotiate additional quotes is unlikely to be that high.
In the fourth chapter we study how the interaction between house prices and mortgage debt may influence reinvestment and mobility decisions. The severe drop in US house prices in 2006-2007, led to a large increase in the number of households facing an outstanding mortgage debt that exceeded the value of their house. The increase in negative equity has increased concern that underwater households are less mobile and decrease housing reinvestment. The large costs associated with the reinvestment and move decisions urge households to consider future expectations. Although this forward-looking behavior may affect reinvestment and mobility decisions, previous studies have often ignored to take them into account. Therefore, we bridge the gap between the reinvestment, mobility and dynamic literature and estimate a dynamic discrete choice model in which an owner-occupier can either reinvest, move or do nothing. Our findings indicate that negative equity reduces reinvestment and mobility among young households, but not among senior citizens. Furthermore, negative equity decreases reinvestment in non-recourse states, but not in recourse states.