There are two important questions in the economy today that may be related: 1) is negative equity causing a decrease in geographic mobility? and 2) why is the Beveridge Curve breaking down? Wait! Don’t stop reading yet, I can explain it in non-econo-jargon, I promise.
House prices have obviously fallen a lot since the peak of the bubble, and this has left many homeowners “underwater”, so to speak, meaning they owe more on their mortgages than their house is worth. This is potentially causing a big decrease in people’s willingness to move, which includes moving for a job. This, in turn, is potentially causing higher unemployment by preventing people from moving away from places where their labor isn’t demanded to places where it is. The question is, how big of a deal is this?
The second question relates to the Beveridge Curve, which shows the relationship between the unemployment rate and the number of job vacancies. The idea is that when unemployment is high, job vacancies should be low, and vice versa. If people are having a hard time finding work, then employers shouldn’t be having a hard time finding workers, since there are plenty of unemployed people looking for work. However, as the graph below shows this relationship has broken down somewhat over the recent recession. This is suggestive of some sort of friction in the labor markets that is preventing employers from finding hires among the vast numbers of unemployed.
A recent study investigates whether house price induced immobility is causing the breakdown in the Beveridge Curve. This is an intuitive and plausible mechanism. House prices fall, homeowners are underwater and can’t move to jobs, so there are unemployed people in one area and job vacancies in another, and negative equity prevents them from moving to those jobs. Supporting their hypothesis, the authors cite a 2010 study by Ferreira, Gyourko and Tracy which found that having negative equity reduced the probability that a homeowner would move by 35%.
The study uses a structural VAR (which is a big regression with multiple dependent variables) to estimate the dynamic relationship between 3 housing market variables and 2 labor market variables. They find that their model predicts 30% of the increase in unemployment observed during the great recession, and in generates a flat or slightly upward sloping Beveridge Curve as observed in reality. The authors are then able to run a counterfactual where they removing shocks to housing preferences, meaning that they see what would have happened to unemployment without the housing bubble. They find that the unemployment rates are in line with the Beveridge Curve, and thus conclude that underwater homeowners can explain the breakdown of the Beveridge Curve. The graphs below show the counterfactual unemployment rates in two scenarios: a high leverage economy, and a low leverage economy.
The authors do caution that their model is a simple one, but the results suggest that housing markets are holding back labor markets. What is also important to note however, is that the model only explains 30% of the increase in unemployment during the Great Recession, leaving plenty of room for aggregate demand led unemployment.