Did home ownership made things worse in the Great Recession?

I have complained several times already that house ownership should not be encouraged by public authorities, mainly because it prevents diversification of risk by households and because there is slim evidence at best that home owners are happier and better contributors to society. It also quite obvious that high ownership rates have contributed to make the last recession worse in the United States. A recent trio of papers studies this last point.

Silvio Rendon and Núria Quella show that higher homeownership rates fed by easy financing lead to higher unemployment rates. This is because homeowners have higher reservation wages through a wealth effect. They find that in the US this has increased the unemployment rate by an incredible 6 percentage points. You may also want to add to this that homeowners are less willing to move for a new job, further increasing the unemployment rate, something the model does not capture.

Ahmet Ali Taṣkin and Firat Yaman look at unemployment duration in the US and find that renter stay unemployed the shortest and homeowners the longest, especially those who do not carry mortgages. Following this result, facilitating home financing would lengthen a little unemployment spells and increase the unemployment rate, under the hypothesis that job losing rates are unaffected.

Stijn Baert, Freddy Heylen and Daan Isebaert show that the unemployment spell length depends on the housing tenure situation in Belgium. The homeowners with mortgages exit the fastest, those without mortgages the slowest and renters lie in between. Easier home financing would thus reduce the unemployment rate here, again assuming it does not affect the rate at which people lose jobs. Keep in mind that Belgium is unique in that unemployment insurance benefits can last forever.

Rent control and home ownership

A new and recent trend in the United States has been the decline in the homeownership rate. While I have mentioned before that homeownership (the "American Dream") is not necessarily a good thing, both privately and socially, it is heavily favored by government policies. And while obviously the recent housing debacle has reduced homeownership, the trend started before that. To understand why the trend is down, it may be of interest to understand how it went up.

Daniel Fetter looks at the period where the nationwide homeownership rate went up the fastest, World War II. Paradoxically, this was a period were home construction was actually severely restricted. Yet, the 10 percentage point increase (half the increase over the entire century) happened in the context of widespread rent control. Exploiting differences in rent reductions through control across cities, Fetter finds that a majority of the increase in the homeownership rate was indeed due to rent control. I suppose it was renters somehow coerced by their landlords to buy the home they lived in, with no alternatives available. Would this mean that imposing rent control now would reverse the decline in ownership? I doubt it, as the market has now segmented between owned homes and rented apartments, and they are not close substitutes for the most part. And you would not want rent control anyway.

Ethnic ghettos and unemployment

Both in Europe and the United States, minorities face significantly higher unemployment rates. In addition, they live in places that are farther from work than others, or at least their commuting options make it more difficult to get to work. Are the two linked? Obviously, if you do not live where the jobs are, unemployment gets more prevalent. But one could also move, and this may be more difficult for minorities, for various reasons. But before going there, one needs to determine how much of the unemployment rate is due to this spatial mismatch.

Laurent Gobillon, Peter Rupert and Etienne Wasmer pick up on a previous paper of the latter two, which I discussed here. In this spatial search-and-matching model, commuting time acts as a friction, but can only explain a fraction of the unemployment rate gap between "majorities" and "minorities". So other factors are clearly at play. The fact that minorities are de facto confined to particular areas certainly plays a role here.

How many mortgage defaults resulted from lofty expectations?

A housing bubble is sustained by expectations of further increases in house prices. There is strong suspicion that this is what happened during the US housing boom preceding the last crisis, and that these expectations have triggered excessive mortgage borrowing. Actually verifying that claim is not that straightforward, though, as one needs to find extensive household level data.

Steven Laufer found this for Los Angeles County with panel data that tracks a property and all its mortgages. He comes to the sad conclusion that only 30% of mortgage defaults there were a result of household level shocks. The rest is all about borrowing and mostly extracted additional cash excessively with the expectation that the loan-to-value ratio would be reduced as house prices continue to grow. When this did not materialize, massive defaults resulted. Using the estimation model, Laufer finds that could have been mostly avoided by imposing the 80% loan-to-value ratio. Although this would have lowered house prices by a considerable 14%, this would have reduced defaults by 28%, as small number given the price drop but a large one considering the number of defaults. And house prices in LA are too high anyway.

Why so much policy focus on home ownership?

Some have blamed the Community Reinvestment Act (CRA) for the too risky lending to US homeowners during the house price run-up. Actual evidence for this is hard to come by, though. In this previous post, I discuss that there was indeed more risk taken, but it is not clear whether that additional risk was priced in or not. And were we to blame CRA, it would show in banks giving loans to neighborhoods that should not have received them for economic reasons, only to satisfy CRA.

Patrick Bayer, Fernando Ferreira and Stephen Ross look at the history of mortgages that they can link to credit scores and demographic characteristics. They find that for the same credit score, blacks and Hispanics were much more likely to run into mortgage trouble. While the authors do not mention this, the CRA was clearly targeting neighborhoods with such populations, and banks had to lend more there to comply. This would indicate that there is at least some truth to the CRA blaming. The authors frame this result rather by writing that this is evidence that favoring homeownership is not a good way to reduce wealth disparities. I would agree, but also because owning a home is very poor diversification, especially when this is all the wealth you can have. And there is no evidence that homeownership is good anyway, to the contrary.

Foreclosure procedures last too long

The handling of foreclosures in the recent housing crisis in the US has been a serious disaster. The drop in household income made that many households could not service their mortgage obligations and had to default. In addition, the drop in house values meant that many mortgages were worth more that the house that serves as collateral. This encouraged owners to walk away from payments. The mass of defaults lead mortgage servicers to resort to automatic treatment of foreclosures, leading to many errors, in particular foreclosing houses that not at issue. The reaction of many US states was to require longer foreclosure delays, first to make sure procedures are properly followed, second to allow owners to renegotiate, recoup and still make payments. The latter did not work out, as reported here previously. Were these state interventions worth it, in the end?

Larry Cordell, Liang Geng, Laurie Goodman and Lidan Yang use extensive databases of foreclosure procedures to quantify the lengthening of foreclosure delays and what this has cost. An important consideration is how foreclosures happen across states. In some, courts need to get involved (judicial states), in others the procedures only follow the stipulations of the mortgage contract (statutory states). In the former, the length of the procedure went from 26 to 44 months, in the latter from 16 to 22 months. During all this time, both parties are left in limbo, owners have incentives not to pay at all and neglect house maintenance, and lenders get no return on investment and may try to find whatever means to get any money out of the house, including reselling the mortgage. Also, there are externalities on neighborhoods as they get blighted. This is costly. The cost went up from 8% to 12% oh house value in statutory states, while it is from 17% to 30% in judicial states. These costs are estimated by adding unpaid property taxes, excess depreciation and unpaid insurance. This is thus the cost to the mortgage servicer, and does not even include capital costs. For a cost to society, one would also have to add the impact on other property values and deduct the fact that owners are living for free in these homes. There is no doubt the costs are considerable.

US local lenders knew about the housing bubble

Among the main culprits of the recent boom and bust in the US housing market that have been identified, the lenders and their excessive pushing of mortgages have been prominently featured. As pushing mortgages to people who cannot afford it seems to be a losing proposition, some pretty weird incentives must be in place for this to work. In other words, their must be some pretty sophisticated scheme in the lending business for some to make a gain from this. Local lenders, though, are not that sophisticated, as they handle most of the steps in the lending process themselves. As it turns out, they saw the debacle coming and pretty much got out of lending mortgages as soon as they felt things were getting excessive.

This is what you can conclude from the the analysis of Kristle Romero Cortés. She finds that where home prices where rising the fastest, the share of local lenders on the mortgage market was declining the fastest. In the subsequent bust, home prices were declining less in areas where local lenders were more present during loan origination. And looking at California only, foreclosures rates were lower where local lending was more prevalent. And all these results are even stronger where local lenders did not securitize the mortgages. What this shows is that there is still good value in homegrown lending, where the lender knows the markets intimately and knows to back off where things are getting dicey. Or, this can also be an indictment of the national mortgage chains like Countrywide Financial that were lending without thinking or had twisted incentives in place.

How to contain housing bubbles

One cannot deny that housing bubbles can lead to nasty consequences, as shown in Japan, the US and Spain, for example. What can a policy maker do? Foremost, it is difficult to identify bubbles on the spot, and even in hindsight. Also, imagine the backlash when the government intervenes to rein in a booming industry. One thus needs a policy rule that kicks in automatically, or some policy that just reduces the volatility of prices. Natural candidates are transaction taxes and capital gains taxes for houses, and such taxes have been proposed not only for real estate markets, but also for financial markets in general (for instance, the Tobin tax).

Nicole Aregger, Martin Brown and Enzo Rossi exploits differences in such taxes across Swiss administrative divisions, as well as corresponding house price indices, to identify whether such taxes work. They do not. The capital gain taxes, especially those that apply to short-term gains, amplify prices movements. Why? Likely because house owners are reluctant to put their house on the market if such penalties apply, which drives prices even higher. As for transaction taxes, they have no impact whatsoever on price fluctuations. Thus such tax policies do not work, unlike you are willing to subsidize capital gains...

Wind farms: Is NIMBY justified?

Personally, I find wind farms to be beauties. They are very elegant and even in large numbers the wind mills offer a sumptuous spectacle. But not everyone shares this point of view, and especially the windmills' neighbors are railing against their visual impact, their shade and their noise. If it is so bad, it must then have an impact on property values. It turns out that it is actually quite difficult to find a significant effect. So is all this NIMBY talk a big fuss for nothing?

Yasin Sunak and Reinhard Madlener point out that the small literature on the topic uses OLS estimates of hedonic models. That is how property valuation studies are usually done, but they find that results can change if one uses geographically weighted regressions, and one takes into account spatial autocorrelation. The analysis is performed for a particular area of Western Germany. It would have been more convincing if this study would have overturned the results of an earlier one. The study also excludes re-sales, for technical reasons. But of course, one could also concentrate on re-sales to see the impact of the windmill proposal and then its construction. Still, their specification can tease out some interesting results. OLS estimates reveal a negative impact that becomes more complex with a more general specification: the negative impact is much stronger in one city compared to the other, all else equal. Some areas even benefit. It would be interesting to see whether this is from internal migration away from the windmills.

On the difficulty of calculating the cost of living

Quality of life indexes are popular in the press. But they are not that easy to compute. While one can easily measure how much one has to work for, say, a loaf of bread, quality of life needs to consider a broader basket of goods. Now, you need to define that basket, which may be very different across locations (and across time if the horizon is long enough). It becomes even more difficult if some of the goods are location specific, such as housing.

John Winters points out that housing rents and house values are typically used for this kind of exercise. But house values can be very misleading, as most of the price of a house contains future services and their price, not current ones. Rents, in contrast, only contain the value of current services. An additional problem is that depending on the location the rental and sale markets may be very segregated and thick. Indeed, rentals are typically small and of lower quality. Winters compares rents and house values for US metropolitan areas and finds that they correlate well, but house values exhibit wide dispersion, making them indeed less reliable. He recommends using only rents, even if few rents are available and may not be necessarily representative of the housing stock and market.

Foreclosure crisis: it is not about irrationality and sneaky bankers

Why has there been a foreclosure crisis in the United States? Two popular explanations are that 1) evil mortgage brokers forced people to take mortgages they could not possibly honor, and 2) those taking the mortgages did not understand what they were doing. As an economist who insists on logic and rationality, it is difficult to adopt these points of view, except that a point could be made about perverse incentives in the mortgage industry where the risk is masked and pushed unto unsuspecting people. But were mortgage holders really that stupid to think they would be able to make it? After all, I know several PhD economists who are still underwater, and they do not look stupid to me.

Christopher Foote, Kristopher Gerardi and Paul Willen come to the rescue. They argue that market participants made perfectly rational decisions given the information they had a the time, and in particular given the beliefs they had. The latter turned out to be too optimistic in retrospect though. Foote, Gerardi and Willen come to this conclusion with an interesting data analysis. They draw 12 "facts" that together contradict the popular explanations. Foremost, it does not appear that there is any correlation between exploding mortgage rates and mounting foreclosures. Also, even borrowers with spotty credit have had a remarkably good repayment history. One should thus not conclude that mortgages were designed to fail. Furthermore, all the instruments and innovations in the mortgage industry were introduced well before the past decade, and there was no significant regulatory change. Market participants knew what they were doing, had plenty of information and understood the risks. They were too optimistic though. Finally, no top-rated mortgage-backed security turned out to be toxic. The same cannot be said about similar bond-based securities.

All in all, there was nothing really wrong with the mortgage market apart from being too optimistic. In other words, there was a bubble, which can be a perfectly rational outcome. So there. But we still need to better cope with the eventuality of a bubble.

Seasonality in house prices

There is a marked seasonal cycle in many housing markets. Sale volumes and house prices are significantly higher in the Summer and lower in the Winter. Evidently there should be some arbitrage, by selling high and buying low and renting in between for those who are genuinely moving or simply holding on to real estate for speculators. Possibly, the transaction costs are too high for this to happen. Or maybe the market for houses is not fluid enough for price not to cycle in a predictable way.

Cemil Selcuk picks up on this second idea and builds a search model where the supply is smaller in the Winter in the sense that the probability of finding an appropriate house is lower. As a result, there are fewer successful matches in the Winter, and they happen with a lower price because of the discount cost of waiting for better opportunities in the Summer and because the matches in the Winter are of lower quality. This is a rather trivial theoretical result, and it would be nice to know whether it approaches quantitatively the seasonal differences that are observed.

The housing bubble: fooled by efficiency

In retrospect, the large rise in housing prices before 2007 looks suspect, many would even call this a bubble. But when you observe it in real time, it is much more difficult to judge whether house price inflation is excessive, although economists have been calling for it. A characteristic of many bubbles is that expect prices to increases forever, while it clearly cannot be true, at least in this magnitude. So somehow people are fooled.

Brian Peterson rationalizes this in a search model where the search frictions allow prices to deviate from their fundamental. The innovation of this model is that people are assumed to believe that price are not affected by the search frictions: they think markets are fully efficient.This means that instead of bargaining over the house price level, buyers and sellers bargain over house price increases. One interesting consequence of this is that turn-over volume and price inflation are then positively correlated. And once you quantify the model, about 70% of he price run-up can be explained by this "foolish" behavior.

House prices and consumption: how model specification matters

What is the impact of a decrease of house prices on consumption? If you go by the wealth hypothesis, one should see that home owners would reduce their consumption significantly because their wealth has suffered, while young renters are barely affected. According to the common factor hypothesis, both agent classes should respond in the same way, because they respond to common factors, for example future income prospects. It should be easy to test one hypothesis against the other and settle this. It turns out that using the exact same dataset, John Campbell and João Cocco could not reject the first, while Orazio Attanasio, Laura Blow, Robert Hamilton and Andrew Leicester could not reject the second. Well, that is embarrassing.

Annalisa Cristini and Almudena Sevilla Sanz replicate these results and show that because the two models are not nested, they cannot test against each other. Worse, it all boils down to the specification: estimate an Euler equation (which uses consumption growth) and the wealth hypothesis wins; estimate a consumption function, with consumption in levels, and the common factors hypothesis wins. So I am afraid it is not sufficient to estimate a single equation, one needs to estimate the whole structural model jointly.

Why more bad mortgages? Too much reliance on credit scores

The current financial crisis is at least partially blamed on lax lending practices in the US mortgage industry. More mortgages were provided to less credit-worthy individuals with smaller down-payments than ever before, until this house of cards fell apart. Of course, this is not the whole story, but at least there is some partial truth to it, right? Now I am not so sure.

Indeed, Geetesh Bhardwaj and Rajdeep Sengupta look at a large fraction of the sub-prime mortgages originated from 2000 to 2006. And they find that the credit-worthiness of their holders, as measured by the FICO score, actually increased (and more so than the general population). How could this be possible? One hypothesis is that mortgage issuers have gradually relied more and more on simple metrics they could enter into some software instead on analyzing other details on an application file. And if you end up relying on a single criterion, the selected applicant will look much better according to this criterion. But if this criterion is not well correlated with actual credit-worthiness and relevant information is neglected, your loan pool becomes more risky.

Near rational agents and house price booms

House price run-ups, especially when they appear excessive, are difficult to explain. It is it even more difficult to explain how they are not coordinated across countries in a globalized world. Indeed, right now house prices are severely depressed in the United States, while you can have strong suspicions of bubbles in China, Norway and Switzerland. Bubbles are substantial deviations from fundamentals that could be due to some deviations from rationality or herd behavior, or both. But "rationalizing" this is a major challenge because of the apparent randomness of the occurrence of such house price booms.

Klaus Adam, Pei Kuang and Albert Marcet think they have a way to explain this using the concept of internally rationally agent. Such a agent, like the economist, does not know the true process of prices but tries to infer it from past observation using Bayes' rule. The belief about prices then becomes part of the state space and leads to some sort of path dependence. With shocks that are not perfectly correlated, it is then possible for different countries to experience different paths for house prices.

Teenage achievement and the house price bubble

The general economic context of where and when you grow up matters. Think, for example, of those raised during the Great Depression in the US or World War II in Europe who are likely to be very careful with their spending, never through anything away and finish their plates. In this regard, what should we expect from those reaching adulthood in the past years?

Daniel Cooper and María José Luengo-Prado study the impact on teenagers of the house price boom before the current crisis in the United States on educational outcomes. Using the Panel Study of Income Dynamics (PSID), they find that a 1% higher house price at age 17 leads to a 0.8% higher income as adult if the parents owned the home, 1.2% lower if they were tenants, after conditioning for socio-economic characteristics. These are big numbers. They can be justified by the observation that higher house prices allows more collateral to borrow for education. Indeed households with a below median non-housing wealth saw even a 1.6% boost in their child's future income. To explain the impact on tenants, I suppose one can explain it with higher tuition in reaction to larger loans, which tenants cannot afford as well.

The consequences from the recent house price crash are daunting in this context. And given that state are disengaging themselves from financing their public colleges, leading to even higher tuition, the outlook is even worse.

Compartmentalized thinking in personal finances

Even before the crisis hit in the United States, there was talk about how foolish it is to get balloon mortgages, with low teaser rates for a few years. Yet people where going for them, either because they had expectations of strong income growth, or because they were time inconsistent or very impatient. Or people do not understand the true cost of the loan.

Johan Almenberg and Artashes Karapetyan document a phenomenon that is in some ways similar in Sweden. Mortgage interest is deductible from taxes for personal loans, but not when a co=op takes a loan. Yet people seem to favor financial situations that shift debt from personal to co-op loans. On average, the equivalent of US$540 a year are left on table. This can be explained by what is termed salience of debt. People only care about the costs they directly see, and the interest payments of the co-op are not itemized in the fees. The authors survey co-op apartment owners on how they think about their finances. It turns out people a very aware of their personal finances, but completely ignorant of the co-op finances. They never considered the trade-off between personal and co-op debt. That last point may indicate that ignorance may be more important than salience, though. This is reinforced by the fact that market price do not seem to reflect the tax difference.

Economists did see the bubble coming

Economists have been lambasted for not alerting the public that a bubble was in the making in US real state, except for a few oddballs. Of course everyone is wiser in hindsight, but what did economists actually say? It never hurts to look at the facts.

Martha Starr analyzes statements in 24 California newspapers from 2002 to 2007. From 1998 to 2005, the state's house prices increased by more than 10% each year. This prompted the newspaper to run 379 stories with 688 statements by economists on house prices. Academics were clearly warning that house prices were not sustainable. Economists employed in the real-industry, however, were resolutely optimistic. What emerges is a mixed message that gave no guidance to the public, which was even reassured by positive messages from the Federal Reserve.

It is entirely possible opinions could have diverged based on the same evidence. But it seems more likely the professionals were not acting in good faith. They had everything to lose from predicting an end of house price growth. The media should have learned not to trust such biased speakers, yet they continue to be interviewed. Now as to why Greenspan and then Bernanke were so optimistic is beyond me. There speeches are definitively strategic and while they may have realized there was a problem, they may have tried to prevent a bubble from bursting too brutally. Then all the credit to them for trying. But one cannot postpone indefinitely a bubble from bursting, and they knew that.

Latin American home owners are happier, unlike US ones

There is a myth saying that owning a home makes people happier and leads them to contribute more to their community. In an earlier report, I pointed out that this idea is a myth for the US homeowner. What about elsewhere?

Inder Ruprah finds that Latin American house owners are indeed happier. This is obtained from a survey where people declare how happy they are, the reliance of which many researchers have called into question. But happiness studies slowly get more acceptance, especially when results are clear cut, like here. Of course, homeownership could be correlated with some unobservables that matter a lot for happiness, for example economic and social standing. There is a variable that could capture this in the regression, "Interviewer assessment of economic situation of the household," but I have no idea how reliable it is.

PS: The pdf file is 7.3MB large. It took me five attempts to download it. There are a few very simple graphs and histograms in the paper, in other words no reason to have such a large file, but for unnecessary front and back covers. But if the IADB is willing to waste bandwidth that way, especially as its target audience in Latin America may not necessarily enjoy fast internet.