Barney Hartman-Glaser, Benjamin Hébert,
Based on: The Insurance Is the Lemon: Failing to Index Contracts DOI: https://doi.org/10.1111/jofi.12856
As house prices fell and unemployment rose during the great financial crisis of 2008, many homeowners defaulted on their mortgages. The result was a massive wave of home foreclosures, causing economic pain for homeowners, banks, bondholders, and local and national governments. Many commenters observed that the foreclosure crisis was not an inevitable consequence of the fall in home prices. The original mortgage loans could have been designed to automatically reduce borrowers' principal balances in response to declines in home prices. Proposals for indexed mortgages of this kind pre-date the great financial crisis and have been resurrected as a way of avoiding similar crises in the future. Recently, the growth in home prices has started to slow after almost a decade of appreciation at the same time that interest rates are rising. Indexed mortgages would help dampen the effect that falling house prices have on existing homeowners.
The advantages of indexed mortgages are clear: they insure the borrower. Insurance, of course, comes at a cost. To compensate lenders for providing this insurance, borrowers must either pay higher interest rates or share some of their house price appreciation with lenders. Standard economic analysis predicts that borrowers and lenders should prefer this kind of risk-sharing arrangement. For many borrowers, their home is a large part of their wealth, and the possibility of house prices falling in their region is a significant source of risk. A well-diversified lender is better positioned to absorb this risk, just as insurance companies are better able to bear fire or disaster risks than homeowners. Yet, despite the advantages of indexed mortgages, such mortgages were and are almost nonexistent in practice.
Economists have several standard explanations for risk-sharing failures, but these explanations are not well-suited to explaining the lack of indexation in mortgages. Sometimes, risk sharing requires complex contracts, and in other cases requires costly information gathering. However, indexed mortgages need not be complicated, and the information necessary to implement them is widely available. Indeed, there are many publically available house price indices at a local level, such as the Zillow Home Value Index or the Case-Shiller Index. Risk-sharing might also change the borrower's behavior in a way that adversely affects lenders (what economists call "moral hazard"). For example, if a borrower has little stake in the home's sale price, she will not put in the effort required to sell at the highest price possible or engage in maintenance to preserve the home’s value. Indexed mortgages avoid this type of moral hazard problem because they specify payments contingent on a house price index beyond the borrower's control. A new explanation is required.
Despite its benefits, indexation is not a panacea. A borrower with an indexed mortgage faces the possibility that the index goes up while her home price goes down; from her perspective, this is the worst possible scenario. This type of risk is called basis risk and is more likely with a low-quality index, that is, one that is not particularly related to the value of the borrower's own home or the housing market more generally, than a high-quality index. Basis risk can greatly reduce the insurance benefit a borrower receives from an indexed mortgage.
In contrast, a lender might prefer to use a low-quality index. When house prices fall, a lender’s existing borrowers are less likely to repay their loans. If a lender makes an indexed mortgage loan to a new borrower using a high-quality index, that new borrower’s payments will fall precisely when the lender’s existing loans are not being repaid. That is, with a high-quality index, the risks of an indexed mortgage loan and the lender’s existing mortgage portfolio are correlated. In contrast, with a low-quality index, these risks are uncorrelated, which explains why a lender would prefer, all else equal, to make an indexed mortgage loan using a low-quality index.
Therein lies the problem. Most borrowers know little about the quality—or lack thereof—of house price indices. They do, however, know that lenders are better informed than they are. Suppose a lender proposed an indexed mortgage and explained that the borrower must pay higher interest rates or give up some home appreciation in exchange for downside protection. In that case, the borrower might wonder: Is this index insuring me against risk, or am I paying for useless insurance? That is, is the insurance a lemon?
Upon asking herself this question, the borrower will realize that it is in the lender's interest to offer expensive but useless insurance. If other lenders are offering standard mortgages, the borrower might reject the offer of the indexed mortgage and go with another lender. Lenders will anticipate this rejection and rationally only offer a standard, non-indexed mortgage. Thus, despite the advantages of indexation, the borrower and lender will settle on a standard mortgage.
This outcome arises from the borrower's belief that the lender is better informed, a form of what economists call "asymmetric information." In the classic analysis of Akerlof, asymmetric information, otherwise known as the lemons problem, can lead to a complete market breakdown. In the mortgage market, asymmetric information about the quality of house price indices leads to a breakdown in the use of those indices—which is to say, the market will consist of only standard, nonindexed mortgages. Borrowers and lenders cannot share risk because lenders cannot convince borrowers that they have a good way to measure it.
This risk-sharing failure is not the only possible market outcome. No single lender or small group of lenders is willing to introduce indexed mortgages on their own for fear of losing business to other lenders. However, if all lenders could coordinate (or were coordinated by the government) to offer a standardized indexed mortgage product, borrowers would be able to compare interest rates across lenders. This kind of competition would give borrowers confidence that they were getting a fair price for the insurance provided and promote the development of indexed mortgages. Coordinated action has the potential to overcome asymmetric information, unlocking the risk-sharing benefits of indexation and perhaps avoiding future financial crises.
Associate Professor of Finance
Anderson Graduate School of Management
University of California, Los Angeles
Associate Professor of Finance
Stanford Graduate School of Business