How to avoid fire sales during a financial crisis

During financial crises, fire sales or forced asset sales are mentioned to aggravate financial fragility because they deplete the balance sheets of market participants. But why do not market participants take actions to mitigate fire sales? And are there any financial structures that can avoid or reduce forced asset sales?

Defining fire sales

Fire sales occur if in a forced sale, assets fetch prices below their fundamental values (Shleifer and Vishny 1992). Price dislocations associated with forced asset sales generate externalities that feedback on asset values and cause price-default spirals, especially in illiquid markets with collateralised lending. If asset liquidations in these markets trigger downward spirals in asset prices and impair the balance sheets of other market participants, an important question is why lenders do not take actions to avoid collateral liquidation. As Shleifer and Vishny (2011) also note, surprisingly, the theoretical and empirical literature on this question is scant.

When are fire sales less likely?

In a recent paper (Favara and Giannetti 2015), we take up the challenge of exploring whether some lenders have incentives to internalise the negative spillovers created by forced asset sales and are, therefore, less prone to liquidate. We argue that lenders that hold a large proportion of the outstanding collateralised debt internalise the feedback effects of liquidation decisions on collateral values and may be more inclined to renegotiate to avoid price-default spirals. Using data on foreclosures and house prices during the 2007-2010 US housing crisis, we find evidence that such incentives are at work and are economically significant.

The 2007-2010 US housing prices collapse has been followed by a dramatic increase in mortgage defaults, which have often led to foreclosures. Empirical evidence shows that foreclosures are associated with price declines of neighbouring houses (Campbell et al. 2011), either because poor maintenance of foreclosed properties affects the quality of nearby houses (Harding et al. 2009), or because foreclosures increase the supply of homes in illiquid markets (Anenberg and Kung 2013). In this way, foreclosures create a negative externality because house price declines trigger further defaults by borrowers with negative home equity (Elul et al. 2010)

We argue that the feedback loop between foreclosures and house prices is mitigated in neighbourhoods where lenders hold larger shares of outstanding mortgages. The extent to which a lender is directly exposed to mortgage losses in a neighbourhood affects the lenders’ incentives to foreclose or renegotiate defaulting mortgages. This, in turn, has an effect on the dynamics of house prices and bank losses in that neighbourhood.

When the provision of credit is dispersed, foreclosure decisions are taken in isolation, and small lenders do not internalise the pecuniary externality that their decisions have on local housing prices. In these markets, liquidity defaults cause house prices declines, and may be followed by strategic defaults, as borrowers with negative equity that can afford to repay find it optimal to default.

In contrast, a lender with a large proportion of the outstanding mortgages in the neighbourhood internalises the adverse effects of liquidation decisions and has stronger incentives to renegotiate defaulting loans. More renegotiations reduce the adverse effects of negative shocks on the demand for housing leading to lower rates of house price depreciation and weakening other borrowers’ incentives to strategically default. As a consequence, in neighbourhoods with high outstanding mortgage concentration, even small banks that foreclose defaulting mortgages experience smaller losses.

We find empirical support for these conjectures. We construct a zip code level index of outstanding mortgage concentration, measuring the incentives of lenders in a neighbourhood to renegotiate. An increase in this index from the bottom to the top decile of the distribution reduces, all else equal, the foreclosure rate by over 40% in an average US zip code. Furthermore, a move in the concentration index from the bottom to the top decile of the distribution leads to 22% lower rate of house price declines in the average US zip code. These results are obtained controlling for a large set of zip code characteristics, proxies for borrower creditworthiness, and standard controls for local housing, income, and demographic characteristics. We present a number of further tests at the zip code level as well as at the loan level that strongly indicate the causal interpretation of these findings.

Policy implications

Our findings have important policy implications.

• First, in taking foreclosure decisions, lenders are affected by the outstanding mortgages on their balance sheets.

When income shocks limit borrowers’ ability to repay, measures favouring the consolidation of impaired lenders with similar exposure may increase the concentration of outstanding debt. These measures are expected to reduce all lenders’ aggregate losses because they tend to strengthen their incentives to renegotiate defaulting loans and because they increase the price at which lenders that do not internalise the negative spillovers of liquidations are able to liquidate. Similar effects may be achieved with the creation of bad banks that collect impaired loans at times of crises.

• Second, by showing that a market with dispersed lenders is more prone to fire sales externalities, we also provide a reason for why competition in the credit market may erode financial stability.

Disclaimer: This paper represents the views of the authors and not those of the Federal Reserve System or its Board of Governors.

References

Anenberg, E and E Kung (2013), “Estimates of the Size and Source of Price Declines due to Nearby Foreclosures”, Finance and Economics Discussion Series 2012-84. Board of Governors of the Federal Reserve System.

Campbell, J.Y, S Giglio, and P Pathak (2011), “Forced sales and house prices”, The American Economic Review 101, 2108-2131.

Elul, R, N Souleles, S Chomsisengphet, D Glennon, and R Hunt (2010), “What Triggers Mortgage Default”, The American Economic Review 100, 490-494.

Favara, G and M Giannetti (2015), “Forced Asset Sales and the Concentration of Outstanding Debt: Evidence from the Mortgage Market”, CEPR Discussion Paper 10476.

Harding, J P,  E Rosenblatt,  and V W Yao (2008), “The Contagion Effect of Foreclosed Properties”,Journal of Urban Economics 66, 164-178.

Shleifer, A and R Vishny (1992), “Liquidation Values and Debt Capacity: a Market Equilibrium Approach”, Journal of Finance 47,1343–1366.

Shleifer, A and R Vishny (2011), “Fire Sales in Finance and Macroeconomics”, Journal of Economic Perspectives 25(1), 29-48.

This article is published in collaboration with Vox EU. Publication does not imply endorsement of views by the World Economic Forum.

To keep up with Agenda subscribe to our weekly newsletter.

Author: Giovanni Favara is a Senior Economist in the Division of Research and Statistics, Federal Reserve Board. Mariassunta Giannetti is a Professor of Finance, Stockholm School of Economics and CEPR Research Fellow.