Bailout policies are interconnected, and one policy can increase or decrease the effectiveness of another. The government is currently considering alternative means of assisting banks: buying assets (including the creation of an aggregator bank) and offering various forms of downside guarantee against loss on some assets. Whatever form of support the government adopts, it will create an interest (either as a buyer or as an insurer) in sub-prime mortgages and related securities.
Thus, it would make sense for the government to set its foreclosure mitigation policy simultaneously with its bank assistance policy. Eliminating the risk of cram-downs, and providing some sort of loss-sharing arrangements for encouraging renegotiation rather than foreclosure could substantially increase the values of mortgages and their related securities. Indeed, merely resolving the uncertainties about government policy towards foreclosure could itself raise those values.
The primary objective of any bailout policy should be to revive the economy. While the costs to taxpayers should be fully taken into account, the benefits should be defined to include the economic growth and wealth creation that an effective bailout would produce. Much of the discussion of bailouts overemphasized the monetary gains that might conceivably be earned by taxpayers on bailout “deals,” which may be creating unrealistic expectations.
Bank and borrower distress and their effects on the real economy are evoking trillions of dollars of government expenditures in TARP funds, stimulus spending, Fed and Treasury assistance to individual firms, and prospective foreclosure mitigation costs. Assistance programs alter future behavior through the incentives they create for recipients, and these incentive consequences are an important component of the ultimate social cost of providing bailouts. The designers of bailouts should consider carefully the incentive consequences of different forms of bailouts, and in selecting policies should consider overall social costs, which consist of both current cash outlays and undesirable future behavior. To illustrate this point, the Committee wants to identify several issues that are particularly relevant in the design of the next round of bailout programs.
Forbearance is a form of bailout policy that has been used in the past to insulate banks or borrowers from the legal consequences of financial distress. Foreclosure moratoria have been used in the US in the 19th and 20th centuries to relieve borrower distress. During the 1980s, savings and loan associations were permitted to pretend that they were solvent for years because this avoided the unpleasant consequences of shutting down insolvent S&Ls. Some observers are calling for similar policies today. Forbearance, however, has very bad incentive consequences. Forbearance leads banks that are insolvent to take on more risk as a gamble for “resurrection,” and tempts borrowers to mismanage their finances and the maintenance of their properties. The suspension of market discipline produces inefficient tolerance for incompetent, as well as excessively risky, borrowers and bankers. A recent example is the excessive delay by the Office of Thrift Supervision in closing IndyMac, resulting in a nearly 30% loss in asset value to be absorbed for the FDIC insurance fund.
Lending to troubled banks (e.g., in the form of loans or injections of preferred stock) can help resolve financial distress, particularly if their problem is one of illiquidity rather than insolvency. But if the problem is one of insolvency, lending is a form of forbearance that produces the same inefficient risk-taking incentives and tolerance of waste.
Other forms of assistance (for example, injections of common equity funds) avoid the risk-taking incentive problems of forbearance and lending, but create potential distortions related to the government’s role as a stockholder. In particular, government ownership interests in banks, or in the extreme case nationalization of banks, is likely to politicize the lending policies of banks, and produce inefficient allocation of credit related to a lack of focus on economic value creation in order to favor particular political constituencies.
Given that different assistance mechanisms are more or less appropriate depending on the circumstances of recipients, uniform treatment of all banks or borrowers during a financial crisis is wasteful. Thus, for example, making loans accessible only to banks that can raise sufficient matching funds in the form of equity may make sense as a policy designed to identify viable banks and to avoid lending to insolvent banks. Any policies that discriminate among borrowers should be based on uniform guidelines that determine access to assistance, which can be defended on efficiency grounds.
It is desirable for government policy to reflect a pre-existing framework in which responses can be predicted and an exit strategy is provided for. Compounding the fundamental uncertainties that have shocked the marketplace over the last year has been considerable uncertainty about the response of federal policymakers and how various financial instruments would be treated. In the case of Fannie Mae and Freddie Mac, this left the private markets unwilling to provide financing.
Bailout policies also can have undesirable consequences if they affect the legal framework for debt resolution. The legal framework underlying mortgage loans is part of a national policy favoring home ownership and is central to the market’s pricing of these loans. A major part of the reason that mortgage financing has so dominated consumer finance over the past two decades is that mortgage loans are generally cheaper for consumers than other consumer loans (e.g., credit cards). That low cost reflects legal protections that are unique to mortgage finance. In particular, under current law, bankruptcy judges are precluded from forcing mortgage lenders to forego the value of their collateral and write down their principal and interest payments on mortgages for primary residences.
Recently, there have been calls to allow bankruptcy judges to modify the terms of existing mortgages on primary residences. This would raise cost of new mortgage finance for several reasons: First, the capital market would require a higher interest rate to loan funds whose terms could later be altered in bankruptcy without the consent of the lender. Second, changing the bankruptcy framework after the fact contrary to contract terms would reduce the value of existing mortgage claims. This will increase the cost of all future credit by making contract enforcement on future loans more uncertain. Permitting such a mortgage “cram-down” will increase the cost of financing of primary residences going forward and thereby reduce the affordability of housing and slow the potential recovery in housing. Third, even if the cram-down option credibly applied only to existing mortgages, it would reduce the likelihood of successful voluntary renegotiations between borrowers and lenders. In addition, retroactive legal changes affecting existing mortgages would invite court challenge as a “taking,” delaying any actual effect.
It would be far more desirable to mitigate foreclosures by establishing government policies that explicitly share the costs of a write down, which would encourage more renegotiations instead of foreclosures. That loss sharing could take many possible forms (e.g., taxpayer absorption of a percentage of write down cost, or in a transaction involving the use of a deed in lieu of foreclosure that would allow homeowners to remain in their home as tenants, a taxpayer-financed rental subsidy). If policymakers believe that the current wave of foreclosures has large social costs that warrant policy intervention, they should be willing to help pay for mitigation, not use destructive legal maneuvers that give rise to long-term market inefficiencies and higher costs to avoid current cash outlays.