Notes from the Vault
Larry D. Wall
September 2017
The bursting of the tech stock bubble in the 1990s and the bursting of the housing bubble of the early 2000s produced similar reductions in the market value of wealth, according to a study by New York University Professor Lawrence J. White.1 Yet the macroeconomic outcome of these two events was very different. The bursting of the tech bubble was associated with only a 2.4 percentage point increase in unemployment, whereas the housing bubble burst was associated with a 5.6 percentage point increase during the Great Recession. Why did these two events that had a similar impact on overall wealth have such very different implications for the economy?
The difference was not due to a relatively stronger monetary and fiscal policy response to the drop in tech prices; if anything, the policy response was far stronger for the housing price decline.2 Professor White argues the difference was in how the investments in equities were financed relative to the way housing was financed. For the most part, the buyers of the tech stocks did not use leverage (debt) to finance their purchases. In contrast, many of the buyers of residential real estate relied heavily on debt to finance their purchases, and this debt was supplied to a large extent by financial intermediaries that were themselves highly leveraged. Thus, the bursting of the housing bubble not only reduced the market value of its owners and lenders, but also it threatened many of the lenders with insolvency and led to runs on important parts of the financial system.
Given this recent experience, one might have expected that one of the major initiatives in the period since the Great Recession of 2007 to 2009 would have been to reduce the use of debt finance by the private sector. Some such measures have been taken. For example, bank supervisors have required the largest banks to increase their equity capital substantially. However, a variety of government programs remain in place that encourage individuals and financial firms to take on more debt. This Notes from the Vault post highlights of some of the major federal government programs that subsidize debt and considers why these programs remain in place. It then reviews some of the broader costs created by these programs and discusses various ways of reducing the subsidy.
Programs that encourage the use of debt
The federal government encourages individuals and firms to use debt finance in a variety of ways. One way is through federal direct lending and loan guarantees to private lenders. These programs seek to facilitate borrowing for congressionally approved purposes on terms more favorable to the borrower than would have been the case if the borrower had obtained funds without government intervention.3 The various direct loan and loan guarantee programs are detailed in the U.S. Office of Management and Budget's report Analytical Perspectives: Budget of the U.S. Government Fiscal Year 2018 and summarized in table 1.
Source: Office of Management and Budget and the author's calculations
Those who borrow under these programs are typically required to pay interest on direct loans and fees for the guaranteed loans. However, even with these payments, the cost of the loan is less than the cost of obtaining funds in the private market.
A second way in which the government encourages the use of debt is by allowing individuals to deduct certain types of interest payments on their income tax returns. According to a paper by University of California at Davis Professor Dennis J. Ventry Jr., prior to the Tax Reform Act of 1986, all interest payments were tax deductible for those taxpayers who filed an income tax return with itemized deductions. However, the 1986 Act disallowed deductions for personal interest expense, leaving only mortgage interest deduction for an individual's primary and secondary residences. Subsequent changes to the law limited the mortgage interest deduction to the first $1 million of principal but, subject to the overall cap, allowed deductions for second mortgages.4 As Ventry explains, the clear purpose of retaining the mortgage interest deduction was to subsidize the purchase of residential real estate. Analytic Perspectives: Budget of the U.S. Government Fiscal Year 2018 (table 13-1) estimates the deductibility of mortgage interest on owner-occupied homes reduced income tax payments by over $61 billion in 2016.
The third way the government subsidizes private sector debt is by explicitly and implicitly guaranteeing the liabilities of many types of financial firms. The Federal Reserve Bank of Richmond produces an annual report of what it calls the Bailout Barometer. Federal Reserve Bank of Richmond economists Liz Marshall, Sabrina Pellerin, and John Walter produced the following 2016 Estimate.
Source: Liz Marshall, Sabrina Pellerin, and John Walter
Some of these guarantees are clearly intended to subsidize certain activities of financial firms, such as the guarantees of Fannie Mae and Freddie Mac that primarily serve to reduce the cost of obtaining a residential mortgage. Some of the other guarantees, however, may serve multiple purposes. For example, deposit insurance has also been seen as serving two other purposes: (a) providing less wealthy households with a risk-free store of value, especially for their transactions accounts, and (b) reducing the risk of costly runs on banks. Regardless of the intent, the effect of these guarantees is to encourage financial firms to rely more heavily on debt relative to equity financing. Additionally, this subsidy may also encourage greater reliance on debt on the part of those borrowing from these financial institutions to the extent that part of the value of the guarantee is passed through to borrowers. For example, one of the reasons for allowing government-sponsored enterprises (GSEs) to have implicit guarantees is they will in turn pass through part of the subsidy to their borrowers.
A fourth way in which the federal government subsidizes the use of debt is by allowing corporations to deduct interest expense but not their return of equity capital in the calculation of income taxes. This provision is not generally viewed as providing a corporate subsidy, but the differential treatment of payments to debt and equity holders has the effect of reducing the cost of corporate debt relative to equity.
Benefits and costs of subsidizing through debt
The variety of debt subsidies and the activities receiving them suggest that linking the subsidy to after-tax interest rates on debt provides some benefits relative to other ways of providing the subsidy. Some of these benefits are primarily economic in nature, although others are more political.
One of the potential economic benefits of supporting activity through debt subsidies is the size of the subsidy is likely to scale with the size of the activity. That is, we would generally expect that in most cases the more debt the borrower takes on, the more the borrower is engaged in the activity being favored by Congress. Another economic advantage arises because the borrower typically has to provide some equity funding to obtain a subsidized loan. This gives borrowers some "skin in the game," which should encourage them to make better decisions.
Additionally, debt subsidies come with some important political advantages. Direct federal lending and debt guarantees can appear on a cash flow basis to be something for nothing in most years—the borrower receives a benefit but very few borrowers default so there is little immediate cost to the taxpayer. The Federal Credit Reform Act of 1990 reduced this bias by requiring the estimated lifetime cost of a new loan or loan guarantee be recorded in the year in which the loan is disbursed. However, the devil is in the details. An Issue Brief by the Congressional Budget Office explains that this could be measured as the fair market value of the expected loss. In practice the cost is measured at the rate on U.S. Treasury securities of a comparable maturity—a rate that does not provide taxpayers with compensation for the credit risk they are bearing. As a result, the true cost to society of these loans and loan guarantees exceeds the budgeted amount of the cost. In other words, the "something for nothing" aspect of the loan programs is reduced but not eliminated, as these loans provide more of value to the borrowers than is recorded as a cost to the taxpayers.
The political benefit of tax breaks takes a different form. The federal budget for each year fully recognizes the cost of tax deductions. However, most of these deductions, including the mortgage interest deduction, are not subject to the annual budget process. Once a deduction is made a permanent part of the code, Congress must affirmatively vote to change the tax code in order to reduce or remove the deduction.
In terms of the costs, there are significant economic costs to providing the subsidy through debt. One of the major economic costs of subsidizing debt, as discussed above, is increased financial fragility. Not only is the government exposed to direct losses on its loans but also to reduced tax revenue and increased social spending during periods of financial system distress. Additional sources of welfare loss include the increases in unemployment and the large reductions in the value of financial and housing assets that occur during periods of financial instability. Even the intended beneficiaries of these programs can become their unintended victims to the extent the programs induce borrowers to take on excessive amounts of debt they cannot continue to service during a crisis.
Another cost of the debt programs is they are often "leaky buckets" for the delivery of government subsidies. That is, part of the gain from the subsidy goes to third parties that are not the intended beneficiaries. For example, those that participate in the lending process for direct and guaranteed loans benefit to the extent that more and larger loans are being made as a result of the federal government's loan subsidies. These intermediaries likely compete away part of the subsidy they receive in the form of lower loan rates, but they are unlikely to lower their rates by the full amount of the subsidy they receive.5
Conclusion
The federal government continues to encourage the use of debt financing through direct loan and loan guarantee programs, through explicit and implicit guarantees of many financial institutions' debt, and through the tax code. Although many of these programs may seem appropriate on their own, collectively they tend to make the financial system and the real economy less robust to shocks.
Various approaches exist to reduce or eliminate these incentives to take on more debt. Which approach is best will depend in part on the specific program under consideration. The incentives created by the federal loan and loan guarantee programs could be eliminated if they are replaced with direct financial support for activities deemed socially desirable by Congress. However, these subsidies could also be reduced with tighter underwriting criteria and increased charges to borrowers. The debt taking incentives created by the tax code could be reduced by eliminating the deductibility of interest (for mortgage interest) but also by various other measures such as lower marginal tax rates, increasing the standard deduction (for mortgage interest), or allowing for deductions to distributions to equity holders (for corporate interest payments).
The issues with eliminating the subsidy of debt for financial institutions is a bit more complicated. A good case can be made for providing some level of explicit deposit insurance so that low- and middle-income individuals can hold their transactions balances in a risk-free account. Still, although there is a strong case for eliminating implicit guarantees, doing so is not a simple task. Investors' belief that some financial firms benefit from implicit guarantees does exist not because of what the federal government has promised to do, but rather from what investors will expect the government to do in the future. Thus, a necessary condition for eliminating implicit guarantees is to establish ex ante a set of procedures that make it credible that these financial firms can and will be resolved when needed without a taxpayer bailout. Until then, the second best substitute is stricter prudential supervision and higher fees levied on the beneficiaries of implicit guarantees.
Larry D. Wall is executive director of the Center for Financial Innovation and Stability at the Atlanta Fed. The author thanks Scott Frame for helpful comments. The view expressed here are the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System. If you wish to comment on this post, please email atl.nftv.mailbox@atl.frb.org.
References
Frame, W. Scott, and Lawrence J. White (2005). "Fussing and Fuming over Fannie and Freddie: How Much Smoke, How Much Fire?" Journal of Economic Perspectives 19, no. 2: 159–184.
_______________________________________
1 The stock price drop from 1999 to 2002 reduced share value by $7 billion and the drop in house prices between 2006 and 2011 reduced values by $6.8 billion.
2 In part because of the bursting of the tech bubble, the Federal Reserve adopted a more accommodative monetary policy, lowering its federal funds rate target from a peak of 6.5 percent in 2000 to 1 percent in 2003. However, the Federal Reserve also lowered rates after the bursting of the housing bubble, from a peak of 5.25 percent in 2006 to a range of 0 to 0.25 percent in 2008. Additionally, the Fed created a series of extraordinary liquidity facilities in 2007 and 2008, and subsequently conducted three Large-Scale Asset Purchases (often referred to as quantitative easing or QE). Also, Congress provided funds through the U.S. Treasury to support Fannie Mae and Freddie Mac, and created the Troubled Asset Relief Program (TARP) to help other distressed financial institutions. Thus, it appears the Federal Reserve and Congress did far more to support the economy and financial system after the housing bubble burst than after the tech bubble burst.
3 Indeed, borrowers would not use these programs if the terms of the loans without government intervention were more favorable to borrowers than those with government intervention.
4 The $1 million cap allows individuals to deduct the full amount of mortgage interest on loans up to $1 million. Any interest paid on principal exceeding $1 million is not deductible.
5 Frame and White (2005) discuss evidence suggesting that a substantial fraction of Fannie Mae and Freddie Mac's shareholder value came from that part of the subsidy they retained rather than passed through to mortgage borrowers.