Download PDF On April 13, 2012, the US Department of the Treasury released new cost estimates for the Troubled Asset Relief Program (TARP) as part of a series of charts describing the combined impact of Treasury, Federal Reserve, and FDIC financial-stability programs. Looking principally at actual and projected contractual cash flows, the document concludes that: "Overall, the government is now expected to at least break even on its financial stability programs and may realize a positive return."
The Shadow Financial Regulatory Committee believes that the Treasury has made four fundamental errors in reasoning. These four errors are not only inconsistent with basic economics and accounting principles, they are also inconsistent with standard practices for government assessments of the costs and benefits of such programs.
First, the cash-basis calculation of cost provided in the Treasury's analysis wrongly deducts Federal Reserve interest revenues as a contra-cost in measuring cost. Second, the cash-basis calculation of cost ignores the projections of the Special Inspector General of TARP, which contemplate larger cash-on-cash losses from TARP lending, and understates prospective losses from government funding of Government-Sponsored Enterprises' (GSE) losses. Overall, taking these two errors into account, the Treasury's estimates understate the true cash-basis cost to taxpayers by at least $200 billion, and probably much more. Third, it is inappropriate to measure taxpayers' costs on a cash basis. Fourth, the Treasury asserts - but does nothing to show - that the benefits and net gains to society from the programs were large.
The first error relates to the inclusion of $179 billion in transfer payments from the Federal Reserve to the Treasury. Federal Reserve transfers do not belong in the calculation of cash-on-cash program returns. Treating Federal Reserve transfer payments as income to the Treasury is a sleight of hand that would make Bernie Madoff proud. Since the Federal Reserve is not permitted to lend to the Treasury directly, it purchases government securities in the open market, effectively replacing Treasury debt held by the public with Federal Reserve debt in the form of commercial-bank deposits held at the central bank. From taxpayers' perspective, to assess the costs of financial stability programs and score the net transfer of funds between the two government agencies, the Federal Reserve balance sheet and income statement should be consolidated with those of the Treasury. The Federal Reserve purchases government debt and receives interest as a transfer payment from the Treasury. The Fed deducts operating expenses from the interest it receives and returns the remainder to the Treasury. On balance, therefore, the Treasury underwrites the cost of Fed operations. It is a mischaracterization to say that the Fed earns a profit on its Treasury holdings. In fact, there is a regular net flow of payments from the Treasury to the Fed.
The second error is the underestimation of prospective cash-on-cash returns for the TARP program. The Treasury's calculation conflicts with CBO estimates and with the testimony of Christy Romero, the Special Inspector General for TARP. Her recent testimony indicates that taxpayers are still owed some $118 billion, and that 351 small banks with roughly $15 billion in outstanding loans face a "significant challenge" in being able to raise funds to repay the government (Wall Street Journal, April 25, 2012, pp. C1-2). Furthermore, the Treasury's projected cost of bailouts for the GSEs is unrealistic; the projected cost of only $28 billion through 2022 assumes a remarkable improvement in GSE performance, given that costs to date are roughly $180 billion.
Third, the true cost to taxpayers is still bigger, because the Treasury does not consider the risks that taxpayers bore in supplying federal guarantees and emergency credit support to troubled firms. Both forms of support were offered on highly subsidized contractual terms. At the time taxpayer funding was extended, the risk was substantial and deserved a high rate of compensation.
Mischaracterizing bailout costs, as the Treasury does, not only flies in the face of economic and accounting principles, it ignores specific reporting requirements embodied in the Emergency Economic Stabilization Act of 2008, which was the empowering legislation for TARP. This Act requires the Congressional Budget Office (CBO) and Office of Management and Budget (OMB) to measure subsidies by including the cost of market risk when estimating the budgetary cost of TARP.
During the height of the crisis, few institutions were willing to commit funds in large amounts even on overnight terms. Acting on behalf of taxpayers, government agencies created new FDIC guarantees, TARP funding, and Federal Reserve lending and guarantee programs. Those programs supported extremely large amounts of financing at below-market terms for substantial periods of time. If those funds and guarantees had been priced at or near their true market value, taxpayers would have been entitled to substantially higher rates of return.
The true cost of credit-support subsidies can be measured in several ways. The simplest way to think about this cost is as a return that private investors would have demanded to be paid for filling in the capital shortfalls that existed at the distressed firms if private investors, rather than taxpayers, had funded the interventions in question. Another way to think about the true opportunity cost is that it would equal the return that taxpayers could have required had they hired a team of properly incentivized private agents to deploy funds in comparable efforts to stabilize the financial sector at the time that TARP and other federal commitments were made. Indeed, studies conducted by the CBO routinely show that the value of the subsidies in these and other deals can be determined by comparing these measures with the much-lower payouts authorities actually contracted for.
Fourth, a full economic analysis of assistance programs such as TARP requires an assessment of benefits as well as costs. Best-practices reporting standards for quantitative and qualitative costs and benefits in regulatory reviews are set by OMB Circular A-4. This circular states that a "good regulatory analysis is designed to inform the public and other parts of the Government (as well as the agency conducting the analysis) of the effects of alternative actions."
Economically, OMB Circular A-4 sets a sterner test than the mere avoidance of financial collapse or macroeconomic depression. It insists that agencies explain how policy actions "are linked to expected benefits." To be sure, during the height of the crisis, the market was demanding extraordinary risk premiums and if bailout support had been priced consistently with then-prevalent market terms, it would have defeated the purpose of the programs. But if private agents had been asked to negotiate such a deal on taxpayers' behalf, to meet the OMB standard the government's contracting authorities would have been obliged to make the private agents establish just how these highly subsidized deals actually promised to produce greater benefits for society. Otherwise, authorities would have had to deem the agents' handling of taxpayer resources negligent or even grossly negligent.
Here and in Europe, bailout deals have left piles of taxpayer money on the deal making table and a substantial proportion of this money ended up in the hands of the traders and other executives whose aggressive risk-taking had turned their firms into basket cases. Whether or not these huge subsidies can be shown to be worth their cost, it is dishonest to pretend that no subsidies existed. Treasury's emphasis on cash flows rather than economic costs egregiously misframes the financial and economic public-policy issues. Rather than trying to mislead the voting and taxpaying public with false claims of bailout profits and clever policymaking, officials should be trying to identify the true costs and benefits of each rescue program so that, here and abroad, future episodes of financial instability can be handled more effectively.