Politicians love government credit subsidy programs. Direct loans, or loan guarantees (where the government will make good on a debt lent by others), provide a quick infusion of capital to favored constituent groups or industries with little or no immediate budget hit.
Often, there is a nice story to accompany the subsidy: a new industrial plant or a production line expansion. Sometimes the sponsors pump up the volume with visions of this or that loan being the spark needed to bring entirely new industries into a state, jump-starting good jobs at good wages for decades to come. There are ceremonies, of course, and press releases, dinners and TV interviews that boost the politician's stature and reelection prospects. Many glowing words of public-private partnerships are usually spoken.
The beauty of the credit program approach is that if (or when, as defaults are fairly common) the loans go bust, years have gone by. The politician who made it all happen may be out of office. Or if still active, he or she may be on to other subjects and not easily pinned to the problem. Quick benefits, little immediate cost to the Treasury, and lots of time between funding and potential fall-out: what's not to like?
Clarifying the fiscal cost of credit subsidies
The reality, of course, is not so sanguine. The credit support often flows to activities with substantial risk. Funding decisions too often rely on political criteria as much or more than economic ones. And subsidy recipients may also have competitors, sometimes even in-state, that are disadvantaged by the programs.
The Federal Credit Reform Act of 1990 (FCRA) was an important step in bringing some fiscal reality to the political happy times, at least at the federal level. The probability of losses associated with each lending program had to be assessed, and the estimated losses included in the budget of the funding agency. Each year, this "credit subsidy" value would be updated based on new information, and in comparison to the Treasury's cost of borrowing. This was a big improvement: loan guarantees would no longer look "free" simply because no cash changed hands at the inception of a deal. Borrower defaults from either the loans or the guarantees would be picked up and linked to the specific lending program. The potential for transparency and accountability was created.
But by using the Treasury's cost of borrowing as a benchmark, and by ignoring the costs of loan administration, FCRA always understated the real subsidies associated with loans. This is because pretty much every borrower had a higher financial risk than Treasury, and the higher risk was not being picked up in loan pricing. With hundreds of billions of dollars in all sorts of lending programs, these measurement problems become quite important. And, it is possible that they may, at last, be getting fixed.
Improving on FCRA rules with "Fair Value" estimates
On October 16th, the Financial Economists Roundtable (FER), a group of senior financial economists from many of the world's premier universities, released a statement on "Accounting for the Cost of Government Credit Assistance." FER was founded in 1993, and Roundtable member Deborah Lucas notes that this was the first statement in the Roundtable's existence on which there were no dissenting votes.
FER concludes that the current government accounting rules used to calculated the budgetary costs of government credit programs "results in the systematic understatement of the cost of federal credit programs." There is no mincing words here:
The apparent cost advantage of government credit assistance over private lenders is, in the opinion of the FER, primarily due to FCRA accounting rules, rather than to any inherent economic advantage of the government.
They continue:
Under FCRA rules the recorded budgetary costs of most federal credit programs are calculated by discounting to the present the projected expected cash flows over the life of the loan or guarantee using current, maturity-matched, Treasury interest rates as the discount factors. Use of Treasury rates as discount factors, however, fails to account for the costs of the risks associated with government credit assistance -- namely, market risk, prepayment risk, and liquidity risk. These costs must ultimately be borne by taxpayers, just as they must be borne by the equity holders (owners) of private lenders that make private loans..."
FER proposes using instead "discount rates that capture all risks borne by taxpayers, and in particular by the use of discount rates consistent with market-based or fair value estimates of cost." The cost of Treasury debt isn't what matters: "Although the government can borrow at low Treasury rates, its full cost of capital when it makes a risky loan is higher because it also includes a fair return to taxpayers and the public, who bear the associated risks." Imagine that: the idea that taxpayers, as providers of large quantities of capital to high risk enterprises, deserve a fair return on their investment. That would be progress, indeed.
This shift in valuation is something I've favored for many years. Back in 1993 when I did my first work on energy subsidies, I insisted on a definition of subsidies as "(1) government-provided goods and services, including risk-bearing [emphasis added], which otherwise would have to be purchased in the marketplace; and (2) reductions in tax burdens compared to standard treatment for a similar activity." Risk-bearing is a central element in many of the subsidies that government provides, and often increases the degree to which subsidies distort marketplace choice. Yet it was too often viewed as being not a "real" subsidy at the time because there were no immediate cash payments.
Practically, my study attempted to capture the fair value of loans, and to deal with the concerns that risk subsidies weren't "real" by establishing a subsidy range. The low-end of the range compared credit pricing to the direct cost to the Treasury of borrowing capital of a similar duration -- basically the FCRA approach plus the administrative costs of the oversight agency. The high-end estimate incorporated the "intermediation" value of having the low-risk Treasury borrowing on behalf of higher-risk enterprises on capital markets, obtaining rates and terms that would not have been possible for the borrower to get itself. This is quite similar to the Fair Value approach being proposed by FER.
My work drew a distinction between the cost of credit support (which drove fiscal impacts on government budgets) and the value of the credit support (which drove inter-fuel distortions in energy markets). What is nice about the FER statement is that they are merging the two: they are saying that the "cost" to the government really does need to pick up these other risk factors, and the value of this intermediation is actually a cost of the program to taxpayers.
Better measurement can lead to better decisions: applying "Fair Value" approach to other intermediation subsidies
FER is careful to state that improved loan costing is a separate issue from whether any particular credit subsidy program is beneficial or not. Nonetheless, much will change in terms of the incentives to use government as a bank if the full value of the subsidies is visible, and we will likely see tighter, better, lending programs as a result. It is also important to note that the exact same logic FER has applied to loans and loan guarantees applies to other subsidy forms in which the government is intermediating in private markets to provide a good or service at a discount.
There is a clear parallel with government-subsidized insurance programs, but the logic likely applies to mandated purchases as well (since the mandates set a demand floor that greatly reduces the risk of large capital investments in production capacity). Government-owned enterprises, whether public power or more complex service organizations like the Nuclear Waste Repository or the Strategic Petroleum Reserve, are other likely candidates. These initiatives usually have a mixture of risk-shifting instruments (grants, tax exemptions, subsidized credit and insurance) that understate the costs of operations and subsidize the product or service being delivered. A fair value approach would be expected to illustrate the cost of these initiatives much more clearly, and in the process demonstrate that there are less expensive routes to achieve similar public policy goals.
