There’s a very wonky dust-up afoot about how the federal government accounts for its lending programs, like student loans. The resolution is important, since if we were to make the change for which some influential folks are calling, our budgets would reflect considerably higher costs for government loans.
Now, mind you: no one—including opponents of the current way we score our loan programs—is arguing that those programs would actually cost the government any more than they do now. Those calling for the change want the budget to reflect higher costs, even though they don’t believe the government will actually face those costs. As I argue below, that makes them phantom costs.
To understand why some are calling for changing from the current method to the so-called “fair-value accounting” (FVA) method, you have to understand how these loans are currently treated in the budget. When the government makes a loan, the cash dispersed to the borrower is considered government spending. But because we expect most borrowers to repay, the budget subtracts the value of what it expects to get back. That subtracted value is reduced to account for the fact that some borrowers will default or pay late and also to account for the estimated value today of interest payments and principal expected to come back to the government over the course of the loan (the “net present value”).
So far, so good. FVA’ers don’t disagree with this part of the deal. But they argue that there’s another cost which we’re leaving out of the picture: the extra costs that would be invoked if a private lender were making the loan.
You see, private lenders face various disadvantages relative to the government and they reasonably want to be compensated for them, so they charge more for their loans. They can’t borrow as cheaply as Uncle Sam, they can’t tax, they can’t print money, they won’t make loans unless they expect to make a profit on them, and for all these reasons, they’re more risk averse. So they add a risk-premium to their interest rate.
The FVAers believe that the government should treat its loans (and loan guarantees) the same way that private lenders treat them, essentially adding the cost of the “market risk” faced by private lenders on top of the present value cost already in the budget.
To folks on the other side of the argument, like me and my CBPP colleagues, this makes no sense. The government as a lender differs in fundamental ways from its private counterparts, and those differences must be reflected in the budget. That doesn’t imply riskless lending. As noted, the full likelihood of defaults or late payments and the probability of changes in interest rates need to be taken account of – as they already are. But to charge the government for market risks it does not face is to create phantom costs, (which in turn, invoke the need for phantom offsets).
Here’s an important wrinkle that I think shows the folly of the suggested change.
The FVAers don’t want the government to actually charge a risk premium to say, student borrowers. They just want us to record the budget costs as if the government had to charge a risk premium but failed to do so. They want government lenders to take advantage of their special characteristics noted above and lend at favorable rates relative to private lenders. But they want us to write down the value of our loan portfolio if we were private lenders.
Why? I’m not 100% sure. The FVAers seem to think the current accounting methods give the government an unfair advantage over private lenders, and want to put them on equal footing. They seem put out, for example, by the fact that our student loan program is scoring as a “negative subsidy”—it’s making money—right now, as the government is taking advantage of favorable spreads between its borrowing and lending costs.
Perhaps the FVAers want to artificially pump up the budgetary costs of the loans because they worry that a government that can lend more cheaply than the private sector will fail to impose enough discipline on its lending programs. While I’m obviously opposed to the shift to FVA, that’s not an unreasonable concern. I don’t want to see the government take advantage of its unique ability to lend at lower costs than private lenders such that it gets into all kinds of different markets. Student loans are different, since they support an important public good. But if this is the FVAer’s motivation, I kinda see where they’re coming from.
It’s just that their solution to the problem is flat-out wrong. Government ≠ private sector—and pretending that inequality doesn’t hold is just plain distortionary.
In this regard, the FVAers seem a bit like Nigel from the band Spinal Tap describing how his amplifier is extra loud because it goes up to 11 instead of 10. To most of us, the fact that someone drew an “11” on a dial that used to read “10” doesn’t change the amp’s maximum volume.
The budget scores loans at a “10” because a “10” is the best estimate of the cash flows, just as the amp recorded its highest volume at 10 before someone changed the dial’s top value to read “11.” The FVAers want us to record an “11” or “12” or whatever because that’s how that’s how (heavily medicated) Nigel perceives volume much like they say the public perceives risk. But the noise that the amp makes at 10 is the same it makes at 11.
Or try this one: if it’s good budgeting to add in the higher cost of private sector risk aversion, uncertainty, and so on to lending programs, then mustn’t we value and add the benefits of reduced uncertainty which our social insurance and safety net programs afford us? As it stands, the future costs of these programs are based on our best guesses about what it will cost to provide the income (e.g., Social Security) and health benefits (Medicare) to those entitled to such benefits. But those projections, especially re medical costs, invoke considerable uncertainty—uncertainty which is borne by the government, not by the beneficiaries. The logic of the FVAers would have us tote up the value of this risk shift from retirees to the government and subtract it from the budgetary costs.
That might help make our budget projections look better but it would be as unjustified as what they’re proposing re the lending programs. So let’s leave things as they are here and not bend ourselves into a budgetary pretzel in order to pretend the government isn’t the government. Even if our amp goes up to 11.