Along with Dean Baker and a few others, I’ve been fighting a lonely crusade against those who insist that TARP “made money.” First of all, TARP wasn’t even the bailout, it represented about 2% of all emergency programs provided to banks. Second, the banks paying the money back they got at 0% interest and could lend out at 3% interest isn’t really an amazing scenario. Third, TARP wasn’t just supposed to save the financial system, it was supposed to restart lending and fix the housing mess, and on that score it failed miserably. But none of this usually gets factored into the bumper-sticker claims about TARP.

But if critics can’t get a hearing on this, perhaps the Government Accountability Office can. In a new report, the GAO faults Treasury for constantly shifting the goalposts on how they report that “profit” from TARP. It’s a slightly different argument, but one that highlights how this illusion of a successful TARP lies in manipulation of the numbers. This is from the executive summary:

Although Treasury regularly reports on the cost of TARP programs and has enhanced such reporting over time, GAO’s analysis of Treasury press releases about specific programs indicate that information about estimated lifetime costs and income are included only when programs are expected to result in lifetime income. For example, Treasury issued a press release for its bank investment programs, including CPP, and noted that the programs would result in lifetime income, or profit. However, press releases for investments in AIG, a program that is anticipated to result in a lifetime cost to Treasury, did not include program-specific cost information. Although press releases for programs expected to result in a cost to Treasury provide useful transaction information, they exclude lifetime, program-specific cost estimates.

Everyone puts their best foot forward, right? Only this has a damaging impact. Pretending that TARP brought back benefits to the taxpayer isn’t just about defending that action, but preparing for the next one. Future bailouts will be justified by the ability of Treasury to win profits through TARP. Except that’s not what happened at all.

As Yves Smith points out, GAO went after the specific problem with TARP – that policymakers lent money with one hand and made sure banks could get rich with the other, which devalues the “success” of the banks being able to lend back. And that has a serious cost.

While Treasury can measure and report direct costs, indirect costs associated with the moral hazard created by the government’s intervention in the private sector are more difficult to measure and assess.

That’s about as close as you’re going to get to government bureaucrats calling bullshit on their colleagues.

The entire report is available in a PDF. Ultimately, GAO recommends that Treasury “enhance” their communications by “consistently including information on estimated lifetime costs, especially when reporting on program results.” That’s actually a pretty narrow recommendation. It should also demand that reporting on “costs” of bailout programs consider the entire range of largesse provided in that bailout, not just the narrow sliver that makes Treasury look good.

Incidentally, GAO does discover that Treasury promised $45.6 billion for housing programs, mainly HAMP, and almost three years after the announcement they’ve spent $2.4 billion. And Treasury is throwing a time bomb into the proceedings: they continue to assert that all $45.6 billion will get spent. When it doesn’t, they will claim to have “saved” billions of dollars through TARP, without telling you that the savings came by not using the allocated housing money that is desperately needed.