Since the publication of an article from Pro Publica’s Jesse Eisinger and NPR’s Chris Arnold about alleged new analyses by Fannie Mae and Freddie Mac finding that principal reductions are good business, the growing community has been on the warpath around demanding that the FHFA, which oversees Fannie and Freddie, allow a principal write-down program to go forward. Organizations like The New Bottom Line and the Campaign for America’s Future have confidently said that the Pro Publica article proves that Ed DeMarco must stop holding up write-downs for delinquent underwater borrowers before they slip into foreclosure. Democrats on the House Oversight Committee Elijah Cummings and John Tierney, who have led the charge in Congress, asked DeMarco in a new letter to deliver the analyses from Fannie and Freddie, “as well as information on a principal reduction pilot program that was cancelled in 2010.”

Indeed there was a pilot principal reduction program at Fannie back in 2010. Fannie alleges that the program was ineffective compared to forbearance, which also lowers the monthly payment without a cut in principal, maintaining the bottom line on the balance sheet. That may have been true in the first month, but the whole point of principal reductions as a superior modification strategy is that they’re built to last – negative equity correlates strongly with foreclosures over time. So Fannie may have pulled the plug too early.

But all of this came before the Treasury Department tripled its financial incentives under HAMP to encourage principal reduction by the GSEs. This is supposed to be what triggered the new analyses at Fannie and Freddie.

Only where are these analyses? Fannie and Freddie did not respond to the Pro Publica piece. There’s no evidence in there other than anonymous sources, and we don’t know if the sources come from the GSEs themselves.

Here’s what we do know. Administration sources, in particular HUD, which has been the public face of housing policy lately, have wanted to channel liberal energy toward demanding principal reductions at Fannie and Freddie, in part to take the pressure off their own policies. You don’t hear about the fact that banks can jack up interest rates on refinancing under the new HARP program. You don’t hear about how banks can get dollar-for-dollar credit for their punishment in the settlement by bulldozing or donating homes or waiving deficiency judgments, all of which they do anyway. No, you just hear that principal reductions by Fannie and Freddie are the key to unlock the housing market. It’s as if Treasury and HUD are driving that conversation themselves. It certainly works out well for them.

Meanwhile, Gretchen Morgenson has an interesting piece where she claims that the only analyses done by Fannie and Freddie on principal reductions are inconclusive, contradicting the Pro Publica report. Privately, I’ve heard that there are no new analyses, just one Power Point presentation from a Freddie Mac official based on his own opinions, which was leaked by someone with an interest in keeping the pressure on DeMarco.

Morgensen also runs through some of the implications of a principal write-down program from the GSEs:

But what the proponents of principal reductions at Fannie and Freddie don’t talk about is what a transfer of wealth from taxpayers (again) to large banks such a program would represent. The fact is, principal reductions by Fannie and Freddie are not the panacea that they may seem [...]

Here’s how: Many banks hold second liens on the same properties for which Fannie and Freddie either own the first mortgage or have guaranteed. If principal amounts on these first mortgages are reduced while leaving the second liens intact, those seconds become much more likely to be paid off over time. With no principal reduction, the banks would have to write off many of those second liens.

As such, principal write-downs are another backdoor bailout for the banks that brought you the mortgage crisis.

Answering his critics, Mr. DeMarco has agreed to approve principal reductions at Fannie and Freddie, but only when Congress passes legislation enabling it. Writing a law to force taxpayers to bail out the banks in this way, however, might anger constituents. So it’s far easier for members of Congress to rail against the one supposedly intransigent man who is preventing the great American housing recovery.

DeMarco putting this back on Congress is funny, because surely he knows that a GOP House will have nothing to do with this.

Additionally, I have some problems with Morgenson’s analysis. I don’t know why she says that “the banks would have to write off many of those second liens” without a principal reduction. They’d be under no obligation, and regulators are in no hurry to force a write-off. And I tend to believe that principal reductions on underwater loans, in the cases where they lead to a prevention of foreclosures, are long-run positive. So I don’t think that taxpayers would be hurt by the principal reductions (unless you’re talking about the HAMP incentives, but they go from one end of the government – Treasury – to the other – Fannie and Freddie – so that doesn’t scan either). But it’s true that banks would be enriched by them if they don’t have to wipe out their second loans. In the settlement, where a first lien with a second is written down, the second only has to be written down on equal terms, which is a violation of standard lien priority and a gift to banks. So this has happened before. If the focus is on writing down firsts at Fannie and nothing gets done to the seconds, it will be an even bigger gift.

This doesn’t get DeMarco totally off the hook. According to the most recent data from CoreLogic, 6.3 million of the 10.7 million underwater homes in America don’t have second liens, accounting for $329 billion in negative equity. DeMarco could allow principal reductions on just those loans, if this is an obstacle, without the bank-enrichment aspect coming into play.

But there is an argument among regulators and banks about what to do with the seconds, one that doesn’t usually get portrayed in the simplistic arguments about Ed DeMarco holding back the recovery by himself. OCC would surely step in to block any principal reduction program that would force the seconds wiped out, for example (and they may have done this with the pilot program at Fannie). The banks still have the seconds on their balance sheets mostly at par, even if they’re worthless. We’re talking about hundreds of billions of dollars in second liens, more than the banks’ capital reserves. Protecting the banks from accepting such losses has been a dominant policy framework of the Administration.

Morgensen also runs through the numbers showing that Fannie and Freddie have consistently delivered twice as many loan modifications as the banks. Of course, I think they have twice as many loans. What is worth noting is that their modifications perform better than the ones given by banks, with lower re-default rates. Banks have not done their job, certainly not at the level of the GSEs, and yet the new sexy thing to yell is “Why doesn’t DeMarco write down GSE loans?” The rallying cry could just as easily be “Why don’t the banks wipe out their seconds?”

UPDATE: Felix Salmon is right, of course. Writing down firsts while leaving seconds intact would defeat the purpose of principal reduction, because the borrower would still be underwater when you add up the first and second lien. But maybe that’s part of FHFA’s analysis. Maybe they’ve been told by banks they can’t touch the seconds, and maybe OCC backed them up. That’s very plausible. Indeed, seeing what was done in the settlement, where the second liens are not wiped out but only reduced in equal measure to the first, I’d say that’s the furthest the Administration is willing to go, and that’s on a mandatory punishment.

Felix is also right that if Fannie and Freddie are doing non-principal reduction modifications that are producing a low re-default rate, that “enriches the banks” as well on their seconds. And Fannie and Freddie could easily start with the underwater loans on their books without second liens. So Morgenson’s defense falls short. But the issue of the seconds isn’t nothing – at least 40% of the market and almost half the negative equity – and nobody has demanded a solution for that. If Felix’ “oh, just give them $5,000″ solution worked, it would have been done a long time ago. This has been an impediment to the housing market because the banks won’t take losses, and the regulators won’t demand it of them, as they could.