Treasury should not bail out Fannie and Freddie’s subordinated debt

Published in Economics 21.

When the U.S. Treasury bailed out Fannie Mae and Freddie Mac in 2008, holders of $13.5 billion in Fannie’s and Freddie’s subordinated debt—debt paid off after senior debt is repaid—were completely protected. Instead of experiencing losses to which subordinated lenders can be exposed when the borrower fails, they got every penny of scheduled payments on time.

The structural reasons for the unusual occurrence should be carefully examined by the Treasury to avoid its repetition.

This outcome was the exact opposite of the academic theories that had for years pushed subordinated debt as the way to create market discipline for financial firms that benefit from government guarantees on their senior obligations. In the event, Fannie’s and Freddie’s subordinated debt produced, and its holders experienced, zero market discipline. So much for the academic theories, at least as applied to government-sponsored enterprises.

“Perhaps surprisingly,” a Federal Reserve 2015 post-mortem study of the bailout says, “the two firms maintained their payments on the relatively small amount of subordinated debt they had.”

In terms of the theory that subordinated debt will be fully at risk, this is very surprising. As economist Douglas Elmendorf, former director of the Congressional Budget Office, wrote in criticism of this bailout detail, “The crucial role of subordinated debt for any company is to create a group of investors who know they will lose if the company fails.” But the Fannie and Freddie bailout was structured so that it didn’t happen.

The $13.5 billion is certainly a material number. While small relative to Fannie’s and Freddie’s immense liabilities, it was a significant part of their overall capital structure. In June 2008, Fannie and Freddie combined reported common equity of $18.3 billion and preferred stock of $35.8 billion, giving them, with the subordinated debt, total capital of $67.6 billion. The subordinated debt was thus 20 percent of reported total capital at that point, but it did not carry out its function as capital.

“In fact, subordinated debt is part of regulatory capital since the Basel I Accord (1988) and was always meant to absorb losses,” says a 2016 study for the European Parliament. Fannie and Freddie were not subject to the Basel Accord, but this nicely states the general theory.

“Market discipline is best provided by subordinated creditors,” wrote banking expert Paul Horvitz in 1983. A Federal Reserve study group produced the report “Using subordinated debt as an instrument of market discipline” in 1999. In 2000, Fed economists considered “a number of regulatory reform proposals aimed at capturing the benefits of subordinated debt” and concluded that it would indeed provide market discipline. “Ways to enhance market discipline… focused in large part on subordinated debt,” as a study by Fannie and Freddie’s regulator observed. Consistent with these theories, in October 2000, Fannie and Freddie committed to begin issuing publicly traded subordinated debt, and did. But bailout practice turned out to be inconsistent with the theory.

The Treasury knew precisely what it was doing for the subordinated debt holders. “These agreements support market stability,” said then-Treasury Secretary Henry Paulson’s Sept. 7, 2008 statement about the bailout, “by providing additional security and clarity to GSE debt holders—senior and subordinated—and support mortgage availability by providing additional confidence to investors…etc.” Treasury slipped that “and subordinated” in there in the middle of the paragraph, without any further comment or explanation.

“Under the terms of the agreement,” Paulson continued, “common and preferred shareholders bear losses ahead of the new government senior preferred shares.” But the government’s new senior preferred shares would by definition bear losses ahead of the subordinated debt. Why have the taxpayers be junior to the subordinated debt?

It appears that the Treasury was trapped as an unintended result of the Fannie and Freddie reform legislation of earlier in that bailout year, the Housing and Economic Recovery Act of 2008. A major battle in the creation of that act was to include the potential for receivership for Fannie and Freddie in the event of their failure—so, in theory, the creditors would have to consider the possibility of loss. Overachieving, Congress in HERA Section 1145 made receivership not just possible, but mandatory, in the event that Fannie and Freddie’s liabilities exceeded their assets and the regulator confessed it.

But in the midst of the financial crisis, the last thing Treasury wanted was a receivership, because the last thing they it wanted was to panic the creditors around the world by the prospect that they would be taking losses on Fannie’s and Freddie’s trillions of dollars of senior debt and MBS. Instead, they wanted to convince these creditors that there would be no losses. Treasury theoretically could have arranged to guarantee all the senior debt and MBS formally, but that would have forced the U.S. government to admit on its official books that it had an additional $5 trillion of debt. That would have been honest bookkeeping, but was an obvious nonstarter politically.

Once having decided against receivership, Treasury had to put the taxpayers’ money in as equity. Officials calculated that if Fannie and Freddie’s net worth were zero, their liabilities would not exceed their assets, so kept putting in enough new equity to bring the equity capital to zero. But once they had done that, there was no way to give the subordinated debt a haircut.

“We have been directed by FHFA to continue paying principal and interest on our outstanding subordinated debt,” Fannie reported. The covenants of the subordinated debt provided that if it were not being paid currently, no dividends could be paid by Fannie and Freddie—that would have included dividends on the Treasury’s senior preferred stock.

One reform idea, the mandatory receivership, had knocked out the previous reform idea that subordinated debt must take losses, and neither happened.

So financial theory notwithstanding, Fannie and Freddie’s subordinated debt achieved nothing. Its holders got a premium yield but were fully protected by the U.S. Treasury. The purchasers had made a very good bet on the financial behavior of governments when confronted with the failings of government-sponsored enterprises.

The 10th year since the bailout of Fannie and Freddie, and of their subordinated debt, will begin in September. Will Treasury now begin to address the structural issues?