- Because government has encouraged 30-year fixed-rate mortgages, a majority are backed by long duration loans
- As we found in the last financial crisis, what the government wants isn't always good for the financial system as a whole
- Did the government learn from the "lending long, borrow short" investment strategy that ultimately sank the S&L in the 1980s?
According to Inside Mortgage Finance, on September 30, 2011, U.S. banks and savings and loans held $1.363 trillion in mortgage-backed securities issued by U.S. government-backed agencies such as Fannie Mae, Freddie Mac and Ginnie Mae (the agency that securitizes FHA insured mortgages). Because the government has long encouraged 30-year fixed-rate mortgages, the vast majority of these securities are backed by long duration loans. Moreover, the interest rates on these loans today average below 5 percent and will likely be near 4 percent by 2013. Rates like this are far below historic levels for mortgages.
Accordingly, these loans will be outstanding for an extraordinarily long time, since the holders will be reluctant to refinance a 4 percent mortgage, or sell the home and buy another at what could very well be double the interest rate. The tendency of homeowners who now have these mortgages will instead be to fix up the old place with a second mortgage and stay in the home.
Since late 2008, and confirmed by Section 343 of the 2010 Dodd-Frank Act, the full amount of non-interest bearing deposits in insured banks - funds that can be withdrawn at any time - are covered by FDIC insurance. This policy, which is supposed to expire at the end of December 2012, has allowed corporations and other large cash holders to park funds temporarily with banks, with the assurance of government backing. This has resulted in a huge increase in non-interest bearing bank deposits, which according to the Wall Street Journal, increased over 30 percent year-over-year in 2011 and now total over $2 trillion.
If the Fed's activities in stimulating economic growth ultimately generate inflation, as many economists expect, 30-year mortgages paying low rates will be worth even less, and the hit to bank capital will be much higher.
This arrangement has been highly favorable for banks. They are receiving huge inflows of non-interest bearing deposits, and are in effect investing the funds in safe and liquid government-backed securities on which they are earning a spread of something like 3 to 3.5 percent. As in past recessions, the banks are gradually being recapitalized by these investments-a process that is accelerated because the Basel risk-weighted capital rules require no capital for carrying securities, like Ginnies, that are explicitly backed by the U.S. government and only 1.6 percent capital for securities issued by Fannie and Freddie.
Some might think all of this makes sense. Indeed, it was probably the reason that the Dodd-Frank Act authorized this deposit free-parking facility. It provided, in the banks and S&Ls, ready-made buyers for securities the government needed to sell and would produce the low interest rates that were supposed to revive the housing market. In other words, by eliminating the cap on FDIC insured non-interest bearing deposits, and utilizing the incentives of the Basel bank capital regulations, the government accomplished four policy goals at once: it created a market for 30 year fixed-rate mortgages, fostered low mortgage interest rates, gradually recapitalized the banks, and provided a safe place for cash-rich companies and individuals to park their funds. Pretty neat.
However, as we found in the last financial crisis, what the government wants isn't always good for the financial system as a whole over the long term, or even for the banks it was trying to help. For example, consider the enormous interest rate risk the banks are taking on by acquiring and holding MBS like Fannies, Freddies and Ginnies, mostly backed by mortgages with 30-year terms.
Let's assume that by 2013 the interest rate on 30-year mortgages increases to, say, seven percent - a rate at or below the average rate for every year from 1971 to 2001. Naturally, the free money the banks were encouraged to take will have disappeared from their balance sheets; the firms and individuals that had parked it on a temporary basis will want it back for more profitable investments if interest rates have risen from their current unprecedentedly low levels.
On the other side of the ledger will be MBS holdings totaling about $1.5 trillion by 2013, backed by mortgages yielding 4-4.5 percent. These will now be heavily discounted, so the banks will have to write down the value of these agency MBS to about $1.35 trillion. Much of this loss would have to be recognized immediately under mark-to-market accounting. To put these losses in perspective, the tier one capital of all U.S. banks and S&Ls is now about $1.217 trillion, so bank capital positions could be reduced by about 12 percent. Although there will always be a market for these securities, if the banks all tried to sell them at the same time the price would be driven down further and the losses will be greater.
Obviously, too, the risk of inflation cannot be dismissed. If the Fed's activities in stimulating economic growth ultimately generate inflation, as many economists expect, 30-year mortgages paying low rates will be worth even less, and the hit to bank capital will be much higher.
One would think the government might have learned from the "lend long, borrow short" investment strategy that ultimately sank the S&Ls in the 1980s. In that case, the government's effort to keep mortgage rates low filled the balance sheets of the S&Ls and banks with 30-year mortgages that were paying 7-9 percent, while even short-term interest rates rose to levels that were about double that. Of course, that was then and this is now. 1970s-style inflation can't happen again, right?
Peter J. Wallison is a senior fellow and Edward J. Pinto is a resident fellow at AEI.