GSEs: balance-sheet turbocharging
[Third of three parts laying the groundwork for dissecting and interpreting the OFHEO's damning report on the Fannie Mae scandal. Part 1 covers primary and secondary mortgage market; Part 2 the GSE earnings engine.] Now that we have a basic framework for assessing secondary mortgage market players, how might a GSE turbocharge its earnings?
Alien not included. By exploiting the combination of:
- Federal implicit guarantee and ‘moral hazard’. The managers’ ability to maximize use of the GSEs’ implicit Federal guarantee.
- Yield curve. That short-term rates are usually lower than long-term fixed rates.
Three different yield curves, one inverted Lender risks: not just default, but prepayment! When a secondary market mortgage securitizer buys a pool of loans, it normally raises capital through a new issue to fund its purchase. The securitizer’s assets are the loan pools it has purchased; its liabilities are the new securities it has issued. Lender balance sheets: assets and liabilities To a lender or mortgage bank, loans are assets, securities are liabilities.
Make sure your assets don’t fall out.
If we as investing lender buy a loan, we would like to receive every payment as scheduled, from the first to the last. In how many different ways might we not? Each of this is a category of lender risk. Lender risks: categories 1. Credit risk. The borrower could default and we’d have to seek recovery or loss mitigation.2. Prepayment or call risk. The borrower could pay off our loan at a time convenient for him.3. Interest rate risk. If we borrowed money to fund our purchase, our cost of money might rise even if the borrower’s loan rate was fixed. (There may be other types as well.)Each of these risks is greater or lesser based on the particular instrument:
- Credit enhancement or collateral reduces credit risk.
- Prepayment lockouts or yield-maintenance penalties reduce call risk.
Prepayment prohibited.
In the topsy-turvy world of investing in loans, lenders not only do not want borrowers to default, they also do not want borrowers to prepay.
A borrower’s down arrow is a lender’s up arrow
Think about it — why would you as a home borrower voluntarily prepay a loan? You might be refinancing at a lower interest rate, and if you are, that means the lender would rather see you not refinance, because the lender is enjoying an above-market rate.
“Don’t wake up, don’t prepay.”
Mitigating prepayment risk: match-funding or borrowing ‘off balance sheet’ Making the loan’s interest rate adjustable decreases our interest rate risk — but let’s suppose that you are a Federally chartered institution with a virtual mandate to provide fixed-interest financing. How can you possibly hedge the prepayment risk? Quite simply: by issuing securities of similar maturity, whose prepayment provisions mirror the loan pool’s prepayment provisions.
Mirror, mirror, on the wall, who’s the fairest of them all?
This passes through the prepayment risk onto the securities holders; the issuer does not take the prepayment risk. (In other words, Fannie Mae could always sell long-term securities that are callable by Fannie Mae if and only if borrowers prepay their loans.) As I posted in my translation (emphasis added) of former Federal Reserve Chairman Alan Greenspan:
Fannie Mae (Fannie) and Freddie Mac (Freddie) essentially run two lines of business:
1. Securitization of mortgage credit,
2. Holding of mortgage and other assets for investment purposes.
[They can only do the second by borrowing on their own credit. -- Ed.]
The first line of business provides substantial benefits for affordable housing through the process of using credit guarantees to turn mortgages into marketable securities that trade in public debt markets. This process creates a wide variety of liquidity benefits, some of which flow to homeowners and mortgage originators.
“To paraphrase our president, a rising tide makes the pie higher.”
Making loans more liquid frees up originators to go back to work, and the competition also lowers the rate spread (between safe rate and what the borrower pays), so customers get cheaper money. (”When lenders compete, you win.”)
Moreover, creating securities from the mortgages extended to nontraditional homeowners is an important step to making mortgage credit more widely available.
Focusing Fannie and Freddie on this type of securitization activity can promote affordable housing without creating significant risks to the [nation's] financial system [and by extension, the taxpayers].
By matching maturities (colloquially called match-funding), an issuer like Fannie Mae can keep itself out of prepayment risk, and the securities issue can be free-standing, with no open-ended risks to the issuer; such an issue is spoken of as ‘off balance sheet.’ Naturally, securities buyers who know they could have their securities called any time interest rates drop will want a higher premium.
It all began under the buttonwood tree … We can see that the risk-return marketplace works — match-funding is much safer, but there’s much less spread between interest Collected and Paid. Mismatching maturities: lending long, borrowing short Instead of reducing risk by match-funding, a long-term lender could deliberately mismatch maturities, borrowing only short-term.
Lending long Lending long means issuing new loans with: · A fixed interest rate (no reset or adjustment)· Level monthly payments· No balloon or acceleration provisionsLending long and borrowing short means that even though the loan portfolio itself will not accelerate, the issuer will have to reissue the securities in the interim. Whatever the yield curve is, the issuer will have to take it. That’s risky — but lucrative whenever the yield curve is positive.
Yield curves prevailing in 2002 and 2003.
When the yield curve is positive (as it normally is), lending long and borrowing short manufactures spread. For instance, let’s take July 2002 from the chart above. Ten-year Treasuries were commanding 5.25%; five-year Treasuries only 4.25%. For an issuer routinely borrowing billions, every billion financed with five-year debt instead of ten would generate another $10,000,000 in annual earnings, every year of the first five years. But why stop with five years? One-year rates were about 2.75%, so issuing one-year securities would generate $25,000,000 in annual earnings. Naturally, you’d have to refinance in twelve months, but that’s a very steep yield curve, and borrowing short would be worth a lot of short-term shekels. Got it? Lending long, borrowing short, turbocharges annual earnings. Meanwhile, it dramatically increases portfolio risk.
We move fast and everything’s under control!
When you lend long, you are giving the borrower the optionality — the borrower can prepay, but you cannot compel the borrower to prepay. If you then borrow short, you are making a bet that over the period the loan is outstanding, average interest rates will be well below your fixed rate. You had best know what you are doing.
Professional financier on closed course!
GSEs’ mismatching maturities: turbocharging In the conventional environment, where the bond buyers are looking at the issuer’s creditworthiness, mismatching of maturities are penalized in rate; in effect, those who lend short cast a cynical eye over the borrower’s balance sheet, leverage, exposure, and management practices.
In case of emergency, change corporate management.
A conventional borrower’s turbocharger has an automatic brake, but the GSEs, with their awfully big advantages including the implicit Federal guarantee, have no brakes at all, as Mr. Greenspan put it succinctly (full text in .pdf):
Under normal circumstances, GSEs are able to easily maintain and grow their large portfolios of mortgage and non-mortgage assets without the significant market checks or balances faced by other publicly traded financial institutions.
A private company trying this trick would be tested for its ability to cover an adverse swing. Since the GSEs are perceived to be protected by Congress, this doesn’t happen, so the market doesn’t charge the GSEs a premium, and hence the market doesn’t check their balance sheet growth.
These large portfolios, while enriching GSE shareholders, do not meaningfully benefit homeowners and do not facilitate secondary market liquidity.
The rate spread compression (cheaper loans) occurs at origination and initial securitization. Buying and selling existing portfolios doesn’t help the consumer. It does make Fannie Mae a ton of money, and it puts taxpayers at risk:
They do add systemic risk to our financial system, which normal market forces are unable to resolve.
“And risk is bad.”
When an actor has insurance (explicit or implicit) against a behavior, he is more likely to take higher risks; economists call the phenomenon moral hazard. Moral hazard means that people with insurance may take greater risks than they would do without it because they know they are protected, so the insurer may get more claims than it bargained for. The Economist’s references offer grim examples:
See also DEPOSIT INSURANCE, LENDER OF LAST RESORT, IMF and WORLD BANK. The key to moral hazard: who’s at the controls? If the market won’t check turbocharging, who will? In a government context, there are two possibilities: a regulator, or the company’s management. For the GSEs, the regulator is OFHEO.
Note how small the regulator is!
Without in any way inviting a comparison with current or past Fannie Mae or Freddie Mac management, I may observe that:
-
Clark Kent would match-fund because it’s safer; Lex Luthor would exploit the moral hazard for all it’s worth. -
Clark Kent would promptly comply with any regulatory request; Lex Luthor would deride, ignore, intimidate and his regulator whenever he could.
“I always say nice thinks about you, Clark!”
