Martin Wolf on Resurrecting Securitization

Martin Wolf, the Financial Times’ economics editor, may have called the demise of securitization prematurely in his article, “Securitisation: life after death.”

This is an odd piece for the normally thoughtful and pragmatic Wolf. On the one hand, he gives a succinct and colorful of assessment of the credit crisis, depicting it as yet another instance of banks-as- lemmings. He stresses that one of the two big causes was that securitization encouraged sloppy lending since (duh) the originator could be cavalier about the risks.

Much of this is familiar ground, but Wolf give a tidy precis and makes a number of acute observations in passing.

But the piece goes a bit pear shaped when it comes to recommendations. Wolf says that securitization offers too many benefits to try to roll the clock back to the old-fashioned model of having banks hold the loans they sourced. Yet his remedies are woefully inadequate. He says that originators of securitized liabilities need to care about their reputations (and he also has a vague exhortation that securitization should become as normal as selling junk bonds).

Reputation? Does anyone care in this day and age when Martha Stewart goes to prison and her business carries on as before with nary a dent? It hasn’t even occurred to two Wall Street firms that pretend to care most about their image, Goldman Sachs and Morgan Stanley, that there might have something unseemly about their role in packaging mortgage paper, particularly collateralized debt obligations. Investors are grown ups, they can do their own due diligence. Caveat emptor.

Similarly, Bank of America invested in Countrywide, an institution that is guaranteed to be up to its eyeballs in liability. A generation ago, no respectable institution would have touched a Countrywide until it was on the verge of collapse and the old management was promptly shown the door. In the bad old days, strategic investors cared more both about litigation risk and the possibility of tarnishing their reputation by association. No more.

So Wolf’s recommendation is uncharacteristically quixotic. Perhaps he is loath to embrace measures that have more teeth, such as having liability extend to anyone who knowingly on-sells a questionable credit. As we noted back in April, citing a Financial Times story, It’s possible to extend the scope of liability in ways that frankly aren’t crazy, given the extended sales chain in many securitizations (particularly given resecuritizations like CDOs), but they have been demonized:

The concept is “assignee liability,” and at least for now, both Democratic committee chairman Barney Frank and the ranking Republican Spencer Bacchus support it:

Spencer Bachus, Mr Frank’s Republican counterpart, has backed an “assignee liability” system that would mean investment banks that repackage mortgages into bonds would be liable to pay compensation to borrowers if loans turned out to have been mis-sold.

Why is this a radical idea? Packagers are currently at risk for the securities they sell, but their liability is to investors, basically to make sure that the securities are what the offering documents say they are. However, Wall Street firms do that by making clear legally how much of the document was provided by them (typically only the offering price and any language that might relate to the offering process, as opposed to the description of the securities or the issuer). The agreement with the issuer also limits the investment bank’s maximum liability to its fees.

The proposed legislation would make packagers like investment banks who acquire subprime mortgages from brokers and banks liable for abusive practices. And it’s almost a given that they won’t be able to limit liability to their fees. (Separately, I wonder how “mis-sold” would be defined.)

From Martin Wolf in the Financial Times:

When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you’ve got to get up and dance. We’re still dancing.” Chuck Prince, Financial Times, July 10 2007.

“A sound banker, alas, is not one who foresees danger and avoids it, but one who, when he is ruined, is ruined in a conventional way along with his fellows, so that no one can really blame him.” John Maynard Keynes, 1931.

The dance has stopped: Mr Prince’s Citigroup has just announced $6bn (€4.2bn, £2.9bn) in write-downs and losses for the third quarter of 2007. He is far from alone. More bad news is no doubt to come. As Keynes foretold, the banks joined Mr Prince’s dance together and are leaving it together. Until the dance ends, nobody knows what a bank’s profits are: bankers report (and pay themselves) on the basis of profits that are normally offset by write-offs when the bad lending comes to light.

What is remarkable about the present crisis is how traditional it is, despite the modern paraphernalia of securitised lending. We are seeing old-fashioned bad lending and old-fashioned mispricing of risk. What is remarkable, in addition, is the severity of the consequences. The US market in asset-backed paper contracted by 21 per cent between August 8 and October 1. The flight from risk also brought about big divergences between interest rates on commercial paper and US Treasury bills and between central bank interest rates and those in interbank three-month markets. This disruption, moreover, has taken place at the core of the world economy: the US housing market and debt markets of advanced countries.

I admit to both surprise and disappointment. No, I am not surprised by the repricing of risk. On the contrary, together with a host of outside observers, I was astonished by the willingness of yield-seeking investors to take on risks for small reward. If anything, the repricing has been remarkably small, at least so far: equity markets are buoyant; spreads over US Treasuries of emerging market bonds in the JPMorgan emerging market composite bond (EMBI Plus) index rose by a mere 39 basis points between July 6 and October 1; on Baa-rated corporate bonds, they rose by 59 basis points; and even on Caa-rated bonds, they rose by no more than 191 basis points.

What I am surprised by is how toxic securitisation of subprime mortgages has turned out to be for the financial markets. I admit that I thought securitisation had attractive features: it should allow banks to remain in the mortgage business as originators and intermediaries without taking too much of the interest-rate, term and liquidity risks on to their own highly leveraged books; it should allow banks to transfer those risks on to investors who want longer-term, higher-yielding assets; and, in the process, riskier borrowers should have access to more credit than before.

In 2005, Alan Greenspan himself, then Federal Reserve chairman, remarked that advances in technology had revolutionised lending: “where once more-marginal applicants would simply have been denied credit, lenders are now able to quite efficiently judge the risk posed by individual applicants and to price that risk appropriately. These improvements have led to rapid growth in subprime mortgage lending.”* Oops!

So what went wrong? There are two chief answers. The first has nothing to do with securitisation itself: it is that a fit of all-too-familiar euphoria overwhelmed both lenders and borrowers at a time of low interest rates and rapid rises in the prices of the underlying collateral (namely, housing). But the second has a great deal to do with securitisation: it is that the process of removing lending from the books of the initiators encouraged sloppy lending (“it is not going to end up on our books”) and greater belief that banks were free of the risk (“this special purpose vehicle has nothing to do with us”) than turned out to be the case.

Why did this happen? As Robert van Order of the universities of Aberdeen and Michigan points out, securitisation necessarily creates a chain of transactors where bank lending interposes just one institution between the borrower at one end and the depositor at the other. Such chains depend on trust or, as he puts it, “reliance on originators and servicers to originate good loans and service them properly”. The trust proved misplaced and has duly vanished: credit means “he (or she) believes”. Alas, he no longer does.

In trust’s absence, ignorance not only of what securitised assets are worth but also of who holds them has dried up asset-backed paper markets. That has forced banks to lend directly to the conduits, special purpose and special investment vehicles they created. The need to fund these has dried up lending and, above all, the provision of liquidity to interbank markets. One result (among many) was the collapse of Northern Rock’s business model in the UK.

An obvious reaction to this debacle is to recommend going back to the old bank-based lending model. But that would be a big mistake. The potential advantages of securitisation, vis a vis “plain vanilla” bank lending, remain, because banks are inherently so fragile. But if these markets are to recover, the errors must be fixed: first, a way must be found to demonstrate integrity of lending; second, transparency of the securities will have to increase; and, third, banks must insure themselves adequately against the need to provide liquidity to their off-balance-sheet vehicles. It is not impossible to sell complex products safely: Boeing and Airbus manage it. But only companies that demonstrably care about their reputations are able to do so. It is up to originators of securitised liabilities to do the same.

A great deal of dust still has to settle in housing markets, financial markets and the world economy. The world that emerges will look different, in many ways. But there is no reason securitisation should not become as normal and reliable an element in financial markets as corporate “junk bonds” and loans to emerging markets have also turned out to be. It is always possible to have too much of a good thing. In this case, the world has had far too much of something that was not as good as it ought to have been. But securitisation is a good thing, all the same. It can re-emerge, provided the lessons of the financial markets’ dance are learnt.

Print Friendly, PDF & Email

One comment

  1. Anonymous

    But there is no reason securitisation should not become as normal and reliable an element in financial markets as corporate “junk bonds” and loans to emerging markets have also turned out to be.

    Corp. junk bonds and emerging markets involve larger entities whom it is relatively easy to identify and research their credit riskiness. It seems to me that this is less true of, say securitized mortgages, even after, esp. after, they have been sliced and diced into different tranches.

    Perhaps if the firms originating them provided insurance where they would be on the hook if more than x% of a tranche defaulted. Or if not the originators, than independent firms (Moody’s, S&P anyone?). The insuring firms would presumably take into account not only their evaluation of a security’s riskiness, but also the reputation of the originator…

    I can already here people saying, “If the market wants this, then someone will appear to supply it… ” Just like pre-depression deposit insurance (which existed, but wasn’t up to that task). Or perhaps someone will devise a trading strategy analogous to portfolio insurance for securitized lilabilities…

Comments are closed.