FT Warns of Profligate Lending and Deteriorating Standards

Posted on by

Unlike its US counterparts, the Financial Times has consistently been on top of the various unsavory elements of the credit market bubble: the near disappearance of risk premia, the growth of leverage on leverage, the lack of investor sophistication.

A piece by John Plender does a very good job of connecting some of the dots – the explosive growth in CDOs, the role of hedge funds, the implications of the lack of mark to market in the CDO market. We’ve added some further commentary from Gillian Tett, the FT’s capital markets editor. It’s a bit lengthly, but very much worth your attention.

The fact that many (most?) hedge funds are using a known-to-be-dubious method of valuing their positions, rather than marking them to a true market value (meaning what they could sell them for) is bad practice verging on fraud (except they are exploiting a technicality, as Plender explains, that makes it kosher). It means they can carry their inventory at inflated values and collect performance fees based on those phony values.

Let’s look at this another way. One of the mysteries of the brave new world of the securitization of nearly everything is that Wall Street has managed to front-load the profits from credit extension and take a disproportionate share of it for themselves as arrangers (versus the stone ages, when banks made loans and made a simple spread over funding costs over time). Now if Wall Street takes most of the profits, who will take the securities if they offer lower spreads? They should be too unattractive to appeal to any buyers.

Well, the Wall Street firms have managed to tranche the underlying assets and cash flows so that some are triple A (investors will accept low yields for triple A paper). Who takes the rest? Apparently, a lot of the buyers are hedge funds who can make these skinny margins work by leveraging them and some financial engineering (oh, and by phony accounting too). And who is lending them the money to leverage these crappy tranches? Many of the very same Wall Street firms.

Now admittedly this is a bit simplistic; who knows what mechanisms the hedge funds are using to achieve leverage? Derivatives? Options? Swaps? Margin lending? But the general point holds: Wall Street, via its prime brokerage, is the big source of credit for hedge funds.

You can see that this looks like yet another slow motion train wreck

It’s reminiscent of a classic Wall Street joke: on a slow day, a trader has decided to open bidding on a can of sardines. A bids $1, B buys it from him at $2, and after a series of trades, E buys it from D for $5. E takes his can of sardines and opens it, only to find the sardines are rotten. E goes back to D to try to get his money back. D rebuffs him, saying, “Those aren’t eating sardines, those are trading sardines.”

The problem is that a lot of buyers of structured products will soon find that all they have is trading sardines.

From Plender’s “A stretched credit cycle, a more savage downturn“:

High finance has never been more sophisticated. Bankers have never been more clever. Yet in the US subprime lending boom, banks fell over themselves to advance 100 per cent loan-to-value mortgages to out-of-pocket deadbeats. According to industry folklore, even an insolvent arsonist was given accommodation.

Lending standards to private equity are collapsing just as risks rise and returns are being competed away. “Cov-lite” loans are the order of the day, meaning that restrictions on a borrower’s interest cover and balance sheet leverage cease to apply. This has prompted Anthony Bolton, Britain’s most admired fund manager, to warn of impending doom. So what is the explanation for such apparently aberrant behaviour? At one level, it is simply that banks no longer have to worry about loan quality in securitised markets where the loans they originate are immediately sold. So the more pertinent question is, why do investors buy from the banks? The answer, as Henry Maxey of the Ruffer fund management group argues in a forthcoming paper for the Centre for the Study of Financial Innovation, is that Wall Street has solved their most pressing problems with its invention of structured products. Take the hedge funds, in which conventional investors such as pension funds invest increasingly via hedge fund of funds. These intermediaries typically aim for positive returns of 1 per cent a month, net of fees, with low volatility. If the hedge funds they back fail to deliver on 3- to 6-month performance figures, they are culled.

The hedge funds need to make about 20 per cent gross a year, before a welter of fees, to provide that 1 per cent a month for their backers. Such a spectacular return can be gained either by market outperformance, which is beyond most fund managers, or by taking on leverage through borrowing, or trading in derivatives. For most, that means adopting leveraged strategies in illiquid assets, to avoid leaving capital values hostage to market volatility.

Structured credit products are tailor-made for this task. Collateralised debt and loan obligations (CDOs and CLOs) invest in poor-quality assets such as subprime mortgages or loans to super-leveraged buy-outs, and sell matching liabilities to investors. Yet the sale involves an alchemical transformation. The package is sliced and diced into high- and low-risk tranches, with usually up to 80 per cent being rated Triple-A or AA and the residue being very lowly rated or unrated.

For pension funds and managers of official reserves, the resulting high-grade paper is a boon in a world where the number of Triple-A corporate borrowers has dwindled to a handful.

For hedge funds the low-grade paper, which provides a cushion against default risk in the high-grade tranches, is likewise a boon, especially since, as Mr Maxey points out, it lends itself to arbitrage whereby hedge funds take long positions in the high-risk tranches and short positions in the low-risk tranches, which are relatively expensive. This ought to increase market efficiency since more investors can buy into a given pool of low-quality credit-enhanced assets.

The peculiarity of this trade is that profit is never arbitraged away in the benign phase of the credit cycle because positions are not constantly marked to market. Their illiquidity requires them to be marked to a model approved by credit rating agencies. As we saw after Enron and the subprime mortgage fiasco, rating agencies, who are paid by those who they rate, do not adjust ratings to reflect deteriorating economics. They close stable doors after profligate horses have bolted.

It follows, as Mr Maxey notes, that relaxing lending standards is perfectly rational. To increase lending volume, banks could either reduce their interest rates or reduce underwriting standards. Given the hunt for yield, this is a no-brainer. So collapsing standards will now stretch out the credit cycle while ensuring the delayed downturn will be more savage when the defaults finally happen. This subverts the argument that structured products uniformly enhance market efficiency. Credit is being mispriced, courtesy of credit rating agencies that are insensitive to market risk. Stand by for systemic consequences in due course.

Gillian Tett highlights another aspect of this situation, how CDOs and CLOs keep pumping out more product:

Once upon a time, it was presumed that the actions of central bankers controlled behaviour in the risky lending world. For if central banks jacked up rates, the argument went, the cost of borrowing would rise – making it harder for highly leveraged groups, such as buy-out funds, to snap up deals.

Now, however, this argument is looking a touch quaint. In the last couple of years, Western central banks have indeed been raising rates. Meanwhile, investors have had to contend with minor matters such as surging oil prices, Middle East turmoil, and now subprime woes. Yet, the credit party has continued, seemingly oblivious – triggering a buy-out frenzy.

So could anything else take the punchbowl away? Some bankers are now starting to mutter quietly about one risk that is not often discussed – the collateralised debt obligation world. For though the CDOs certainly do not have the debating glamour of a small war or central bank, they have helped power the credit bubble. Thus the question now is whether trends in this sector could also now deliver a jolt.

First, however, a quick finance recap: a CDO essentially is a pool of debt assets, in which investors take stakes with different levels of risk, a little like the way mutual funds operate in the equity world. They have existed for many years, particularly in the US. However, in the last couple of years the sector has exploded, particularly in Europe, where collateralised loan obligations – which buy risky loans — have spread like wildfire.

This, unsurprisingly, has roiled credit markets. After all, if a hundred new well-funded mutual funds suddenly appeared, it would not be hard to imagine the impact on stock markets. So, too, the sudden proliferation of asset-hungry CDOs has raised debt prices, making borrowing increasingly cheaper for buy-out groups. Last week alone, for example, another $4.5bn new CDOs came on tap wanting to buy assets – and another $57.7bn are now in the pipeline, according to JPMorgan.

But now there are ominous rumblings from CDO land. Rumours are circulating that some funds have suffered losses from the recent subprime debacle. While no funds have folded as a result, this has the potential to dent iconfidence (or at least prompt them to demand a higher price when they invest in these funds). Indeed, I am told some smart money is already furtively creating vehicles designed to feed on sickly CLOs. This week in London, Park Square Capital created a new credit fund which publicly declared that it expects to see a CLO shakeout – and prey on this.

But even without losses, there is another problem that might slow the pace of CLO creation. The reason why these instruments have recently spread like wildfire is that their leveraged structure typically promises returns better than traditional fixed income funds, but lower (and more stable) than those of hedge funds. But the maths only works if credit spreads are at a certain level – say, 250-275 basis points above Euribor for leveraged loans with European CLOs.

It used to be easy to get this spread. But demand for debt assets is so high that average spreads are now dropping below 225bp. The CLO industry is killing its own golden goose. This may mean that CLO launches are quietly shelved, or scaled back. In theory, this trend should be self-correcting: if fewer CLOs emerge, loan demand will fall, then spreads will rise, and more CLOs will appear. And hopefully this benign scenario will play out. But it may not, given that markets have a nasty habit of over-correcting. Worse still, the CLO sector is opaque, and pipelines operate with a long timelag. It is thus entirely possible to imagine a scenario where CLO activity suddenly collapses, producing a shock for debt prices, which could be a tipping point for credit markets that are already overstretched.

Let me stress, this is not my central, short-term prediction. But the moral is clear: to understand the financial imperatives behind the buy-out boom, do not just blame central bankers (or the Chinese, for that matter). Instead, bone up on the structured finance world as well – with all its alphabet soup.

Print Friendly
Tweet about this on TwitterDigg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone