Collateral Damage

Reddit

‘The Great Moderation’ is the name given to the time leading up to the 2008 financial crisis. (Different people give it different starting dates.) The idea is (now was) that regulation and central banking had given rise to stability and growth.

Part of this moderation was attributed to derivatives. By allowing companies to hedge the risks inherent in their primary business, derivatives were supposed to make those primary businesses safer.

A company dealing in lots of foreign cash could secure an exchange rate by buying a foreign exchange derivative. Pension funds could insure their portfolios by buying options. Insurers could insure themselves against wild weather by buying a weather-linked derivative.

But the practice didn’t just allow risk to be mitigated. Someone had to take the opposite side of each trade. So, pretty quickly, derivatives got turned into financial gambling steroids. They allowed leveraging of positions without actual leverage. Heck, they changed the very definition of ‘leverage’. It used to mean using debt. Now it describes bets that move in multiples to the underlying asset’s movement.

Aside from derivatives, which are of course a major factor, there is another story that allowed The Great Moderation to take hold and then blow up. And it’s one we haven’t really read about. At least not with this angle.

When people engage in transactions, they often demand collateral from their counterparty. The idea being to reduce the risk of something going wrong. ‘If you don’t pay, then I have the right to sell your house.’ Or something like that.

Banks and financial institutions often use collateral to back up promises. And many companies hold safe assets on their balance sheet as a type of collateral to creditors. If the company fails, there are plenty of assets to go around. It may not be collateral in the strict sense of the word, but the effect is much the same: to reduce counterparty risk… the financial fallout of the other person not doing what they are supposed to under the agreement.

But using collateral exposes you to a whole new risk. If housing is used as collateral, then house prices are a risk. If collateralised debt obligations (CDOs) are used as collateral, then you are exposed to CDO price risk. If an implied government backing, or ‘too big to fail’ status is used as a type of collateral, you are exposed to political indecision risk. (Ask Lehman Brothers creditors about this one.) If access to the central bank’s discount window is used as a type of collateral, you expose yourself to risks in the discount rate.

In other words, reducing counterparty risk by using collateral, or something that has the effect of collateral, may leave you with a new type of risk that can infect the rest of the transaction or relationship. Collateral can add to your overall risk instead of reducing it.

So, what do you do? Not use collateral? More on that below. First, let’s look at just how prevalent this issue is. Based on Ben Bernanke’s comments at the Financial Crisis Inquiry Commission, the Tri-party Repo Market in 2008 was a prime example of how collateral can turn on you. With no small consequences.

‘ … runs in the tri-party repo market, where what we used to think was very stable funding, which is funding through repurchase agreements where the investment banks would put out assets overnight and use that as collateral, they thought that was a pretty much foolproof form of short-term funding.’

To clarify, a repurchase agreement, or ‘repo’ is like a short-term loan with collateral. You sell assets with the agreement to buy them back. If you can’t finance the repurchase, the holder of the collateral is left holding assets. Which is much better than a claim on assets as with normal lending agreements. Bernanke continues:

‘But in a crisis where people began to doubt the liquidity or the value of those assets, the haircuts went up and you got into a vicious cycle which led to the Bear Stearns collapse and was important in the Lehman collapse as well.’

In other words, the collateral used in the repo agreements added risk instead of reducing it. The value of the assets fell, making the repo transactions dubious. That interrupted the funding structure of two major investment banks, marking the onset of a major financial crisis.

Many of the assets used in these repo agreements were linked to real estate and mortgages. This is the crucial link between sub-prime lending, securitisation and the collapse of financial institutions. They used securitised loans as collateral in their repos. Assets that were conveniently rated AAA by credit ratings agencies, which is a common requirement for collateral in repo markets.

Had Bear Stearns borrowed in straightforward lending agreements without collateral, the falling asset prices of the collateral used may not have caused the collapse of the company.

This is of course an over-simplified view, as Bear Stearns would have had the assets used as collateral on its balance sheets instead of in the repo market. But the risk would have been within the company, which changes the nature of the risk. For example, by using housing-linked assets as collateral for so much of its funding (enough to cause it to fail), Bear Stearns was playing the game with an open hand. Its creditors knew the assets Bear Stearns held, as they themselves held them as collateral. The famous secrecy with which investment banks conduct their affairs allows them to hide such structural weaknesses.

Also, when the repo market dried up, Bear was forced to sell exactly those assets that were being used as collateral in the repo market. This pushed the price down further, exacerbating the problem. The predictability of this sequence of events from the perspective of Bear’s repo counterparties is a major factor to the sudden loss of willingness to lend to Bear.

If Bear had had access to a lending facility that was not as directly linked to the assets it held on its balance sheet, its funding crisis may have slowed significantly. Whether it could have been saved is pure speculation.

It could be argued that by relying on funding that is directly linked to assets held (by using them as collateral), a company is asking for trouble. If something goes wrong with the assets held, not only is the balance sheet in trouble, but funding the balance sheet is too.

Remembering the point made above about derivatives providing leverage, this is much the same concept, but applied to corporate finance. It is not traditional leverage, although that forms part of the equation. It is a speculative type of leverage, which changes the odds of the game.

By tying funding (via collateral) and profit (via assets held) to the same set of assets, you get leveraged profits. In a good year, the price of the assets go up, improving the quality of your collateral, which lowers funding costs. The gains in the value of assets held also show up as revenue. A double-sided benefit. In a bad year, the collateral is perceived risky and funding costs rise, while the falling price of assets pushes down revenue. Good and bad are exacerbated. Much like with the derivatives used to earn multiples of returns, rather than the inherent return of the underlying asset.

The idea that collateral can make things safer by reducing risk is a remarkably similar fraud to the derivatives story in many other ways. But an even closer parallel is likely to be found in your own portfolio.

How many times has your financial advisor or broker told you to hold a diversified portfolio? Well, aren’t you just assuming a larger number of different types of risk each time you buy a stock different to the ones you already hold?

If you are optimistic on retail and your portfolio consists of retail stocks only, then you are exposed to risks affecting retail. If you buy mining stocks, you take on the risk of mining as well. You may be less exposed to risk in retail, but you now have two sectors of the economy for which bad news will affect you. To do well, you have to be right on both counts.

As for the mathematical justification for diversification, your editor has studied it. And it relies on many theories that are now considered laughable by anyone who has followed the news during the financial crisis. But even before 2008, the idea that diversification was a good idea never really held up. For example, when your broker shows you how well the All Ords has performed over the long term, ask them what happens to the companies that fail and drop out of the index. What if you had bought and held them? How would your portfolio look compared to the index?

The source of capitalism’s strength is that everyone is not in it together. People can dissent. This allows those who are prudent to do well, and those who aren’t to be taken over by better decision makers. If you force all to sink or swim together by assuming each other’s risk, you lose this dynamism.

The purpose of collateral is to soften those effects of capitalism. Some would call it risk management. But, as mentioned above, it often fails miserably.

The solution is painfully obvious. Use collateral that is inversely correlated to your existing risk. Cancel the risk out. Do the opposite of firms like Bear Stearns. But now we are back to where we started – derivatives. The agreements designed to hedge risk, which can be abused to gamble. Collateral can be used in much the same way.

But many people do use derivatives well and responsibly. So collateral can be used effectively too. And it should have similar attributes to the way derivatives are used well. The assets used as collateral should be inversely correlated to your inherent risk.

Otherwise, don’t call it collateral.

So, we’ve had derivatives and collateral blow up. What’s next? What asset class is currently being designated as ‘risk mitigating’ when it is actually increasing risk.

In keeping with the idea that the economy is going from bad to worse and that the risk has gone from the balance sheets of the banks to the balance sheet of governments, bonds are the place to look. US treasuries are used by finance professionals as the risk-free benchmark. And what could better blow up than a risk-free asset?

But how? Well, like the risk-mitigating abilities of derivatives and collateral encouraged bad behaviour, so too do government bonds. Designating the bonds ‘risk free’ created massive demand for them, which has allowed governments to spend and borrow their way to presidencies and prime-ministerships. But, like with derivatives traders and collateral users, the concept was taken too far.

Now the so-called risk-free asset class looks like it could bring down the system completely. It will make derivative and collateral collapses look boring when it does. Japan, Europe and the US are playing a combination of spin the bottle and Russian roulette on this one.
Nick Hubble
For Daily Reckoning Australia

Nick Hubble
Nick Hubble is a feature editor of The Daily Reckoning and editor of The Money for Life Letter. Having gained degrees in Finance, Economics and Law from the prestigious Bond University, Nick completed an internship at probably the most famous investment bank in the world, where he discovered what the financial world was really like. He then brought his youthful enthusiasm and energy to Port Phillip Publishing, where, instead of telling everyone about The Daily Reckoning, he started writing for it. To follow Nick's financial world view more closely you can you can subscribe to The Daily Reckoning for free here. If you’re already a Daily Reckoning subscriber, then we recommend you also join him on Google+. It's where he shares investment research, commentary and ideas that he can't always fit into his regular Daily Reckoning emails.
Reddit

Leave a Reply

2 Comments on "Collateral Damage"

Notify of
avatar
Sort by:   newest | oldest | most voted
Stewart
Guest

That’s a fantastic article – will share it with my Dad!

Thanks!

Roy Zimmerhansl
Guest
I must disagree with several aspects of this article. The author mixes and matches issues as if they are interchangeable. Safe assets held on a company’s balance sheet aren’t collateral, they are part of the company’s assets and subject to a credit assessment process. Unlike “real” collateral that is held by creditor counterparties, from the moment the balance sheet is printed, it is possible that any of the “safe” assets could be disposed of rendering them meaningless as even an indirect form of collateral. While I agree that holding collateral does expose the holder to risks associated with that collateral,… Read more »
wpDiscuz
Letters will be edited for clarity, punctuation, spelling and length. Abusive or off-topic comments will not be posted. We will not post all comments.
If you would prefer to email the editor, you can do so by sending an email to letters@dailyreckoning.com.au