In the wake of the Lehman minibonds debacle a lot has been said. A vicious blame game has begun between the banks and their investors — should investors be blamed for not reading the danger signs, or were they wrongly informed by bank salespeople? Amid the discussion, a common consensus that has emerged though. To prevent such a fiasco from repeating itself, some have suggested some kind of agency be established to rate the safety financial products in the future, so consumers can make better informed decisions about the risk they make. This has been raised in the ST forums, as well as by Andy Ho in yesterday’s column.
If only life were so simple.
As logical as this sounds, actually creating a successful agency isn’t as simple as it seems. In fact, it’s probably impossible to come up with satisfactory rating system that rates products accurately, like every other thing in Singapore.
The reason for this is the risk linked to financial products is ever-changing and almost impossible to quantify. Andy Ho has compared such financial products to “toys” and “microwaves” lacking proper safety measures, but this is a false comparison. The risk inherent in a microwave exploding is very much unlike the risk inherent in a bond defaulting, for two reasons.
The first is complex interdependency of financial markets. To borrow George Soros’s theory of reflexivity vaguely, markets are the result of the constant feedback between participants biases and their actual behaviour: thousands of interactions taking place in an instant, affecting each other interdependently. With an infinite number of possible permutations, this makes calculating risk accurately a nearly impossible task.
Furthermore, the second challenge in calculating risk is the human factor in markets. One of economic’s greatest assumptions — the assumption of rational behaviour — also is one of the subject’s greatest flaws. Hence, while there are so many experts in the field of statistics, econometrics and finance soothsaying about the market, only one Warren Buffett exists. As Keynes rightly pointed out, the market is often governed by emotions and gut feelings (“animal sentiments”) and not rational behaviour assumed in models. So tell me: how do you articulate animal sentiments into a mathetical equation?
The point here is that is almost impossible to gauge an appropriate level of risk. Risk in itself is a constantly changing thing. I often take ratings by agencies like Moody’s with a pinch of salt — witness how many times “AAA” products have turned foul within moments. The logical response is to perhaps factor in the fluctuations arising from market uncertainties, but if this method is used, then only ultra-conservative estimates are likely to result. This would defeat the value of even using ratings in the first place, with all products automatically downgraded.
Instead of establishing a ratings agency, what should be present are a clear set of laws which prevent abuses of consumers. By abuses, I mean portraying incorrect information, or in some cases, incorrectly portraying correct information. In the first case, for example, I personally know someone who was told by the bank last year that these minibonds were “risk-free”. This is portraying incorrect information, as even the fine print did not guarantee a full return of principal. In the latter case, sometimes financial product sellers only reveal select pieces of information, painting an incomplete picture. This too is considered consumer abuse. Laws need to make sure transaction processes are more transparent, so that both sellers and buyers of these products are properly informed of the products they buy, and in the case of abuse, proper penalties are awarded. Other than this, there’s really little else that can be done.
-A.S.
2 responses so far ↓
chue // November 9, 2008 at 12:33 pm |
For all that has been said about the futility of calculating risk, one should be reminded that we are living in what is perhaps an once-in-a-century economic occurrence, where human emotions extend beyond the historical quantifiable mean range (explaining the unprecedented volatility within financial markets).
It is of course possible to extend any risk model to take into account such extreme human exuberance and pessimism, but that would ultimately create a final risk value that is artificially high and inapplicable to 99.99…% of all economic events.
I believe that the crux of the issue lies beyond the revamping of existing risk models or regulation, but rather on the ability of financial engineers and economists to switch between various risk models and the transition process (which itself can lead to artificial non-human driven volatility within the financial system).
Cuthbert // November 25, 2008 at 5:33 pm |
Two things I’ve read recently about ratings and risk struck me.
The first is what Robert Skidelsky describes thus: ‘…the redefinition of uncertainty as risk… was the main achievement of mathematical economics. Whereas guarding against uncertainty had traditionally been a moral issue, hedging against risk is a purely technical question.’ (Article: http://www.koreatimes.co.kr/www/news/opinon/2008/11/137_34874.html)
There is only one sure thing about uncertainty, and It undermines any attempt at quantification. When uncertainty became risk and risk became a calculable probability, no-lose situations become mathematically possible, as long as the volume is available. I.e., if there’s more risk, hedge more, because although hedging is expensive, if you hedge when things get bad and limit your losses and do well during the good times, the numbers work out in your favor. Doing this is difficult, but that probably makes it even easier to live by. It is also false, because of the reasons pointed out in the post, and because of the nature of uncertainty in itself. Skidelsky writes about the excesses and moral hazards such a mindset leads to.
But even while writing the above, the second thing I had in mind was gnawing away at me, because all the above discussion seems idle in the light of the possibility, or fact, that: the ratings are bunk.
“For instance, he knew that the big Wall Street investment banks took huge piles of loans that in and of themselves might be rated BBB, threw them into a trust, carved the trust into tranches, and wound up with 60 percent of the new total being rated AAA.
“But he couldn’t figure out exactly how the rating agencies justified turning BBB loans into AAA-rated bonds. “I didn’t understand how they were turning all this garbage into gold,” he says. He brought some of the bond people from Goldman Sachs, Lehman Brothers, and UBS over for a visit. “We always asked the same question,” says Eisman. “Where are the rating agencies in all of this? And I’d always get the same reaction. It was a smirk.” He called Standard & Poor’s and asked what would happen to default rates if real estate prices fell. The man at S&P couldn’t say; its model for home prices had no ability to accept a negative number. “They were just assuming home prices would keep going up,” Eisman says.” (Article: http://www.portfolio.com/news-markets/national-news/portfolio/2008/11/11/The-End-of-Wall-Streets-Boom%3Fprint%3Dtrue#page5)
Again, the ratings are a technical question, since the pension funds and other types of funds I’m not familiar with have regulations that only permit them to invest in safe, risk-free AAA bonds. As to the credibility of the claim made in article, read it and judge for yourself.
The ratings system would ideally function as an expression of a complex range of information and factors in a simplified form. Ideally, the simplified expression would be accurate for potential users’ purposes. I suppose there’s a conflict of interests as well. The companies submitting products to be rated depend on having good ratings, to increases potential profits or to gain an edge over competitors. Usually the volumes of money involved are so large that they have a huge incentive to… affect… the ratings they are given. For buyers of rated products, if the ratings happen to be wrong, it’s their own money they’re losing, and ratings were only ever a loose benchmark anyway, although I suppose technical factors might magnify the relevance of ratings somewhat. The buyers have interests at stake, but the incentive to ensure some sort of accuracy in ratings isn’t huge, because their interests are too diverse and diluted. Ideally, ratings agencies view all interests objectively.