Risk Management, Or Risk Manipulation

Value at risk (VaR) financial models are the latest game being played by those on Wall Street who profess to manage risk, a troubling trend detailed superbly by Joe Nocera in a January 2 New York Times Magazine article. Such models give bankers a false sense of confidence in their risk control while, in reality, they increase the level of risk for society as a whole.

But Nocera understates the problem. The risk management groups on Wall Street are actually engaging in risk manipulation, risk distortion, and risk amplification — anything but risk management.

Public perception is that Wall Street didn’t do much risk management over the past decade, or perhaps longer, resulting in the profound credit crisis that wiped out many financial firms and left others precariously hanging on. But the problem is not that Wall Street didn’t have people monitoring risk. Almost every firm hired scores of risk managers during the last several years, with some being paid millions of dollars a year. The problem was that the more people they hired and the more VaR financial models they ran, the worse their understanding and assessment of risk became.

Why so? There are two main reasons. First, the structure of VaR models is not based in reality. They place too much faith in the fantasy of mathematical algorithms to explain the behavior of human beings. They assume human behavior can be modeled as accurately as launching a rocket — that we can predict its path and outcome 100% correctly. It’s no coincidence risk managers are often called rocket scientists — they treat people like physical objects. Is human behavior really that predictable? Are risk managers so crazy as to think human beings behave like a mindless, computer predefined rocket? Does human behavior obey math principles or is it the other way around?

Most financial models rely on theories of probability and statistics. In modern physics, quantum mechanics relies heavily on statistics as a way to explain cause and effect. But the financial world is no science experiment; everything is for real. You can never go back to do it “right” and repeat an “experiment.” Things might work one time but may not work the next time. When a physics-like approach is applied to financial products whose value is heavily tied to human actions, like mortgage prepayments, it becomes a computer game of garbage in and garbage out.

Or worse, it becomes a self-fulfilling prophecy. As risk managers used financial models to come up with VaR for toxic products, iterating to arrive at what they believed were successively more accurate estimates, they developed a false sense that they were actually in control. They believed they could accurately predict every possible cash flow scenario for a mortgage-backed security, as well as its probability distribution. The CDOs and the credit default swaps created through this process embedded a level of overconfidence which killed the whole industry. You can always fool many people for a long time, especially when you become a fool yourself.

For a time the VaR model seemed to “work,” but it failed exactly when it was needed the most. As hedge fund manager David Einhorn said in Nocera’s article, VaR is “relatively useless as a risk-management tool and potentially catastrophic.” Why so? Because we will never be able to understand and assess the true nature of supposedly rare catastrophic events. Statistically this is the “fat tail,” an event which happens a lot more often than we perceive and put into VaR models. Second, when it happens, its consequences are catastrophic, potentially putting everyone out of business. Computer models cannot handle this kind of discontinuity, which is a little like a number divided by zero. As Nicolas Nassim Taleb said in the article, “In the real world, the magnitude of errors is much less known.” If you don’t know the true probability and potential damage, you might as well throw the whole VaR model into the garbage. To instead use it to manage risk is absurd.

But it is worse than Nocera described. The second reason for the failure of risk management is that financial models were all based on assumptions. It was too easy to twist a few of them to produce the desired outcome. Risk managers felt they are infallible, to the point of feeling like gods. They justified any rating for their CDOs or predicted any MBS default probability and payment schedule they wanted. If too much risk was calculated by the model, no problem; they just twisted a few assumptions in the Monte Carlo simulation of the VaR model and then re-ran it. Suddenly the distribution graph showed the exact curve they needed. This transformed a game of false but honest assumptions into much more insidious risk cover-up.

Most of the time, common sense dictates whether you are adding or reducing risk, without even running any models. For example, when a former high level executive of Citigroup (C) pushed the firm to get into the exotic derivative areas of MBS, CDOs and CDSs, even naïve observers knew Citigroup was adding risk to its portfolio. But by using some “magic” financial models, the risk management group and their “renowned” consultants were able to show the Board of Directors that Citigroup was not taking any more additional risk and, even if it was, it could be diversified away through their global supermarket portfolio. Risk managers twisted the model to produce the desired future outcome, and they used financial models to justify a huge amount of risk that has since wiped out their shareholder value many times over. In another example, after AIG (AIG) repeatedly assured investors there was no risk at all from their CDS portfolio, with a risk model to back up their counterintuitive assertion, a very small financial product group ultimately wiped out the financial conglomerate.

The seductive elegance, overconfidence and abuse inherent in financial modeling are at least part of the reason for the current credit crisis. The more risk managers hired on Wall Street in the years running up to the crisis, the riskier the firm proved to be. Just look at Citigroup. How many of its employees and consultants have been, and are still, doing risk management one way or another? When top management relentlessly pursues quick profits by taking on more risk, risk managers become puppies. Eager to please their managers, they use their expertise to cover up risk rather than expose it. Computer models become their prime weapon.

Outside of risk management, financial modeling is also heavily used in portfolio return analysis and forecasting. For most of the last ten years of the Greenspan era, a big myth — or “theory” — was that low cost of capital (which Greenspan achieved by relentlessly driving down interest rates) would lead to improved return on equity (ROE). Many people used financial models to justify or “predict” a value for the Dow of 36,000 or even 100,000, a so-called paradigm shift of ROE. Suddenly companies got all the free capital they wanted, leveraging their ROE (ROE is a leveraged factor in the capital structure). The sky was the limit for the return to shareholders and for their stock prices. And it was supposed to go on forever. No longer human beings living on Earth, investors became in their own minds powerful angels who could do no wrong, led by the maestro Greenspan. When too many people (and their computer models) told the same lie, the lie itself became the truth. How could Greenspan and so many other very smart people suddenly forget the very basic economic rule that low cost of capital will eventually lead to zero return on equity? That is a fundamental principle of capitalism.

Another myth of the last decade was that using financial models in dynamic asset allocations could improve performance. The Yale and Harvard endowment funds used dynamic asset allocation to invest in private equities, hedge funds, real estate and timber. Other endowments followed their lead to “diversify” and “rebalance” their portfolios whenever dictated by their computer models. But they failed to realize that most of those assets are illiquid, and when everyone is dumping them at the same time, it is a downward spiral or worse, and there may be no way out. Computers are notoriously bad at modeling liquidity. This was a critical lesson of the program trading and dynamic hedging that caused the 1987 Black Monday market crash. As Jeremy Grantham of GMO has said, in the long run, human beings learn nothing from history, and 1987 was just two decades ago.

In a certain sense, the liquidity crisis of the last six months was inevitable. Wall Street got complacent with computer models, and nature came back to punish them (and the rest of us) for shrugging off the resistance to modeling of a key factor: liquidity. Computer models depend on the assumption of a continuous market, with a balanced equilibrium between buyers and sellers. A situation where all the liquidity is sucked out of the market destroys the value of all those exotic paper products. We do not need a bunch of highly paid math geeks to run millions of Monte Carlo simulations to tell us that. A computer can never replace common sense.

Now we have another Fed Chairman who only knows how to print more money, then print some more, and expand the Fed’s balance sheet ever wider. Bernanke drops the money at only one location, Wall Street. Being an economist and renowned monetarist, he must know that excessive printing will eventually lead to zero value of the fiat currency, the US dollar, just as low cost of capital eventually leads to zero ROE. If that is the inevitable outcome, the government should drop money on the middle class and the poor, not the super-rich bankers on Wall Street. Since ten times zero is still zero, what difference does it make? In addition to being a politically popular move, this might even avoid a few incidents of social unrest.

So-called “extreme” events with “low” probability happen more often than people perceive in risk management. When they occur, an unforeseen tsunami of incalculable magnitude results, destroying wealth on a scale from which it may take a generation or two for the economy to fully recover. Meanwhile, you can pretty much throw risk management models out the window. They do more harm than good.

Comments

[Amouna:]

Great article, thank you very much!

Nassim N Taleb has been a notorious critic of the VaR approach to risk management with his Black Swan theory, which advocates exactly the type of scenario we are in today. At the same time, I would have to change career too...:)

Jan 30 11:12 AM

[Tony Petroski:]

In a bull market, all the models work well. In a bear market...

Jan 30 11:32 AM

[terryt:]

VaR model is used to measure risk in a normal market. for a market like this, VaR, or perhaps any model, will be out of whack given most of them are still using Gaussian normal distribution to derive probabilities.

the bottom line is that, to simply blame models is pointless, they are just math formulas. The key is to know how and when to use them. It's people/organization issue, not a model issue.

Traditionally, the reward system on Street is heavily tilted toward risk taking. Risk managers are powerless even they have identified the potential risk exposure, because the business heads would simply ignore the warnings. Remember, they are not getting paid by how much losses COULD be potential avoided but by how much money WILL be made.

Jan 30 11:52 AM

[Geoff Consi...:]

Sir:

Your perspective is a common one--and Nassim Taleb has been banging this drum for quite some time. There are two questions that need asking:

1) Do the models work if used correctly?
2) Do people have an incentive to misuse the models?

Your answer to both appear to be no--but you provide no evidence that actually refutes (1). Certainly, we know that many firms have used risk management to obsure risks--to their own benefit. But there is a severe logical flaw in then assuming that you have addressed #1.

When we have people like David Swensen and Ray Dalio who have consistently made money over decades relying on good quantitative portfolio management (although both took their lumps in 2008), we would be fairly foolish to ignore these guys and simply jump on the bandwagon that risks are not quantifiable and that portfolio analysis is worthless.

Frankly, I will throw in my lot with Dalio and Swensen over Taleb any day of the week.

Jan 30 12:51 PM

[Thomas D:]

Great article.

The underlying issue with most risk management practices today is that the objective has changed. Risk management primary focus was to guarantee the financial survival of the company at any time, especially against unforeseen events. It then became a methodology to better leverage capital and maximize investment return. A completely different goal. Add insult to injury, when the new models (like VaR) don't work or produce different results, the explanation is that it was caused by unforeseen or unpredictable events!

The fact that so many financial institutions have either go bankrupted or asked for government bailout demonstrate that the new risk management philosophy is widely used.

Jan 30 12:55 PM

[Heaven, Hel...:]

Excellant article which goes to the heart of the matter and highlights what has always bothered me, namely that the models break down just when you need them the most. This is part of man's age old yearning to reduce factors which are inherently "grey" into "black and white". Qualitative elements can not be distilled down into a mathematical equation.

I have been shocked more than once by the answers to questions I have posed to gatekeepers of VAR models, my favorite inevitably being that VAR models assume that the portfolio can be liquidated within a short 1-8 day time period, depending on the algorithm, and that this will occur at a normalized bid-ask spread.

What a false sense of security they indeed have given. A true "shortcut to thinking" for the end users of the data.

Jan 30 01:08 PM

[jayjayjay1212:]

I agree with this article in general. VaR, if used blindly, is indeed quite useless. However, I would like to add a few points, if I may, to give a few counter-arguments to the ones you made:

1. VaR models are ex-post models of risk. Everyone knows that you can't exclusively use the past to predict the future. However, the past can give indications of general patterns in asset returns over time, which can help. Because of that, most if not all firms use VaR in conjunction with many ex-ante (ie forward-looking) risk models. Those models normally integrate implied volatility curves, correlations and many other qualitative inputs to create a picture of the future risk portrait. Monte Carlo simulations are also widely used.

2. VaR models are not unique and differ in quality. Indeed, there is no definitive, objective and straightforward way to run your VaR model. The assumptions made are a major aspect of the end result. This means some are better than others, and that the quality of the management team will determine whether the VaR model is realistic or not. Obviously, companies with poorer VaR models (and poorer risk management in general) did much worse than the ones with better models. The better ones incorporated the real risks associated with options writing and selling credit default swaps. Poorer ones viewed them as steady gainers, where the losses fell in the ''unnormal'' part of the distribution. Knowledge of financial products along with common sense made it clear that CDS and options increase risk. Which leads to my next point:

3. The Board of directors of many firms lacked the knowledge and/or courage to question risk models and decisions made by management. The best example is with Lehman Brothers ; their board members were mostly from a non-financial background, were friends of the CEO, were advanced in age and inexperienced with the characteristics of the exotic products the firm was venturing into. Corporate governance was as much if not more to blame than risk management failures in Lehman's case.

Bottom line is VaR alone is a horrible way to manage risk. However, quality VaR models, in combination with ex-ante risk models, competent management teams and good corporate governance are all important factors in successfully managing the risks of a financial institution.

Jan 30 01:18 PM

[Tradememe:]

It's not that Wall Street didn’t do much risk management, but rather it took too much risk. The derivatives are supposed to enable the sharing of risks, thus making the system more robust. Instead, it resulted in too much risk-taking.

Jan 30 01:32 PM

[Chris Monoki:]

Mr. Tan:

One of the best summaries of the risks (ironically) using value-at-risk modeling. Thank you.

Modern portoflio theory an mean-covariance matices have their place, but this episode highlights some serious flaws that will ultimately be rehashed, with in our profession and in academia. Perhaps other thoughts, perhaps coming from Mandelbrot and Taleb, even Soros (reflexivity) will be raised in seriousness. They seem to be paving a better path than the Bauchalier group.

Again, thank you.

Keep pressing,
Chris Monoki

Jan 30 01:47 PM

[WhoWantsARi...:]

VAR critics often use vacuous terms like "competent management teams and good corporate governance" as an intellectual crutch. Oh the brilliance of hindsight.

Jan 30 01:56 PM

[Nick Waddell:]

Great article. I highly recommended Niall Ferguson's "The Ascent of Money" which is showing on PBS right now. Niall interviews George Soros on these matters and it would be a nice companion to this article...

Jan 30 02:01 PM

[riskmgr:]

Every time some idiot steps up to the plate to talk about the shortcomings of VaR, which are considerable, they consistenly neglect to mention sensitivity analysis as a necessary component of overall risk management. Sensitivity analysis can revalue securities based upon shifts in the market, with out without using a covariance matrix. A simple parallel shift of equities / commodities to -20% would give you a pretty good idea of how your book would perform during an outlier event that Taleb is so obsessed with running his mouth about. At outlier events, correlations turn to 1, and a simple parallel shift is very effective at predicting the magnitude of losses given extreme market movements. But oh, let's not talk about that, let's talk about how VaR sucks. If you aren't a sophisticated enough investor to understand the full range of tools available to risk managers or if you are a simpleton who expects to look at one number and be able to discern a complicated aggregate risk picture, you and everyone else using this novice approach will lose all their money to more sophisticated investors in Darwinian fashion. So yeah, go ahead and throw out your risk models, and you can hand me your wallet while you're at it.

Jan 30 02:42 PM

[Dkiral:]

In 2006, a Economics Professor that I know from U Chicago, stated that when US data was entered into the models used by investment bankers to evaluate developing markets, the models predicted that the US would be next Argentina. U C Economists opinions on the impact of the findings was divergent. Those in the International School, believed that the models preductions were correct. Those in the monetarist school believed that since the US holds a unique position due to its currency, that those models could not be relied upon. He stated that at the time, there was so much divergence of opinion, that the Economists did not even know where to begin discussions. So, perhaps it wasn't the models, but Friedman's free market dogma that did us in.

Jan 30 02:47 PM

[Chuch:]

everything is manipulated now a days, should be a scary week next week, see here crashmarketstocks.com

Jan 30 03:21 PM

[morph366:]

I enyoyed reading this article very much. As you have suggested financial time series data cannot be modeled with the same tools designed to study inanimate objects.
Market liquidity/illiquidity is one of the real enigmas in this and I have never encountered a satisfactory explanation from anyone in academic finance as to how to anticipate when any market will lose all of its normal fractiousness and suddenly become completely illiquid.

Jan 30 03:22 PM

[Xxantu:]

Use of a risk models itself is the inherent flaw. Regardless of how well they work during stable periods the market is inherently unstable and their scope is limited. The models are essentially a dark, dense group of trees that keep us from seeing the forest.

And even if we could see through all of the trees that we keep planting in front of our faces, who can keep track of them all?

Jan 30 03:27 PM

[Chris B:]

"First, the structure of VaR models is not based in reality. They place too much faith in the fantasy of mathematical algorithms to explain the behavior of human beings."

To create a frequency distribution, you use historical data. Thus, it is ironic that the companies that used VaR were the same companies that printed "past results are not indicative of future returns" on their prospecti and account statements. They apparently did not believe that was true. As we see now, black swan events happen somewhere every day.

"When top management relentlessly pursues quick profits by taking on more risk, risk managers become puppies. Eager to please their managers, they use their expertise to cover up risk rather than expose it. "

And why would top management have an interest in pumping up short term earnings while taking on massive risk? Could it have anything to do with short-term stock options being used as compensation for short-term managers? It's a game of musical chairs where the economy always loses.

"As Jeremy Grantham of GMO has said, in the long run, human beings learn nothing from history..."

We certainly didn't learn the lessons of our grandfathers in the depression. In 1999, we repealed the Glass-Steagal act of 1933, allowing the combination of investment and commercial banking for the first time since the depression. A few years later, the results are the same: investment losses undercut commercial lending, leading to a credit crunch, leading to layoffs, bankruptcies, and bailouts.

Meanwhile, investors again failed to learn to ignore pundits and advisors who call diversification a dumb idea and pushed complex, counterparty-intensive debt instruments as safe.

Jan 30 03:36 PM

[Whitehawk:]

Good article, but please reconsider having Bernanke drop that printing press cash on the middle class instead of Wall Street in a fiat of modern Keynesianism. It won't prevent malaise as you suggest. We need a contraction of gov't spending, corporate tax rate cuts, and a reduction of monetary regulation. The Chinese must be watching all this mess over here with some degree of amusement.

Also: as a physicist I disagree that models can't work in modeling risk, the problem lies with unphysical inputs and assumptions: one cannot assume asset prices will keep ever increasing without considering the possibility of rapid disinflation due to a bubble being pricked. Yes, many have not learned from history, even yet. That means better opptys for the rest of us.

Jan 30 04:16 PM

[morph366:]

Re Whitehawk
I'm not sure what point you are making when you say that the models cannot work because "the problem lies with unphysical inputs". It would seem to follow that you are then suggesting that the mathematics and underlying probability assumptions are valid for measuring risk but not for risks where human beings involved.
If this is the point let's just admit that there is no model for quantifying risk in financial markets - because at least at present humans are involved. If at some future time some smart genetic engineers design creatures with a new kind of algorithmically deterministic consciousness the world will be safe with a VaR risk model.

Jan 31 05:50 AM

[prudentinve...:]

Another case of not seeing the forest because of the trees.

Any model output, depends on its underlying assumptions. "Garbage in garbage out" is an old saying. Employing physics or math geniuses who can create sophisticated models is worthless if the assumptions are false.

The assumption, that home prices can rise indefinitely at a rate faster than GDP per capita is manifestly absurd, as anyone who has a calculator with an exponential function (and know how to use it) can discover. It does not math genius to conclude that lending based on such an absurd assumption is a bad idea.

Jan 31 08:59 AM

[Dividend Tree:]

The author fails to mention that a fundamental level, models were used to justify risk taking (by purposely hacking to show low risk). Models are as good as one wants it to be. If wall streets (and people like you) are hell bent on negating the modeling tool, why the heck wall streets engineer water tight pay/bonus packages for themselves. Why was not linked to the company performance ? Model are great tools, depends on how one wants to use it. A screw driver can be used to plug a hole (or it could be used to make a hole). Wall street used it to "make a hole". They took what came out of that hole, and ran when it became a deluge.......

Jan 31 10:02 AM

[foxy 44:]

Now we have another Fed Chairman who only knows how to print more money, then print some more, and expand the Fed’s balance sheet ever wider. ""Bernanke drops the money at only one location, Wall Street"". Being an economist and renowned monetarist, he must know that excessive printing will eventually lead to zero value of the fiat currency, the US dollar, just as low cost of capital eventually leads to zero ROE. If that is the inevitable outcome, the government should drop money on the middle class and the poor, not the super-rich bankers on Wall Street. Since ten times zero is still zero, what difference does it make? In addition to being a politically popular move, this might even avoid a few incidents of social unrest.

we hope so Obama correct the target, if not, good luck América!!!!!

Jan 31 10:25 AM

[Moon Kil Woong:]

Risk assumptions presume a standard calculation for risk usually based on historic trends. They can't and don't account for uncertainty, and most certainly don't properly account for catastrophic fraud and default risk. Why should they? How can they? If they did, economics would grind to a halt because the cost would always be very very high making it impossible for a lot of standard business from being done. Spreads would widen...

What am I talking about. Just look around at what's happening today. Now that's not risk, it's uncertainty and it's causing a lot of economic problems.

Thanks for the good article Thomas Tan.

Jan 31 02:01 PM

[E.D. Hart:]

"Computer models depend on the assumption of a continuous market, with a balanced equilibrium between buyers and sellers."

This is the crux of the problem, in my view. Assumptions about equilibrium are mostly true, until they are wrong. Markets behave as adaptive complex organisms that never really achieve a state of true equilibrium.

Risk modeling works most of the time, under the vast majority of conditions. Until they don't, then all @$#5 breaks loose.

Jan 31 05:06 PM

[littleBonap...:]

i was working in the credit/market risk area for nearly 2 decades.
i am a mathematician and specialized in derivatives - and i saw how the VAR-mania
started. Now what is VAR?..and why it was so successful?

How does VAR work?...in principle VAR is not an invention of
our clever bankers..but the idea is hundreds of years old,
and comes from gambling. Lets assume you toss a coin and you
have only 2 events.. heads or tails.

If you do that 1'000 times - theoretically it should be 500 hits
for both .. of course it will be quite likely not the same, but maybe
we will see 490 heads and 510 tails. If we see 300 heads and 700
tails, then there is sth definitely wrong with the coin.
But with a certain probability (confidence .. usually 95% or 99%) the
number of heads will be within a certain interval. This 95% confidence interval tells
you that only in 5% of the cases the number heads
should be outside of that interval -- it is a scenario "to be expected"

For derivatives the idea is the same, but the calculation is far more complex.
VAR is usually used for short term trading positions (liquidity time horizon
is approx 20 days)

The reason why VAR was welcome so quickly:

1) it gave the people the impression that the risk could be always expressed
in handy numbers (and in many cases - it cant !!!)
.. it was handled like a black box, like the oracle of delphi

2) it opened door for riskier trades - it was a kind of toy, which really works
under certain conditions

just that there were basically 2 types of people handling this:

a) those people which did not really understand what they are dealing with.
Mathematical and Physical models work under certain assumptions (called boundary
conditions - these can be market indicators, too). These people did
not understand how risk really works. (this is also how the toxic papers
emerged !!).

If a woman gives birth to 1 child between 20 and 30, and gives
birth to 3 children between 30 and 40, then this does not imply
that she will give birth to 729 children when she is between 80 and 90.
The example is ridiculous i know - but this is exactly how these guys
"calculated"

b) people (mostly some jerks sitting at the top) which ignored all warnings -
and simply misused the models or even completely ignored them. (these were exactly
those kind of people which turned around later and accused the models .. "our models were wrong")
When big profits were in sight - almost all these guys closed both eyes.
.... risk is an illusion in good times.

Adding all up - it was an ideal mixture of ignorance and arrogance(driven by greed)

or as an Irish saying goes: God is good, but never dance in a small boat.

Jan 31 06:05 PM

[Obama is Bu...:]

Mr Tan,

As a number of other comments have already noted, VaR has been criticized by Nassim Taleb (and many others) for years. Many finance models, including the Black-Scholes option model, use a set of assumptions designed to make the math into an equation that can be solved with existing math techniques-- rather than assumptions that reflect reality. Black Scholes relies on no transaction costs (commissions and bid/ask spreads do not exist?), no taxes (guess we are left with just death as inevitable?), and Gaussian distribution (which does not fit historical distributions, but has solvable equations).

I wish you had touched more on the human / behavioral side of why risk management has problems. Most Wall Street firms operate as though sales and trading are 110% of the firm. All back office functions -- including risk management -- are cost centers that need to be minimized and marginalized.

An extra dollar spent on risk management hurts the bottom line, and the CEO's paycheck. On the other hand, there is no way to quantify the benefit of spending that extra dollar --- and certainly no incentive for CEOs to focus on risk. By the time big banks got the bills for the lack of risk management-- the CEOs had long since collected hundreds of millions in pay. So what if the bank collapses? That does not effect the long term compensation of the CEO one bit

Jan 31 09:04 PM

[Dave Wrixon:]

The paradox is that the problem was in fact a very simple one. It did not need a complex computer to work it out. The eminent professor Roubini predicted it, and, even though I am a humble civil engineer, I made predictions to this effect 12 months in advance of the event (you will have to refer to domaining forums to see evidence of this, even got kicked out of few of them because my views were so unpopular with patriotic Americans.)
Predicting the occurrence of the problem is simplicity itself. It is just a question of what level of systemic debt is no longer sustainable. The complex psychological bit is required to determine when the rather over-rated blind herd that stampedes around Wall Street will actually catch on. No rational model is ever going to predict that!
The biggest problem in my view is that of the American psyche. If you actual put forward an opinion that suggests anything other than the USA being anything other than a Utopia for commerce, such views are generally not just ignored but are trigger a violent reaction. I think in the higher echelons of the GOP such views would have been deemed heretical. It would have been like confronting the Queen of Hearts from Alice's Adventures in Wonderland.

Feb 01 07:39 AM

[Dave Wrixon:]

Or to put it another way.

They got the wrong answers because they didn't ask the right questions.

Feb 01 08:07 AM

[777:]

I agree with Mr. Tan on risk manipulation.That is to say the market should have sold us more secrities, but pitifuuly the world financial market have sold us more toxic products and more risk. The IPO system is the major souce for the worldwide finacial crisis and economic receesion. But not China or the USA. The Wall Street should change its policy to sell out
more toxic ETF and non-profitable stocks in the market delisting the most
dangerous equties and ETF. That is risk manipulation, really.丨

Feb 01 08:56 AM

[Charles A...:]

In the 1960's as computer systems became more available, manufacturing inventory management was an early, very large and important application. Over time mathematical models of all sorts were developed to attempt to "optimize" production. When we learned that the Japanese car makers were running their factories by exchanging actual "Bin cards" to stimulate a production order; using what we eventually understood as "Just in Time" inventory management, we balked at it in disbelief. Theirs was a world of the real ours a world of the abstract. Real won.

Feb 01 09:37 AM

[Simon Smelt:]

Nice article, thank you. So, after the "China made me do it" excuse, we now have "the computer made me do it" excuse.

I'm old enough to recall when "high powered" computers (wouldn't cut it as a notebook these days) became available for macro-economic analysis. At last, really sophisticated models of the economy with lots of variables could be run. But, strangely these did no better (and sometimes worse) than much simpler models. This was for three reasons:
First, you're only as good as your assumptions - and the more complex the model, the more assumptions it embodies. Second, with increasing complexity, a black box effect sets in so that nobody quite grasps what is going on or the implications of some of the dynamics occuring within the model. Third, stochastic effects tend to increase the effect of any modelling "error" over time until a tipping point is reached and the model fairly abruptly fails.
Looks like the same story now, just a lot more costly.

Feb 02 03:50 AM

[Chris B:]

littleBonaparte,

Another problem is that even if the odds are in your favor, if you repetitively play the equivalent of double-or-nothing by reinvesting past proceeds, you will eventually hit zero. This is true even if you have a 99% chance of winning, because when that 1% outcome happens, it's game over.

Perhaps the models correctly predicted that a better-than-breakeven outcome was most likely. Perhaps that correct analysis persuaded the investment bankers to bet the farm - over and over again. The inevitable result of a series of such decisions is an eventual catastrophic loss. The decisions are individually logical, but in a series they are doomed.

That's why day traders and casino gamblers rarely succeed. Even if they win several times in a row, they will eventually hit a loss or losses - probably losing principal and prior earnings.

Additionally, it is human nature that we cannot walk away from something that has rewarded us. Show me a day trader who has made 7 very profitable trades in a row and I guarantee you he will make an 8th attempt. He cannot help it - he MUST. It's the same dynamic with bankers and big investors. We can only stop playing when we finally lose, and our early victories only set us up for repeated gambling culminating in an eventual loss.

Perhaps the only way to avoid this psychological trap is to make money the old-fashioned way - through interest or buy-and-hold.

Feb 02 09:24 AM

[Between The...:]

Good points Geoff. My take:
1) Do the models work if used correctly? Yes.
Almost by definition models work fine if used correctly. Models are like a toaster - they make toast without issue, maybe a bagel, but will electrocute you if you stick a fork in.
1A) Were the models used correctly? No.
Most VaR models are attempting to predict the max amount of money that would be lost in 99% of trading days. Even if the assumptions on which these models were predicated were correct (they weren't - most used historical data of only a few years of stable interest rates, low default, and a bull equity market, plus they don't model day-to-day correlation like an 18 month bear-market), the model still expects greater losses than the given value on 1% of trading days. It became too easy to dismiss that 1% chance as insignificant and plow on. However, it would be foolish to do so - there is a 20% in a month and a 92% chance in a year of an event outside the 99%. So even the naive model is being misused if it suggests confidence. With the "outlier event" a near certainty over the course of time, it would border on criminal negligence to ignore it and fail to plan for what clearly was capable of bringing down the entire system.
2) Do people have an incentive to misuse the models? Yes.
Design of incentive programs like salary and bonus are nearly all tied to short term goals of profitability. Risk is a method to keep unrealized liabilities off the books, and risk-less above market returns do not exist in the long run. The more risk that could be crammed into that 1% un-modeled tail, the better the return that could be earned without moving VaR. This, to me, is the systemic failure of distorting incentives that shareholders, board members, and management all bear the blame for creating.


On Jan 30 12:51 PM Geoff Considine wrote:

> Sir:
>
> Your perspective is a common one--and Nassim Taleb has been banging
> this drum for quite some time. There are two questions that need
> asking:
>
> 1) Do the models work if used correctly?
> 2) Do people have an incentive to misuse the models?
>
> Your answer to both appear to be no--but you provide no evidence
> that actually refutes (1). Certainly, we know that many firms have
> used risk management to obsure risks--to their own benefit. But
> there is a severe logical flaw in then assuming that you have addressed
> #1.
.....

Feb 02 02:44 PM

[Burton A. J...:]

Risk is best managed by three simple rules:

1. Have a reliable method of determining value.

2. Never pay too much for that value.

3. Never make an investment decision based on predicting the future.

Feb 10 07:39 PM

[User 356205:]

Thomas-

Interesting article. Models don't work in a catastrophe. Markets are a subset of human behavior and that is eratic in a catstrophe. In one instance we have $147 oil and $35 oil within a few months of each other. In another we elect a President who has never been in charge of a lemonade stand. In another instance a jet lands on the Hudson river and everybody patiently walks out and waits on the wing for a rescue. You can't model the time it takes to shovel snow in Miami. By the way, we are essentially going to print our way out this mess (and yes create inflation and sink the dollar, but since everybody else is doing it we should be fine). Just in case I suggest you take advantage of your 2nd amendment rights and store up on SPAM (protein and high energy that can be shelved for a long time, tasty in a greasy way as well [try it fried]). Any thoughts on real estate?

Feb 12 10:30 PM