The real cause of the financial crisis -- An MIT Blackjack Team perspective | Hacker News Reader

The real cause of the financial crisis -- An MIT Blackjack Team perspective

semyon.com

185 pointsprakash60 comments

"...The only solution is to forbid money management as we know it..."

That's a bold and provocative statement. I'd feel a lot better if the author actually proposed some way of ending "money management as we know it" instead of just throwing the blurb out willy-nilly.

Nobody disagrees about the premise: markets will always seek to find loopholes in the system and in the human condition. Some people see this as a source of growth -- after dozens or scores of such blowups, beginning way before the Dutch Tulip Bubble in 1637, markets have continue to grow and become more and more sophisticated. Whatever the fundamental flaws, it's obvious they haven't prevented outstanding growth over the last several hundred years or so.

To propose a dramatic new solution, it's your job to make the case that the benefits outweigh the drawbacks. Otherwise it's like speaking out against eating because of all of the cases of food poisoning -- you're missing the point.

Though he doesn't suggest it in so many words, I understood him to be proposing that money managers must be paid a flat rate rather than by commission based on returns or total money managed, that it would remove their incentive to take short-term risks to run up their personal reward.

He is suggesting that top executive and trader pay be based on the long term performance of the fund. In particular, their own money should be invested in everything they do, and likely also for a number of years after they leave.

One interesting thing to me is that even if you do this (correlate compensation with longer term results) a lot of the same problems still happen. Take, for example, the first big hedge fund blowup: Long Term Capital. Shortly before they blew up, they kicked most of the third party money out of the fund and even levered up their own individual positions (see the book "When Genius Failed").

I'm not saying that misalignment of incentives doesn't play a significant role in a lot of what we are seeing now, but never underestimate hubris.

If they hadn't paid out that dividend, LTCM would have had enough in reserves to survive, but no one seems to have learned from that, least of all Alan Greenspan.

We can't solve hubris, but perhaps we can highlight it to investors.

Funds that too hard to value or are highly leveraged should switch from regulation by the SEC to regulation by state gambling commissions. It would be tricky to find the right cutoff, though, but if an investor would have to travel to a Native American reservation to buy credit default swaps, they might start thinking about whether they really understand what they're investing in.

How about the following regulation:

No restrictions on what funds can do, no government bailouts, but full disclosure so that clients may impose their own restrictions.

A lot of strange things happen when the players have diffrent amounts of information.

EX: Would you bet 1 billion dollars with slightly less than even odds?

Well if you don't have a billion dollars then your real risk could be much lower vs. your potential gain. The real question is who would trust you to actually have a billion dollars? That's the real sucker, unless he also lacks a billion dollars.

That doesn't work.

If I'm paid a flat fee to manage money, the people handing me money to manage are still going to look for the manager with the highest returns (that's human nature).

So, I still have an incentive to chase riskier short-term gains; I see less of the upside than if I were also getting a cut of the gains, but the incentives that arise from competition with other money managers remains in place.

I think lots of people are ready to forego higher returns for higher security. That's what I ask from my banker.

For a short while I was looking for a bank that would just have a portfolio that follows the market instead of trying to beat it, and that would take less fees because it took less effort to manage. I contacted two other banks, but they didn't have anything like that. I found it was extremely tiring to speak to bankers, they all seem to want to drown you with words. I eventually stayed with my original bank just to stop wasting time.

I thought such things were common?

http://en.wikipedia.org/wiki/Index_fund

They are common.

The problems they don't solve is how to allocate the assets between currencies. Also part of my money is in fixed income papers. So even when not trying to beat the market, there are some decisions I would like to leave to a professional.

Dimensional Fund Advisors (http://en.wikipedia.org/wiki/Dimensional_Fund_Advisors) diversify among currencies, markets, and asset classes according to some math I don't really understand. They seem pretty sharp, though, and they're low-load for the buy-and-hold version (they have market timing funds designed to lower the fluctuation at the cost of lowering the long-term return).

Yes, maybe after a market crisis more people look for security, but I think that over time they forget the last crises and move back to chasing returns until it gets bad enough for the next crises to occur.

I really don't think Seymon has a real solution. Money managers who think they have a foolproof scheme will willingly bet their own money. Sure limiting compensation is a good thing, but not a cure. The best suggestion that I have seen is a requirement for the issuers of derivatives to maintain a reserve. Regular insurance is a kind of derivative and insurance companies are required to maintain a reserve to pay off policies. The same should be true of other kinds of derivatives. A reserve requirement slows down the explosion of derivatives, because eventually the total reserves are greater than the value of the underlying instrument. We might still have failures of individual firms, but total collapse of the financial system would be less likely.

There is still a need to put realism back into lending, because we are getting these crises every few years: loans to South American dictatorships, the savings and loan crisis, the Asian financial meltdown and the current mortgage and derivatives failure, all in the past 40 years.

Well, according to WaPo, 40% of all the "profits" in 2007 came from Wall Street. Did those profits actually exist? Were they short-term book-cooking so that they got their bonuses?

By 2007, when Wall Street's profits amounted to an astonishing 40 percent of all American profits, the business of American finance was no longer American business -- providing loans for domestic production, technological innovation, that sort of thing -- but swapping bets and hedges on bets and hedges, all for hefty commissions.

http://www.washingtonpost.com/wp-dyn/content/article/2009/01...

Apparently $70 Billion of the bailout package has gone out as bonuses to the same idiots on Wall Street who got us into this mess.

I won't say this is the best thing I've ever read on the subject, since that would imply I have sufficient knowledge of the financial markets to accurately evaluate what I'm reading.

Instead, let me say this seems, or feels like the best thing I've read on the topic.

In addition to relating the crisis to the Martingale betting system, the author describes how an ecosystem of bettors tends to converge on a Martingale-like system when playing with other people's money, and taking a percentage of the stake as a fee.

And limited liability on the down side!

I don't have a lot of faith that outlawing performance-based money management yields a stable system either. (I do agree that it is likely to yeild one less to synchronized catastrophic implosions, until it naturally devolves back into what I've seen twice in my 4 decades on the earth.)

Suppose we "agree" that 6% is the safe rate of return, and no money manager is allowed to collect fees or bonuses based on performance. The next logical step for anyone with a relatively large amount of money is to try to build their own in-house (or in-family) money management arm to generate 7% or 8%. If that attempt is successful, by whatever means including blind luck, that money manager can go take his expertise elsewhere with a track record of "proven high returns", or get a not exactly fund performance based raise or bonus. (Call it a corporate profit-sharing bonus, or retention bonus, etc.)

Logically, once one group is doing it, others will try to follow. No wealthy family is going to sit by generating 6% when someone in their same situation is generating 8%. Those that don't have large enough investment nuts to start with will pool with other not-quite wealthy investors, and eventually this will trickle down to the common mutual fund. And then we'll see "funds of funds" (or funds of fund, as we saw recently) that will take money, invest in the latest 8% fund, pay out 7% (still beats the 6%) and pocket 1% for expenses.

Trying to legislate away greed is like pissing in the ocean.

IMO, you need people to feel like risky investments are risky investments. If they instead feel like they own an implied governmental put on their risky investments, you just get more and more comfort with risk. People who make bets that they can't cover need to feel the pain, and for those fallouts to become public in the form of bankruptcies.

Banishing an entire category of financial work (fee-based money management) because of some crooks, some bad bets, and some large failures is not the path to a productive and efficient means of distributing capital to the businesses who need it to generate growth, jobs, and new products, IMO.

I worry that a logical outcome of this proposal would be an end to pooled-risk VC funds, and a substantial impairment to the liquidity and transparency of the public markets. It makes for a good blog entry though, as long as you don't think about it too hard.

This essay makes some good points, but ultimately it fails to address the titular 'real cause', due in part to an over-reliance on the personal experience of the author. In this last respect it is typical of the genre; Michael Lewis, for example, tends to focus on miscalculation of risk, and particularly on the defects of the Black-Scholes model for pricing options---based in large part on his experience as a Big Swinging Dick at Salomon Brothers in the '80s.

Based on my reading of a bunch of these economic diagnoses, I'd like to suggest a two-part litmus test for addressing the 'real cause' of the crisis:

  (a) Explain the system of central banks and fractional reserve banking.

  (b) Connect (a) to the present crisis.

I've become convinced that any explanation that falls short of this test invariably misses the forest for the trees.

People will always move their money into the places that give the best return over a few years, no matter how many times they are warned with the disclaimer that "past performance is no indication of future returns." And eventually the crisis that results will reach global dimensions beyond the means of a government bailout, especially if part of the risk managing strategy becomes counting on bailouts happening every decade or so.

The only solution is to forbid money management as we know it.

There are a lot of assumptions, first and foremost that all investors are trying to maximize their returns with no regards to risk. I know many people who choose investments that are less risky and thus have less of an interest rate.

People always look at money management from the investment side (ie, what is my return) and don't think about what is actually going on... ie funding new ideas/projects, if you outlaw money management then drastically alter the way that people can get funding -- greatly increasing the misallocation of capital.

A better solution would be to tell people to research what they're investing in, if it doesn't look right or you can't understand it don't invest in it.

> A better solution would be to tell people to research what they're investing in, if it doesn't look right or you can't understand it don't invest in it.

Unfortunately for most people, that's almost everything.

If we didn't have an monetary policy of inflation, merely depositing money in a savings account would be a viable option.

I don't see that to be true. There are plenty of low-risk, low reward options out there if all you want to do is preserve your value.

Inflation has its own issues, but as I see it they are mostly social class related.

> There are plenty of low-risk, low reward options out there if all you want to do is preserve your value.

No there aren't. To preserve real purchasing power monetary inflation FORCES you to speculate. Straightforward saving has been a loser's game for many decades now. This was not always true and there were long periods of price deflation and good interest rates before the establishment of the fed.

Straightforward saving has been a loser's game for many decades now.

18% interest was a loser's game? http://www.hsh.com/indices/6mocd80s.html Even as recently as 1989, CD rates were over 10%. In 2000, they were over 7%, and in 2007, they were over 5%. http://www.hsh.com/indices/6mocd00s.html

Yes, and go look at annualized monthly CPI increases during those periods.

Annualized monthly CPI increases were over 18% for 6-month stretches? How about 5 years? There are 5-year CD's. From 1980 to 1985, the CPI went up only 30.6%. http://cost.jsc.nasa.gov/inflateCPI.html From 1989 to 1994, the CPI went up only 19.5%. From 2000 to 2004 (as late as that calculator goes), the CPI went up only 9.7%.

Right now, America, and the world, are experiencing deflation. Yet, 6-month CD's are paying over 2% interest.

A few quarters here and there with decent real CD rates is not reflective of how people actually go about saving. You have consistently lost purchasing power if you plowed cash into savings accounts, CDs, and money markets funds.

When all factors are considered, including understated CPI figures, I believe real returns to cash in the period in the 80s you point to were not even very good. Especially when you consider the opportunity cost of your five year CD and what played out in other asset classes during that five years.

Especially when you consider the opportunity cost of your five year CD and what played out in other asset classes during that five years.

What other zero-risk asset classes were we comparing against?

Hedge fund managers get paid more on their performance than on size. Their % of assets under management is generally about enough to keep the lights on. And many hedge fund managers are heavily invested in their own funds.

What he describes with " And money managers are rewarded based on the size of their fund, or the level of returns. The managers do not risk their own money. " sounds more like a mutual fund, though selling out of the money puts does not.

Hedge funds usually carry losses over from one year to the next as well. They're definitely designed to align the interest of the manager and the limited partners much better than mutuals.

In fact, in the long run hedge funds outperform mutual funds by a substantial margin for exactly that reason. He's right that neither have pure investor/manager alignment. The only way to do that is to lengthen the term from one year to 10 or something.

Very interesting argument.

I understand the article to imply that we need more government regulation because the market cannot correct itself for pyramid schemes without severe devastation to the economy (the Albania example).

For me, this argues conclusively that Fannie Mae and Freddie Mac may have been the triggers for the current crisis but not the cause (in themselves, they were not sufficient to cause the crisis). From the article's analysis, the real cause of the current crisis was the lack of regulation of mortgage swaps. The trillion-dollar mortgage swaps is what amounted to the Martingale system.

Are you familiar with the Albania example? I wish the author elaborated more on it.

Here's a Wikipedia article on the 1997 Albanian Crisis that was initiated by a ponzi scheme: http://en.wikipedia.org/wiki/1997_unrest_in_Albania

Thanks! If we are to believe globalsecurity.org, which is quoted verbatim as a "primary source" in the article, it seems like the UN played an important role in restoring Albania's order.

Semyon gives Albania as a cautionary tale, but a cursory look at WP reveals pitifully little. Anyone with more knowledge on the matter, aside of it being simply a collapse from an overgrowth of unregulated Ponzi schemes? Is it that simple? Finally, is this a textbook example in economic policy?

How about making perfomance based payouts only on longish returns, say 10 years? That is, money manager has to be consistently producing positive returns over 10 years, before starting to collect comissions. Now, this can be gamed too, but to much lesser extent than the current system.

He doesn't take into account the misguided financial policy of the Federal Reserve as well as deregulation measures.

Perhaps the SEC needs to act more like the FDA. Before new financial instruments are allowed to be sold to the public, they need to be vetted and understood.

Drugs are complex and drug makers would love to release their drugs early in order to make a profit. However, as a society, we realize that would be detrimental.

Regulation of financial instruments perhaps would stifle financial "innovation", however, it would hopefully help prevent the spread of toxic instruments as well.

This is the best explanation of how to get short term gain at the cost of long term ruin I have seen.

However, let's not get caught up in the details. These bankers knew what was going on. Wall Street told shareholders around the world that they had their best interests at heart, when in fact they were selling them up the river. There is a word for this kind of deception: fraud.

And yes, it is still illegal, even if "everyone is doing it".

The SEC should have stopped this a long time ago, and there is a name for that too: corruption.

> The only solution is to forbid money management as we know it

Why it is impossible that people voluntarily do not gamble their money with slick money managers? That is what I'm doing...

You can avoid funds. You can even get out the stock market altogether. But you can't get out of the global economy... (unless you have a spaceship, or go live in the woods)

Great case but...

> The only solution is to forbid money management > as we know it.

Terrible conclusion. Human nature is stupid, but powerful government is worse.

Let the markets rise and fall. Let people get burnt and learn from their mistakes. Let the market emerge from the current situation with investment firms that thrive because they give their customers confidence that they're not doing dumb things.

Another direction one could take concerning this problem is change the reward model of money-managers, similar to vesting in startups.

If fund-managers would not be paid their whole money annually, but subsequently over a longer period of time, they might be more interested in the long-term success of the assets they are managing. For example, if a fund-manager would be entitled to 10M$ for one year, he could get paid 2M$ + interests annually over five years. This way, accountability could be strengthened.

I think one major cause of the problem is that every financial show/book/article for the last 20 years has been telling people that funds are a safe, smart thing to invest in. And they have been, but they have been boring and have modest returns. It's not surprising that people are trying to get a better return. The solution is what it has always been "Let the buyer beware". If Madoff's customers hadn't skipped their due diligence, they would have been fine.

I have a great idea. Let's just bail them out when they loose the 100k for $100 at blackjack. See, that way, they never learn!

I've read a few books on junk bonds, and if they were a Martingale / pyramid, how come Nelson Peltz, Carl Icahn, MCI, et al, became (and still are) billionaires? They made out like bandits.

Only the winners are remembered and reported. Likewise is the world of "famous" fortune tellers.

But as far as I can tell, there were more junk bond successes than failures. But I admit, when Michael Milken was arrested, it all came tumbling down for Drexel, and I can't remember any book saying exactly who failed with Milken's attentions.

So yeah, I think you're right, and even books can dodge the truth.

"Den of Thieves" by James B. Stewart covers the Milken, DBL and Boesky years. It's a decent read--I just finished it about a month ago.

A pyramid scheme always has a few huge winners. They're the ones who get into the pyramid when it's small. Depending on how broad the pyramid is, oftentimes everyone except the bottom rung (the retail investors) ends up as a winner.

What people forget is that capitalism is based on fear and greed. If you look at a particular way, everything is a "pyramid scheme". For example: "Why didn't people see that a company based on selling copies of music is not sustainable?" -- this may be the excuse used by everyone when most music music recording companies fail.

Any capitalist activity can be seen as a suckers game, because price is subjective. Financial markets are just a faster, larger scale version of other markets.

So, a generalized successful capitalist formula:

  1 - Use greed to motivate Other People
  2 - So you can use Other People's Money
  3 - Get in at the beginning, get out early with the money
  4 - Leave behind suckers with worthless junk

Use your profits to begin the cycle again.

I prefer Warren Buffet's approach:

  1 - Look for good people producing genuine value
  2 - Make a bet on them

The difference between the two is really in how cynical the observer is, not in anything that they're doing. For example, one of Buffett's first (and most profitable) investments was in See's Candies. They make chocolate, at fairly high prices. Where's the value in that?

Then he used the money to buy companies like Coca-Cola (they sell carbonated sugar water, and in developing nations often steal much needed water from indigenous people to do it) and furniture stores (which purchase furniture from cheap laborers and sell it after a 4x markup).

This really makes perfect sense when you consider that "greed" is a synonym for "what people want", with the only difference being that "greed" implies that what they want is frivolous, unnecessary, or destructive. So Paul Graham's maxim of "Make something people want" could, if you're cynical, translate to "Exploit people's greed." The end result is the same - you have something they desire - but the connotations are very different.

See's pays careful attention to freshness of ingredients. This is not only my experience, but also the opinion of a panel of chocolatiers and food critics hired by Consumer Reports to evaluate various brands of chocolates. For me, they are a much better value than Godiva. I prefer to give See's boxes as gifts. Giving my friends and family pleasure is valuable to me.

Furniture stores - There is a value add to being able to go somewhere and try out furniture. If you think 4X markup is way too much, then you have a business opportunity on your hands. Get information to the consumer and connect them to furniture makers. The consumer pays less, the manufacturer gets more, and you get rich off of your cut.

Coca-Cola - I don't buy it. Others do. To them, it has some sort of value.

It's a stretch to equate "Make something people want" and "Exploit people's greed." That Venn diagram is two overlapping circles, not just one.

Junk bonds are not a pyramid scheme. Junk bonds are really high yield bonds, and they were perfectly useful and successful in unwinding the strange corporate conglomerations made during the 60's and 70's.

Even if they were a pyramid scheme, the Peltzes, Icahns, etc didn't hold the bond themselves. They would want to take over a company and sell the parts. So they would front some capital, and get other people to buy the bonds which was used to fund the takeover. So they were running the pyramid, not actually in it (if you think junk bonds were in fact a ponzi scheme).

When the junk bond/leveraged buyout strategy started, it was a reaction to the fact that corporate managers who had lived through the great depression were too risk averse when it came to balance sheet debt, along with the fact that there is a tax asymmetry between providing returns to suppliers of capital in debt form as opposed to equity form. In the early days, these factors, along with low stock prices, permitted the financiers to buy companies and either restructure or dismember them to produce a profit at fairly low risk. Early profits on these deals attracted competition, eventually resulting in extremely high prices and debt levels. By the time of RJR/Nabisco, the debt levels were too high to be safe in an economic downturn. I expect that there was a similar life cycle for collateralized debt instruments. The people that made out well were the early adopters who understood what they were doing, and the people who got burned were the ones who simply had to be in the business because they saw that other people were making so much money.

The financial crisis has little to do with probabilities or risk analysis. It's not like blackjack. It was more like seeing that a ball was falling and would eventually hit the ground. Everyone has latched onto discussions of risk analysis and reward structures because it conveniently excuses widespread idiocy and greed.

"Why did I just lose a fortune? Oh it's not my fault, it was a Black Swan! And my fund manager's incentives were not properly regulated." No acknowledgment is ever given to the notion that the financial markets have worked exactly as they should: fools have been separated from their money.

When you have to define "fools" as "almost everyone in the world", I'm not sure you have a workable definition.

When you define "almost everyone in the world" as the <5% that own stock, then I'm not s.......

Typical mathematician knee-jerk reaction (including, often for me). Just because money management is not perfect does not mean that it does not have a high positive return. In fact, the largest money managers, the banks, are absolutely critical to a productive economy.

Reply on news.ycombinator.com