Tuesday, June 16, 2009

Black Swan: Fallen from the pure faith

This story in today's Wall Street Journey has me scratching my head. (I know, it was reported a couple weeks ago as well, I'm a busy blogger). I read Black Swan. I thought it was absolutely brilliant. Nassim Nicholas Taleb has truly come up with one of the most useful and original ideas about markets that I've read in my entire career.

All that being said, what I'm reading about Universa betting on inflation is total incongruous with what I thought was Taleb's philosophy: basically that the big events in life, history, and markets are almost totally unpredictable. Given this, the best way to invest is to bet on change. Not to bother with long-term predictions based on fundamentals as we see them today, because by the time the long term comes around, fundamentals will be wildly different.

Now I'll admit. I'm not convinced Taleb's method of investing (doing nothing but buying out of the money options) is a good trading strategy. But I am convinced that his general way of thinking is a critical addition to any trader's mindset. Don't assume that the world as we know it today will exist tomorrow. Don't assume that just because you can't think of a reason why the world will radically change, doesn't mean it won't. Just because you make an accurate prediction about a certain event, doesn't mean you'll be able to accurately predict the myriad of consequential events. Just because something seems impossible given what we currently know, doesn't mean its actually impossible.

Taleb and his colleagues have never been more vindicated than now, when what seemed like a backwater portion of the mortgage market (sub-prime) touched off a series of events that almost crashed the whole capitalist system. Even among those that saw sub-prime as a problem, few imagined how this crisis would play out. And that's was Taleb's whole point. Its nearly impossible to see the Black Swan coming.

So now... Mark Spitznagel, who runs Universa, the hedge fund Taleb advises, is betting on inflation? I'm not going to rehash my own view on inflation, that's not the point. The point is that Taleb is known specifically for not betting on markets and economic events. Any market/economic outcome that you can see ahead of time, even as a lesser possibility, it isn't a Black Swan at all. Many people see inflation accelerating. Almost every client meeting I attend, I get a question asking about the consequences of hyperinflation. I don't know that even an extreme version of a commonly held view can be called a Black Swan.

Second of all, Taleb and Spitznagel aren't economists. They are mathematicians and traders. I don't know that they are especially equipped to make an inflationary call.

So it makes you wonder. Is Spitznagel doing this because its his view of markets or his view of what's marketable? I have a feeling its more of the later than the former. He's brought in all kinds of new assets based on Universa's huge returns in 2008. Now he wants to continue the trend. He knows that hyperinflation is a common fear, and his fund is known for thriving during a period of disaster. Its an easy fit.

It just doesn't fit his purported philosophy. I remember an interview Taleb gave on CNBC during the winter when he derided the money bank executives had made, specifically arguing that they managed their banks to enrich themselves, not their shareholders... Mmmm...

Our methods are not as different as you pretend. I am but a shadowy reflection of you.


Unknown said...

im pretty sure the wsj article got universa's strategy wrong. i think the new funds are betting on increased volatility in macroeconomic variables. they are not making directional bets on anything. the bet on volatility is consistent with taleb's philosophy.

cap vandal said...

My first impression on reading Taleb's books was that his "insight" would be hard to monetize. Corporations have limited liability and can't go below zero. As astroid would make it all irrelevant. So an event needs to be extreme enough for you to collect but not so extreme that nobody collects.

His basic assertion that tail risk is underpriced seems logical and true.

Here's a couple of links about his reporting of his financial success:



He didn't make any money in 2001, which included the WTC, etc.

He has a lot of insights but I find him too annoying and arrogant in his books, which badly need editing.

He doesn't have much evidence that his approach is profitable over longer periods and any assertion to that effect is basically unfalsifiable to throw in a bit of Popper.

I'm sure he has made some money trading, but letting a typo go that indicates that it was $20B is a problem.

Most things that are referred to as "black swans" are grey swans at best. That is, people understand that there is risk but they underprice it. A black swan should be reserved for those events that are a true surprise. I would put the WTC in that class, since the idea in the US of pricing for a "man made" property catastrophe was never given a thought.

cap vandal said...

Maybe these will work.....



Janet Tavakoli -- spent some time trying to verify his record.

John (Ad Orientem) said...

Conceding that the future is rarely predictable, I have recently become something of a fan of the approach to investing proposed by the late Harry Browne. I posted a short piece on the subject over at SA. The link for those interested is...


I think his concept of permanent portfolio has stood up reasonably well against the test of time.

Feanen said...

The events of the past year were predicted by quite a few people. It doesn't count as a black swan event when simple logic allows you to predict the events ahead of time. This event was only unpredictable if you limited yourself to looking at the historical data that was available in computerized formats (which is what most banks did). Shiller's work on housing prices basically proved that house prices have never gone up more than 1-2% a year above inflation except in bubbles. Once you assume that house prices had to return to where they were 10 years ago (inflation-adjusted) then everything else followed naturally. That same logic says that we are barely even halfway through this mess. And that's before you account for what's going on with commercial real estate. The Fed is trying to cause inflation now because the only way to save the banking system is to raise price levels (and wages) in general until incomes catch up to house prices and commercial real estate. Which still requires about 50% inflation over the next 2-3 years. The only other option is continued waves of mass-defaults.

I don't think it is possible to have 10-20% annual inflation for a few years and then bring inflation back under control. Which means the Fed has an impossible task. It would actually help the US a lot if foreign central banks would dump their dollars. That would fix most of the problems in the US economy, although it would destroy every country that dumped the dollars (because we would stop importing from them due to bad exchange rates).

Accrued Interest said...

I suppose you could say a true Black Swan (such as 9/11) are extremely rare. But I'd also say that the financial crisis was Black Swanish not because no one predicted elements of it, but because no one saw how it would play out. I think Taleb would say that there were too many moving parts to ever see exactly how each domino would fall. Betting on vol was the only way to win.

Feanan: 50% inflation? The home prices are down about 32% according to Case-Shiller. Where does 50% come from? The Fed will not try to inflate us back to where we once were in home prices. There is no need. It would be better to slog through 5-10 years of mediocre growth while consumers rebuild a decent balance sheet than to play chicken with hyperinflation.

Bill Moore said...

I too loved the book. I don't think Universa is specifically saying, "I think inflation is going up, so let's bet on it." My understanding is that they believe that way out-of-the-money options and derivatives related to inflation (or in this case hyper-inflation) are extremely cheap given their probability.

Feanen said...

The amount they've already fallen doesn't matter. Real house prices in most cities are still about 50% higher then they were before the bubble. To get back to trend, wages have to go up another 50%, house prices have to fall another 33%, or some combination of the two. The more house prices fall, the more loans will default. With the amount that house prices have already fallen, the mortgage defaults may already be more than the banking system can withstand. The only stable solution is to keep house prices where they are (or even push them up a little), which means raising wages 50%. If you look at the loan reset/recast schedules, this needs to happen in the next 2-3 years to prevent a major banking crisis.

Olive Zeng said...

Nothing new from Mr. Black Swam to me though, Soros made that even clearer.

However, guess all should be attributed to Mr. Karl Popper, all i meant.

Mr. Taleb? Don't see nothing new, at all from him to be honest.

In Debt We Trust said...

The sub prime meltdown was predictable by many (including me). I even wrote a working paper about it last summer.


What was not predictable was Paulson's decision to nuke Lehman and destroy the credit markets.
Only the people in that NY Fed Bank room knew WHY they choose to do that.

*I took it down for further revisions but left the title page up.

In Debt We Trust said...

Regarding inflation, this article should tell us that we are very near a short term top. When you see things like this start to appear, the goldbugs are due for a reversal.


GS751 said...

Take into account that Taleb came up with his theories while trading currency options he could base his fund on large swings in the currency markets.

Donald Pretari said...

"The latest wager is more of a directional bet( NB DON ) that regulators' efforts to prop up the financial sector and the broader economy will spark inflation."

This story was hype. The recent story on Russia and US Treasuries was hype. Is this position even controversial? Hell, I believe, and hope, that QE and the stimulus will lead to some inflation. And, for QE to work, people need to believe that there will be inflation, so he's doing my view a favor in proclaiming this.

But somebody, anybody, tell me what's so newsworthy in this point.

Don the libertarian Democrat

Salmo Trutta said...

"Taleb's philosophy: basically that the big events in life, history, and markets are almost totally unpredictable"

Completely false:

There is no such thing as “long and variable monetary lags”. Monetary Lags for real-GDP & inflation are fixed.

Historically, they never vary in length. Mathematically then, it’s impossible to miss forecasts using (1) rates-of-change in bank reserves in conjunction with (2) rates-of-change in bank debits (impossible up to a period of about 1 year).

Obviously, all price increases (resulting from a long-term excessive flow of money relative to the volume of real output of goods and services offered in the markets), don’t occur within a single month, but the majority of reference, or distribution points, are clustered, or skewed, within a single month. That month then becomes the immovable lag.

Milton Friedman’s & Anna Schwartz’s research [Monetary History of the United States (1963)] never went further than specifying that a range of months (“initially 6-9 months, with inflation another 6-9 months”), was associated with a given growth rate of the money supply.

I.e. with bank debits, a very high proportion of the number of months used (in the lags & data for GDP), have already passed by before estimating the next quarterly GDP data (and its 2 subsequent revisions). Why wait 2 more months to plug in “final GDP” into the Taylor or policy rule?.

GDP data is ex-post. GDP data is measured quarterly. MVt is measured was measured more frequently-monthly (or up until 1941 weekly). Monetary flows are ex-ante.

Unfortunatley, the data for both reserves & bank debits are no longer available. When Ed Fry retired who managed the series at the FED, the series was discontinued (because it was a regulatory burden).

As the FED has never been cooperative in supplying data, the whole series was overwritten at the close of the series.

The data revision had errors. Even so you can still use the St. Louis historical data to check the answer for yourself.

Horace Kent said...

I thought Black Swan could've been a chapter of fooled by randomness personally......

Who considers massive inflation a black swan? The fed and policy makers.

Salmo Trutta said...

There is more "noise" to data than people realize. The late William Bretz of Juncture Recognition repeatedly found reporting errors in the FED's data series.

If it weren't for his attention to detial, some of the reporting errors in the debit series would never have been uncovered.

The only other party involved was the Bank Credit Analyst & they weren't very agreesive.

Salmo Trutta said...
This comment has been removed by the author.
Salmo Trutta said...

I picked up a book on investing at the book store. In the forward the author stated that no one could pick tops & bottoms in the market.

Well, he obviously couldn't. But I bet most economists aren't traders. And if you knew how to pick tops & bottoms you would have an incentive not to tell anyone.

Joseph Granville of the Granville letter got so popular that his followers anticipated his call in Sept 01. The followers bought & moved the market so that the call never was made.

The point is that if anyone knows how to make predictions, you aren't going to hear about it.

In Debt We Trust said...

Can you do a special post about the California situation?