«Be paranoid on one side, and risk seeking on the other»

Nassim Nicholas Taleb, risk specialist and author of bestsellers like «the Black Swan», explains how not to be fragile. He fears for Switzerland's success.

Mr. Taleb, since the financial crisis, the US economy has recovered and the financial markets have staged an impressive comeback. How fragile is the world economy and the financial system today?
We are much more fragile than we were ever before. The reason is the high level of government debt and the record low interest rates. This time, governments and central banks have less room for maneuver when things start to fall apart, because rates are already low and debt levels are too high.

How worrying is the beginning of tighter US monetary policy?
Rising US rates are more dangerous for other markets than for the US. I remember in 1994 when former Fed Chairman Alan Greenspan tightened just a bit and Italy still had the Lira.

We all thought that yields in the US would go up sharply. But it was the higher yielding Italian local bond market that suffered the most from higher US rates as capital flowed back to the US.

What is your investment advice today based on that experience?
The equivalent to Italy at that time would be emerging markets today. I also like the idea of shorting junk bonds because the spreads are too narrow and will rise when the Fed starts to raise rates.

Things have gone wild and people are going to go bankrupt, because the generation in charge today has not seen risks. They have written too many books and attended too many classes about it but they still don’t know what risk means.

But haven’t the banks learned their lesson and changed the way they think about risk? After all, your book «the Black Swan» was also heavily debated in the financial industry.
No, they have not. Let me talk about the idea of the black swan. It is an illusion to think that you can predict rare events. Complex environments are very unpredictable.

To predict where the stock market is in one year or what kind of tail event is going to happen is completely nonsense. I have been repeating this since 1990. But people keep telling: We know, but we have better methods now. And then they all blow up in the same way as before.

Could you give us examples of this repeating phenomena?
The victims of the financial crisis blew up the same way the Metallgesellschaft blew up in 1994 or LTCM in 1998. They all said, well these events happen very seldomly. Or take the mortgage company Fannie Mae. I wrote in the book «Black Swan» that Fannie Mae will go bankrupt, because I saw the accelerating losses. They just said: We have 50 PhDs.

What is so wrong about employing smart people?
With 50 PhDs you can go to the moon but you cannot predict the stock market. There are some mathematical impossibilities, something called the curse of dimensionality.

It is much easier to be robust than trying to predict rare events. It is much easier also because we can measure robustness and fragility. Trying to predict black swans is a very nice exercise but it is not going the help to avoid ruin.

That means that the banks are not safer than 2007?
No, they are more fragile than ever. Just look at the London whale scandal. J.P. Morgan lost $ 6 billion on one single trade.  They said this should only have happened every 10 billion years and that they would change their Value at Risk. But I tell you, it will happen to someone next year.

The only thing you can predict is that those who think that they can predict rare events will blow off, because their portfolio is very fragile and they rely on flawed quantitative and theoretical principles. I called winners of the Nobel prize in economics charlatans. Things like the Garch-Modell or Value at Risk do not work. You cannot make people rely on them. These models do not show you the error rate.

But why are people holding on to these models of modern finance?
Because they rather want to make steady returns and risk big losses in an accident than to make steady small losses and gain when things collapse. When they lose they just write an apology letter but don’t send a cheque. They keep on doing it. People are hiding risks using the same methods which we have known since 1987.

This is why the next banking crisis is inevitable. What should have been done in terms of bank regulations?
I am not too much worried about the banks and the stability of the banking system. Banks are part of the past, of an old story, when they were the only source of credit.

And the overall financial industry? How can it become less fragile?
We have to use a different paradigm in finance.

What would that be?
Less PhD and top-down engineering, more trial and error and grey hair. Banking was once an apprenticeship thing, a business for people with grey hair, skeptical of mathematical stories and focused on robustness. This is how it should be.

I argue that the only people who understand risk are the old-fashioned bankers and traders with grey hair. But what happened to them? They got fired or retired. You only might find them in places like Geneva or Zurich.

Are you more positive for the Swiss financial industry?
Yes, I have more hope for you guys. I worked for Swiss banks in the US. When I began, the bank was extremely risk-averse and mathematics did not persuade them.

But then the Swiss went from not wanting to trust these models to be enamored of them. From being risk averse they started giving money to anyone, and they even put money into the subprime market. Now they are going back to where they were before.

Why do you always focus on improbable events? Because they can be very costly?
Ruin is not a renewable resource. Because the cost of ruin is effectively infinite, cost-benefit-analysis  is useless. It doesn’t matter if you try to pick up pennies or 1000 franc bills in front of a steam roller.

How exactly do you define and detect fragility and its opposite, what you call antifragility?
Fragile is everything that cannot take a big shock, what is disproportionally more harmed from a black swan than from regular events. Take a coffee cup, it does not like random events.

A common element of the fragile is accelerating losses. If it goes down 1% you lose a million, if it goes down 2% you lose an extra 10 million. For the fragile, shocks bring higher harm as their intensity increases. One large stone harms more than 1000 pebbles.

And what is antifragile?
If there is the fragile there must also be the opposite. Most people would say robustness is the opposite but it is only neutral. If fragile is what is harmed from disorder and volatility, the opposite is everything that benefits from disorder and volatility.

I like the idea of a parcel on which is written, «please mishandle» instead of «handle with care». Because there exists no word for this property I call it antifragile.

How can one remove the fragile?
What does not like volatility does not like time. The life expectancy of a young child is higher than the one of an old man. With ideas, technologies, concepts, or even companies it’s the opposite. What has been there for years will stay for years. What is three years old or less, will not be here for more than three years.

With this very simple rule you can remove the fragile. It is also a simple approach to analyze companies: A company that had a lot of trouble and came back is more probable to survive much longer. If something has handled an earthquake, that tells me something.

You say size matter: the bigger the more fragile. But what about economies of scale?
If it existed the leading companies from the 1980ies should today control the world, but they are not even in the S&P 500 (S&P 500 3'636.50 +0.21%). When companies merge, the resulting company is not necessarily better. The problem with size is, that when you have a problem it costs more and more the larger you are. The costs are non-linear.

When Société Générale (GLE 17.36 +0.56%) had to liquidate € 50 billion, it cost them € 5 billion. The liquidation of € 5 billion would have cost them almost nothing and not a tenth of the 5 billion. If you are large you suffer more from rare events.

How can investors implement the concept of antifragility?
One way is to look for strategies with optionality. An option has the following particularity: It is a convex bet. You make more when you are right than you lose when you are wrong. Avoiding to be fragile means being long volatility. The convex strategies of buying tail events paid off in 1987, 1998, and of course in 2008. Nobody ever made money shorting volatility.

But this convex strategies requires a lot of discipline, especially when everyone around you is making gains.
You need someone to monitor you. Me too, I hated the strategy. It was horrible to be betting on rare events when everybody made steady money. I instructed a colleague to keep buying options for me or I associated myself with people that are rigorous. I always have to fight against my emotions. But visibly the strategy works.

What is the best portfolio strategy to come closer to the antifragile?
I call it the barbell strategy, it is a dual strategy: Be paranoid on one hand and risk loving on the other. And I think it is very Swiss: That means you put 80% of your money into an ultra-safe asset such as Cash or Treasury bonds. With the other 20% you can take risky but uncorrelated bets.

The maximum loss of such a barbell strategy is 20% but you have a huge upside. Having 80% in liquid safe assets has another advantage: You have powder dry to buy when everybody is forced to sell. The last great opportunity was in 2009. That is why cash has an optionality, too.  The barbell strategy exists also in nature. I found many biological examples. Or look at Google: 70% of its business is bread and butter, the other 30% of the resources are going to find to new ideas.

What specific investments suit the requirements of the risk bucket?
Investment that offer good risks. I prefer to express bets via options. I have some convex bets in real estate. Investment in technology is also fundamentally long volatility.

A typical investment with optionality are companies where mistakes or a side effect make you rich, like the invention of Viagra. A lot of the pharmaceutical business is based on trial and error. You try and try and make small mistakes, but when it works the gains are huge.

What about being long oil?
Oil has no optionality. Oil is losing importance in the energy market. A better risky investment with optionality in the energy sector would be companies that specialize on the storage of solar power.

In your last book «Antifragile» you called Switzerland the best example of a country that is antifragile? What exactly did you mean with that?
Switzerland is the most stable country in the world. And it is not only stable, it is also antifragile, because the country benefits from disorder and volatility. It is doing especially well during a crisis, when everybody is looking for a safe haven.

What’s behind the success?
Just ask anybody who the president of your country is. Most people don’t know the name. This has nothing to do with the education level, it shows that the central government and its president here have very limited power. Switzerland is the opposite of a top-down system.

It is a bottom-up system, it works by trial and error, the Brooklyn way, no central planning.  It is not a nation-state, but rather a collection of small municipalities left to their own devices. As such it may be always harmed by small shocks, but that makes the system more robust. The mistakes are too small to ruin the whole country.

In your lecture you also mentioned Germany as a successful, decentralized country. So the German success is not the result of some special cultural values?
The decentralization was the result of the attempts of the Allies after World War II to weaken Germany politically.

They saw that Bismarck and Hitler had a lot of bad ideas in Berlin, so they shut down Berlin and forced Germany to decentralize. Exactly this decentralization which was meant to weaken Germany politically, made Germany economically stronger. Forget about special cultural values.

It seems that Switzerland’s antifragility has precarious side effects such as an overvalued currency and rising property prices. Is Switzerland so antifragile that it becomes fragile again?
Switzerland has been a very successful place indeed but the best part is over. I am not too worried about the loss of competitiveness due to the stronger franc. Switzerland has always found ways to sell its products. Look at this watch, people will not stop buying Swiss watches, even if prices rise further.

The bigger risk to the successful model of Switzerland is that it is overrun by wealthy foreigners. The US is an immigration country too, but we get the poor and the smart who drive the economy. In Switzerland it is more like in London, primarily the rich and the saturated come here. Is that what you want? Is this the immigration that was the engine of Switzerland’s success?

Thanks to rich foreigners we pay lower taxes and can afford high-level infrastructure.
That income is not sustainable. The wealth of the foreigners causes envy and rising rents and unaffordable housing increases the risk of social tensions. Only a few sellers profit from surging house prices. The money of the rich foreigners can cause corruption. London has become more corrupt. In Monte Carlo it is even worse.

Would you call for stricter immigration rules?
There are many ways to deal with this, but a more restrictive immigration policy is dangerous. You still want the intelligentsia, people like Lenin and his opponent Nabokov who come here, because they are free to do what they want. The best way to deal with the problem is probably a ban on foreigners to buy property here.

In your book you draw a very idealistic picture of the Roman empire. Should it be the standard model of the European Union?
The Roman empire was the archetype of a decentralized political system. An empire is not a nation-state. Rome only collected taxes and guaranteed security: Pax romana. The Model of a bottom up empire is today still alive in the US.

Also (ALSN 237.50 +2.37%) Europe has now the principle of subsidiarity, but they don’t apply it. If they do it will bring us back to the Roman empire. Subsidiarity is the most important thing – it means  that nothing should be done on a higher level if it can be done on the lower level. But today Brussels regulates even  the speed of your windshield wiper.

But doesn’t an integrated market need common standards?
Let the private sector deal with that. It is much more important to have a similar legal system, for example the US-style common law like in Hong Kong or Singapore.

In a recent paper you claim that the precautionary principle should be used to prescribe severe limits on genetically modified organisms (GMO). Why are you so concerned?
Because the risk of GMOs is systemic. Engineered modifications to ecological systems are different from bottom-up ones like nature does. Absent humanity the biosphere engages in natural experiments due to random variations with only local impacts.

Top-down modifications bypass the evolutionary pathway and unintentionally manipulates large sets of interdependent factors at the same time with dramatic risks of unintended consequences. The burden of proof about absence of harm falls on those proposing GMOs, and not on those opposing it.

What about nuclear energy?
Different to GMOs the potential harm from nuclear energy is localized. The risks are better understood and not systemic.