Opinion

Felix Salmon

Is it possible to hedge tail risk?

By Felix Salmon
July 13, 2010

Pine River Capital Management has just launched a new hedge fund. You’ll like the fee structure: there’s no incentive fee at all, which makes for a welcome change from the standard structure where the fund manager takes 20% of the profits. But you might not like the performance: it’s designed to lose between 12% and 18% per year playing in the options market.

Why would anybody invest in such a product? As insurance: the idea behind the fund is that it will soar in the event of extreme market chaos. It’s a productized form of tail risk hedging, and it gives a pretty good indication of how difficult and how expensive true tail-risk hedging really is. Especially since there’s no guarantee that the fund will actually work as hoped.

Deutsche Bank’s Ken Akoundi has a great 21-page primer on tail risk hedging, which lays out the various options. For those of us who rely on old-fashioned diversification across asset classes, there’s this handy cut-out-and-keep chart:

correlation.tiff

What you’re looking for here is numbers less than zero. At zero, there’s no correlation at all: for instance there’s no correlation between U.S. bonds and managed commodity futures. Less than zero, and you’re likely to at least partially make up in one asset class’s gains what you lose in another asset class: for instance, if you’re long U.S. equities and also long volatility, then when stocks crash and volatility spikes, you’ll do better than if you just held stocks on their own. The correlation between stocks and volatility is very low, at -0.65.

The other asset class which has a negative correlation with stocks is, again, those managed commodity futures — which are not to be confused with commodities themselves. Those have a positive correlation. It’s worth noting that diversifying internationally doesn’t seem to help at all: U.S. stocks have a +0.93 correlation with foreign stocks.

The problem with trying to invest in asset classes like volatility or managed commodity futures, of course, is that it’s expensive and difficult to do so. You can’t just go out and buy an ETF. Deutsche Bank has its own proprietary products, with names like ELVIS and EMERALD, which try to give pension funds the ability to invest in these asset classes, but again it’s hard to know whether they’ll work ex ante. As everybody knows, in a crisis, correlations all tend to zoom towards 1.

So what other options are there for hedging tail risk? Akoundi presents a pretty long list. There are complex things like variance swaps, inflation floor agreements, and tail risk protection indices, but there are also simpler ideas like buying out-of-the-money index puts, or buying credit protection in the CDS market. And then there are the kind of strategies which ask “if there’s a crisis, what’s likely to happen to certain assets”: Akoundi cites as examples the “sovereign risk commodity hedge” of buying calls on gold and puts on oil and aluminum, or the “sovereign risk rates hedge” of buying something known as a  low‐strike receiver swaption in USD.

I’m not a huge fan of these strategies, because they only work if (a) there’s a crisis like the crisis you think might come, and (b) it plays out in the way that you think it will. Crises, of course, have a way of being unexpected, both in terms of where they come from and how they play out. That’s why someone like Peter Schiff could position his investors for the coming crisis, see the crisis materialize, and still lose his investors lots of money.

There’s another way to deal with tail risk, and that’s the Nassim Taleb approach: put 90% of your money in Treasury bills, and then invest the other 10% in options and other instruments which normally go down but which are likely to pay off massively if there’s a crisis or a major spike in inflation. If they don’t, well, at least you still have 90% of your money in Treasury bills. But that kind of strategy is much harder to pull off if the bulk of your money is in stocks rather than risk-free investments.

Ultimately, tail risk is something that’s very expensive to hedge, and attempting to do so might well fail. It’s worth thinking about, but some things, while great in theory, just don’t work so well in practice. And I don’t think there are any tried-and-tested tail-risk hedging strategies.

Comments
13 comments so far | RSS Comments RSS

I’m not persuaded that there’s any systematic way to ‘play’ tail risk in a market-oriented strategy– when everything’s going non-linear, correlations can have any value, e.g., somewhere between -1 and +1. To be more concrete I’d like to see a comparison between this and a ‘value’ strategy of being a short-seller.

Posted by MattF | Report as abusive
 

Took particular glee in reading that link to MISH’s takedown of Peter Schiff and Euro Pacific (ya, it was from Jan09 but every bit as relevant today as then!) Now if only every one of these ‘Schiff for Senate’ and Ron Paul-loving num-nums would take 15 minutes and read that whole post America would be a much better place; probably should be required reading for Teabaggers too.
For those not familiar with Mike Shedlock, he is quite a bear but a true analyst – it’s not about saving face or being well known, it’s about making $$ and having a 0 correlation strategy. Cheers!

Posted by CDNrebel | Report as abusive
 

Warren Buffett’s approach to hedging tail risk is to own your assets outright and have plenty of cash on hand.

Then when crisis hits, those who levered up on assets are forced to disgorge them at fire sale prices while you clean up.

Posted by DanHess | Report as abusive
 

Sadly, no. BRK.A’s drawdown and recovery off 2007 highs looks just like the S&P 500′s. If Buffet hedged tail risk and cleaned up at fire sale prices, then all of corporate America did, too.

The real problem with tail risk is that in a true tail event, as Brad Delong does not tire of repeating, the only things that have value are bottled water, sewing needles, and ammunition.

Posted by wcw | Report as abusive
 

@wcw –

Actually, I have long marveled that Berkshire Hathaway together with its many wholly owned subsidiaries would be an awesome post-apocalyptic island in the event of the unthinkable.

Berkshire and its subsidaries are in everything from energy to food, housing, clothes, shoes, transportation and everything in between. As any autocrat could tell you, it is nice to control the productive apparatus of a country when stuff goes down.

If financial markets collapse, those things can keep on chugging. Heck he could issue his own scrip backed by his large real output and not miss a beat.

Posted by DanHess | Report as abusive
 

“I don’t think there are any tried-and-tested tail-risk hedging strategies.”

Sure there are. But they are expensive. That’s the problem.

If you always own a bunch of out o the money puts (and even calls), you are protected against the huge move. Like I said, it’s too costly for most people.

Posted by MarkWolfinger | Report as abusive
 

Technically the absolute value of the correlations of returns goes to 1 in a crisis.

You nailed why these “black swan” funds tend to massively underperform – because knowing what is going to crash in a crisis is hard to know in advance and staying solvent whilst making those bets is hard too. Making a non-time sensitive, more generalised bearish bet is near-on impossible in vanilla products and using a customised product exposes you to counterparty risk – assuming you can even make the trade. The fact is that to get any sort of consistent super-normal return you need to take risks. The key is knowing what risks you are taking and having a view on the correct price for that risk.

Posted by Danny_Black | Report as abusive
 

By the way there is one sure fire way to make money, short an asset class when Pension Funds start to pile in and double up that short when those products start to get sold to the retail public!

Posted by Danny_Black | Report as abusive
 

“If they don’t, well, at least you still have 90% of your money in Treasury bills. But that kind of strategy is much harder to pull off if the bulk of your money is in stocks rather than risk-free investments.”

i find it deeply troublesome that you think T-bills are “risk free.” free from credit risk, sure. but inflation? not even close.

Posted by misterxroboto | Report as abusive
 

“The correlation between stocks and volatility is very low, at -0.65.”

What? According to my understanding of correlations that’s pretty high, but negative and that’s why – as you said:
“when stocks crash … volatility spikes”

Posted by ToulouseLautrec | Report as abusive
 

You omit the other big problem in tail risk, which is credit risk.

If everything *really* goes off the rails will the institutions on the losing end of these bets still be solvent?

Also, someone truly serious about managing tail risk should probably have a decent civil defence plan for their family with stores of food and water (and some places, heating oil) to manage natural disaster. Sure it’s unlikely to be needed – but there’s a reasonable chance that at some point in your lifetime it will be. Financial crises aren’t the only things with major impact that strike rarely.

Posted by sean_from_NZ | Report as abusive
 

What if there is a lengthy period of time without a low probability disaster? This tends to happen for years or decades between disasters, so allocating a portion of assets to this and watching it disappear every six years waiting for an event sounds throwing good money after bad. Accept it will happen and have plenty of liquid capital that can be used then to take advantage of this event when it happens. Rather than spend it on risk mitigation of an event that isn’t likely to happen perhaps put that money aside for the pounce on the opportunity when it does.

Posted by Jwsampair | Report as abusive
 

Wearing a bullet proof vest everyday is a sure way to reduce the risk of a mortal bullet wound to the chest, however, staying inside all day would even be better.

Of course this analogy does not profit you much…the better strategy is learning how to “dodge” bullets. My 8 year old understands moving averages (not to start a discussion on MAs) and this could have curtailed a lot of loss if one simply “dodged” the bullet.

These discussions can be of great help to new investors, but some times its best to keep it simple.

Posted by Nicoli1020 | Report as abusive
 

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/
  •