Tail Risk and Hedge Fund Returns
Here’s a good new paper on tail risk and hedge fund returns. Few things in this business are more poorly understood than alpha creation and risk adjusted returns. Most investors have no idea whether their fund manager is actually generating returns that justify the risk levels or even if the manager is providing a service that justifies a fee that is higher than a correlated index. More often than not, we find that investment managers are mirroring an index or taking on more risk than their returns justify, but the investors don’t realize or understand how to properly benchmark their manager so they continue paying for something that is totally unjustified. AiCIO recently interviewed the authors of the paper:
“We want to distinguish between genuine expertise of a hedge fund manager rather than outperforming by taking on extra risk — anyone can take on extra crash risk and get a higher return,” Kelly told aiCIO following the recent release of his academic paper on the topic, which he co-wrote with Erasmus University Professor Hao Jiang.
If investors want to measure risk correctly, they must be more careful about calculating the returns generated by hedge fund managers’ willingness to take tail-risk positions, exposing them to downside risk, according to Kelly’s paper titled “Tail Risk and Hedge Fund Returns.” The main thrust of the research: hedge fund managers tend to be more willing to take on positions that have a high likelihood of crashing in order to earn higher expected returns.
Paper abstract:
“We document large, persistent exposures of hedge funds to downside tail risk. For instance, the hardest hit hedge funds in the 1998 crisis also suffered predictably worse returns than their peers in 2007-2008. Using the conditional tail risk factor derived by Kelly (2012), we find that tail risk is a key driver of hedge fund returns in both the time-series and cross-section. A positive one standard deviation shock to tail risk is associated with a contemporaneous decline of 2.88% per year in the value of the aggregate hedge fund portfolio. In the cross-section, funds that lose value during high tail risk episodes earn average annual returns more than 6% higher than funds that are tail risk-hedged, controlling for commonly used hedge fund factors. These results are consistent with the notion that a significant component of hedge fund returns can be viewed as compensation for selling disaster insurance.”
Cullen Roche
Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.
Follow Me:
13 Comments
This is old news. There have been countless studies on nonlinear risks and hidden optionality in hedge fund returns. The bottom line is, most hedge funds make money by being short volatility, and nonlinearly so. David Hsieh and Andy Lo have numerous papers on the subject.
Not that I am necessarily disagreeing, but isn’t it only “old news” once these lessons have been learned and “priced in” by participants? It appears to me that most investors are falling for the same tricks in funds that bear a different name….
You are absolutely right about that. In fact, pension funds now are betting big on hedge funds without fully understanding the implications.
I tell you one thing though. I have worked with hedge funds a lot, and it is very difficult to figure out what they do just by looking at their returns history and without a transparent snapshot of the positions.
A perfect example of ‘latent’ risks, that out of the money naked call write that does not show up until it does …
Yes, and as LVG mentioned, the survivorship bias in the HF data is through the roof. So who really knows what the value add is here anyhow? My guess is the numbers aren’t too much better than the MF industry. Of course, there are funds that are great and add value, but compared to what?
Bottom line is, know your fund, its proper benchmark, its true risk levels and how it fits into your overall investment strategy.
The thing that bugs me about hedge fund returns is the extreme survivorship bias. Most of these studies don’t even include the thousands of funds that have blown up or gone under over the last 10 years. It’s even worse than the mutual fund business.
Isn’t the same sort of tail-risk also present when you invest in an index fund? or is it even worse for hedge funds?
Tail risk is just a 3 std dev move so the tail risk for an index fund might be different than the tail risk for a hedge fund. Apples and oranges in most cases. The tail risk for an index fund is likely much lower than it is for most hedge funds.
There are reports that do include graveyard funds. HFR does a decent job at trying to minimize survivorship bias. On the flipside, there are a lot of good funds that have not reported returns.
Only consultants like Albourne Partners or solid FoFs (like PAAMCO) who get access to daily transparency positions and transactions (typically via managed accounts) are in a position to really evaluate true alpha generation. Most funds don’t pass the test…at most they are disguised beta chasers.
But some do creat alpha consistently, and if you can construct a portfolio of some really good managers across strategies, products and geo mandates, while getting favorable liquidity and fee terms through structures such as managed accounts/seed deals (or even commingled funds), solid and persistent returns are possible via hedge funds.
There have been some great funds who have kept their net exposures tight, max drawdowns low over the last five years of tumultuous markets, and have maintained low correlations and betas to global benchmarks, etc. Look at Bridgewater, Tudor and Brevan for examples.
great new site by the way
Tail Risk gets amplified through the use of leverage especially when the leverage is applied to supposedly safe assets to take advantage of the carry trade.
When I worked at Bernstein we had a hedge fund called the Multi-Income Strategy Fund in A, B, C, flavors which differed in their use of leverage to correspond with a supposedly conservative, neutral, or aggressive investor. In theory, during a full market cycle this fund was supposed to return excess alpha of at least 4 percent versus equivalent duration treasuries with the same level of risk.
What happened next?
You guessed it, those “safe assets” were mainly Triple AAA rated, illiquid MBS’s which formed the majority of the holdings. This fund proceeded to blow up in late 2007 at a rate that literally took my breath away – the aggressive portfolio lost 60% in a single month in Jan 08 after having already suffered massive losses. Now that’s aggressive!
All the while, the quants who were running the fund kept saying it would come back because the models they used said it would. In their defence the VAR calculations did indeed show that these funds were not risky until they became so. That is the message, the strategy works until the environment changes and we get those so-called once in a universe events that seem to happen so regularly. Apparently the irony of this recurring statistical imponderable never seems to stop the shills from continuing to promote the mathematical safety of their funds.
The major lessons I learned from this debacle is that the more impenetrable the prospectus, the more likely it is to blow up and secondly not to rely upon the sponsor’s marketing description of the fund.
I was so disgusted by this episode that I quit. I know it was repeated at every major fund so it was nothing unique to Bernstein which is what made it even more disturbing for me. This type of groupthink from the supposed “best minds in the universe” has made me a cynic of the industry ever since. I know there are good companies, however it is a fee driven industry and that is what drives behavior. They win no matter what.
End of diatribe!
Actually, global leverage leverage levels are quite low for hedge funds…averaging less than 2x. Prime brokers don’t like this of course. Couple that with low volumes, it’s a tough time in hedge fund land.
Compare hedge fund leverage levels to that of prop desks during the early 2000s, hedge funds would seem like safe havens.
There are memorable exceptions such as LTCM which leveraged as high as 100x. Its relative value fixed income arbitrage strategy depended on mean reversion that never panned out and was rightly described as “vacuuming up pennies in front of a steam roller”.
That’s why I only look at very liquid strategies like market neutral equity, CTA/Managed Futures, certain macro strategies, trading mostly Type 1 assets and some Type 2, but never Type 3 which are very illiquid(according to FAS157).
Hey MCL,
Do you find that leverage and even vol can be quite low on the whole, however, the tail risk can still be quite high? This has been my conclusion in most hedge funds compared to correlated indices. Strategies are often opaque and concentrated leading to the perception that vol is low and that risk adjusted returns are high, but the black swan event is often all that’s needed to destroy years of returns.
Actually this is not entirely correct – different strategies are using different levels of leverage for example the quants/stat arb guys are still using upwards of 10x.