Practical analysis for investment professionals
03 August 2020

Creating Anti-Fragile Portfolios

Long or Short Volatility?

I once worked as an equity derivatives intern at Credit Suisse First Boston in London. As at other investment banks, the team had three distinct types of members: salespeople, traders, and structurers. The latter were almost exclusively polymaths from the top French engineering schools who had few job opportunities in Paris but surprisingly well-compensated ones across the English Channel.

Their core role was to create innovative new products that they first pitched to the team during daily 7 am meetings. In one such meeting, the managing director asked if the proposed product was long or short volatility.

Subscribe Button

The structurer was stumped and could not run through the complex derivative solution quickly enough. So he blushed and mumbled that he would revert later with an answer.

The question has stuck with me ever since. It is not a common one in the asset management industry. Most investors are pretty much the same across asset classes and their investment philosophy is relatively easy to grasp after a short conversation, whether they allocate capital to stocks, bonds, or real estate. They buy something because it is cheap, they follow trends, or invest in quality.

In contrast, speaking with someone who works on a derivatives desk is an almost alien encounter. It is all about gamma, delta-hedging, and similar terminology borrowed from the Greeks.

But after years in the investment industry in roles varying from real estate investor to hedge fund manager, I’ve found the question of whether a portfolio is long or short volatility has risen almost to the top when it comes to long-term asset allocation.

Let me make the case.

Ad for The VIX Index and Volatility-Based Global Indexes and Trading Instruments

Paper Diversification

Most asset classes are bets on economic growth. Companies struggle to grow earnings when growth is declining and corporate and sovereign bond default rates rise.

Some asset classes — private equity or real estate, for example — ostensibly provide diversification benefits. But that is only on paper. Their returns are calculated using lagged and smoothened valuations. Try selling that private company or commercial building at its recent valuation when the economy is falling into recession.

In what looks almost like a gigantic Ponzi scheme, everything depends on the global economy’s continued expansion.

So, what drives economic growth? Broadly speaking, it’s the change in productivity and the working-age population. The former is a vague concept, the latter crystal clear.

In theory, the technological innovations of recent years should have led to massive productivity increases. But economists haven’t been able to make a statistical case for this.

By contrast, population trends are easy to understand. The populations of most developed countries and many emerging ones are shrinking. For example, China is expected to lose 400 million people between now and 2100. That’s more than the current population of the United States.

The lack of productivity and population growth leads to lower or maybe even negative global economic growth in the medium to long term. Japan serves as a real-life case study. To be sure, my own view may be skewed from having lived in Japan for years. But I’ve walked through entire villages that have been entirely deserted due to population declines. Against such structural headwinds, the unconventional monetary policies of recent years seem wholly inadequate.

From this perspective, endowment-style portfolios that are diversified across asset classes are more or less all the same flavor ice cream. They require economic growth and benefit from low or falling economic and market volatility.

Put another way: They are short volatility.

Ad for The Productivity Puzzle

Long Volatility Strategies

Naturally, some strategies do exhibit low correlations to traditional asset classes. The hedge fund universe comes to mind. But most hedge funds have either high correlations to equities (long-short equities), tend to fail in crises (merger arbitrage), provide little alpha over long time periods (equity market neutral), or are not hedged at all (distressed debt). And almost all are expensive.

The managed futures category is one notable exception. Managed futures have structurally low correlations to stocks and bonds, are supported by an abundance of academic research, and are available as low-cost mutual funds and exchange-traded funds (ETFs).

However, low correlation to equities and doing well when economic and market volatility increases or stays elevated for years are not the same thing. Eurekahedge, a hedge fund data provider, constructs indices for funds that focus on tail risk and long volatility strategies. Both strategies shared some performance trends over the last 15 years — which is to be expected given similarities in portfolio construction — but also some differences.

Tail risk and long volatility funds generated high returns during the COVID-19 pandemic in 2020 and therefore delivered crisis alpha. But long volatility strategies did better amid the global financial crisis (GFC) in 2008 and the high-volatility years that followed. Although the Long Volatility Index also lost money when volatility declined due to quantitative easing post-2011, returns were far less negative than those of tail risk funds.

Since we’re looking for a strategy that benefits from rising and structurally higher volatility rather than singular extreme market events, this analysis will focus on long volatility strategies.


Long Volatility and Tail-Risk Strategies vs. VIX

Source: Eurekahedge, FactorResearch

For the last 30 years, bonds offered attractive diversification benefits when equities declined. But those days are over. Bonds have become much less useful in asset allocation since yields in most developed markets are low or negative. The end of the fixed-income bull market also dampens the return outlook for such leveraged asset classes as private equity and real estate, which rode high amid declining interest rates.

But most critically, none of these asset classes can be expected to perform well in a reduced growth world. After all, they offer similar exposure to what we’ll call the economic factor. As such, portfolios diversified across these public and private asset classes are short volatility and essentially fragile.

So how do investors create anti-fragile portfolios geared for a world of reduced economic growth where fixed income no longer serves its traditional role in portfolio construction? Long volatility strategies may be an option. Their correlations to the S&P 500 and bonds were -0.32 and 0.26 between 2004 and 2020, respectively, and they delivered uncorrelated returns. Of course, their performance suffered amid periods of declining volatility. And at times they have been painful to hold. Still, the same can be said for any other asset class. Equities certainly were no picnic during the bear market of 2007 to 2009.


Long Volatility Strategies vs. US Equities and Bonds

Chart depicting Long Volatility Strategies vs. US Equities and Bonds
Source: Eurekahedge, FactorResearch

Adding Long-Volatility Strategies to a 60/40 Portfolio

How would a traditional US equities and bonds portfolio have performed with an allocation to long volatility strategies? We looked at the 16 years between 2004 and 2020, a period that includes multi-year bull markets in equities and bonds as well as two severe stock market crashes.

Although a 20% allocation to long volatility strategies slightly reduced the annual return of a 60/40 portfolio, volatility declined even further, therefore increasing the risk-adjusted returns. But the real benefit of a less-fragile portfolio is demonstrated by calculating the maximum drawdown, which decreased by almost 50%.


Adding Long Volatility Strategies to a 60/40 US Equities-Bond Portfolio

Source: FactorResearch

Constructing portfolios that are less sensitive to the economic factor makes intuitive sense and the simulated results show the attractive diversification benefits for traditional equity-bond portfolios. But hedge fund indices are prone to various biases that tend to overstate returns and understate risks. The Eurekahedge Long Volatility Index currently only has 10 constituents, which is more than it has had in the past, which means investors need to be wary of the historical returns.

Fund managers tend to start reporting their returns to databases when they’re doing well and stop when performance tails off. We can partially correct for this reporting bias by reducing the annual returns of long volatility strategies between 250 and 750 basis points (bps) per annum. Although this reduces the performance of an anti-fragile portfolio, it does not change the significant reductions in maximum drawdowns during the GFC or the coronavirus crisis.


Long Volatility Strategies Adjusted for the Reporting Bias: Max Drawdowns

Source: FactorResearch

However, the favorable risk characteristics of the Eurekahedge Long Volatility Index could be due to a single manager and therefore more the product of luck than skill and not captured by most other managers. We do not have the data to answer this, but it would be worth further evaluation.

Further Thoughts

The COVID-19 crisis reminds us just how fragile the world is. Unfortunately, other events could have similarly devastating effects. Solar bursts could take out the energy grid and satellite communications. Massive volcano eruptions in Mexico City or Naples could envelop North America or Europe in clouds of dust for months. An earthquake could hit the Bay Area of California — the list goes on.

But protection against natural disasters isn’t the only rationale for anti-fragile portfolios. Weak demographics may inhibit global economic growth and create social unrest. What if underfunded pensions funds in the United States and Europe start declaring insolvency and cutting benefits?

Financial Analysts Journal Ad

And if that does not create meaningful issues for society, then there are always purely human-created disasters on the horizon. Argentina was once one of the world’s wealthiest nations, Myanmar was the richest country in Southeast Asia, and Venezuela has the largest oil reserves on earth.

The future of humanity is bright. But it won’t be a smooth ride. Over the last three decades, thanks to economic and productivity growth across the globe, investing has been like driving the German Autobahn. There have been a couple twists and turns, but it’s largely been a quick, steady, and uneventful drive.

But that is changing. The journey over the next decade will have its share of bandits, potholes, and broken glass. So better invest in some insurance and a vehicle that can handle the bumpy road ahead.

For more insights from Nicolas Rabener and the FactorResearch team, sign up for their email newsletter.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.


All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©Getty Images / George Fairbairn

About the Author(s)
Nicolas Rabener

Nicolas Rabener is the managing director of FactorResearch, which provides quantitative solutions for factor investing. Previously he founded Jackdaw Capital, a quantitative investment manager focused on equity market neutral strategies. Previously, Rabener worked at GIC (Government of Singapore Investment Corporation) focused on real estate across asset classes. He started his career working for Citigroup in investment banking in London and New York. Rabener holds an MS in management from HHL Leipzig Graduate School of Management, is a CAIA charter holder, and enjoys endurance sports (100km Ultramarathon, Mont Blanc, Mount Kilimanjaro).

11 thoughts on “Creating Anti-Fragile Portfolios”

  1. Norbert Mittwollen says:

    Hi Nicolas, thank you for your very valuable article. Active long volatility may be perfectly complementing trend-following managed futures for mitigating equity portfolio risks during most diverse black swan scenarios without opportunity costs. This is because it seems to react much surer, faster and stronger to events like COVID-19 within days due to its stronger long volatility bias than active trend. Active long volatility also tends to keep its crisis alpha gains better or to even add up to them during rebound phases as long as volatility remains elevated.

    However, during usually extended times of well performing equities with low volatility this defensive asset struggles to just break even against its cost of carry. But how persistent is this ability of some but not all long volatility funds with quite diverse options and futures strategies, which showed this ability quite homogeneously over the past rather calm 8 years? And (how) can they be systematically selected ex ante?

    Complementary to long vol., active trend can gain much more during pronounced equity bull markets and during slow and less volatile crises like dot.com. This provides for its equity-like level of non-correlated return long-term proven for many decades. However active trend lags more or less for weeks into a short crisis like in March to build up anti-correlation and crisis alpha returns and looses quite substantially of them during rebounds.

    Altogether, if active long volatility succeeds to persistently provide long-term profitability over complete market cycles like active trend besides excellent crisis alpha it may become the second really beneficial alternative asset for long-term buy&hold investors as an effective and cost-efficient protection against expected more turbulent times. Chris Cole from Artemis strongly recommends it already for the really long-term of a century in his Dragon Portfolio besides the only useful other alternative strategy of active trend.

    Indeed, more effective, robust and profitable diversification is hardly imaginable. Good to know from these two gentlemen in which two small corners to look for real value in the vast HF space!

    1. Hi Norbert, thanks for sharing your perspective. Fund selection is challenging across asset classes and likely even more so when the complexity of portfolio construction increases. There is no easy solution for this.

      It’s worth highlighting that many long vol funds shared trends in performance, which indicates that they are harvesting a systematic risk premia. Given this, it should be possible to replicate their pay-off profile systematically, which is a topic on our R&D list, so stayed tuned.

  2. Kirk Cornwell says:

    As we contemplate the possibility and meaning of forever low interest rates, the 60/40 portfolio becomes a retreat from advisor/manager responsibility (as is buying a .56% 10-yr bond for “income”). Yes, a partial answer may be found by hedged allowance for volatility, but even though “growth” may be harder to find, the fact remains – through recent and future black swan events – that the real winners will be those who take the risk and find it.

  3. Johnathan Nichols says:

    Thank you Nicholas, for your cogent and concise observations. The trick, as ever, is finding a robust solution available to non-institutional investors. Usually they are way too expensive, too new (little/no robust track record), or both. Would be intrigued to hear solutions to this issue.

    1. Hi Jonathan, unfortunately we’re not aware of any liquid alternative mutual funds or ETFs that offer the same exposure as long vol funds that are available to institutional investors. We’re currently researching if these can be replicated with liquid instruments and will provide an update on our analysis when available.

  4. Norbert Mittwollen says:

    @Kirk: The real winners will be those, who combine both correlated risks of secular growth as well as non- to anti-correlated risks of secular change and disruption. The first risks are easy to take through passive indexing. However, micro-efficient markets render popular active search for presumably more attractive growth risks beyond pure indexing useless.

    Thus, it is much more rewarding to find and take good alternative risks, particularly in the spaces of active long volatility and trend. This is because they exploit macro-inefficiencies of mostly under-priced tail risks, which are robustly available in all markets.

    @Johnathan: As Nicolas and Chris mentioned, the Eurekahedge Long Volatility index is a good starting point for solutions. As long volatility suffered during the past 8 years, those funds, which managed to just break even over at least 5 hard years after costs of carry and fees, may have proven their competitive performance sufficiently already to be eligible for allocation.

    This is because all of them performed equally well during volatile crises, no matter what strategies they applied. However, during the long low volatility phases before the Covid-19 crisis, low performance dispersion occurred only with those few competitive long volatility funds, which succeeded to break even.

    I conclude from this homogeneous behaviour of the industry leaders that active long volatility may provide an at least equally robust and profitable systematic risk premium as active trend. This is certainly due to the strongly negative correlation between equity volatility changes and returns.

    Thus, I think it is the right time for prudent investors to start with tentative allocations. At least for those, who successfully allocated active trend already. Both allocations are mainly for securing their beneficial risk mitigation properties for rough times to come as well as their good chances to generate attractive long-term returns, at least better than non-yielding long-term bonds.

    As with all active approaches, careful due diligence, based on thorough understanding, is mandatory but worthit. Would be interesting how you see this.

    1. Kirk Cornwell says:

      Norbert – I submit that the market is “micro-efficient” except when it isn’t. Being willing to underperform indexes is part of the game for some of us — and makes big winners sweeter.

      1. Norbert Mittwollen says:

        Kirk – ok, this is fine for the beta style of a behavioural portfolio, preferred by the individual investor. For all investors it is equally important to make their portfolios antifragile, as discussed here.

  5. Victor Kamendrowsky says:

    Anti-fragility was a concept discussed by Nassim Taleb in a book by that title.. In it the author hinted at what such a portfolio would be like. As I recall, he had in mind risk rather than volatility. Would mind commenting on that?

    1. Norbert Mittwollen says:

      Volatility is usually exploding when left tail risk is materializing during severe crises, such as in 2008 or in March 2020. During times of presumably low risk, when market valuations are skyrocketing, volatility is usually low and thus under-priced.

      Thus, buying volatility, when it is under-priced during boom phases, quasi as insurance against left tail risk, in order to sell it at exploded prices, when this risk materializes during deep crises, is quite a smart defensive strategy, originating partly from Taleb.

      Thus, with competitive long-volatility trading strategies you can get a kind of insurance that is not only free but can earn you a profit like equities over a whole market cycle, if the Eurekahedge index is a useful index for this alternative return potential.

      The price you have to pay for that is the risk that the payout is not a guaranteed hedge of the nominal losses in equity value during the next crisis.

  6. Victor Kamendrowsky says:

    Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *



By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

Close