Back to news list

1 Sep 2015

QLAB Invest - Update August 2015

Multi-Asset Strategy Performance as of 31-August-2015

RISK (10Y)
QLAB Asset Allocation
QLAB Dynamic Allocation

RETURNS: Annualised if >1Y
RISK: Annualised standard deviation or volatility, daily data
MDD: Maximum drawdown, peak to trough
Strategy indices live since 1-Jan-2011, net of trading costs, gross of product fees. Investable products may have lower returns and shorter track records


August saw a return to levels of volatility not witnessed for several years. The S&P500 sold off more than -10% in just 5 days before rebounding 6% in the next 2 days. All assets with exposure in the investment universe posted negative returns, including the 2yr and 5yr US Treasury bonds, as rates crept up in the last few days of the month. Crude Oil was the only asset to post a gain for the month but with huge volatility, down -18% before rallying over 20% in the last 3 days.

The result was a negative month for both strategies; however YTD there is still a positive return. Over 1 year the return turns negative due to a particularly high close in August 2014. The current drawdowns are the same as the MTD figures as an all-time-high was set on 31 July. Over a horizon of 3 years and upwards, returns and realised volatility are in line with expectations.

To put this into perspective we compare the strategy returns of the last month and year-to-date versus a few benchmarks.

QLAB Asset Allocation is compared to the Naïve Market Portfolio which we define as the same investment universe held at maximum risk exposure, which is approximately equally weighted. In a single month such as August one can see there was little place to hide and the loss was similar to the Naïve portfolio. However YTD the results are better due to the active management over time.

QLAB Dynamic Allocation is compared to US equities and commodities (CCI Index) as the strategy can take up to 105% exposure to each asset class. Again the results YTD are better due to active management, but over one month there was weaker performance.

This serves as a reminder than market shocks can occur at any time and diversification can also let you down at the same time. That is why the QLAB risk management process accounts for left-tail risk, using a non-normal distribution model that also assumes stressed correlations in its portfolio construction process. When combined with the rules-based asset allocation process, keeping emotions out of the investment decisions pays off over the medium to long term.

It is also worth reminding ourselves that any strategy exposed to risky assets must follow the law of “statistical gravity” and that losses in excess of 2 times the monthly standard deviation can happen every few years. One of the aims of the QLAB investment process is to offer an improvement in return distribution characteristics over the underlying risky assets which on a buy-and-hold basis show larger left-tail risk or negative skew in this respect.

In response to a statistically weak momentum signal from US equities versus the 5yr US Treasury the QLAB models will exit equities and turn off the leverage in the dynamic allocation strategy. Exiting equities can come with an opportunity cost, potentially missing out if there is a subsequent rally, but in January 2008, December 2000, September 1998 and October 1987 it turned out to be very prudent. So whilst the models are not predictive, but reactive, if there are further equity declines, the portfolios are protected. If equities recover, then the strategies will re-risk automatically. It is this high conviction and non-emotional asset allocation process, coupled with robust portfolio construction that offers investors an asymmetric return profile.

To illustrate this, the chart below shows the 1yr rolling returns of the QLAB Asset Allocation strategy against the Naïve Market Portfolio. When the market is performing well, the strategy also does well showing high participation, or beta. The true alpha can vary, just as with a traditional active strategy managed against a benchmark. When the market is performing negatively, an ETF on the market or a traditional benchmark product would also perform negatively. However, the QLAB strategy is able to deliver consistently higher returns by not holding the worst performing assets. This “crisis alpha” can be higher than the true alpha, and this strong relative performance in a crisis allows the QLAB strategy to exhibit lower drawdowns and outperform the market over time.

This investment process also means that correlations of returns are generally low against benchmark portfolios, and the QLAB strategies provide good diversification in an overall portfolio context as a result.

QLAB Asset Allocation and QLAB Dynamic Allocation are accessible in two formats via our product partners:
  • RPM Risk and Portfolio Management (www.rpm.se) manage two Luxembourg funds: QLAB Convexity Fund and QLAB Convexity DL Fund  available to professional investors
  • Neue Helvetische Bank (www.nhbpro.ch) manage two listed certificates traded on the SIX stock exchange available to Swiss domiciled retail and professional investors

Download this Newsletter in PDF
 format: QLAB Update August 2015

Close disclaimer

DISCLAIMER: This document does not constitute an offer, a solicitation, an advice or a recommendation to purchase or sell any investment products associated with the material described herein. The purpose of this document is to describe the principles, research and ideas behind the QLAB Invest strategy indices. Prior to an investment in any product tracking a strategy index, you should make your own appraisal of the investment risks as well as from a legal, tax and accounting perspective, without relying exclusively on the information provided by QLAB Invest. Investment products tracking the indices must be issued or/and marketed by a regulated company. This document is strictly for informative purpose. The single source of the underlying asset data is Thomson Reuters Datastream and QLAB Invest cannot guarantee the correctness of the underlying asset data and cannot be held legally responsible in this regard. Any references made to historical performance up to the official live inception do not reflect actual live performance and can be subject to selection, curve fitting and other statistical biases. Performance in investment products linked to the indices may be reduced by the effect of commissions, fees or other charges in excess of those already factored into the index calculations. The level of the indices will fluctuate due to the volatility of the underlying exposures and past performance or volatility is not necessarily indicative of future results.