AUM growth as a signal of hedge fund quality - Is the investor herd right?

It is hard to determine whether one manager is better than another when looking at performance numbers. The sample sizes are often too small to distinguish return differences, so investors often looking for other signals that can be used to suggests one manager is better. One that is often used is growth of AUM. Call it the "wisdom of crowds" signal. If an investor cannot distinguish the return performance between two managers, he will place weight on the dollar opinion of others. If the herd is investing in manager X, perhaps they know something that others don't. The investor will free ride on the due diligence of others and invest with the manager who is growing faster. 

Unfortunately, this wisdom is a noisy signal. Growth may tell us something about past performance, but it may not be an indicator of future gains. It could be just dollar votes for past gains. There also is a strain of literature that past performance is not predictive of the future because the flow of funds will generate diseconomies of scale. There has been shown a link between past performance and the flow of funds, but we are asking a deeper question of whether fund flows tells us something about future performance that is not included in the past numbers. Additionally, is a growth strategy a tool by the management company to signal quality?

We believe that managers want to grow assets for a number of reasons beyond just the immediate gain of added revenue. Growth may signal quality, so managers will be willing to cut fees in order to grow and show the market there is strong demand for their services.  Discounts in price and the added cost of aggressive marketing may lead to a positive feedback loop that will generate more new revenue. Given this signaling effect, managers may be willing to suffer the risk of some diseconomies of scale and forgo maximizing income over the shorter run in order to generate the quality signal. 

This signaling is especially important if performance is within some tight range relative to other managers. 

• Growth at any cost is important for the small manager who wants to break-out from the pack of other small managers where the dispersion in returns is large. In this case, aggressive pricing and terms are used to gain the AUM signal even though costs for running the firm may not be fully covered. Spending on marketing is critical. 
• For the medium size manager, who may have already covered costs, high growth can accelerate investor interest even if performance is similar to other firms. Here, a strong marketing budget may eat into profit margins but generate longer-term business gains. 
• Large firms who have limited worries concerning fixed costs should continue to aggressively market products in an effort to popularize the firm for new alternatives and revenue. Given economies of scale, marketing is a inexpensive way of add revenue in a measured manner. Fee revenue may be more important than incentive optionality. Similarly, firms will work hard to maintain AUM to avoid the negative signals associated negative growth.

The link between asset growth and signaling may seem obvious for some marketing experts, but we believe this is especially critical when product quality via returns is harder to distinguish. To date, we do not believe that growth-pricing strategies with hedge funds have been fully explored as signals beyond performance.  

Is the herd right? This is a testable hypothesis. Do above average money flows signal better future performance after accounting for any past return effect?  Even a weak signal of quality is worth exploiting, if past performance is also a noisy signal and future relative returns are uncertain. 

Trend-following - A century of data suggests there is value

Consistency is critical for any investment style or factor, and it seems that trend-following seems to show it better than most other alternative strategies. It is now almost amusing that when efficient markets ran supreme as the paradigm of choice for market behavior no one was able to find these results, but now that there has been a shift in our views the scales have been lifted from our eyes and we find trend following is accepted through the ages. See the latest research from AQR Capital Management, "A century of  evidence on trend-following investing." which has been updated through December 2016.

I will not say that trend-following is perfect for all times, but the evidence says that it is an effective core strategy that can always be used by any investors. Prices are primal and their trend behavior is a good foundation for any discussion how to form an effective portfolio.

The research numbers shows that trend-following is effective across all major market sectors and long time periods. The Sharpe ratios net of fees are all positive for every decade since the 1880's. Additionally diversification with equities is strong with correlation with stock indices ranging between -.34 and .33. Both short and long-term trend-following is effective but not necessarily at the same time. The research shows that long-term trend-following is not sensitive to a lagging or delay of the signals. 

There is consistency with the benefits from trend-following which just reinforces the same story that marketers of managed futures funds have been using:

1. There is a non-liner relationship with equity returns, a performance smile whereby trend-following does better during market extremes.

2. The trend-following portfolio will do well during times of market stress. The research focuses on comparison between trend-following and a 60/40 stock/bond mix which is more representative of an investor's portfolio than 100% equities. Trend-following will add return at critical times.

3. When added to a 60/40 stock bond portfolio, trend-following will add return, chop volatility, improve the Sharpe ratio, and cut the maximum drawdown.

Questioning current trend-following performance should be expected, but location decisions should be based on prior evidence. The current drag on performance form holding a trend-following manager is real, yet the long-term evidence suggests that a trend-following is useful and patience should be a guide when making allocation changes. 

Is "directional volatility" needed for trend followers?

What is needed is volatility coupled with price trends, which Ivarsson refers to as “directional volatility”. “What we at RPM look for is ‘directional volatility’, meaning volatility that drives markets in a certain direction”. - RPM’s executive Vice President Per Ivarsson

The concept of directional volatility is elusive. It combines two concepts, the path of prices with the price spread away from the average. The quote is focused on the critical need for a trend with volatility for trend-follower to profit. Volatility is necessary but not sufficient for strong trend-following profits. It is necessary to have prices move across a range in a discernible path, but a wide price range can still be without trends. Trends may occur if there is low volatility but the level of profits will be smaller and the trends will be harder to identify. 

The depiction of trend following as a look-back straddle is based on volatility and the ability of the trend follow to capture the range. The look-back option is the potential max that a trend follower may achieve for a single trade within a time span. Nevertheless, if we think of a stochastic process for a price series, we want a mean change plus wide dispersion. It is not enough to just have volatility, the journey expressed in the volatility is critical. 

We can graph the simplest case of a Monte Carlo simulation of a normal distribution with zero mean and a 10% annualized volatility to generate 10 paths for 252 days (one trading year) to show possible paths for prices. We have also included a set of paths with 10% volatility and a 5% annual mean. A given volatility does not ensure that that there will be a path that can reach extremes over simulated time period.  Though not surprising, a 10% volatility with zero mean will generate paths that look like there is no trend. However, there will also be some paths that do seem to generate trend behavior even though there is no mean. Similarly, the blue lines which simulate paths with a 5% annual mean show paths that do not seem to have any trends. Volatility will not guarantee you a path that will trend higher or lower. 

Directional volatility is good short-form phrase, but it is important to understand the parts that potentially can drive trends. Trend followers want spread in price but they need paths for profit.

ECB CISS index; there is no trend in stress - Be happy

Don't worry be happy and without stress. There is declining stress in the EU as measured by the ECB Composite Index of Systematic Stress (CISS). While there is a big disclaimer with the ECB risk dashboard that this is not a early warning system, the declining trend tell a story of stability. 

This index serves as a European equivalent of the stress indicators that are used by the Federal Reserve Banks; however, there is a greater emphasis on cross-claims country claims, flows, and banking risks. The CISS numbers tell us that Euro area stress has been less volatile and reaching all time lows after spiking during the BREXIT vote.  The US stress indices have also been trending down in 2017. This down trend is consistent with EU equity volatility measures.

The VSTOXX volatility index below has shown a consistent downtrend albeit there have been spikes that suggest liquidity may be at times be strained in equity markets.

The ECB should be likely influenced by these positive stress numbers. A tapering of bond purchases would seem natural given the current state of stress, the lowered deflationary fears, the lower unemployment in most countries, and positive growth. A tilt to a policy change seems appropriate. Unfortunately, global macro managers who often thrive on stress and market dislocations have not fully taken advantage of this less stressful environment.