There are few performance industries that are more obsessed with data than investment management. Data analytics is at the core of everything we do. The data we study broadly fall into three categories: 1) input data for decision-making and risk management 2) output data (e.g., performance) and 3) investment decision data.
How much time does the average fund manager spend on each of these categories? In my experience, very little time is spent on studying investment decision data. It is often overlooked, and in many instances, the data are not even collected for analysis.
This stands in stark contrast to another data-obsessed industry, Formula 1 — where during the course of any given race, hundreds of sensors on the cars are streaming real-time data to the pits about a car's performance and the driver's decisions. A team of engineers analyzes the data and feeds back information to the team principal and the driver. Investment management as an industry is about 20 years behind Formula 1 in terms of understanding its own investment decision data.
Here are four questions that asset allocators should ask (and fund managers should be able to answer) around core investment skill areas where active portfolio managers can add value, position sizing, trading acumen, selling discipline and timing of investment decisions.
How did your portfolio perform vs. an equal-weighted portfolio over the same period?
This question can be answered by comparing the returns of an actively weighted portfolio to an equal-weighted portfolio. The difference between the two portfolios is the excess positive or negative return generated from the portfolio manager's sizing decisions. The frequency of rebalancing the equal-weighted portfolio depends on the time frame of the analysis and the turnover of the investment strategy. This is similar to looking at tracking errors for funds with a defined benchmark. The measure provides an indication on the investor's skill in effectively allocating capital.
How did your portfolio perform vs. a no-turnover portfolio?
This analysis can be achieved by comparing actual portfolio returns with a “static” or no-turnover portfolio. The difference between the two portfolios' returns is the amount the fund manager is adding or subtracting to returns through active trading of positions. If a fund manager is choosing to actively trade around positions, it is an investment decision. They should be able to demonstrate that the decision adds value or amend the behavior if it does not. This analysis can point to some other common investment behavioral problems such as overtrading.
Would a 10% stop-loss have helped or hurt your performance in the past 12 months?
Stop-losses are one of the first-line defense tools in risk management. However, portfolio managers' rules around stop-losses often are extremely vague. Alarmingly, almost every professional investor that I have asked has not been able to answer the question above.
We can think of stop-losses as a percentage move on a position, as a stop on a percentage of NAV or even a nominal loss amount. Either way, if a portfolio manager chooses not to use a stop-loss, they should know if that decision is helping or hurting their performance and by how much.
This analysis is the first step in looking at stop-loss optimization for a particular investment strategy. It might be worth conducting further analysis if there are substrategies with different volatility characteristics.
One can further analyze selling skill by conducting a similar analysis for profit taking. The question then would become: Would a rule to take profits on positions up 15% have helped or hindered performance of the portfolio? This question should be modified to the investment strategy. If the portfolio manager sets price targets for securities, then one could ask: How would a take-profits rule at 85% of the price target have helped or hindered performance?
Generally speaking there is far less time spent on the discipline of selling and thinking about selling strategies for both winning and losing positions than there is time spent on assessing securities for buying.
Timing of entry / exit
How would your performance have been affected if you had waited 15 days before entering the position?
This question is more applicable to higher turnover investment or trading strategies. A common complaint that is often heard on trading floors is “I was going to put the position on but it started to move.” Conversely another oft-heard complaint is, “I am typically very early to the trade.” These questions can be analyzed by replicating a portfolio's performance by lagging or leading the actual trade dates by the variable time horizons you want to test for. The exercise can be conducted on both entries and exits.
Worth the investment
Analyzing investment decision data can be complex and rather expensive. However, there are third-party software solutions that can help portfolio managers assess their core skill sets as an investor and so we can begin to measure skill over time. Once we can measure investment skill, we can work to improve skill and, hence, investment returns. Both allocators and investment managers should be investing time and resources toward investment decision analytics.
Josh Jacobson is chief operating officer-equities at Cheyne Capital LLP, London. This content represents the views of the author. It was submitted and edited under P&I guidelines, but is not a product of P&I’s editorial team.