Tag Archives: benchmark

The New Definition Of Investment Manager Success: How To Tell Who’s Winning

It’s become self-evident recently that peer groups suffer from “loser bias” because the majority of active managers underperform their benchmark. Beating the losers is not a “win.” Peer group comparisons simply don’t work anymore. Beating the benchmark is a good beginning, especially when combined with a statistical test of significance called a “Success Score.”. If intermediaries continue to use peer groups, as is likely the case, investors will continue to be disappointed because they’ll continue to hire losers. In the “good old days,” investment managers had two shots at winning. They could beat their index or they could beat the median manager in their peer group. That peer group thing doesn’t work anymore. Due to the popularity of passive ETFs and the emergence of Robo Advisors, there is only one pertinent yardstick – beating the benchmark. Unfortunately , less than 20% of active managers achieve this measure of success. This active manager failure renders peer groups worse than useless. It is now well-understood that peer groups suffer from “loser bias,” in addition to survivor and classification biases. Loser bias is the reality that more than 80% of the managers in a peer group are losers since they fail to beat their benchmarks. Beating the losers is like winning the prize for best ballerina in Waco. Investors need to demand better. So the new definition of “success” is beating the benchmark, but there’s more to winning than this simple measure. We want to know that success is not just luck, that it is likely to repeat in the future. That’s where statistics and “Success Scores” come in. We call it a “win” if the outperformance of the benchmark is statistically significant. Success Scores are the statistical significance of benchmark outperformance. A facsimile of a peer groups is created by forming all the portfolios that could be formed from the stocks in the index. A ranking against these Success Scores in the top decile is significant at the 90% confidence level – we can be 90% sure that it wasn’t just luck. Success Scores are bias free and available a day or two after quarter end. It’s not enough to beat the benchmark. An investment manager needs to beat his benchmark by a significant amount to be a true winner. Success Scores are especially worthwhile for hedge fund managers since peer groups of hedge funds are just plain silly. The tradition of disappointment in active managers will continue if clients (investors) allow it to continue. Clients deserve better,but they need to know how to get it. Investors need to understand their advisor’s due diligence process and to be concerned if it includes peer group comparisons. In other words, investors should seek out advisors who employee contemporary due diligence tools if they are relying on their advisor to select good investment managers. Here are some facts every investor should know: Based on Dr. William F. Sharpe’s “Arithmetic of Active Management”, 50% of active managres should beat their benchmark. The fact is only 20% beat their benchmark, far below expectations. The search for “alpha” uses regression analysis. “Alpha” is the Greek letter for the intercept. It is well-documented that it takes at least 50 years for a manager with “average” skill to deliver a statistically significant alpha. By contrast, “Success Scores” can provide significance for very short periods, like one year. 70% of managers are active, not passive. Towers Watson, a prestigious investment consulting firm, says this number should be closer to 30%. There are too many active managers. Approximately 40% of funds in a peer group don’t belong because they’re different. This problem is called Classification bias. For hedge fund peer groups, most funds don’t belong because hedge funds are unique, which by definition means without peers. Knowledge is power. Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.

Beyond The Benchmark: Tracking Error Vs. Active Share

Summary We have reservations about using tracking error to gauge “active investing” because it relies on historical volatility data versus a benchmark to draw conclusions about risk. Active share, in our view, provides a clearer picture of how active a fund manager is as compared with drawing conclusions from standard deviation calculations. We believe the fund has to be meaningfully different to its benchmark to create an opportunity to deliver alpha. We believe active share more clearly shows how a fund and benchmark differ, a key to delivering alpha. By Rob Stabler, Product Director Active share, a tool for demonstrating how a fund’s portfolio differs from its respective benchmark, has been a common term among active investors over the last few years. Tracking error, which has a much longer history, is often regarded as another tool that does the same job. But the differences between the two measures affect how Invesco’s Global Opportunities investment team views their effectiveness and usefulness for investors. Tracking error: Useful from returns perspective Tracking error – the divergence between price behaviors of a portfolio and its benchmark – is a backward looking tool, using historical data to show the volatility of the fund’s returns versus that of its benchmark. It’s useful in demonstrating how closely a portfolio follows its benchmark from a returns perspective. However, it’s important to consider these two questions: What’s the benchmark? A fund with a low tracking error versus a volatile benchmark may not produce the return profile investors seek. Are upside and downside volatility equally important to investors? The most common method of assessing tracking error involves calculating the standard deviation of the fund and benchmark returns, which reflects both upside and downside volatility. In our experience, however, investors have been more concerned about the implications of downside volatility. More importantly, as active investors, our team’s main reservation about tracking error is acceptance of the benchmark as the right reference point for measuring volatility and, by implication, risk. In contrast, the investment world doesn’t revolve around the benchmark for our fund managers. We define risk as the potential for permanent loss of capital, using maximum drawdown and downside volatility as indicators. And we often view volatility – at least in the short term – as an opportunity to exploit valuation anomalies in the stock market. Active share: Looks at holdings and weightings Active share is a much simpler calculation that provides a snapshot in time. It measures how different a portfolio is from its benchmark by comparing the fund’s holdings and their weightings with those of the benchmark. We believe active share provides a clearer picture of how active a fund manager is than drawing conclusions from standard deviation calculations. In simple terms, a tracker fund that perfectly replicates its benchmark will have an active share of 0%, while an active fund that owns no constituents of its reference benchmark will have an active share of 100%. This measure is increasingly important, given the rise of passive investing and the need to differentiate between quasi-passive and genuinely active managers. Origin of active share The concept of active share was introduced in research by Martijn Cremers and Antti Petajisto, which indicated that portfolios with a high active share were, on average, likely to outperform their benchmarks, suggesting a positive correlation between performance and active share. 1 Additional research by Cremers and fellow economist Ankur Pareek 2 combined active share analysis with portfolio managers’ stock holding period, where long duration is defined as more than two years. The research shows clear outperformance, on average, of those strategies that combine high active share and long duration, or low turnover, of stocks. Of course, past performance does not guarantee future results. Earlier this year, Invesco published a white paper examining the historical outperformance of active management , using active share as the measuring stick for active management. Because high active share offers no performance guarantee, it’s possible to have a high active share portfolio that underperforms its benchmark. However, our team believes that to outperform a benchmark, portfolio construction needs to differ from the benchmark, and active share is a reliable, easy way of measuring this. So while active share doesn’t guarantee performance, we believe it’s a prerequisite – if you aren’t different, then you can’t hope to achieve a different result, good or bad. By-product of investment philosophy While we don’t explicitly target a high active share in the Invesco Global Opportunities strategy, it’s a by-product of our investment philosophy – concentrated and flexible investing that views risk as absolute, not relative. The result is an active share that is typically high, currently at 95%. Put simply, to create an opportunity to deliver alpha for our investors, we believe the fund has to be meaningfully different from its benchmark. In addition, we see no evidence to suggest a direct link between the strategy’s tracking error and performance. Sources “How active is your fund manager? A new measure that predicts performance,” Aug. 7, 2006. Patient Capital Outperformance: “The Investment Skill of High Active Share Managers Who Trade Infrequently,” Sept. 19, 2014. Important information Alpha refers to the excess returns of a fund relative to the return of a benchmark index. Standard deviation measures a portfolio’s range of total returns and identifies the spread of a portfolio’s short-term fluctuations. Drawdown is the largest cumulative percentage decline in net asset value as measured on a month-end basis. Absolute return refers to the return an asset achieves over a certain period of time, without comparison to another measure or benchmark. The information provided is for educational purposes only and does not constitute a recommendation of the suitability of any investment strategy for a particular investor. Invesco does not provide tax advice. The tax information contained herein is general and is not exhaustive by nature. Federal and state tax laws are complex and constantly changing. Investors should always consult their own legal or tax professional for information concerning their individual situation. The opinions expressed are those of the authors, are based on current market conditions and are subject to change without notice. These opinions may differ from those of other Invesco investment professionals. NOT FDIC INSURED MAY LOSE VALUE NO BANK GUARANTEE All data provided by Invesco unless otherwise noted. Invesco Distributors, Inc. is the US distributor for Invesco Ltd.’s retail products and collective trust funds. Invesco Advisers, Inc. and other affiliated investment advisers mentioned provide investment advisory services and do not sell securities. Invesco Unit Investment Trusts are distributed by the sponsor, Invesco Capital Markets, Inc., and broker-dealers including Invesco Distributors, Inc. PowerShares® is a registered trademark of Invesco PowerShares Capital Management LLC (Invesco PowerShares). Each entity is an indirect, wholly owned subsidiary of Invesco Ltd. ©2015 Invesco Ltd. All rights reserved. Beyond the benchmark: Tracking error versus active share by Invesco Blog

Alpha Wounds: Passive Management Is Not Passive

By Jason Voss, CFA Alpha wounds are decisions made by the investment industry that hurt active investment managers. It is my belief that there is still plenty of alpha left to be harvested by discerning research analysts and portfolio managers. So far, I have discussed the deleterious effects of managing to, rather than from, a benchmark ; poor evaluative methodologies by investment industry adjuncts; and the poor diversification of the human resources portfolio at active management houses. This month I point out a fact hiding in plain sight: Passive management is not passive. One of the tremendous and rarely discussed ironies in the active vs. passive debate is that passive management is thought of as the opposite of active management. That is, it is perceived as a ship set adrift in an ocean with no compass heading and no crew. Passengers are on board and left to fend for themselves. I politely disagree. Passive management is not blind, deaf, or dumb. In fact, for every index and for every fund or exchange-traded fund (ETF) designed to track it, human choice is involved. As I have discussed before in an entirely different context, choices are actions , that is, activity. That is, we are talking about active investing. To be fair, passive investing is not exactly “active” investing. It is really more like “less active” investing. Given a) the consistent inability of active managers to beat benchmarks, and b) the fact that passive investing actually involves active choices, maybe it makes sense to see what the indices are doing, right? . . . Right? Case Study: The S&P 500 Let’s consider one very famous index, the Standard & Poor’s 500. I hope it is indisputable that the S&P 500 is among the best-known indices and hence a proxy for stock market activity in the United States. Is an index fund or ETF that tracks the storied S&P 500 truly passive? Absolutely not. Many do not realize that a small committee at Standard & Poor’s oversees and makes decisions about the index. Specifically: “S&P Dow Jones U.S. indices are maintained by the U.S. Index Committee. All committee members are full-time professional members of S&P Dow Jones Indices’ staff. The committee meets monthly. At each meeting, the Index Committee reviews pending corporate actions that may affect index constituents, statistics comparing the composition of the indices to the market, companies that are being considered as candidates for addition to an index, and any significant market events. In addition, the Index Committee may revise index policy covering rules for selecting companies, treatment of dividends, share counts or other matters.” To me this sounds very similar to a description of the activities of an investment committee at an actively managed mutual fund. Yes, there is certainly a demure, passive tone. No doubt. But there are decisions being made here. Which brings me to my next point. Perhaps active managers would be wise to examine the nature of the decision criteria made by this committee in order to improve their own results. This is especially true if, like many funds, the S&P 500 is their benchmark. Put another way: What is this committee doing so incredibly right so as to best a majority of those competing against it? Here are the criteria that the US Index Committee consider: Market capitalization Liquidity Domicile Public float Sector classification Financial viability Treatment of IPOs A list of eligible securities Additionally, there are criteria for deleting an issue. Some of the above may seem simple on the face of things, but let’s drill a little deeper. The Hidden Story Inside Market Capitalization Market capitalization is indicative of some unique characteristics of a business. For example, a large market capitalization is likely the result of a highly successful business with in-demand products, well-established markets, a strong competitive position, that is professionally managed, well capitalized financially, and for which all of these things have been true over a long period of time. Heck, it is also more likely than not that the business pays its shareholders back with share buybacks, or – gasp! – dividends. In other words, large market capitalization is a natural outcome of running a successful business. The Remedy for the Alpha Wound: Could “active” managers also consider such criteria in conducting fundamental analysis? Could active managers actually roll up their sleeves and engage in some good old-fashioned fundamental analysis? Low Turnover Like most indices, the components of the S&P 500 do not change very frequently . A review of the historical data from 2002 through November 2015 shows 69 additions (and, hence, deletions) from the index. That works out to a turnover ratio of just 1.06% [(69 changes ÷ 13 years = 5.31 changes per year on average) ÷ 500]. Compare that with the average turnover ratio of 124.6% in the United States in 2012 (the last year for which data is available), and an average of the major global equity markets of 89%. Is there any possibility of actually understanding the companies in which you have placed your investors’ cash in these circumstances? Said differently, US investors have 117.5 times the turnover of the S&P 500. Given that most of the trading is likely in S&P 500 stocks, that the turnover of the index is so low, and that active managers have underperformed, does it seem like a possible self-inflicted alpha wound? In the most positive light, this is a trading desk enrichment program. The Remedy for the Alpha Wound: Could an “active” manager perform better by reducing its turnover? Diversification Another possible lesson to be learned from looking at indices is that each of them represents a diversified portfolio within a given context. For the record, I am personally against what I and many others call “deworsification”. Forthcoming research from C. Thomas Howard, CIO of Athena Investment Management, and a brokerage firm I cannot mention quite yet, entitled Why Most Equity Mutual Funds Underperform and How to Identify Those That Outperform, demonstrates that most fund managers are horribly diversified – as in overly so. The researchers estimate that for every one-decile increase, that over-diversification subtracts 13.5 basis points (bps). Also, they estimate that for every one-decile increase in closet-indexing, that performance is negatively affected by a whopping 31.6 bps. So as managers r-squared relative to their benchmark increases, performance decreases. It is important to remember that originally indices were created not as investment vehicles, but as a way of summarizing the performance of an entire market in one number. No one is likely to have originated the idea of investing in 500 companies. One benefit of being fully invested in each component of the S&P 500 is you end up buying every winner. But you also end up buying every loser. One simple strategy, and I am surprised that it is not deployed more frequently, is to buy the S&P 500 but to conduct fundamental analysis of its components and identify the handful of firms you believe have the highest probability of performing poorly. Then either exclude these from your index-like fund or short them. The Remedy for the Alpha Wound: Could it be that active managers are hurting alpha by over-diversifying and closet-indexing? “Passive” Investing Free Passes Passive investing gets three massive free passes. First, frequently risk-adjusted returns are calculated relative to the benchmark. This means that because benchmarks are both the numerator and the denominator in such calculations, their risk is always cancelled out. This implies that benchmarks have no risk. Clearly this is bogus. What is needed is a neutral way of evaluating risk to which both the benchmark and the active manager are compared. Second, benchmark returns are always gross of fees. Yet, if you read through the S&P Dow Jones report I referenced above, you get the sense that there is a large team making these decisions. What is the expense of creating and maintaining these indices? Also, the expense of buying and selling the securities from the benchmark is excluded. Yes, the turnover is low, but for a true apples-to-apples comparison, shouldn’t these be included? As a proxy, many investment industry adjuncts evaluate index funds tracking a particular benchmark in order to estimate these expenses. This is clearly fairer to active managers. The third and likely largest of the free passes handed to passive investors is the massive momentum effects of their “buy lists.” Indices are effectively “buy” lists. For the larger indices this means that there are huge momentum effects embedded into the strategies. So passive investors benefit considerably from non-fundamental factors when their performance is evaluated. To my knowledge, there is no agreed-upon method for how to back these factors out. In conclusion, passive investing is not truly passive. It is more like less active management. Looked at in this way, it makes obvious certain innate characteristics of smart investing that “passive” investors take advantage of. Maybe active managers could learn a thing or two from these strategies. Disclaimer: Please note that the content of this site should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute.