Indexology^® Blog

As the importance of ESG investing grows, especially in the U.S., the ability to quantify and measure the impact of an ESG-incorporated portfolio will become more relevant.  In evaluating performance, traditional investors focus on standard metrics such as return, risk, tracking error, and other familiar modern portfolio theory statistics; however, ESG investors require all of these metrics plus more.  They seek ways to quantify the impact of their ESG investing; therefore, it’s beneficial to know the basics of how providers are reporting impact.

There is a difference between impact reporting and impact measurement.  Impact reporting refers to a holdings-based calculation using company-level ESG data and scores provided by the major ESG ratings providers.  Some providers calculate ESG ratings based on a numerical score from 0 to 100, while others employ a letter-based ranking scale from AAA to CCC.  Those using the numerical method calculate the overall portfolio score using the below formula and report the weighted average portfolio and benchmark ESG scores.  The improvement in the portfolio’s ESG score compared with the benchmark’s constitutes the percentage improvement, or the “ESG save.”

The letter-based ratings approach reports ESG improvement in a binary fashion, whereby stocks are bucketed into two categories: “AAA to A” or “BBB to CCC.”  Accordingly, the percentage of stocks in the top bucket (e.g., 60% of the portfolio) is then compared to that of the benchmark index (e.g., 38% of the benchmark), which is then stated as a percent improvement in the ESG score (22%, for our example).  However, one should note that whether they use the numerical or letter-based rating system, providers do not generally disclose the distribution ranges or statistics of their scores, which would be useful and serve as a helpful reference.

The methods described above are considered impact reporting, which is meaningfully different from impact measurement.  Impact reporting gauges the sustainability practices of a company and represents an operational perspective, while impact measurement is an outward assessment aiming to quantify the impact of a company beyond the realm of the environment to include its broader impact on society.

Initiatives such as Principles for Positive Impact Finance, launched earlier this year in Paris, are working to incorporate impact measurement in alignment with the 17 Sustainable Development Goals (“SDGs”).  The endeavor itself is honorable but presents a host of considerable challenges.  Global companies have a multitude of businesses and can find themselves making numerous products—some that are deemed positive and others deemed negative.  An attempt to arrive at an overall score that quantifies such an impact can be problematic; perhaps, at least in the early stages, the emphasis should be on working out whether an investment has a broadly positive or negative impact, rather than trying to calculate a number.

Impact reporting is a step in the right direction, and as investors become more familiar with gauging impact, the momentum to assess, evaluate, and measure impact will only continue to grow.

In an earlier blog post, we provided a brief survey of recent monetary policy cycles in the U.S., showing that a higher Fed funds rate doesn’t necessarily affect the yield on Treasury bonds in the same way.  Policy rate changes affects short-term bond yields much more directly than longer-term yields (see Exhibit 1).  We argued that the difference in impact is likely a result of other macroeconomic factors that affect longer-term rates and segmentation in the market.  In this follow up note, we focus our attention on the shape of the yield curve and returns over various tightening cycles.

But what does curve flattening mean for returns?  As a bond’s yields increase, the price of the bond drops.  Consider the yield curve just before the beginning of the 2004 tightening cycle versus the curve shape at the end of that cycle (see Exhibit 2).  It seems that the price of a 20-year bond should be little changed, and longer-term bonds generally outperform shorter term bonds—at least in terms of yield.

Over this period, the S&P U.S. Treasury Bond 10+ Year Index did indeed outperform, providing a total return of 9.7% versus the S&P U.S. Treasury Bond 1-5 Year Index, which returned 3.5%.  Total return, however, is comprised of both price change and interest income.  The 10+ year index provided 12.5% of interest income over the period, cushioning the 2.5% price decline.  The short-term index, on the other hand, provided 7.9% of interest income, offset partially by 4.1% price decline.

So, longer duration bonds are a good thing when rates are rising?  There are many underlying market conditions and cycle characteristics that can affect returns.  Factors including the starting level and shape of the curve, as well the speed and magnitude of the policy changes, need to be considered.  Concerns about a brewing housing bubble had driven the front end of the Treasury curve 100 bps higher (June 2003 to June 2004) and the curve remained steep.  A higher starting point meant more interest income to cushion against price depreciation.  The steepness of the curve increases the chance of a more meaningful flattening.  The protracted cycle (consecutive 17 meetings) was gradual (25 bps/meeting), which meant less abrupt changes in price.

With the exception of the one-off surprise hike in 1997, there have been four tightening cycles, each slightly different in starting points and cycle characteristics.  The Treasury curve flattened in all cases, however in the 1994 and 1999 cycles, the outset of the cycle caught markets by surprise; adjustment was much quicker and the curve didn’t flatten sufficiently, with longer-term bonds underperforming in price terms.  Exhibit 3 compares returns of the S&P Sector Indices over the policy cycle periods.

The pace of the current hiking cycle and the modest flattening of the yield curve over the past two years have been positive for long duration bonds.  Market participants continue to expect two additional hikes in 2017.  Technical drivers may become more relevant for medium- and long-term yields as the Federal Reserve begins plans to reduce its balance sheet (unwinding QE).

In the inaugural publication of the Journal of Portfolio Management in 1974, Nobel Laureate Paul Samuelson wrote that there is no “brute fact” that “there could exist a subset of decision makers in the market capable of doing better than the averages on a repeatable, sustainable basis.”[1]  That article partly inspired John Bogle to launch the first index mutual fund.[2]  Even though over four decades have passed since Paul Samuelson penned the article, his words remain relevant to today’s world of investing.

This year marks the 15^th anniversary of S&P Dow Jones Indices publishing the SPIVA^® U.S. Scorecard.  A lot has changed in the asset management industry since we started reporting on the active versus passive debate.  The proliferation of index-linked investment products and the low-cost, efficient ways they can provide exposure to desired asset classes and markets have been disruptive, adding a tremendous amount of pressure on active managers in terms of fees and performance.  At the same time, the line between what has traditionally been considered beta versus alpha has blurred, with the passive side continuing to innovate and replicate strategies that once sat predominantly in the active realm.

Amid that shifting landscape, the SPIVA Scorecard has maintained its objective voice as an independent scorekeeper of the active versus passive debate.  It has demonstrated that over a long-term investment horizon, average active managers across market cap segments and styles have underperformed their respective benchmarks.  Exhibit 1 shows the rolling three-year relative performance of actively managed domestic large-, mid-, and small-cap funds against their respective benchmarks[3].

Moreover, the outperformance that winning active managers have generated has been shown to be fleeting.  Through a number of studies, we have found that managers that outperform their benchmarks in a given year or are in the top quartile of their peer groups are unlikely to repeat their winning streak repeatedly.[4]  For example, we studied over 789 large-cap, 383 mid-cap, and 511 small-cap funds on average on a rolling quarterly basis from March 31, 2003, to Sept. 30, 2016 (see Exhibit 2).  We found that out of the 20%-30% of funds that outperformed their benchmark in a given year, only a small subset were able to repeat that outperformance in the subsequent three years (see Exhibit 2).[5]

Taken together, these findings indicate that market participants may face substantial difficulty in identifying a winning manager in advance.  They also pose the challenging question of how one should go about measuring the success of active management.  If net-of-fees returns (and, in many equity markets, gross-of-fees returns)[6] are not high enough to overcome benchmark returns across all market cycles and the outperformance produced is fleeting, what should constitute the evaluation framework for active management?

With that, the active versus passive debate turns a new chapter.  Given that fees contribute partly to managers’ underperformance,[7] it seems clear that closet indexing at high costs cannot be sustainable.  For active management to add value and separate themselves from passive strategies, focusing on differentiated portfolio construction or strategies in which managers take compensated bets, along with outcome-oriented investment solutions, can serve as ways to promote their skills.

Only time will tell whether these differentiated strategies can deliver higher risk-adjusted returns than their passive counterparts.  It seems clear that the active versus passive debate is evolving and heading in a new direction in which active management could become more active.

[1]   Paul Samuelson, “Challenge to Judgement”. The Journal of Portfolio Management 1974.

[2]   John C Bogle, “Lightning Strikes: the Creation of Vanguard, the First Index Mutual Fund, and the Revolution It Spawned”. The Journal of Portfolio Management. Special 40^th Anniversary Issue

[3] See the SPIVA U.S Year-End 2016 Scorecard

[4]   See the Persistence Scorecard.

[5]   See Fleeting Alpha: Evidence from the SPIVA and Persistence Scorecards.

[6]   See Institutional SPIVA Scorecard – How Much Do Fees Affect the Active Versus Passive Debate?

[7]   See Institutional SPIVA Scorecard – How Much Do Fees Affect the Active Versus Passive Debate?

Returns are definitely one of the key influencers of investment decisions.  However, poor governance can result in failures, as too much emphasis on returns may result in ignoring the need to understand and manage the potential risk.  It is important to understand the risks associated with an investment strategy.  Risk can have different interpretations depending on the situation; it could be seen as an opportunity for one and an avoidable situation for another.  Many portfolio strategies target the dual objective of growth and prevention of loss of capital.  Hence, it is imperative to adopt superior risk management processes.

Assessing the risks associated with portfolio strategies or investment decisions has multiple benefits. Risk can be quantitative or qualitative in nature, and assessing the different types can help manage volatility or limit downside potential.  A well-defined risk management approach is beneficial during market turbulence, when impulsive decisions can lead to losses and overlooking the scope of long-term wealth restoration or creation.

There are various tools that are used for risk management, such as diversification, balanced asset allocation, and monitoring of the goals and the portfolio.  Diversification is not considered a guarantee or a perfect solution, but rather it is a tool to manage risk and reward by choosing a mix of assets to help spread the risk, compared with a concentrated strategy that can be overweight in certain asset classes or sectors.

Passive, or index-based, investing can be a great way to diversify a portfolio or strategy.  Indices can be used to represent an asset class in the asset allocation process, and historical index data helps in probability analyses.  To demonstrate diversification via sectors, we can view the various sectors of the S&P BSE Indices that display trends over the past 10 years.  The S&P BSE SENSEX, a well-diversified market benchmark, is designed to measure 30 stocks that represent all 10 key economic sectors in India.[1]  This is where we can observe a diversified strategy.  There are sectors that have outperformed the diversified S&P BSE SENSEX, but some sectors have underperformed.  A historical analysis helps to identify the underperforming sectors, but when portfolio management strategies are constructed, we cannot know what certain sectors, regions, or strategies may outperform in the future.  Therefore, diversification can be a useful tool to help balance a strategy.

Strategies that bear investment and longevity risk because they hold funds for a longer amount of time may need to adopt sound risk management practices in order to obtain better protection against potential risk.

Exhibit 1: S&P BSE Sector Indices

Source: Asia Index Private Limited.  Data from March 2007 to March 2017.  Past performance is no guarantee of future results.  Chart is provided for illustrative purposes.

Exhibit 2: S&P BSE Sector Indices 

Source: Asia Index Private Limited.  Data as on 31 March 2017.  Past performance is no guarantee of future results.  Table is provided for illustrative purposes.

[1]   Note: The S&P BSE Indices sectors are based on the BSE sector classification.

It is undeniable that an individual investor would need considerable skill (or luck) to navigate optimally between the various single-factor equity strategies.  If the goal is to outperform the benchmark, then simply choosing between a quality, value, momentum, or low volatility strategy may present the biggest risk.  While they all have been shown to hold a unique systematic risk premium that may provide returns over the long term, how can the shorter-term drawdowns of each be avoided?  How could an individual investor expect to have the foresight to know which single factor will outperform in the coming years—especially since they all tend to perform differently depending on the market environment and economic cycle?

One solution to this conundrum could be in choosing not to choose.  Diversification is often described as the only free lunch in finance, so why not grab a plate and help yourself to a multi-factor buffet?  The results are compelling.

Because the active returns of single-factor strategies often tend to have negative or low correlations, a simple equal-weight allocation to the four key single-factor strategies of the S&P 500^® can lead to significantly higher probabilities of outperformance.  Using a simple multi-factor “index of indices” approach, we can show that the frequency of risk-adjusted outperformance can be just as good, if not better, than picking the best-performing, single-factor strategy.

If this simple, yet effective, multi-factor index of indices approach is not satisfactory, an investor could decide to go one step further; choosing to combine multiple factor scores at the stock level.

This approach selects the top quintile of the S&P 500 based on the average of its multi-factor scores, targeting stocks that have the highest combination across the desired factors.  In practice, some compromises may be necessary; stocks with high exposures to all factors are rare.  However, the stock-level approach seeks to select the stocks with highest attainable factor combinations in an attempt to improve the portfolio’s overall factor exposures.

The new S&P 500 Quality, Value & Momentum Multi-Factor Index is an example of this type of stock-level approach.  Exhibit 3 shows the average risk/return figures over 15-year rolling windows for both multi-factor approaches and their corresponding single-factor strategies.  Take a look at the top-left of Exhibit 3, where the highest returns and lowest risk are represented; you may stop to re-think putting all your eggs in just one single-factor basket.

For a more detailed overview on S&P DJI’s approach to multi-factor indices, please see “The Merits and Methods of Multi-Factor Investing.”