Get Indexology® Blog updates via email.

In This List

Factor Based Indices in India

Oil Gains This Big Only Happen Around Bottoms

Pure Style Eliminates the Muddle in the Middle

Recession Rumors

Details of the Two-Factor Model

Factor Based Indices in India

Contributor Image
Utkarsh Agrawal

Associate Director, Global Research & Design

S&P Dow Jones Indices

The recent turmoil in the Chinese market has taken the global market on a turbulent ride since the beginning of 2016. India was no exception. During the first two weeks of 2016, the S&P BSE SENSEX lost almost 6.37%, making some investors jittery. Foreign investors took out almost INR 34.84 billion from the Indian stock market, while mutual funds added close to INR 28.18 over the same time period. While the broader stock market suffered a huge draw down through January 15, 2016, let’s see how a new series of factor based indices that were recently launched by Asia Index Private Limited fared over the same time frame based on back-tested data.

A high-level summary of the characteristics of these indices is as follows.

  • All four of the factor based indices include 30 stocks from the S&P BSE LargeMidCap. The constituents differ depending on the rules of the index.
  • The S&P BSE LargeMidCap is a size sub-index of S&P BSE AllCap. Its methodology is designed to cover approximately 85% of the total market cap of the S&P BSE AllCap and is a float-market-weighted index.
  • The S&P BSE Enhanced Value Index consists of “best value” stocks according to a proprietary methodology. “Value” is measured using the price-to-book, price-to-earnings, and price-to-sales ratios of companies.
  • The S&P BSE Low Volatility Index consists of the least-volatile stocks measured over a one-year period.
  • The S&P BSE Momentum Index consists of the stocks within the S&P BSE LargeMidCap with the highest momentum measured over a one-year period. Momentum is measured as risk-adjusted price performance over time.
  • The S&P BSE Quality Index consists of the “highest quality” stocks according to a proprietary methodology. Quality is measured using the return-on-equity, financial leverage, and accrual ratios of companies.

Over the past two years ending Jan. 15, 2016, the S&P BSE Quality Index had a cumulative total return of 55.46%, which was the highest among the four factor indices (see Exhibit 1). It was followed by the S&P BSE Low Volatility Index, which had a cumulative total return of 52.58%. Over the same period, the S&P BSE Enhanced Value Index performed lower than the S&P BSE LargeMidCap. While the S&P BSE Quality Index and the S&P BSE Low Volatility Index displayed lower volatility in comparison with the S&P BSE LargeMidCap (see Exhibit 2), the S&P BSE Enhanced Value Index was 80% more volatile than the S&P BSE LargeMidCap over the same two-year period.

Over the past one-year period ending Jan. 15, 2016, the S&P BSE Low Volatility Index, the S&P BSE Momentum Index, and the S&P BSE Quality Index were in the black, even though the S&P BSE LargeMidCap ended in the red (see Exhibit 1). The S&P BSE Low Volatility Index and the S&P BSE Quality Index even displayed lower volatility than the S&P BSE LargeMidCap (see Exhibit 2). Over the same period, the S&P BSE Enhanced Value Index lost 24%, which was more than two-times the decline suffered by the S&P BSE LargeMidCap (see Exhibit 1).

During the six-month period ending Jan. 15, 2016, all of the indices delivered negative performance (see Exhibit 1). The S&P BSE Enhanced Value Index remained the most volatile and suffered the largest decline of the four of the factor based indices (see Exhibit 2).

Overall, the markets have been in the bear mode, and we can notice that factor based indices displayed their own risk/return characteristics.

Ex - Cumlative Returns

 

 

 

 

Ex - Annualized Volatility

 

 

 

 

 

 

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Oil Gains This Big Only Happen Around Bottoms

Contributor Image
Jodie Gunzberg

Former Managing Director, Head of U.S. Equities

S&P Dow Jones Indices

The S&P GSCI (WTI) Crude Oil posted a 3-day gain of 14.4% ending Feb. 17, 2016. This is the biggest 3-day gain in about 6 months for the index, and gains of this magnitude have only happened near oil bottoms.

Source: S&P Dow Jones Indices.
Source: S&P Dow Jones Indices.

After the index hit its lowest since Nov 4, 2003, on Jan. 20, 2016, it capitulated, reaching near that level again, a few days ago on Feb. 11. Then oil spiked on news of the possibility of a production freeze from OPEC and non-OPEC producers, but quickly waned after hopes diminished and the reality of the lack of potential impact hit.  OPEC has the ability to be the swing producer given its large market share, spare capacity, low production costs and capability of acting alone or in a cartel; however, U.S. inventories need to be low for it to matter.

Source: Bloomberg, IEA.
Source: Till, Hilary. Does OPEC Spare Capacity Matter? Modern Trader Magazine. May 16, 2016. Data Sources are Bloomberg and IEA.

The production coordination of U.S. shale producers is difficult since there are too many players that are always aiming to produce as much oil profitably as possible. Today, the American Petroleum Institute  reported an unexpected decline in U.S. inventories that excited the market.

Source: http://www.investing.com/economic-calendar/api-weekly-crude-stock-656
Source: http://www.investing.com/economic-calendar/api-weekly-crude-stock-656. Feb 17, 2016

Without the decline in inventories from the U.S., the potential impact of an OPEC production freeze is far less. This makes the U.S. inventory arguably more important than the production decisions of the swing producer, Saudi Arabia, but as the U.S. inventory drops, the picture may shift the oil price power back towards Saudi Arabia.

 

 

 

 

 

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Pure Style Eliminates the Muddle in the Middle

Contributor Image
Shaun Wurzbach

Managing Director, Head of Commercial Group (North America)

S&P Dow Jones Indices

Financial advisors who implement style investing seek to outperform the blended benchmark whenever they believe that market conditions will favor either a growth or a value approach.  Their view could be influenced by research, fundamental factors, or technicals.  To implement such a view does not require 100% conviction in either growth or value.  In fact, some prefer to “tilt” toward growth or value by maintaining a reduced position in the less-favored style.  As a hypothetical example, advisors have shown me portfolios with a 60% weighting in value and 40% in growth.  Such an example overweights value stocks compared to the blended benchmark.

If we accept that some style investors “tilt” as their implementation technique, then we recognize that these advisors are creating a mixed basket of stocks that have a combination of growth and value characteristics.  In accepting this mix, many advisors overlook that the overlap in styles is more profound than they may expect based on a “technical” term that I call “the muddle in the middle.”

This “muddle in the middle” is a side effect of the way most index providers create style baskets.  Using the S&P 500® as an example, Exhibit 1 shows how the S&P 500 Growth and S&P 500 Value are constructed.

Capture

While this index construction technique has existed for many years and is widely practiced, it is less understood that the stocks in the middle, or blend basket, are assigned to both the growth and value indices.  This “muddle in the middle” currently leads to 172 overlapping stocks in the S&P 500 Style Indices that don’t have strong growth or value characteristics (see Exhibit 2).

Capture

The S&P Pure Style Indices were created to measure concentrated style expressions while eliminating the blend basket.  The S&P Pure Style Indices measure, select, and retain only the strongest growth and value stocks and weight the constituents in accordance with their style score.  The two approaches to basket selection can be compared in Exhibit 3 with the S&P Style Indices on the left and the S&P Pure Style Indices on the right.

Capture

The S&P Pure Style Indices seek to create style baskets without any stock overlap.  The S&P Pure Style Indices eliminate the “muddle in the middle” problem for style investing (see Exhibit 4).

Capture

The ramification for advisors employing a style tilt is that the same 60% value and 40% growth tilt imagined before now results in zero stocks overlapping in the resulting portfolio, instead of 172.  S&P Dow Jones Indices has over 10 years of live index history for the S&P Pure Style Indices (launched on Dec. 16, 2005) that can be shared with advisors in the S&P 500, S&P MidCap 400®, and S&P SmallCap 600® size classifications.

Sam Stovall, Equity Strategist at S&P Global Market Intelligence, plans to cover S&P Pure Style Indices as Smart Beta in one of his “7 Rules of Wall Street Strategies” at our FA Forum in Miami on Feb. 24, 2016.  Advisors may register at no cost to attend in person or by live streaming.

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Recession Rumors

Contributor Image
David Blitzer

Former Managing Director and Chairman of the Index Committee

S&P Dow Jones Indices

One sign that people are worried about a recession is getting questions about one on days the market rises.  There seem to have been more down days than up days since the start of the year and the recession question is being asked more on all days.

The current recovery is weaker than other recent rebounds. The first chart compares the last several recoveries and expansions by indexing each one to 100 at the quarter of the pre-recession peak.  The horizontal axis is the time dimension measured in quarters. The thick black line at the bottom is the current recovery. The low point took a bit longer to reach than some other recession/recovery cycles; the track is below all the other cycles shown on the chart.

To paraphrase Mark Twain, predicting recessions is quite difficult, especially about the future.  One theory popular in the press is that as recoveries get older they get weaker and eventually collapse into the next recession. Though this may be an appealing analogy to nature, the statistics don’t support it.  Research, including a recent note published by the San Francisco Federal Reserve Bank, shows that since the end of the Second World War, there is no link between longevity and the timing of the next recession.  Then there is the stock market which Paul Samuelson noted successfully predicted nine of the last five recessions. Doubters can examine the second chart which shows the S&P 500 and the recessions since 1947.

In fact, most recessions weren’t even recognized until a few months after they began.  At the moment recession fears seem to be overdone – the unemployment rate is falling, weekly initial unemployment claims are far below any danger zone and consumers are busy buying houses and cars.

I will speak more on recessions and the outlook during my keynote at the upcoming CBOE Risk Management Conference.  Register: http://www.cboermcus.com/

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Details of the Two-Factor Model

Contributor Image
Hong Xie

Former Senior Director, Global Research & Design

S&P Dow Jones Indices

After identifying value and low volatility as factors that can effectively explain the return and volatility of an investment-grade corporate bond portfolio, we proposed a two-factor model to capture the security selection process of active corporate bond managers.

The underlying universe for our study is the S&P U.S Issued Investment Grade Corporate Bond Index from June 30, 2006, through Aug. 31, 2015.  To enhance the liquidity profile and tradability of the index portfolio, we first derived an investable subuniverse, applying seasoning and outstanding amount as criteria.  Of that investable subuniverse, the bonds were then divided into groups based upon their effective duration and credit rating.  Within each group, bonds were selected by credit spread and low volatility factors; bonds with Libor OAS wider than the median level of the group were ranked by yield volatility, and only the 20% of those ranked bonds with the lowest volatility were then selected.  The weights of selected bonds were designed in such a way that the weight for each grouping matched that of the underlying base universe.

Utilizing a volatility factor is a key step in the index component selection process.  Without the overlay of this factor one is simply picking the cheapest bonds with the widest OAS and, therefore, most likely piling on credit risk.  Screening first by OAS results in a pool of bonds that have higher potential for spread tightening and better carry.  The subsequent low-volatility screening is designed so that bonds with less risk, as demonstrated by their trading pattern, are selected, while duration and credit rating are held equal.  One can think of our selection process as identifying the cheapest bonds with wide credit spreads that are not justified by their historical trading volatility in their respective duration and credit groups.

Exhibit 1 shows the improved risk-adjusted return of the two-factor portfolios versus the underlying universe.  The Sharpe ratio for the two-factor model improves to 1.15 (monthly rebalancing) or 1.05 (quarterly rebalanced) from 0.87 of the base universe.  The two-factor model also provides better downside protection, demonstrated by a maximum drawdown of 13.28% (monthly rebalancing) or 13.69% (quarterly rebalanced) compared to 14.57% of the base universe.

Capture

The posts on this blog are opinions, not advice. Please read our Disclaimers.