Indexology^® Blog

Many key factors are currently driving oil down and the bloodshed might not be over. As mentioned in a recent CNBC interview, from current levels, the S&P GSCI Crude Oil Total Return could drop another 45% before surpassing energy’s worst historical loss of 78.4% that happened in 2008-2009.

The headwinds of the strong dollar from the Greek crisis, the seasonal slowing of gasoline demand post July 4th, increase in rig counts despite high inventory levels, possible lifting of Iranian sanctions that could ramp up production, and slowing Chinese crude demand plus their crashing stock market are all bad news for oil.

It is not only bad news for oil but the weak Chinese economy and stock market is a huge influence for copper given China consumes about half of the world’s copper. Since early 2011, copper has been dying a slow death but the loss of 3.5% on July 6, 2015  was the biggest one day drop in six months, since January 14, 2015.  Also, the weaker dollar from the Greek crisis is impacting all economically sensitive commodities – and copper is known to be one of the most sensitive to growth. Like oil, copper has about 50% left to fall before hitting new index lows.

However, Dr. Copper is not always so smart. Historical correlations only run about 0.4 with the Chinese GDP growth and copper actually performs well in weak recessions. It is only in strong recessions where copper tanks. Maybe this is the signal for the start of more than just a weak recession.

According to Jim Cramer, these things need to happen for another global financial crisis: 1. The Greek deal needs to be delayed further, causing the euro to go lower and the dollar to get stronger, which ultimately hurts U.S. exports (and commodities – but not all); 2. China needs to collapse from the weight of its stock market and trillions in bad loans; and 3. The Fed needs  to continue discussing raising rates, even though employment isn’t as strong as most thought and industrial production is still not great.

The problem is the worst is potentially just around the corner, coming in Q4, according to commodity seasonality.  Despite all the headwinds, the summer generally helps many of the commodities used for travel, construction and even grilling, which kept commodities flat through Q2. Now that the peak driving season has passed, a major catalyst is out of gas, and mid-July is precisely when the worst drawdown in history that lasted from July 14, 2008 to February 18, 2009 started. During that period, the majority of days (16 of 23) with the biggest losses (>4.0%) were in Q4 of 2008.

An investor might be wondering now if it is even possible to hold a basket of commodities without getting slaughtered. Maybe an active manager could pick out the winners, but probably not, given the results from last year’s conference poll where sugar (-16.9%), copper (-12.2%) and nickel (-28.2%) were cited as the best picks for the year going forward. All of those are performing worse than the composite benchmark, the S&P GSCI, that is down 7.3% YTD.

Maybe the smart answer is to simply use an equally weighted index like the Dow Jones Commodity Index (DJCI) rather than the world-production weighted S&P GSCI. To answer that, it must be possible for agriculture and precious metals to overpower the economically sensitive energy and industrial metals. After all, agriculture and livestock have historically very low correlations to the dollar – that is to be expected given people need to eat and these commodities are much more sensitive to weather. Even better now, is the extreme weather from El Nino may be the biggest driving force for food commodities, and on average, the positive impact on agriculture has accelerated with every El Nino since 1982. In the midst of all the current turmoil, wheat and sugar were positive on July 6, 2015, and 5 of 8 commodities in the sector remain positive for the year. Unfortunately, the bad news is that the weighting alone is not powerful enough to carry a commodity basket. Year-to-date through July 8, 2015, the DJCI, S&P GSCI Light Energy, S&P GSCI Risk Weight and S&P GSCI Roll Weight Select have lost 6.4%, 7.1%, 6.3% and 7.7%, respectively.

Another set of indices called “modified roll” addresses contract selection to help commodity performance during down markets. Historically it has been shown there may be a premium for giving up liquidity at the front end of the curve for later dated contracts, but it depended on the time period. Unfortunately, during this time period rolling is not working, and that is because there has been backwardation in commodities this year that has only recently flipped to contango. Year-to-date through July 8, 2015, the S&P GSCI Dynamic Roll, S&P GSCI Enhanced Commodity, S&P GSCI Multiple Contract and S&P GSCI 3 Month Forward have lost 8.7%, 7.0%, 7.5% and 8.2%, respectively.

Again, now is a difficult time for commodities that might get even harder.  Is it possible to get smart enough to hold a basket that does not require picking single winners? Unfortunately, no long-only index has proven smart enough to generate a positive return during the worst down days of the past 10 years. That includes all days that dropped more than 4.0% for a total of 35 days. Not even our smartest long-only index, Dow Jones – RAFI Commodity Index, that combines smart rolling and weighting has proven successful (it is down 7.6% YTD.)

However, there is a set of broad-basket commodity indices that is smart enough to generate positive returns in the worst of times. These indices are a subset of a long/short family called “strategic futures” and they are strategic for a reason. The broader strategic futures include financial futures as well as commodity futures, but only the subset using purely commodity futures is examined for this analysis. Three indices, namely, the S&P GSCI Dynamic Roll Alpha Light Energy, the S&P Dynamic Commodity Futures Index (S&P DCFI) and the S&P Systematic Global Macro Commodities Index can be considered smart enough to perform in the worst commodity periods.

In the worst drawdown of the S&P GSCI that lost 70.9% from July 2008 – Feb 2009, the S&P GSCI Dynamic Roll Alpha Light Energy and the S&P Dynamic Commodity Futures Index (S&P DCFI) gained 10.7% and 14.1%, respectively, but despite out-performance over the S&P GSCI the S&P Systematic Global Macro Commodities Index still lost, but only 3.4%.

Extending this analysis by examining the 10-year time period into the best 35 days and worst 35 days gives more insight into the time when protection is needed most.  On average, the worst 35 days classified by all days with a greater than 4.0% loss in the S&P GSCI, generated a return of -5.3%. The modifications of long only indices helped but not enough to get positive results. In spite of the worst historical time, the long/short strategic futures indices were smart enough to be positive with average daily returns of between 28 and 129 basis points with some significant daily returns up almost 8.0%. Source: S&P Dow Jones Indices. Ten years of daily data ending July 7, 2015.

On the flip side, when the S&P GSCI posted its 35 biggest daily gains, all more than 3.5%, the strategic futures were basically flat. This combination of protection on the downside with flat performance in commodity bull markets preserves capital. Over long periods of time, smart indices that roll and/or weight differently have also performed well with reduced losses rather than positive returns in down periods. However, for real risk management and the chance for positive returns during the worst of times, the strategic futures have done the job best with all of the independence, transparency, liquidity and cost efficiency indices provide.

The companies in the S&P 500 are often described as leading companies in leading industries.  The S&P 500 is the go-to index for analyzing, tracking and understanding large cap stocks in the US market.  Its history goes back over 50 years to 1957 and its predecessor indices extend back to 1923 and 1926. Only the Dow claims a longer heritage. With about 50% of the revenues of the S&P 500 companies coming from overseas, the index is also a leading benchmark for developed markets worldwide.  Until now one thing has been missing – many of these 500 companies raise capital in the fixed income markets as well as in the equity markets.

S&P Dow Jones Indices is introducing a bond index covering the 500 companies in the S&P 500 (equity) index.  The index will include all publicly traded bonds issued in the US by companies in the S&P 500.  This is about 4,760 issues covering some $3.7 trillion of debt. The combination of the stock and bond indices makes possible consistent analyses of the two most important asset classes for investors. By anchoring both indices in the same list of leading companies the two indices will provide a more complete picture of the financial condition of large cap corporations in the US.  The distribution of debt and equity among companies and sectors differs and a comparison of the indices highlights these differences. On the bond side the largest sector measured by the market value of bonds is financials, followed by consumer discretionary.   Financials are also high on the equity list at number two by market value. However, the largest in equities, information technology, is the seventh by size in bonds.  Health care is the third largest sector in both the S&P 500 and the S&P 500 Bond index.  Telecommunication services is the smallest equity sector and the second smallest bond sector.  The chart compares the sector shares in the two indices.

In fixed income analyses of bonds, a primary division is between financial and nonfinancial corporations. Using this division and data from the Federal Reserve, bonds issued by nonfinancial companies in the S&P 500 represent about 59% of nonfinancial corporate bonds while bonds issued by financial companies represent about 23% of the financial bond market. On the equity side, the S&P 500 represents about 80% of the total market value of equity of US corporations.

There are many ways to divide or classify the issues in the S&P 500 Bond index that can’t be done with equities. These can add to our understanding of both bonds and stocks.  The bond ratings and the distribution of the ratings gives a picture of the overall credit quality of the S&P 500.  About 94% of the market value of the debt is rated investment grade, a good report card for this group of companies. Possibly less inspiring is the percentage of market value that receives the highest rating of AAA: 1.4%.   The maturities range from one month to almost 96 years.  By market value 41% is less than five years, 31% is between five and ten years and the balance are over ten years.  There are nine issues with maturities longer than 80 years including one at almost 96 years. These long lived issues include a bank, a railroad, a health care company and some industrials.

The debate on the merits of active versus passive fund management can be a contentious topic and most recently was so for S&P DJI’s most recent SPIVA India Scorecard.  To that end, it might be helpful to review some of the provisions of the CFA Institute’s 2010 edition of the Global Investment Performance Standards (GIPS) regarding calculation methodology and composite construction, including the following.

• Calculation methodology
  • A.1: Total returns must be used.
  • A.6: Composite returns must be calculated by asset-weighting the individual portfolio returns using beginning-of-period values or a method that reflects both beginning-of-period values and external cash flows.
• Composite construction
  • A.6: Terminated portfolios must be included in the historical performance of the composite up to the last full measurement period that each portfolio was under management.

For the India SPIVA report, we use total returns for comparison purposes, since these include income from dividends, and we include merged or liquidated funds. We choose the benchmark indices to be representative of the segment in which each fund invests. We also take the oldest share class per fund to avoid double counting.

To compute survivorship, we calculate the number of funds that were either merged or liquidated during the evaluation period. For example, for the five-year period ending December 2014, we found that the survivorship rate was 80.30% for the Indian large-cap actively managed mutual funds. What this means is that out of the total available funds, which were 102 at the beginning of the period, 20 funds were either merged or liquidated during the five-year period.

To compute outperformance, we compare each individual fund’s returns with those of the relevant benchmark, after removing those funds for which data is not available. We use the total returns of the benchmark for comparison purposes. For example, for the five-year period ending December 2014, we found that 52.94% of the Indian Equity Large-Cap actively managed mutual funds could not keep up with the S&P BSE 100. What this means is that out of the total available funds, which were 102 at the beginning of the period, 54 funds underperformed the S&P BSE 100. The returns of each individual fund were compared with those of the S&P BSE 100 to arrive at this statistic. This has nothing to do with equal-weighted or asset-weighted returns. These 54 funds also include the 20 funds that were either merged or liquidated because excluding them would inflate the perception of success.

In addition, we calculate equal-weighted fund returns based on the monthly returns of each fund. We did not include the asset-weighted returns, given there is limited data available to calculate asset-weighted returns since only the quarterly average of the assets under management is available for evaluation from the Association of Mutual Funds of India.

Nevertheless, let’s look at asset-weighted returns based on the quarterly average assets under management (see Exhibit 1). In this case, the share class with the highest quarterly average assets under management at the beginning of the period was identified for each fund and then the asset-weighted returns were calculated.

Source: S&P Dow Jones Indices LLC, Morningstar, Association of Mutual Funds of India.  Data as of Dec. 31, 2014.  All returns in INR.  Past performance is no guarantee of future results.  Table is provided for illustrative purposes. 

We can observe that the five-year asset-weighted returns for the actively managed Indian Government Bond funds and the Indian Composite Bond funds are less than their respective benchmarks, the S&P BSE India Government Bond Index and the S&P BSE India Bond Index. In the case of the actively managed equity mutual funds, all the fund categories have higher five-year asset-weighted returns than their respective benchmarks. The equal-weighted returns presented in the India SPIVA Year-End 2014 report told the same story, except for the Indian Equity Large-Cap funds.

We can observe that the asset-weighted returns for the actively managed Indian Equity Large-Cap funds were almost 1.56% more than S&P BSE 100, indicating that the larger funds led the rally. Let’s break down the statistics of this fund category into quartiles by the asset size (Exhibit 2). The calculation used the share class with the highest quarterly average assets under management at the beginning of the five-year period for each fund for which the average asset under management was available.

Source: S&P Dow Jones Indices LLC, Morningstar, Association of Mutual Funds of India.  Data as of Dec. 31, 2014.  All returns in INR.  Past performance is no guarantee of future results.  Table is provided for illustrative purposes. 

We can notice that, even in the first and second quartile, which consists of the largest funds by asset size, the S&P BSE 100 outperformed 40% and 32% of the funds over the five-year period ending December 2014. Over the same period, within the third and fourth quartile, the majority of the funds were outperformed by their respective benchmark. The equal-weighted asset returns of the first quartile were less than the second quartile by 44 bps whereas the asset-weighted returns of the first quartile were more than the second quartile by only 12 bps over the same five-year period. Even the survivorship percentage of the second quartile was better than the first quartile.

Therefore, it is evident that not all the funds outperformed the benchmark on the basis of having the largest asset size

Each year about 20 to 25 stocks leave the S&P 500 and are replaced with other stocks that meet the guidelines for index membership.  These changes are usually announced five trading days before the index change. Given the size of the asset pool that tracks the S&P 500 — ETFs, index funds and institutional portfolios – about 11% of the float shares of each stock in the index is in index tracking portfolios.

The idea that 11% of a stock’s shares may be bought over five days suggests to some people that buying will surge and push the price up.  The reaction of stock prices to index adjustments has been studied several times dating back to the mid-1980s.  The extent of any price increase varied in the last 30 years and has declined in the past ten years.  Recently the effect is a few percentage points, or less. Moreover, any bump up is temporary. A few weeks after a stock joins the S&P 500 there is no last effect and its price, adjusted for overall market movements, is back to normal.

Still, the idea that one could profit from buying a stock as soon as possible after an addition is announced and then selling it when it enters the index is tempting.  To see if this could work, I looked at the 13 adds to the S&P 500 in the first half of 2015. There are 13 stocks added to the index.  Since the announcements are made at 5:15 PM New York time, the strategy is buying at the open on the next trading day after the announcement and selling at the close on the day the stock enters the index.  The closing price is the price S&P DJI uses to adjust the index divisor – if an ETF trades at that price they avoid any tracking error. How did it work? Of the 13 additions, six would have lost money and seven would have made money. The odds of coming out ahead aren’t very good. If you bought and sold one share of each addition you would have ended up with a profit of $8.81. However, you would have needed capital of about $540 because four adds had overlapping dates.  The return on your capital would have been 1.7%.  Doing this analysis completely would require a detailed event study covering much more than six months and adjusting stock prices to isolate the effect of index fund buying from overall market movements. Such studies have been done and they suggest, as did the example, that there isn’t any free money lying around.

The test case just discussed assumes that an index fund portfolio manager only buys at the closing price on the implementation date.  A portfolio manager running an S&P 500 fund is not constrained to buy only at the close on the day the stock is added to the index. He or she can buy at any time; can spread their purchases over a few days or weeks if they believe they will get a better price.  The flexibility gives the portfolio manager an opportunity to time the buying to his advantage and use his understanding of the index.  Some hedge fund managers and speculators do buy when an index add is announced and expect to sell to an index fund a few days later.  Some may also forecast (guess) what stock will be added next and buy in advance.  All this activity increases the liquidity of the stock being added and can dampen any price movements.

To conclude, and as the old saying goes, there ain’t no free money to be found in the stock market. And to believe that index fund managers are at the mercy of hedge funds or alleged front runners, well that’s just a tall tale.

Greece has voted “no” to a referendum on further austerity measures proposed by the ECB. Greek PM Tsipras wanted this result, in order to gain more bargaining power with creditors. The ECB reacted accordingly, and placed steeper haircuts on collateral for emergency funding.   This puts additional pressure on a banking system that is already in deep distress. Today, finance ministers from across Europe are meeting to discuss next steps.

On Monday, European government bond markets reacted as expected, with a flight to quality and a sell-off of riskier assets. The reaction was more subdued outside of Greece, than one would expect, as a Grexit seems more probable. The S&P German Sovereign Bond Index rallied, with yields tightening 3bps, while the S&P Portugal Sovereign Bond Index sold off, with yields widening 10bps. The S&P Spain Sovereign Bond Index widened 10bps, and the S&P Italy Sovereign Bond Index widened 8bps. The Greek Sovereign Bond Index continues its descent, with yields widening 452bps to close at 20.74%.

The performance of these markets indicates that systemic contagion is not a huge concern. What is a concern, however, is the precedent that will be set for other Eurozone countries with struggling economies. If Greek’s hardball tactics end up giving them a degree of debt relief, and/or looser austerity measures, there could be incentive for Portugal, Spain and Italy to change their compliance with austerity measures. This has serious implications for the euro’s stance as a reserve currency.

A growing concern in Europe, is a possible shift in power to anti-austerity political parties outside of Greece. Portuguese Deputy PM Paulo Portas felt the need to comment Monday, saying that his government is committed to austerity plans and intends to stay on track. Portas was responding to Portugal’s socialist leader Antonio Costa, who is calling for Portugal to show more solidarity with Greece. Spain has a very outspoken, rising anti-austerity party called the Podemos. The Podemos, the second largest political party in Spain, is gaining momentum as it appeals to a society that has engaged in years of unpopular austerity measures. Spanish elections are set for late fall, and the ECB’s actions over the next few days will surely be scrutinized.