Outperforming the S&P 500 by picking individual stocks is a challenging exercise; however, many observers comment that certain sectors outperform the index at different times. This article will look at how the different sectors are correlated and how the have performed over the last couple years. Are there some sectors more or less correlated to the S&P 500? How did they perform during the recent downturn? What happened to their correlations - the hypothesis would be that everything became increasingly correlated during the downturn. The following sectors will be compared to the S&P 500 as tracked by the SPDR S&P 500 Trust ETF (SPY):
|SPY||SPDR S&P 500 Trust ETF||95.3||1.7%||14|
|XLY||Consumer Discretionary Select Sector SPDR||2.3||1.4%||16|
|XLP||Consumer Staples Select Sector SPDR||4.0||2.6%||15|
|XLE||Energy Select Sector SPDR||10.5||1.3%||13|
|XLF||Financial Select Sector SPDR||8.0||1.0%||13|
|XLV||Health Care Select Sector SPDR||3.4||1.7%||15|
|XLI||Industrial Select Sector SPDR||4.5||1.7%||16|
|XLB||Materials Select Sector SPDR||3.0||2.9%||14|
|XLK||Technology Select Sector SPDR||7.6||1.3%||15|
|XLU||Utilities Select Sector SPDR||4.3||3.9%||13|
Source: Yahoo!Finance downloaded on May 15, 2011.
The initial hypothesis would be that defensive sectors including consumer staples, health care, and utilities would show the lowest correlations to SPY and have good returns during down markets. The first analysis will look at correlations over different time periods - 10 years, 5 years, and November 2007 to April 2009 during the downturn and March 2009 to present during the upturn.
Defensive Sectors Show Lowest Correlations as Expected
As expected, the defensive sectors consistently showed the lowest correlations to SPY across all time frames. Furthermore, these sectors became less correlated to SPY during the downturn when compared to their 5 year correlation while the more exposed sectors like consumer discretionary and industrials became more correlated.
|Ticker||Sector||10 Year||5 Year||Downturn||Upturn|
Source: Monthly closing split and dividend adjusted prices from Yahoo!Finance. Correlations based on monthly returns, not prices.
The real insight here is that the correlations do not really cover large ranges. The correlations for consumer staples is usually within 5-10% points of consumer discretionary which means doing a wholesale sell of discretionary and purchasing staples to become more defensive is probably not really a great play. To capture the benefit of low correlations, an investor would benefit from over weighted XLU and XLE.
Actual Sector Returns show Unexpected Results
Similarly one would expect returns in the defensive sectors to be relatively higher during the downturns. Consumer staples should also outperform consumer discretionary. Benefiting from hindsight, Financials will show the biggest decline and also the strongest recovery. Materials and Energy should also show strong recoveries. This two latter observations would probably not generalize going forward.
|Ticker||Sector||10 Year||5 Year||Downturn||Upturn||Since Nov 2007|
Source: Yahoo!Finance. Note that the Since Nov 2007 and Downturn and Upturn returns will not mathematically align since I have two months of overlap between the downturn period and the upturn period.
The first observation is that as expected Financials were hit the hardest and had one of the strongest recoveries. Industrials also followed this pattern. However, this trend is not consistent since Utilities and Energy also had large downturns, but much weaker recoveries. Consumer Staples also generally followed the trend with the best performance in the downturn followed by a weaker recovery. To me, the interesting observation is that the top performing sectors from November 2007 when many people thought things were great have been Consumer Discretionary and Consumer Staples. Excluding Financials, the drags have been Utilities, Materials, and Industrials. These returns seem to point to the notion that the U.S. is still facing a manufacturing deficit relative to the world and that it has been greatly exacerbated by the downturn. I've noted earlier here and here that the downturn swept across all indices across the world. I reordered the above data to focus on 10 year returns and returns since Nov 2007.
|Ticker||Sector||10 Year||Since Nov 2007|
Global growth has driven commodity prices much higher over the past 10 years and Energy and Material sectors have benefited. The two consumer sectors continue to show strong returns over 10 years and recently. However, Utilities and Industrials while posting above market returns for 10 years are showing weaker returns since November 2007.
The clear issue with this analysis is that it looks at broad sectors. It could easily be argued that more focused ETFs could provide a broader range of correlations allowing more decisive defensive plays. Furthermore, picking individual stocks would also very easily. As often noted, McDonald's Corporation (MCD) and Wal-mart (WMT) had strong performances during the downturn.
The second issue that became apparent with some of the returns review would be to further subdivide the time periods. In a repeat analysis, I would also exclude any overlap between the downturn period and upturn period. The overlap had been used since the analysis was focused on correlations.
So what should an investor do with this information?
- Based on this historical analysis, it seems the notion of broadly switching between Consumer Discretionary and Consumer Staples is more of a market timing activity. If you get it right there will be big payoffs, but staying put performed equally well. Looking forward, I don't think this is good play at the sector level if you anticipate a downturn.
- Based on correlations, Utilities look like a good defensive play. However, their recent performance of -9.6% (which includes dividends) is quite disappointing. Looking forward, Health Care, which also had low correlations, could be a better standing defensive play.
- Income investors with substantial long positions in Utilities might also want to investigate Materials and Consumer Staples. While both have lower dividends and higher correlations to SPY, they should be more resilient to declining U.S. manufacturing and Industrial activity while Utilities might not be. The difference in dividends is material but not enormous with Materials offering an average of 2.9% and Consumer Staples at 2.6%. Materials also offer a global economic growth play while Utilities do not.
- Specific stock selections within each sector may provide added benefits or more focused exposures. In any of the above conclusions, specific stocks may prove this general sector level thoughts wrong.
- I think this also highlights the importance of diversification. An investor in Financials would have enormous losses. Today, an investor who has been in Materials and Energy has done quite well but will it continue? Technology has been out of favor for 10 years but has the 4th best performance since November 2007.
- Extreme market conditions also raise the notion of behavioral finance which has gained substantial traction and defies the notion of efficient markets. To me, behavioral finance is about fear and greed while longer term investing success is about patience, discipline and hard work. While this obviously is with hindsight, all but two sectors have now posted positive returns since November 2007. If you are gripped with terror about stocks, the reasonable sell point has probably long since passed. These are the moments that try my own investing resolve. My personal experience is that the deeper the fear the more likely I should buy and the greater my euphoria the more likely I should sell. Part of my thought process is to avoid thinking about losses and gains on specific stocks and securities but more around the exposures I want to have. Should I be increasing or decreasing my exposure?
Disclaimer: This article is for informational and educational purposes only and shall not be construed to constitute investment advice. Nothing contained herein shall constitute a solicitation, recommendation or endorsement to buy or sell any security.
Disclosure: I am long SPY.