Should You Attempt A Sector Rotation Strategy?

|
Includes: FV, IWV, XLB, XLE, XLF, XLI, XLK, XLP, XLRE, XLU, XLV, XLY, XTL
by: Kurtis Hemmerling
Summary

Sector rotation theory is over 150 years old.

Common sector rotation methods include momentum and volatility filtering.

I test various sector rotation strategies and find most do not add additional benefit beyond holding all sectors equal-weight.

I test a low turnover sector rotation strategy with interesting results.

It sounds attractive - buy the best sectors of the market to juice returns and lower risk. But is beating the market this simple? I want to examine a few approaches to sector rotation and let you be the judge.

Breaking Down the Market into Sectors

As we talk about sectors, you should know that there isn't any one universal standard. There are a variety of classification systems such as the Thomson Reuters Business Classification, Global Industry Classification Standard and Industry Classification Benchmark and more. GICS is my personal preference, and this is classification system which Sector SPDRs are based on.

We currently have 11 sectors as Real Estate has left the home of Financials. At the earliest point in my testing, there will be 9 ETFs as the Telecom and Real Estate sector ETFs were not available - but these sectors are still represented, albeit, mixed in with other sector ETFs.

Ticker

Name

(XLB)

Materials Select Sector SPDR

(XLE)

Energy Select Sector SPDR

(XLF)

Financial Select Sector SPDR

(XLI)

Industrial Select Sector SPDR

(XLK)

Technology Select Sector SPDR

(XLP)

Consumer Staples Select Sector

(XLRE)

The Real Estate Select Sector

(XLU)

Utilities Select Sector SPDR

(XLV)

Health Care Select Sector SPDR

(XLY)

Consumer Discretionary Select Sector Fund

(XTL)

SPDR S&P Telecom ETF

This chart will be our benchmark for the following sector rotation strategies.

Holding Sector SPDR ETFs Equal-Weight

If we held all available sector SPDRs equal-weight since 1999, our annualized return would be 7.3%. If we compare this to the Russell 3000 ETF (IWV), we see that this 'dumb' equal-weight sector strategy beats the market long-term on both volatility and performance. This is not the purpose of this article but an interesting side point nonetheless.

Choosing a Sector Rotation Strategy

Next comes the hard part - what are our criteria for choosing one sector over another? One of the most common approaches is to use momentum. The principle behind momentum is that assets which have the highest trailing price performance over a certain look-back period (typically 3 to 12 months) often continue to outperform over the subsequent period.

The following tests will retain the top 5 ETFs on a 4-week rotational cycle using a variety of momentum periods.

Test from 1999 to 2018

Momentum Lookback Period

Compound Annual Return

Max Drawdown

Sharpe

All sectors equal-weight (bench)

7.30%

-53.30%

0.46

13 weeks

5.79%

-48.75%

0.37

26 weeks

6.68%

-48.56%

0.43

52 weeks

6.76%

-47.08%

0.43

My momentum tests do not show any additional out-performance by holding the highest momentum sector funds. At least, there isn't an improvement in the risk to reward profile as highlighted by the Sharpe Ratio.

Sector Rotation Based on Low Volatility

Perhaps we are attacking this the wrong way. Maybe we want to hold the sectors with the lowest volatility. Slow and steady wins the race.

I will look at a variety of ways to filter based on volatility.

Test from 1999 to 2018

Sector Rotation Method

Compound Annual Return

Max Drawdown

Sharpe

All sectors equal-weight (bench)

7.30%

-53.30%

0.46

Beta 3 Year

7.95%

-48.47%

0.54

Trailing 100-day Average True Range

6.44%

-46.81%

0.44

Standard Deviation

7.23%

-46.93%

0.5

It does appear that there is a slight advantage to holding sectors which have lower volatility. It is debatable as to how much additional benefit there is - but some are present. But do we really need to employ sector rotation for this? Couldn't we just pick sectors that have some defensive properties and get similar results?

Holding 3 defensive sector ETFs (healthcare, utilities, and staples) would have resulted in annual return of 6.4% since 1999, a max draw-down of -40.5% and a Sharpe Ratio of 0.47.

Modern Portfolio Theory?

What about Modern Portfolio Theory? Perhaps we keep the 5 ETFs with the lowest trailing 3 months of volatility and then apply a mean-variance optimization. This is where we blend the ETF weights to get the highest return for the volatility risk.

Modern Portfolio Theory Mean-Variance Sector Blending

The annual return is 8.1% with a maximum draw-down of -45.26% and a Sharpe of 0.55. This is very similar to our low-beta strategy.

What About Lower Liquidity?

Here is a novel approach that I have discovered to be useful... but as of yet, I cannot fully explain why it works. So take this with a grain of salt and skepticism as to whether it will continue to work. I find that the sector funds with the lowest dollar volume turnover seem to perform the best.

Why might this be the case? Perhaps the sector ETFs garnering the most attention are the most risky - the most potential upside and downside. And maybe the gamble doesn't pay off on average. Maybe it is a fluke. I would really appreciate your thoughts and opinions as to why this might be so. Does low dollar volume represent a lack of interest and thus a 'boring sector'?

The strategy holds the 5 sector funds with the least amount of turnover over the past 20 days. It replaces SPDRs every 4 weeks.

Low Turnover Sectors

I run a further test which places these sector ETFs into 3 portfolios or buckets based on their turnover. The leftmost bucket is the Russell 3000 index. The next (blue bar) is the highest turnover sort and the far right (yellow) is the lowest turnover portfolios. These portfolios are resorted every 4 weeks since 1999. The one-third of sectors in the highest turnover bucket have 2.3% less annual return than those in the lowest turnover portfolio.

Bucketed Portfolio Sorts Based on Dollar Volume Turnover of Sector ETFs

What I like about this strategy is that there is very little turnover or rotation of ETFs needed. It makes little difference whether you rotate every month or year.

These are the rankings from lowest turnover to the highest.

Ticker

Name

20 Day Turnover

XTL

SPDR S&P Telecom ETF

Lowest Turnover

XLRE

The Real Estate Select Sector SPDR

XLB

Materials Select Sector SPDR

XLP

Consumer Staples Select Sector SPDR

XLY

Consumer Discretionary Select

XLV

Health Care Select Sector SPDR

XLU

Utilities Select Sector SPDR

XLI

Industrial Select Sector SPDR

XLK

Technology Select Sector SPDR

XLE

Energy Select Sector SPDR

XLF

Financial Select Sector SPDR

Highest Turnover

I should note that the Telecom and Real Estate ETFs are fairly new and maybe this is why they have lower turnover at this point in time.

Other Sector Rotation Ideas

Are there not other sector rotational ideas and funds we could employ? Absolutely!

First Trust Dorsey Wright Focus 5 ETF (FV) employs a momentum technique on their sector strategy. But since inception, I do not see any additional benefit as opposed to holding all the sectors equal-weight. Pass.

And what about PowerShares DWA Tactical Sector Rotation (NASDAQ:DWTR)? I am even less impressed. From December 2015 until today, it has an annual return of 3.85% with a Sharpe Ratio of 0.65. First Trust did over 9% CAGR with a Sharpe of 0.95 and our 'all sectors equal-weight' did 11.7% CAGR with a Sharpe of 1.71. Pass.

So I throw it out to you - the informed reader - is it really worth spending the additional fees and costs to get someone to manage a sector rotation strategy for you?

What do you think about ETF-based sector rotational strategies?

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 (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.