Beyond the Random Walk

A not so random walk through the world of finance.

Industry Momentum Trading Strategy

Posted by Vijay Singal on May 27, 2008

Background

Industry momentum refers to a phenomenon where industries that have done well in one period also do well in the next period. This pattern in returns has been documented and verified through extensive research by practitioners and academics over the last several decades.

One industry may be hot today and another may be hot next month depending on changing fads, individual preferences, national requirements, or political expediency. For example, after the terrorist attacks on September 11, 2001, the defense and security industry did exceptionally well, as demand for such equipment and personnel increased. Similarly, bad weather or continued drought can affect production of commodities (increase prices) and hurt industries that use those inputs. If changes in an industry are permanent, then a change in stock prices should reflect all of those industry changes promptly with no room for a predictable price pattern. However, if changes in an industry occur gradually, then stock prices may also change gradually. Evidence tends to support gradual movement in industries and in returns of industry portfolios. This means that compared to an industry that underperforms, an industry that does well in one period is more likely to do well in the next period. Such predictability of returns enables the creation and implementa­tion of profitable trading strategies. Industry portfolios constructed in this manner generate returns that may be much larger than the S&P 500 return, at not necessarily higher risk.

To be sure, there is evidence of momentum in individual stocks. However, trading in individual stocks can become very expensive decimating any profits from a momentum trading strategy.

Fidelity Sector Funds

I use Fidelity Sector funds for the trading strategy because they are single-industry mutual funds designed to capture industry movements with a few stocks. The advantage of Fidelity Sector funds over homemade industry portfo­lios is that the cost of trading and maintaining a portfolio of Fidelity Sector funds is relatively small. Trading among Fidelity sector funds can be costless if the portfolio is held for at least thirty days and traded via the Fidelity website. The short-term trading fee is 0.75 percent if the fund is held for less than thirty days.

Alternatives to Fidelity sector funds are available such as exchange-traded funds or sector funds with unlimited trading created by Rydex and Profunds. I prefer the Fidelity funds because they have a long history of almost two decades, so they are easy to evaluate. And the greater flexibility provided by alternative vehicles is not useful because my strategy does not require more frequent trading than permitted by Fidelity.

However, there are several limitations of Fidelity sector funds. First, these portfolios reflect gains or losses due to active management by the fund manager. The fund manager will alter the portfolio de­pending on his perception of the prospects of a particular company. If the manager is uncertain about the whole industry, he may re­duce exposure by holding cash, making exposure to the industry incomplete. Second, the fees and expenses associated with the portfo­lios can be large. The expense ratio, which includes administrative costs and management fees, is about 2 percent per year. Addition­ally, nearly all sector funds have annual turnovers in excess of 100 percent. This means that, on average, all stocks are traded at least once a year. Such frequent trading will affect returns negatively. The transaction costs, such as the bid-ask spread and brokerage fees, are not part of the expense ratio. In spite of these shortcomings, Fidel­ity sector funds are best suited for industry-momentum-based strategies.

Trading strategy set up

There are 43 Fidelity sector funds in 21 unique industries. Fidelity has 5 funds in health care: medical delivery, medical equipment, pharmaceuticals, biotech, and health care. I consider all of them in the same industry. On the other hand, there is a separate sector fund for gold, also an industry; for leisure, also an industry. When multiple sector funds constitute an industry, the amount to be invested is equally divided among the sectors.

In order to minimize portfolio risk through diversification, my recommendation is to invest in 10 out of 21 industries. Fidelity requires a minimum investment of $2,500 for each sector fund. Given that some industries have more than one sector fund, a minimum investment of $50,000 is required. However, $100,000 is desirable. All trades can be made through Fidelity or another brokerage firm that trades Fidelity funds. However, non-Fidelity brokerage firms may charge a fee for trading Fidelity funds.

Initial purchases of mutual funds are not marginable. However, once a mutual fund has been held for 30 days, it becomes marginable. Exchanges of margined mutual funds within the same family continue to be marginable and are not subject to a new 30-day waiting period.

Trading Strategy Implementation

There are two periods to consider: an estimation period to determine past performance and a holding period for which the recommended portfolio should be held. In this strategy, I have found that a 5-week estimation period and a 5-week holding period have been the easiest to implement and give relatively good results.

I intend to create a portfolio every weekend, which will be announced by morning of the first trading day of the week (usually Monday morning) so that funds can be bought at the close of the first trading day. Portfolio 1 is being announced this weekend (for trading on May 27th). Portfolio 2 will be announced for trading on June 2, portfolio 3 for trading on June 9, portfolio 4 for trading on June 16, and portfolio 5 for trading on June 23.

Since the holding period is 5 weeks, portfolio 1 will be reconstituted on June 30. Sector funds no longer in the new portfolio will be sold and exchanged for sector funds that need to be added to the portfolio. At that time, performance relative to other benchmarks will also be reported.

Though the strategy is designed for an investor in only one portfolio, you may participate in all 5 portfolios as a way of further diversification.

Details

Cost and Frequency of Trading: Trades will take place once every 5 weeks. Trades on Fidelity’s web site are free. Other brokerage firms may charge for trading of mutual funds. Assuming $10 per trade and 50% turnover in the portfolio holdings, you can expect to pay $1,000 a year [$10 times exchange of 10 funds every 5 weeks for 10 times in a year] in trading costs with a non-Fidelity brokerage house.

Taxes: Note that this trading strategy relies on short-term trading. When compared with a buy and hold strategy, industry momentum will result in higher taxes. You can reduce your tax liability by trading in a tax-deferred account. In any case, it is important for you to compute your after-tax returns to determine whether or not the extra returns are sufficient to overcome the extra taxes.

Past Performance

The annual past performance of this strategy is given below for an 18-year period, from 1990 to 2007. The graph is followed by a spreadsheet. The first column of the spreadsheet contains the year, the next three columns contain S&P 500 performance, and the last three columns contain the performance of a sector rotation strategy. The sector rotation strategy outperforms the S&P 500 by approximately 4.5% a year, on average. An investment of $100,000 grows to $660,000 with a buy and hold strategy and to $1,380,000 with sector rotation. Neither strategy considers tax liability.

The risk of the sector rotation strategy is marginally lower than the risk of S&P 500. The strategy outperforms the S&P 500 in 12 years out of 18 years. Of course, past performance is not necessarily a good indicator about future performance.

Comments are always welcome on this blog.

4 Responses to “Industry Momentum Trading Strategy”

  1. Praveen Chawla said

    Do you think results will be better if you focus more i.e. focus on 5 industries rather than 10?
    I would guess volatility would be more but returns higher.

  2. vsingal said

    Very good question and something I have experimented with earlier. The volatility is definitely higher with a 5-sector portfolio but the returns are practically unchanged looking at the last 15 years. A 3-sector portfolio has slightly higher returns though, but with much higher volatility. Another possibility to invest only when the returns are positive – that one is worse both ways: higher volatility and lower return.

  3. Advait said

    Hey Vijay,
    Great post, very interesting analysis of the momentum strategy. I enjoy following your blog posts from time to time as well.

    I work for http://www.wikinvest.com and am personally approaching high quality bloggers such as yourself to join the Wikinvest Wire(www.wikinvest.com/blogger/Wikinvest_wire). The Wire is an invitation only blogwire, which links bloggers across the financial space, and media partners such as CNET News, ABC News, SF Chronicle, USA Today, etc. The eventual reach of the Wire is slated to be around 400 million. Its a very exciting venture and I do hope you’re interested. Beyond the random walk is a great blog and we’re hoping to have you on the Wire.

    If you are, let me know at advait@wikinvest.com and we should talk more.
    Look forward to hearing from you

    Thanks,
    Advait

  4. Laurent Giroud said

    Hi, thanks for this very interesting article.

    I am currently Benjamin Graham’s “The intelligent Investor” and this strategy reminds me a lot of the rebalancing one he suggests adopting between stocks and bonds depending on their respective returns. I had thought to study the returns brought by a permanent rebalancing between individual stocks in a diversified portfolio but the sector approach is obviously much safer. I will be definitely buying your book soon.

    Looking at the graphic, I notice that during the two periods during which the S&P globally changes direction the strategy seems to bring even bigger returns than usual. Could that indicate that tracking momentum allows to anticipate the change of direction and benefit from it ? Of course 20 years is not a big period of time and this could prove to be a statistical anomaly but I guess this might make an interesting research subject.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Connecting to %s

 
Follow

Get every new post delivered to your Inbox.