Market Timing: Look To Large Caps For The Canary In The Coal Mine

| About: SPDR S&P (SPY)

Summary

Long-term trend following can provide better performance than buy and hold, at least for the S&P 500.

Small cap and mid cap indexes do not seem to provide their own best signals for exit and entry.

But small caps and mid caps often can outperform large cap stocks over long periods of time.

This performance can be even better if looking to the S&P 500 for buy/sell signals.

The SPDR S&P 500 Trust ETF (NYSEARCA:SPY), which tracks the S&P 500 index, may be a good indicator of the direction of the entire U.S. stock market, but it's not the only way to try to play the long-term trends in domestic stocks. For those who invest in domestic equities outside of the large caps, they may be interested (and surprised) to learn that keeping an eye on more than the major index isn't just more to keep tabs on; it's also potentially misleading.

In the first article in this series on market timing, a strategy of attempting to follow long-term trends in the S&P 500 index was compared against buying and holding the index for three historical time periods of twenty years each. In each case, trying to catch the upward trend while avoiding a large drawdown led to some underperformance in bull markets and strong outperformance in bear markets, for net positive long-term gains and less volatility.

Let's take a look, then, at a few of the oldest and largest domestic ETFs and see how performance would have fared from 2001 to 2016 under three different strategies: buy and hold, following the trend of the ETF itself, and using the trend of the S&P 500 as an indicator.

The period of 2001 to 2016 is chosen because it is the period in which data is available for these ETFs. The strategy will not perform nearly so well in other fifteen year periods as it did during the violent swings of the last fifteen years, but that's why we looked at some time series in the more remote past in the last article. This article is focused on gauging the exits and entries for long-term trend following with different domestic ETFs.

Simple Moving Averages Help Measure Trends

All trends are measured using the 50-day and 200-day simple moving averages, but the degree to which the 50-day moving average must cross above or below the 200-day moving average is adjusted to optimize for performance with that particular ETF. Such parameters can reduce the number of "noisy" signals that quickly reverse and also can be used to increase the tolerance for volatility within the long-term trend. (You can think of it as working a little like a thermostat, which doesn't kick in exactly when the target is crossed but only a little above or below.)

For the S&P 500 and when using the movement of the S&P 500 as a signal for other domestic ETFs, the buy signal occurs when the 50-day moving average is 1% above the 200-day moving average, and the sell signal occurs when the 50-day moving average is 1% below the 200-day moving average. There is nothing magical about these numbers or about simple moving averages, of course. They're suited to observing medium-to-long-term trends in the rear view mirror, but they're not the only way to do so.

All backtests in this article are calculated based on daily market close data with dividends reinvested.

SPDR S&P 500 Trust ETF

Click to enlarge

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $20,834 for an annualized return of 5.07% with a 55% maximum drawdown. The long-term trend-following investor ends the fifteen year run with $32,460 for an annualized return of 8.26% with a 17.3% maximum drawdown.

SPDR S&P MidCap 400 ETF (MDY)

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $30,840 for an annualized return of 7.89% with a 55% maximum drawdown. The long-term trend-following investor in MDY, using the buy/sell signals described above for SPY when switching to and from a cash position, ends the fifteen year run with $39,464 for an annualized return of 9.7% with a 23.3% maximum drawdown.

However, the long-term trend-following investor, using the SMA 50/200 crossovers for buy/sell signals applied over the data for MDY, ends the fifteen year run with only $24,138 for an annualized return of 6.12% with a 25.2% maximum drawdown. Attempts to adjust the parameters (not shown) do not produce better results than buy and hold. Such results, using the data for MDY as signals for buying and selling MDY, generally become better to the extent that they more closely approximate the buy and hold strategy, with later exits and earlier entries. Attempting to follow the long-term trend of MDY itself (when measured by the SMA 50/200 crossovers) carries a significant risk of underperforming the buy and hold strategy.

iShares Russell 2000 Index (IWM)

Click to enlarge

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $26,372 for an annualized return of 6.76% with a 58.6% maximum drawdown. The long-term trend-following investor in IWM, using the buy/sell signals described above for SPY when switching to and from a cash position, ends the fifteen year run with $42,884 for an annualized return of 10.32% with a 23.3% maximum drawdown.

However, the long-term trend-following investor, using the SMA 50/200 crossovers shown above for buy/sell signals applied over the data for IWM, ends the fifteen year run with only $21,951 for an annualized return of 5.44% with a 46.5% maximum drawdown. The results are worse if the parameters are not adjusted to reduce sensitivity to volatility (as one might expect when following small caps). Here all the same caveats apply here, as they did before with the mid caps and MDY. Attempting to follow the trend of the lower-cap ETF itself led to underperformance.

Vanguard Total Stock Market ETF (VTI)

VTI is a capitalization-weighted ETF covering most of the U.S. stock market. Here we see some of the themes from the previous three charts coming together in one illustration: small caps and mid caps can add to performance, but using their price data as a buy/sell indicator can add to the noise.

Click to enlarge

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $25,144 for an annualized return of 6.41% with a 55.5% maximum drawdown. The long-term trend-following investor in VTI, using the buy/sell signals described above for SPY when switching to and from a cash position, ends the fifteen year run with $34,498 for an annualized return of 8.71% with an 18.5% maximum drawdown.

Using a particular setup for following the long-term trend of VTI itself resulted in a final figure of $31,368 for an annualized return of 8.01% with a 21.8% maximum drawdown. Adjusting those parameters (not shown), however, proves them to be relatively brittle and capable of underperforming a buy and hold strategy. Simply following the trend of SPY provides better and more robust backtested performance.

Some of the outperformance could also be attributed to the inclusion of stocks on the Nasdaq in VTI.

PowerShares QQQ ETF (QQQ)

This ETF tracks the 100 largest companies on the Nasdaq, weighted by capitalization. It is tilted towards technology stocks and biotechnology stocks.

Click to enlarge

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $24,982 for an annualized return of 6.37% with a 60.7% maximum drawdown. The long-term trend-following investor in QQQ, using the buy/sell signals described above for SPY when switching to and from a cash position, ends the fifteen year run with $40,836 for an annualized return of 9.95% with a 17.3% maximum drawdown.

The investor who tries to follow the long-term trend of QQQ, as indicated above, ends the fifteen year run with $39,668 for an annualized return of 9.74% with a 26.6% maximum drawdown. This hypothetical backtested return is very sensitive to changing the parameters for the buying and selling signals. Also, the maximum drawdown is considerably higher than it is when using the SPY signals as a buy/sell indicator. While this ETF (along with SPY) is the best of those tested for following its own trend, it seems safer just to follow SPY as an indicator in this case as well.

SPDR Dow Jones Industrial Average ETF (DIA)

Click to enlarge

(Chart from 3/13/2001 to 1/15/2016. Copyright © Peter Kirby.)

The buy and hold investor ends the fifteen year run with $21,859 for an annualized return of 5.41% with a 51.9% maximum drawdown. The long-term trend-following investor in DIA, using the buy/sell signals described above for SPY when switching to and from a cash position, ends the fifteen year run with $30,109 for an annualized return of 7.72% with an 15.8% maximum drawdown.

Attempting to follow the long-term trend of DIA itself, using the buy/sell signals above, ended with $22,461 for an annualized return of 5.61% with a 15.8% maximum drawdown.

What this may show is that the Dow Jones no longer has the breadth needed to capture the overall trend and sentiment of U.S. stock market. It also might just indicate the rise to prominence of the S&P 500 as a barometer of the stock market, over the Dow Jones, that occurred in the late 20th century. Backtesting over a historical data series going back to 1930, not shown here, would demonstrate that the Dow Jones Industrial Average has been more useful for indicating its own trend in the more remote past.

Apparently, for Market Signals, Size Does Matter

Originally I attempted to apply a trend-following technique to each individual domestic ETF, for which there were some decent results (DIA, QQQ - both large cap ETFs) and some poor results (MDY, IWM - both non-large cap ETFs). The inconsistency of these results led me to attempt to come up with a better thesis when it comes to finding entry and exit signals in US equities for this time period.

The hypothesis that SPY provides entry and exit signals provided good results across the board, for all the domestic ETFs considered. This suggests that a person attempting to apply a strategy of following the long-term trends in US equities can actually get better results just by following the S&P 500, both when investing in SPY or when investing in other domestic equity ETFs.

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.