When Tech Diverges

Jacob Hess

November 29, 2020

• Stock Market
• Technology
• Data Analysis

The stock market has been on a tear. Despite a global pandemic reemerging in the second half of the year, investors seem to be okay with buying on expectations of an improvement in the economy in 2021 because that seems to be the only real reason for bulls to win out in the market. Unemployment is still at excessive levels with more than 700,000 claims for unemployment being filed every week (though continued claims are slowly edging lower). The bounce in output growth was strong over the summer but momentum seems to be slowing in retail sales, manufacturing, and industrial output. Despite this, indexes reach all-time highs.

That is even more so true for the Nasdaq which has rocketed past its pre-pandemic highs. The index tracking the top tech companies seems to have left the S&P 500 and the Dow Jones Industrial Average in the dust. Based on trading from February 24th to November 27th, the Dow and the S&P 500 have grown 6.97% and 12.79% while the Nasdaq has surged 32.37%. Some might say a divergence like this demands a correction because of the cyclicality of the market, but is that what typically occurs?

Using 5-year return data of ETFs that track the DJIA (DIA), S&P 500 (SPY), and Nasdaq (QQQ), it can be clearly demonstrated that these indexes have high correlations over long periods of time. Over the past 5 years, the correlations have been strong (Pearson's r is calculated). Even looking at the most recent YTD return data provides a similar conclusion. It's only recently that there has been a major divergence between two indexes, the DJIA and the Nasdaq (similarities between the S&P 500 and Nasdaq make the previously mentioned relationship more useful). Since October 1st, the returns of QQQ and DIA have had a correlation of just 0.56, much lower than a longer-term correlation that appears to be around 0.87.

This isn't necessarily an unusual occurrence. The Nasdaq's correlation with the other two indexes has varied quite a bit over time. In particular, the correlation between the DJIA and Nasdaq has seen several dips over the past five years. The chart below shows the correlation of a moving window of 100 days of returns. The data has several local minimums, points where the divergence between the DJIA and Nasdaq was strongest in near-term periods, including two instances of a correlation less than 0.5 and six instances of a correlation less than 0.7. These local minimums were almost always followed by a snap back to a more normal correlation.

Unfortunately, these "snap backs" didn't necessarily help predict how the indexes would perform in the future. 25-day, 60-day, and 100-day returns following each local minimum were not correlated with how strong the divergence was. In individual returns for each index (DJIA, S&P 500, and Nasdaq), there were practically no correlation coefficients above 0.15. The only returns that posted weak correlations were:

• Nasdaq - DJIA returns (Nasdaq excess returns) 60-days after local minimums had a 0.27 correlation coefficient with local minimum values.
• Nasdaq returns 100 days after local minimums had a 0.17 correlation coefficient with local minimum values.

There is not much statistical significance in these observations, but it does suggest that tech stocks have a weak propensity to underperform in mid-term and long-term periods following strong periods of divergence. This would make sense in instances of tech stocks cooling after sprinting past other sectors. An interesting observation does arise in short-term returns. In the 25-day return periods following the minimums, there is only one return period above 5% (Nasdaq following 6/23/2016) with almost half of those short-term returns in the negative.

So when tech diverges, what can investors expect? Results are not significantly significant, but they suggest that stocks, no matter the sector, are more likely to take a pause especially if the divergence is strong. As of November 27th, the DJIA and Nasdaq correlation over the last 100-days stands at 0.62 and is trending lower. Are bearish days ahead? It's hard to say, but it might be fair to point out that the downside risk for tech stocks is rising quickly as tech divergence intensifies.