Looking at Market Correlations

Here’s some data I’ve been working on that I thought I’d share with you. Below is a table showing the correlations of the daily changes of the 10 S&P 500 sectors. (I apologize for the crowded data.) I’ve also included the daily changes for the S&P 500 and for our Buy List. The data only covers calendar year 2015.

Date Ene Disc Spl Fin HC Ind Tech Mats Tele Utes S&P 500 BL
Energy 60.3% 56.8% 65.2% 53.8% 69.1% 59.0% 79.2% 57.0% 38.6% 73.1% 65.0%
Discretion. 60.3% 81.8% 86.6% 80.3% 85.7% 84.3% 74.4% 70.8% 49.7% 92.9% 91.6%
Staples 56.8% 81.8% 80.8% 75.3% 83.9% 78.0% 70.4% 73.0% 63.8% 88.3% 85.7%
Financials 65.2% 86.6% 80.8% 76.1% 87.7% 83.2% 77.3% 75.2% 46.5% 93.3% 91.9%
Healthcare 53.8% 80.3% 75.3% 76.1% 76.4% 76.8% 67.2% 58.1% 47.9% 87.0% 84.0%
Industrials 69.1% 85.7% 83.9% 87.7% 76.4% 86.0% 84.8% 76.0% 54.3% 94.2% 91.3%
Tech 59.0% 84.3% 78.0% 83.2% 76.8% 86.0% 74.7% 71.2% 45.8% 92.6% 90.0%
Materials 79.2% 74.4% 70.4% 77.3% 67.2% 84.8% 74.7% 67.0% 44.9% 84.8% 79.3%
Telecom 57.0% 70.8% 73.0% 75.2% 58.1% 76.0% 71.2% 67.0% 52.2% 78.2% 74.7%
Utilities 38.6% 49.7% 63.8% 46.5% 47.9% 54.3% 45.8% 44.9% 52.2% 56.6% 52.3%
S&P 500 73.1% 92.9% 88.3% 93.3% 87.0% 94.2% 92.6% 84.8% 78.2% 56.6% 96.5%
Buy List 65.0% 91.6% 85.7% 91.9% 84.0% 91.3% 90.0% 79.3% 74.7% 52.3% 96.5%

A few things to note.

First is how strongly stocks are correlated with one another. At least in the near-term. Even a low daily correlation is around 60% or 70%. This means that just by being a stock, you’ll get tossed around with whatever the market’s doing that day. This is actually good news for stock-pickers because it means that good stocks can get knocked down with everything else, thereby creating good bargains. The shorter the time period, the greater the impact of the overall market.

I also think it’s interesting how the Industrials have the highest correlation with the broader S&P 500. I’ve noticed this in previous years as well. In other words, the Industrials stand out by being the most similar to everyone else.

Oddly, I find that this makes sense yet I’m not sure why. Perhaps it’s that the Industrials lie at the middle C of the economy. While sectors like Energy or Finance can diverge from the broader economy, the Industrials are unlikely to because…well, that is the economy. I’m not exactly sure, but it’s not due to size. Industrials are big but not the biggest.

But this also means that if you wanted to build a cheap-but-somewhat-decent index, you could probably use as few as five major industrial stocks. By this, I mean companies like DuPont or 3M. Sure, it would be far from perfect, but considering how small it is, it would be pretty darn good.

Naturally, over the short term, a stock will reflect the broader market and its sector. But as time goes on, the value specific to each company would gradually move to the surface and the higher categories would lose importance. On the table, you can see how strongly correlated the Buy List is with the S&P 500 on a daily basis, yet after 12 months, we had no trouble separating ourselves from the market last year.

It’s interesting to see sector relationships in the market. For example, Energy and Materials are often closely related, yet Materials are usually closer to Industrials than Energy is. Staples and Healthcare are often close, but not so much last year. Tech is often a lone wolf, not much correlated with anyone. In that regard, it’s the opposite of Industrials.

I suspect that the numbers on the table for Utilities are unusually low. Last year was probably an aberration for the Ute sector. Typically, I’d expect a higher correlation with defensive sectors like Staples and Healthcare.

Finally, don’t overthink this table. I think it’s interesting but only in the sense that it gives us a glimpse at how the market behaves. None of these correlations is stable. The major lesson is that a powerful market trend can carry off any stock.

Posted by on January 11th, 2016 at 8:35 am


The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.