A fun fact for US equity market watchers who enjoy etymology: the modern English word “Fang” comes from old Norse/Germanic roots and originally meant “to seize or capture”. That is neatly appropriate for Wall Street, where the FANG stocks (Facebook, Amazon/Apple, Netflix, Google) have certainly captured investor attention. And, of course, plenty of incremental market capitalization as well.
As a result, there is a common perception that FANG dominates the entire US equity market. On a performance basis, that’s largely true of course. In aggregate, FANG is 9.8% of the S&P 500 and those stocks are up anywhere from 13% (Apple) to 98% (Netflix, even after today) on the year. Without them, the S&P 500 would be closer to flat on the year instead of up 5%.
We got to wondering what the FANG names’ actual price correlations to the S&P 500 might be. Are they really 1:1 tied at the hip to the overall US equity market? And are there any differences between them? Here’s what we found by looking at 90-day price return correlations between the FANG stocks and the S&P back to the start of last year:
#1. Google consistently shows the highest correlation to the S&P 500. That is surprising, if only because GOOG/GOOGL have a slightly lower aggregate weighting in the index (3.03% total) than Amazon (3.09%) and much lower than AAPL (3.95%).
This tight linkage only kicked into high gear in mid-February 2018; before that it was reliably 0.50 – 0.60 all the way back to the start of last year. Now, the correlation is 0.83 and was as high as 0.88 in May. For those readers with a statistical bent, the r-squares were 30% in 2017 but more like 70% now. It is a notable difference, from well below 50% to well above it.
The big question here is “Why the increase?” Our best explanation: Google has become an especially crowded trade in 2018 and sees its performance wax and wane with daily market moves.
#2: Amazon is the second most correlated FANG stock to the S&P. Its current reading is 0.75, but unlike Google its correlations to the market are still climbing. In May, for example, AMZN’s correlation to the market was 0.67. In June, it was 0.70. Like Google, however, correlations for Amazon are significantly higher than 2017’s 0.50 – 0.60 levels.
We chalk up the pop in correlations in 2018 to the crowded trade syndrome as well. Not that this has hurt the stock, mind you…
#3. In third place is Apple, with a 0.68 correlation. Like the first two names, its correlations now are much higher than in 2017 (also 0.50 – 0.60 here). Unlike Google and Facebook, correlations are now dropping more noticeably from their highs earlier this year (0.79).
#4/#5. Currently at the low end of the correlation spectrum are Netflix (0.62) and Facebook (also 0.62). While that may not seem appreciably lower than Google’s 0.83, the difference in r-squares tells the story better. NFLX and FB show a 38% r-squared to the overall US equity market, but Google is at 69%.
So do these lower correlations make these names safer to own? Yes and no… Facebook’s price correlation to the S&P has been generally steady at current levels for the last 18 months. Netflix’s has been climbing, much like Google and Amazon.
Just on this math, Facebook seems less of a crowded trade than Netflix. That makes sense if only because FB’s run-ins with regulators (past, present and future) likely scares away some investors. And Netflix’s Lazarus act today after its subscription miss shows the crowd hasn’t given up on this name just yet.
Summing up: we embrace the old adage that “Correlation is not causation”, so we would never argue that US equities generally are inextricably tied to the FANG stocks. The recent increases in correlation for names like Amazon and Netflix, however, do make us worry about who owns those names.