Market correlations
/So, I have been worried about my portfolio these days (which mostly consists of index funds), because of all the volatility around the trade war and Trumps tweets. Not to mention rising recession fears. But it turns out that the S&P 500 really hasn’t been that bad, down just 1.75% since its high, up 20% since the beginning of the year and still above where it has been for most of the year.
Surprisingly, the Vietnamese ETF, VNM, that I follow the closest (because it has the best data) has been a mess. It’s up 6.6% YTD, but down 8.5% from its high and below where it has been for most of the year.
If I was starting from first principles, I would have thought the reverse. While both performed not great in 2018 (the end of the year was a disaster), the S&P index is up 50% in the past 5 years while VNM fell 31%. Given a predilection to revert to the mean, you would think that VNM would do much better in 2019. And the story about Vietnam is just so strong: real GDP growth of almost 7%, main beneficiary of the US-China trade war, and an increasing political spotlight because of the US-N. Korea summit. Plus, no real inflation concerns, which can really hurt equity returns.
SOURCE: YAHOO FINANCE DATA, VIETECON.COM
Well, I would have thought wrong. Still, I wanted to see how correlated the S&P500 and Vietnam’s stock market are. With the transmission mechanism being that Vietnam is a big trading partner with the US, gets lots of capital from the US, and, frankly, lots of stock markets internationally seem to be correlated. Based on data in this article, though, the S&P500 is not correlated well with an index of international equities when it is in a bull market, but is correlated when it is in a bear market (which is unfortunate for US investors that was diversification).
US & Vietnam stocks are loosely correlated: This data isn’t necessarily true for Vietnam. I ran a regression of changes in the S&P Index vs changes in VNM for the past 5 years. The adjusted-Rsquared is pretty low (0.3), but has a very low P-value, meaning that it is probably somewhat accurate. This equates to a the correlation found in times when the S&P is in a bull market. But it hasn’t been in a bull market over the past 5 years - both 2015 and 2018 were bad. So Vietnamese equities are somewhat different from international equities as a class.
But the markets do move in the same direction most of the time! I also did another test. I wanted to see if at least directionally the markets move together. When VNM is positive, is the S&P positive? Remember, over this period, VNM was down 31%, but the S&P was up 49%. So you wouldn’t think they would be moving together much. Well, you (and I) would again be wrong: They moved together 2/3rds of the time. I counted each time when both were up and when both were down.
That is strange. Not sure what that actually means, but it thought it was interesting. I wonder, if you just put a small bet on the S&P every day given what happened with the VNM, would you actually make money? It looks like it, or you would have over the past 5 years.
What I’m telling you is that if you use this proprietary data to make money, you owe me 2 and 20.
Not really. Good luck. It’s your grave.