Diversification Doesn't Work?

How to manage around diversification flaws cited by one well known hedge fund manager.

Long time listener, first time caller. I own Bitcoin, Ripple and Riot Blockchain (RIOT). Am I diversified?

This morning, Mark Sptiznagel from Universa Investments was on Bloomberg TV with some interesting things to say. I've mentioned Universa before. The strategy, its a hedge fund, is to bet on extreme events or outcomes such that with events like the one that played out over the last couple of days, they would make a killing. Waiting for those events can be expensive but the idea is that the occasional big gains will more than offset the small losses along the way.

First just some random comments that might be of interest before getting to the main point of this post. He is in the business of risk mitigation, he said timing crashes is impossible. "If you need to predict crashes to do well you're doing it wrong." He found it interesting that VIX moved up 20 points the other day which caused a 4% drop compared to when the S&P 500 fell 9% one day in 2008 but VIX only rose 14 points.

The blow up in the short-volatility was an incredibly naive trade in his opinion. Someone else equated the trade to picking up nickels in front of a steam roller. He also thinks the huge amount of assets in passive ETFs is a trouble spot but that was not explored any further.

Now to his comments on diversification. In trying to mitigate risk, people think diversification is the answer, but he doesn't think it is effective because it lowers returns and when things go wrong, correlations have the tendency to go up which would indeed dilute the benefits someone might think they are getting from building a diversified portfolio. The context of a diversified equity portfolio seemed to be owning a lot of stocks and/or equity funds.

He offered sound advice in saying don't do anything too complex, keep it simple. He thinks a lot of people are exposed to derivatives who shouldn't be, including some professionals. He was asked what then should an investor do and he offered that "the way to hedge the S&P is to own less of it."

We have been exploring this exact topic for many years. I typically modify diversified portfolio with the word properly to account for what I think is an obvious observation, during large market declines, all stocks should be expected to go down a lot. From peak to trough in the financial crisis, the S&P 500 fell close to 60% and long time client holding Johnson & Johnson (JNJ) fell a little under 30%. That 30% is still down a lot, obviously much better than the market and would contribute to outperforming a large decline. While I think that offers something of a place to hide, not everyone would agree with that.

While I would tell you that you should expect the typical stock to be a proxy for the stock market, there are plenty exchange traded products that are not proxies for equities and offer a better chance at reducing correlation. Yesterday, I wrote about the AGFiQ US Market Neutral Anti-Beta Fund (BTAL) which has a negative correlation to the S&P 500 per ETFreplay.com. Inverse funds have a negative correlation to equities. Merger Arbitrage funds like MNA have low but varying correlations to the S&P 500. Gold, like client holding SPDR Gold Trust (GLD) often but not always has a negative correlation to the S&P 500. The WisdomTree Managed Futures' (WTMF) ranges from negative to modestly positive.

Anyone who's been reading my posts for a while knows what I am going to say. Tactical use of these products when risks of a bear market are elevated (my indicators are the 200 day moving average, an inverted yield curve and the 2% rule) combined with a couple of sales (it doesn't take too many) can go a long way toward avoiding the full brunt of a serious decline. This would be my idea of being properly diversified. It is a dynamic process but of course markets are dynamic.

I am not sure, but the context of the Spitznagel interview seemed to imply that the last few days was a serious market event. It was serious for inverse VIX ETNs but that's it, at least for now. I think of it more of a litmus test for certain hedging products like the ones above except maybe for managed futures which are based on longer term trends so a five day event could yield any sort of result.

There are no guarantees for any of this to work in the next serious event. I believe they will but they may not which is why I would want to use several of these in small doses as opposed to a big bet on just one.

Comments