Asymmetry Not Complexity

Let's have some fun with portfolio theory thanks to a note from Howard Marks and a short paper from Man Institute, both of which came via Idea Farm. 

Page three from Marks really has the goods. 

"You can't simultaneously emphasize both preservation of capital and maximization of growth...it becomes clear the goal should be optimization not maximization...the goal should be to achieve the best relationship between return and risk."

I take this to be about having a long term view focusing on a long term horizon. At 50 years old, making a behavioral mistake in your portfolio is far more likely to have a long term impact than some sort of big win. Mistakes are most likely to be caused by some sort of panic sale or making some sort of big bet that goes wrong. 

Man Institute tried to answer three questions, two of which were particularly interesting, and they shared some interesting nuggets along the way. 

1 Should you bother owning anything but stocks?

Their answer was obviously yes. One interesting nugget though was that by their work, trend (managed futures) has compounded at 11% annually versus 10% for stocks. Also interesting is that gold has compounded at 8%. This chart of gold versus the S&P 500 might surprise you, it did me. The chart is shorter than the time frame they were referring to though.

The paper made an interesting behavioral point in answering question 1 which is that investors' time horizons shorten considerably during bear markets or other types of scary declines. 

They believe in leveraging up to get a higher Sharpe Ratio (this is a measure of risk adjusted return where higher is better). This is a version of risk parity. They land on an allocation of 30% stocks, 60% bonds and 60% alternatives as being an optimal expression of this thought and we can backtest the idea.

All three portfolios allocate 30% to Invesco S&P 500 Momentum (SPMO), 60% to iShares Aggregate Bond ETF (AGG) and you can see how the 60% alternative sleeves are built out. I tried to choose two very different types of alts that we regularly use for blogging purposes. 

The results are surprising insomuch as the weight to equities is so low and Portfolio 3 does have a noticeably lower Sharpe Ratio and standard deviation even if not night and day different while being right in the mix CAGR-wise. Portfolio 3 also has the best Calmar Ratio (higher is better) and the best kurtosis (lower is better), dramatically better than VBAIX. We can backtest the leverage but we'd have very limited choices implementing it though. You could take out AGG, which I wouldn't want anyway, allocate 30% to SPMO, 30% to each of the two alts and the remaining 10% to cash and you'd have a CAGR of 8.52%, standard deviation of 7.50%, the Calmar is the highest I've ever seen at 4.77 but the kurtosis isn't as good as the original Portfolio 3. 

The other question from Man to mention is should we bother trying to time stocks? Where the context is usually all in or all out, the answer from me is clearly no. Markets to swing to extremes occasionally though and in those instances it can make sense to adjust net long exposure up or down a little bit. A simple example is when the S&P 500 gets more than 20% away from its 200 day moving average. That has shown to be unsustainable. Increasing net exposure when the index is 20% below the moving average makes sense and decreasing net exposure when it is 20% above the moving average. I'm talking like 2-4% changes in exposure so the consequence of being wrong isn't disproportionate. 

Also buying a little more after a large decline is often a good idea. The idea is not to think you'd be buying the bottom but that you're buying much cheaper. The S&P 500 will get to 10,000 at some point. Maybe it will take many years or maybe it will come quickly. It will happen, we just don't know when or the path to get there. With the S&P 500 at 6000 today, knowing it will be at 10,000 at some point, would you think buying a pullback to 4800 would be a good idea? Sure it might go to 4000 after 4800 but 10,000 looks pretty good from 4800. 

And a bonus from the FT who says that university endowments have struggled with performance lately due to huge weightings in private equity. Specifically, Yale and Princeton were called out as lagging smaller endowments with the FT noting they have "at least 40%" in private equity and/or venture capital. It's not quite right to call private equity uncompensated complexity but I do think calling it unnecessary complexity for retail sized investors just trying to make it to and through retirement successfully is fair. 

To actually access private equity requires high dollar amounts and the willingness to lock the money up. Buying funds with ticker symbols in your brokerage account is complicated. Here are two examples. ARKVX just trades an NAV and DXYZ is a closed end fund that owns some of the right names but where you see the price is around $40, the NAV is at $5.15. 

We've talked once or twice about the ETFs that might be coming to the private equity space but there will be complexity there too with how they mark to market to provide liquidity. I also have doubts about the quality of the names the ETFs will own. We talk a fair bit about this but I think it is better and simpler to just buy one of the management companies that is selling access to private equity investments. The operating companies tend to do very well but with a lot of volatility in both directions. 

In a related note, the UK Pension is allocating 3% to Bitcoin. It's a bet partially on asymmetry. We've been talking about Bitcoin as asymmetry for years and it has generally gone up but with a lot of volatility. I think the endowments could get more bang for the buck going the asymmetric route, which I would argue is also simpler than putting a ton, 40% is a ton, into private equity. 

The above are not rebalanced. The point is not that anyone actually did this but they are clear examples of the concept. 1% Bitcoin, the rest in cash came pretty close to 100% in the S&P 500. I'm not so sure that Bitcoin as an asymmetric bet came out of nowhere but maybe Nvidia did come out of nowhere. The practical application of this that I think can work for individuals and endowments is to build out a simpler portfolio than what the bigger endowments are doing but carving out maybe three percent to make small allocations to two or three asymmetric bets. Bitcoin might be one of those. Some people think uranium offers asymmetric potential, you'd need to do the work to figure out where the opportunity is. Maybe you're right, if so you come out way ahead. 1% Bitcoin/99% S&P 500 unrebalanced compounded at 18.36% over the same period. If instead of going up a ton in that period, Bitcoin went to zero, the 99% invested in the S&P 500 would have grown from $9900 to $23,593. 

I am not saying it is easy to find asymmetric winners but I am saying anyone so inclined can put in the work to look. 

The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.