To Drift…or Not to Drift?
Like street racers, portfolio managers also like to drift…and that’s about where the similarities end. Everyone know what drifting means in Fast and Furious, but what does it mean for portfolios and why is it so important here?
What is Drift in Portfolios?
Imagine you have a big box of Legos and you’ve got all kinds of pieces. Now, let’s say you want to keep your Lego box balanced. So, you decide you want half of your Legos to be red and the other half to be blue.
But every time you play with your Legos, you end up adding more pieces. Sometimes you add more blue than red, and sometimes more red than blue. After a few days of playing, you realize you’ve got a lot more blue Legos than red in the box. Your box isn’t balanced anymore — it’s drifted away from the starting point of half and half.
Portfolio drift is a lot like this. Imagine your money is like the Legos. You invest your money in different things, like $BTC and $ETH. When you first invest your money, you might have a starting point, like putting $1 in $BTC and $1 in $ETH (a 50–50 portfolio). But over time, let’s imagine your $ETH allocation grows to $3, while the $BTC portion only grows to $2. So, after a while, you might have more of your money in $ETH than you planned — just like having more blue Legos than red. This is what we call “portfolio drift.”
So Why is it Important?
Drift impacts returns by quite a bit. You can see that just by comparing the S&P 500 index to the S&P 500 Equal Weighted Index over the last 10 years (one allows more drift than the other). Obviously, market direction, the type of assets, etc., all play a massive role here. For example, in a bull run, do you want to sell out of the outperformers and buy some of the underperformers? Do the same rules and thought processes apply for cryptos as stocks? These are all questions that need to be asked when evaluating whether or not drift makes sense.
How Should You Think About This for Cryptos?
That’s the million dollar question. Because the decision to allow drift can’t just be thought of in isolation. Alongside that comes the question of rebalancing, and what to rebalance that drifted position to. Compared to stocks in the S&P 500, cryptos are way, way, WAY more risky. Even compared to companies in the High Yield index (companies that are more at risk of bankruptcy and defaults), cryptos are way more risky.
This is a chart from Fitch (one of the big rating agencies) showing what default rates are for High Yield companies. While we can’t tie this 1-to-1 with companies going bankrupt and bust, it’s a pretty decent proxy. Looking at the numbers here, you can see that the typical default rate for high yield companies sits at around 1–3% per year. I think we would all be very happy if the number of crypto projects dying out per year was only around 1–3%.
Many people think of investing in cryptos as investing in start-ups, where 90% of start-ups fail. Does it actually make sense to think of cryptos the same way? If so, then that will impact my rebalancing strategy. Once my position drifts, should I be rebalancing to a benchmark asset like $ETH? Should I chance it with a token that’s been underperforming?
The answer unfortunately isn’t as straightforward.
So Should I Drift?
The answer unfortunately isn’t that clean cut. And the reason is because to sell something, we need to buy something else. That said, most rebalancing strategies will outperform a buy and hold strategy (or a strategy with FULL drift) here in the crypto space. We saw that quite clearly, where the median return of a buy and hold strategy was 116% compare to simply rebalancing to ETH (to be touched on below), which returned 300–400%.
So yes, we can say that we shouldn’t just let portfolios aimless drift. But to rebalance means every time we sell an overweight token, we need to buy up a different token. What should that token be?
Depending on how you rebalance, this will drastically impact whether or not you should encourage drift in your portfolio.
Let’s start by taking a look at one of the more standard rebalancing method that people employ — rebalancing to a benchmark asset like ETH.
Rebalance to ETH:
All we’re doing here is whenever a token drifts to a certain level, we sell out of the extra bit and buy up ETH. We ran hundreds of historical Monte Carlo simulations on 471 random tokens to get a sense of what the most likely results were and what we found was quite interesting (full details can be seen in the footnotes):
We looked at three different rebalancing thresholds: 1.05x, 2.0x, and 2.5x. In other words, for a starting size of 10% in a portfolio, we looked at the impact of rebalancing a token to ETH if it hit 10.5% of portfolio value, 20% of portfolio value, or 25% of portfolio value (and of course, what happened if we didn’t rebalance at all). To make it more akin to a live trading environment, we made sure to add a trading cost of 5%/trade to penalise overtrading.
What you see, interestingly enough, is that as you wait longer to rebalance to ETH, the performance actually drops. In other words — if your rebalancing strategy is to sell tokens to buy ETH, you should actually do it sooner rather than later.
This is a simple and illustrative example, but is one of the core things we focus on here at Fyde. Common principles such as portfolio drift are often overlooked, but can have a sizable impact on your overall return.
One thing that we touched on here is that there are different types of rebalancing as well. For example, should we rebalance to a benchmark token like ETH? Should we rebalance to the most underweight token instead? These are the questions we’ll hit next.
