How I Think About Portfolio Rebalancing
Jake Hallac (@jakeh) wrote a great memo about portfolio rebalancing that I wrote an embarrassingly long response to in a comment. This is worth thinking about.

Why should you rebalance your portfolio? Jake gives a good answer:

"Empirical evidence has shown that disciplined rebalancing tends to reduce risk and enhance return."

This is true- but the problem is the use of the word risk. When reading papers on portfolio theory, risk is almost always measured by a thing called beta.

Beta is a statistical measure of the volatility of a stock versus the overall market.

Volatility gets used as a proxy for risk in studies because academics need a number for their calculations that is objective, can be looked at historically, and extrapolated into the future.

But just because it's easy to measure doesn't mean it's telling you the right thing.

I would argue that volatility is not the same thing as risk.

In my opinion, risk is the permanent loss of capital. That's what I really care about.

Post media

Think about this: A stock that goes from $60 to $90 is likely to have the same statistical volatility as a stock that goes from $60 to $30. But which was riskier? "Empirical evidence" would say that they were both equally risky, but we all know that it's important to describe these two situations as being very different. I don't care if my stock goes up and down a lot, as long as it ends up at $90 instead of $30.

The risk is that it ends up at $30, not the path it took to get there.

Therefore, I tend to think focusing on portfolio rebalancing as asking the wrong question. Portfolio rebalancing will lead you to sell companies that have been outperforming (which is likely due to excellent execution) and buy companies that are underperforming. This is a bit like benching your best basketball player because you don't want to score too many points all at once.

I may be guilty of 'anchoring bias' due to the fact that I've never rebalanced my portfolio in seven years and that has worked out really well for me- so take my thoughts with a grain of salt. But it's worth thinking about for yourself if you're ok with more volatility in pursuit if higher returns.
Ambrose's avatar
I agree with @nathanworden. Volatility just means how much the price change. It's a double-edged sword. It is up to the person to decide whether change is risky. I view higher volatility better as it increases my chances of more gain

I remembered talking about how diversity decreases volatility. To me, low volatility itself is a risk. Your gains will typically mimic an ETF and 1 bad event or company is going to ruin your entire gains.

So is higher volatility better?
I do not think the correct question to ask what should your portfolio volatility be. Since it is dependent on the type of companies you invest, smaller cap stocks will probably have higher volatility than larger caps.

To me, volatility does not matter as long as I know the company I invest in is good. Does it really matter that $TSLA drops or jumps 10% if you know it is going to be a trillion-dollar company?
Volatility is such a short-term take on investing. Buy-sell buy-sell just because it is volatile? No one is an oracle. Higher can go higher, lower can go lower. Rebalancing because your stock gains are too high and risky is probably the wrong reason to rebalance your portfolio.

Like what MF and I said previously, winners keep winning. Instead of selling the stocks that are performing, the best stocks to buy are the ones you already own.
Nathan Worden's avatar
💯 Very well put Ambrose. @zebo

I love this line you said: “To me, low volatility is itself a risk”

^More people should be coached to understand this.
Jake Hallac's avatar
Yes - I like this @nathanworden

First, I think it’s v important to differentiate between risk (permanent loss of capital) and uncertainty (vol). This took me to a while to internalize.

Second, if you’re a bottoms up stock picker, I agree it makes sense to let winners ride. This is the only way you’ll get to benefit from the snowball effect of compounding returns.

That said, I still think portfolio rebalancing makes sense in certain contexts: (I) you only invest in factor-based etfs and you don’t want to be overly exposed to a given factor, (ii) you invest across asset classes (e.g., DM stocks, emerging market bonds) and based on your long term outlook, you decide you need more/less exposure to a given asset class, (iii) you’re managing someone else’s money and your factor/asset class exposure goes outside of your previously agreed upon exposure ranges.

My recommendation for the fundamental analyst with a stock portfolio would be to have a quarterly “checkpoint” where you reassess your conviction level across all your positions (e.g., has my thesis changed?). Quarterly is a good cadence given earnings release timing.

In this context, “rebalancing” is less formulaic, but more dependent on your bottoms up research. It could even result in you adding to your highest conviction names.

  • Jake
Nathan Worden's avatar
I love the nuance and thought that goes into this- yep, I think the contexts you outlined do make sense for portfolio rebalancing, especially the factor risk strategy where your goal is to not be overly exposed to any one given factor. I am wanting to write more about that topic because I'm a big fan of the work Fama and French have done.
Jake Hallac's avatar
On that last note, before initiating a position, my PM would have me list out the 3-5 likely scenarios/events/changing metrics that would impact my conviction level in a a position.

I find thinking through these “signposts” ahead of time very useful because you’ll be more equipped to effectively “rebalance” when things change.