• With Tesla (TSLA) stock shooting to the sky recently, I had an interesting debate on Twitter with @GerberKawasaki and @SamTalksTesla as to whether stock holders should trim or hold.
• @GerberKawasaki and @SamTalksTesla are successful money managers who are required to take a conservative view in the interests of their clients.
• Individual investors do not have these constraints, so I decided to investigate whether rebalancing would improve or impair performance for the average growth investor.

Experience tells us that in a portfolio of growth stocks, some will do very well, with, perhaps, 30% plus returns, whereas others will be total duds returning zero or less. To understand the effect of rebalancing we do not need to know which stocks will be winners, we just need to assume that some will be.

I decided to start my exercise with a simple model. I constructed a representative growth portfolio with one 30% grower, two 20% growers, two 15% growers, two 10% growers, two 5% growers and one 0% grower. Bear in mind that in a real situation all of these would have been picked in the hope of high growth, but realistically, we know that with growth stocks only a few reach their full potential.

Under this simple model, the portfolio with no rebalancing returns 28% more than the rebalanced portfolio over 10 years. The problem with this model is that it assumes constant compound growth each year, but in practice stock values are volatile, and the values of growth stocks can be excessively volatile.

To extend the model I decided to add a random fluctuation to the stock price. If we take the ‘on track’ price of a stock with compound growth, at any one time the actual price will usually be either higher or lower than this price.

Adjusting each stock price by a random amount for a single calculation of the portfolio value would be misleading, as one or two outlier values will produce an unrepresentative result. To counteract this I carried out a ‘Monte Carlo’ analysis. Under this scenario the calculation is carried out many thousands of times, and on each iteration a new offset is randomly allocated. (I adjusted the value using a normal/Gaussian distribution specifying the offset in terms of one standard deviation.)

The analysis threw up some interesting results. If we allow the stock prices to deviate from the ‘on track’ prices using a 20% offset (representing one standard deviation) the unbalanced portfolio outperforms the rebalanced portfolio by 6% over ten years. However, if we increase the volatility allowing stock prices to deviate by an amount corresponding to 40% (at one standard deviation), the rebalanced portfolio outperforms the static portfolio by 9%. This was a surprise to me as I did not expect a circumstance where the rebalanced portfolio would outperform the static ‘buy and hold’ strategy.

From this we can conclude that under extreme volatility a rebalanced portfolio can outperform a static portfolio.

Before moving on to the next stage of my analysis, I decided to make one further adjustment. Stocks that grow slowly are usually less volatile. So I set the volatility offset to 30% for the 30% grower, 15% for the 15% and 20% growers, and 10% for the 0% and 10% growers. Under these circumstances, the static portfolio reasserted itself, outperforming the rebalanced portfolio by 26%.

At this stage I was becoming confident that under most circumstances, rebalancing a growth stock portfolio will give lower returns than simply letting your winners run. But theory is not reality and relies on a number of assumptions. So I decided to carry out one last exercise.

I took a random portfolio of growth stocks that I have invested in over the last ten years and decided to carry out a backtest, calculating the returns that would have been achieved using both a simple buy and hold strategy, and also a rebalancing strategy. In order to make sure that the portfolio was fairly representative of real performance I deliberately included a couple of duds (Westport Innovations and Clean Energy Fuels). I tracked the performance from 2010 to 2020, rebalancing annually taking into account the stock price on 1st January.

On the backtest, the static portfolio outperforms the rebalanced portfolio by 16% over ten years.

On that basis I think it is fair to conclude that under most circumstances, for a growth stock portfolio, a simple buy and hold strategy will outperform a rebalancing strategy.