As usual, we have to first ask ourselves if it’s actually useful to rebalance a portfolio. I always try to be as impartial as possible when thinking of these topics, I intend to state both logical sides.
There are a few things in favor of rebalancing. Mainly, periodically (monthly, quarterly, daily, yearly) rebalancing a portfolio ensures that the owner is in complete comfort and calm (allegedly). Secondly, it removes emotions from the process. Rebalancing goes in line with the investment plan designed (in a rational state) and setting it in advance makes your future self avoid acting based on momentary irrationality.
To the contrary, rebalancing also counts with some disadvantages I notice. Winners will continuously tend to exceed the threshold established and rebalancing would imply us to sell that excess and add it to companies that have gone below their desired weight. Consequently, in many scenarios, it can turn to a potential “trimming winners and adding to losers” strategy. Rebalancing assumes the weights established are ideal, because one would implicitly be reallocating capital based on these arbitrarily selected weights, leaving no room for mistakes in the weight selection process.
It's important to keep in mind most (I think) professional portfolio managers include in their portfolio several asset classes (bonds, stocks, currencies, etc). They also, in most occasions, build clients’ portfolios by assigning weights to each particular asset class and get exposure to it through diversified vehicles. The weight assignment is done with a particular ‘risk objective’ (standard deviation objective). If the investment policy agreed by portfolio manager and client establishes this risk objective, it’s appropriate for PMs to rebalance portfolios.
Since most portfolio managers follow an asset (in the actual sense) allocation approach, it seems correct for them to rebalance. They do not have intrinsic or fundamental single position risk. However, if an investor has a portfolio fully built with equities, he/she does have a unique position risk (not under volatility parameters). This risk literally depends on the fundamentals of the different and unique companies. By rebalancing, we would be de-estimating such intrinsic characteristics.