@tartufo:
Well, my intuition would say that the rebalancing premium is very small if all the assets in the portfolio have high correlation to each other (as would be the case if the portfolio is 99% stocks). Would you agree?
It would indeed work better with negatively or uncorrelated assets. In two out of three portfolios I added the maximum amount of gold I could (7% I believe) to achieve that. I’ve switched to 1% cash for the time being, will give this a spin until April and then consider a higher cash allocation.
From the best paper I’ve read so far on the subject [1]:
We have established the surprising result that these strategies generate port-
folio growth rates in excess of the individual asset growth rates, provided
some volatility is present. As a consequence, even if the growth rates of the
individual securities all have mean zero, the value of a fixed-mix portfolio
tends to infinity with probability one
I also found your old post on Why we rebalance and other tales about volatility. Great stuff, I’ve been reading Hollerbach / Abdelmessih / Eifert for years and may or may not have got an inspiration or three from The Great Age of Rebalancing Begins… Birds of a feather, and all that.
So I’ll leave you with a related paper from the absolute goats MacLean, Ziemba & Li [2] that may be a little more immediately useful for financial independence & early retirement than pumping vol in 3a pillars you cannot reach for some decades:
A dynamic Bayesian fractional Kelly strategy, where the investor
rebalances the portfolio based on the performance to date, is shown to be optimal
assuming that the risky assets are jointly lognormally distributed. The strategy
minimizes the expected time to reach a wealth goal while maintaining a high
probability of reaching that goal before falling to a subsistence level of wealth.
[1] Dempster, Evstigneev, Schenk-Hoppé: Growing Wealth with Fixed-Mix Strategies (2009)
[2] Maclean, Ziemba, Li: Time to wealth goals in capital accumulation (2005)