Moore, Digital Wealth

Robo-advisors have become increasingly popular wealth management services, although they still represent just a sliver of the wealth managed by human advisors. Automated, algorithmic programs can construct portfolios, rebalance them, and even harvest tax losses. They cost a fraction of what a typical financial advisor charges, and they do a better job of asset allocation and fund selection. What, of course, they can’t do is to hold the nervous investor’s hand in times of trouble, although they can send soothing messages.

Simon Moore is the CIO of FutureAdvisor, one of the largest digital asset managers. In Digital Wealth: An Automatic Way to Invest Successfully (Wiley, 2016) he explains the basics of constructing a diversified, cost efficient, tax efficient portfolio using ETFs. And, as might be expected, he argues that financial software is superior to human beings in setting up and managing an investment portfolio.

What should be of interest to all investors who have not committed to a digital service, whether they manage their own money or rely on a financial advisor, is how the FutureAdvisor algorithms go about rebalancing a portfolio. Moore argues that “if and when any rebalancing does occur, it should ideally be combined with other portfolio considerations such as tax efficiency, cash and dividend investment, and consideration of whether the initial fund selection is still valid in the presence of current expense ratios, commissions, and bid-ask spreads.” (p. 123)

There are two basic ways to go about rebalancing: a time-based system and a threshold system. In the former, the portfolio is rebalanced every quarter, let’s say, quite independently of market conditions. Threshold-based rebalancing, by contrast, “makes moves when they are large enough to matter to the portfolio. … “[I]f your portfolio is set up to rebalance once a quarter, then you’ll get a rebalance once a quarter whether the market is virtually flat or experiencing the largest volatility in history. However, in those same environments threshold-based rebalancing would avoid rebalancing in a virtually flat market and potentially balance more than once during a period of high volatility.” (pp. 123-24)

FutureAdvisor goes a step further, using a tiered rebalancing approach. “[I]t appears optimal to use threshold-based rebalancing but to implement it in a tiered fashion so that trades are made only when there is a deviation that matters at the level of portfolio construction. However, the need for rebalancing also interacts with other features of the algorithm in minimizing trading costs, so often rebalancing trades is combined with other trading goals, resulting in less portfolio turnover. For example, sometimes rebalancing can effectively be done for free if a tax loss harvesting trade is occurring or there is sufficient cash in the account to be invested and the algorithm can see those opportunities and take advantage of them.” (p. 125)

I have no doubt that wealth management algorithms will become omnipresent, whether as stand-alone products or as must-have tools for human financial advisors. It’s time for investors to understand them better and decide what role they should play in their own long-term wealth management.