The Tax Harvesting Oasis

The Tax Harvesting Oasis

A Response to Michael Edesess

September 23, 2014

by Daniel Egan and Boris Khentov

PDF Print Email Reminder Share

Previous 1 (current) 2 3 4 Next

This article is in response to Michael Edesess’ article, The Tax Harvesting Mirage, which appeared Aug. 12. This article is also part of an ongoing conversation about Edesess’ article on APViewpoint, which you can view here. If you are not a member of APViewpoint, you can join here.

Michael Edesess’ response to this article appears here as well as in the APViewpoint conversation.

Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.

What is the value of tax-loss harvesting (TLH) to individual investors? Automated investment services have democratized investing strategies that were once the exclusive domain of highly sophisticated investors and planners. TLH is a prime example of such a strategy, and with greater exposure comes greater awareness among the investing public.

The automated strategy itself can be complex enough – far more so is any attempt to produce a reasonable estimate of its long-term value. Any such analysis is hugely dependent on a sizable list of assumptions with respect to general market conditions over the time period, investor-specific cash flows and individual tax circumstances extrinsic to the portfolio. Varying just one of these assumptions could dramatically alter the estimate.

It is therefore not surprising that attempts to pin down a set of reasonable assumptions have been met with alternate analyses that reach different results. In a recent article, poetically titled “The Tax Harvesting Mirage,” Michael Edesess referenced our firm’s Betterment TLH+ service and the performance estimate we have published (77 bps annually) and attempted to estimate the value of TLH on his own. He determines the benefit (14-17 bps annually) to be notably lower.

For reference, peer-reviewed studies found higher figures than the ones Edesess reported. Arnott, Berkin and Ye found a 50-basis-point post-liquidation ‘tax alpha’ from tax-loss harvesting over a 25-year Monte Carlo simulated period.¹ Smith and Smith found a 37-basis-point premium from loss harvesting combined with rebalancing over a 40-year period.²

Edesess’ effort is admirable, and he clearly understands many of the nuances of long-term TLH strategies. However, as a starting point, he assumed an algorithm far less sophisticated than the one powering our TLH+ service (which our white paper describes in detail). Then, in running his own analysis, he diverged significantly from the assumptions made in ours, with no discussion as to why his are preferable.

Edesess suggested that purveyors of automated TLH services gravitate towards assumptions that overestimate the benefit of TLH. We would argue that he did the opposite – he failed to factor in a number of circumstances that we believe apply to a majority of investors. In doing so, he understated the potential benefit of TLH+.

His piece, and others like it, yielded lively discussion amongst advisors and investors on the value of automated TLH, in light of its higher profile. We took great care to publicly disclose how we arrived at our estimates. We would like to take the opportunity to highlight where our assumptions differ from Edesess’ and why we believe ours are appropriate.