Tax Loss Harvesting Is Overrated
Mar. 16, 2015 8:47 AM
- Robo-advisors are exaggerating the benefits of tax harvesting.
- It mostly benefits equity-heavy portfolios.
- Majority of tax alpha comes from big down years like 2001-2002 and 2008.
After reviewing the literature on tax efficiency and loss harvesting, we find that a well-diversified and risk managed portfolio implicitly promotes tax efficiency by reducing the number of big down years, and as a result, minimizes the amount of tax available to be harvested. We also find that claims by robo-advisors of a 1-3% annual gains from tax harvesting are overestimated because of 1) bias in the timeframe, 2) bias in the portfolio composition, 3) tax deferrals misrepresented as gains, and 4) exclusion of operational costs from the analysis. Actual gains from loss harvesting may not be economically significant or may even be negative. Therefore, we recommend advisors to harvest only the largest and most obvious losses at year-end by employing a switch-back strategy, and only if the gains from harvesting can significantly offset the opportunity cost of switching.
Robo-advisors like Betterment claim that tax loss harvesting can add as much as 1-3% in annual returns to an investor's portfolio. However, these claims are based on an equity-heavy portfolio (70% equity) that has experienced two large bear markets. Based on Betterment's own figure (See Chart: Annual Tax Offsets with Tax Loss Harvesting+ ), harvesting tax losses in most years produces little to no alpha. Most of the tax alpha comes from big down years in 2000-2002 and 2008-2009 - the amount is small even in a volatile year like 2011. A diversified and risk managed portfolio is not driven by any single market and tends to avoid the big down years in the first place, leaving little remaining tax alpha to be harvested.
Kitces (2014) also shows that tax loss harvesting doesn't produce any net tax alpha, but a tax deferral. This is because harvesting a loss generates current tax savings, but also reduces the cost basis of the investment, triggering a potential gain in the future that may offset most or all of the loss harvesting benefit. A tax alpha of 6.43%
is only able to leave investors with a 0.44% after-tax deferral value.
The remaining deferral value from loss harvesting may be more than offset by operational costs. The existence of the wash-sale rule introduces significant business overhead to tax loss harvesting. Getting around the rule requires either switching back-and-forth between a primary and alternate ETF for each asset type or holding two similar ETFs in parallel at all times (also not all asset types have more than one ETF). The extra holdings introduce operational hurdles including 1) higher transaction costs to the portfolio, 2) higher account minimums to achieve a fully invested portfolio and, 3) present unnecessary complexity to a portfolio around the time when capital is contributed/withdrawn from the account.
Tax alpha mostly comes from harvesting losses for equity-heavy portfolios, and the amount is only economically significant in big down years. A diversified, risk managed portfolio enjoys little to no tax alpha and may have a negative effect when business overhead costs are factored in. The advisor can choose to harvest the obvious losses at year's end by employing a switch-back strategy to avoid the wash-sale rule - i.e. selling the positions with large losses and buying a highly correlated ETF, if one exists (and switching back to the original ETF after 30 days). However, the alternate ETF may have a higher management fee or tracking error so one should only harvest losses if the tax alpha more than offsets the opportunity cost of switching.
Kitces, Michael (2014). Evaluating The Tax Deferral And Tax Bracket Arbitrage Benefits Of Tax Loss Harvesting.
Kitces, Michael (2013). Is Capital Loss Harvesting Overvalued.
Kashner, Elisabeth (2014). Inside Robo Advisor Tax Loss Harvesting.
Betterment (2014). White Paper Tax Loss Harvesting+.
Susko, P. M. (2003). Turnover Rates and After Tax Returns. The Journal of Wealth Management, 6(3), 47-60.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.