back to Home Page
Client-Only Articles
Email:
Password:
 

 

Reviewed February 2021

TAX LOSS HARVESTING

Abstract:  Tax loss harvesting may be of value when harvested proceeds are not reinvested or are reinvested later in the same asset class and never sold before an investor dies.  When a fund is harvested to realize tax losses, then repurchased, that sets the cost basis on the fund or its equivalent lower and if it is sold at some point in the future the additional tax paid will equal the tax saving initially created and could be greater if tax rates rise.  As of 2020 tax rates are historically low.  We recommend letting fund managers manage tax issues internally within the funds they manage and DFA offers several funds with that specific objective.


Every Fall in the fourth quarter of the year the financial press and industry begin promoting tax loss harvesting as a way to enhance investment returns and add "tax alpha", usually as if it it were a new insight.  Harvested capital losses can be used against realized capital gains in the same year or carried forward into subsequent years.  Both active and passive managers use the strategy with some claiming to add 1% or more to annual returns in tax alpha.  As is the case with much financial advice advisor promises and investor outcomes may differ.

On the surface tax loss harvesting makes sense.  Seldom mentioned though is that tax loss harvesting can only be certain to add value when the harvested proceeds from the loss are never again reinvested in the asset class from which the harvest was taken or the owner is deceased and the cost basis has been reset at death.  This isn't usually the way it's done because most investors will reinvest the proceeds in the same asset class and fund, usually after waiting 31 days.  If this replacement starts at a lower cost basis after harvesting and more taxes will be due on it in the future if it is ever sold.  If the price of the harvested asset has climbed before it is repurchased then the gain in value before repurchase will reduce the potential profit versus not harvesting.

Typically, harvested proceeds are immediately reinvested into a similar asset class or reinvested 31 days later in the same asset class in order to avoid violation of the "wash" rule.  Harvesting doesn't generally lower taxes longer term because it lowers the cost basis on the asset and recaptures taxes later when it is sold. It does defer taxes.  Assuming that at some future time the asset will be sold for a profit and capital gain and that future capital gains rates are the same as today's the future gain on which a tax is computed will be increased by an amount exactly equal to the tax originally saved during the initial tax loss harvesting.  If capital gains tax rates have increased, and they are low historically as of 2019, an investor may end up paying more in taxes then if they had not initially done tax loss harvesting.

As an example of harvesting, let's assume we hold $1 million in DFA's passive equity funds after they have declined $500,000 from an original purchase price of $1.5 million.  We'll omit State capital gains taxes since they vary greatly from state to state and don't alter the principles underlying this analysis.  We decide to harvest $500,000 in losses, saving $100,000 by tax deferral and taxes not paid against anything sold for a gain of $500,000.  The tax loss can be carried forward in time indefinitely and assumes the Federal government's current 20% capital gains tax rate.  In a state like California with high capital gains taxes an investor taxed in the maximum Federal and California brackets, about 33% combined, will defer and save about $165,000.  However, in the example here if we decide to sell our equity positions that replace the harvested losses at some point in the future after they have risen $500,000 and are again worth $1.5 million, our cost basis is not $1,500,000 but $1,000,000 and we owe more in taxes if they are sold.  We now have an additional $500,000 in capital gains taxed that we wouldn't have had if we had not initially harvested $500,000 of losses in the portfolio.  Our tax loss "harvests" have been exactly offset by additional capital gains when the replacement assets are sold sometime later since the basis has been reset lower by the same amount of our loss.  And, that assumes taxes will not rise in the future.

Many investors will probably sell most if not all of their equities in taxable accounts before they die.  Of course, if the replacement assets are never sold and the original owner dies and the cost basis is reset at date of death heirs might benefit from avoiding taxes on the asset if the price has risen.  There is little certainty regarding the outcome of harvesting with one exception.  Since capital losses may also be applied against up to $3000 of income in any given year and since the maximum current Federal income tax bracket is 37% in 2019 and Federal capital gains taxes are 15%, 20% or 23.8%, the advantage for most investors will be 20% on $3000 or $600 per year, a very small reduction in income taxes for most investors.  Of course, again assuming that at some future time gains are realized in taxable accounts, we will have used up capital losses against $3000 of income per year rather than having them available to offset future capital gains.  Investors will have gained little, if any, advantage from tax loss harvesting.

Tax loss harvesting presents other drawbacks.  It incurs transaction costs.  If the proceeds are invested immediately the similar but not identical replacement fund may not track exactly with the harvested fund and may underperform the fund that was harvested.  If the proceeds from the harvest are kept out of the market for 31 days before being put into an identical fund prices can move upwards and the replacement fund will have lost some return versus the harvested fund due to an increased cost basis at the repurchase.  Another issue is tax code.  EAM has spoken with various IRS representatives, accountants, and tax attorneys and been told that the IRS makes it clear that the sale of anything solely for the generation of a tax loss to be used against a future tax increase is questionable.  This topic is rarely discussed in the financial community.  We are not aware of any of our clients having had their tax loss harvesting questioned by the IRS but in passive portfolios this could become an issue because it's hard to make the case that a position was sold if the explicit purpose of a passive portfolio is to practice a long term "buy and hold" investment strategy in a passive and index portfolio.  And, future capital gains tax rates may well be higher than today's historically low rates.  A Federal capital gains tax rate of 35% to 40% has been the historical average.  On the plus side tax loss harvesting produces extra capital today that might be reinvested in some other asset that might grow over time more than if an investor had not harvested tax losses and kept the money in the original assets.

In February 2018 Dave Twardowski, Ph.D. at DFA put together an analysis for EAM on tax loss harvesting and reached similar conclusions to those above.  Tax harvesting does not necessarily add "tax alpha", additional return, and is really tax deferral.  If the money deferred from paying taxes by tax harvesting is invested in something other than the harvested investment the advantage of the harvest will come only if that new asset outperforms the harvested asset.  However, typically the harvested asset is quickly replaced with a similar or the same asset after 31 days in passive portfolios.  DFA ran hypothetical harvested portfolios for various time frames and found that when returns are positive, the benefit of periodically harvesting losses declines dramatically over time in a portfolio that does not receive fresh cash flows.  DFA found that taxable gains could be substantial in a portfolio that has been harvesting losses over time.  DFA concluded that under certain circumstances harvesting can defer taxes and allow tax deferred growth on a portion of the portfolio but that it is only one aspect of tax management and is by no means certain to be of benefit to investors.  Changes in future tax rates, future interest rates, and the unpredictable lowered cost bases and future growth in the asset that is a replacement for a harvested asset all determine whether the harvesting adds or subtracts value and can't be known in advance.

Another factor investors should consider in deciding to tax loss harvest or not is that passive and index portfolios have inherently low yearly capital gains pass throughs because portfolio turnover is inherently very low.  DFA's equity portfolios generally pass through between 0% and 2% of total share price per year as cap gains versus about 15% in active portfolios and sometimes that can over 50%.  DFA also offers tax-managed and tax-advantaged portfolios where losses and gains are offset internally within the funds, not by investors selling and rebuying a fund.  DFA offers core/vector portfolios where stocks are allowed to move between asset classes and capital gains are very infrequently realized and sometimes never realized in a given year.  We recommend letting DFA manage taxes within their funds rather than doing so by investor initiated tax loss harvesting.  We recommend avoiding tax loss harvesting unless an investor is unlikely to ever sell the equities they have repurchased with the proceeds from the harvesting.  Our philosophy of portfolio management is "candid but never dogmatic" and the benefits of tax loss harvesting are uncertain.  EAM will tax loss harvest whenever clients request and in Fall quarter of 2008 after large declines in almost all asset classes we did extensive tax loss harvesting for clients.

Site Navigation