Learn more / Tax-loss harvesting
Tax-loss harvesting

What is tax-loss harvesting?

Robo-advisory platforms build customized portfolios for users and then monitor and rebalance the portfolios periodically for low and affordable management fees. The low cost and low minimums are due to the use of algorithms that require minimal human involvement. One of the numerous services that some robo-advisors offer through their systems is tax-loss harvesting (“TLH”).

Activating the TLH algorithm is a way to potentially increase your after-tax returns by selling investments that have declined in value and replacing them with similar investments. The realized tax losses can potentially lower your end tax bill, and be reinvested in your investment portfolio.

Interactive Advisors has developed a TLH algorithm that is available to all clients who are invested in our taxable custom Asset Allocation portfolios. Clients who wish to avail of tax-loss harvesting in their taxable accounts may choose to opt in to the TLH program.

Tax-loss harvesting image

What are the benefits of tax-loss harvesting?

The benefits of TLH will vary depending on the unique tax situation of each client. While we expect that the TLH algorithm will add value in general, the value added from TLH will depend on a variety of factors. Among others, these include:

Tax-loss harvesting in Interactive Advisors Asset Allocation portfolios

Performing a tax loss harvest on an ad-hoc basis can be tedious and complicated for the average investor, which is why many robo-advisors have included this value-added strategy as part of their services. Robo-investment platforms have an algorithm in place that incorporates computational rules like the 30-day IRS wash-sale[1] rule and typically applies it on portfolios composed of ETFs. The rationale is that it is easier to find replacements for ETFs than it is for individual stocks that carry idiosyncratic risks.

Whether the TLH algorithm is activated or not, periodic rebalancing ensures that the portfolio is always in balance. During Interactive Advisors quarterly rebalance, trades are done in order to ensure that the portfolio is balanced and at target exposure. By doing so, IA ensures that the portfolio has target risk-return allocation.

In addition to the regular automated rebalancing that applies to all clients, clients who activate TLH may see additional selling, whereby an ETF is sold entirely to harvest losses. After a sale is made, in order to keep the portfolio balanced and maintain exposure to the same risk and return sources, the system will purchase another similar ETF to replace the sold one.

Activating TLH may result in additional trading and therefore, TLH algorithm should only be activated by clients for whom the expected benefit exceeds the expected cost. As such, every investor should examine their income and tax situation before electing it.

Tax-loss harvesting algorithm

During quarterly rebalancing, the TLH algorithm checks to see which ETFs have unrealized net losses. Consider ETF 'A', which has unrealized losses in the custom Asset Allocation portfolio.

The algorithm has logic in place to factor in the 30-day IRS wash-sale rule. If 'A' has been purchased in the last 30 days in the Interactive Advisors account, the TLH algorithm is not invoked, since selling 'A' would result in a (partial) wash-sale. The algorithm then checks to see if there is 'sufficient' loss. Since long-term investments (held for > 365 days) are taxed at a lower rate than short-term investments, harvesting short-term losses is more valuable than long-term losses. The TLH algorithm incorporates this logic. If the loss threshold is met, the position of ETF 'A' is sold entirely and the proceeds are used to buy the replacement ETF 'B'. During periods of high volatility, we may choose to cancel or postpone swapping the ETF 'A' with 'B'. For simplicity, the replacement swap is a complete one, meaning we will never have positions in both the original 'A' and replacement security 'B'. Note that the algorithm may be revised in the future to accommodate partial swaps.

When selecting ETFs for our Asset Allocation products, we consider liquidity, expense ratios, and our assessment of how well the ETF can represent an asset class. When considering replacement ETFs, we consider the same criterion, and also require a high degree of correlation of the replacement ETF with the primary ETF. For example, in order to gain exposure to US Large Cap stocks, the primary ETF we consider is VTI Vanguard Total Stock Market ETF (total assets > 200 billion, expense ratio of 0.03%, tracks CRSP US Total Market Index) while the replacement ETF we consider is IVV iShares Core S&P 500 ETF (total assets > 200 billion, expense ratio 0.03%, tracks S&P 500 Index) which has a correlation of 0.99 with the primary ETF.

Asset Class ETF Replacement ETF
Large Cap VTI Vanguard Total Stock Market ETF IVV iShares Core S&P 500 ETF
Mid cap VO Vanguard Mid-Cap ETF IJH iShares Core S&P Mid-Cap ETF
Small cap VB Vanguard Small-Cap ETF IJR iShares Core S&P Small-Cap ETF
Dividend VYM Vanguard High Dividend Yield ETF SCHD Schwab US Dividend Equity ETF
Energy VDE Vanguard Energy ETF XLE The Energy Select Sector SPDR® Fund
REITS VNQ Vanguard Real Estate ETF SCHH Schwab US REIT ETF
Developed VEA Vanguard FTSE Developed Markets ETF IEFA iShares Core MSCI EAFE ETF
Emerging VWO Vanguard FTSE Emerging Markets ETF IEMG iShares Core MSCI Emerging Markets ETF
US Govt BND Vanguard Total Bond Market ETF AGG iShares Core U.S. Aggregate Bond ETF
Non-US Govt BWX SPDR Bloomberg International Treasury Bond ETF IGOV iShares International Treasury Bond ETF
Corporate Investment grade LQD iShares iBoxx $ Investment Grade Corporate Bond ETF USIG iShares Broad USD Investment Grade Corporate Bond ETF
Corporate High yield USHY iShares Broad USD High Yield Corporate Bond ETF HYG iShares iBoxx $ High Yield Corporate Bond ETF
Municipal Bonds MUB iShares National Muni Bond ETF VTEB Vanguard Tax-Exempt Bond ETF
Inflation protected Bonds SCHP Schwab US TIPS ETF SPIP SPDR® Portfolio TIPS ETF SPIP

While many traditional financial advisors only run a tax-loss harvest once a year due to the time-consuming and labor-intensive process, robo-advisors can run these processes more frequently. The frequency with which the TLH algorithm at Interactive Advisors is activated is typically quarterly; quarterly activation avoids the likelihood of triggering the wash-sale rule, where losses cannot be utilized if you buy the same security 30 days before or after the sale (period of 61 calendar days) in any account, including ones that are not visible to Interactive Advisors. Given that we rebalance our portfolios on a quarterly schedule, we have decided to sync up TLH with our rebalancing schedule since the risk of a wash sale increases when rebalancing trades are on a different schedule from TLH trades.

We feel that pursuing very aggressive TLH where we do not have a threshold for harvesting losses, or where we harvest very frequently, only makes sense when one makes very optimistic assumptions about tax differentials, expected returns, investment horizon and other factors.

Illustrative example to showcase potential value added from TLH

As mentioned above, whether and how much of a benefit TLH provides depends on the investment horizon, expected portfolio return, as well as the difference between the current and future tax rates. We shall try to illustrate this with a simple model.

Consider a $10k investment that experiences a loss of negative 20% in the first year; assume that the investment then grows at 6% annually. Assume that the effective tax rate at the point of harvesting the losses was 40%, and the effective tax rate at withdrawal is 25%.

As we mentioned earlier, the tax rate differential (difference between tax rate when harvesting versus tax rate at final withdrawal date) plays an important role in calculating the benefit of using a TLH algorithm. Consider the following scenarios.

Scenario 1
TLH, harvested taxes reinvested at 6% and tax rate at withdrawal: 30%
Scenario 2
TLH, harvested taxes reinvested at 6% and tax rate at withdrawal: 25%
Scenario 3
TLH, harvested taxes reinvested at 6% and tax rate at withdrawal: 20%
Scenario 4
TLH, harvested taxes reinvested at 6% and tax rate at withdrawal: 40%

The NAV for these over different investment horizons is as follows:

Whereas annualized excess returns in each scenario look like this:

Clearly, the TLH benefit will be different depending on the assumptions. For example, when the tax differential works against you (Scenario 4), the excess returns can be significantly less and even negative.

Looking at the red line (Scenario 2) in the graph above, for a base case scenario where the current tax rate is 40%, future tax rate is 25%, expected return on investment portfolio is 6%, horizon is 10 years, and all realized losses can be fully utilized when filing taxes, the annualized excess returns generated by tax-loss harvesting are around 0.5%.

Appendix
Buyer Beware: The Reality of Tax-Loss Harvesting Benefits
Evaluating The Tax Deferral And Tax Bracket Arbitrage Benefits Of Tax-Loss Harvesting