ETF Tax Loss Harvesting is Decadent and Depraved

Recently I’ve been listening to the Tim Ferriss podcast on my commute. It’s a mix between tech, entrepreneurship, and lax bro. The podcast is free and so listeners get 5 minutes of very well manicured sponsored Ads at the beginning of each episode, including one for Wealthfront, the ETF Robo Advisor.

Wealthfront is an automated investment service which charges low management fees to invest in ETFs for you according to a strategy based on Modern Portfolio Theory. It also happens to be a FinTech startup approaching a $1 Billion Dollar valuation. In the podcast ad, Wealthfront singled out it’s Tax-Loss Harvesting strategy to show why “the value of Wealthfront is in the automation of habits and strategies that investors SHOULD be using on a regular basis but normally aren’t.” Tax-Loss Harvesting is the very first strategy they mention in the ad. And I thought to myself, well that’s a load of bullshit.

You see, there’s a difference between the tax-loss harvesting in a textbook and the shady ETF Portfolio Tax Loss Harvesting that many Robo Advisors employ to convince customers that their plans outperform the market. The textbook tax-loss harvesting strategy can allow investors with an extended investment time horizon to write off some taxable income or reduce future tax burden by harvesting the capital losses of securities in a portfolio that are down in a tax year. ETF Portfolio Tax-Loss Harvesting, on the other hand, uses the opaqueness of an ETF wrapper and a lack of regulatory enforcement to take a sensible strategy and bastardize it. While Wealthfront claims to democratize investing, this practice only benefits those wealthy enough to afford the service, as Wealthfront uses ETF Portfolio Tax-Loss Harvesting to upsell their higher net worth clients into higher fee investment plans.

So what is Tax Loss Harvesting? Well, the textbook Tax-Loss Harvesting strategy generates capital losses by selling that year’s negative performers and putting the sale proceeds into a highly correlated, but different, security. So an owner of Exxon Mobil in a poor year might decide to harvest those losses by selling Exxon ($XOM) and purchasing British Petroleum ($BP) for a few months to avoid wash sale rules. Pretty straightforward example, an investor can potentially reap a small tax benefit by incurring some market risk. After all, XOM and BP are two different stocks with different management teams, different earnings, and different fundamentals. Sure, they’re correlated, but investors still bear the risk on 100% of the difference between XOM and BP. Straightforward enough.

Before I compare the above with an ETF Portfolio Tax Loss Harvesting strategy, I want to touch briefly on wash sale rules. Above I mentioned that an investor would have to own the different security for ‘a few months’ to avoid wash sales. Wash Sale rules were put in place many years ago after investors became aware of the potential for harvesting capital losses and would sell a stock and then immediately buy it back. So Wash Sale rules were created which basically force investors to not hold a ‘substantially identical’ security for 60 days, otherwise the capital loss is disallowed. But what does ‘substantially identical’ really mean? It basically means that you can’t buy an ETF based on the exact same index, and you can’t buy a derivative position in the same security, so you can’t sell 100 shares of a stock and then just buy 1 deep in the money call option, for example. But frankly, what it really means is that it’s open to interpretation by the IRS and they put the burden on the individual to stay afoul of the having their capital loss disallowed.

Wealthfront and other Robo advisors go to extreme lengths to get as close to substantially identical as possible without waking up the IRS, and they even advertise the funds they do it with. So let’s look at those funds.

To illustrate, let’s check out their Developed Market ETF choice. Their ‘Primary’ Developed Mkt ETF is Vanguard’s VEA, which tracks the FTSE Developed Markets Ex-North America Index. Their ‘Secondary’ Developed Mkt ETF which they use to maintain Developed Market equity exposure while harvesting a capital loss without triggering wash sale rules is Schwab’s SCHF, which tracks the FTSE Developed Markets Ex-US Index. Same index provider, same geographic focus, nearly the same expense ratio, and a stated 99% correlation, with the only difference being one fund doesn’t invest in Canada, and the other does (Neither the US nor Mexico are in either fund). So that leads to the logical question of, well, just how much of this fund’s index is invested in Canada to cause it to be not ‘substantially identical’? Let’s take a look:

Only 7.36% of the fund’s index is invested in Canada, so yes, it’s not identical. But the other 20+ countries are not only exactly the same, but because it’s the same index provider, they hold exactly the same securities right down to the stock ticker, number of stocks per country, and weightings of those stocks (adjusted to account for the Ex-North America fund not holding Canadian stocks). To wrap that all up, even though over 92.5% of the stocks are literally the definition of substantially identical, that 7.36% non-substantially identical difference allows Wealthfront to avoid a wash sale and incur a capital loss on 100% of the portfolio. The Fintech revolution, indeed.

Look, this is just one example. You can go through all the other ETFs they hold, and compare the difference between their Primary and Secondary ETFs by looking at the fund’s Portfolio Composition Files. It’s all publicly available. I haven’t and so maybe my numbers are off by a little, and maybe some examples are less egregious than others, but the basic fact remains that they’ve put a nice fancy wrapper called “Fintech” around what’s essentially a cat and mouse game with the IRS, and called it innovation.

Maybe the advertisement I heard on the Tim Ferriss show isn’t bullshit. Maybe this sort’ve thing is what investors “should” be doing, and I’m the sucker for finding it to be an abuse of the system. But I’d like to think that there’s more to financial innovation than some securities lawyers taking their public sector buddies out to dinner and asking them what they can get away with.

Anyway, that’s all I got. I appreciate anyone taking the time to read. Would love any feedback and happy to chat, and overall I think Robo Advisors are doing good things. I just hope they get rid of the bullshit.

Thanks for reading,

Eric Mustin