Let the Robots Do Your Taxes
A while ago, I wrote a piece about the opportunities and challenges facing the fast-growing “robo-adviser” industry (which, by the way, really should be called “robo-managers”; the logic of industry buzzwords eludes me).
In that piece, I claimed that the main advantage of robo-advisers was behavioral. By preventing people from trying to time the market and pick stocks, robo-advisers can bring individual investor performance up to the market average -- in other words, a huge improvement. The only problem, I said, was that this improvement might be a tough sell. If investors were smart enough to use a robo-adviser, I asked, might they not eventually get smart enough to simply invest their money themselves, without paying 0.25 percent of their wealth each year (the fee charged by Wealthfront, the industry leader)?
I still think that robo-advisers will create huge benefits for retail investors by cutting out middlemen and by reducing behavioral biases. But there was one important part of the robo equation that my earlier piece overlooked.
That piece is good old Uncle Sam. Taxes, it turns out, are a far trickier beast than passive asset allocation. A beast so tricky that only a robot might be able to defeat it.
Those who work in the finance industry, of course, will already know what tax-loss harvesting is. For the uninitiated, it’s the ability to delay the taxes that you pay on your investments.
When you sell an investment that went up in value from when you bought it, you owe capital gains taxes. As long as investments generally go up over time, you’d rather pay taxes later instead of now -- if you can delay your taxes, the money you would have paid in taxes stays invested, and grows over time.
This is a problem for investors, who have a number of reasons to sell pieces of their investments in any given year. For instance, suppose you want to rebalance your stock portfolio. You want to keep 60 percent of your money in U.S. stocks and 40 percent in foreign stocks, but since U.S. stocks went down this year and foreign stocks went up, you need to sell some of your foreign stocks and buy U.S. stocks to maintain that 60-40 ratio.
When you sell the foreign stocks, you’re going to pay capital gains tax. But you can delay that tax with a technique called tax-loss harvesting. Basically, you find some stocks in your portfolio that went down -- and since stocks are volatile, there will almost always be some of those -- and you sell them. This gives you a tax loss, which you can use to reduce or even eliminate the capital gain that you would have had to pay taxes on.
Now, selling those losing stocks leaves you with a chunk of cash. You would like to turn right around and buy back the stocks you sold, so that you can restore your old allocation. But the Internal Revenue Service’s “wash sale rule” prohibits you from doing that.
But there’s a workaround. Instead of buying back the stocks you sold, you buy back some combination of stocks that mimics the stocks you sold. Voila -- you have an allocation essentially the same as your old one, and now you don’t owe taxes this year.
Of course, you will eventually have to pay taxes on the stock you bought, assuming it goes up. But that’s far in the future. In the meantime, the money you didn’t pay in taxes this year will be compounding itself.
That’s one way to game the tax system. There’s another. The IRS lets you use losses on your investments to offset as much as $3,000 of ordinary income each year. Ordinary income is taxed at a much higher rate than capital gains, so tax-loss selling helps you dodge a bit of income taxes each year. You can even carry the tax loss forward from year to year, reducing your income tax each year!
If this sounds too complicated for the average investor to do by hand, realize that it’s even more effective if you do it every day instead of every year. Because stocks are volatile, doing daily tax-loss harvesting gives you a lot more opportunities for finding those useful losing stocks.
Naturally, this would be a nightmare for a human. But it’s no problem for a robot.
And this is where the final big advantage of robo-advisers really makes itself felt. The algorithms to pick a diversified portfolio of exchange-traded fund and index funds -- the basics of investing -- can be written in an afternoon. The algorithms to minimize the amount of taxes with daily tax-loss harvesting are orders of magnitude harder. Of course, tax-loss harvesting isn't without risks -- for example, the possibility of being in a higher capital gains tax bracket when you eventually do sell your stock. It takes sophisticated algorithms to minimize that risk.
This is a big reason why robo-advisers really are technology companies, not just service companies. By staying ahead of the curve with advanced tax-minimizing algorithms, robo-adviser companies such as Wealthfront and Betterment can beat the average investor without beating the market. Wealthfront estimates that its tax-minimization algorithms can increase investors’ returns by almost 2 percent a year. That’s quite a haul over the long run. Even if the true number were only half that, it would still be more than enough to justify the company’s 0.25 percent annual fee.
In other words, expect the robo-adviser -- or robo-manager -- business to grow. I predict it won’t be long before every asset manager adopts tax-loss harvesting algorithms similar to the ones these companies are developing. By the time, they do, though, the money management world may well belong to the robots.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Noah Smith at email@example.com
To contact the editor on this story:
James Greiff at firstname.lastname@example.org