[zeroer]

I'm really talking about what happens after you've been invested with them for over a decade. Wealthfront has not yet existed for 10 years, so I know you haven't had your money invested with them for that long. The fact that tax loss harvesting can be beneficial is not in question.

Once you tax loss harvest once, you lower your cost basis on the investment to less than you originally paid. You can only subsequently tax loss harvest on the same security to the extent that the value of the investment is lower than your new lower cost basis. This will become harder as time goes on and you have previously tax loss harvested many securities.

Their white papers all use a timeframe of 10 years to show that Wealthfront is cost effective. I'm pretty sure they don't want customers thinking through all the implications of longer investment time horizons.

[hesdeadjim]

Oh gotcha with the cost basis raising over time. I wonder what strategies you could use to mitigate that. Perhaps buying a similar asset, holding that instead of the other, wait for the original to go down, and then re-purchase. Seems fragile and risky of course...

[zeroer]

> Perhaps buying a similar asset, holding that instead of the other, wait for the original to go down

If you're presupposing the assets are similar, this is unlikely to happen to any significant degree.

The standard way to increase your odds of being able to tax loss harvest is to own as many different uncorrelated securities as possible. You can take this to mean a fund per industry (as Betterment and Wealthfront do) or even to the extreme of only owning individual companies. That way, some are up and some are down, and you can TLH. The more slices you divide your portfolio up into, the longer you'll be able to do it. But I think realistically (especially factoring in inflation), this strategy will stop giving results before 15 years.

[hesdeadjim]

What about the effect of a (let's be honest, inevitable) market crash? You'd be able to realize quite a bit of loss as that is happening by selling assets. Then as the market recovers and assuming those assets are actually worth more than the crash-adjusted value, you would be able to harvest losses again on that asset.

[zeroer]

Take a look at this chart of the SP500 over time: http://www.macrotrends.net/2324/sp-500-historical-chart-data

(Make sure to turn off inflation-adjusted)

I think you'll see that in the last 90 years, even the worst market crashes don't take the index down to a level lower than what it was 15 years prior. To put it another way, pick any time in the past 90 years, the S&P 500 is always higher 15 years later than that date and every day afterwards. Maybe slicing up your investments into finer-grained categories than the entire S&P 500 will help ... but I'm very skeptical that anyone can TLH for long periods successfully.

If you want to pay a perpetual 25 basis points per year for Wealthfront or Betterment, go ahead, but it seems unlikely to me you'll come out ahead of a simple index fund if you are investing long term.

[hesdeadjim]

I suppose there is no harm then in milking the tax loss harvesting for as long as it is profitable and then reevaluating performance in ten or fifteen years to see if it makes sense to switch to a lower cost provider like Vanguard.

Like I said though, I have seen very significant returns from my loss harvesting. Even without a yearly source of capital gains, it definitely does not hurt to collect the losses and use them later in life (for instance if you sell an investment property).

[zeroer]

There is a harm. In 15 years, if you decide you don't like Wealthfront, in order to move, you will have to sell all your assets and incur the capital gains.