Well, while having a discussion with Andy and the Wealthfront team, we had a discussion about tax-loss harvesting. Disclosure: Some of the features to be discussed came out of our discussion, so I'm naturally biased towards the product.
Here's the deal: most of us do naive, year-end-based tax-loss harvesting. In other words, at the end of the year, we check our balances, and if we see losses, we harvest them and then buy equivalent securities, or just sit in cash for the wash-sale period (31 days) and then buy back the original securities. According to Wealthfront's back-testing, this kind of tax-loss harvesting nets an additional 50 basis points (0.5%).
You can take a more sophisticated approach to this. For instance, Parametric Portfolio Management advocates a strategy where you build an approximate index and then tax-loss harvest individual securities. The biggest problem with that approach is if you eventually end up with no more tax losses to harvest, you're stuck with 500 or so securities in a separately managed account, and you'd find yourself wondering "what the heck do I do now?!" As far as I can tell, that's why few Googlers went with Parametric's approach.
Wealthfront's approach is much less headache inducing. The idea is that instead of just naively tax-loss harvest at the end of the year, you can tax-loss harvest at any time, as long as the cost of taking the losses is lower than the volatility of the asset in question. Now, once you do that, you have to take into account what happens if new money gets added to the account (not a problem), and when to switch back. Wealthfront's back-testing indicates that over the last decade (which has been very volatile), this would have netted an additional 100 basis point (1%) in performance a year!
Before you dismiss this as tiny, think about it this way. If your portfolio averaged 10% gain a year, an additional 1% is a performance improvement of 10%! (It's very hard to get 10% a year, by the way --- most realistic numbers are in the 6-8% range, making an incremental 1% huge) By the way, if your financial adviser has you in actively managed funds, then it's tough for him to do tax loss harvesting for you because it'd be hard to get equivalent funds to trade into. Historically, Wealthfront's approach of continuous tax-loss harvesting was only applied to separately managed accounts.
This new feature makes Wealthfront a compelling option for managing money. The additional 50 basis points gained from continuous tax-loss harvesting more than compensates for Wealthfront's 25 basis point management fee. And of course, if you're the kind of person who doesn't even do the naive tax-loss harvesting the number's even better. Even more importantly, these numbers are derived using existing, historically low tax rates. If tax rates go up in the future, Wealthfront's win will be even bigger. And of course, anyone who's been in the financial markets over the last 10 years knows that we've seen signs of increasing volatility: 2 asset bubbles and at least one Minsky moment. This sort of situation make automated rebalancing and tax-lost harvesting ideal. Wealthfront will give you statements at tax reporting time so that you don't have a reporting nightmare on April 15th. I suggested that they give you reports timed for the IRS estimated tax payment deadlines as well.
After I saw Wealthfront's presentation, I went home and opened a Wealthfront account. I do not foresee moving all my assets there because of the above-described reasons, but I can see myself moving liquid assets there over time. I've endorsed Wealthfront in the past, but this time, I'm actually putting (some of) my money where my mouth is. (Disclosure: Wealthfront waived management fees on my account because my previous product endorsement gave them many new customers --- this fee waiver predated the tax loss harvesting feature, and did not move me to sign up as a customer until I saw the tax loss harvesting presentation)