evaluating the benefits of Wealthfront's "Direct Indexing" product - offering my real-life data

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Topic Author
buffalo
Posts: 31
Joined: Mon Apr 16, 2007 6:12 pm

evaluating the benefits of Wealthfront's "Direct Indexing" product - offering my real-life data

Post by buffalo »

A recent post by LegendaryRed about leaving Wealthfront offers this observation:
There is some really severe tracking error with their Direct Indexing - it is currently holding more MSFT than AAPL and AMZN combined, which is completely not the market cap, mainly because I invested a good amount before AAPL took some dips (so AAPL was tax loss harvested), while MSFT mainly went straight up since I put money in. A lot of their assumptions assume you are constantly pouring more money into Wealthfront, which isn't what I'm doing now. Last year they started pushing risk parity which is also very sketchy, and I managed to turn that off in time.

Direct Indexing is actually bad because it prevents tax-efficient rebalancing: Whereas ETFs can perform share redemptions and creations in order to rebalance by cap weight or to add or remove stocks, you as a direct indexer, cannot.
I'd like to explore the Direct Indexing product in more depth. I've been a Wealthfront user for a few years, and I've been happy with the service. In December 2018, I enabled Direct Indexing, mostly because I was interested in seeing how it worked in real life, but the truth is that I haven't been able to figure out an appropriate way to measure its performance relative to a vanilla investment in VTI.

I'm hoping that mathematically-inclined people here can use my real data to perform an evaluation of the product.

For those unfamiliar with Direct Indexing, the idea is that Wealthfront essentially builds its own index fund on my behalf, saving the (minimal) expense ratio that would come with VTI/SCHB/ITOT and more importantly allowing me to claim tax losses on an individual-security basis. Instead of holding assets in a total market fund, it buys shares in up to 100/500/1000 large-cap companies and then purchases VB/VOO/VXF in order to complete the index. As individual stocks move up and down, Wealthfront buys and sells my holdings. I'm currently participating at the 100-company level (completing the index with VOO and VXF) since my balance isn't large enough to qualify for the 500/1000 level.

Wealthfront performs stock-level tax loss harvesting on a pretty regular basis, with transactions occurring (for me) over the past four-week period on 5/9, 5/8, 5/6, 5/2, 4/30, 4/29, 4/26, 4/25, 4/22, 4/18, 4/17, 4/16, 4/12, and 4/9. And that's without me adding any additional funds to my account during that timeframe.

My specific holdings, as of today, are shown below. I'd be happy to share trade confirmations (with my personal details removed, as in the sample below) in PDF format with anyone who is interested in testing the Direct Indexing product...I imagine you'd be able to quickly import trade data into a spreadsheet to run calculations. My problem is that I have no idea how to fairly approach those calculations myself.

My major question: is the tracking error inherent in an approach like this indeed outweighed by Tax Alpha, as suggested by Wealthfront's literature, when looking at real-life data from the last six months?

I'm currently in the 32% federal bracket, in a no-income-tax state, but I'm particularly curious to see how this would play out across all the federal tax brackets.

Thanks to anyone who is willing to tackle this. Reply here or PM me, please.

Current holdings:
Image

Individual stock holdings:
Image
Image
Image

Sample trade confirmation:
Image
NancyABQ
Posts: 299
Joined: Thu Aug 18, 2016 3:37 pm

Re: evaluating the benefits of Wealthfront's "Direct Indexing" product - offering my real-life data

Post by NancyABQ »

buffalo wrote: Thu May 09, 2019 12:08 pm My major question: is the tracking error inherent in an approach like this indeed outweighed by Tax Alpha, as suggested by Wealthfront's literature, when looking at real-life data from the last six months?
I don't have any profound analysis to offer, but my immediate thought on seeing this was that you don't have enough invested in the direct indexing (less than $20K) for it to work well? I am surprised you can do it with that amount of money, I thought their lower limit was $100K -- but I haven't looked at the whole thing in several years, maybe things changed.

With less than $20K, you have only a couple shares of each individual stock, which must lead to a large tracking error out of the gate, and would only get worse with TLH?

If you aren't adding any new money, they still do have dividends to contribute to rebalancing, but that probably wouldn't put a dent in it if the tracking error is large.

It's interesting to see some real-world info on how this works.
Post Reply