A lot has been happening lately for retail sized investment accounts. Most of the big online brokerage firms no longer charge commissions for individual stocks or ETFs (there are a few exceptions) and Charles Schwab announced it will allow investors to buy fractional shares of stocks, not ETFs, and it is a reasonable bet that the other brokers will follow suit on the fractional shares.

A separate conversation for months has been potential changes to the indexing business with custom indexes and direct indexing. Custom is pretty straight forward, direct essentially means bypassing the middle man, the fund company, to own all 500 or 1000 stocks directly. Nate Geraci has a good primer to learn more about direct indexing.

It is not clear to me at this point that custom indexing or direct indexing will meet the need of the typical retail sized investor (here I mean do-it-yourselfers and advisory clients). The mechanics of building an index yourself in the right weightings and then making changes when the index does would be monumental. Nate notes that there are companies that will do it for you and charge 25-35 basis points. That's cheap but funds tracking large cap, domestic indexes are mostly (entirely?) below 10 basis points. Why pay more?

If you own all 500 or 1000 stocks directly, there is the possibility for tax loss harvesting. This is true but I doubt too many people would take on what amounts to a full time job selling the losers and then tracking when they can be bought back as well as create a much bigger job for their accountants at tax time doing all that spreadsheet work. I can't see this making economic sense for someone with a few hundred thousand dollar or even a couple of million dollars. Think about, for a $2 million portfolio an index component with a 0.30% weight would be $6000, if it lost $600 are you likely to sell to realize the $600 loss and then wait the month to buy it back?

Nate does bring up the interesting idea that direct indexing can allow someone to avoid overlapped holdings that might occur in working for a large publicly traded company for many years and accumulating a lot of stock. Someone still working may not be able to divest the shares so in a extreme example if they have 50% of their money in their company stock (and they can't sell) and 50% in an S&P 500 ETF, direct indexing would be an effective way to avoid doubling up on the risk. For people working for Microsoft (MSFT), Apple (AAPL) or Amazon (AMZN), avoiding the overlap might make sense, those are the only three companies in the S&P 500 with a greater than 2% weight.

An Employee of Medtronic (MDT) in the situation outlined above probably doesn't need to worry about something bad happening to the stock in the context of index-level damage. MDT has a 0.56% weighting in the S&P 500. If there was news tomorrow that took it down to zero immediately, the impact on the index would be de minimis. This person would be in a world of hurt from the stock decline of course but direct indexing out the stock wouldn't help at all.

Where we are talking about evolution/innovation in indexing and how securities are traded, it makes sense to expect there will be further evolution/innovation that would better hedge out single stock risk as we're discussing here. We have talked in past blog posts about ProShares suite of ex-sector funds, these funds own the S&P 500 except for one sector. So our MDT employee could buy S&P 500 Ex-Health Care ETF (SPXV) and rely on their MDT position as a proxy for their healthcare exposure or add in a couple of other healthcare stocks or a sector ETF that one way or another avoids MDT.

Fractional share purchases are a little more interesting primarily for high priced stock shares (meaning a high price in nominal terms, not talking about value). The stigma and expense of owning less than 100 shares of stock is long gone which seems to have made stock splits less frequent. There are many stocks whose shares trade for many hundreds of dollars and a few north of $1000 per share.

Someone just starting to invest who opens an account with $25,000 might be comfortable buying one share of Alphabet (GOOG) for $1300 in terms of position size relative to the account, remember we are now commission free. For someone starting out with $5000, buying a share of GOOG for $1300 might be too much relative to the account size but buying 1/4 of a share or a 1/3 of share would be easier to do and take on less concentration risk. Fractional shares would also allow for more precise portfolio construction. Usually, when I implement a new client a 2% or 3% weighting (typical for my clients) means rounding up or rounding down such where the position is not exactly the 2% or 3% I think ideal. Using fractional shares allows a process that has been mostly spreadsheet work to be entirely spreadsheet work (talking about implementation not security selection).

A crucial idea to never get too far away from with any sort of innovation is whether it, so for now we're talking about direct or custom indexing and fractional shares, help keep you on the path to where yo need to be, whether they bring something to the table in terms of improving your chances for success or making the path easier or whether they are noise or some other sort of distraction. Based on what I know so far, fractional shares could very well be the former but it looks like direct and custom indexing would fall into the the category of distraction, at least for retail sized accounts.