The Inside Story on Trading Costs

Some things are easy to measure, like the internal costs of a mutual fund or exchange traded fund (ETF) or how much we spend in commissions at our custodian to buy or sell stocks, bonds, mutual funds and ETFs.

It’s much harder to measure the market impact of your trading, or how much you pay as a price-taker (like us) between the price to buy a security (the offer) and how much to sell a security (the bid).

For example, if the market to buy XYZ stock is 99.98 – 100.00, you want to know that you aren’t paying more than $100.00 when you are buying or receiving less than 99.98 when you are selling.  You know that the market maker will earn some spread, but you want to make sure that it isn’t too much.

Last year, for the first time, we hired an outside firm to do a trade cost analysis (TCA) of our stock and ETF trading (mutual funds all trade at one price after the market and bonds are a whole different story).

The TCA firm looked at all of our trades for the last two quarters of 2014 and said that we were actually making money from our trades, meaning that we’re inside the 99.99 – 100.00 bid/offer that I described earlier.

As I mentioned in Daily Insights last year, which you can find by clicking here, I thought that was more of a data issue or a fluke than an accurate picture of our trading (no offense to our awesome trading desk).

This year, we did the analysis again for all of 2015 and got a more understandable and still excellent result: the market component of our trading costs was less than half of one basis point, or 0.0038 percent.

We did 11,607 stock and ETF trades with a total notional value of $223.2 million dollars and the cost of our market impact was $8,451 according to the analysis.

As small as our trading impact is, it is actually a little overstated.  This analysis looks at every trade and we can see that three small cap stock trades were about $1,000 of the $8,451, even though they were small trades.  These were not things that we bought specifically because they are illiquid and costly to trade, but a client brought them in and we wound up selling them for some reason.

For all of the hype about high frequency trading (HFT) and other blood sport traders sucking money from small investors, they’re not doing it to our clients (well, a little bit, but it’s pretty insignificant).

To some extent, we protect ourselves by using limit orders, which means that we say that we won’t pay more than 100.00 in that previous example.  The problem with limit orders is that prices change all the time and sometimes you don’t get a trade done because the price went away from you and you had to pay more.

It turns out that this practice of using limit orders cost us a little bit more than using market orders, but not by very much.  Also, since more than 90 percent of our orders were limit orders, we might not have a good enough sample size to really say.

If we had done more market orders, perhaps we would have been picked off by market makers and HFT traders more.  Fortunately, the differences are pretty small, so we don’t have to worry much about it.

Overall, I am very pleased with the analysis.  It shows that we are fulfilling our fiduciary obligation to get best trade execution for our clients, which is something that we take seriously.

A representative from Schwab told us once that we could just rely on their best execution reports and didn’t need to pay for the TCA analysis, but I do feel better having a disinterested party tell us the good news.