Robinhood.io isn’t that big of a deal

A new upstart in the stock broker-dealer market Robinhood.io recently announced it’s pre-existence, spotlighted by an investment from Google and Andreessen Horowitz, promising the miracle of completely free stock trading. Some cheered to celebrate an end to investing fees while others are left scratching their heads wondering what the catch is.

It is entirely possible that Robinhood is completely genuine in their claim to “always get you the best price possible”. However, “possible” can be pretty broadly defined. As is often repeated: if you’re not paying for the product, then you’re not the customer; you are the product.

Robinhood claims that API fees and margin accounts will provide sufficient revenues, but what they leave unsaid is more telling. Read some of this discussion on HN carefully and you’ll see some additional theories emerge. Most interesting is the idea retail brokers can “sell” their trading flow from unsophisticated investors (aka: the small fish) to the more sophisticated market makers (aka: sharks) who use high frequency algorithms to nearly guarantee profits by carefully managing the bid/ask spread.

The central issue with playing the markets is how opaque the internal workings of the markets are and how difficult it is to tell who’s getting rich off of whom. The lack of transparency in market systems should be enough to give anyone pause. For a humorous opinion on this read James Altucher’s “who makes money on Wall Street”. Once you get past the snark and fatuously antagonistic tone, you’ll find someone with experience in trading as well as algorithmic trading strongly arguing that this is a losers game for all but the closest of insiders, richest of the rich, or fastest among algorithms. While I don’t completely agree with these generalizations, they do seem to be mostly true. To go a little deeper, I’d also recommend The Wall Street Code. The basic premise is that even amongst the fastest algorithms, the table appears to be rigged towards insiders.

Of course, the markets aren’t necessarily “rigged” per-se, but there is, at the very least, subtle and hidden fees being collected by all the internal deal makers who get their cut by “helping” make deals with your money. This is entirely separate from the other fees and commissions you do (or don’t in the case of Robinhood) pay. This arguably creates a kind of “drag” which works against your outcomes, particularly if you trade often. Trying to make short term gains on risky bets can work doubly against you if you’re often on the losing side of those bets. The market makers know this game well and are effective at using it against you.

Let’s say, for the moment, you feel I’ve got it all wrong (entirely possible) and plan to go for it anyways. Well then, I’d ask, why wait for an unproven upstart with an unclear business model? You can do this right now with Interactive Brokers (no affiliation), who are well known, modern and low-cost—It’s not free, but $0.01/share seems pretty good—For quants wanting to test their algorithms they have an API with libraries on Github (how’s that for the old fashioned finance). Additionally, you could also check out Quantopian. They allow you to build and backtest algorithms for free against historical minute-level data. Their live trading beta lets you then connect with an Interactive Brokers account so you can play for “realz realz”. Again I want to stress I’m not recommending this, just showing that the future is already here, and even available in Canada for a change. Robinhood.io’s pricing model and modern touch may be novel, but it doesn’t seem all that groundbreaking.

In the end, it’s wise to remember there’s no free lunch. If you want to be a winner in the markets, you can either become an insider (aka: cheater) or, like the rest of us, build your wealth slow and steady sticking to proven long term strategies. The latter is clearly harder but is still, and by far, the most statistically reliable method. Everything else is tantamount to gambling, and like most casinos the house always wins.

Thanks to Roham Gharegozlou for providing feedback on an early draft of this post.