Locate Costs versus Slippage
By Dave Mabe
Here's a question from Oleg on short locate costs versus slippage in a backtest (name used with permission, edited for clarity):
Oleg:
Wouldn't accounting for HTB (hard-to-borrow) cost in a backtest be a bit different than slippage? We don't know how much they will be when the time to place a trade comes, but unlike slippage (outcome of which we find out only after the trade is placed), with HTB locates we can decide before the trade is placed.
As we have to locate shares before trading, we can clearly identify that if the price of the locates is higher than the profit from the ideal TP, then there is no point in actually trading. Yes, we can't do this in backtests and have to take everything assuming there were locates available, but in real life, we have an additional step to make a decision. How do you trade it live? Do you have a specific price per share / max R% / anything else to make this decision?
Dave:
You're right - there is an additional decision to make in live trading if you're trading a short strategy and use a broker that offers short locates.
How much are you willing to pay for the locate if the stock is hard to borrow?
But for a backtest, you should ignore locate costs, just as you should not spend a lot of time modeling slippage.
Similar to slippage, the right place to consider this is before you go live with a strategy.
And I would use the same calculation as I would for slippage (more info here).
The question is: how much shortfall from the backtest is too much for the strategy to be profitable?
Shortfall could be slippage and/or locate costs.
The threshold you're looking for should be some percentage of average profit per trade - at your current sizing.
So, ideally, as you size up your strategy, the locate threshold can automatically adjust accordingly.
When in doubt, start with a lower threshold and skip the larger locate costs, but - and this is important - keep track of the ones you skip in a log.
In the Trade Client with MabeKit (the tool I use to automatically execute my trades), there is a log that keeps track of locate costs.
You'll want to report on them later once you have enough data to explore.
One question you might want to ask is:
Do stocks with higher locate costs perform better in my strategy?
The answer to that question might lead you to adjust your threshold up or down after you amass enough data to start drawing conclusions.
Great question, Oleg - thanks for sharing with the group!
-Dave
P.S. "The process of becoming a more systematic trader honestly mirrors the journey of becoming a trader at all—lots of ups and downs, frustration, second-guessing, and moments where you want to quit. And then every once in a while, there’s a real moment of clarity and forward progress. Using the AFL code generator in MabeKit feels like one of those moments." - Mike A. Get Instant Access