What is Best Execution?
From the SEC website “Brokers are legally required to seek the best execution reasonably available for their customers’ orders.” You will notice they don’t actually define best execution; they just state the legal obligation of brokers to seek it.
FINRA takes a stab at it “Best execution is a significant investor protection requirement that essentially obligates a broker-dealer to exercise reasonable care to execute a customer’s order in a way to obtain the most advantageous terms for the customer. As the circumstances of each order and trading environment vary, so does the determination of what is best execution.”
So what is an investor to do. The SEC won’t define it but says brokers are legally bound to it and FINRA says “it depends”. In this article, we attempt to shoulder this crusade and outline exactly how you can measure best execution.
Arrival Price (aka Implementation Shortfall; IS; Slippage)
Performance to Arrival Price is like the Puff Daddy (who changed his name 9 times in his career) of the execution arena. But don’t be fooled or confused by the various monikers and convoluted titles; it’s really quite simple:
WHAT WAS THE PRICE OF THE STOCK WHEN YOUR BROKER RECEIVED YOUR ORDER? ARRIVAL
WHAT PRICE DID YOUR BROKER EXECUTE YOUR ORDER? EXECUTION PRICE
WHAT WAS THE DIFFERENCE? PERFORMANCE
An example usually helps to illustrate.
You want to buy 10,000 shares of Apple (AAPL) and you send your order to your broker at 10:00 AM. At that time, AAPL is trading at $150.44. An hour later you receive an execution price from your broker of $151.6534. Your performance can be measured a few different ways.
Absolute price difference: $150.44 – $151.6534 = -$1.2134
Now the absolute price difference, while helpful, because stocks trade at different prices is only meaningful for that specific stock. $1.21 of slippage is much different for TSLA whose price ($1162.94) is nearly 10 times AAPL and for Cemex (CX) which trades at less than 1/10th of AAPL ($6.97).
Enter Basis Points or “Bps”. Basis points are simply another form of percent but instead of multiplying by 100 (0.1*100 = 10%), you multiple by 10,000 (0.001*10000 = 10bps). This is used when changes are too small to be represented in percent. If a stock moves from 10 to 11 or 0.10, we can easily represent that in percent: 10%. But if it moves from 10 to 10.01 representing that in basis points, 10bps, is easier to comprehend than 0.001 or even 0.1%.
Basis points (bps), because it measures a relative movement, can be used to compare performance across various stocks. In the example above, AAPL moved 80.7bps which is easier to understand than 0.807% and MUCH easier than 0.0081. Calculation of performance to arrival in bps [(150.44 – 151.6534)/150.44*10000].
| At 0 to 1 Analytics, we take the executed price immediately after your arrival time (or the time you sent your order to your broker) down to the millisecond and use that for the arrival price benchmark to calculate your slippage. This ensures that your broker “is on the clock” immediately after receiving your order. |
Another relative measure of performance is in spread sizes or how many spreads was your slippage. Using the example above, let’s say that the average spread (bid price minus ask price) of AAPL for the past 20 days was $0.173. Your slippage in spreads was $1.2134/$0.173 or 7.01 spreads.
If you have made it this far, I have good news. You are now an expert in best execution or “transaction cost analysis” (TCA). But we aren’t done.
Philosophical interlude
Because equities is such an evolved asset class, we take for granted that an arrival price actually exists and that we can simply make an API call with an identifier and a time stamp and get back “arrival”. If we look at other asset classes or even other brokerage markets, arrival price is a much more elusive benchmark. Take freight brokerage for example. A shipper pays a broker $2.55/mile for a shipment from St. Louis, MO to Olympia, WA on November 9, 2021. Forget about milliseconds, what was the last price paid for that lane in the last week? Is there even data available? (Turns out there is “market data” for freight brokerage: DAT) Was that a refrigerated load and worse a “hazmat”? The point is, let’s not lose sight of the importance of the arrival benchmark just because it is so readily available.
VWAP (Volume Weighted Average Price)
Because equity orders are typically executed over a certain time period or duration, another meaningful benchmark is VWAP or the volume weighted average price over that time period. It’s basically measuring how well your broker tracked the prevailing prices during the time of execution. Did the broker submit the order on the bid at 10:00 AM, go for coffee while the stock price went up and then an hour later come back to their desk and simply amend the order to market?
In this scenario, if the stock price had an upward movement for the whole hour, the VWAP price would be much lower and thus the broker’s performance to VWAP would be terrible. Only looking at performance to arrival price, which may be within the band of acceptable, would not expose the broker’s true (under)performance.
Despite its mystique, VWAP is actually an easy calculation (provided you have the proper market data). It is simply the product sum of shares traded times the price traded for each transaction divided by the sum of shares traded in a certain time period. Here’s a very simple example:
AAPL transactions:
Shares Price Time
10,000 150.45 10:01:37.378
12,700 150.53 10:01:42.793
VWAP for those two transactions: (10,000*150.45 + 12,700*150.53)/(10,000 + 12,700) = 150.4948
Using our original example, let’s say VWAP from 10:00 AM to 11:00 AM for AAPL was 150.7589. Your broker’s performance to VWAP can be measured in absolute price difference:
$150.7589 – $151.6534 = -$0.389 or
in bps: [(150.7589 – 151.6534)/150.44*10000] = -21.2bps
Your knee jerk response would be to say the broker did much better versus VWAP than the Arrival but again relativity is key. Because the VWAP measure stretches over the entire period, it is a much less volatile benchmark than its cousin Arrival price which is simply a snapshot and much more erratic. A 21bp underperformance to VWAP is actually a poor outcome relative to industry standards while an 81bp underperformance to Arrival for just one order must be taken in the context of market conditions and may not be that bad.
| At 0 to 1 Analytics, we collect all your orders and calculate aggregate performance metrics across many different benchmarks throughout the life of the order (Previous Close, Open, Arrival, VWAP, Close, Next Day Close). We use industry leading market data providers so you can be sure your performance metrics are accurate. |
There are many other benchmarks including participation rate, duration and reversal to take under consideration but by reviewing your broker’s overall performance against the Arrival price and VWAP, you are on the right track to conducting your best execution review!
Check out our TCA Now™ broker performance application promotional video. Or call us at 833-201-0211 for a consultation.
