T2W Trading Community

Traders go to great lengths optimising market strategies and trading systems to obtain good profitability. Unfortunately, those profits often deteriorate substantially after factoring in order fees. This article examines the effect of various broker commissions on portfolio performance and determines the optimal investment size needed depending on broker fees. The choice of broker greatly determines if a trader survives successfully in this business or quickly gives up in frustration.

Brokers use various methods to determine their pricing structures. They are often based on either block pricing (i.e. number of stocks per trade, investment amount, trading activity) or fixed pricing, better known as flat rates (fees independent of number of stocks traded or investment amount). That’s why investors should determine trading magnitude and frequency before making a short list of brokers. Of course, commissions are only one aspect to be weighed when considering a broker, but as will see it is a very important one.
The author publicly trades a small model portfolio (ca. $8000.00) on his web page using a well-known discount broker and a defined trading system. When analysing the results, it became apparent that order commissions substantially reduced profits.

Performance during a five-month period from the beginning of May 2004 through the end of September 2004 amounted to a respectable 23.44%. The actual profit after subtracting commission fees was 15.90%, a reduction of more than 32%. Even the most robust of strategies can hardly make up for this “loss”. It became apparent that a more thorough analysis of order costs was needed.

Peripheral Factors
The above mentioned model portfolio contains only US stocks traded directly in New York, so the broker’s current commissions for the USA were used in the calculations. Additional costs (such as exchange fees) were not considered.

Eight stock baskets containing ten stocks each were traded equally resulting in 80 trades. The original historical data of the model portfolio can be viewed under www.value-trading.de/trades.html.

Looking at the results on the web page, readers will note a couple interesting differences. There are two reasons behind this:

1. Because of account fluctuation through profit and loss and the strict money management rules used, the eight baskets were allocated varying amounts of investment capital. However, capital was divided equally among the ten stocks within each basket (in the performance simulation presented here, the same amount of capital is allocated to each basket). A maximum 50% (5% per stock) of total available capital was invested as a rule.
2. Two different commission fees were applied in the analysis because the broker used for the test changed its pricing schedule during the sample timeframe effective June 10, 2004.

The influence of this change on the study had to be eliminated if various brokers were to be compared on the same basis. For that reason it was necessary to determine the basic performance of the trading system alone without consideration of order fees or money
management rules.

Performance Simulation
A simulation was conducted on the basis of basic system performance using an excel spreadsheet. For each broker a basket containing ten stocks was traded equally and the commissions for each transaction calculated. This was repeated using various levels of investment capital. The stocks and their buy and sell prices can be reviewed in the sample portfolio. The results are listed in figure 1.

Note that with varying portfolio performance the actual numbers diverge, that is to say a parallel displacement of the curves in figure 4 takes place either up or down, but the fundamental effect of order fees remains the same.