Transaction Cost

A transaction cost is a cost of making an exchange that is separate from the price of the thing exchanged. In financial markets it covers everything a trader gives up in buying or selling a security: the fees paid to intermediaries, taxes on the trade and the less visible cost of dealing at a price worse than the market's midpoint. Transaction costs reduce investment returns directly and determine which trading strategies are viable.

The concept comes from economics, where it has a wider meaning. Ronald Coase argued in 1937 that the costs of searching, negotiating and enforcing contracts explain why firms exist, and Oliver Williamson later built a theory of economic organization on the idea. Both received the Nobel Memorial Prize in Economic Sciences, in 1991 and 2009. This entry concerns the narrower sense used in trading.

Explicit costs are charged separately and appear on a trade confirmation.

• Commissions. The commission is the broker's fee for handling the order, charged per trade, per share or as a percentage of value.
• Exchange and clearing fees. Trading venues and clearing houses charge per transaction. Some venues pay a rebate to orders that add liquidity and charge those that remove it.
• Taxes and levies. Examples are stamp duty on share purchases in the United Kingdom, a financial transaction tax in several European countries, and regulatory fees on sales in the United States.
• Financing and borrowing. Positions held with borrowed money carry interest, and short selling requires a fee to borrow the security.

Implicit costs are embedded in the price obtained and must be estimated.

• Spread. A buyer pays the ask and a seller receives the bid, so a round trip costs the bid-ask spread. Each side pays roughly half.
• Market impact. A large order moves the price against itself as it consumes the available liquidity. Market impact is usually the largest cost for institutional orders.
• Delay and slippage. The price may move between the decision to trade and the execution. The resulting difference is slippage.
• Opportunity cost. An order that is not completed, for example a limit order that never fills, forgoes the profit the trade would have made.

The sum of these, measured against the price at the moment of the investment decision, is known as implementation shortfall.

A share is quoted at $49.98 bid and $50.02 ask, a midpoint of $50.00. A trader buys 1,000 shares at the ask and pays a $5 commission. Relative to the midpoint, the half-spread costs $0.02 \times 1{,}000 = 20$ dollars. On a trade worth $50,000, that is 4 basis points, and the commission is 1 basis point. Selling later on the same terms costs the same again. The round trip costs $2 \times (20 + 5) = 50$ dollars, or 10 basis points of the position.

The figure looks small, but it compounds with activity. A strategy that turns over its whole portfolio 50 times a year would lose $50 \times 0.10 = 5$ percent a year to costs before earning anything. Costs therefore weigh most on strategies with high turnover.

For large orders, practitioners often use an empirical rule in which impact grows with the square root of the order's size:

• $\sigma$ is the daily volatility of the security.
• $Q$ is the size of the order and $V$ the average daily trading volume.
• $Y$ is a constant of order one, estimated from data.

For a share with daily volatility of 2 percent, an order equal to 1 percent of daily volume, and $Y = 1$, the estimated impact is $2 \times \sqrt{0.01} = 0.2$ percent, or 20 basis points. Quadrupling the order doubles the impact. The rule is an approximation, and actual costs depend on how the order is worked.

Transaction costs set the boundary of what can be traded profitably. Many patterns reported in academic studies yield returns smaller than the cost of capturing them, particularly in small, illiquid shares. Costs also limit a strategy's capacity, because impact rises as more money is deployed, and they limit arbitrage, allowing related prices to differ by up to the cost of trading between them. A realistic cost model is an essential part of backtesting.

Investors manage costs by trading less, by using limit orders, by splitting large orders with an execution strategy, and by measuring results through transaction cost analysis. Brokers are obliged by best execution rules to take costs into account when handling client orders.

Explicit costs have fallen sharply. The move to quoting United States shares in cents in 2001 narrowed spreads, electronic trading reduced commissions, and in 2019 the major United States retail brokers cut commissions on share trades to zero. Zero commission does not mean zero cost: brokers may be paid by market makers through payment for order flow, and the spread and impact remain.