Trade execution benchmarks

Trade execution benchmarks

Published September 28, 2026

Trade execution benchmarks

On this page

Learning Outcome Statement

Compare benchmarks for trade execution.

On the exam: You get a mandate, order characteristics, decision-price information, liquidity or volume pattern, urgency, and candidate benchmarks. You compare their fit and recommend a yardstick in a short constructed response or selected-response question.

Your portfolio manager has already decided to trade. You choose the execution benchmark before observing the trade, so the completed trade cannot choose a yardstick that flatters it. A benchmark is a yardstick for evaluating execution, not a trading algorithm. The mandate sets active risk through its tracking-error budget, while order characteristics, urgency, liquidity, information leakage, capacity, and turnover shape a fair comparison.

Call the decision-time price $P_0$ and the average execution price $P_{\mathrm{exec}}$.

BenchmarkReference pointMain limitation
Arrival priceDecision-time price $P_0$Misses unfilled opportunity cost
VWAPVolume-weighted average priceVolume pattern may be unrepresentative
TWAPTime-weighted average priceIgnores volume concentration
Implementation shortfallDecision price plus execution outcomesNeeds price for missed quantity

The anchor decides what the trade is judged against. For a buy, paying above $P_0$ is adverse, while a sell reverses the price direction.

Implementation shortfall is the decision-price measure that combines the cost of what you filled with the opportunity cost of what you did not fill. For a buy, the filled quantity, $Q_{\mathrm{filled}}$, creates a price difference of $P_{\mathrm{exec}}-P_0$. The unfilled quantity, $Q_{\mathrm{unfilled}}$, creates an opportunity cost measured against the later comparison price, $P_T$. Add explicit costs, $C_{\mathrm{explicit}}$, to get the total, $\mathrm{IS}_{\mathrm{buy}}$:

$$ \begin{aligned} \mathrm{IS}{\mathrm{buy}} &= Q{\mathrm{filled}}\times(P_{\mathrm{exec}}-P_0) + Q_{\mathrm{unfilled}}\times(P_T-P_0) + C_{\mathrm{explicit}} \end{aligned} $$

The filled term rises when you pay above $P_0$. The unfilled term rises when the later price $P_T$ is above $P_0$, which is the opportunity cost of waiting or failing to complete the order. $C_{\mathrm{explicit}}$ raises the investor’s cost, and a sell order reverses the price differences.

Use this order when the stem offers several plausible benchmarks:

Loading diagram…

Choose the benchmark before execution so later fills cannot shift the standard to favor the realized result. Arrival price isolates movement from the decision point, while VWAP and TWAP summarize different market patterns. Implementation shortfall adds the consequence of quantity left unfilled, which matters when urgency or a risk budget makes waiting costly. Read the mandate and order first, then select the yardstick that captures the relevant trade-off.

Worked example

You are the portfolio manager of a pension fund. An allocation change creates a public-market active equity rebalance with an order value of USD 1,000,000. The mandate seeks long-term equity alpha within a tracking-error risk budget and imposes capacity and turnover constraints. Choose the benchmark before execution and recommend how to interpret the cost information.

InputGiven
Portfolio actionRebalance active equity
Order valueUSD 1,000,000
Basis-point divisor10,000
Dollar value per basis pointUSD 100.0000
Cost ratesSpread 8 bps, impact 12 bps, commission 2 bps, delay 5 bps
Component costsSpread USD 800.00, impact USD 1,200.00, commission USD 200.00, delay USD 500.00, total USD 2,700.00
MandateLong-term equity alpha
Risk budgetTracking error
ConstraintsCapacity and turnover
PerspectiveInvestor, adverse costs positive
Candidate benchmarksArrival price, VWAP, TWAP, implementation shortfall
Patient choiceRisk 4 bps, alpha 12 bps
Balanced choiceRisk 8 bps, alpha 18 bps
Aggressive choiceRisk 15 bps, alpha 22 bps

The line uses execution risk proxy on the x-axis and alpha retained on the y-axis, following the patient, balanced, aggressive rows in the table. It shows an implementation trade-off, not a benchmark definition. Read the three points as choices under the same mandate: greater risk accompanies greater retained alpha in these illustrative inputs. This relationship helps explain why a benchmark must fit the investor’s objective. It does not prove that the most aggressive choice is best; capacity, turnover, urgency, and missed opportunity still shape the judgment.

Loading diagram…

The next bar is illustrative and shows the rising cost sequence. The table below records the cumulative implementation cost in USD, so the dollar amounts remain readable. Follow the bars from start through spread, impact, commission, and delay to see each component added to the running total. This makes the separate contributions visible, while the accompanying amounts let you check each step in the investor’s cost calculation. Keep the rates and dollar values distinct when explaining the result.

Loading diagram…

Read the heights as cumulative amounts: each new bar includes the earlier costs as well as the next component. The sequence reaches USD 2,700 after delay, which should match the component sum in the worked calculation. This visual makes the accumulation clear, while the rows that follow give exact labels and dollar figures for each point in the sequence.

PointCumulative cost
StartUSD 0
SpreadUSD 800
ImpactUSD 2,000
CommissionUSD 2,200
DelayUSD 2,700

Read the cumulative rows as a running sum rather than four independent final costs. Spread contributes USD 800, then impact brings the total to USD 2,000; commission raises it to USD 2,200, and delay completes the USD 2,700 total. This check connects the visual sequence to the component calculations and helps you explain how execution evidence accumulates. The investor perspective treats each adverse component as a positive cost.

  1. Convert the order value into a dollar value per basis point. Call that amount $u$ and divide by the basis-point divisor: $u=\frac{1,000,000}{10,000}=100.0000$ USD per basis point. This puts every cost rate on the order’s dollar scale.

  2. Multiply each cost rate by $u$ because each rate is a basis-point share of order value. The resulting spread, impact, commission, and delay costs are USD 800.00, USD 1,200.00, USD 200.00, and USD 500.00.

  3. Add the components from the investor perspective to show cumulative exposure. The total is $(800.00+1,200.00+200.00+500.00)=2,700.00$ USD, and the bridge records cumulative costs of USD 800, USD 2,000, USD 2,200, and USD 2,700.

Use the tracking-error risk budget, capacity constraint, and turnover constraint to judge the cost. These mandate facts make missed alpha relevant, so implementation shortfall is the primary benchmark. The risk-return figure illustrates an implementation trade-off, not a benchmark or algorithm.

Recommend implementation shortfall as the primary yardstick and arrival price as a companion, using USD 2,700.00 as implementation evidence for the mandate.

Changed condition

Only the delay rate changes in the pension rebalance, rising from 5 bps to 12 bps. All other facts remain fixed. Predict the cost movement before you read the table.

ItemBaseChanged
Delay rate5 bps12 bps
Delay costUSD 500.00USD 1,200.00
Total implementation costUSD 2,700.00USD 3,400.00

Only delay moves. With $100.0000$ USD per basis point unchanged, delay cost rises with the rate and total cost rises by the same dollar amount. The benchmark recommendation does not automatically change unless mandate urgency, liquidity, information leakage, capacity, or turnover changes. Before reading on, recompute the delay cost and total cost from the changed delay rate.

Common error

You may choose VWAP or TWAP because each sounds fair across its chosen pattern. That feels right because both are familiar ways to spread an order, but it confuses a benchmark with a trading algorithm. When the stem asks for a preselected yardstick and supplies volume pattern, urgency, and decision-price facts, compare benchmark fit rather than declaring either method universally superior.

Check yourself

Practice

Which comparison is correct before observing the trade?

Revision snapshot

Choose the execution benchmark before observing the trade. Arrival price is decision-time based. VWAP is volume-weighted. TWAP is time-weighted. For a buy, $\mathrm{IS}{\mathrm{buy}}=Q{\mathrm{filled}}\times(P_{\mathrm{exec}}-P_0)+Q_{\mathrm{unfilled}}\times(P_T-P_0)+C_{\mathrm{explicit}}$. Implementation shortfall links the decision price to execution outcomes and opportunity cost. Match the benchmark to mandate urgency, order characteristics, liquidity, information leakage, capacity, and turnover. A benchmark is a yardstick for evaluating execution, not a trading algorithm. The trap is treating VWAP or TWAP as universally superior. Volume pattern, urgency, and decision-price facts signal benchmark fit.

Unlock the full lesson and continue reading

Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more

Scan to download app

Scan to download
UNLOCK FREE ACCESS

Practice Questions: Trade execution benchmarks

  1. 1. A bond priced at 101 pays a 4 annual coupon and is expected to be priced at 102.5 in one year if the yield curve is unchanged. The calculated one-year return is closest to:

    • A. 3.96%
    • B. 5.45%
    • C. 6.44%
    Show answer

    Answer: B. 5.45%

    > Key point: One-year return equals coupon income plus the price change, divided by the initial bond price. Why 5.45% is correct - The coupon is 4, and the expected price gain is 102.5 − 101 = 1.5. - Total return is (4 + 1.5) / 101 = 5.45%. Working | Step | Working | Result | |---|---|---| | Price change | 102.5 − 101 | 1.5 | | Total gain | 4 + 1.5 | 5.5 | | One-year return | 5.5 / 101 | 5.45% | Answer: 5.45% Why the other options are wrong - 3.96%: 4 / 101 gives the coupon yield only, omitting the 1.5 price gain. - 6.44%: Using 100 instead of the purchase price 101 gives a price gain of 2.5 and return of (4 + 2.5) / 101 = 6.44%.

  2. 2. Apex Asset Management centralizes trading for several institutional equity portfolios. Portfolio managers transmit orders to the trading desk after completing their investment decisions. The desk records the market price when each order is released, all execution prices, explicit commissions, and the closing price for any quantity that remains unexecuted. The head trader uses implementation shortfall to evaluate total execution cost because the measure includes both completed trades and the opportunity cost of unexecuted shares. Exhibit 1 presents a buy order released when the security was quoted at USD 50.00 per share. The entire 600,000-share order was included in the manager's paper portfolio at that decision price. Only 450,000 shares were executed during the day, and the remaining shares were not cancelled for investment reasons. For implementation-shortfall measurement, the desk values the unexecuted quantity at the closing price. There were no taxes, fees, or delay costs other than the stated commission. The head trader next reviews three prospective buy orders shown in Exhibit 2. Atlas is small relative to average daily volume, and the portfolio manager has no material short-term information. Beacon is a large order associated with a time-sensitive earnings-revision signal; its price is rising, and failing to execute would create significant opportunity cost. Crest is also large, but the manager's signal has a long expected life. Crest has a wide quoted spread, and another institution has expressed nonbinding interest in selling a block at a negotiated midpoint price. To reduce broker selection bias, Apex also employs an algorithmic trading wheel that randomly allocates a portion of orders among approved brokers and tracks each broker's average fill rate, price improvement versus arrival price, and an information-leakage score, where lower scores indicate less adverse price movement attributable to the broker's routing. Exhibit 3 summarizes recent wheel performance. Separately, for orders such as Crest that have available crossing interest, the desk can route through a dark pool that matches buyers and sellers at the midpoint of the prevailing bid-ask spread without pre-trade display of order size. The desk evaluates executions using different benchmarks depending on the purpose of the analysis. Arrival price is available for all orders and represents the market price when an order reaches the desk. Closing price is also recorded, and volume-weighted average price is calculated using eligible market trades during the execution interval. The chief investment officer wants a benchmark that measures how effectively the trader preserved the value of the portfolio manager's decision. The compliance officer additionally notes that passive limit orders can face adverse selection. A limit order supplies liquidity but may execute disproportionately when another market participant possesses superior short-term information. All orders in this case are agency orders, and the desk receives no payment for order flow.

    • 3. Hartwell Asset Management operates a benchmark-relative active equity fund. The investment team uses company fundamentals and short-horizon earnings revisions to generate expected returns. Portfolio managers then translate those forecasts into positions, subject to limits on active risk, liquidity, turnover, and unintended factor exposures. Because several signals decay quickly after publication, the realized value of research depends materially on execution quality. The trading desk evaluates orders using implementation shortfall measured from the portfolio manager's decision price. For a purchase, implementation shortfall equals the actual cost of acquired shares, including explicit commissions and fees, minus the cost of acquiring the same shares at the decision price. Exhibit 1 presents a completed purchase order. Because all shares were executed on the decision day, there is no opportunity-cost component for unfilled shares. The desk does not use a single algorithm for every order. Instead, it considers order size relative to market volume, expected alpha decay, liquidity, intraday volume predictability, information leakage, and the performance benchmark specified by the portfolio manager. Exhibit 2 describes three pending orders. The alpha order is particularly time sensitive, so delaying execution can create substantial price risk even though immediate execution may increase market impact. The index-rebalance order has low urgency and will be assessed against full-day volume-weighted average price. The chief investment officer separately reviews three portfolio construction proposals in Exhibit 3. She wants forecast conviction to influence active positions but requires an implementable portfolio. Expected transaction costs should therefore be incorporated before trading, and constraints should prevent unintended concentration, excessive factor exposure, and positions that cannot be liquidated within the fund's risk limits. Exhibit 4 records comments made during a trading desk meeting. The committee evaluates each comment under normal market conditions. It recognizes that displayed venues may provide execution certainty but can expose trading intentions, whereas non-displayed venues may reduce information leakage but introduce execution uncertainty and adverse-selection concerns of their own. The desk also reviews a separate order worked with a percentage-of-volume (POV) algorithm and evaluated against arrival price rather than implementation shortfall, shown in Exhibit 5, to test execution quality under a different benchmark mechanic.

      • 4. The Meridian Public Employees Fund is reviewing both its passive and active domestic equity mandates. The passive mandate tracks a broad capitalization-weighted index containing 1,800 securities. Approximately 12% of the index weight consists of small, relatively illiquid securities. The investment committee permits ex ante tracking error of up to 0.40% and directs the manager to minimize total ownership cost, including turnover, commissions, bid–ask spreads, and market impact. Dividend income must be reinvested promptly, and no derivatives may be used. Three portfolios proposed for the passive mandate are presented in Exhibit 1. Full replication requires holding every constituent at its benchmark weight. The stratified sampling proposal groups securities by sector, market capitalization, and investment style, then selects representative securities from each group. The optimization proposal uses historical return covariances and transaction-cost estimates, subject to portfolio-level risk constraints. The committee recognizes that historical covariance estimates may be unstable when the number of securities is large. The fund also employs an active equity manager with a 3.0% ex ante tracking-error budget. Current sector positioning and risk data are shown in Exhibit 2. Expected active returns are the manager's forecasts relative to the sector components of the benchmark. The manager may adjust sector active weights without changing total portfolio value. A second active manager proposes a systematic portfolio using value, quality, momentum, and earnings-revision signals. The manager ranks approximately 1,200 securities each month, applies a multifactor expected-return model, and uses an optimizer to control sector and style exposures. The strategy targets 3.0% tracking error, holds approximately 250 securities, and permits only limited discretionary overrides when model inputs are judged unreliable. Finally, Meridian's fundamental manager uses bottom-up research to identify 40 high-conviction securities. Performance attribution indicates that the manager's security-selection decisions have added value, but unintended sector, size, and value-factor exposures have generated substantial volatility without positive average return. The chief investment officer wants the portfolio construction process to retain the manager's stock-selection insights while reducing uncompensated sources of active risk. Position limits, liquidity screens, and the existing benchmark will remain unchanged.

        • 5. A credit position has an expected credit return (excluding government-yield changes) of −1.10%, carry of 1.10%, spread duration of 5.0, and expected loss of 0.20%. The expected change in spread required to produce this return is closest to:

          • A. 20 bps narrowing
          • B. 40 bps widening
          • C. 60 bps widening
          Show answer

          Answer: B. 40 bps widening

          > Key point: Credit return equals carry minus the spread-duration effect and expected loss. Why 40 bps widening is correct - Set the expected return equal to carry minus spread duration times the spread change, minus expected loss. - Solving −1.10% = 1.10% − 5.0 × spread change − 0.20% gives a spread change of 0.40%, or 40 bps widening. Working | Step | Working | Result | |---|---|---| | Set up | −1.10% = 1.10% − 5.0 × Δspread − 0.20% | −1.10% = 0.90% − 5.0 × Δspread | | Solve | 5.0 × Δspread = 2.00%; 2.00% ÷ 5.0 | Δspread = 0.40% | Answer: 40 bps widening Why the other options are wrong - 20 bps narrowing: A −0.20% change gives 1.10% − 5.0 × (−0.20%) − 0.20% = 1.90%, not −1.10%. - 60 bps widening: A 0.60% change gives 1.10% − 5.0 × 0.60% − 0.20% = −2.10%, not −1.10%.

        • 6. Hawthorne University Endowment is replacing a domestic equity manager whose portfolio has deviated from the institution's revised investment policy statement. The endowment has a long investment horizon but must support annual university spending and capital projects. Consequently, the committee requires that at least 85% of the domestic equity allocation be held in securities offering daily liquidity. The committee believes active management can add value in domestic equities but does not want a highly concentrated portfolio to dominate total-fund risk. The investment policy assigns the mandate an expected tracking-error range of 3%–5% relative to the broad domestic equity benchmark. The committee will not waive either the liquidity minimum or the tracking-error range solely to select a manager with a higher stated expected return. Exhibit 1 summarizes three finalists using comparable forward-looking assumptions. After selecting a manager, the endowment will appoint an independent transition manager. The legacy and target portfolios contain some of the same securities, while other names appear as sales in the legacy portfolio and purchases in the target portfolio. The transition manager can cross compatible orders internally at an independently observed market price before routing residual orders to external markets. The endowment's objectives are to preserve benchmark exposure during the transition and minimize total implementation cost. Exhibit 2 presents the results for one buy order released at the decision price. Implementation shortfall is measured relative to that decision price. For a buy order, the calculation includes execution-price slippage on completed shares, commissions on completed shares, and opportunity cost on unexecuted shares using the closing price. No taxes, fees other than the stated commission, or overnight price movements apply. A separate rebalance order represents approximately 3% of the security's expected daily volume. The order is not information-driven, has low urgency, and can be traded throughout the day. The security has a stable and predictable U-shaped intraday volume pattern. The committee specifies full-day volume-weighted average price as the performance benchmark and wants to limit market impact rather than obtain immediate execution.

          • 7. An equal-weighted three-stock portfolio begins with USD 100 million in each stock. Returns are 20%, −10%, and 0%. At rebalancing, the trade in the first stock is closest to:

            • A. Sell USD 16.67 million
            • B. Sell USD 20.00 million
            • C. Buy USD 3.33 million
            Show answer

            Answer: A. Sell USD 16.67 million

            > Key point: Rebalancing to equal weights means trading each holding to one-third of the portfolio’s post-return value. Why Sell USD 16.67 million is correct - After returns, the holdings are USD 120 million, USD 90 million, and USD 100 million, totaling USD 310 million. - One-third of USD 310 million is USD 103.33 million; the first stock is USD 16.67 million above that target, so sell that amount. Working | Step | Working | Result | |---|---|---:| | Post-return portfolio value | 120 + 90 + 100 | USD 310 million | | Equal-weight target per stock | 310 ÷ 3 | USD 103.33 million | | Trade in first stock | 103.33 − 120 | Sell USD 16.67 million | Answer: Sell USD 16.67 million Why the other options are wrong - Sell USD 20.00 million: This removes the first stock’s gain and leaves USD 100 million, below its USD 103.33 million target. - Buy USD 3.33 million: This is the amount needed to raise the third stock from USD 100 million to its target; the first stock is already above target.

          Flashcards: Trade execution benchmarks

          1. 1. Best Execution

            Show answer

            Four characteristics of best execution: - Cannot be judged independently of the investment decision - Cannot be known with certainty ex ante (before the fact) - Can only be assessed after the fact but not a trade by trade basis (can't judge base on one hand) - Relationships and practices are integral to best execution

          2. 2. Algorithmic Trading Strategies

          3. 3. Types of Trades

            Show answer

            Principal Trades - Broker commits capital for the prompt execution of the order - Suitable when order size is large and more urgent Portfolio Trades - Order to trade a specific basket of securities - Low cost on a relative basis

          4. 4. - High-touch approaches 1. Principal trades. 2. Agency trades.

            Show answer

            - High-touch approaches involve high levels of human involvement.\ This is usually required for large trades (known as block trades) 1. Principal trades (broker risk trades). This is where dealers or market makers assume all or some of the risk relating to executing the order, Quote-driven, over-the-counter (OTC), or off-exchange markets, request-for-quote (RFQ) markets. 2. Agency trades. This is where the broker finds the other side of the trade, and risk for order execution remains with the portfolio manager/trader

          5. 5. characteristics of bond with lower liquidity (5) - credit ratings, - time since issuance, - issuance size, - issuance outstanding, - time to maturity

            Show answer

            Low credit ratings,\ longer time since issuance,\ smaller issuance size,\ smaller issuance outstanding,\ longer time to maturity

          6. 6. Which Algorithm for trade execution? Scheduled algorithms\ Liquidity-seeking algorithms\ Arrival price algorithms\ Dark strategies/liquidity aggregators\ SORs

            Show answer

            Scheduled algorithms - small orders in liquid markets - less urgency (i.e., greater risk tolerance for longer execution periods) - concerned with minimizing the market impact Liquidity-seeking algorithms - larger orders in less liquid markets - higher urgency - mitigate the market impact. - concerned that displaying limit orders may lead to information leakage, - when liquidity is typically thin with sporadic episodes of high liquidity. Arrival price algorithms - small orders in liquid markets - for managers who believe prices are likely to move against them during the trade horizon, and therefore wish to trade more aggressively (e.g., a profit-seeking manager). - appropriate for more risk-averse managers who want to minimize execution risk. Dark strategies/liquidity aggregators - large orders in illiquid markets, - arrival price or scheduled algorithms would likely lead to high market impact. - Since there is a lower chance of execution in dark pools, these strategies are for managers that do not need to execute the full order immediately. SORs - small market orders with low market impact where the market can move quickly, - small limit orders with low information leakage where there are multiple potential execution venues.

          Everything you need for CFA prep

          Get full Oncourse access with lessons, practice questions, flashcards and AI study tools.

          GET STARTED FOR FREE
          75% OFFLimited time offer
          GET 75% OFF