Evaluating trading procedures, disclosures and record keeping

Evaluating trading procedures, disclosures and record keeping

Published September 28, 2026

Evaluating trading procedures, disclosures and record keeping

On this page

Learning Outcome Statement

Evaluate a firm’s trading procedures, including processes, disclosures, and record keeping with respect to good governance.

On the exam: You receive a mandate, benchmark, risk and cost constraints, pre-trade and post-trade procedures, disclosures, and records. You evaluate governance and recommend a route in constructed-response or selected-response form.

Good governance is the firm’s ability to show that a trade was permitted, executed fairly, disclosed properly, and reviewed for the client. Start with the mandate, not the manager’s attractive alpha claim, because a route can promise more active return while spending risk or cost the client did not authorize. The evidence must let an independent reviewer reconstruct the decision.

Good governance requires documented processes, appropriate disclosures, auditable records, and independent oversight.

Mandate termControl testRequired evidence
MandateActive bet permittedApproved mandate
BenchmarkReturn comparison fixedBenchmark policy
Alpha sourceThesis supports active returnResearch record
Active riskTracking-error ceiling respectedPre-trade risk report
Risk budgetRisk and cost allocationBudget approval
CapacityOrder size fits trading capacityCapacity review
TurnoverTrading allowance respectedTurnover report
Implementation constraintCost, liquidity, timing limitsPre-trade test

Expected active return is the reward claim, while tracking error consumes the active-risk budget. Capacity and turnover test whether the portfolio action can be implemented within the mandate’s constraints.

For the cost test, let $B$ be the mandate’s cost budget in basis points of traded notional. Let $C$ be the route’s pre-trade cost estimate, positive from the fund’s perspective. The remaining cost headroom, $H$, is the budget left after the estimate.

$$ H=B-C $$

Use the same subtraction for tracking error, replacing the cost budget with the tracking-error limit and the cost estimate with the tracking-error estimate. A larger $B$ leaves more room, while a larger $C$ consumes room. Positive $H$ supports the route against that guardrail. Negative $H$ signals a breach before the trade is observed.

Good governance follows the trade from authorization to independent review. Each responsible party should leave evidence for the next phase.

Process phaseResponsible party and controlDisclosure and record
Pre-tradePortfolio manager and compliance test mandate limitsConflicts disclosed. Approval retained
ExecutionTrader follows approved route and benchmarkBroker choice, allocation logic, order trail
Post-tradeCompliance and independent reviewer test exceptionsAllocation, exception, and review records

Conflicts and allocation choices belong in disclosures, not only in internal conversations. The reviewer should be able to trace the mandate decision through execution and post-trade records.

Read the process table as a handoff of accountability. The portfolio manager and compliance establish permission before the order exists; the trader then follows an approved route and leaves a trace of broker choice and allocation logic. After execution, compliance and an independent reviewer examine exceptions. A control is useful only when its evidence survives that handoff and lets the reviewer connect the original mandate to the final allocation.

The figure makes the handoffs visible: authorization and benchmark setting lead into pre-trade checks, execution, retained records, and independent review. The sequence fixes the benchmark and control tests before the order is observed, so later performance cannot redefine whether the route was acceptable. At each step, ask who owns the decision and what record proves it happened. Missing evidence at one handoff makes the next reviewer rely on an unsupported claim.

Loading diagram…

Worked example

The flowchart gives you a practical reading order for the case: establish the mandate and benchmark, test controls, execute, then review the records. Do not begin with the attractive return claim. First determine whether each proposed route fits the limits and whether its decisions can be reconstructed. This order keeps the approval standard independent of the outcome and gives the reviewer a clear trail from authorization through execution.

You are handed a mandate with a stated cost budget and tracking-error limit. Routes are proposed, and you must evaluate their governance before recommending one. The controlled route has documented approvals, conflict disclosures, allocation logic, and post-trade review. The discretionary route has a disclosure gap and weaker records.

Case factValue
Cost budget10 basis points of traded notional
Tracking-error limit0.005 decimal tracking error
Controlled cost8 basis points of traded notional
Discretionary cost13 basis points of traded notional
Controlled tracking error0.0030 decimal tracking error
Discretionary tracking error0.0080 decimal tracking error
Controlled expected active return0.0040 decimal expected active return
Discretionary expected active return0.0060 decimal expected active return

Read the chart together with the case facts: the bars show tracking error, and the line shows expected active return. The controlled route uses less of the risk allowance and has a lower return estimate; the discretionary route claims more return while using more than the permitted tracking error. The chart does not show cost or disclosure quality, so check those separately in the table and records. Treat the return line as a claim to assess only after the guardrails pass.

Loading diagram…
  1. Subtract each route’s pre-trade cost estimate from the budget before observing the outcome. The controlled route gives $10-8=2$. The discretionary route gives $10-13=-3$. These results are basis points of traded notional.

  2. Apply the same control to tracking error because it consumes active-risk budget. The controlled route gives $0.005-0.003=0.0020$. The discretionary route gives $0.005-0.008=-0.0030$. A positive result leaves room, and a negative result breaches the limit.

  3. Compare expected active return only after the constraints pass. The discretionary route’s return claim is higher, but its risk and cost results fail the mandate.

  4. Inspect the records and disclosures against the process table. The controlled route supplies the required evidence, while the discretionary route leaves a disclosure and record gap.

FindingGovernance implicationEvidence and corrective action
Controlled route within limitsEligible for approvalHeadroom 2 and 0.0020. Retain review
Discretionary route breaches limitsOutside risk budgetHeadroom -3 and -0.0030. Reject or remediate
Discretionary disclosure gapIndependent review weakenedAdd conflict and allocation evidence
Weak discretionary recordsAudit trail incompletePreserve exception and review records

Recommend the controlled route because it respects the mandate and risk budget, stays within implementation cost, and can be independently reviewed.

Changed condition

Now broker capacity is strained, so only the discretionary route’s expected cost changes. All other risk, return, disclosure, and record facts remain unchanged.

MeasureBefore strainAfter strain
Discretionary cost1318
Discretionary cost headroom-3-8
Discretionary risk headroom-0.0030-0.0030

Before seeing the result, predict that cost headroom will fall because capacity strain raises expected cost. Risk headroom stays unchanged because no risk input moved. The governance breach worsens even though expected active return and procedural gaps do not move. Before reading on, write the route decision and the evidence you would require.

Common error

You approve the discretionary route because it promises the higher expected active return. That choice feels right because the manager's alpha claim appears to justify taking more active risk. The giveaway is the tracking-error or cost breach, combined with missing conflict disclosures, weak allocation evidence, or records that cannot support independent review: those facts expose a governance failure.

Check yourself

Practice

Which procedure best demonstrates good governance for a proposed trade?

Revision snapshot

Start with the mandate, benchmark, active risk, risk budget, capacity, turnover, and implementation constraints. Good governance requires documented processes, appropriate disclosures, auditable records, and independent oversight. Choose the execution benchmark and control tests before observing the trade. For a cost budget $B$ and pre-trade estimate $C$, headroom is $H=B-C$. Positive headroom supports approval against that guardrail. Negative headroom signals a breach. The trap is choosing higher expected active return while ignoring tracking-error or cost breach, missing conflict disclosures, weak allocation evidence, or records that cannot support independent review. On the exam, recommend the route that fits the mandate and cite process, disclosure, record, and independent-oversight evidence.

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: Evaluating trading procedures, disclosures and record keeping

  1. 1. 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.

    • 2. 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

      40 bps widening - Expected credit return equals carry minus the spread-duration effect minus expected loss: −1.10% = 1.10% − 5.0 × spread change − 0.20%. - The required spread-duration loss is 2.00%, so the spread change is 2.00% ÷ 5.0 = 0.40%, or 40 bps of widening. 20 bps narrowing - Narrowing of 20 bps would produce a 1.00% gain from spread duration, not the loss needed to reach a negative return. - Substituting −0.20% gives 1.10% − 5.0 × (−0.20%) − 0.20% = 1.90%, not −1.10%. 60 bps widening - Widening of 60 bps would produce a 3.00% spread-duration loss, overshooting the required 2.00%. - Substituting 0.60% gives 1.10% − 5.0 × 0.60% − 0.20% = −2.10%, not −1.10%.

    • 3. 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

      Sell USD 16.67 million - After the returns, the stocks are worth USD 120 million, USD 90 million, and USD 100 million, for a USD 310 million portfolio. - Equal weighting sets each holding at USD 310 million ÷ 3 = USD 103.33 million, so the first stock requires a USD 16.67 million sale. Sell USD 20.00 million - Selling USD 20 million removes the first stock’s entire USD 20 million gain, returning it to its original USD 100 million value. - The equal-weight target is USD 103.33 million, not USD 100 million, because the portfolio’s total value has changed. Buy USD 3.33 million - USD 3.33 million is the purchase needed for the third stock: USD 103.33 million − USD 100 million. - Applying that purchase to the first stock confuses the holdings; at USD 120 million, the first stock is already above its equal-weight target.

    • 4. 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%

      5.45% - The one-year return includes the 4 coupon and the expected price gain of 102.5 − 101 = 1.5. - Dividing the total gain by the initial price gives (4 + 1.5) / 101 = 5.45%. 3.96% - 4 / 101 = 3.96% calculates only the coupon yield. - It omits the expected 1.5 price gain when the bond moves from 101 to 102.5. 6.44% - Using 100 instead of 101 to calculate the price gain gives (4 + 102.5 − 100) / 101 = 6.44%. - The bond was bought at 101, so its expected price gain is 1.5, not 2.5.

    • 5. 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.

      • 6. 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.

        • 7. 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.

          Flashcards: Evaluating trading procedures, disclosures and record keeping

          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