CIPM Level I common mistakes
CIPM Level I · Understand the traps and how to avoid them.
When computing a time-weighted return, the candidate chains sub-period returns using the arithmetic average of the sub-period returns instead of geometrically linking them (multiplying (1+r) factors and subtracting 1).
Candidates learn that TWR 'removes the effect of cash flows' and then reach for the familiar arithmetic mean from statistics, especially when sub-period returns are small and the two answers look close. The arithmetic mean only equals the linked return when all sub-periods are identical, so the error hides in easy-looking numbers.
What to do instead
Write the linking formula explicitly every time: R = [(1+r1)(1+r2)...(1+rn)] - 1. If a question gives three or more sub-periods, compute the product of growth factors on paper rather than averaging, and sanity-check that the linked result is below the arithmetic mean whenever returns vary.
The candidate classifies a large client-directed cash flow as an external flow in a period when it actually occurred at the start or end of the sub-period, or mis-sets the sub-period boundary so the flow is treated as occurring mid-period, distorting the TWR.
Candidates focus on the return arithmetic and treat the timing of flows as a minor detail. In TWR, the whole point of revaluing at each large external flow is that the flow must be placed at a boundary; putting it inside a sub-period reintroduces the timing effect the method is designed to remove.
What to do instead
Before calculating, list every large external cash flow with its date and draw the sub-period boundaries at those dates. Confirm whether the firm's policy values the portfolio before or after the flow and apply it consistently. Then state in one sentence why the flow sits at a boundary, which forces you to notice if it does not.
The candidate selects money-weighted return (IRR) when the question calls for a return that is not influenced by the manager's control over cash flow timing, or selects TWR when the question asks for the return actually earned on the capital invested over the period.
Both measures are 'the return,' and candidates memorize definitions without anchoring them to the decision the number serves. They miss cue words such as 'manager cannot control contributions' (TWR) versus 'dollar-weighted experience of the investor' (MWR/IRR).
What to do instead
Build a two-line decision rule and apply it before any calculation: TWR evaluates the manager and neutralizes flow timing; MWR/IRR measures the investor's actual experience and is sensitive to flow size and timing. Scan the stem for who is being evaluated and whether flows were within the manager's control.
When computing a portfolio's return from segment returns, the candidate averages the segment returns equally instead of weighting each segment by its beginning market value (or the firm's defined weighting basis), so a small high-return segment dominates the answer.
Candidates see 'combine the segments' and default to a simple mean. They also mix weighting bases within one calculation, for example beginning values for some segments and ending values for others, without noticing.
What to do instead
Write the weighting basis in the formula before substituting numbers: R = sum(wi * ri) with wi = segment beginning value / total beginning value. Verify the weights sum to 1. If the question supplies both beginning and ending values, choose one basis, state it, and apply it to every segment.
In attribution, the candidate assigns an allocation effect to the interaction term, or reports selection plus allocation that do not reconcile to the total active return because the interaction residual is dropped or double-counted.
Different attribution frameworks (Brinson-Fachler versus Brinson-Hood-Beebower) treat the interaction term differently, and candidates memorize one formula without tracking which framework the question implies. The residual then vanishes or is assigned to the wrong effect.
What to do instead
State the framework up front and write all three components before computing: allocation, selection, and interaction. Then verify allocation + selection + interaction = active return exactly. If the question asks for a two-component decomposition, explicitly fold interaction into selection (or allocation) as the chosen convention and say so.
The candidate interprets a positive allocation effect as proof the manager overweighted an outperforming sector, without checking the sign convention for underweights and the direction of the benchmark-relative comparison.
The arithmetic of (portfolio weight - benchmark weight) times (benchmark segment return - total benchmark return) produces signs that feel counterintuitive. Candidates compute correctly but narrate the result backwards, which is exactly what many item stems test.
What to do instead
After computing each effect, translate it into a sentence and test the sentence against the weights: 'Overweight in a sector that beat the benchmark total should give a positive allocation effect.' If your number and your sentence disagree, recheck the sign of the benchmark-relative term, not the weights.
When asked to appraise performance, the candidate declares a manager skilled (or unskilled) from a single period's excess return and ignores risk adjustment, benchmark choice, and statistical significance.
The word 'appraise' invites a verdict, and candidates supply one from the raw return. They forget that appraisal requires asking whether the excess return is large relative to tracking error or volatility, and whether the benchmark is appropriate.
What to do instead
Structure every appraisal answer in three steps: (1) is the benchmark appropriate for the mandate, (2) what is the risk-adjusted excess return (Sharpe, information ratio, or alpha), and (3) is it statistically meaningful given the sample length. Never conclude skill from an unadjusted return alone.
The candidate uses the Sharpe ratio to compare two portfolios with different risk targets or uses the information ratio when the question is about total risk rather than active risk, and reports the ratio without noting its assumptions.
Both ratios divide excess return by a risk measure, so they blur together. Candidates also forget that Sharpe uses total volatility and the risk-free rate while the information ratio uses tracking error and active return relative to a benchmark.
What to do instead
Label the numerator and denominator before computing: Sharpe = (Rp - Rf)/sigma_p; information ratio = active return/tracking error. Match the ratio to the question: total-risk decisions use Sharpe; benchmark-relative, active-management decisions use the information ratio. State the assumption that returns are approximately normal when relevant.
In a GIPS context, the candidate treats the standards as a performance calculation methodology and asserts that compliance guarantees the accuracy of the returns, rather than describing them as ethical and disclosure standards governing how performance is presented and reported.
GIPS appears alongside return math, so candidates assume it prescribes formulas. The standards actually require full disclosure, fair representation, and consistent policies; they do not certify the correctness of any particular return calculation.
What to do instead
Memorize the framing sentence: GIPS are ethical standards for presenting performance, requiring firms to define the firm, document policies, include all actual fee-paying discretionary portfolios in composites, and disclose required information. When a stem claims compliance 'proves' returns are correct, identify that as the error.
The candidate states that a firm 'claims compliance' in marketing material without the required conditions, or omits that a firm must have its claim verified or that verification is optional, and confuses verification with a requirement for compliance.
Compliance and verification sound like the same thing, and candidates assume the stronger-sounding term is mandatory. They also overlook that the claim must be supported by policies, records, and a compliant presentation for each composite.
What to do instead
Separate the two: a firm may claim compliance only if it meets all applicable requirements of the standards; verification by an independent third party is recommended but not required to claim compliance. In any item, check whether the stem says 'verified' where it should say 'compliant,' and flag the substitution.