CFA Level I common mistakes
CFA Level I · Understand the traps and how to avoid them.
Answering ethics items by asking 'what would a decent person do?' and then picking the choice that matches that instinct, without first naming which of the seven Standards is engaged and what its sub-paragraph literally permits or forbids.
The Standards are a codified rulebook with specific gates - written consent, disclosure of conflicts, from whom, and by when - that often differ in small but decisive ways from everyday ethical intuition. Intuition gets you to two plausible answer choices and then fails.
What to do instead
Before reading the answer choices, write the Standard number and the operative verb on scratch paper ('Standard III(B) - fair dealing - must disseminate'; 'Standard VI(A) - must disclose to employer, not just refrain'). Eliminate any choice that satisfies intuition but skips a required consent, recipient, or timing step.
Computing a test statistic or confidence interval and then comparing it to 1.96 (or 1.645) as a fixed cutoff regardless of sample size, and stating the p-value as 'the probability the null hypothesis is true.'
The 1.96 anchor is memorized early and never updated; candidates also carry over the misreading of p-values from general reading, and both errors survive because the arithmetic that produces the statistic is correct.
What to do instead
Always write 'df = n - 1' explicitly before choosing a critical value, and look up the t-value rather than substituting the z-value. State the p-value definition verbatim from memory: the smallest significance level at which the null can be rejected, given the null is true.
Reading an exchange-rate quote and applying an interest-rate differential in the direction that 'feels right' - e.g., concluding the currency with the higher interest rate will appreciate in the forward market.
The quote convention (price currency per base currency) and the arbitrage logic of covered interest parity point in opposite directions to casual intuition, and the candidate silently flips the base and price currency mid-calculation.
What to do instead
Write the quote in words first: 'units of price currency per one unit of base currency.' Apply CIP as F/S = (1 + i_price)/(1 + i_base), then check the sign: the higher-interest-rate currency trades at a forward discount. If your answer says forward premium, the quote has been inverted.
Building CFO from net income using only the visible line items, forgetting to add back depreciation and amortization and to remove gains or losses on asset sales, and misclassifying interest or dividends received between operating and investing.
Candidates anchor on the income statement and treat the indirect-method adjustment list as a small patch rather than a systematic reconciliation; the gain-on-sale subtraction is especially easy to omit because the gain already sits in net income.
What to do instead
Use a fixed checklist on every indirect-method question: non-cash charges (+), gains on disposals (-), losses on disposals (+), increases in operating assets (-), increases in operating liabilities (+). For interest and dividends, apply the classification rule for the standard being tested (US GAAP vs IFRS) explicitly rather than defaulting to operating.
Estimating the cost of debt from the coupon rate printed on the company's outstanding bonds rather than from the current yield to maturity, and weighting the capital structure with book values because they are the numbers given.
The coupon is the most visually salient number in an exhibit and book values are easy to source from the balance sheet, while the required return on debt and target market-value weights require an extra, less obvious step.
What to do instead
Before computing WACC, state the rule: cost of debt = YTM on currently traded debt (or a synthetic rating-based spread), and weights = market value of equity plus market value of debt. If market values are not given, use the target capital structure stated in the question; use book values only if the question says so.
Feeding trailing dividends into a forward-looking valuation, or plugging a growth rate into the Gordon growth model without testing that it is below the required return, then reporting the resulting (negative or infinite) number.
D0 and D1 look nearly identical in a question stem and the difference is a single superscript; the g < r stability condition is usually checked as an afterthought, if at all.
What to do instead
Label the dividend in the stem as D0 or D1 out loud before substituting, and force any needed growth: D1 = D0(1+g). Compute g < r as an explicit inequality in the margin first; if it fails, the model does not apply and a multistage or spread-based answer is required.
Quoting a computed present value as the bond price when the settlement date falls between coupons, i.e., comparing a full price directly with a dealer's clean price quote and calling any difference an arbitrage or a mispricing.
Textbook problems usually assume settlement on a coupon date, so accrued interest never enters the calculation; the candidate then carries that same habit into questions that deliberately place settlement mid-period.
What to do instead
After computing a present value, always ask 'is settlement on a coupon date?' If not, accrue interest from the last coupon to settlement and use full price = clean price + accrued interest. State which one the dealer quotes and which one the present value represents.
Treating the value of a forward or futures contract as if it were the terminal payoff, or flipping the sign of the gain on the short side of an interest-rate or currency forward.
Both quantities are labelled 'value' in the reading and both involve the futures price, so the distinction between marked-to-market value at time t and payoff at expiry collapses under time pressure.
What to do instead
Write the contract value formula for the long position as the discounted expected spread, then set t = T explicitly when the question asks for payoff. For sign, restate the position as a cash-flow sentence ('I have agreed to buy at X; the market price is now Y'), which makes the short side a mirror image without memorization.
Computing hedge fund or private fund fees in the wrong order or off the wrong base - taking the incentive fee on the beginning-of-year gross value, or applying the management fee to ending value, or ignoring a hurdle or high-water mark.
Three interacting conventions (fee base, fee ordering, hurdle) are each individually simple, but the question stem supplies only some of them, and the candidate silently assumes the default.
What to do instead
Before calculating, write three labels at the top of the working: management fee base (typically beginning or committed capital), incentive fee base (net of management fee, on the excess over the hurdle), and high-water mark status. Compute management fee first unless the stem says otherwise, and state the assumption in one line.
Annualizing a multi-period return linearly (multiplying by the number of periods) while annualizing the accompanying standard deviation with the square-root-of-time rule, or mixing a monthly standard deviation with an annual return in the Sharpe ratio.
The two annualization rules are learned at different times and recalled as separate formulas, so the pairing error is rarely noticed; both numbers look plausible in isolation.
What to do instead
Derive both from the same horizon in one line: return scales with t, variance scales with t, so standard deviation scales with sqrt(t). Before dividing, write the units next to both numerator and denominator - if one says 'monthly' and the other says 'annual,' convert first.