KASNEB CIFA Advanced common mistakes
KASNEB CIFA Advanced · Understand the traps and how to avoid them.
Selecting the project with the highest IRR when comparing mutually exclusive projects of different scale or duration, and then declaring it the better investment.
IRR is a percentage, so it flatters small, short projects; candidates learn 'accept if IRR > cost of capital' as a rule and apply it mechanically to ranking decisions without noticing the scale/size mismatch. Reinvestment-rate assumptions are also silently ignored.
What to do instead
Before computing anything, classify the decision as accept/reject (IRR acceptable) or ranking (use NPV). Always compute NPV alongside IRR for mutually exclusive options and justify the choice on NPV; if IRR is required, compute the incremental IRR on the cash-flow difference between the two projects.
Subtracting inflation from the nominal rate as a straight subtraction (r = n - i) when the question supplies high or variable inflation, and mixing nominal cash flows with a real discount rate.
The approximation is taught first and works for low inflation, so candidates stop writing the Fisher equation down. The deeper error is failing to keep the numerator and discount rate on the same basis (both nominal or both real).
What to do instead
Write (1 + nominal) = (1 + real)(1 + inflation) on the script before substituting, then state the basis of the cash flows and the discount rate in one line: 'nominal flows discounted at nominal rate'. Convert one side fully before discounting, never midway.
Building project cash flows that omit working-capital injection and its recovery at the end of the project, and taking depreciation itself as a cash outflow instead of using the tax shield.
Depreciation and working capital appear as non-cash or 'invisible' items in the income statement, so candidates copy accounting profit rather than reconstructing free cash flow. Working capital is often treated as a sunk cost sitting outside the project.
What to do instead
Use a fixed four-line layout for every project: operating cash flow (before interest, after tax, with depreciation added back or tax shield added), capital expenditure, change in working capital, terminal recovery. Reconcile the total to 'free cash flow' explicitly before discounting.
Pricing a semi-annual bond with an annual coupon and annual discounting, and quoting the answer as an absolute currency amount when the market convention asked for a price per 100 nominal.
Textbook examples are usually annual; candidates carry the annual template into semi-annual questions and forget to halve the coupon, halve the yield and double the periods. Price basis (per 100 vs absolute) is treated as presentation rather than substance.
What to do instead
Write the timeline first with the period count n = years x payments per year, coupon per period = nominal coupon / payments per year, and per-period yield = YTM / payments per year. State the answer as 'per KES 100 nominal' and reconcile to the absolute amount only at the end; note whether accrued interest must be added to get the dirty price.
Computing EPS by dividing earnings by the year-end number of shares, or ignoring the retrospective restatement of last year's EPS after a bonus issue or share split.
Bonus issues and splits raise no cash and are treated as bookkeeping, so candidates use the share count on the face of the statement of financial position. The comparative-year restatement requirement is easy to overlook when the question only asks for 'the EPS'.
What to do instead
Always build a weighted-average share schedule with dates and time weights, and apply the bonus/split factor to all periods presented (including the comparative). For a rights issue, apply the bonus factor only to the pre-rights portion and use the theoretical ex-rights price in the adjustment factor.
Valuing a share with the dividend growth model by using the current dividend D0 in the numerator instead of the next dividend D1, or applying a growth rate greater than the required return and reporting a negative or nonsensical value.
Formulas are memorised as P = D/(ke - g) without the time subscript. Candidates also take g from the question uncritically, even when ke has been estimated close to it, and then force an answer rather than flagging the breakdown of the model.
What to do instead
Label the dividend explicitly as D0 or D1 before substituting, and if D0 is given, apply (1 + g) first. Check the model's validity condition ke > g and state it in the answer; if it fails, switch to a two-stage or exit-multiple approach and say why the constant-growth model cannot be used.
Computing portfolio variance or standard deviation by averaging the individual asset standard deviations (or averaging betas) instead of using value weights and the covariance term.
The linear 'weighted average' pattern from expected return and beta is over-generalised to risk measures, where it is wrong. The 2w1w2Cov12 cross-term is dropped because it looks like a detail rather than the whole point of diversification.
What to do instead
Write the full variance expression with the covariance term before substituting numbers, and compute the covariance or use the correlation coefficient given. Sanity-check the answer: portfolio sigma must lie between the weighted average of the sigmas and the lowest individual sigma, so any result outside that band signals an error.
Using standard deviation as the risk measure in the Treynor ratio or Jensen's alpha, and mixing a portfolio's total risk with its systematic risk across the three performance measures.
Sharpe, Treynor and Jensen are memorised as three similar-looking fractions; the distinction between total risk (applicable to a fully diversified portfolio) and systematic risk (beta, applicable to a sub-portfolio within a larger fund) is not linked to which ratio is being used.
What to do instead
Before computing, write the risk measure each ratio demands: Sharpe = sigma (total risk), Treynor and Jensen = beta (systematic risk). Then ask whether the portfolio being appraised is the investor's entire holding or one component of a diversified fund, and let that answer determine the appropriate ratio; justify the choice in one sentence.
Pricing a treasury bill with the coupon formula, or applying the discount rate to the price paid rather than to the face value, and failing to distinguish the discount yield from the add-on (investment) yield.
Money market instruments are quoted on a discount basis over a 360-day year, but the bond pricing template is more familiar. Candidates also convert a 91-day yield to an annual figure by dividing by 4 rather than by 365/91.
What to do instead
For discount instruments first compute price = face value x (1 - discount rate x days/360), then derive the add-on yield from that price as (face - price)/price x 365/days. State the day-count convention and the quotation basis used in one line at the top of the workings, and never apply the coupon formula to a zero-coupon instrument.
Calculating the theoretical ex-rights price by averaging the cum-rights price and the subscription price without weighting by the number of shares, or computing the value of the right from the ex-rights price instead of from the cum-rights price.
The rights issue is treated as a simple averaging problem, and the cum-rights versus ex-rights distinction is glossed over because both prices are quoted in the same question. The value of the right is then derived from the wrong base.
What to do instead
Build a small table: existing shares x cum-rights price, plus new shares x subscription price, divided by total shares after the issue, to give the TERP. Then value the right as cum-rights price minus TERP (for an existing shareholder) and confirm that the shareholder is neither better nor worse off by holding the right versus selling it.