JSE FX Trader common mistakes
JSE FX Trader · Understand the traps and how to avoid them.
Computing a forward FX rate by adding the interest rate differential to the spot rate instead of applying it to the spot as a multiplicative adjustment over the actual day count.
Candidates carry over simple-interest intuition from bond yield questions and forget that covered interest parity prices the forward as spot times (1 + r_quote × days/basis) divided by (1 + r_base × days/basis), so they add instead of multiply and use 360 or 365 inconsistently.
What to do instead
Write the CIP formula before every calculation, fix the quote/base currency direction first, and check that the forward points have the correct sign (premium in the currency with the lower interest rate). Sanity-check with a rough estimate: forward points should be close to spot × rate differential × days/basis.
Confusing real and nominal share counts in an index calculation, or rebalancing weights without applying the divisor adjustment, so the index level jumps when a constituent changes.
Candidates treat a market-cap-weighted index as a simple average of prices and forget that the divisor must be recalculated after a constituent change, rights issue, or share count update to keep the index continuous.
What to do instead
Always compute the divisor as (total adjusted market cap) / (index level) before and after the corporate action, and confirm the index level is unchanged at the moment of the change. Practise identifying whether a corporate action changes market cap (new shares, rights) or only price (split).
Valuing a warrant with the Black-Scholes call formula and then failing to adjust for the dilution effect of new shares issued on exercise.
Candidates apply the standard option pricing model mechanically because warrants look like call options, ignoring that warrant exercise increases share count and dilutes existing holders, so the per-warrant value is lower than the equivalent call.
What to do instead
Use the warrant pricing adjustment: divide the total equity value (existing shares × share price plus warrant proceeds) by the fully diluted share count, then solve for the warrant value, or apply the dilution factor to the call value. State explicitly in the answer whether dilution is being modelled.
Reporting a sample standard deviation when the population standard deviation is required, or mixing up n and n-1 in the denominator without checking the question wording.
Candidates memorise one formula and apply it automatically; exam questions often specify 'the sample of returns' or 'the population of returns', and the difference in the divisor changes the answer enough to eliminate the correct multiple-choice option.
What to do instead
Read the wording for 'sample' versus 'population' before computing, and note the divisor explicitly in working. For a large sample the difference is small, but for small n (under 10) always check both and match to the answer options.
Calculating ETF tracking difference as the simple difference between fund return and index return without separating fees, transaction costs, and dividend withholding tax effects.
Candidates conflate tracking error (volatility of the return difference) with tracking difference (the cumulative return gap), and they ignore that the fund's total expense ratio and securities lending revenue both affect the gap.
What to do instead
Define both terms explicitly: tracking difference is the annualised fund return minus index return; tracking error is the standard deviation of that difference. When asked to decompose, list TER, transaction costs, sampling, and tax drag as separate line items.
Applying the clean price to a bond duration or accrued interest calculation when the dirty price is required, or forgetting to add accrued interest when settling a trade.
Candidates see the quoted clean price in the question and use it directly in present value or duration formulas, but bond pricing and duration are based on the dirty (full) price including accrued interest from the last coupon date.
What to do instead
Always compute accrued interest as coupon × days since last coupon / days in coupon period, add it to the clean price to get the dirty price, and use the dirty price in duration and convexity calculations. Check the day-count convention (actual/actual, 30/360) stated in the question.
Treating a preference share as a fixed-rate bond and ignoring the call feature, cumulative dividend rights, or the fact that dividends are paid from after-tax profits and are not tax-deductible.
Candidates see a fixed dividend and a par value and default to bond valuation, but preference shares have equity-like features (perpetual life, optional deferral, call schedules) and a different tax treatment that changes the effective cost of capital.
What to do instead
Identify the specific preference share features first (cumulative vs non-cumulative, callable, participating, convertible), then choose the valuation model: perpetuity for non-callable, yield-to-call for callable. Remember preference dividends are paid from net profit after tax, so the cost of preference capital is not tax-adjusted.
Computing a futures fair value without including the cost of carry for the correct asset class, e.g. ignoring storage and convenience yield for commodities or ignoring dividends for equity index futures.
Candidates learn one carry formula and apply it across asset classes, but commodities require storage costs and convenience yield, equity index futures require the dividend yield, and FX futures require the interest rate differential.
What to do instead
Build a carry checklist per asset class before calculating: equities (dividend yield), commodities (storage minus convenience yield), FX (rate differential), fixed income (coupon income minus financing). Write the formula with each term labelled so a missing term is visible.
Mixing up the bid and offer when calculating the cost of a round-trip FX or equity trade, or applying the mid rate to a client-facing transaction.
Candidates read the two-way price and use the mid or the wrong side because they do not fix the dealer's perspective: the dealer buys at the bid and sells at the offer, and the client does the opposite.
What to do instead
State the dealer's position explicitly ('dealer buys base at bid, sells base at offer') before selecting a rate, and compute the spread cost as the difference between the rate paid and the mid. For a round trip, apply the bid on the exit and the offer on the entry (or vice versa) consistently.
Calculating a traded security's yield or return using the nominal coupon rather than the yield to maturity or the holding-period return, and ignoring the reinvestment assumption.
Candidates default to the coupon rate because it is the most visible number, but the question asks for yield to maturity, current yield, or realised compound return, each of which uses a different formula and reinvestment assumption.
What to do instead
Before computing, circle the requested measure: current yield (coupon / price), YTM (IRR of all cash flows), or holding-period return (including reinvested coupons). For YTM, state the reinvestment assumption explicitly and use the correct compounding frequency.