Series 16 common mistakes
Series 16 · Understand the traps and how to avoid them.
Approving a draft report after reading the prose and checking the disclosure block, but never re-deriving the price target from the valuation method described in the body, so a report goes out with a Buy rating and a target below the current share price, or a target implying an exit multiple the stated comparable analysis cannot support.
Reviewers treat the price target and rating as narrative content produced by the analyst rather than as a computed output, and under time pressure they skim for language and formatting instead of re-running the math.
What to do instead
Before sign-off, rebuild the price target from the stated methodology (base year estimate x stated multiple, or DCF output), compute implied upside from the date-of-report price, and confirm the result falls inside the band the firm's published rating definitions assign to that rating. Recompute, do not eyeball.
Letting the summary, headline, or bullet teaser drift away from the body during editing, so the front page shows an upgrade or a new target while the rating table, the prior-rating line, and the risk paragraph in the body still reflect the old call.
Reports are assembled in pieces by different people and the summary is usually written last-minute or reused from the previous report; no single person owns the consistency check.
What to do instead
Run a line-by-line reconciliation of the five elements that must match exactly: rating, price target, earnings estimates, prior rating/prior target line, and valuation/risk discussion. Do it as a formal checklist step after the final edit, not before.
Modeling revenue on a gross basis for a company that acts as an agent (marketplace, broker, consignment, or pass-through distributor), thereby overstating revenue and understating the reported margin percentage, and then applying a revenue multiple to that inflated top line.
Candidates default to the income-statement top line as printed and never ask who controls the good or service before transfer and who bears inventory/credit risk, which is the actual test for principal versus agent treatment.
What to do instead
For every company modeled, document whether it is principal (gross) or agent (net) using control indicators, and check the accounting policy note and the gap between revenue and gross profit. Restate historicals on the same basis before forecasting.
Projecting NOPAT and free cash flow by carrying forward the reported effective tax rate, when that rate was depressed by a one-time item (audit settlement, valuation-allowance release, foreign mix shift), or carrying it forward when it was inflated by a nondeductible charge.
The effective rate is the line the model pulls automatically from the tax footnote, and it looks like a real assumption rather than the residue of unusual items.
What to do instead
Reconcile the effective rate to the statutory rate item by item; use a normalized marginal rate (statutory plus expected permanent differences) for forecasts, and separately model discrete items only in the quarters management has guided to.
Mismatching the numerator and denominator of a multiple: dividing enterprise value by net income or earnings per share, dividing equity value (market cap) by EBITDA, or computing EV/EBITDA without first bridging from market cap by subtracting cash and adding debt, preferred, and noncontrolling interest.
Multiples are memorized as pairs of labels rather than as claims on the capital structure, and spreadsheet cell references make it easy to point a formula at the wrong total.
What to do instead
Before computing any multiple, state in writing whether the metric accrues to all capital providers (revenue, EBITDA, EBIT) or to equity holders only (net income, book value, FCFE). Put the EV bridge on its own worksheet with a labeled check that EV equals market cap plus debt plus preferred plus NCI minus cash.
Building WACC with pre-tax cost of debt, with book-value rather than target market-value weights, or with a levered beta taken from the company's current capital structure and applied unchanged to a valuation that assumes a materially different one; equally, discounting nominal cash flows at a real discount rate.
The WACC tab is built once and copied across models, so the assumptions embedded in it stop being visible; the tax shield on debt is conceptually understood but forgotten at the keyboard.
What to do instead
Write out each WACC input with its source and basis. Use after-tax cost of debt, target (not current) weights, and relever/unlever beta with the Hamada or comparable formula whenever the assumed structure changes. Match nominal-to-nominal or real-to-real for both cash flows and the discount rate.
Producing quarterly estimates by dividing the annual growth rate evenly across the four quarters, which produces a Q1 that contradicts the company's historical seasonality, channel-fill patterns, and segment mix, and then anchoring the full-year number to that smooth quarterly path.
Straight-lining is fast and looks defensible on a chart, and seasonality requires going back to historical quarterly data that is not on the summary page.
What to do instead
Build the top line from volume x price (or units x ASP) by segment, apply historical quarterly seasonality factors explicitly, and reconcile the resulting quarterly sum to the annual estimate and to management guidance ranges before publishing.
Publishing a price target with no stated time horizon or valuation basis, or changing the rating without changing estimates or the target, so the rating moves for reasons that appear nowhere in the report and the target cannot be reproduced by a reader.
Ratings are often driven by relative preference, sentiment, or a sector call, but the report only documents the absolute valuation math, leaving the actual reasoning undocumented.
What to do instead
Every target must state the metric, the multiple (or DCF date and assumptions), and the horizon (e.g., 12 months). When the rating changes, state explicitly in the first paragraph whether estimates, the target, the sector view, or relative positioning drove the change, and make sure the target and horizon support it.
Producing the disclosure block from memory or from a prior template, omitting the rating distribution (percentage of covered companies rated buy/hold/sell), the percentage of those that are investment banking clients, the price chart with historical ratings and targets, or the market-making/ownership and compensation disclosures because the covered company 'doesn't have a banking relationship.'
Disclosures are treated as boilerplate conditional on the deal, and templates are reused across names without checking which items are unconditional.
What to do instead
Maintain a disclosure matrix with every required item in FINRA Rule 2241 (which consolidated the former NASD Rule 2711 and NYSE Rule 472) marked as either always required or event-triggered, and tick each item against the current fact pattern rather than against the previous report on that name.
Treating the permitted review of a draft by non-research personnel, or the post-offering waiting period, as looser than the rules allow: routing a draft to investment banking with an invitation to comment on the rating or target, or publishing initiation research inside the post-offering quiet period (generally 10 calendar days after an IPO and 3 calendar days after a follow-on offering for a managing underwriter).
The exceptions are learned as headlines ('bankers can see drafts' and 'there is a quiet period') without the limits attached to each, and commercial pressure to publish early is strongest exactly when the restriction bites.
What to do instead
Confine any non-research review to verification of factual accuracy and log it; never let banking personnel see a draft before a rating or target is final, and never allow them to influence it. Put offering dates and the applicable quiet-period end date on the research calendar at the time the mandate is signed.