Analyzing Income Statements

This reading covers how revenue and expenses are recognized, from the five-step revenue model to the choice between capitalizing and expensing, and how nonrecurring items and accounting changes are reported. The calculations to master are revenue on long-term contracts, basic and diluted EPS, and common-size income statements with their margins.

LOS 28.a — Revenue recognition: principles, applications and analysis

The accrual idea

Revenue belongs in the period in which it is earned, that is, when the company has done what it promised. This need not be the period in which cash arrives.

• Credit sale: revenue is recognized at the sale, and an asset, accounts receivable, is recorded.
• Cash received in advance (subscriptions, deposits, gift cards): no revenue is recognized yet. The company records cash and a liability, unearned (deferred) revenue. As the goods or services are delivered, the liability is reduced and revenue is recognized.
• Revenue is reported net of expected returns, discounts and allowances. Exam convention: estimated warranty provisions count as one of the allowances subtracted from revenue when the sale is made. Current practice: a standard (assurance-type) warranty is recognized at the sale as an expense with a matching provision, and a warranty sold as a separate service is a performance obligation whose share of the price is deferred.

The converged standard (IFRS 15 and ASC 606)

The IASB and FASB jointly developed this converged revenue standard. It is principles-based. Instead of many industry-specific rules, it rests on one core principle: recognize revenue when (or as) control of a promised good or service passes to the customer, at the amount the firm expects to be entitled to. The principle is applied through a five-step model:

Indicators that control has transferred are that the customer has physical possession, has accepted the item, bears the risks and rewards of ownership and holds legal title, and that the seller has a present right to payment.

Exam convention: revenue is recognized only when a later reversal is highly unlikely, so a firm may have to record a refund liability, with an asset for its right to the returned goods, when revenue from a sale cannot be estimated reliably. Current practice: under IFRS 15 and ASC 606, every sale with a right of return gives revenue only for the amount not expected to be refunded, a refund liability for expected refunds and an asset for the goods expected back.

Revenue over time (long-term contracts)

Progress can be measured by inputs (cost incurred as a share of total expected cost) or by outputs (milestones, units delivered). Costs of securing a long-term contract, such as sales commissions, are capitalized and then expensed over the life of the contract.

A building is usually one performance obligation. Site preparation, foundations, wiring and the roof are not distinct promises, because each is an input to a single combined output that the customer contracted for.

• Year 1: $8/25=32\%$, so revenue is $0.32\times 30=\$9.6$ million.
• Year 2: $(8+7)/25=60\%$, so cumulative revenue is $\$18.0$ million and Year 2 revenue is $18.0-9.6=\$8.4$ million.

The formula always uses the transaction price as currently estimated. Had the bonus become includable by the end of Year 2, cumulative revenue would be $0.60\times 33=\$19.8$ million and Year 2 revenue $19.8-9.6=\$10.2$ million. Year 1 revenue is not restated, so the bonus share of the Year 1 work is recognized in Year 2.

The amounts billed or collected do not affect these figures. The result is the same as under the older percentage-of-completion method, although the converged standard no longer uses that name.

Specific applications

Disclosures and analysis

Firms must disclose contracts by category, contract assets and liabilities (and changes in them), remaining performance obligations with the price allocated to them, and the judgments used on timing and amount. Analysts should watch the choices that move revenue between periods: the measure of progress, estimates of variable consideration, contract modifications, and principal-versus-agent classification.

LOS 28.b — Expense recognition, capitalization and expensing

How expenses are recognized under accrual accounting

An expense is a decrease in economic benefits (assets used up or liabilities incurred) that lowers equity other than through distributions to owners. Under the accrual basis, an expense is recognized when the benefit is consumed, whether or not cash has been paid. Three routes achieve this:

Production or purchase costs stay on the balance sheet as inventory until the goods are sold. If nothing is sold in a period, none of those costs reaches the income statement.

A policy that recognizes expenses later is aggressive because it raises current income. A policy that recognizes them sooner is conservative, so expensing an outlay is more conservative than capitalizing it.

Recording a credit sale

When goods costing 70 are sold on credit for 100, receivables rise by 100 and inventory falls by 70, so total assets rise by 30. Retained earnings (equity) rise by the profit of 30, and liabilities are unaffected (ignoring taxes).

Inventory cost-flow methods

Specific identification assigns each unit its own actual cost. It suits goods that can be tracked one by one, such as vehicles recorded by serial number. When individual units cannot be tracked this way, the firm assumes a cost flow:

Revenue is unaffected by the method, so differences in COGS pass directly into margins. Analysts comparing firms must adjust for different methods.

Depreciation methods

The depreciable base is cost minus salvage value. Straight-line spreads it evenly over the useful life.

Intangible assets

• Finite-life intangibles (patents, licenses, customer lists) are amortized over their useful lives.
• Indefinite-life intangibles (acquired goodwill, some trademarks or licenses renewable at little cost) are not amortized. They are tested for impairment at least annually under both IFRS and US GAAP.
• Most internally generated intangibles are expensed. Research costs (aimed at new scientific or technical knowledge) are expensed. Under IFRS, development costs (turning research findings into a plan or design for a new product or process) may be capitalized. Among other criteria, the firm must show that it is able to finish the asset and plans to use or sell it. Under US GAAP, R&D is generally expensed; the main exception is software developed for sale, which is capitalized after technological feasibility is established.
• When one firm capitalizes development costs and a peer expenses them, to make the two comparable, an analyst expenses the period's development spending and removes the amortization of earlier capitalized amounts from the income statement. On the balance sheet the capitalized asset is removed, which lowers assets and equity. In the cash flow statement the outlay moves from CFI to CFO, so CFO falls.

Capitalizing and expensing

An outlay expected to bring economic benefits over more than one period is capitalized. If the future benefit is unlikely or highly uncertain, it is expensed. The chart shows the decision and the route by which each kind of capitalized cost later reaches the income statement (see the figure below). A capitalized asset's cost includes the purchase price plus the costs needed to get the asset ready for use (freight, installation, non-recoverable taxes). Except for land and indefinite-life intangibles, that cost is then spread over the asset's life as depreciation, depletion or amortization. Later outlays that extend the asset's life or add capacity are also capitalized. Routine repairs, maintenance and staff training are expensed.

Total cash flow is the same under both choices, provided the tax treatment does not depend on the accounting choice; only the classification between CFO and CFI differs. If the tax deduction follows the accounting choice, expensing brings the whole tax saving into the year of the outlay. Total cash flow is then higher in that year under expensing, CFO is higher in later years under capitalization (depreciation keeps reducing tax), and cumulative cash is the same by the end of the asset's life.

• Capitalize: depreciation is $4,000, so net income is $(25,000-4,000)(1-0.25)=\$15,750$ every year. CFO is $15,750+4,000=\$19,750$ every year, and CFI is −$20,000 in Year 1.
• Expense: Year 1 net income is $(25,000-20,000)(1-0.25)=\$3,750$, which is also Year 1 CFO. Years 2–5 net income and CFO are $25,000\times 0.75=\$18,750$.
• Year 1 total cash flow is −$250 when capitalizing and $3,750 when expensing. Capitalization then gives $1,000 more CFO in each of Years 2–5, and five-year cash totals are $78,750 either way.
• Cumulative net income over the five years is also $78,750 under both choices $(5\times 15,750=3,750+4\times 18,750)$. The choice moves income between years; it does not change the total.

The table describes a single outlay. A firm that capitalizes similar costs every year keeps reporting higher income than an expensing peer for as long as the amount capitalized each period exceeds that period's depreciation or amortization of the capitalized costs.

Capitalized interest

Interest incurred while a firm builds an asset for its own use must be capitalized as part of the asset's cost under both IFRS and US GAAP. In limited cases this also applies to an asset built for sale. Capitalized interest is not shown as interest expense. It later reaches income through depreciation (asset held for use) or COGS (asset held for sale), and the cash paid appears in investing cash flow. As with any capitalized cost, net income is higher while the asset is being built and lower in later years. In those later years EBIT is also lower, because the interest now sits inside depreciation, an operating expense. Exam convention: interest paid that is not capitalized is an operating outflow under US GAAP and an operating or financing outflow under IFRS. Current practice: under IFRS 18, effective for annual periods beginning on or after 1 January 2027, a company whose main business is not financing reports interest paid in CFF.

For interest coverage, analysts add the interest capitalized in the period to interest expense. They also add the depreciation of previously capitalized interest back to EBIT, because it is economically an interest cost rather than an operating cost.

Analyzing estimates

Bad-debt, warranty, useful-life and salvage estimates give management room to shift income. Analysts should ask why an estimate changed and compare it with peers. The footnotes and MD&A disclose accounting policies and significant estimates.

LOS 28.c — Nonrecurring items and accounting changes

Analysts forecast sustainable earnings, so they need to know which income statement items are likely to repeat. The standards give certain items special placement or treatment.

Unusual or infrequent items

Unusual or infrequent items are material items that are unusual in nature or infrequent in occurrence. Examples are:

• gains or losses on selling assets or parts of the business outside ordinary operations;
• impairments, write-downs and write-offs;
• restructuring charges.

They are reported within income from continuing operations, before tax, often as a separate line. Analysts should judge whether they are really one-off: a company with "unusual" charges almost every year is in effect reporting a recurring cost.

Discontinued operations

A discontinued operation is a component that management has decided to dispose of (and that is classified as held for sale) or has already disposed of during the period. Exam convention: the business must be physically and operationally separable from the rest of the firm in its assets, operations, and investing and financing activities. Current practice: IFRS 5 also requires a separate major line of business or geographical area (or a subsidiary bought only for resale), and US GAAP requires a strategic shift with a major effect on operations, so selling a minor product line or individual assets is not a discontinued operation.

• The measurement date is the date a formal disposal plan is adopted. The phaseout period runs from then to the actual disposal.
• The operation's income or loss, and any gain or loss on disposal, is reported after income from continuing operations, net of tax. Prior-period income statements presented are restated to show it separately.
• Once the component is held for sale, its assets are measured at the lower of carrying amount and fair value less costs to sell. An expected loss on disposal is therefore recognized immediately as a write-down. An expected gain is recognized only when the sale is completed.
• Exam convention: on the measurement date the firm accrues the estimated operating losses of the phaseout period along with any estimated loss on the sale. Current practice: IAS 37 prohibits provisions for future operating losses other than for onerous contracts, and US GAAP reports the results of a discontinued operation in the periods in which they occur. Under both frameworks, phaseout-period operating losses are therefore reported in discontinued operations as they are incurred.
• Analysts usually exclude discontinued operations when forecasting earnings, although the disposal itself can signal something about future cash flows.

Accounting changes

The depreciation of $100,000 per year for Years 1–3 is not restated; the new amount applies from Year 4 onward.

Changes in scope and exchange rates

Acquisitions (changes in scope, which alter the size of the combined entity) and currency movements (which change the reporting-currency value of foreign sales, purchases and subsidiaries' results) can distort year-on-year comparisons, even though the standards do not require their effects to be isolated. Analysts should adjust for them where possible.

Where items appear

The forecast therefore starts from income from continuing operations of $25.5 million. If the restructuring is truly a one-off, recurring after-tax income is $25.5+12\times (1-0.25)=\$34.5$ million; if Brackwell restructures most years, the $25.5 million figure stands.

LOS 28.d — Basic and diluted earnings per share

Earnings per share (EPS) shows how much of the period's profit belongs to each common share. Only companies with publicly traded shares must report it, and it is always expressed per common share. There is no EPS figure for preferred shares.

Simple and complex capital structures

Straight (nonconvertible) debt and nonconvertible preferred stock cannot create new common shares, so they do not make a capital structure complex. Other features, such as a call feature, a cumulative dividend or a sinking fund, do not change this.

Exam convention: a company with a simple capital structure reports basic EPS only. Current practice: that is the US GAAP rule (ASC 260). IFRS (IAS 33) requires every company within its scope to present both basic and diluted EPS; with no dilutive potential shares the two figures are equal.

Basic EPS

• The numerator is income available to common shareholders. Subtract the current year's preferred dividends.
• Common dividends are not subtracted. They are a distribution of income that already belongs to common shareholders.
• Convertible securities do not enter basic EPS at all: there is no add-back in the numerator and no extra shares in the denominator.

Weighted average shares outstanding

Each block of shares is weighted by the fraction of the year it was outstanding. Strictly, the weight is days outstanding divided by days in the year; exam questions normally use months.

• New shares (sold for cash, issued in an acquisition, or treasury shares reissued) count from the issue date.
• Repurchased (treasury) shares drop out from the repurchase date.
• Stock dividends and stock splits, including a reverse stock split, are applied retroactively. Every share outstanding before the event is restated as if the event had happened at the start of the year, and prior-year figures are restated too, so EPS does not appear to fall because of a largely cosmetic change in the share count. Shares issued or repurchased after the event are not adjusted for it.
• A split or stock dividend brings in no new resources and leaves each holder's percentage of the company unchanged. Its date within the year therefore does not change how the shares outstanding before it are restated: a 2-for-1 split in late March and one in early November both double the opening shares for the full year.
• A 3-for-2 split and a 50% stock dividend have the same effect: each old share becomes 1.5 new shares. A 1-for-4 reverse split multiplies pre-event shares by 0.25.
• Issuing preferred shares has no effect on the common share count.

The $50,000 of common dividends plays no part in the calculation. When comparing EPS across years, check whether a change comes from earnings or from the share count: repurchases can raise EPS while net income and margins stay flat.

Diluted EPS and the if-converted method

Diluted EPS shows the worst case for existing shareholders: what EPS would be if every dilutive security had been converted or exercised at the start of the year, or on its issue date if it was issued during the year.

Convertibles are handled with the if-converted method, which assumes conversion even if the terms did not allow it during the period.

Convertible preferred stock. If the shares were converted, the preferred dividends would not be paid. Add them back to the numerator (they were subtracted to get income available to common) and add the conversion shares to the denominator. There is no tax adjustment, because preferred dividends are not tax-deductible.

Convertible bonds. If the bonds were converted, the interest would not be paid, but the tax deduction on that interest would also be lost. Add back the after-tax interest, $\text{Interest}\times (1-t)$, and add the conversion shares. If the bonds were issued part-way through the year, weight both the interest add-back and the shares by the fraction of the year the bonds were outstanding.

Options and warrants. These are handled with the treasury stock method. There is no numerator adjustment, because exercise saves no expense. Assume exercise, and assume that the cash received is used to repurchase shares at the average market price for the period:

where $N$ is the number of shares issuable, $X$ is the exercise price and $\bar{P}$ is the average market price. Options are dilutive only when $X<\bar{P}$, that is, when they are in the money on average. The year-end price plays no part.

Testing for dilution

A security is dilutive if including it lowers EPS, and antidilutive if including it would raise EPS or reduce a loss per share. Antidilutive securities are excluded from diluted EPS, so diluted EPS can never exceed basic EPS.

A quick test for a convertible is to compute its per-share effect and compare it with basic EPS:

If the ratio is below basic EPS, the security is dilutive; if it is above, the security is antidilutive. Each potentially dilutive security is tested on its own.

When a company reports a net loss available to common shareholders, adding shares would make the loss per share smaller. All potential shares are then antidilutive, and diluted EPS equals basic EPS.

Basic EPS $=(3,000,000-300,000)/900,000=\$3.00$.

Including the bonds would raise EPS to $3,120,000/1,200,000=\$2.60$, which confirms that they are antidilutive. The $58 year-end price plays no role.

LOS 28.e — Common-size income statements and margin analysis

Vertical common-size income statement

In a vertical common-size income statement, every line is divided by revenue (sales) for the same period, so revenue $=100\%$. This removes the effect of size and lets an analyst compare:

• the same firm across years (time-series or trend analysis), and
• different firms in the same period (cross-sectional analysis), however different their scale.

A common-size figure shows the cost per unit of revenue. It does not show the absolute amount. If cost of sales goes from 55% to 60% of revenue, it has risen relative to sales; whether it rose in currency terms depends on what happened to revenue.

Tax is the exception. Tax expense is more informative relative to pretax income:

The rate can be computed directly from common-size percentages because revenue cancels out. For example, tax of 6% of sales divided by pretax income of 24% of sales gives $25\%$.

Margin ratios

Any subtotal divided by revenue is a margin:

The differences between successive margins isolate each cost layer, all as a percentage of revenue:

Gross margin improves when a firm raises prices or cuts unit production costs. Product differentiation (brand, technology, patents) supports higher prices. Administrative costs do not affect gross margin. Net margin reflects every cost, including financing and tax. Both should be compared over time and against peers.

The table below shows what moves gross margin when there are no fixed production costs. A product sells for $50 and costs $30 per unit to make, a 40% gross margin.

Common-size statements are also a starting point for strategy analysis. A firm that spends a larger share of revenue on research or marketing may be building the product differentiation that later shows up as a higher gross margin.

• COGS is a steady 58% of revenue.
• Operating expenses fell from 26% to 22% of revenue, so the whole improvement comes from cost control below gross profit.
• Nonoperating expense is a stable 2%, so financing costs are unchanged relative to sales.
• The effective tax rate is $3.5/14=4/16=4.5/18=25\%$ every year.