Equities · Reading 46

Financial Statement Forecasting in Equity Valuation

CFA Level I · Equities · Reading 46 · about 46 min

Equities notes as a PDF125 pages · 1.8 MB · free, no sign-up · All PDFs

What you'll learn

LOS 46.a Explain why and how analysts forecast revenues, expenses, assets and financing needs in a disaggregated (financial statement) valuation model.
LOS 46.b Evaluate how a company's life-cycle stage (start-up, growth, mature, decline) shapes the construction of its equity valuation model.
LOS 46.c Estimate and interpret the value of an equity security from the outputs of a financial statement forecast model.

Module 46.1

Financial Statement Forecast Models

This reading explains how analysts build a disaggregated financial statement forecast from revenue, expense, asset and financing models, including top-down revenue growth and financing needs. It then matches the valuation approach to the company's life-cycle stage, with venture capital pre-money and post-money values, growth-stage terminal values and probability-weighted values in decline, and values equity from the model's FCFF and FCFE.

LOS 46.a — Why and how analysts build financial statement forecast models

Single-stage and multistage present value models compress a company's future into one or a few growth rates. A disaggregated valuation model instead forecasts full financial statements line by line. Revenues, expenses, assets and financing can then each follow their own assumptions, and the analyst can bring in industry projections, the competitive environment and company-specific information. The model improves the estimate of future cash flows. It does not change how the discount rate is estimated, and it still needs a terminal value.

A financial statement forecast has four building blocks: revenue modeling, expense modeling (or profit modeling), asset modeling and financing modeling.

Revenue modeling

Key concept

Approach	Starting point	Best suited to	Typical inputs
Historical approach	The company's own past revenue growth (total or by operating segment)	Mature firms with long records and stable business models	Past growth rates, segment disclosures
Bottom-up approach	Company-specific business drivers	Firms whose revenue can be built from units, prices, capacity or outlets	Unit volumes × average selling prices; same-store sales growth plus planned new store openings; capacity (e.g., available seat kilometers) × utilization × price per unit of capacity
Top-down approach	A macroeconomic variable, usually expected nominal GDP growth	Firms whose sales track the economy or an industry	GDP growth and the expected relationship between it and company sales; industry sales growth and market share

Two common top-down calculations:

Growth relative to GDP. When company revenue is expected to grow $x %$ faster than nominal GDP, with the premium stated as a percentage of GDP growth, the company's growth rate is $g_{G D P} (1 + x)$ . A premium stated in percentage points (for example, "2 percentage points above GDP growth") is added instead.
Market growth and market share. Start from the forecast growth of industry sales, then apply the company's expected market share. Company revenue growth is market growth plus the proportional change in market share:

Key concept

g_{r e v e n u e} = g_{ma r k e t} + (\frac{new share}{old share} - 1)

Example. An industry is expected to grow 4%, and a firm's share rises from 20% to 22%. Forecast revenue growth is $4% + (22/20 - 1) = 4% + 10% = 14%$ . If the firm's current sales are $30 million, industry sales are $30/0.20 = $150$ million now and $150 \times 1.04 = $156$ million next year, and firm sales are $0.22 \times 156 = $34.32$ million, growth of 14.4%.

Exam convention: the market growth and market share method adds market growth to the proportional change in share, which gives 14% here; use this figure when a question asks for the forecast growth. Current practice: computing next year's sales in levels gives the exact growth, 14.4% here. The gap is the product of the two changes ( $4% \times 10% = 0.4%$ ), so it matters only when both changes are large.

Expense (profit) modeling

Historical approach. Each expense is linked to revenue using past data, often with a regression that separates fixed and variable components:

Δ Expense = b_{0} + b_{1} \times Δ Revenue

The intercept $b_{0}$ captures the fixed component and the slope $b_{1}$ the variable component. Regressions need a minimum sample size to be reliable, so this approach can be used only for established companies with enough historical data.

Itemizing expenses. Economically significant line items are forecast one by one. This is the approach to use when new information must be reflected, such as a new wage agreement, new supplier terms or a new lease.
Direct profit margin approach. A margin is assumed directly without disaggregating costs. It works well where COGS dominates total cost and moves with revenue:

COGS = forecast revenue \times (1 - gross margin)

Margins are hard to forecast when there are one-off charges such as restructuring costs or impairments. Some items, such as interest expense and interest income, tend not to vary directly with revenues, so they are usually forecast separately.

Example. A regression gives $b_{0} = 0.03$ and $b_{1} = 0.70$ for the change in SG&A. If revenue grows 8%, SG&A grows $0.03 + 0.70 \times 0.08 = 8.6%$ . In a separate case, a firm has revenue of $80 million, expected to grow 6%, and a 30% gross margin. Next year's revenue is $84.8$ , gross profit is $0.30 \times 84.8 = 25.44$ , and COGS is $84.8 - 25.44 = $59.36$ million.

Asset modeling

A constant growth rate model assumes that assets and revenue grow at one common rate, so total asset turnover stays constant.
Alternatively, individual asset categories are forecast separately, possibly with a regression like the one for expenses. Regressions are limited by the timing of capital expenditures, because fixed assets must be in place before they generate revenue.
Only assets and liabilities with cash flow implications belong in the model. A goodwill impairment is a non-cash accounting charge and is left out. Purchases of equipment and changes in inventory involve cash and are included.

Financing modeling

Key concept

financing needs = Δ operating assets - Δ non-debt liabilities - Δ retained earnings

Operating assets here means all assets other than cash.
Non-debt liabilities are liabilities that carry no interest, for example accounts payable and accrued expenses.
The change in retained earnings is net income minus dividends.

If $Δ$ operating assets $-$ $Δ$ non-debt liabilities is positive, the company has a funding need. It can be met from existing cash, retained earnings, borrowing or issuing stock. If the difference is negative, the company has a surplus that can fund dividends or share buybacks, reduce debt or add to cash. With retained earnings included in the formula, a positive result is the amount of new financing needed and a negative result is the size of the surplus.

The financing assumptions need reasonableness checks. The forecast debt ratio should stay within the company's normal range, so a sudden unexplained drop in leverage is a warning sign. Stock issuance or repurchases should be assumed only if the company's history supports them. Cash should be neither implausibly high nor below the company's minimum or target level.

Example. Operating assets rise $50,000, non-debt liabilities rise $18,000, net income is $20,000 and dividends are $5,000. Financing needs are $50, 000 - 18, 000 - (20, 000 - 5, 000) = $17, 000$ .

Common exam traps

Adding the change in market share in percentage points to market growth, instead of the proportional change in share. In the example, adding the 2-point share gain to 4% market growth gives 6% instead of 14%.
Adding a premium to GDP growth when the premium is stated as a percentage of GDP growth.
Reporting gross profit, or this year's COGS, when next year's COGS is asked; applying the gross margin itself as the COGS ratio. In the example, gross profit is $25.44 million and this year's COGS is $56.00 million, while next year's COGS is $59.36 million.
In financing needs: using the level of retained earnings instead of the change, including the change in cash, or adding the change in non-debt liabilities. In the example, adding the $18,000 rise in non-debt liabilities gives $53,000 instead of $17,000.
Confusing bottom-up (prices × volumes, outlets) with top-down (GDP, industry growth and share).
Thinking a constant growth asset model makes asset turnover grow. Turnover stays constant.

Exam shortcuts

The exam-convention market-share growth differs from the exact growth computed in levels only by the product of the two changes, so the two answers are close unless both changes are large.

Bottom line

A disaggregated valuation model forecasts full financial statements through revenue, expense, asset and financing modeling; it improves the estimate of future cash flows but does not change how the discount rate is estimated and still needs a terminal value.
Revenue can be forecast with the historical approach (the company's own past growth, for mature firms with stable business models), the bottom-up approach (company drivers such as units × prices, same-store sales plus new stores, or capacity × utilization × price) or the top-down approach (nominal GDP growth, or industry growth and market share).
A revenue premium stated as a percentage of GDP growth gives growth of $g_{G D P} (1 + x)$ , while a premium stated in percentage points is added to GDP growth.
Exam convention: forecast revenue growth is market growth plus the proportional change in market share, $g_{ma r k e t} + (new share / old share - 1)$ ; current practice: computing next year's sales in levels gives the exact growth.
Expenses can be linked to revenue by a regression $Δ Expense = b_{0} + b_{1} \times Δ Revenue$ for established companies with enough data, itemized when new information must be reflected, or set by a direct margin, with $COGS = forecast revenue \times (1 - gross margin)$ .
A constant growth asset model keeps total asset turnover constant, and financing needs equal the change in operating assets minus the change in non-debt liabilities minus the change in retained earnings, a positive result being new financing needed and a negative one a surplus.

Quick check

Question 1Core

Compared with a present value model that applies one or two growth rates to a single cash flow measure, the main benefit of a disaggregated valuation model built on forecast financial statements is that it:

Show answer and explanation

Correct answer: B

A disaggregated model forecasts revenues, expenses, assets and financing separately, each with its own assumptions. That flexibility lets the analyst reflect industry projections, competition and company-specific information, which improves the cash flow forecast.

Why the other options are wrong

A. A forecast model still covers only a finite horizon, so a terminal value is still needed to capture value beyond it.
C. The model refines the cash flow estimates. The required return is estimated separately.

Key takeaway Disaggregated models improve the numerator (cash flows). They leave the denominator (discount rate) to be estimated separately, and they still need a terminal value.

Module 46.2

Models Based on Company Characteristics

LOS 46.b — Matching the valuation model to the company's life-cycle stage

The right model depends on where the company is in its life cycle.

Key concept

Stage	Typical characteristics	Typical valuation approach
Start-up stage	Low or no revenue, negative earnings and cash flow, little or no debt (so equity value = firm value), little operating history, high chance of failure, funded by venture capital (VC)	A multiple of anticipated future sales (a price-to-sales (P/S) multiple), discounted with the VC investor's ROI multiple
Growth stage	Some historical data; cash flows fund growth objectives; sales scale up and profit margins improve	Forecast margins and growth; most value comes from the terminal value, usually an earnings (P/E) multiple on forecast net income
Mature stage	Stable growth in earnings and cash flow, ample historical data	Constant growth model (e.g., Gordon growth DDM or constant growth FCFF/FCFE)
Decline stage	Declining profit margins, focus on returning capital, possible financial distress	Adjusted historical data, higher discount rates, probability-weighted value of going-concern and distress outcomes

Start-up stage: venture capital arithmetic

VC investors require very high returns because start-ups have a high likelihood of business failure. The required return is expressed as an ROI multiple over the investment horizon:

ROI multiple = (1 + r)^{n} ⟹ r = (ROI multiple)^{1/ n} - 1

where $n$ = years to the VC investor's exit and $r$ = the annual required rate of return. An ROI multiple of 3x means a total profit of 200%.

Key concept

post-money value = \frac{expected revenue \times ( P / S )}{ROI multiple}

pre-money value = post-money value - VC investment

f = \frac{VC investment}{post-money value}

The post-money value is the firm's value immediately after the VC investment, and the pre-money value is its value before the investment. $f$ is the VC investor's fractional ownership.

Example. A start-up needs $5 million. In six years it expects revenue of $40 million and a P/S of 3x, and the VC investor requires a 10x ROI multiple. The post-money value is $40 \times 3/10 = $12$ million, the pre-money value is $12 - 5 = $7$ million, and $f = 5/12 = 41.7%$ . The implied annual return is $1 0^{1/6} - 1 = 46.8%$ .

Growth stage

Forecast revenue, costs and margins to the end of the high-growth period, estimate net income in that year, and apply an earnings multiple to get the terminal value. Discount it to today at the required return, or divide it by the target ROI multiple if the required return is stated as a multiple.

T V_{n} = N I_{n} \times P / E, V_{0} = \frac{T V _{n}}{( 1 + r ) ^{n}} or V_{0} = \frac{T V _{n}}{ROI multiple}

With no debt, $N I = revenue \times operating margin \times (1 - t)$ . If revenue and costs grow at different rates, grow each separately before computing profit.

Example. Year-5 revenue is $25 million and the net margin 16%, so $N I_{5} = $4.0$ million. A P/E of 15 gives $T V_{5} = $60$ million, and at 12% $V_{0} = 60/1.1 2^{5} = $34.05$ million.

Mature stage

Stable growth in earnings and cash flow makes a constant growth model appropriate: $V_{0} = C F_{1} / (r - g)$ .

Decline stage

Financial distress is the inability to meet financial obligations as they fall due. Analysts adjust historical data for the deteriorating business, and use higher discount rates because both debt and equity investors face more risk. Terminal values based on multiples of healthy, stable peers are likely to overvalue a declining firm. When the going-concern assumption is in doubt:

value = P (going concern) \times going-concern value + P (distress) \times distress sale value

When equity is being valued, the distress value is what shareholders recover: sale proceeds net of all creditor claims and liquidation costs. If distress means the assets pass to creditors, the equity's distress value is zero.

Example. Going-concern equity value is $20 million. There is a 40% probability of distress, in which case shareholders would recover $5 million after creditors are paid. Value $= 0.60 \times 20 + 0.40 \times 5 = $14$ million.

Enterprise value and EV multiples

EV = market value of common and preferred stock + market value of debt - cash and short-term investments

Enterprise value (EV) multiples divide EV by an operating income measure such as EBITDA (or by revenue or FCFF). Operating income belongs to all capital providers and is unaffected by capital structure, and EBITDA is less often negative than net income. Common equity value can be backed out as EV less net debt (debt minus cash and short-term investments) and less preferred stock.

Common exam traps

Using revenue (instead of post-money value) as the denominator of fractional ownership; switching the P/S and ROI multiples; reporting post-money when pre-money is asked (or adding the investment). In the example, dividing the $5 million by revenue gives 12.5% instead of 41.7%, and the post-money value of $12 million is not the $7 million pre-money value.
Converting an ROI multiple to an annual return by simple division, or treating 3x as a 300% profit.
Treating a target ROI multiple (e.g., 8x) as a percentage discount rate (8%).
Forgetting to discount the terminal value, or discounting it for only one year. In the example, leaving the $60 million undiscounted, or discounting it one year to $53.57 million, overstates the $34.05 million value.
In a probability-weighted value: ignoring the distress value, or swapping the probabilities. In the example, ignoring the distress value gives $12 million and swapping the probabilities gives $11 million instead of $14 million.

LOS 46.c — Valuing equity from forecast model outputs

Valuation tools from earlier readings are applied here to the outputs of a forecast model.

Free cash flow to the firm (FCFF): $F C F F = N O P A T - net investment$ , with $N O P A T = E B I T (1 - t)$ . Firm value $= F C F F_{1} / (W A C C - g)$ ; equity value $=$ firm value $-$ market value of debt.
Free cash flow to equity (FCFE): $F C F E = N I - investment in net long-term assets - increase in working capital + net borrowing$ . Equity value $= F C F E_{1} / (r - g)$ ; divide by shares outstanding and compare with the market price. Intrinsic value below the price means the stock is overvalued.
Sustainable growth rate based on FCFF:

Key concept

S G R_{F C F F} = (1 - \frac{F C F F}{N O P A T}) (\frac{N O P A T}{beginning book value of total capital})

where the beginning book value of total capital = book value of debt + book value of equity. This gives the sustainable growth rate of FCFF. The first term is the reinvestment rate and the second is the return on capital.

Reading the assumptions. If revenue is forecast to grow faster than expenses, margins are expected to rise. If most operating expenses are forecast to grow faster than revenue, the operating profit margin is expected to fall; the effect on gross and net margins needs more information. A forecast rise in interest expense signals more borrowing, so the funding shortfall is at least partly financed with debt.
Revenue inputs often come from the top-down calculations of LOS 46.a.

Example. EBIT is $90 million and the tax rate 25%, so NOPAT $= 67.5$ . Net investment is 22.5, so FCFF $= 45$ . With a beginning book value of capital of 750, $S GR = (1 - 45/67.5) (67.5/750) = 0.333 \times 0.09 = 3.0%$ . If next year's FCFF is 45, WACC 8.5% and $g = 3%$ , firm value $= 45/0.055 = 818.2$ ; with debt of 200, equity $= 618.2$ .

Common exam traps

Forgetting to subtract debt from firm value, or valuing FCFF with an extra year of growth when it is already a next-year forecast. In the example, reporting firm value gives 818.2 instead of equity of 618.2, and growing the 45 one more year gives equity of 642.7.
Using $F C F F / N O P A T$ instead of $1 - F C F F / N O P A T$ as the reinvestment rate, or dividing NOPAT by the market value of debt instead of by total book capital. In the example, $0.667 \times 0.09$ gives 6.0% instead of 3.0%.
In FCFE, forgetting (or adding back) the increase in working capital, or ignoring net borrowing.

Exam shortcuts

Read margins from the growth assumptions: revenue forecast to grow faster than expenses means rising margins, and most operating expenses growing faster than revenue means a falling operating profit margin, though the effect on gross and net margins needs more information.

Bottom line

A start-up is valued with a P/S multiple of anticipated sales discounted by the VC investor's ROI multiple, a growth-stage firm mainly through a terminal value from a P/E on forecast net income, a mature firm with a constant growth model, and a declining firm with adjusted historical data, higher discount rates and a probability-weighted value.
$ROI multiple = (1 + r)^{n}$ , post-money value is expected revenue × P/S divided by the ROI multiple, pre-money value is post-money value minus the VC investment, and the VC's fractional ownership is the investment divided by post-money value.
A growth-stage terminal value is $T V_{n} = N I_{n} \times P / E$ , discounted at r for n years or divided by the target ROI multiple when the required return is stated as a multiple.
When the going-concern assumption is in doubt, value is the probability-weighted average of the going-concern value and the distress sale value, where the equity's distress value is what shareholders recover after creditor claims and liquidation costs.
From a forecast model, $F C F F = N O P A T - net investment$ gives firm value $F C F F_{1} / (W A C C - g)$ , and equity value is firm value minus the market value of debt.
The sustainable growth rate of FCFF is the reinvestment rate $1 - F C F F / N O P A T$ times the return on capital, NOPAT divided by the beginning book value of total capital (debt plus equity).

Quick check

Question 2Core

A venture capital fund targets an ROI multiple of 5x on a start-up investment that it expects to exit after six years. The annual required rate of return implied by this target is closest to:

Show answer and explanation

Correct answer: A

The ROI multiple compounds the annual required return over the investment horizon: ROI multiple $= (1 + r)^{n}$ . Solve for $r$ .

5 = (1 + r)^{6} \Rightarrow r = 5^{1/6} - 1 = 1.3077 - 1 = 30.8%

Calculator: 5 [y^x] 6 [1/x] [=] gives 1.3077.

Why the other options are wrong

B. 34.8% treats a 5x multiple as a 500% profit, i.e., a multiple of 6x: $6^{1/6} - 1$ . A 5x multiple is a 400% profit.
C. 66.7% divides the 400% total profit by six years ( $(5 - 1) /6$ ), ignoring compounding.

Key takeaway $r = (ROI multiple)^{1/ n} - 1$ . A multiple of $M$ means a total profit of $M - 1$ .

Practice Questions

Question 3Core

Tomasz Wilk is building a forecast for Halvorsen Aluminum. Halvorsen has just signed a four-year energy supply contract that fixes the electricity price for its smelters at a level well above what it paid in the past. Wilk wants his expense forecast to reflect the new contract terms. Which approach to forecasting expenses is most appropriate?

Show answer and explanation

Correct answer: A

Itemizing forecasts economically significant expense lines individually, so new information, such as the terms of a new supply contract, can be built directly into the affected line. Wilk can raise the energy cost line to reflect the contracted price while forecasting other costs separately.

Why the other options are wrong

B. A historical approach (for example, a regression of each expense on revenue) assumes the past relationship between expenses and revenue will continue. The new contract breaks that relationship for energy costs, so history would misstate them.
C. A direct profit margin approach assumes a margin without breaking costs into components, which makes it hard to capture the effect of one specific contract on one cost line.

Key takeaway New, specific information about a cost, such as a labor agreement, supplier terms or a lease, calls for itemizing.

Question 4Core

An investor estimates that Wexcombe Textiles has an 85% probability of surviving as a going concern, in which case its equity would be worth $46 million. If it does not survive, the investor expects a distress sale to leave shareholders $10 million after all creditor claims and sale costs are paid. The value of Wexcombe's equity is closest to:

Show answer and explanation

Correct answer: B

For a firm whose going-concern assumption is in doubt, equity value is the probability-weighted average of the going-concern equity value and the distress value to shareholders (here $10 million, already net of creditor claims).

value = 0.85 \times 46 + 0.15 \times 10 = 39.1 + 1.5 = $40.6 million

Why the other options are wrong

A. $46.0 million is the going-concern value alone; it ignores the 15% probability of distress.
C. $15.4 million swaps the probabilities, weighting the going-concern value by 15% and the distress value by 85%: $0.15 \times 46 + 0.85 \times 10$ .

Key takeaway value = P(going concern) × going-concern value + P(distress) × distress sale value; the two probabilities sum to 1.

Question 5Core

An analyst's forecast for Carrow Retail assumes that revenue will grow 3% a year, while most of Carrow's operating expenses, including SG&A and depreciation, will grow 5% to 6% a year. Based only on this information, which of the following is most likely?

Show answer and explanation

Correct answer: C

When most operating expenses grow faster than revenue, operating expenses take a larger share of each dollar of sales, so the operating profit margin is expected to fall. No calculation is needed.

Why the other options are wrong

A. Gross profit margin depends only on revenue and COGS, and the other operating costs listed do not affect it. The stem does not say how fast COGS will grow, so a steady gross margin cannot be concluded.
B. Net profit margin also depends on non-operating items such as interest income and expense, which are not given. A falling operating margin, if anything, pushes it down.

Key takeaway Compare growth rates. Revenue growing more slowly than operating costs means a lower operating margin; revenue growing faster means a higher one.

This reading has 54 questions in the full bank. Practice all of them.

Key Takeaways

Disaggregated models improve the numerator (cash flows). They leave the denominator (discount rate) to be estimated separately, and they still need a terminal value.
New, specific information about a cost, such as a labor agreement, supplier terms or a lease, calls for itemizing.
$r = (ROI multiple)^{1/ n} - 1$ . A multiple of $M$ means a total profit of $M - 1$ .
value = P(going concern) × going-concern value + P(distress) × distress sale value; the two probabilities sum to 1.
Compare growth rates. Revenue growing more slowly than operating costs means a lower operating margin; revenue growing faster means a higher one.