Aller au contenu
Corporate & Investment Banking Mon parcours English

Accueil / Corporate Finance / Valeur et investissement / Module 23

23Measuring value creation

How to tell whether management really made shareholders richer

Savoir si la direction a vraiment enrichi les actionnaires.

Lecture
20 min
Sections
11
Exercices
3
Questions de rappel
6
Questions de QCM
12
Simulateurs
3

Le texte de ce module est encore en anglais : la version française arrive. Les outils interactifs, eux, sont déjà en français.

What you will be able to do

  • Sort any performance measure into accounting, hybrid, cash or market, and say what it rewards.

  • Compute economic profit and show why its discounted sum equals NPV.

  • Explain why EPS growth and ROE growth can accompany value destruction, with numbers.

  • Compute total shareholder return and judge it against a fair benchmark.

  • Design a bonus measure and list how a manager could game it.

1Four families of measures, one underlying truth

Before any formula, a map. Measures differ by where their numbers come from, and that decides how easily they are distorted.

Measures differ by their source, and the source decides how they fail

Accounting, hybrid, cash and market measures use different raw material

Accounting measures take figures straight from the income statement and balance sheet. Hybrid measures start from accounting figures and add a charge for the capital used. Cash-based measures count money that moved. Market measures use share prices, so they reflect what investors expect rather than what has happened.

Family Examples Raw material Main weakness
Accounting EPS, net income, EBIT, EBITDA, ROE, ROCE Reported profit and book capital Ignore the cost of capital; depend on accounting choices
Hybrid Economic profit (EVA), cash value added Accounting profit minus a capital charge Need restated capital; easy to shorten the horizon
Cash NPV, cash flow return, free cash flow yield Forecast or actual cash flows Forecasts are opinions; rarely published
Market Market value added, total shareholder return Share prices Move with the whole market, not only with management

Four families of measures. EBITDA has no standard definition in IFRS, so two companies can report different figures under the same label.

The measures agree on value only in one idealised world

Suppose a company's value is the present value of all its future cash flows, discounted at the required return. Then each new project adds its own NPV, and nothing else adds anything. A measure is good if it moves with the sum of those NPVs. It is bad if it moves for another reason, such as the choice of depreciation method or the mix of debt and equity.

The real world adds a complication: managers know more than investors, and the measure that guides the company is often the one that pays them. Once a number is attached to a bonus, it stops being a neutral gauge. Part 4 returns to this.

Marc runs a family farm. He judges his year by the profit in his accounts and by whether his land is worth more. Explain to him, using a savings account as your comparison, why he should also ask what the money tied up in land and machines could have earned elsewhere.

      Marc's profit says what he earned. It does not say what the land and machines could have earned if sold and the money placed elsewhere. If his savings account would pay 3% on the value of his land and he earns 2% on it, he is working hard to be poorer than if he had sold up.

      The right question is whether the farm earns more than the best alternative use of the same money at the same risk.

      Value creation is a spread, and the spread must be positive

      Return on capital minus the cost of capital is the test

      Think of a company as a pool of capital that earns a return. Investors who provided the capital want at least the cost of capital, for Harmony 9% (WACC, from CF25). The difference between what the company earns and what investors want is the spread. A positive spread means value is created. A zero spread means the company is fair value. A negative spread destroys value even when profit is positive.

      Spread = ROCE after tax − WACC

      ROCE after tax is operating profit after tax divided by capital employed, the money tied up in the business at book value (CF5). For Harmony in year N, operating profit is 12.0, tax at 25% takes 3.0, leaving 9.0. Capital employed is 84.0, so ROCE is 10.7%. The spread is 10.7% − 9.0% = 1.7 points.

      1

      1. Which family of measures is the only one whose figures come from share prices?

      2. Harmony's after-tax ROCE is 10.7% and its WACC 9.0%. What does the 1.7-point spread tell you?

      3. A firm retains its profit and reinvests it at 5% while investors require 9%. What happens?

      2Accounting measures: easy to read, easy to bend

      EPS, ROE and ROCE are the numbers most often quoted. Each is useful in its place and misleading outside it.

      EPS growth is not value growth

      Earnings per share ignores how much capital produced the earnings

      Earnings per share (EPS) is net income divided by the number of shares. Harmony earns 7.5 million on 10.0 million shares, so EPS is 0.75. A company can raise EPS by earning more, by shrinking the share count, or by replacing equity with debt, which lifts EPS as long as the business earns more than the after-tax interest rate, while adding risk that EPS does not show. Only the first is certain to help shareholders.

      A debt-financed acquisition lifts EPS even when it destroys value

      The rule that falls out is simple. A deal paid for with debt raises EPS whenever the target's earnings yield, its net income divided by the price, exceeds the after-tax cost of the debt. For a deal paid with shares, EPS rises when the target's earnings yield exceeds the acquirer's own, which is the inverse of its P/E ratio. Harmony's earnings yield is 0.75 / 11.40 = 6.6%, its P/E of 15.2 inverted.

      Neither rule mentions value. The target can be overpriced, riskier than Harmony, or slower growing, and EPS still rises. Accretion is a statement about arithmetic. Module CF39 returns to it for share-financed mergers.

      a small deal simulator with sliders for the price paid, the target's net income, the interest rate and the share of the price paid in new shares. It displays EPS before and after, and next to it the value created or lost compared with a fair value that the learner can also set. The learner finds deals where EPS rises while value falls, and deals where EPS falls while value rises.

      A share buyback can lift EPS without any gain for shareholders

      Take the same logic in reverse. Suppose Harmony borrows 10.0 million and uses it to buy back its own shares at 11.40. It retires 10.0 / 11.40 = 0.877 million shares, leaving 9.123 million. Interest after tax is 10.0 × 5% × 0.75 = 0.375, so net income falls to 7.125. EPS is 7.125 / 9.123 = 0.781, up from 0.75.

      The company is not worth more. Its equity value fell by the cash paid out, and remaining shareholders hold a more indebted company. CF31 explains when buybacks are a sound way to return cash.

      ROE and ROCE reward the wrong behaviour at the margin

      ROE rises with debt, so it cannot compare differently financed firms

      Return on equity (ROE) is net income divided by book equity. Harmony earns 7.5 on equity of 54.0, an ROE of 13.9%. After the buyback above, equity falls to 44.0 and ROE becomes 7.125 / 44.0 = 16.2%. Nothing in the business changed. Only the financing did. ROE is the right measure for a bank, whose business is leverage, and a risky one for most others. ROCE ignores the financing mix and is better for judging operations.

      ROCE can tell managers to refuse good projects and keep bad assets

      A manager paid on ROCE has two quiet ways to improve it. One is to refuse any project that earns less than the company's current ROCE, even when it earns more than the cost of capital. The other is to let the asset base age.

      So a good rule compares a project's return with the cost of capital, not with the company's average return. A company that never replaces old machines can post a splendid ROCE while its plant decays.

      2.1 Test the accretion rule

      Harmony proposes buying a small competitor, Cadence, that earns 1.2 million after tax. The price is 24.0 million, paid entirely in new Harmony shares issued at 11.40.

      (a) How many new shares are issued? (b) What is EPS afterwards, with Harmony at 7.5 million of net income on 10.0 million shares? (c) A banker says the deal is dilutive and so destroys value. Is the reasoning correct?

      (a) 24.0 / 11.40 = 2.105 million new shares.

      (b) Net income becomes 7.5 + 1.2 = 8.7 on 12.105 million shares: EPS is 0.719, a fall of 4.2% from 0.75.

      (c) The reasoning is wrong. The target's earnings yield is 1.2 / 24.0 = 5.0%, below Harmony's 6.6%, so the deal dilutes EPS. But dilution alone says nothing about value. If Cadence's cash flows are worth more than 24.0, for example because it grows fast, Harmony's shareholders gain. Only the NPV of the deal decides.

      1. A company cuts its share count by 10% with borrowed money. Name two reasons EPS may rise and shareholders gain nothing.

      2. Why is ROE a poor yardstick to compare a heavily indebted firm with a cash-rich one?

      3. A division manager is paid on ROCE and turns down a project earning 11% when the cost of capital is 9% and divisional ROCE is 14%. Who gains and who loses?

      1. Earnings fall by less than the share count, and the extra debt raises risk, which EPS does not show. Shareholders own a smaller, more indebted company worth the same in total.

      2. ROE rises with debt even if operations are unchanged, so a heavily indebted firm looks better only because it is riskier.

      3. The manager looks better on ROCE and the company is poorer. The project earns 2 points above the cost of capital, yet it would pull divisional ROCE down from 14%. Shareholders lose the project's NPV.

      2

      1. A debt-financed acquisition raises EPS whenever…

      2. Harmony borrows 10.0 to buy back shares, and ROE rises from 13.9% to 16.2%. Why?

      3. Why can a manager paid on ROCE destroy value?

      3Economic profit: charging for the capital

      The hybrid family fixes the main flaw of accounting measures by subtracting a cost for the capital used. It links ratios to NPV exactly.

      Economic profit is the profit left after investors are paid

      Net income charges for debt but not for equity

      Net income deducts interest, the cost of debt, but not any cost for equity. A company can show a profit while investors, who supplied equity, earn less than they require. Economic profit, popularised as EVA (economic value added) from the 1990s, corrects this by charging for all capital.

      Economic profit = operating profit after tax − WACC × capital employed = capital employed × (ROCE − WACC)

      Economic profit by division shows where capital works and where it idles

      The measure extends naturally to parts of a company. Suppose Harmony splits its 9.0 of after-tax operating profit between two divisions: complete bikes and accessories. The table shows why EVA is used to allocate capital inside groups.

      Year N, € million Capital employed Profit after tax ROCE Capital charge at 9% Economic profit
      Complete bikes 40.0 5.6 14.0% 3.60 +2.00
      Accessories 44.0 3.4 7.7% 3.96 −0.56
      Harmony 84.0 9.0 10.7% 7.56 +1.44

      The group's positive total hides a division that destroys value.

      Accessories earns a profit and looks healthy in the accounts. It earns less than the capital it ties up costs. Management should either improve its return, shrink its capital or release it to bikes.

      3.1 Where should the next 10 million go?

      Next year Harmony has 10.0 million to invest in either bikes or accessories. Bikes could earn 12% after tax on it. Accessories could earn 8%. (a) Compute the economic profit added by each use at a 9% cost of capital. (b) What should happen to the 10.0 million, and what does a group ROCE target of 10.7% suggest?

      Bikes at 12%: 10.0 × (12% − 9%) = +0.30 a year. Accessories at 8%: 10.0 × (8% − 9%) = −0.10 a year.

      Put the money in bikes. A group ROCE target of 10.7% would say the same here, but it would also reject an accessories project earning 10%, which beats the cost of capital. Economic profit would accept it.

      Discounted economic profits add up to NPV

      The two methods measure the same thing at different speeds

      Economic profit is not a rival to NPV. Add up the discounted economic profits of a project over its life and you obtain its NPV exactly. NPV tells you the surplus in one number up front. EVA tells you each year how much of it is arriving, which is why it suits annual targets.

      The identity holds only if capital employed is measured sensibly. Book capital is distorted by old assets, written-off goodwill, expensed intangibles and leases. Since IFRS 16, leased assets enter capital employed, lease liabilities enter net debt and lease costs leave operating costs, so ROCE and EVA must be restated on the same basis as the cost of capital. These restatements are the method's weak point.

      Cash-based variants avoid the depreciation problem

      Cash flow return measures, usually called CFROI, compare the cash a business generates with the gross cost of its assets, before depreciation has eroded them. A simplified version divides EBITDA by gross capital. This avoids the ageing trap above, but on its own ignores that assets wear out and must be replaced. The full method allows for this by treating the return as a rate earned over the assets' expected life.

      a table of the two-year project with a toggle that switches the capital-employed line between historical cost and cost net of depreciation, and a slider for the cost of capital. The learner sees yearly ROCE and EVA move with the toggle while the discounted total of EVA always equals the NPV.

      Nadia manages a hospital department. She is told it will be judged on economic profit. Explain to her, with her own equipment budget as an example, why a department can show a surplus in its accounts and still be charged a negative economic profit.

          Suppose her department uses 2 million of equipment and rooms, and the hospital's cost of capital is 5%. The capital charge is 2 × 5% = 0.10 million a year. If the department shows a surplus of 0.06 million, its economic profit is −0.04 million.

          The accounts say it is in surplus. Economic profit says the resources it holds could have produced more elsewhere in the hospital.

          3

          1. How is Harmony's economic profit of 1.44 million computed?

          2. Under what condition does the discounted sum of a project's economic profits equal its NPV?

          3. Harmony's accessories division has positive profit but negative economic profit. What should management do?

          4Market measures and the link to pay

          Share prices summarise what investors expect. They are the final judge, but a noisy one, and they are what most pay plans use.

          Market value added shows what investors believe, not what happened

          MVA is the gap between what the market pays for the business and what was put in

          Market value added (MVA) compares the value of the company in the market with the capital that investors have put in at book value. In theory it is the present value of all future economic profits.

          MVA = market capitalisation + net debt − capital employed ≈ market capitalisation − book equity

          MVA is a stock, not a flow. It moves when expectations change even though nothing in the business has: after a crash it collapses for nearly every company.

          Total shareholder return is what an investor actually earns

          Total shareholder return (TSR) adds the capital gain and the dividends received, as a percentage of the price paid at the start. It is what a shareholder who held the stock experienced, and it is the only measure in the module that corresponds to money investors actually received.

          TSR = (end price − start price + dividends received) / start price

          TSR needs a benchmark. A 14.5% return sounds fine, but if peers returned 18% a year, management lagged even though shareholders gained. TSR also depends on the start date: measured from a crash it flatters, from a peak it punishes.

          4.1 Harmony against its peers

          Harmony's shares were bought at 12.00. One year later the price is 13.20 and a dividend of 0.30 has been paid. A peer group of comparable listed bike makers returned 18% over the same year.

          (a) Compute Harmony's one-year TSR. (b) Did it beat the peers? (c) Give one reason this single year proves little about management.

          (a) (13.20 − 12.00 + 0.30) / 12.00 = 1.50 / 12.00 = 12.5%.

          (b) No. The peer group returned 18%, so Harmony lagged by 5.5 points even though shareholders gained.

          (c) One year of prices is dominated by market noise and timing. A multi-year window against a fair benchmark is needed.

          Pay linked to value works only if the yardstick resists gaming

          Every measure attached to a bonus will be optimised

          Agency theory says managers act in their own interest unless contracts align them with shareholders (CF1, CF35). Bonus and share plans are those contracts, but a measure chosen for pay becomes a target, and a target gets managed.

          Yardstick What it rewards How it is gamed Best use
          EPS growth Higher earnings per share Buybacks, cheap acquisitions, cost cuts that hurt the future Short-term sanity check
          ROCE or ROE Efficient use of book capital Refusing projects, ageing assets, leverage Operating units with stable assets
          Economic profit Return above the cost of capital Short horizons, capital restatements Annual bonus and capital allocation
          Free cash flow Cash generation Delaying capex and payments Mature businesses
          Relative TSR Beating peers in the market Short-term price support, luck Multi-year share plans

          Good plans combine horizons: an annual metric on economic profit or cash, a multi-year share plan on relative TSR, and a holding requirement so managers keep shares after vesting. Some plans add non-financial targets such as emissions. These are judgements about what the board values, and must be audited like any figure.

          a decision game in which the learner is the remuneration committee. They choose a bonus metric from the table, then watch a simulated manager over five years take the actions that maximise it. The learner sees the metric improve while value, shown as NPV, rises or falls, and discovers which combinations of metrics leave the fewest loopholes.

          1. Why does a sharp rise in the whole market lift a company's MVA and its CEO's pay even if management did nothing?

          2. A plan pays on three-year relative TSR against ten peers. Name one benefit and one weakness.

          3. Why can a firm with positive net income and positive ROCE still have negative economic profit?

          1. MVA is market value less book value, and the market rises with sentiment. A plan measured in absolute price or MVA pays for the market, not for management, so benchmarks matter.

          2. Benefit: it filters out market-wide movements. Weakness: with ten peers one outlier distorts the ranking, and managers can be rewarded for luck within a short window.

          3. Net income is positive after interest and ignores a charge for equity. Economic profit subtracts the cost of all capital at WACC, so a return below WACC gives a negative figure even when ROCE is positive.

          4

          1. Harmony after listing: market capitalisation 150.0, net debt 15.0, capital employed 96.5. What is its MVA?

          2. Why must TSR be judged against peers and over several years?

          3. Which bonus design leaves the fewest loopholes?

          Summary and cheat sheet

          Les formules du module

          Touchez un terme pour voir ce qu’il veut dire et sa valeur chez Harmony.

          L’écart qui crée la valeur

          = −

          Profit économique

          = −

          Même résultat par l’écart : 84,0 × (10,7 % − 9,0 %) = 1,44.

          Valeur ajoutée de marché

          = + −

          Rendement total pour l’actionnaire

          = ( − + ) / P₀

          Idea What to remember
          Four families Accounting, hybrid, cash and market measures differ by source and by how they fail.
          The test Return on capital minus cost of capital. Positive spread, value created.
          NPV The only exact measure. Hard to publish, so others stand in for it.
          EPS Ignores capital and risk. Can rise through debt, buybacks or overpaid deals.
          Accretion rule Debt deal: EPS up if target earnings yield beats after-tax interest. Share deal: if it beats the acquirer's earnings yield.
          ROE Rises with leverage. Fine for banks, poor for comparing other firms.
          ROCE Ignores the cost of capital and drifts up as assets age.
          Economic profit Profit after tax minus WACC times capital employed. Equals capital employed times the spread.
          EVA and NPV Discounted economic profits sum to NPV when capital is measured consistently.
          MVA Market capitalisation minus book equity (when net debt is at book). Shows expectations, not results.
          TSR Price gain plus dividends over cost. Needs several years and a peer benchmark.
          Pay Mix an annual hybrid or cash measure with a multi-year market measure.

          The one-sentence version

          A measure of performance is only as good as its link to net present value: accounting numbers ignore the cost of capital, economic profit charges for it, and market numbers need a benchmark and a long window.

          Key terms

          Term Plain-language meaning
          EPS Net income divided by the number of shares.
          Capital employed Money tied up in the business at book value: fixed assets plus working capital.
          Spread Return on capital minus the cost of capital.
          Economic profit (EVA) Operating profit after tax minus a charge for all capital employed.
          Earnings yield Net income divided by price, the inverse of the P/E ratio.
          Accretive / dilutive A deal that raises or lowers earnings per share.
          Market value added Market capitalisation minus the book value of equity.
          Total shareholder return Share price gain plus dividends, as a percentage of the starting price.
          Relative TSR Total shareholder return compared with a peer group over the same period.

          Going further

          • Vernimmen, Quiry, Le Fur — Finance d'entreprise (Dalloz), chapters on value and corporate finance and on measures of value creation.

          • Brealey, Myers, Allen — Principles of Corporate Finance, chapters on measuring performance and incentives.

          • Stewart, G. B. — The Quest for Value (HarperBusiness, 1991), on economic profit and EVA.

          Independent educational material. Not investment, legal or tax advice. All companies and figures in the running case are fictional.