McKinsey on Finance

Perspectives on Corporate Finance and Strategy Number 15, Spring 2005

Do fundamentals—or emotions—drive the stock market? 1 Emotions can drive market behavior in a few short-lived situations. But fundamentals still rule. The right role for multiples in valuation 7 A properly executed multiples analysis can make financial forecasts more accurate. Governing joint ventures 12 Better oversight isn’t just for wholly owned businesses. Merger valuation: Time to jettison EPS 17 Assessing an aquisition’s value by estimating its likely impact on earnings per share has always been flawed. Now it’s likely to be flat wrong.

McKinsey on Finance is a quarterly publication written by experts and practitioners in McKinsey & Company’s Corporate Finance practice. This publication offers readers insights into value-creating strategies and the translation of those strategies into company performance. This and archived issues of McKinsey on Finance are available online at www.corporatefinance.mckinsey.com. Editorial Contact: [email protected] Editorial Board: James Ahn, Richard Dobbs, Marc Goedhart, Bill Javetski, Timothy Koller, Robert McNish, Dennis Swinford Editor: Dennis Swinford Design and Layout: Kim Bartko Design Director: Donald Bergh Managing Editor: Kathy Willhoite Editorial Production: Sue Catapano, Roger Draper, Thomas Fleming, Scott Leff, Mary Reddy Circulation: Susan Cocker Cover illustration by Ben Goss Copyright © 2005 McKinsey & Company. All rights reserved. This publication is not intended to be used as the basis for trading in the shares of any company or for undertaking any other complex or significant financial transaction without consulting appropriate professional advisers. No part of this publication may be copied or redistributed in any form without the prior written consent of McKinsey & Company.

Correction: In the Autumn 2004 issue, “The scrutable East” included an exhibit that forecast GDP in billions of dollars; this unit of measure should have been trillions. Our apologies.

1

Running head

Do fundamentals—or emotions— drive the stock market?

Emotions can drive market behavior in a few short-lived situations. But fundamentals still rule.

Marc Goedhart,

There’s never been a better time to

Timothy Koller, and

be a behaviorist. During four decades, the academic theory that financial markets accurately reflect a stock’s underlying value was all but unassailable. But lately, the view that investors can fundamentally change a market’s course through irrational decisions has been moving into the mainstream.

David Wessels

1 For an overview of behavioral finance, see Jay

R. Ritter, “Behavioral finance,” Pacific-Basin Finance Journal, 2003, Volume 11, Number 4, pp. 429–37; and Nicholas Barberis and Richard H. Thaler, “A survey of behavioral finance,” in Handbook of the Economics of Finance: Financial Markets and Asset Pricing, G. M. Constantinides et al. (eds.), New York: Elsevier North-Holland, 2003, pp. 1054–123.

With the exuberance of the high-tech stock bubble and the crash of the late 1990s still fresh in investors’ memories, adherents of the behaviorist school are finding it easier than ever to spread the belief that markets can be something less than efficient in immediately distilling new information and that investors, driven by emotion, can indeed lead markets awry. Some behaviorists would even assert that stock markets lead lives of their own, detached from economic growth and business profitability. A number of finance scholars and practitioners have argued that stock markets are not efficient—that is, that they don’t necessarily reflect economic fundamentals.1 According to this point of view, significant and lasting deviations from the intrinsic value of a company’s share price occur in market valuations. The argument is more than academic. In the 1980s the rise of stock market index funds, which now hold some $1 trillion

in assets, was caused in large part by the conviction among investors that efficientmarket theories were valuable. And current debates in the United States and elsewhere about privatizing Social Security and other retirement systems may hinge on assumptions about how investors are likely to handle their retirement options. We agree that behavioral finance offers some valuable insights—chief among them the idea that markets are not always right, since rational investors can’t always correct for mispricing by irrational ones. But for managers, the critical question is how often these deviations arise and whether they are so frequent and significant that they should affect the process of financial decision making. In fact, significant deviations from intrinsic value are rare, and markets usually revert rapidly to share prices commensurate with economic fundamentals. Therefore, managers should continue to use the triedand-true analysis of a company’s discounted cash flow to make their valuation decisions. When markets deviate

Behavioral-finance theory holds that markets might fail to reflect economic fundamentals under three conditions. When all three apply, the theory predicts that pricing biases in financial markets can be both significant and persistent. Irrational behavior. Investors behave irrationally when they don’t correctly process all the available information while forming their expectations of a company’s future performance. Some investors, for example, attach too much importance to recent events and results, an error that leads them to overprice companies with strong recent performance. Others are excessively conservative and underprice stocks of companies that have released positive news.

2

McKinsey on Finance

Systematic patterns of behavior. Even if individual investors decided to buy or sell without consulting economic fundamentals, the impact on share prices would still be limited. Only when their irrational behavior is also systematic (that is, when large groups of investors share particular patterns of behavior) Behavioral-finance theory argues that should persistent patterns of overconfidence, overreaction, price deviations occur. and overrepresentation are common to Hence behavioralmany investors and that such groups can finance theory be large enough to prevent a company’s argues that patterns of overconfidence, share price from reflecting underlying overreaction, and economic fundamentals overrepresentation are common to many investors and that such groups can be large enough to prevent a company’s share price from reflecting underlying economic fundamentals—at least for some stocks, some of the time. Limits to arbitrage in financial markets. When investors assume that a company’s recent strong performance alone is an indication of future performance, they may start bidding for shares and drive up the price. Some investors might expect a company that surprises the market in one quarter to go on exceeding expectations. As long as enough other investors notice this myopic overpricing and respond by taking short positions, the share price will fall in line with its underlying indicators. This sort of arbitrage doesn’t always occur, however. In practice, the costs, complexity, and risks involved in setting up a short position can be too high for individual investors. If, for example, the share price doesn’t return to its fundamental value while they can still hold on to a short position—the so-called noise-

Spring 2005

trader risk—they may have to sell their holdings at a loss. Momentum and other matters

Two well-known patterns of stock market deviations have received considerable attention in academic studies during the past decade: long-term reversals in share prices and short-term momentum. First, consider the phenomenon of reversal—high-performing stocks of the past few years typically become low-performing stocks of the next few. Behavioral finance argues that this effect is caused by an overreaction on the part of investors: when they put too much weight on a company’s recent performance, the share price becomes inflated. As additional information becomes available, investors adjust their expectations and a reversal occurs. The same behavior could explain low returns after an initial public offering (IPO), seasoned offerings, a new listing, and so on. Presumably, such companies had a history of strong performance, which was why they went public in the first place. Momentum, on the other hand, occurs when positive returns for stocks over the past few months are followed by several more months of positive returns. Behavioralfinance theory suggests that this trend results from systematic underreaction: overconservative investors underestimate the true impact of earnings, divestitures, and share repurchases, for example, so stock prices don’t instantaneously react to good or bad news. But academics are still debating whether irrational investors alone can be blamed for the long-term-reversal and short-termmomentum patterns in returns. Some believe that long-term reversals result

3

Do fundamentals—or emotions—drive the stock market?

merely from incorrect measurements of a stock’s risk premium, because investors ignore the risks associated with a company’s size and market-to-capital ratio.2 These statistics could be a proxy for liquidity and distress risk.

the number of available Palm shares was extremely small after the carve-out: 3Com still held 95 percent of them. As a result, it was extremely difficult to establish a short position, which would have required borrowing shares from a Palm shareholder.

Similarly, irrational investors don’t necessarily drive short-term momentum in share price returns. Profits from these patterns are relatively limited after transaction costs have been deducted. Thus, small momentum biases could exist even if all investors were rational.

During the months following the carveout, the mispricing gradually became less pronounced as the supply of shares through short sales increased steadily. Yet while many investors and analysts knew about the price difference, it persisted for two months—until the Internal Revenue Service formally approved the carve-out’s tax-free status in early May 2002. At that point, a significant part of the uncertainty around the spin-off was removed and the price discrepancy disappeared. This correction suggests that at least part of the mispricing was caused by the risk that the spin-off wouldn’t occur.

Furthermore, behavioral finance still cannot explain why investors overreact under some conditions (such as IPOs) and underreact in others (such as earnings announcements). Since there is no systematic way to predict how markets will respond, some have concluded that this is a further indication of their accuracy.3

Two well-documented types of market deviation—the mispricing of carve-outs and of dual-listed companies—are used to support behavioral-finance theory. The classic example is the pricing of 3Com and Palm after the latter’s carve-out in March 2000.

Additional cases of mispricing between parent companies and their carved-out subsidiaries are well documented.4 In general, these cases involve difficulties setting up short positions to exploit the price differences, which persist until the spin-off takes place or is abandoned. In all cases, the mispricing was corrected within several months.

In anticipation of a full spin-off within nine months, 3Com floated 5 percent of its Palm subsidiary. Almost immediately, Palm’s market capitalization was higher than the entire market value of 3Com, implying that 3Com’s other businesses had a negative value. Given the size and profitability of the rest of 3Com’s businesses, this result would clearly indicate mispricing. Why did rational investors fail to exploit the anomaly by going short on Palm’s shares and long on 3Com’s? The reason was that

A second classic example of investors deviating from fundamentals is the price disparity between the shares of the same company traded on two different exchanges. Consider the case of Royal Dutch Petroleum and “Shell” Transport and Trading, which are traded on the Amsterdam and London stock markets, respectively. Since these twin shares are entitled to a fixed 60–40 portion of the dividends of Royal Dutch/Shell, you would expect their share prices to remain in this fixed ratio.

Persistent mispricing in carve-outs and dual-listed companies

2 Eugene F. Fama and Kenneth R. French,

“Multifactor explanations of asset pricing anomalies,” Journal of Finance, 1996, Volume 51, Number 1, pp. 55–84. 3 Eugene F. Fama, “Market efficiency, long-term

returns, and behavioral finance,” Journal of Financial Economics, 1998, Volume 49, Number 3, pp. 283–306. 4 Owen A. Lamont and Richard H. Thaler,

“Can the market add and subtract? Mispricing in tech stock carve-outs,” Journal of Political Economy, 2003, Volume 111, Number 2, pp. 227–68; and Mark L. Mitchell, Todd C. Pulvino, and Erik Stafford, “Limited arbitrage in equity markets,” Journal of Finance, 2002, Volume 57, Number 2, pp. 551–84.

��� 4 ��������

McKinsey on Finance

Spring 2005

�������������� ��������������������������������������� ���������

smaller. Furthermore, some of these share structures (and price differences) disappeared because the corporations formally merged, a development that underlines the significance of noise-trader risk: as soon as a formal date was set for definitive price convergence, arbitrageurs stepped in to correct any discrepancy. This pattern provides additional evidence that mispricing occurs only under special circumstances—and is by no means a common or long-lasting phenomenon.

����������� ������������������������������������������������������������������ ������������������������������������������������������

���������������������������

��

��

��

��

��

��

��

��

��

��

�

�

���

���

���

���

���

���

��� ����

�����

�����

�����

�����

�����

�����

��� ����

�����

�����

�����

�����

�����

�����

Markets and fundamentals: The bubble of the 1990s

�����������������������������������������������

���������������

Over long periods, however, they have not. In fact, prolonged periods of mispricing can be found for several similar twin-share structures, such as Unilever (Exhibit 1). This phenomenon occurs because large groups of investors prefer (and are prepared to pay a premium for) one of the twin shares. Rational investors typically do not take positions to exploit the opportunity for arbitrage.

5 US corporate earnings as a percentage of GDP

have been remarkably constant over the past 35 years, at around 6 percent. 6 Marc H. Goedhart, Timothy M. Koller, and

Zane D. Williams, “The real cost of equity,” McKinsey on Finance, Number 5, Autumn 2002, pp. 11–5. 7 Marc H. Goedhart, Timothy M. Koller,

and Zane D. Williams, “Living with lower market expectations,” McKinsey on Finance, Number 8, Summer 2003, pp. 7–11.

Do markets reflect economic fundamentals? We believe so. Long-term returns on capital and growth have been remarkably consistent for the past 35 years, in spite of some deep recessions and periods of very strong economic growth. The median return on equity for all US companies has been a very stable 12 to 15 percent, and long-term GDP growth for the US economy in real terms has been about 3 percent a year since 1945.5 We also estimate that the inflation-adjusted cost of equity since 1965 has been fairly stable, at about 7 percent.6

Thus in the case of Royal Dutch/Shell, a price differential of as much as 30 percent has persisted at times. Why? The opportunity to arbitrage dual-listed stocks is actually quite unpredictable and potentially costly. Because of noise-trader risk, even a large gap between share prices is no guarantee that those prices will converge in the near term.

We used this information to estimate the intrinsic P/E ratios for the US and UK stock markets and then compared them with the actual values.7 This analysis has led us to three important conclusions. The first is that US and UK stock markets, by and large, have been fairly priced, hovering near their intrinsic P/E ratios. This figure was typically around 15, with the exception of the highinflation years of the late 1970s and early 1980s, when it was closer to 10 (Exhibit 2).

Does this indict the market for mispricing? We don’t think so. In recent years, the price differences for Royal Dutch/Shell and other twin-share stocks have all become

Second, the late 1970s and late 1990s produced significant deviations from intrinsic valuations. In the late 1970s, when investors were obsessed with high

5

Do fundamentals—or emotions—drive the stock market?

short-term inflation rates, the market was probably undervalued; long-term real GDP growth and returns on equity indicate that it shouldn’t have bottomed out at P/E levels of around 7. The other well-known deviation occurred in the late 1990s, when the market reached a P/E ratio of around 30—a level that couldn’t be justified by 3 percent long-term real GDP growth or by 13 percent returns on book equity. Third, when such deviations occurred, the stock market returned to its intrinsicvaluation level within about three years. Thus, although valuations have been wrong from time to time—even for the stock market as a whole—eventually they have fallen back in line with economic fundamentals. Focus on intrinsic value

What are the implications for corporate managers? Paradoxically, we believe that such market deviations make it even more important for the executives of a company to understand the intrinsic value of its shares. This knowledge allows it to exploit any deviations, if and when they occur, ��� to time the implementation of strategic �������� decisions more successfully. Here are some �������������� examples of how corporate managers can ������������������������������������������������������������������������������������� take advantage of market deviations. ���������

�������������������������������� ����������������������������������������������� ����

����

����

��������������������

�

��

�������������������

�

��

�������������������������

�

��

������������������������������������������������

� ��������������������������������������������

���� �� ��

�� ��

��

��

• Issuing additional share capital when the stock market attaches too high a value to the company’s shares relative to their intrinsic value • Repurchasing shares when the market underprices them relative to their intrinsic value • Paying for acquisitions with shares instead of cash when the market overprices them relative to their intrinsic value • Divesting particular businesses at times when trading and transaction multiples are higher than can be justified by underlying fundamentals Bear two things in mind. First, we don’t recommend that companies base decisions to issue or repurchase their shares, to divest or acquire businesses, or to settle transactions with cash or shares solely on an assumed difference between the market and intrinsic value of their shares. Instead, these decisions must be grounded in a strong business strategy driven by the goal of creating shareholder value. Market deviations are more relevant as tactical considerations when companies time and execute such decisions—for example, when to issue additional capital or how to pay for a particular transaction. Second, managers should be wary of analyses claiming to highlight market deviations. Most of the alleged cases that we have come across in our client experience proved to be insignificant or even nonexistent, so the evidence should be compelling. Furthermore, the deviations should be significant in both size and duration, given the capital and time needed to take advantage of the types of opportunities listed previously.

6

McKinsey on Finance

Provided that a company’s share price eventually returns to its intrinsic value in the long run, managers would benefit from using a discounted-cash-flow approach for strategic decisions. What should matter is the long-term behavior of the share price of a company, not whether it is undervalued by 5 or 10 percent at any given time. For strategic business decisions, the evidence strongly suggests that the market reflects intrinsic value. MoF

Spring 2005

Marc Goedhart ([email protected])

is an associate principal in McKinsey’s Amsterdam office, and Tim Koller (Tim_Koller@McKinsey .com) is a partner in the New York office. David Wessels ([email protected]),

an alumnus of the New York office, is an adjunct professor of finance at the Wharton School of the University of Pennsylvania. This article is adapted from the authors’ forthcoming book, Valuation: Measuring and Managing the Value of Companies, fourth edition, Hoboken, New Jersey: John Wiley & Sons, available at www.mckinsey.com/valuation. Copyright © 2005 McKinsey & Company. All rights reserved.

7

Running head

The right role for multiples in valuation

A properly executed multiples analysis can make financial forecasts more accurate.

Marc Goedhart,

Senior executives know that not all

Timothy Koller, and

valuation methods are created equal. In our experience, managers dedicated to maximizing shareholder value gravitate toward discounted-cash-flow (DCF) analyses as the most accurate and flexible method for valuing projects, divisions, and companies. Any analysis, however, is only as accurate as the forecasts it relies on. Errors in estimating the key ingredients of corporate value—ingredients such as a company’s return on invested capital (ROIC), its growth rate, and its weighted average cost of capital—can lead to mistakes in valuation and, ultimately, to strategic errors.

David Wessels

We believe that a careful analysis comparing a company’s multiples with those of other companies can be useful in making such forecasts, and the DCF valuations they inform, more accurate. Properly executed, such an analysis can help a company to stress-test its cash flow forecasts, to understand mismatches between its performance and that of its competitors, and to hold useful discussions about whether it is strategically positioned to create more value than other industry players are. As a company’s executives seek to understand why its multiples are higher or lower than those of the competition, a multiples analysis can also generate insights into the key factors creating value in an industry.

Yet multiples are often misunderstood and, even more often, misapplied. Many financial analysts, for example, calculate an industry-average price-to-earnings ratio and multiply it by a company’s earnings to establish a “fair” valuation. The use of the industry average, however, overlooks the fact that companies, even in the same industry, can have drastically different expected growth rates, returns on invested capital, and capital structures. Even when companies with identical prospects are compared, the P/E ratio itself is subject to problems, since net income commingles operating and nonoperating items. By contrast, a company can design an accurate multiples analysis that provides valuable insights about itself and its competitors. When multiples mislead

Every week, research analysts at Credit Suisse First Boston (CSFB) report the stock market performance of US retailers by creating a valuation table of comparable companies (exhibit). To build the weekly valuation summary, CSFB tracks each company’s weekend closing price and market capitalization. The table also reports the projections by CSFB’s staff for each company’s future earnings per share (EPS). To compare valuations across companies, the share price of each of them is divided by its projected EPS to obtain a forward-looking P/E ratio. To derive The Home Depot’s forward-looking P/E of 13.3, for instance, you would divide the company’s weekend closing price of $33 by its projected 2005 EPS of $2.48. But which companies are truly comparable? For the period covered in the exhibit, Home Depot and its primary competitor, Lowe’s, traded at nearly identical multiples. Their P/E ratios differed by only 8 percent, and their enterprise-value-to-EBITDA (earnings

������������ ��������� 8 McKinsey on Finance �������������� �����������������������������������������������

Spring 2005

��������

�������������������

��������������� ���������������

������� ���������������� ���������

������������������� �������� �

���������������� ���������������

����

����

�������

����

���������������� �������������� ������

����� �����

������ ������

���� ����

���� ����

��� ���

���� ����

���������������� ����������������� ����������������

����� �����

������ �����

���� ����

���� ����

��� ���

���� ����

�������������������� �������� �������������������

����� �����

������ �����

���� ����

���� ����

��� ���

���� ����

��������

���

�����

�����

���

����

��������������� �������

������������������������������������������������������������������������������������ �������������������������������������������������������������������������������������������������������������

�������������������������������������������������������������������������

before interest, taxes, depreciation, and amortization) ratios1 by only 3 percent. But this similarity doesn’t extend to a larger set of hard-lines retailers, whose enterprise multiples vary from 4.4 to 9.9. Why such a wide range? Investors have different expectations about each company’s ability to create value going forward, so not every hard-lines retailer is truly comparable. To choose the right companies, you have to match those with similar expectations for growth and ROIC .

1 Enterprise value equals market capitalization

plus debt and preferred shares less cash not required for operations. 2 Nidhi Chadda, Robert S. McNish, and Werner

Rehm, “All P/Es are not created equal,” McKinsey on Finance, Number 11, Spring 2004, pp. 12–5.

A second problem with mutiples is that different ones can suggest conflicting conclusions. Best Buy, for instance, trades at a premium to Circuit City Stores when measured using their respective enterprisevalue multiples (6.3 versus 4.4) but at a discount according to their P/E ratios (13.8 versus 22.3). Which is right—the premium or the discount? It turns out that Circuit City’s P/E multiple isn’t meaningful. In July 2004, the total equity value of this

company was approximately $2.7 billion, but it held nearly $1 billion in cash. Since cash generates very little income, its P/E ratio is high; a 2 percent after-tax return on cash translates into a P/E of 50. So the extremely high P/E of cash artificially increases the company’s aggregate P/E . When you remove cash from the equity value ($2.7 billion – $1 billion) and divide by earnings less after-tax interest income ($122 – $8), the P/E drops from 22.3 to 14.9. Finally, different multiples are meaningful in different contexts. Many corporate managers believe that growth alone drives multiples. In reality, growth rates and multiples don’t move in lockstep.2 Growth increases the P/E multiple only when combined with healthy returns on invested capital, and both can vary dramatically across companies. Executives and investors must pay attention to growth and to returns on capital or a company might achieve its growth objectives but forfeit the benefits of a higher P/E . The well-tempered multiple

Four basic principles can help companies apply multiples properly: the use of peers with similar ROIC and growth projections, of forward-looking multiples, and of enterprise-value multiples, as well as the adjustment of enterprise-value multiples for nonoperating items. 1. Use peers with similar prospects for ROIC and growth

Finding the right companies for the comparable set is challenging; indeed, the ability to choose appropriate comparables distinguishes sophisticated veterans from newcomers. Most financial analysts start by examining a company’s industry—but industries are often loosely defined. The company might list its competitors in its

9

The right role for multiples in valuation

annual report. An alternative is to use the Standard Industrial Classification codes published by the US government. A slightly better (but proprietary) system is the Global Industry Classification Standard (GICS) recently developed by Morgan Stanley Capital International and Standard & Poor’s.

3 A note of caution about forward multiples:

some analysts forecast future earnings by assuming an industry multiple and using the current price to back out the required earnings. As a result, any multiple calculated from such data will reflect merely the analyst’s assumptions about the appropriate forward multiple, and dispersion (even when warranted) will be nonexistent. 4 Jing Liu, Doron Nissim, and Jacob K.

Thomas, “Equity valuation using multiples,” Journal of Accounting Research, Volume 40, Number 1, pp. 135–72. 5 To forecast the price of a company, the authors

multiplied its earnings by the industry median multiple. Pricing error equals the difference between the forecast price and the actual price, divided by the actual price. 6 Moonchul Kim and Jay R. Ritter, “Valuing

IPO s,” Journal of Financial Economics,

Volume 53, Number 3, pp. 409–37.

With an initial list of comparables in hand, the real digging begins. You must examine each company on the list and answer some critical questions: why are the multiples different across the peer group? Do certain companies in it have superior products, better access to customers, recurring revenues, or economies of scale? If these strategic advantages translate into superior ROICs and growth rates, the companies that have an edge within an industry will trade at higher multiples. You must become an expert on the operating and financial specifics of each of the companies: what products they sell, how they generate revenue and profits, and how they grow. Not until you have that expertise will a company’s multiple appear in the appropriate context with other companies. In the end, you will have a more appropriate peer group, which may be as small as one. In order to evaluate Home Depot, for instance, only Lowe’s remains in our final analysis, because both are pure-play companies earning the vast majority of their revenues and profits from just a single business. 2. Use forward-looking multiples

Both the principles of valuation and the empirical evidence lead us to recommend that multiples be based on forecast rather than historical profits.3 If no reliable forecasts are available and you must rely on historical data, make sure to use the latest data possible—for the most recent

four quarters, not the most recent fiscal year—and eliminate one-time events. Empirical evidence shows that forwardlooking multiples are more accurate predictors of value. Jing Liu, Doron Nissim, and Jacob Thomas, for example, compared the characteristics and performance of historical and forward industry multiples for a subset of companies trading on the NYSE , the American Stock Exchange, and Nasdaq.4 When they compared individual companies against their industry mean, the dispersion of historical earnings-to-price (E/P) ratios was nearly twice that of oneyear forward E/P ratios. The three also found that forward-looking multiples promoted greater accuracy in pricing. They examined the median pricing error for each multiple to measure that accuracy.5 The error was 23 percent for historical multiples and to 18 percent for one-year forecasted earnings. Two-year forecasts cut the median pricing error to 16 percent. Similarly, when Moonchul Kim and Jay Ritter compared the pricing power of historical and forecast earnings for 142 initial public offerings, they found that the latter had better results.6 When the analysis moved from multiples based on historical earnings to multiples based on one- and two-year forecasts, the average prediction error fell from 55.0 percent, to 43.7 percent, to 28.5 percent, respectively, and the percentage of companies valued within 15 percent of their actual trading multiple increased from 15.4 percent, to 18.9 percent, to 36.4 percent, respectively. 3. Use enterprise-value multiples

Although widely used, P/E multiples have two major flaws. First, they are systematically affected by capital structure.

10

McKinsey on Finance

Spring 2005

For companies whose unlevered P/E (the ratio they would have if entirely financed by equity) is greater than one over the cost of debt, P/E ratios rise with leverage. Thus, a company with a relatively high allequity P/E can artificially increase its P/E ratio by swapping debt for equity. Second, the P/E ratio is based on earnings, which include many nonoperating items, such as restructuring charges and write-offs. Since these are often one-time events, multiples based on P/E s can be misleading. In 2002, for instance, what was then called AOL Time Warner wrote off nearly $100 billion in goodwill and other intangibles. Even though the EBITA (earnings before interest, taxes, and amortization) of the company equaled $6.4 billion, it recorded a $98 billion loss. Since earnings were negative, its P/E ratio wasn’t meaningful.

generate misleading results. (Despite the common perception that multiples are easy to calculate, calculating them correctly takes time and effort.) Here are the most common adjustments.

One alternative to the P/E ratio is the ratio of enterprise value to EBITA . In general, this ratio is less susceptible to manipulation by changes in capital structure. Since enterprise value includes both debt and equity, and EBITA is the profit available to investors, a change in capital structure will have no systematic effect. Only when such a change lowers the cost of capital will changes lead to a higher multiple. Even so, don’t forget that enterprise-value-to-EBITA multiples still depend on ROIC and growth.

• Excess cash and other nonoperating assets. Since EBITA excludes interest income from excess cash, the enterprise value shouldn’t include excess cash. Nonoperating assets must be evaluated separately. • Operating leases. Companies with significant operating leases have an artificially low enterprise value (because the value of lease-based debt is ignored) and an artificially low EBITA (because rental expenses include interest costs). Although both affect the ratio in the same direction, they are not of the same magnitude. To calculate an enterprisevalue multiple, add the value of leased assets to the market value of debt and equity. Add the implied interest expense to EBITA . • Employee stock options. To determine the enterprise value, add the present value of all employee grants currently outstanding. Since the EBITAs of companies that don’t expense stock options are artificially high, subtract new employee option grants (as reported in the footnotes of the company’s annual report) from EBITA .

4. Adjust the enterprise-value-to-EBITA multiple for nonoperating items

Although the one-time nonoperating items in net income make EBITA superior to earnings for calculating multiples, even enterprise-value-to-EBITA multiples must be adjusted for nonoperating items hidden within enterprise value and EBITA , both of which must be adjusted for these nonoperating items, such as excess cash and operating leases. Failing to do so can

• Pensions. To determine the enterprise value, add the present value of pension liabilities. To remove the nonoperating gains and losses related to pension plan assets, start with EBITA , add the pension interest expense, deduct the recognized returns on plan assets, and adjust for any accounting changes resulting from changed assumptions (as indicated in the footnotes of the company’s annual report).

11

The right role for multiples in valuation

Other multiples too can be worthwhile, but only in limited situations. Price-to-sales multiples, for example, are of limited use for comparing the valuations of different companies. Like enterprise-value-to-EBITA multiples, they assume that comparable companies have similar growth rates and returns on incremental investments, but they also assume that the companies’ existing businesses have similar operating margins. For most industries, this restriction is overly burdensome. PEG ratios7 are more flexible than

traditional ratios by virtue of allowing the expected level of growth to vary across companies. It is therefore easier to extend comparisons across companies in different stages of the life cycle. Yet PEG ratios do have drawbacks that can lead to errors in valuation. First, there is no standard time frame for measuring expected growth; should you, for instance, use one-year, twoyear, or long-term growth? Second, these ratios assume a linear relation between multiples and growth, such that no growth implies zero value. Thus, in a typical implementation, companies with low growth rates are undervalued by industry PEG ratios.

7 PEG multiples are created by comparing

a company’s P/E ratio with its underlying growth rate in earnings per share.

For valuing new companies (such as dotcoms in the late 1990s) that have small sales and negative profits, nonfinancial multiples can help, despite the great uncertainty surrounding the potential market size and profitability of these companies or the investments they require. Nonfinancial multiples compare enterprise value to a nonoperating statistic, such as Web site

hits, unique visitors, or the number of subscribers. Such multiples, however, should be used only when they lead to better predictions than financial multiples do. If a company can’t translate visitors, page views, or subscribers into profits and cash flow, the nonfinancial metric is meaningless, and a multiple based on financial forecasts will provide a superior result. Also, like all multiples, nonfinancial multiples are only relative tools; they merely measure one company’s valuation compared with another’s. As the experience of the late 1990s showed, an entire sector can become detached from economic fundamentals when investors rely too heavily on relativevaluation methods.

Of the available valuation tools, a discounted-cash-flow analysis delivers the best results. Yet a thoughtful analysis of multiples also merits a place in any valuation tool kit. MoF Marc Goedhart ([email protected])

is an associate principal in McKinsey’s Amsterdam office, and Tim Koller (Tim_Koller@McKinsey .com) is a partner in the New York office. David Wessels ([email protected]),

an alumnus of the New York office, is an adjunct professor of finance at the Wharton School of the University of Pennsylvania. This article is adapted from the authors’ forthcoming book, Valuation: Measuring and Managing the Value of Companies, fourth edition, Hoboken, New Jersey: John Wiley & Sons, available at www.mckinsey.com/valuation. Copyright © 2005 McKinsey & Company. All rights reserved.

12

McKinsey on Finance

Governing joint ventures

Better oversight isn’t just for wholly owned businesses.

James Bamford and

Corporate governance has become a top

David Ernst

priority for executives of public companies. Yet too few of them have raised the bar for governing joint ventures, whose financialmanagement systems, most executives tell us, just aren’t as good as those of wholly owned businesses. Such systems, we hear, don’t regularly incorporate joint ventures into the standard corporate-planning and review process, and parent companies don’t pay enough attention to them. Where standards exist at all, they are informal and vary quite widely. This neglect is risky. Most large companies today have ten or more sizable joint ventures accounting for 10 to 20 percent of their annual revenues, income, or assets. And in our experience, the effects of weak governance—chronic underperformance, a failure to adapt and evolve, and excessive managerial costs—help sink many such partnerships. More than a decade ago, the California Public Employees’ Retirement System (Calpers), hoping to improve the performance of corporate boards, established a set of corporate-governance guidelines. Applying them to joint ventures, of course, calls for adjustments. Nonetheless, we believe that such guidelines will not only help companies (and perhaps

Spring 2005

their public shareholders) to assess the governance of their existing joint ventures more clearly but also make their executives better informed when they enter into new joint ventures. Indeed, guidelines such as these can make the difference between good and bad governance—between governance that is fast, accountable, and transparent, on the one hand, and prone to gridlock, weak performance management, mistrust, and stagnation, on the other. Our experience with many large joint ventures suggests that improving their governance can help their corporate parents to change their strategy, scope, financial arrangements, and operations and to identify and reduce the risks they face. Those risks can be considerable: in a recent case, an industrial conglomerate discovered, 18 months after the fact, that one of its joint ventures had exposed it to a $400 million liability. Understand the challenge

Shareholders of public companies are typically united by a common desire to optimize returns—in the form, for example, of dividends and share price levels—and to manage risks. By contrast, the governance of joint ventures, which receive ongoing operational resources from a few large shareholders, is much more complicated (exhibit). In joint ventures, for example, board members (and others involved in the governance system) must manage what may be the divergent strategic and economic interests of the parent companies, which often have very different constraints, views of the market, and means of profiting from the business. Likewise, the boards of joint ventures must secure and oversee the flow of operational resources (including

������� Governing joint ventures ������������� �������������� ��������������������������������������������������������������������� ��������

��������������������������������������������������������������� ������������������������������������������������� ����������������������������������������������������������� ����������������������������������������������������� ��������������������������������������������������������������

of the basic tenets of good corporate governance can—and should—be applied to both. The principles discussed here might be seen as the minimum needed to promote accountability, speed, transparency, and, ultimately, performance.1

�������������������������������������������������������� ������������������������������������������������� ����������������������������������������������������� ����������������������������������������������������� �����������������������������������������������������������

Appoint at least one outside director

����������������������������������� ������������������������������������

���������������������������������������������������

��������������� ����������������� �������������������������������������������������� ������������������������������������������������ ���������������������� ������������������������������������������������ � ����������� ����������������� �������������������������������������������������������� ������������������������������������������������ ��������������������������������� ������������������������������������������������ ��������������� ��������������� �������������������������������������������������� ����������������������������������������������� ���������������������� ���������� �

��������������������������������������������������������������� ��������������������������������������������������� ����������������������������������������������������������� ���������������������������������������������������������� �������������������������������������������������������������� �����������������������������������������������

������������� ��������������� ������������������������������������������������ ����������������������������������������������� ������������������� ����������

���������������������������������������������������� ��������������������������������������������������� �������������������������������������������������������������� ���������������������������������������������������������� ����������������������������������������������� �����������������������������������������������

����������� ������������� ������������������������������������������������ �������������������������������������������������� ������������������� ���������������������������������������

���������������������������������������������������� �������������������������������������������������������� �������������������������������������������������������������� ����������������������������������������������������� ����������������������������������������������� �����������������������������������������������������������

technology, raw materials, and staff) between them and their parent companies. These boards must also navigate through other operational problems, such as the creation of incentives for employees of the parent companies (such as engineers and sales reps) who interact with but aren’t employed by the joint venture.

1 These guidelines are a subset of McKinsey’s

draft joint-venture governance guidelines: more than 30 principles that companies should apply to their joint ventures. The guidelines have grown out of our client work with more than 500 alliances and, more directly, out of McKinsey’s October 2004 roundtable for CEO s and directors. 2 When Exxon merged with Mobil in 1999, the

new company installed a board composed entirely of internal directors.

13

What’s more, the members of the boards of joint ventures—almost always employees of their parent companies—have to overcome inherent conflicts between the specific interests of each of the parents and the overall interests and health of the venture. As one former board member of a large aerospace joint venture explained, “It was quite a dishonest system. You went into a meeting wearing two hats, as both supplier and owner. Each partner’s prime motivation was to look after its own interests.” Toward a better model

Despite the differences between wholly owned businesses and joint ventures, many

Outsiders are now extremely rare on the boards of joint ventures. Yet an outsider who is explicitly charged with promoting the interests of such a business and asking tough questions about its performance and long-term direction can dramatically improve its transparency, its bottom-lineperformance orientation, and its overall returns. Such an outside director is also in a position to argue on behalf of its strategy when the parents have diverging interests. In the infrequent cases where joint ventures have appointed outside directors, the experience has been quite encouraging. Consider, for example, the case of Aera Energy, a multibillion-dollar upstream oil joint venture that Mobil and Royal Dutch/ Shell formed in 1996. The venture, initially a wholly owned spinout of Shell, was built on the premise that to compete in the mature and highly competitive heavycrude business in California, Aera needed to operate in a manner very different from that of its major-oil-company parents. It therefore had to be quite independent. A key part of establishing this independence was the appointment of outsiders to the board. In 1996, when Shell created a 60-40 joint venture with Mobil’s upstream California assets, the partners agreed to retain the outsiders on the joint venture’s board and have them continue to chair the audit and compensation committees.2 Instead of choosing oil industry veterans, the parents sought out highly respected executives from other industries. Eugene

14

McKinsey on Finance

Spring 2005

Voiland, the CEO of Aera, believes that the outside directors brought a fresh perspective to the business, promoted transparency and frank discussion, and advanced the interests of the joint venture as a whole (rather than optimizing one parent’s interests).

the consortium between board meetings by working with the CEO to review and challenge the content of all board presentations and by serving as a sounding board for the management team on key strategic and operational issues. The chair must bring real credibility to the role: Bilous assumed it after a career at IBM , where, as a general manager of a major division of IBM Microelectronics, he had structured and served on the boards of four very large semiconductor-manufacturing joint ventures. He is not, however, currently employed by Sematech or any of its investors, so he is an independent nonexecutive chairman.

Designate lead directors or a strong chairperson

Partners in a joint venture typically approach the design of its board by picking three or four members each, without specifying their individual roles. Such board members thus act as generalists instead of contributing specific skills. Ideally, however, companies should adopt a highly specialized model of joint-venture governance by appointing board members with individual expertise who can provide real oversight and guidance in important areas such as finance, manufacturing, and regulatory affairs. As a first step, each parent company should appoint a board member to function as its lead director. This makes at least one member a peer of the joint venture’s CEO, with the power to challenge the management team and the responsibility for securing resources from the parents and for managing the relationship with them. One prominent global chemical company insists that in every major joint venture, each partner must appoint a lead director who spends at least 20 days a year on its affairs (the norm for nonlead directors is 5 days). In some cases—especially multiparty joint ventures and consortia—a strong chair with oversight of strategy and budget issues can serve the same purpose. Consider Sematech, an 18-year-old research consortium formed by more than a dozen leading semiconductor companies. Its chairman, O. B. Bilous, devotes time to

Review and reward the performance of board members

Contrary to today’s standard for corporate governance, few joint-venture board members are evaluated on their individual performance, and compensation is only obliquely, if at all, linked to the performance of the venture they oversee. Companies should rethink the way they review and compensate the members of the boards of joint ventures. For starters, in annual reviews board members should be evaluated by such criteria as their impact in shaping the joint venture’s strategy, their success in fulfilling their risk-management responsibilities, their track record in securing resources and attracting good people, and their ability to secure timely decisions from the corporate parents. Moreover, companies should consider linking the directors’ compensation directly to the profit-and-loss statement or to other performance targets. To align the interests of the joint venture with those of the inside directors and to ensure that they have some skin in the game, at least 5 percent of their total compensation

15

Governing joint ventures

should be linked to a common metric for the performance of joint ventures. In fact, these businesses could steal a page from nonequity alliances that reward their board members and managers for exceptional performance by drawing down sums from a monetary fund whose size is linked to financial targets.

a contrast. Each year, its board hired a leading accounting firm to audit its books. The accountants were asked to pay particular attention to the parent companies’ transactions, such as crude-oil supply and off-take agreements, thereby ensuring that materials and services were fairly priced and that the board understood the total economics of the business.

Sponsor an external audit

Large joint ventures do a fairly good job of generating basic financial and operating data—for instance, the cost of goods sold, plant utilization levels, and product defect rates. But they tend to be less effective at understanding their own economics (including transfer price profits) from the perspective of their corporate parents and at generating the second-level management data3 so essential to grasping the real issues and prospects of a business. The resulting lack of transparency about the economics and transfer prices can be startling. One industrial joint venture’s parents extracted essentially all of their value through the sale of product subsystems to it—not through its dividends, which were negligible. At no time did its board understand whether it was truly profitable, because the governance system made the parents’ actual costs in building these subsystems opaque. Indeed, the issue of transfer-pricing profits created deep tensions in the relationship. The parents referred to the annual meeting on prices as “the poker game,” and the parents’ auditors called themselves “the liars’ club” because they had to smoke out each other’s pricing bluffs.

3 For example, information on the profitability

of different customer or product segments, assessments of key business risks, and detailed comparisons with the performance of competitors.

One US downstream oil industry joint venture that also depended extensively on its parents for key inputs, such as crude oil and administrative services, provides

Create a real challenge process

Too often, and for many reasons, joint ventures are shielded from thorough performance scrutiny. For one thing, they exist on the corporate periphery, outside their parents’ normal reporting processes. Likewise, their parents may be reluctant to invest scarce managerial resources in efforts to oversee them properly merely to capture, say, only half of the resulting benefit. Indeed, the returns not only seem lower but also may take more effort to secure: changing an underperforming joint venture in a significant way typically requires agreement by the parents, and that can be difficult—if not impossible—to achieve. Nonetheless, a company making a large investment in a joint venture should oversee it with the same level of intensity that would be devoted to other businesses of the same size. One important consideration is how to create a rigorous reporting and review process that engages all of the corporate parents without subjecting the joint venture to “double jeopardy”—that is, to full and separate reporting to all of its corporate parents, forcing it to meet different data and format requirements and to report for different calendar years. Here, the experience of one consumer-productfinancing joint venture is instructive. To avoid double jeopardy, it was linked directly into one parent company’s corporatereview process and treated like one of

16

McKinsey on Finance

Spring 2005

that company’s wholly owned business units. The key difference is that the board members from both parents participate in these meetings and challenge proposals from the parents’ perspective. In the oil industry, the board of a multibillion-dollar joint venture established an independent review process, including a separate and very strong finance and audit committee as well as the aggressive use of outside auditors to benchmark the performance of the business.

have similar expectations about the joint venture’s performance.

Let the venture’s CEO run the business

We recognize that the board of a joint venture may, from time to time, have to intervene in its operational affairs or strategy decisions—for example, during a major downturn in its performance or a fundamental change in its geographic or product scope. But we also believe that a board must empower a joint venture’s CEO to operate as its true general manager, not only for the sake of fast and objective decisions, but also to attract and motivate strong leaders. All too often, CEOs of joint ventures lack the authority to run them, while board members act as quasi-operators who intervene haphazardly in tactical decisions. One thing companies can do is to grant the CEO the power to hire and fire all employees, or at least a veto. Another is to give the CEO a reasonably high degree of sign-off authority on capital expenditures and to establish beforehand what percentage of the dividends the joint venture will retain to invest. In addition, the board should develop and endorse a performance contract for the CEO — essential not only to define the boundaries between the CEO and the board but also to make sure that the corporate parents

The experience of a four-partner joint venture in the electric utility industry is illustrative. Although the board had endorsed a strategic road map for the first two years, it was exceedingly general and hadn’t been codified consistently or even formally approved by the board. Individual directors, trying to steer the business in one direction or another, made many backchannel requests to the CEO between board meetings, so that the original road map became less and less relevant. The joint venture lost its focus, missed a key customer window, and nearly drove one of its parents to exit. To improve its business discipline, the board recruited an experienced general manager as CEO. Before accepting the job, he insisted that the board endorse a three-page memo identifying six objectives for his first year. Each objective came with three or four pinpoint deliverables and a relative weighting for evaluation. The result: the parents started to walk in lockstep during year three, and the joint venture became more disciplined in business and operational matters.

Few if any joint ventures now follow such guidelines. But those that do have an opportunity to improve their performance materially. MoF Jim Bamford ([email protected]) is a

consultant and David Ernst (David_ Ernst@ McKinsey.com) is a partner in McKinsey’s Washington, DC, office. Copyright © 2005 McKinsey & Company. All rights reserved.

17

Running head

Viewpoint Merger valuation: Time to jettison EPS Assessing an acquisition’s value by estimating its likely impact on earnings per share has always been flawed. Now it’s likely to be flat wrong.

Richard Dobbs,

In any acquisition, it’s difficult to predict

Billy Nand, and

future cash flows and synergies. Managers, boards, and analysts in the United States and Europe have therefore generally tested the relative attractiveness of a transaction by measuring its positive or negative impact on earnings per share (EPS). Simplistic and flawed as this approach may be, executives could argue that it was valid as long as accounting rules supported it.

Werner Rehm

1 The US GAAP changes also removed the

pooling accounting approach, which allowed companies to avoid goodwill entirely. See Neil W. Harper, Robert S. McNish, and Zane D. Williams, “Shed no tears for pooling’s demise,” McKinsey on Finance, Number 1, Summer 2001, pp. 17–20. 2 The European Union switched to the

International Financial Reporting Standards ( IFRS ) in January 2005. Now more than 60 countries require IFRS . An additional 7 countries (Bahrain, China, Kazakhstan, Romania, Russia, Ukraine, and the United Arab Emirates) require some domestic companies to use IFRS , and a further 24 allow all of them to use it. Australia’s AIFRS (Australian Equivalents to IFRS ) resembles IFRS with respect to the amortization of goodwill. 3 Before the change, in 2001, the US GAAP had

permitted up to 40 years of useful life for goodwill.

That should have changed for US executives two years ago, when companies using US generally accepted accounting principles (GAAP) stopped amortizing goodwill.1 Under the new rules, nearly every acquisition shows a positive, or accretive, impact on EPS before the cost of restructuring—making the EPS test completely unreliable as an indicator of value created. Yet news releases and public comments from US executives and Wall Street analysts continue to discuss and assess acquisitions in terms of EPS accretion or dilution. In addition, remuneration committees continue to evaluate management teams on their EPS performance. For European executives, the rules on amortizing goodwill have only recently changed.2 With basically the same standard for the amortization of goodwill now established

in so many countries, it’s time for companies to drop, once and for all, the flawed EPS accretion/dilution test as a measure of an acquisition’s value. At best, the test is inaccurate; at worst, it thoroughly misleads investors. A better proxy, if executives need a new rule of thumb, would be an acquisition’s impact on the acquirer’s economic profit—another imperfect measure but one that is better than EPS because it takes into account an acquisition’s full economic cost. A flawed test

Goodwill is the amount paid for acquired companies above the fair value of their book assets. Under previous International Accounting Standards, it was placed on the acquiring company’s balance sheet and then amortized over the duration of its useful life or for 20 years, whichever was less.3 This amortization in effect penalized the company for the cost of the acquisition and could therefore restrain managers who might try to increase earnings through acquisitions that would destroy shareholder value. Assessing the impact of acquisitions on EPS rather than on total net income corrects for the dilutive effect of acquisitions that involve the issuing of shares. When a company uses cash to complete acquisitions, its earnings are reduced by the loss of interest income on the cash used in the transaction or of the interest on the additional debt. When it issues additional shares to complete a transaction, EPS declines mathematically, since earnings are spread across a greater number of shares. But the EPS accretion/dilution test doesn’t reflect the actual value created by an acquisition. Writing off goodwill over 20 years may penalize companies too much—particularly in industries with

��� ������������������������ 18 McKinsey on Finance �������������� ��������������������������������������������������������������� �������������������������������������������������

Spring 2005

���������

they must write it down and take a charge against earnings. If it hasn’t been impaired, the goodwill is assumed to have an indefinite life span and isn’t amortized.

����������������������������� ���������������������������������������� �������������������������� �������������������� �������������� ��������������������� � ������������

������������� ������������

���������

��

�������

��

��������

�� ������� ������������

�� �� �� ������� ������������

���������������������������������������������������������������������������������������������������������������������������

������������������������������������������������������������������������������������������������������

��������������������������������������������������������

�����������������������������������������������������

intangible assets like brands, which can endure longer than 20 years. Or it may penalize companies too little—particularly in cash deals when the EPS calculation doesn’t charge for the acquisition’s full cost of capital. As a result, there is no correlation between the financial markets’ impression of the value created by a deal and its accretive or dilutive nature. In an analysis of 117 transactions larger than $3 billion by US companies from 1999 to 2000, we found no correlation between the capital markets’ reaction to a deal and the deal’s impact on the acquirer’s EPS (Exhibit 1).

4 Most of the ten EU accession countries either

permit or require filing in IFRS . Companies that already report in the US , Canadian, or Japanese GAAP can continue to do so until 2007. 5 IFRS rule 3/IAS 36/38 is required for business

combinations with an agreement date on or after March 31, 2004. Under US GAAP, SFAS 142 is required for fiscal years after December 15, 2001. 6 The sample, including transactions valued

at more than $3 billion, excluded the telecommunications, high-tech, and software companies that drove the market bubble of that period.

The EPS test under new rules

Starting January 1, 2005, countries that had been EU members in 2002 are requiring public companies to adopt a consistent accounting methodology: IFRS.4 As the US GAAP did earlier, IFRS has changed the rules for the treatment of goodwill.5 Under the new rules for both of these accounting standards, acquirers record it as an asset but don’t amortize its value over time. Instead, companies must test the value of goodwill at least annually. If the value has been impaired,

The new rules have no effect on the actual cash flow of the combined companies or on the value created by a deal. They do, however, have a significant impact on EPS accretion/ dilution in acquisitions that create goodwill: because it isn’t amortized, most acquisitions will now result in EPS accretion. This further undermines the EPS accretion/dilution test as an indicator of the value in a deal. It also creates a new danger for boards that use EPS to measure the performance of executives: these boards could be encouraging them to undertake value-destroying deals that increase EPS as measured under the new accounting standards. We illustrate this point with a sample set of goodwill-creating transactions by US companies in 1999 and 2000.6 On average, under the old accounting standards requiring the amortization of goodwill, the earnings dilution for these deals was 24 percent in the first fiscal year after the acquisition and 12 percent in the second. Only 2 of the 11 deals were accretive in the second year (Exhibit 2, part 1). Under the new rules, the picture would be very different: 9 of the 11 deals would have been accretive during both the first and the second years (Exhibit 2, part 2). Across all of these deals, the average accretion would have been 13 percent in year two, as compared with a 12 percent dilution in the same year under the old accounting rules. The new standard has, in effect, completely undermined the traditional rule of thumb. Why? In essence, under the new rules the impact on the acquirer’s EPS doesn’t

��� ������������������������ Merger valuation: Time to jettison EPS �������������� ����������������������������������������������������������������

���������

��������������������������������������������������������� �������������������

�� �������������������� ��������������������������� ������������������ ��������������������������

���

��������������������������

���

������������������������� ����������� � �

�� �������������������� ������������������������ �������������������������� ��������������������������

�� ��

������������������������� ����������� � �

����������������������������������������������������������������������������������������

����������������������������������������������������������������������������������������������������������� �����������������

7 The pretax cost of debt X (1 – tax rate) with a

tax rate of 35 percent. 8 This example assumes that the capital

markets don’t penalize the acquirer and that the exchange ratio can be set in relation to the preannouncement share price plus the 25 percent acquisition premium.

reflect the full economic cost of the acquisition. Consider a hypothetical example (Exhibit 3). A company evaluating the acquisition of a business worth $400 million in the capital markets is willing to pay a 25 percent premium, or $500 million in cash. For simplicity’s sake, let’s assume that the deal will create no synergies and that the acquisition target has a net income of $30 million. The acquiring company decides to finance the deal by raising debt at a pretax interest rate of 6 percent. Obviously, this deal destroys value because the company is paying $500 million for an entity that is worth only $400 million (remember, no synergies). Even so, next year’s earnings— and therefore EPS —increase to 2.3, from 2.0, because earnings from the acquired company exceed the after-tax interest payments that are required for the new debt.

19

This proposed acquisition would have returns higher than the cost of the debt used to finance it but lower than the overall weighted average cost of capital, which includes the required return for equity holders. Raising debt to finance the deal places a sizable burden on the company’s shareholders without compensating them for the additional risk. Only when the return on invested capital is greater than the company’s overall cost of capital is each investor appropriately compensated. In our example, earnings of $30 million on an investment of $500 million would bring a mere 6 percent return on invested capital. While this exceeds the after-tax cost of financing the debt at 3.9 percent,7 it is below the overall cost of capital. If instead the acquiring company exchanged shares to pay for the business it acquired, it would need to issue 12.5 million new shares to provide the 25 percent acquisition premium to the target company’s shareholders. After the deal, the company would have 52.5 million shares outstanding and earnings of $110 million.8 The new company’s EPS would rise to 2.1, so the deal would again be accretive. Regardless of the actual value created, a deal will always be earnings accretive if the acquirer’s P/E ratio is greater than the target’s P/E ratio, including the acquisition premium. No substitute for fundamentals

Therefore, when boards, executives, and investors look at acquisitions, it is no longer possible for them to rely on the EPS accretion/dilution test as a proxy for value creation—if indeed they ever could. Remuneration committees in particular must be wary of using EPS targets to evaluate management. In many countries, the executives would have an incentive to pursue value-destroying deals to make the

��� ������������������ 20 McKinsey on Finance �������������� ������������������������������������������������

Spring 2005

���������

������������������������������������� �������������������������������������������������������������������������������� �����������

������������� ��������

������

���������

�����������

���������������������������

��

��

��������������������������������

����

����

���������������������

��

��

������������������������������

����

����

����������������������������������

�����

�

����������������������������� ���������

����

�����

������������������������

����

����

�������������������

���

����

�������������������������

����

����

�������

���

���

�����������������

���

���

���������������������������

�

�

������������������������������

��

��

�������������������

����

����

������������������������������ ����� ���������

���

����������������������������

���

���������

for goodwill at the full cost of capital.9 This analysis is fairly well established. In fact, many companies already use economic profit for internal decision making. A dialogue about how long a deal will dilute economic profit forces managers to ask how long it will take the company to begin earning its cost of capital on the acquisition. As with any measure, this one can’t be applied unthinkingly: some deals create substantial value through the long-term growth of the business while diluting economic profit in the short term. Understanding the impact of an acquisition on the dilution or accretion of economic profit could, however, provide a good basis for communication and discussion.

������������������������������������������

EPS targets. Instead, boards, shareholders,

and executives must fully understand every transaction’s fundamentals, including a detailed perspective on how synergies will create value and an evaluation using a discounted-cash-flow analysis. 9 Economic profit = invested capital X (return

on invested capital – weighted average cost of capital). See Tim Koller, Marc Goedhart, and David Wessels, Valuation: Measuring and Managing the Value of Companies, fourth edition, Hoboken, New Jersey: John Wiley & Sons, 2005, available at www .mckinsey.com/valuation.

While there is no perfect metric to prove whether a deal creates value, one measure companies could explore in their dialogue with investors is a deal’s impact on economic profit, calculated with a charge

Assessing the value of an acquisition by estimating its impact on EPS has always been questionable. Under recent changes in accounting standards, that approach has become misleading and risky. Companies, boards, and investors must take note. MoF Richard Dobbs (Richard_Dobbs@McKinsey .com) is a partner in McKinsey’s London office; Billy Nand ([email protected]) and Werner Rehm ([email protected]) are consultants in the New York office. Copyright © 2005 McKinsey & Company. All rights reserved.

Amsterdam Antwerp Athens Atlanta Auckland Bangkok Barcelona Beijing Berlin Bogota Boston Brussels Budapest Buenos Aires Caracas Charlotte Chicago Cleveland Cologne Copenhagen Dallas Delhi Detroit Dubai Dublin Düsseldorf Frankfurt Geneva Gothenburg Hamburg Helsinki Hong Kong Houston Istanbul Jakarta Johannesburg Kuala Lumpur Lisbon London Los Angeles Madrid Manila Melbourne Mexico City Miami Milan Minneapolis Montréal Moscow Mumbai Munich New Jersey New York Orange County Oslo Pacific Northwest Paris Pittsburgh Prague Qatar Rio de Janeiro Rome San Francisco Santiago São Paulo Seoul Shanghai Silicon Valley Singapore Stamford Stockholm Stuttgart Sydney Taipei Tel Aviv Tokyo Toronto Verona Vienna Warsaw Washington, DC Zagreb Zurich

Copyright © 2005 McKinsey & Company