The Role of Real Options in Capital Budgeting: Theory and Practice1

Most discussions of capital budgeting take for granted that discounted cash flow (DCF) and real options valuation (ROV) are very different methods that are meant to be applied in different circumstances. Such discussions also typically assume that DCF is “easy” and ROV is “hard”—or at least dauntingly unfamiliar—and that, mainly for this reason, managers often use DCF and rarely ROV. This paper argues that all three assumptions are wrong or at least seriously misleading. DCF and ROV both assign a present value to risky future cash flows. DCF entails discounting expected future cash flows at the expected return on an asset of comparable risk. ROV uses “risk‐neutral” valuation, which means computing expected cash flows based on “risk‐neutral” probabilities and discounting these flows at the risk‐free rate. Using a series of single‐period examples, the author demonstrates that both methods, when done correctly, should provide the same answer. Moreover, in most ROV applications—those where there is no forward price or “replicating portfolio” of traded assets—a “preliminary” DCF valuation is required to perform the risk‐neutral valuation. So why use ROV at all? In cases where project risk and the discount rates are expected to change over time, the risk‐neutral ROV approach will be easier to implement than DCF (since adjusting cash flow probabilities is more straightforward than adjusting discount rates). The author uses multi‐period examples to illustrate further both the simplicity of ROV and the strong assumptions required for a typical DCF valuation. But the simplicity that results from discounting with risk‐free rates is not the only benefit of using ROV instead of—or together with—traditional DCF. The use of formal ROV techniques may also encourage managers to think more broadly about the flexibility that is (or can be) built into future business decisions, and thus to choose from a different set of possible investments. To the extent that managers who use ROV have effectively adopted a different business model, there is a real and important difference between the two valuation techniques. Consistent with this possibility, much of the evidence from both surveys and academic studies of managerial behavior and market pricing suggests that managers and investors implicitly take account of real options when making investment decisions.     

Valuation in Emerging Markets: A Simulation Approach

Most of the foundations of valuation theory have been designed for use in developed markets. Because of the greater, and in some cases different, risks associated with emerging markets (although recent experience might suggest otherwise), investors and corporate managers are often uncomfortable using traditional methods. The typical way of capturing emerging-market risks is to increase the discount rate in the standard valuation model. But, as the authors argue, such adjustments have the effect of undermining some of the basic assumptions of the CAPM-based discounted cash flow model. The standard theory of capital budgeting suggests that estimates of unconditional expected cash flows should be discounted at CAPM discount rates (or betas) that reflect only “systematic,” or “nondiversifiable,” market-wide risks. In practice, however, analysts tend to take what are really estimates of “conditional” expected cash flows—that is, conditional on the firm or its country avoiding a crisis—and discount them at higher rates that reflect not only systematic risks, but diversifiable risks that typically involve a higher probability of crisis-driven costs of default. But there is almost no basis in theory for the size of the increases in discount rates. In this article, the authors propose that analysts in emerging markets avoid this discount rate problem by using simulation techniques to capture emerging-market risks in their estimates of unconditional expected cash flows—in other words, estimates that directly incorporate the possibility of an emerging-market crisis and its consequences. Having produced such estimates, analysts can then discount them using the standard Global CAPM.     

Valuing Companies with Cash‐Flow@Risk

The classic DCF approach to capital budgeting—the one that MBA students in the world's top business schools have been taught for the last 30 years—begins with the assumption that the corporate investment decision is “independent of” the financing decision. That is, the value of a given investment opportunity should not be affected by how a company is financed, whether mainly with debt or with equity. A corollary of this capital structure “irrelevance” proposition says that a company's investment decision should also not be influenced by its risk management policy—by whether a company hedges its various price exposures or chooses to leave them unhedged. In this article, the authors—one of whom is the CFO of the French high‐tech firm Gemalto—propose a practical alternative to DCF that is based on a concept they call “cash‐flow@risk.” Implementation of the concept involves dividing expected future cash flow into two components: a low‐risk part, or “certainty equivalent,” and a high‐risk part. The two cash flow streams are discounted at different rates (corresponding to debt and equity) when estimating their value. The concept of cash‐flow@risk derives directly from, and is fully consistent with, the concept of economic capital that was developed by Robert Merton and Andre Perold in the early 1990s and that has become the basis of Value at Risk (or VaR) capital allocation systems now used at most financial institutions. But because the approach in this article focuses on the volatility of operating cash flows instead of asset values, the authors argue that an internal capital allocation system based on cash‐flow@risk is likely to be much more suitable than VaR for industrial companies.     

Accounting for Risk and Return in Equity Valuation

Standard valuation models forecast cash flows or earnings, add a growth rate, and discount the cash flows to their present value with a discount rate that typically reflects the cost of capital. But as the author argues, projecting the long‐term growth rate is essentially speculative; and along with uncertainty about the growth rate, analysts generally do not have a good grasp of the discount rate either. Thus, instead of reducing uncertainty, these two features effectively compound uncertainty in valuations in the sense that slight changes in the growth rate or discount rate can change the valuation considerably. In this article, the author proposes an alternative approach that views the investor's problem as one of challenging the speculations that are built into the current market price, particularly the speculation about growth. Rather than building in a speculative growth rate (and thereby treating it as if it were a certainty), the author's approach turns the problem on its head by using an accounting analysis of the firm's current earnings and cash flows that provides a basis for recognizing the speculative component of the current stock price. More specifically, the author's analysis identifies the future earnings growth path that is implied by the market price, which can then be evaluated with the question: Do I want to pay for this growth? Because growth expectations are risky, additional analysis can be used to provide an understanding of the risk and return to buying growth, and of the upside and downside if risk growth expectations are not realized. By taking such an approach, investors incorporate their understanding of risk not by increasing the discount rate, but by recognizing that the primary risk in investing is the risk of overpaying for growth.     

Understanding Stock Price Volatility: The Role of Earnings

In an efficient capital market, asset prices vary when investors change their expectations about cash flows, discount rates, or both. Using dividends to measure cash flows, previous research shows that the aggregate dividend‐price ratio varies due to changes in expected discount rates (returns) rather than expected cash flows. In contrast, using accounting earnings instead of dividends as a measure of cash flows, this paper shows that as much as 70% of the variation in the dividend‐price ratio can be explained by changes in expected earnings. Moreover, the paper documents a significant negative correlation between expected returns and expected earnings, suggesting that variations in a common factor to both may generate significant price volatility. The results are consistent with the dividend‐policy irrelevance hypothesis.     

Estimating cost of capital in firm valuations with arithmetic or geometric mean – or better use the Cooper estimator?

We test the extent and determinants of bias effects of the arithmetic as well as the geometric mean estimator and the estimator of Cooper [1996. Arithmetic versus geometric mean estimators: Setting discount rates for capital budgeting. European Financial Management 2 (July): 157–67] regarding discount rate estimation for firm valuation by way of a bootstrap approach for 13 different countries. The Cooper estimator is superior to both the geometric and the (conventional) arithmetic mean estimator. However, a ‘truncated’ version of the arithmetic mean estimator leads generally to better estimation outcomes than the Cooper estimator. This means that, in order to reduce problems of upward-biased firm value estimates, expected cash flows beyond a certain time horizon are completely neglected in terminal value estimation. Such an approach seems particularly reasonable for the valuation of young growth companies as well as for companies from quickly developing countries such as Brazil, China, or Thailand, because the bias in terminal value estimation is increasing in the growth rate of future expected cash flows.     

Implementing Fischer Black's Simple Discounting Rule

Corporate managers typically estimate the value of capital projects by discounting the project's expected future net cash flows at the cost of capital. The capital asset pricing model (CAPM) is generally used to estimate that cost. But, as anyone who has worked on the finance or business development staff of a public company can attest, there are major challenges in applying the CAPM, including largely unresolved questions about what constitutes the “market portfolio,” how to estimate market risk premiums, and how to estimate the betas of projects. In a short article published in Financial Management in 1988, Fischer Black proposed a valuation “discounting rule” that avoids all these problems—one that involves discounting a relatively certain (as opposed to an expected or average) level of operating cash flows at the risk-free rate. But Black's article does not address the question of how to calculate these “certainty equivalent” or “conditional” cash flows. In this article, the authors propose a way of implementing Black's rule that involves estimating the “conditional” cash flows in a three-step procedure:

• • Find a benchmark security that correlates with the project's cash flows;
• • Estimate the percentiles of the distribution in which the benchmark return equals the risk-free rate over different investment horizons;
• • Use information from corporate managers to assess the cash flows that define the same percentiles in the cash flow distributions.

As the authors point out, the virtue of Black's rule is that it shifts the focus of the analyst away from the assessment of discount factors and puts it squarely on the more challenging, and arguably more relevant, problem of estimating the project's cash flows.     

Discounting Mean Reverting Cash Flows with the Capital Asset Pricing Model

Discounting cash flows requires an equilibrium model to determine the cost of capital. The CAPM of Sharpe and the intertemporal asset pricing model of Merton (1973) offer a theoretical justification for discounting at a constant risk adjusted rate. Two problems arise with this application. First, for mean reverting cash flows the risk adjustment is unknown, and second, if the present value is compounded forward then the distribution of future wealth is likely right skewed. I develop equilibrium discount rates for cash flows whose level or growth rate is mean reverting. Serial correlation also largely eliminates the skewness problem.     

Accounting for Sovereign Risk When Investing in Emerging Markets

In theory, political risk is project‐specific and should be accounted for in the estimation of the expected investment cash flows. But in practice, the political risk associated with this type of investment is typically accounted for implicitly by adjusting the investment's required rate of return or the discount rate. As the authors discuss in the article, this approach disguises the specific assumptions being made about the risk of expropriation and so makes it difficult to assess this risk properly. While defending some aspects of current practice, the authors argue that corporate executives should consider some changes. For example, although a project analysis that is shared with the host government could incorporate a risk adjustment to the discount rate, the authors suggest that more explicit analysis of the anticipated risk of expropriation should be incorporated into the analysis of expected project cash flows. This analysis could involve making specific assumptions about the “term structure” of expropriation risk over the life of the investment. Finally, the authors note that the political risk of making investments in emerging economies can be managed to some extent. Investments can be structured in ways that reduce political risk by structuring project cash flows in ways that better align the incentives of the project sponsor and the government of the host country.     

How Fundamental are Fundamental Values? Valuation Methods and their Impact on the Performance of German Venture Capitalists

This paper studies how the use of alternative valuation methodologies affects investment performance for a sample of 53 German venture capitalists. We measure investment performance by the amount of investments they need to write off and by the number of companies they take public. We find that a significant number of investment managers use discounted cash flow (DCF) techniques, but only a minority appears to use a discount rate related to the cost of capital. The majority applies DCF using subjective discount rates. We present evidence that the use of DCF is correlated with superior investment performance only if applied in conjunction with an objectifiable discount rate. Also, funds that invest with a longer horizon perform better. The use of multiples is not significantly correlated with investment performance. We conclude that a focus on fundamental values confers an advantage.     

Real Asset Valuation: A Back‐to‐basics Approach

Different valuation methods can lead to different corporate investment decisions, and the conventional “static, single discount rate” DCF approach in particular is biased against many of the kinds of decisions that corporate managers tend to view as “strategic.” Reducing the bias from valuations involves two main tasks: treating risk in a way that is consistent with observed market pricing, and accounting for the ability of companies to make decisions “dynamically” over time. The authors propose two separate tools, market‐based valuation and complete decision tree analysis, for accomplishing these two improvements in valuation. The authors also suggest working with the full distribution of future cash flows, one possible realization at a time, rather than working with the aggregate measure of expected cash flow. From a technical perspective, it is necessary to work with the full distribution to value real options properly. Valuing the cash flows one realization at a time also leads to a much better understanding of the interaction between economy‐level, systematic risks and local asset‐level, technical risks. Just as important, the proposed approaches support an effective division of labor between local asset managers, who are better positioned to model technical considerations and other asset specifics, and the central finance staff, who can ensure the consistent treatment of economy‐wide risk and to create the rules of engagement for evaluating opportunities. After presenting an overview of both the valuation and the organizational issues, the authors present a case involving a corporate investment in carbon capture and storage that illustrates both the application of the proposed methods and the various sources of bias in the typical DCF analysis.     

Is Country Diversification better than Industry Diversification?

In this paper, we develop a framework in which one can examine the source of industry and country diversification by examining their underlying return components. We find that the global cash flow factor explains on average 39% of the variation of country cash flows and global discount rates explain 55% of the variation of country discount rates. These are much less than the explanatory power of the two factors over industry cash flow and discount rate variations, which are 72% and 78% respectively. This suggests that global factors are much less important for return components at country level than at the industry level. As a result, both better diversification of expected returns and cash flows across countries determine the larger benefits of country diversification versus industry diversification. Moreover, emerging markets tend to have much smaller co‐movements of both dividends and expected returns with those of the world, suggesting a lower degree of integration with the world goods and financial markets. Our results cast doubt on the prevailing wisdom that country diversification should be replaced by industry diversification.     

Risk-adjusted discount rates and capital budgeting under uncertainty

This paper is concerned with the valuation of multiperiod cash flows in a world where prices are determined according to the Sharpe-Lintner-Black model of capital market equilibrium. We find that the current market value of any future net cash flow is the current expected value of the flow discounted at risk-adjusted discount rates for each of the periods until the flow is realized. The discount rates are known and non-stochastic, but the rates for the different periods preceding the realization of a cash flow need not to be the same, and the rates relevant for a given period can differ across cash flows. The risk adjustments in the discount rates arise because of uncertainties about reassessments through time of the expected value of a flow and the relationships between these reassessments and the corresponding reassessments of the expected cash flows of all firms.     

Estimating the Cost of Capital Using Stock Prices and Near‐term Earnings Forecasts

Earnings‐based valuation models, although long used by finance practitioners, have become increasingly popular among finance academics as well. Among the most important reasons for academics' increased acceptance of earnings‐based valuation is the well‐documented claim that earnings over a short (three‐ to four‐year) forecast horizon tend to capture a large fraction—as much as 80%—of today's value, much more than is captured by near‐term forecasts of free cash flow, the measure long advocated by finance theorists as the basis for DCF valuation. But most important for the purposes of this article, the recognition that such a large percentage of the current values of many public companies is captured within a short forecast horizon has led to a large academic literature that uses earnings‐based valuation models together with current stock prices to “back out” estimates of the companies' implied expected rates of return and costs of equity capital. The effectiveness and precision of such reverse engineering depend on the reliability of the forecasts both within a finite forecast horizon and beyond. And although the models tested in academic work, which are based on large samples of forecasts and hard‐to‐verify assumptions about earnings beyond the forecast horizon, often do not appear to provide useful estimates, the author argues that such reverse engineering of the valuation models should become straightforward and workable once reliable forecasts of earnings are obtained—say, from the corporate (or investment) analysts who are familiar with the operations of the companies they work for (or cover).     

Valuation of Corporate Innovation and the Pricing of Risk in the Biopharmaceutical Industry: The Case of Gilead

Much of a firm's market value derives from expected future growth value rather than from the value of current operations or assets in place. Pharmaceutical companies are good examples of firms where much market value comes from expectations about drugs still in the development “pipeline.” Using a new osteoporosis drug being developed by Gilead Sciences, Inc., the author combines discounted cash flow methods values and real option models to value it. Alone, discounted cash flow (DCF) calculations are vulnerable to the assumptions of growth, cost of capital, and cash flows. But by integrating the real options approach with the DCF technique, one can value a new product in the highly regulated, risky and research‐intensive Biopharmaceutical industry. This article shows how to value a Biopharmaceutical product, tracked from discovery to market launch in a step‐by‐step manner. Improving over early real option models, this framework explicitly captures competition, speed of innovation, risk, financing need, the size of the market potential in valuing corporate innovation using a firm‐specific measure of risk and the industry‐wide value of growth operating cash flows. This framework shows how the risk of corporate innovation, which is not fully captured by the standard valuation models, is priced into the value of a firm's growth opportunity. The DCF approach permits top‐down estimation of the size of the industry‐wide growth opportunity that competing firms must race to capture, while the contingency‐claims technique allows bottom‐up incorporation of the firm's successful R&D investment and the timing of introduction of the new product to market. It also specifically prices the risk of innovation by modeling its two components: the consumer validation of technology and the expert validation of technology. Overall, it estimates the value contribution per share of a new product for the firm.     

Two Different Ways of Treating Corporate Cash in FCF Valuations—and the Importance of Getting the Cost of Capital Right

The methods for calculating free cash flow presented in texts on financial statement analysis and valuation appear to be very different from those in corporate finance texts, causing some confusion among academics as well as practitioners. Financial statement analysis and valuation texts generally begin by valuing just the enterprise operations—that is, the entity that engages in the firm's primary revenue‐generating activities—and then adding back the value of its cash holdings and other financial assets. The corporate finance approach is typically to value all the assets together, including financial assets that are not used in the production of the goods and services provided by the firm. Using a simple example, the authors show that the valuation of the equity ownership of the firm should be the same for both methods of calculating free cash flow, provided the analyst makes the appropriate adjustments to the method for calculating the cost of capital (WACC) used to discount forecasted free cash flows to a present value.     

AN EMPIRICAL TEST OF AN EX-ANTE MODEL OF THE DETERMINATION OF STOCK RETURN VOLATILITY

The assumption that changing expected cash flows and discount factors affect a security's return is at the foundation of many financial models. This study examines empirically the hypothesis that expected stock return variability is a function of cash flow and discount rate uncertainty. Maximum likelihood estimation techniques and expectational data are employed. Strong, positive relationships are found, verifying the foundations of the ex-ante models with ex-ante data and providing a better understanding of security markets by explaining, in part, the causes of expected stock price variability.     

Managerial discretion to delay the recognition of goodwill impairment: The role of enforcement

Under International Financial Reporting Standards, managers can use two approaches to increase the estimated fair value of goodwill in order to justify not recognizing impairment: (1) make overly optimistic valuation assumptions, and (2) increase future cash flow forecasts by inflating current cash flows. Because enforcement constrains the use of optimistic valuation assumptions, we hypothesize that enforcement influences the relative use of these two choices. We test this hypothesis by comparing a sample of 1,958 firms from 36 countries that are likely to delay recognizing goodwill impairment (suspect firms) to a sample of control firms. First, we find that firms in high‐enforcement countries use a higher discount rate to test goodwill for impairment than firms in low‐enforcement countries. We also find a more positive association between discount rate and upward cash flow management for suspect firms than for control firms. This result is consistent with suspect firms substituting optimistic valuation assumptions with inflated current cash flows. Second, we find that, relative to control firms, suspect firms exhibit higher upward cash flow management in high‐enforcement countries than in low‐enforcement countries. Third, we show that suspect firms in high‐enforcement countries are more likely to eventually impair goodwill.     

To beat or not to beat? The importance of analysts’ cash flow forecasts

We investigate the implications of firms’ benchmark-beating patterns with respect to analysts’ quarterly cash flow forecasts for firms’ current capital market valuation and their future performance. We hypothesize that nonnegative earnings surprises are more likely to be supported by real operating performance and signal higher earnings quality if they are achieved via higher than expected cash flows or lower than expected accruals. We show that firms beating analyst earnings forecasts have larger positive capital market reactions and larger earnings response coefficients if they beat analyst cash flow forecasts or report lower than expected accruals. We also demonstrate that these firms’ superior future performance may provide an economic justification for their more favorable market response. Our findings suggest that firms’ ability to beat analyst cash flow forecasts is informative regarding the quality of their earnings surprises.     

INCORPORATING COUNTRY RISK IN THE VALUATION OF OFFSHORE PROJECTS

Offshore projects, especially those in emerging economies, are generally viewed as more risky, and thus as contributing less to shareholder value, than otherwise comparable domestic investments. Emerging economies are typically more volatile than the economies of industrialized countries. They also present a greater array of risks that are (perceived as being) primarily of a downside nature, such as currency inconvertibility, expropriation, civil unrest, and general institutional instability. Further, because such risks are relatively unfamiliar to the investing companies, the companies are likely to make costly errors in early years and to require more time to bring cash flows and rates of return to acceptable steady-state levels. To reflect these higher risks and greater unfamiliarity, many companies include an extra premium in the discount rate they apply to offshore and, particularly, emerging-market projects. However, the basis for these discount rate adjustments is often arbitrary. Such adjustments do not properly reflect objective information available about either the nature of these risks, or about the ability of management to manage them. Nor do they take into account the reality that the risks stemming from unfamiliarity fall over time as the firm progresses along the learning curve. As a result, companies often “over discount” project cash flows in compensating for these risks, and thus unduly penalize offshore projects. More important, adjusting for country risk using arbitrary adjustments to the discount rate fails to focus management's attention on strategic and financial actions can be taken to reduce risk—notably, actions capable of transferring some of the company's exposures to specific risks to different parties with comparative advantages in bearing those risks. This paper outlines a four-step procedure for assessing overseas risks that integrates these various aspects:

• ? Classify risks in terms of various stakeholders' comparative advantage in risk-taking based on their: (a) existing portfolio of assets; (b) access to information; and (c) capabilities for reducing risk.
• ? Allocate risk through project structuring and financial engineering to exploitthese comparative advantages.
• ? Adjust resulting cash flows (relative to their most-likely levels) for (a) the impact of “asymmetric” risks; (b) learning effects; and (c) potential competitive options and/or barriers to entry resulting from comparative advantage in dealing with risks.
• ? Discount resulting expected cash flows at a risk-adjusted discount rate that reflects the covariance of the cash flows with the benchmark portfolio.