The Engineering Economist, 50: 2553 Copyright 2005 Institute of Industrial Engineers ISSN: 0013-791X print / 1547-2701 online

e DOI: 10.1080/00137910590917026

MAKING USE OF REAL OPTIONS SIMPLE: AN OVERVIEW AND APPLICATIONS IN FLEXIBLE/MODULAR DECISION MAKING

Lenos Trigeorgis
Bank of Cyprus Chair Professor of Finance, University of Cyprus, Nicosia, Cyprus and President, Real Options Group This article focuses on how the use of real options can be made simple, providing an overview of the power of exible and modular decision making and its use in various applications across industries. After common real options are discussed through a comprehensive example, the article reviews the key lessons and implications of real options thinking for exible decision making. It then proceeds to propose a modular problem structuring approach that allows simplifying of complex real option problems by decomposing them into a few basic building-block option types (reviewed) connected by some basic decision operators. The resulting problemstructuring option map is depicted in a range of illustrative applications in various industries. Past areas of application of real options as well as research challenges ahead are also discussed.

INTRODUCTION In an increasingly uncertain and dynamic global market place, managerial exibility has become essential for rms to successfully take advantage of favorable future investment opportunities, respond effectively to technological changes or competitive moves, or otherwise limit losses from adverse market developments. Thinking of future investment opportunities as real options has provided powerful new insights that in many ways revolutionized modern corporate resource allocation. Real options emphasizes the importance of waiting (e.g., McDonald and Siegel [34]) or staging exibility, suggesting that managers should either wait and see until substantial uncertainty is resolved and the project is more clearly successful,
Address correspondence to Lenos Trigeorgis, Bank of Cyprus Chair Professor of Finance, University of Cyprus, 75 Kallipoleos, P.O. Box 20537, CY 1678, Nicosia, Cyprus. E-mail: lenos@ucy.ac.cy

26

L. Trigeorgis

requiring a premium over the zero-NPV, or they should stage the decision so that they can revise the situation at critical milestones to either proceed to the next stage or abandon. During the waiting or staging period, new information can be revealed that might affect the desirability of the project; if future developments turn out worse than expected, the rm has implicit insurance protecting it against downside losses by choosing not to proceed with the project. Real options also introduces a new insight with respect to the role and impact of uncertainty on investment opportunity value that runs counter to conventional thinking. Since management is asymmetrically positioned to capitalize fully on upside opportunities while it can limit losses on the downside, more uncertainty can be benecial for option value. More can be gained from opportunities in highly uncertain or volatile markets because of the exceptional upside potential and limited downside risk that result from managements exibility to proceed or not proceed with the project. From a strategic perspective, of course, it may not always be benecial to wait and see. For example, by making an early strategic R&D investment, a rm may not only develop more cost-efcient or higherquality products or processes that can result in a sustainable cost or other competitive advantage, but may be able to positively inuence competitive behavior and earn a higher market share down the road. In some cases a rm anticipating competitive entry may make a strategic investment commitment (e.g., in excess production capacity) early on such that it can preempt competition altogether. Therefore, optimal investment timing generally involves a trade-off between wait-and-see exibility and the strategic value of early commitment. Moreover, early investing may itself open up a set of new options embedded in the commercial project (e.g., to later expand, abandon, or switch to alternative uses), whose value may also be enhanced by higher uncertainty, but is realized through early investing. Thus, the presumed depressive impact of uncertainty on investment is not that clear-cut. The above new considerations of investment under uncertainty suggest the need to adopt an expanded or strategic NPV criterion, able to capture managements exibility to alter planned investment decisions as future market conditions change as well as the strategic value of competitive interactionsbesides the value of expected cash ows from committed assets (e.g., see Trigeorgis [50, 52, 55, 56]). The rest of the article is organized as follows: The next section provides an overview of some common real options through a comprehensive example. Key lessons we have learned about real options in terms of main insights and implications are then discussed, followed by discussion on simplifying real-life problems and reducing them to a basic problem structure. A review of real options applications is given next, while the nal section catalogues challenges that future research must focus on more.

Making Use of Real Options Simple

27

OVERVIEW OF COMMON REAL OPTIONS: A COMPREHENSIVE EXAMPLE The following example involving a natural resource extraction and processing facility serves to review many of the most common options encountered in long-term capital investment opportunities. A large natural resources company has a one-year lease to start extracting minerals on undeveloped land with potential reserves. Initiating the project may require certain exploration costs, to be followed by construction of roads and other infrastructure outlays. Planned investment outlays are indicated as It , while Vt indicates the value of the projects expected operating cash ows at time t . The initial investment in exploration is I0 , and the investment in roads and other infrastructure in the rst period is I1 . This is expected to be followed by capital outlays, I2 , for the construction of a new processing facility. Extraction can begin only after construction is completed; i.e., cash ows are generated only during the operating stage that follows the last outlay. During construction, if market conditions deteriorate, management can choose to forego future planned outlays past the current stage. Management may also choose to reduce the scale of operation by c%, saving a portion, IC , of the last outlay ( I2 ) if the market is weak. The processing plant can also be designed up front such that, if mineral prices (or the quantity of reserves) turn out unexpectedly high, the rate of production can be enhanced by x % with a follow-up outlay of I E to install extra capacity. At any time, management may salvage a portion of its investment by selling the processing plant and equipment for their salvage value or switch them to an alternative use value, At . An associated renery plant, which may be designed to operate with alternative sources of energy inputs, can convert the raw mineral into a variety of rened byproducts. We enumerate hereafter the real options embedded in this type of project. 1. The option to defer investment. The lease enables management to defer investment for up to one year and benet from the resolution of uncertainty about mineral prices during this period. Management would invest I1 (i.e., exercise its option to extract the mineral) only if mineral prices (or reserves) are sufciently high, but would not commit to the project, saving the planned outlays, if prices (or reserves) are low. Just before expiration of the lease, the value added will be the greater of the net value created (that is, V1 I1 ) or $0, with value added represented as max (V1 I1 , $0). The option to defer or invest is thus analogous to an American call option on the gross present value of the completed projects expected operating cash ows, V1 , with

28

L. Trigeorgis

the exercise price being equal to the required outlay, I1 . Since early investment implies sacricing the option to wait, this option value loss is like an additional investment opportunity cost, justifying investment only if the value of cash benets actually exceeds the initial outlay by a substantial premium. The option to wait is particularly valuable in such resource extraction industries, as well as in farming, paper products, and real estate development, due to high uncertainties and long investment horizons. 2. The option to stage investment. In most real-life projects, the required investment is not incurred as a single up-front outlay. The actual staging of capital investment as a series of outlays over time creates valuable options to continue with the project or abandon it at any given stage (e.g., after exploration if the reserves or mineral prices turn out very low). Thus, each stage (e.g., building necessary infrastructure) can be viewed as an option on the value of subsequent stages by incurring the next cost outlay (e.g., I1 ) required to proceed to the next stage and can therefore be valued similar to compound options. This option is valuable in all R&D-intensive industries, especially biotechnology and pharmaceuticals; in highly uncertain, long-development capital-intensive industries, such as energy-generating plants or large-scale construction; acquisition or market entry strategies; high-tech start ups; and venture capital. 3. The option to expand. If mineral prices, reserves, or other market conditions turn out more favorable than expected, management can accelerate the rate or expand the scale of production (by e%) by incurring a follow-up cost outlay (IE ). This is similar to a call option to acquire an additional part (e%) of the base-scale project, paying IE as exercise price. The investment opportunity with the option to expand can be viewed as the base-scale project plus a call option on future investment [i.e., V + max(eV IE , 0)]. Given an initial design choice, management may deliberately select a more expensive technology with built-in exibility to expand production if and when it becomes desirable. As discussed further below, the option to expand may also be of strategic importance, particularly if it enables the rm to capitalize on new or future market growth opportunities. When the rm buys vacant undeveloped land, or when it builds a exible plant in a new geographic location (domestic or overseas) to position itself to take advantage of a developing potentially large market, it essentially acquires or puts in place an expansion or growth option. This option, which will be exercised only if future market developments turn out favorable

Making Use of Real Options Simple

29

at a future date, but not otherwise, can oftentimes make a seemingly unprotable (based on passive NPV) base-case investment worth undertaking. 4. The option to contract. If market conditions turn out weaker than originally expected, management can operate below capacity or even reduce the scale of operations (by c%), thereby saving part of the planned investment outlays (IC ). This exibility to mitigate loss is analogous to a put option on part (c%) of the base-scale project, with exercise price equal to the potential cost savings (IC ), giving max(IC cV, $0). The option to contract (just as the option to expand) may be particularly valuable in the case of new product introductions in uncertain markets. The option to contract may also be important in choosing among technologies or plants with a different construction-tomaintenance cost mix, where it may be preferable to build a plant with lower initial construction costs and higher maintenance expenditures in order to acquire the exibility to contract operations by cutting down on maintenance if market conditions turn out unfavorable. 5. The option to temporarily shut down (and re-start) operations. Actually, the plant does not have to operate (i.e., extract the mineral) in each and every period automatically. In fact, if mineral prices are such that cash revenues are not sufcient to cover variable operating (e.g., maintenance) costs, it might be better not to operate, temporarily. If prices rise sufciently, operations can start up again. Thus, operation in each year can be seen as a call option to acquire that years cash revenues (C) by paying the variable costs of operating (IV ) as exercise price, i.e., max(C IV , 0). If switching costs (between the operating and idle modes) are substantial, delays in switching may arise (hysteresis). Options to alter the operating scale (i.e., expand, contract, or shut down) are typically found in natural resource industries, such as mine operations, facilities planning and construction in cyclical industries, fashion apparel, consumer goods, and commercial real estate. 6. The option to abandon for salvage value. If the quantity of reserves turns out low, if mineral prices suffer a sustainable decline, or if the operation does poorly for some other reason, management does not have to continue incurring the xed costs. Instead, management may have a valuable option to abandon the project permanently in exchange for its salvage value, that is the resale value of its capital equipment and other assets in secondhand markets. This option can be valued as an American put option on current project value with exercise price the salvage or

30

L. Trigeorgis

best alternative use value, entitling management to receive V + max(A V, 0) or max(V, A). Naturally, more general-purpose capital assets would have a higher salvage and option abandonment value than special-purpose assets. Valuable abandonment options are generally found in capital-intensive industries, such as in airlines and railroads, in nancial services, and in new product introductions in uncertain markets. 7. The option to switch inputs or outputs. Suppose the associated mineral processing operation can be designed to use alternative forms of energy inputs (e.g., fuel oil, gas, or electricity) to convert the raw mineral into a variety of output by-products. This would provide valuable built-in exibility to switch from the current input to the cheapest future input, or from the current output to the most protable future product mix, as the relative prices of the inputs or outputs uctuate over time. In fact, the rm should be willing to pay a certain premium for such a exible technology over a rigid alternative that confers no or less exibility. Indeed, if the rm can in this way develop more uses for its assets relative to its competitors, it may be at a signicant comparative advantage. Generally, process exibility can be achieved not only via technology (e.g., by building a exible facility that can switch among alternative energy inputs), but also by maintaining relationships with a variety of suppliers, changing the mix as their relative rates change. Subcontracting policies may allow further exibility to contract the scale of future operations at a low cost in case of unfavorable market developments. A multinational company may similarly locate production facilities in various countries in order to acquire the exibility to shift production to the lowest-cost producing facilities as the relative costs, other local market conditions, or exchange rates change over time. Process exibility is valuable in feedstock-dependent facilities such as oil and minerals, electric power, chemicals, crop switching, and supplier relationships in many industries. Product exibility, enabling the rm to switch among alternative outputs, is more valuable in industries such as automobiles, consumer electronics, toys, or pharmaceuticals, where product differentiation and diversity are important and/or product demand is volatile. In such cases, it may be worthwhile to install a more costly exible capacity to acquire the ability to alter product mix or production scale in response to changing market demands. 8. Corporate growth options. As noted, another version of the earlier option to expand, of considerable strategic importance, is corporate growth options that set the path of future opportunities.

Making Use of Real Options Simple

31

Suppose, in the above example, that the proposed processing facility is based on a new, technologically superior process for mineral renement developed and tested internally on a pilot plant basis. Although the proposed facility in isolation may appear unattractive, it could be only the rst in a series of similar facilities if the process is successfully developed and commercialized and may even lead to entirely new mineral by-products. More generally, many early investments (e.g., R&D, a lease on undeveloped land or a tract with potential oil reserves, a strategic acquisition, or an information technology network) can be seen as prerequisites or links in a chain of interrelated projects. The value of these projects may derive not so much from their expected directly measurable cash ows, but rather from unlocking future growth opportunities (e.g., a new-generation product or process, related mineral reserves, access to a new or expanding market, strengthening of the rms core capabilities or strategic positioning). An opportunity to invest in a rst-generation high-tech product, for example, is analogous to an option on options (an inter-project compound option). Despite a seemingly negative NPV, the infrastructure, experience, and potential by-products generated during the development of the rstgeneration product may serve as spring boards for developing lower-cost or improved-quality future generations of that product, or even for generating new applications into other areas. But unless the rm makes that initial investment, subsequent generations or other applications would not even be feasible. The infrastructure and experience gained can be proprietary and can place the rm at a competitive advantage, which may even reinforce itself if learning cost curve effects are present. Growth options are found in all infrastructure-based or strategic industries, especially in high tech, R&D, and industries with multiple product generations or applications (e.g., semiconductors, computers, pharmaceuticals), in multinational operations, and in strategic acquisitions. In a more general context, such operating and strategic adaptability represented by such a set of corporate real options can be achieved at various stages during the value chain, from switching the factor input mix among various suppliers and subcontracting practices, to rapid product design and modularity in design, to shifting production among various products or countries rapidly and cost-efciently in a exible system or multinational network.

32

L. Trigeorgis

LESSONS AND IMPLICATIONS FOR FLEXIBLE DECISION MAKING The insight gained from viewing investment opportunities through a real options lens can be quite powerful. Real options analysis provides a number of insightful lessons and implications, summarized below. 1. Uncertainty and exibility are two key determinants of the value of an asset or rm. Their understanding calls for an expanded valuation criterion. The traditional valuation paradigm based on cash ows from expected plans under the implicit assumption of passive management has proven inadequate in dynamic settings. The role of uncertainty in the presence of managerial exibility is not necessarily penalizing, as conventional wisdom would have us believe. Greater variability of potential outcomes around the expected (mean) result may be benecial in the presence of options and an asymmetric managerial position. Managerial exibility to revise future decisions when there are deviations from the expected plans introduces benecial asymmetry in the distribution of project value returns by enabling upside (value-creation) opportunities to be exploited fully while limiting downside losses by choosing not to proceed or abandon. The resulting skewing of the probability distribution of expected project returns toward a more positive outcome calls for an expanded (strategic) NPV criterion to also capture the additional value of managerial operating exibility and other strategic interactions: Expanded (or Strategic) NPV = passive NPV + Option Premium (ROV) (Flexibility value + Strategic value) Based on this expanded criterion, it can be seen that it may now be justied to accept projects with negative (passive) NPV of expected cash ows (if this is offset by a larger option premium or real option value as a result of additional exibility and strategic value), or delay investment with positive NPV until a later time when expanded NPV would be maximized under uncertainty. 2. Managerial exibility or real option value (ROV) may be higher (other things the same) for rms or industries facing higher uncertainty; for investment opportunities with longer horizons or that can be delayed longer; when (real) interest rates are higher; or for multi-stage (compound) options.

Making Use of Real Options Simple

33

3. Higher uncertainty tends to increase the value of the option to defer (a single, irreversible, proprietary) investmentprovided there are small or no early exercise benets (analogous to dividends), strategic interactions or other embedded options. The exibility to delay or wait and see enables acquiring more or better information and making a more informed future decision, potentially avoiding a mistake from premature investment in case things would develop unfavorably. This higher value to wait and see necessitates a higher critical investment threshold: the critical project value, V , must be at a signicant premium above the required investment cost, I , before it is justiable to invest and sacrice the option to wait. The implication of this is that higher uncertainty would presumably lead to investing less or later (other things the same), with potentially signicant macroeconomic implications. I should caution, however, that this holds under the provisional conditions made above (and relaxed later) and may be different in different contexts, so it is questionable whether empirical studies on investment based on macro data can verify the presumed depressive role of uncertainty on investment (in terms of lower or delayed investment) if the provisional conditions are not conrmed to be carefully satised. 4. If one can reverse a decision (with ease or little cost), it is easier to make it (e.g., invest) in the rst place. A multinational corporation would nd the decision to enter a new foreign country easier if it can get out with limited damage in case of unfavorable developments. This principle holds in general contexts beyond business investment. For example, the decision to get married might be easier in societies where it is easier to receive divorce (in Islamic parts of India men can obtain divorce simply by proclaiming the word 3 times). In deciding whether to move from the United States to a smaller, more risky country, it helps to make the decision if one has U.S. citizenship rights. These rights give the individual the option to reverse the decision and operate in the best of the two countries over time, just as locating plants in several countries enables a multinational corporation to operate in the best subset of several countries and shift production from one country to another to take advantage of uctuations or differences in exchange rates, labor costs or other production inputs, tax regimes, etc. 5. Under uncertainty, it is prudent to stage an investment or proceed with decision plans in stages. Staging the investment or decision plans provides valuable exibility to continue to the next stage (receiving the option value from continuing) or to abandon (exit)

34

L. Trigeorgis

midway. Continuation (e.g., nancing of subsequent stages in venture capital) should be contingent on the success of earlier stages. 6. Multistage opportunities may have signicant growth (compound) option value that may justify making strategic investments despite having negative NPV. Consider, for example, a two-stage growth option. The rst stage involves investing in a manufacturing facility in Spain to introduce a new product that is expected to generate moderate cash ows from the Spanish market. The second stage would involve a 10-fold expansion into the broader European market 3 years later. The rst-stage NPV of the expected cash ows from investing in the Spanish market is negative, and committing now to enter the European market on an expanded scale seems ten times as bad. But the company does not have to commit to European expansion now. Instead, it has an option to wait and see how the Spanish and European demand develops and expand to the European market if and only if it appears favorable to do so 3 years from now. The opportunity to expand in Europe, valued as an option, may well offset the negative NPV of the rst-stage investment (in effect the option premium or exercise price that needs to be paid to acquire the European expansion option) and can justify making this strategic multistage investment on strategic grounds. Empirically, companies in industries with higher uncertainty that involve multistage (compound) options tend to have a higher proportion of their stock price deriving from growth opportunities (PVGO/P), providing an indirect conrmation of the validity of real option theory predictions. 7. If investing would open up or create other options within the project (e.g., to later expand, abandon, or switch to alternative uses), then more uncertainty would also increase the exibility value of these other embedded optionsincreasing the value of early investing in the rst place. For this reason (and the other reasons listed below), higher uncertainty would not necessarily suppress or delay investment. 8. In the presence of competition in an oligopoly setting, early investment may have strategic value by inuencing the equilibrium actions (quantity or price setting) of competitors in a way benecial to the investing rm or even by preempting competitive entry altogether in some cases. Thus, the option value of waiting must be traded off against the strategic commitment value of early investing. Again, the impact of higher uncertainty on investment is not clear-cut; in fact, it may not even vary monotonically

Making Use of Real Options Simple

35

with demand as shifts in demand may lead to shifts in the type of equilibrium games and different market structure outcomes (e.g., from a Nash duopoly to a Stackelberg leader/follower game or a monopoly) in different demand zones leading to value discontinuities as a function of demand. The value of the strategic investment and the optimal competitive strategy (e.g., to invest now or wait) depends on whether the resulting benets of the investment are proprietary or shared and whether they are damaging or beneting the competitor, as well as on whether competitive reaction is expected to be contrarian (opposite to the action of the investing rm) or reciprocating (similar to the action of the investing rm). When the investment benets are proprietary and the pioneer can get stronger at the expense of its competitor, it should commit to an early investment (aggressive) strategy if the competitors reaction is contrarian; e.g., if it will retreat and cut its market share under quantity competition as the pioneer expands its own market share. However, when the benets are shared, thereby beneting the competitor as well, and a contrarian competitor would respond aggressively, taking advantage of the pioneers accommodating position, the rm should follow a exible wait-and-see strategy rather than subsidizing an aggressive competitor while itself paying the full cost. The above can be reversed under reciprocating (price) competition. If the benets are shared and will benet a competitor who will reciprocate when treated nicely (e.g., by maintaining high prices) the optimal strategy might be to invest early (but not aggressively). On the contrary, if the benets are proprietary and will hurt a competitor who will retaliate by entering into a price war, it may make better sense to wait or not invest. 9. Competitive pressure may induce rms (e.g., in a winner takes all innovation race) to invest prematurely, resulting in a suboptimal prisoners dilemma situation. Each of the two rms (like prisoners), being afraid that it may be preempted by the other and lose all (the most severe punishment), would rush to invest prematurely (give in), rather than wait (hold out), which may be the preferred outcome. A joint research venture may enable the two rms to more fully appropriate the exibility value from waiting (avoiding the prisoners dilemma) by coordinating and jointly optimizing against demand uncertaintybesides sharing and saving on the investment cost. A limitation is that, in collaborating, a rm gives up the possibility to outwit its rivals and gain a competitive advantage or strategic value over the other rm.

36

L. Trigeorgis

10. Multiple options embedded in a project may interact; i.e., option value additivity may break down. The presence of a later option enhances the value of the underlying asset for a prior option, while exercising an earlier option may alter the scale of (and in the case of the option to abandon, may extinguish) a later option. The value of a portfolio or combination of embedded options typically is less than the sum of separate or independent option values. Therefore, using an analytic formula like Black-Scholes to determine the values of separate options and then add them up may be misleading. The error from adding up separate option values may be of the same order of magnitudebut in the opposite directionas the error from ignoring options altogether. That is, a wrongly executed options analysis can be as dangerous as a na ve NPV analysis. 11. Options to switch (among the cheapest of several inputs, best of several outputs, or most protable countries of operation) provide valuable exibility and risk management value. Traditional mean-variance portfolio theory based on the notion that risk is undesirable and must therefore be minimized (for a given expected return) is inadequate; it needs to be extended for portfolios of (potentially interdependent) options, incorporating higher moments. The exibility to adjust plans when deviating from expectations by improving the upside potential while limiting the downside risk adds asymmetry or skewness (third moment), while potential volatility dependence on project value, competitive jumps, and technological disruptions may introduce fat tails and kurtosis (fourth moment). The very notion and role of risk must also be revisited when exibility is present. With options to choose the best of several alternatives (or on the maximum or minimum of several assets) or options to switch from one mode of operation or being to another, lower correlation tends to increase the relative volatility and option value of a exible system or network. When the value of one alternative drops, an option to choose the best or switch to another alternative is worth more if the value of that second alternative tends to increase (that is, if it has negative correlation with the rst). For this reason, multinational corporations (MNCs) operating in several countries would prefer to select the next strategic location (to be added to their multinational network portfolio) to have lower correlation (with the existing structure), not so much in order to diversify and reduce risk, but rather in order to increase the relative volatility and option value of the exible network. Risk is not necessarily something to be avoided or be penalized for, but rather can be

Making Use of Real Options Simple

37

seen as a window of opportunity for the more exible and innovative corporations to create more value by leveraging their opportunity choices while limiting losses. 12. When switching among operating modes or strategies, the presence of signicant switching costs (e.g., to enter, exit, or shut down) may induce a hysteresis, inertia or delay/lag effect. Even though immediate switching may be attractive based on short-term cash-ow considerations, it may be long-term optimal to wait, e.g., due to a high switching cost or probability of switching back later. Examples involving hysteresis effects include continuing operations of a currently unprotable mine or oil eld despite temporarily suppressed prices; the Japanese auto producers who, once they entered the U.S. market in protable times, kept hanging on in the United States despite incurring losses in subsequent years; lags in hiring and ring by companies as business moves to an up and down cycle; and delays in seeking divorce despite an unhappy marriage. All these cases involve irreversible or costly-to-reverse decisions that justify delaying a switching decision for a while since a re-switch back to the current situation is either infeasible or would occur only after a costly impairment of infrastructure, goodwill, etc. MODULAR STRUCTURING AND SIMPLIFYING OF COMPLEX REAL OPTION PROBLEMS Most real-life problems involve more complex combinations of the above (and occasionally other) options. However, one can simplify a complex investment decision problem structure by decomposing it into a few basic building-block option types (such as the standard options in the previous section) connected by some basic decision operations. Figures 1AD review some basic or common types of real options. In the gures, options are illustrated with a hexagonal symbol, shown in the second column, with the relevant payout shown at the bottom depending on option type (e.g., C + V for a call option; for shorthand, the max( , 0) is implied by the option symbol itself and is omitted). A hat () over the investment cost to be paid (C ) or the (gross) PV of project cash ows (V ) indicates that they are uncertain (stochastic). The third column in Figure(s) 1 presents a relevant authors analytic model (e.g., Black-Scholes for the call option to wait in Figure 1A). The last column lists actual case applications (mostly from the authors experiences) most appropriate for that type of option. Figure 1A reviews the options to wait (call on project value V ) and to expand (call on e% of V ) using the Black-Scholes [7] formula (on V or eV ,

38

L. Trigeorgis

Figure 1A. Basic/common option types.

respectively), to abandon or contract (put on c% of V), and Margrabes [30] option to exchange or switch one asset for another (here cost C for project value V). Figure 1B focuses on compound options, including the simple compound (or pure growth) option by Geske [12], sequential two-stage

Figure 1B. Basic option types: Compound options.

Making Use of Real Options Simple

39

Figure 1C. Basic option types: Max options.

compound exchange option, and an N -stage compound exchange option representing a numerical generalization of the above. Figure 1C reviews options involving the max, including the Stulz [45] and Johnson [18] models on European options to pay cost C by a given maturity to

Figure 1D. Basic option types: And options.

40

L. Trigeorgis

acquire the max of two or more assets, and the generalized max (OR) operator on the best of 2 or more option alternatives (e.g., applicable in analyzing the choice among the best of two product standards, the digital and analog, in the case of Philips Electronics). Combinations of the above types of options and their possible interactions leading to non-additivity of separate option values are discussed in Trigeorgis [53]. Figure 1D illustrates basic AND type options. They include the sum of N parallel (or strategically independent) options shown vertically, such as expanding geographically in continents A, B, and C or introducing new services D, E, and F, each involving the right to pay (stochastic) cost Ci by time ti to acquire underlying asset value Vi (i = 1, . . . , N ). They also include modeling manufacturing operations of a plant (e.g., Shell Chemicals exible plastic manufacturing plant) as the sum of T European operating options across time (shown horizontally), provided switching costs are negligible. Each year of operation may be modeled as a European option to exchange a variable operating cost (C ) for the value of generated cash ows (V ), using the Margrabe option to exchange one asset for another, or as an option to incur the variable cost (C ) to obtain the maximum revenue from producing the best of two (or more) products, using the Stulz or Johnson options on the maximum of several assets. In general, this approach may accommodate a generic option payout of the form cC + aV 1 + bV 2 + function (V 1, V 2. . . ) where C , V 1, V 2, . . . can be uncertain (stochastic) variables and constants a , b, c can be of any sign. Such generic options (hexagons) can occur in any combinations over multiple stages, connected with one or more of four basic decision operators (with the OR, AND, AVG operators shown in squares), to represent the basic option structure of most real life problems. The four basic decision operations commonly encountered are illustrated in Figure 2. They involve (a) the choice of the best among several mutually exclusive alternatives (OR or MAX); (b) the sum of several (parallel or strategically independent) options (AND); (c) the probabilistic average (AVG) of several follow-on options across various technical scenarios weighted by their corresponding actual probabilities, or investing in a portfolio (weighted average) of several technological or other options by allocating a portion of a given budget in each of these options; and (d) recursive multistage or compound/sequential options (COMP) that may provide not only their own cash ows but also provide follow-on options. In all the above basic decision operators, both the cost of exercising an option (C ) and the present value of resulting project cash ows (V ) may be uncertain. Although Figure 2 illustrates decision operators involving 2 decision branches or two stages, these are readily extended to N decision branches or to N stages (as illustrated in subsequent gures).

Making Use of Real Options Simple

41

Figure 2. Basic decision operations.

Valuation would proceed in a recursive (even modular) manner starting from the end and moving backward following the standard risk-neutral option valuation procedure. Figure 3A illustrates a standard basic problem structure (combination of standard options and decision operators) for two real-life applications taken from two entirely different industries. The energy industry case involves a staged power plant construction with options to abandon midstream and later expand. The pharmaceutical industry case involves valuing R&D during the clinical trials phase and determining the value of the patent rights for a new drug (also involving options to abandon and to expand into a related niche). The basic problem structure and related option maps look remarkably similar, despite differences in problem context and industry characteristics. In both situations each rm faces a compound (COMP) or multistage option to start (in an early stage) and then complete (at a later stage) a development process (building a plant or completing clinical trials), followed by an option on the best alternative (OR), either to continue with commercial production operations (also involving a later option to expand) or to abandon for a salvage or sale value. Since the interim decision on the option to complete development/continue or abandon partly depends on the value of the subsequent option to later expand, that option must be determined rst, starting from the end, and then used in the backward valuation process along the way to make the earlier choice (to continue or abandon) in the previous step, and so on.

42

L. Trigeorgis

Figure 3A. Similar basic structure from different industry applications.

Figure 3B. Details on Glaxo case application.

Making Use of Real Options Simple

43

Figure 4. Eli Lilly case application problem structure.

Figure 3B illustrates the problem structuring option map for the above pharmaceutical case (Glaxo) with more details on the timing, staged costs, base project value (present value of cash ows from launching the oral version of the drug), expansion costs and factors, salvage (sell) value, etc. Figure 4 illustrates the problem structuring for another pharmaceutical company (Eli Lilly), involving a committed decision to launch a basic version of a drug (PTCA) and options to pursue two extensions (AMI and Angina). As shown in the top gure, the second drug extension (AMI) may be introduced alone or following the rst extension (Angina), in which case it may benet from both structural synergies (since the decision to launch AMI would benet from prior knowledge of the success of the preceding Angina introduction) as well as from economic synergies (as the costs of convincing doctors of the benets of the AMI extension would be lower, and the market expansion factor higher, following prior introduction of Angina). The lower gure casts the above managerial decision problem as involving the choice between a sequential (compound option) vs. a parallel (sum of independent options) marketing expansion strategy. Figure 5 illustrates the problem structuring for two HBS case applications involving probabilistic options (AVG). In the Arundel case (top panel), the success of a rst movie will be revealed in the theatres by year 1 with probability p , leading to an (call) option to make a sequel movie within the subsequent 2 years. In the Antamina case (lower panel), the

44

L. Trigeorgis

Figure 5. Probabilistic (AVG) case applications.

winner of a government bid will have 2 years to decide whether to incur the development cost, C2 , to receive the value of copper and zinc reserves, V , which depends on the stochastic copper and zinc prices ( Pc , Pz ). The probabilities of high (H), medium (M) and low (L) reserves, accounted for through the AVG decision operator, would weigh the resulting option values conditional on nding high, medium, or low reserves in year 2. Figure 6A illustrates use of the AND operator to access the growth option value of a high-tech IPO. Tiscalis existing business at the time of the IPO was only xed-line telephony, but its management envisioned plans to rst offer e-commerce services in Italy and then expand them in

Figure 6A. Tiscali (AND) case application: Growth option value of hightech IPO.

Making Use of Real Options Simple

45

Figure 6B. Details on Tiscali case application.

the broader European market (a compound option). It also expressed plans to acquire a UMTS license that would enable it to offer integrated 3rd generation mobile services. More details on Tiscalis problem structuring and option map are given in Figure 6B. Figure 7 illustrates the option map for staging network (NW) infrastructure for a leading UK telecom operator. Each staged network investment (e.g., NW1 ) provides the sum of 3 options: (a) expand by acquiring new business clients by the way the network would be physically laid out, (b) switch customers initially using a low-bandwidth technology to a high-bandwidth one, and (c) proceed to the next stage (NW2), involving similar such options, and so on. The reader can appreciate that there is no analytic, closed-form solution or an easy modeling approach for such problems, but they can readily be handled through the recursive and modular structure proposed herein. PAST AREAS OF APPLICATION Besides theoretical developments, real option applications have been receiving increased attention. Real options valuation has been applied in a variety of contexts, such as in natural resource investments, land development, leasing, exible manufacturing, government subsidies and regulation, R&D, new ventures and acquisitions, foreign investment and strategy,

46

L. Trigeorgis

Figure 7. Telecom case application problem structure: Staging network (NW) infrastructure.

and elsewhere. The review below is just indicative and is in no way meant to be exhaustive or capture all the important contributions. Natural Resource Investments Early applications naturally arose in the area of natural resource investments due to the availability of traded resource or commodity prices, high volatilities, and long durations, resulting in higher and better option value estimates. Brennan and Schwartz [8, 9] utilize the convenience yield derived from futures and spot prices of a commodity to value the options to shut down or abandon a mine. Paddock, Siegel, and Smith [37] value options embedded in undeveloped oil reserves and provide empirical evidence that option values are better than actual DCF-based bids in valuing offshore oil leases. Trigeorgis [50] values an actual minerals project considered by a major multinational company involving options to cancel during construction, expand production, and abandon for salvage. Bjerksund and Ekern [5] value a Norwegian oil eld with options to defer and abandon. Morck, Schwartz, and Stangeland [36] value forestry resources under stochastic inventories and prices. Laughton and Jacoby [27] examine biases in the valuation of real options and long-term decision making under a mean-reversion price process. Kemna [20] shares her experiences with Shell in analyzing actual cases involving the timing of developing an offshore oil eld, valuing a growth option in a manufacturing venture, and the abandonment decision of a rening production unit.

Making Use of Real Options Simple

47

Land Development Titman [47], Williams [57], and Quigg [40] show that the value of vacant land should reect not only its value based on its best immediate use (e.g., from constructing a building now), but also its option value if development is delayed and the land is converted into its best alternative use in the future. It may thus pay to hold land vacant for its option value even in the presence of currently thriving real estate markets. Quigg [39] reports empirical results indicating that option-based land valuation that incorporates the option to wait to develop land provides better approximations of actual market prices. In a different context, McLaughlin and Taggart [35] view the opportunity cost of using excess capacity as the change in the value of the rms options caused by diverting capacity to an alternative use. Grenadier [14] developed a model of real estate development, offering an explanation for observed market behavior in land development, such as overbuilding. Flexible Manufacturing The exibility provided by exible manufacturing systems, exible production technology, or other machinery having multiple uses has been analyzed from an options perspective by Kulatilaka [24, 25], Triantis and Hodder [48], Kulatilaka and Trigeorgis [26], and Kamrad and Ernst [19], among others. Kulatilaka [25] values the exibility provided by an actual dual-fuel industrial steam boiler that can switch between alternative energy inputs (natural gas and oil) as their relative prices uctuate, and nds that the value of this exibility far exceeds the incremental cost over a rigid, single-fuel alternative. Baldwin and Clark [2] study the exibility created by modularity in design that connects components of a larger system through standard interfaces. Leasing Contracts Copeland and Weston [10], Lee, Martin, and Senchack [28], McConnell and Schallheim [33], and Trigeorgis [54] value various operating options embedded in leasing contracts. Grenadier [13] uses real options to develop a model that can be used to price different types of leasing contracts. R&D/Innovation Kolbe, Morris, and Teisberg [23] discuss option elements embedded in R&D projects. Option elements involved in the staging of start-up ventures

48

L. Trigeorgis

are also discussed in Sahlman [41] and Willner [58]. Grenadier and Weiss [15] examine the situation in which a rm has projects that involve sequential innovation. Their valuation incorporates the learning that assists the valuation and decision-making pertaining to future innovations. Security Analysis Real options have been advocated as an approach to valuing companies in security analysis. For example, Mauboussin [32] discusses how real options valuation can be used to supplement traditional valuation approaches. He suggests that real options valuation is particularly useful in valuing companies that are R&D intensive as well as Internet companies. Kester [21] estimates that the value of a rms growth options is more than half the market value of equity for many rms, even 7080% for more volatile industries. Similarly, Pindyck [38] suggests that growth options represent more than half of rm value if demand volatility exceeds 20%. Berger et al. [4] empirically access investors valuation of the abandonment option. Foreign Investment Baldwin [1] discusses various location, timing, and staging options present when rms scan the global marketplace. Bell [3], and Kogut and Kulatilaka [22], among others, examine entry, capacity, and switching options for rms with multinational operations under exchange rate volatility. Hiraki [17] suggests that the Japanese bank-oriented corporate governance system serves as the basic infrastructure that enables companies to jointly develop corporate real options. Other Applications Strategic acquisitions of other companies also often involve a number of growth, divestiture, and other exibility options, as discussed by Smith and Triantis [44]. Luchrman [29] discusses viewing strategy as a portfolio of real options. Smit and Ankum [42] and Smit and Trigeorgis [43] propose a game-theoretic approach to corporate investment strategy. Mason and Baldwin [31] value government subsidies to large-scale energy projects as put options, whereas Teisberg [46] provides an option valuation analysis of investment choices by a regulated rm. Various other option applications can be found in areas ranging from shipping [6] to environmental pollution and global warming (e.g. [16]). The potential for future applications itself seems like a growth option.

Making Use of Real Options Simple

49

RESEARCH CHALLENGES AHEAD Despite signicant progress in recent years, some long-standing gaps and challenges remain. Here is a list of challenging issues that future research must still address: 1. Studying more actual case applications and tackling real-life implementation issues and problems. 2. Studying investments (such as in R&D, pilot or market tests, or excavations) that can generate information and learning (e.g., about the projects prospects) by extending/adjusting option pricing and risk-neutral valuation. 3. Exploring in more depth endogenous competitive counteractions and a variety of competitive/market structure and strategic issues using a combination of game-theoretic industrial organization with option valuation tools. 4. Better modeling of the various strategic and multistage growth options. 5. Extending real options in an agency context recognizing that the potential (theoretical) value of real options may not be realized in practice if managers, in pursuing their own agenda (e.g., expansion or growth, rather than rm-value maximization), misuse their discretion and do not follow the optimal exercise policies implicit in option valuation. This raises the need to design proper incentive contracts by the rm (taking also into account asymmetric information) and develop a more dynamic, option-based extension of economic value added. 6. Better recognizing that real options may interact not only among themselves but with nancial exibility options as well, and understanding the resulting implications for the combined, interdependent corporate investment and nancing decisions. 7. On the practical side, applying real options to the valuation of exibility in related areas, such as in competitive bidding, information technology, or other platform investments, international nance options, and so on. 8. Using real options to explain empirical phenomena that are amenable to observation or statistical testing, such as examining empirically whether the management of rms that are targets for acquisition may sometimes turn down tender offers in part due to the option to wait in anticipation of receiving better future offers. We also need more empirical studies to conrm other qualied predictions of options theory, such as the impact of uncertainty on investment.

50

L. Trigeorgis

9. Doing more eld or survey studies to test the conformity of theoretical real option valuation and its implications with managements intuition and experience, as well as with actual data when available. 10. Developing a more credible general portfolio theory for (possibly interdependent) options under budget or other constraints that recognizes the potentially benecial role of uncertainty in the presence of exibility to select the best subset among alternative options. Finally, applying this to address important strategic portfolio problems in various contexts, such as management of a portfolio of start-up ventures by a venture capitalist, development of a pipeline portfolio of R&D opportunities by a pharmaceutical company, or selection of a subset of technologies to invest in by a telecom company, and dynamic revision of the subset of invested technologies as uncertainties and their relative merits change over time. REFERENCES
[1] Baldwin, C., Competing for capital in a global environment, Midland Corporate Finance Journal, Vol. 5, 1987, pp. 4364. [2] Baldwin, C. and K. Clark, Modularity and real options, Working Paper, Harvard Business School, 1993, Boston, MA. [3] Bell, G., Volatile exchange rates and the multinational rm: Entry, exit, and capacity options, In L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions, New York, Praeger, 1994. [4] Berger, P.G., E. Ofek, and I. Swary, Investor valuation of the abandonment option, Journal of Financial Economics, Vol. 42, 1996, pp. 257287. [5] Bjerksund, P. and S. Ekern, Managing investment opportunities under price uncertainty: From last chance to wait and see strategies, Financial Management, Vol. 19, 1990, pp. 6583. [6] Bjerksund, P. and S. Ekern, Contingent claims evaluation of mean-reverting cash ows in shipping, in L. Trigeorgis (ed.), Real Options in Capital Investment: New contributions (NY: Praeger, New York,). 1994. [7] Black, F. and M. Scholes, The pricing of options and corporate liabilities, Journal of Political Economy, Vol. 81, 1973, pp. 637659. [8] Brennan, M. and E. Schwartz, A new approach to evaluating natural resource investments, Midland Corporate Finance Journal , 1985, pp. 3747. [9] Brennan, M. and E. Schwartz, Evaluating natural resource investments, Journal of Business, Vol. 58, 1985, pp. 135157. [10] Copeland T. and J. F. Weston, A note on the evaluation of cancellable operating leases, Financial Management, Vol. 11, 1982, pp. 6067. [11] Dixit, A., Investment and hysteresis, Journal of Economic Perspectives, Vol. 6, 1992, pp. 6787.

Making Use of Real Options Simple

51

[12] Geske, R., The valuation of compound options, Journal of Financial Economics, Vol. 7, 1979, pp. 6381. [13] Grenadier, S.R., Valuing lease contracts: A real-options approach, Journal of Financial Economics, Vol. 38, 1995, pp. 297332. [14] Grenadier, S.R. The strategic exercise of options: Development cascades and overbuilding in real estate markets, Journal of Finance, Vol. 51, 1996, pp. 1653 1679. [15] Grenadier, S.R. and A. M. Weiss, Investment in technological innovations: An option pricing approach, Journal of Financial Economics, Vol. 44 No. 3 1997, pp. 397416. [16] Hendricks, D., Optimal policy responses to an uncertain threat: The case of global warming, Working Paper, Harvard University Kennedy School of Government, Boston, MA, 1991. [17] Hiraki, T., Corporate governance, long-term investment orientation, and real options in Japan, in L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions (Praeger, New York), 1994. [18] Johnson, H., Options on the maximum or the minimum of several assets, Journal of Financial and Quantitative Analysis, Vol. 22, 1987, pp. 277284. [19] Kamrad, B. and R. Ernst, Multiproduct manufacturing with stochastic input prices and output yield uncertainty, In L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions (Praeger, New York), 1994. [20] Kemna, A., Case studies on real options, Financial Management, Vol. 22, 1993, pp. 259270. [21] Kester, W. C., Todays options for tomorrows growth, Harvard Business Review, Vol. 62, 1984, pp. 153160. [22] Kogut, B. and N. Kulatilaka, Operating exibility, global manufacturing, and the option value of a multinational network, Management Science, Vol. 40, 1993, pp. 123139. [23] Kolbe, A. L., P. A. Morris, and E. O. Teisberg, When choosing R&D projects, go with long shots, Research-Technology Management, 1991, pp. 3540. [24] Kulatilaka, N., Valuing the exibility of exible manufacturing systems, IEEE Transactions in Engineering Management, Vol. 35, 1988, pp. 250257. [25] Kulatilaka, N., The value of exibility: The case of a dual-fuel industrial steam boiler, Financial Management, Vol. 22, 1993, pp. 271279. [26] Kulatilaka, N. and L. Trigeorgis, The general exibility to switch: Real options revisited, International Journal of Finance, Vol. 6, 1994, pp. 778 798. [27] Laughton, D. G. and H. D. Jacoby, Reversion, timing options, and long-term decision-making, Financial Management, Vol. 22, 1993, pp. 225240. [28] Lee, W., J. Martin, and A. Senchack, The case for using options to evaluate salvage values in nancial leases, Financial Management, Vol. 11, 1982, pp. 33 41. [29] Luehrman, T., Strategy as a portfolio of real options, Harvard Business Review, Vol. 76, 1998, pp. 8999. [30] Margrabe, W., The value of an option to exchange one asset for another, Journal of Finance, Vol. 33, 1978, pp. 177186.

52

L. Trigeorgis

[31] Mason, S. P. and C. Baldwin, Evaluation of government subsidies to large-scale energy projects: A contingent claims approach, Advances in Futures and Options Research, Vol. 3, 1988, pp. 169181. [32] Mauboussin, M. J., Get real: Using options in security analysis, Frontiers of Finance, Vol. 10, 1999, pp. 230. [33] McConnell, J. and J. Schallheim, Valuation of asset leasing contracts, Journal of Financial Economics, 1983, pp. 237261. [34] McDonald, R. and D. Siegel, The value of waiting to invest, Quarterly Journal of Economics, Vol. 101, 1986, pp. 707727. [35] McLaughlin, R. and R. Taggart, The opportunity cost of using excess capacity, Financial Management, Vol. 21, 1992, pp. 1223. [36] Morck, R., E. Schwartz, and D. Stangeland, The valuation of forestry resources under stochastic prices and inventories, Journal of Financial and Quantitative Analysis, Vol. 24, 1989, pp. 473487. [37] Paddock, J., D. Siegel, and J. Smith, Option valuation of claims on physical assets: The case of offshore petroleum leases, Quarterly Journal of Economics, Vol. 103, 1988, pp. 479508. [38] Pindyck, R., Irreversible investment, capacity choice, and the value of the rm, American Economic Review, Vol. 78, 1988, pp. 969985. [39] Quigg, L., Empirical testing of real option-pricing models, Journal of Finance, Vol. 48, 1993, pp. 621640. [40] Quigg, L., Optimal land development, In L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions (Praeger, New York), 1994. [41] Sahlman, W., Aspects of nancial contracting in venture capital, Journal of Applied Corporate Finance, Vol. 1, 1988, pp. 2336. [42] Smit, H.T.J. and L.A. Ankum, A real options and game-theoretic approach to corporate investment strategy under competition, Financial Management, Vol. 22, 1993, pp. 241250. [43] Smit, H.T.J. and L. Trigeorgis, Strategic Investment: Real Options and Games, Princeton University Press, Princeton, NJ, 2004. [44] Smith, K.W. and A. Triantis, The value of options in strategic acquisitions, In L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions (Praeger: New York), 1994. [45] Stulz, R., Options on the minimum or the maximum of two risky assets: Analysis and applications, Journal of Financial Economics, Vol. 10, 1982, pp. 161 185. [46] Teisberg, E., An option valuation analysis of investment choices by a regulated rm, Management Science, Vol. 40, 1994, pp. 535548. [47] Titman, S., Urban land prices under uncertainty, American Economic Review, Vol. 75, 1985, pp. 505514. [48] Triantis A. and J. Hodder, Valuing exibility as a complex option, Journal of Finance, Vol. 45, 1990, pp. 549565. [49] Trigeorgis, L., A conceptual options framework for capital budgeting, Advances in Futures and Options Research, Vol. 3, 1988, pp. 145167. [50] Trigeorgis, L., A real options application in natural resource investments, Advances in Futures and Options Research, Vol. 4, 1990, pp. 153164.

Making Use of Real Options Simple

53

[51] Trigeorgis, L., A log-transformed binomial numerical analysis method for valuing complex multi-option investments, Journal of Financial and Quantitative Analysis, Vol. 26, 1991, pp. 309326. [52] Trigeorgis, L., Real options and interactions with nancial exibility Financial Management, Vol. 22, 1993, pp. 202224. [53] Trigeorgis, L., The nature of option interactions and the valuation of investments with multiple real options, Journal of Financial and Quantitative Analysis, Vol. 28, 1993, pp. 120. [54] Trigeorgis, L. Evaluating leases with complex operating options, European Journal of Operational Research, Vol. 91, 1996b, pp. 315329. [55] Trigeorgis, L., Real Options: Managerial Flexibility and Strategy in Resource Allocation, The MIT Press, Cambridge, MA, 1996. [56] Trigeorgis, L. and S. P. Mason, Valuing managerial exibility, Midland Corporate Finance Journal, Vol. 5, 1987, pp. 1421. [57] Williams, J., Real estate development as an option, Journal of Real Estate Finance and Economics, Vol. 4, 1991, pp. 191208. [58] Willner, R., Valuing start-up venture growth options, In L. Trigeorgis (ed.), Real Options in Capital Investment: New Contributions. Praeger, New York, 1994.

BIOGRAPHICAL SKETCH
LENOS TRIGEORGIS is the Bank of Cyprus Chair Professor of Finance at the University of Cyprus (lenos@ucy.ac.cy). He is also President of the Real Options Group (www. rogroup.com). He previously taught at Boston University, Columbia University, and the University of Chicago. He holds a Ph.D. (DBA) from Harvard University. He published widely in numerous journals on corporate nance, competition and strategy, and has written a number of books on real options with MIT Press, Oxford University Press, Princeton U. Press, and others. He is internationally known as the author of Real Options (recently translated in Japanese), which is considered path-breaking for the eld. He recently also published a co-authored book, Strategic Investment. His consulting experiences include British Petroleum, the Fiat Group, Cable & Wireless, Swisscom, Andersen Consulting/Accenture, Ernst & Young, Morgan Stanley, and the U.S. government.