,

CHAPTER 9

REAL OPTIONS FOR INVESTMENT VALUATION

This chapter considers investments that can be staged such that decisions can be made at each stage whether or not to continue investment for the next stage. This multistage approach provides decision makers with considerably flexibility. This flexibility enables hedging against the downside possibility that an investment no longer makes sense because either an earlier stage has failed or the opportunity for which the investment was targeted no longer makes sense.

Framing investment decisions in this way results in each decision to proceed with the next stage as equivalent to purchasing an option on the subsequent stages. The option is the right, but not the requirement, to invest in the next stage at a later time. If the results of the current stage are positive, and the opportunity embodied in the next stage still looks attractive, then one can exercise this option and proceed.

The decision to proceed with investing in the next stage may, in effect, result in the purchasing of an option on the subsequent stages. In this way, the series of investment decisions may be represented as a stream of option purchase decisions, followed by option exercise decisions. Only at the last stage is the asset deployed in the marketplace. However, as later discussion shows, this deployment can also be represented as creating another option for future market offerings.

This chapter proceeds as follows. First, the nature of options is elaborated. This includes the logic of options, option pricing models, and computational methods, with emphasis on Black–Scholes formulations. Strategic metrics for characterizing the worth of investments are then discussed, building systematically on the discussion in Chapter 8. Investment decision making then becomes the focus, including the notions of investment portfolios and discussion of a wealth of investment case studies. Finally, the construct of value-centered investing is elaborated where the core notion is option-based strategic thinking about the contingent needs of an enterprise.

9.1 NATURE OF OPTIONS

An option provides the right to do something. Purchasing an option provides a “chit” that can later be used, or exercised, if one decides that doing what the option enables still makes sense. For example, one might buy an option that enables buying a fixed number of shares of an enterprise for a given price at any time within the next five years. If the market value of these shares exceeds the option price, it is said that the option is “in the money” and worth exercising. Depending on the type of option (see below), one might use the chit and buy the shares for an immediate gain.

Options do not always pertain to shares of stock. One might buy an option to buy a fixed amount of commodities such as corn, pork bellies, or fuel at a particular price at a specific point of time. In this way, one could hedge the downside risks of these commodities becoming much more expensive and undermining the competitiveness of one’s products and services. If the price of one’s products and services cannot be raised, perhaps due to intense competition, then such an option greatly decreases the risk of having to sell products and services for less than the costs of producing and/or delivering them.

The models and methods discussed in this chapter originated in the financial industry. This industry deals with a variety of types of options or, in general, derivatives that attach value to the right to something rather than the thing itself. A “call” option is the right to buy something on or before a particular point in time for a particular price. In contrast, a “put” option is the right to sell something on or before a particular point in time for a particular price.

There are European and American versions of calls and puts. European options can only be exercised at the end point of the time period during which the option is valid. An American option can be exercised at any time up to and including that point in time. European call options are the best representation of the types of investments discussed in this chapter, as well as in Chapter 8.

An overarching question for any of these types of options is their value. This determines how much one should be willing to pay for an option. Black and Scholes (1973) and Merton (1973) developed an analytic solution for European call options. There was widespread adoption of these models throughout the financial industry within one year, a rather amazing rate of diffusion of an innovation.1

In the 1990s, the idea of “real” options received considerable attention—see Dixit and Pindyck (1994), Trigeorgis (1996), Luenberger (1997), Amram and Kulatilaka (1999), and Boer (1999). The idea was to use the models, methods, and tools applied for financial options to investment problems associated with tangible assets such as factories, products, and technologies.

9.1.1 Multistage Options

Figure 9.1 illustrates two multistage real options. The top example includes two stages. The first stage involves, for example, an investment in R&D to create a new technology to enable some function of importance to the investor. If this R&D is successful, and the need for the function continues, the investor may choose to exercise the option gained by funding the first stage and, therefore, invest in the second stage.

The bottom example in Fig. 9.1 includes three stages. The first stage involves investing in research to create a potential new function. If this research is successful, and the function now possible remains attractive, the investor may decide to invest in the second stage to develop the research outcomes into a full capability. If this capability is successfully developed, and the functionality enabled still remains desirable, the investor may invest in the third stage to deploy the capability.

The options associated with the first stages of both of the examples in Fig. 9.1 have economic value regardless of whether or not these options are exercised in later stages. This may seem counterintuitive. Why would one buy something that one does not subsequently use? This possibility may better align with intuition if one thinks about options such as insurance. Owning an option ensures that one has the right to an attractive opportunity if the opportunity emerges in the future. Life insurance is attractive if one dies, but few people would regret not exercising their life insurance last year! Nevertheless, people are usually glad to have an option—an insurance policy—that meets the contingent need if one were to die.

9.1.2 Fundamental Logic

There is a fundamental logic for addressing the types of investments embodied in the two case studies introduced in Chapter 8 (Rouse et al., 2000). The idea is that technology and process investments create contingent opportunities for later solution investments. In other words, they create options that can later be exercised and result in deployment of upgraded or new market offerings. Thus, these technology and process investments not only yield new processes and technologies but also yield options for later investments—all for the costs of the initial investments.

There are usually significant uncertainties regarding likely cash flows from later investments to exercise options and deploy upgraded or new offerings. If there were not, one would likely commit to deployment up front. However, the greater the uncertainty or volatility, the more attractive it is to own options rather than being committed to the uncertain future now. The options give one the right but not the requirement to make the deployment decision later.

Thus, delaying investment decisions—rather than deciding now—can be of substantial value. The farther into the future that a decision is to be made, the more valuable the associated option. Intuitively, one would rather not make decisions now that could be made (much) later, especially when there is significant uncertainty. Therefore, the value of an option (OV or Option Value) typically increases with delay time for exercising it.

In summary, the OV increases with projected cash flows should one exercise the option, uncertainty (volatility) associated with these cash flows, and time until the decision to exercise it need be made. It may seem unusual that uncertainty and time increase value, but the key point is that they increase the OV on a future that may not materialize. That is exactly why one prefers to own an option.

9.2 OPTION PRICING THEORY

The question we address here is how to estimate or assess the economic OV on uncertain future cash flows that may or not be realized. Black and Scholes (1973) addressed this by envisioning a “replicating portfolio” that consists of some number of owned shares of stock and borrowed capital with interest paid at the risk-free rate. This replicating portfolio will have the same payoff as the call option at expiration and therefore, by the fundamental theorem of finance, the portfolio value must equal the call option value. They constructed this portfolio to be entirely self-financing and thus deterministic.

This conceptual insight led them (Black and Scholes, 1973), with contributions from Merton (1973), to derive the now-famous Black–Scholes equation starting with

(9.1)

(9.2)

where S is the price of the underlying security, z is a standard Brownian motion or Wiener process over [0, T], and B is the value of a risk-free bond carrying an interest rate of r over [0, T].

The stochastic process employed to represent the time variation of stock prices is often referred to as geometric Brownian motion or exponential Brownian motion where the logarithm of the random variable follows a Brownian motion process. It is commonly used as an approximation of stock price dynamics. µ is termed the percentage drift and σ the percentage volatility, and both values are assumed constant.

An important implication of assuming geometric Brownian motion is the lognormality of the random value, which means that the probability distribution of the variable is skewed and cannot take on negative values. This is appropriate because negative asset values will result in an option not being exercised as the value of the option is zero. Thus, and again essentially by definition, options are only exercised when they are “in the money.” Otherwise, they are discarded.

A security that is derivative to S is one where its price is a function of S and t, given by f(S, t), and can be obtained by solving the Black–Scholes equation (Black and Scholes, 1973)

(9.3)

The solution of this equation for European call options—which is the designation used for options that can only be exercised at time T—is given by the Black–Scholes call option formula as follows:

(9.4)

(9.5)

(9.6)

where S is the net present value (NPV) of the asset of interest, K is the NPV of the option exercise price (OEP), and N(⋅) is the cumulative distribution function of the standard normal distribution. The NPVs noted require a discount rate that reflects the investor’s cost of capital.

To provide another perspective on valuation of options, consider the following formulation by Smithson (1998). The OV equals the discounted expected value of the asset (EVA) at maturity, conditional on this value at maturity exceeding the OEP, minus the discounted OEP, all times the probability that, at maturity, the asset value is greater than the OEP. Net option value (NOV) equals the option value calculated in this manner minus the discounted option purchase price (OPP). In equation form, OV is thus given by

(9.7)

As noted, the Black–Scholes solution is for European call options. What if the investment problem at hand does not fit this formulation? Different formulations may require computational rather than analytic solutions. Common computational methods include Monte Carlo simulation, finite difference methods, and binomial and trinomial lattices. See Dixit and Pindyck (1994), Trigeorgis (1996), and Luenberger (1997) for treatments of these methods.

9.2.1 Example—Transforming Shipbuilding

Assume that the U.S. Navy would like to transform the way it acquires ships and, therefore, proposes several changes that will streamline the development and design process and reduce rework. Thus, the Navy has the option to transform its ship acquisition enterprise. To determine whether or not the Navy should initiate transformation, an option model was developed (Pennock et al., 2007).

To mitigate technical risks of unsuccessful transformation, it was assumed that there would be a three-stage process:

• Stage 1: Concept development and feasibility analysis. This stage is relatively short and inexpensive. If the transformation idea proves to be infeasible in this stage, the Navy can terminate the project at no additional cost.
• Stage 2: Pilot testing the changes on the acquisition of a single ship. If the project fails in this stage, rework costs will be required to rectify the situation and complete the acquisition of the ship.
• Stage 3: Implementing the transformation across the whole shipbuilding enterprise. If the transformation fails in this stage, a substantial cost in rework is incurred.

Table 9.1 summarizes the staging parameter values for this example.

TABLE 9.1. Stage Parameter Values

Note that as a three-stage model, the solution can only be approximated using the Black–Scholes approach. Thus, Pennock employed the binomial lattice method. A binomial lattice is a large decision tree that represents the decision problem as a series of stages in time. At each stage, the asset price can either increase or decrease by a fixed amount, somewhat larger on the upside than the downside due to upward drift of the assumed geometric Brownian motion. The full lattice extends from t = 0 to T, the expiration time of the option. Once the full tree is formulated, one works backwards from t = T to 0, using backward induction to determine the best decision at each stage.

Using the binary lattice model developed, Pennock found the NOV of this transformation option to be approximately $0.61 billion. If one were to calculate the traditional NPV when considering this technical risk, they would find that the value of the transformation project is approximately −$6.43 billion. This means that one would expect to incur a substantial loss by initiating this project. Here one can see the discrepancy between the NOV and the NPV. The NPV is too conservative because it fails to account for the risk mitigation inherent in staging. So, in this example, a decision maker using NPV as the decision criterion would reject a potentially beneficial program.

The example can be expanded by introducing increased market risk, that is, allowing for uncertainty in cash flows. Option values will inherently increase because options will be exercised only if the upside occurs. If the downside occurs, options will simply not be exercised. The resulting NOV is $5.94 billion, a value that is almost 10 times greater than that without the market risk. Hence, risk can be quite valuable if one can take advantage of the upside while also avoiding the downside.

9.3 OPTION CALCULATOR

Table 9.2 illustrates a Black–Scholes option calculator programmed in Microsoft Excel. The leftmost column includes labels for the rows. The next column (to the right) includes the model parameters and all the Black–Scholes calculations. All of the rows to the right of this column provide the inputs to the NPV and Black–Scholes calculations. Note that Equation 9.4 is employed using S equals the asset value NPV and K equals the option exercise NPV. The parameters for Equations 9.4 to 9.6 are shown above the NPV results.

TABLE 9.2. Black–Scholes Calculator

The example in Table 9.2 employs the data from Table 8.3 for the R&D investment case study. Whereas the NPV in Table 8.3 was $435 million, the NOV in Table 9.2 is $694 million. The difference of $259 million represents the value of being able to hedge the downside risk that either the R&D is not successful or the projected market demand for the new offerings does not emerge as projected.

As shown in Fig. 9.2, NOV is quite sensitive to volatility and discount rate. Increasing volatility leads to increasing NOV. It may seem counterintuitive for value to increase with uncertainty. However, increased volatility results in greater upside potential for an option. Of most importance here, the downside does not matter because, if it occurs, one need not exercise the option. In other words, the option hedges the downside risk while preserving the possibility of upside gain.

Increasing the discount rate, used to calculate S and K in Equations 9.4 and 9.5, leads to decreasing NOV. This should not be surprising as the discount rate has the same effect on NPV. As shown in Fig. 9.2, this effect is quite strong. Thus, it is important that one employs a discount that appropriately reflects the cost of capital. At the same time, one should not increase the discount rate to reflect the uncertainty represented by the volatility. In this way, option models provide us with a better way to address uncertainty than is possible with traditional discounted cash flow models.

9.3.1 Monte Carlo Analysis

It is quite common for people to feel uneasy about their projections of option purchase costs, option exercise costs, and free cash flows in terms of the difference between revenues and operating costs. We need to better understand the impact of these uncertainties.

This understanding can be gained by representing input data as probability distributions rather than point estimates. These distributions can then be sampled, and each sample can be used to compute NOV. The results of each computation can be compiled into a resulting probability distribution for NOV. Typically, 1000 samples are used in this compilation.

This approach is termed Monte Carlo simulation or analysis because of its use of random number generators to create the samples of the input data for each calculation. Table 9.3 shows a spreadsheet-based option calculator with Monte Carlo analysis capabilities. The 1000 samples of the Monte Carlo simulation occur in a second worksheet (not shown) using standard functions within Microsoft Excel.

TABLE 9.3. Option Calculator with Monte Carlo

As inputs to the Monte Carlo simulation or analysis, one chooses the mean and standard deviation of the probability distributions for option purchase, option exercise, and asset value. These are expressed as percentages. The mean and standard deviation percentages define a normal distribution from which random samples are drawn. These samples are then multiplied by the appropriate NPVs (i.e., option purchase, option exercise, or asset value) to define the inputs to the Black–Scholes calculation, along with the other parameters from the upper left of the spreadsheet.

Note that use of the normal distribution for this type of Monte Carlo is reasonable for small to moderate variations of parameters. For large variations, for example standard deviation of 30% to 50% or more, the normal distribution presents problems in that negative percentages can result, which, of course, are meaningless. If such large variations are of interest, one should employ something like an exponential or lognormal sampling distribution.

The approach to Monte Carlo analysis just described involves varying the three NPVs that are input to the Black–Scholes calculation. One could instead vary each of the point estimates toward the top of the spreadsheet, or perhaps vary the finer-grained point estimates that served as inputs to these estimates. Monte Carlo analysis can be pursued at any level of detail one chooses. This can require, of course, much more detailed modeling of uncertainties in input parameters.

The bottom of Table 9.3 shows the probability distribution for NOV on the left, with the cumulative distribution on the right. These histograms reflect means of 100% and standard deviations of 10% for all three NPVs. Choosing means of 100% results in the mean values being the same as the above point estimates. Thus, the average NOV calculated from the Monte Carlo results is almost exactly the same as that calculated without random variations, that is, $692 versus $694. However, the 10% standard deviations yield an NOV standard deviation of $122. Thus, small component variations can add up to very substantial overall variations.

Figure 9.3 portrays the sensitivity of NOV to percentage means that are greater than 100%. Increasing the option purchase NPV decreases the NOV, as one would expect; whereas increasing the option exercise NPV has much less effect. Increasing the asset value NPV has a dramatic effect, again, as one would expect. This suggests that the investment analyst should pay careful attention to projections of the asset value.

Figure 9.4 portrays the sensitivity of NOV to percentage means less than 100%. Decreasing the option purchase NPV increases NOV, as one would expect; decreasing the option exercise NPV has much less effect. Decreasing the asset value NPV has a dramatic effect, again, as one would expect. This adds emphasis to the suggestion that the investment analysis pay careful attention to the projections of the asset value.

It is very important to note that the effects of varying the NPVs in this way are heavily dependent on the particular parameters of the overall model. Thus, one should not assume that the specific sensitivities portrayed in Figs. 9.3 and 9.4 will hold for other problem formulations. Therefore, one should pursue such analyses for each investment problem independent of previous analyses.

This discussion thus far illustrates how straightforward it can be to formulate option pricing models and create associated option calculators. What is less straightforward is the appropriate framing of options. The case studies presented later in this chapter illustrate the range of types of investments that can be formulated as option investments. Later discussion presents some guidelines for framing of options.

9.4 STRATEGIC METRICS

In Chapter 8, NPV and related metrics were discussed. This chapter has discussed NOV—option value minus the cost of purchasing the option. Which metric should we use? In situations where the lion’s share of the overall investment occurs up front, NOV will approach NPV. In this case, we should use the simpler metric.

In other situations, one might use both metrics. Boer (1998, 1999) has argued for a composite metric, such as strategic value (SV), given by

(9.8)

(9.9)

To avoid double counting, the NPV should be based on the financials of the existing lines of business while NOV should reflect new lines of business that may, or may not, be entered in the future based on available options.

To illustrate, a large corporation had a small subsidiary in the wireless LAN business in the commercial sector. They were concerned with whether they should sell this business because the NPV of the financials for their existing lines of business equaled $3 million—pretty small potatoes for a $20-billion corporation.

Using the methods and tools discussed in this chapter, it was estimated that the NOV associated with the possibility of entering the consumer market for wireless LAN equaled $300 million. Thus the SV for this subsidiary was $303 million. This insight resulted in the corporation retaining this small subsidiary due to the attractive SV.

Consider the efficiency case study investment summarized in Table 9.4. In this example, the firm invests $300 million in years 1 and 2 to perform the engineering and acquire the equipment and facilities to enable substantial efficiency increases. These capabilities are deployed in year 3 and lead to increasing profitability in the subsequent years. The bottom of Table 9.4 shows the metrics from Chapter 8 as follows:

TABLE 9.4. Example Efficiency Investment

• NPV is a modest $55 million, with an IRR of 12%.
• CAPM is the same as for the R&D investment, 25%.
• CFROI is $1000 million due to the assumed ability to monetize assets.
• EVA is −$172 million due to the asset-intensive nature of this investment.

There is no option for this investment as the $600-million commitment in years 1 and 2 represents all of the investment needed. There is no downstream investment needed to deploy the capabilities.

In contrast, revisiting Table 9.2, this R&D investment represents a classic real option using the data from Table 8.3. As indicated earlier, the staging of this investment results in an additional $259 million in economic value. Staging of the efficiency investment might have increased its valuation, although such staging may not have made sense technically. For example, if the technologies and methodologies involved are quite mature, then there is little chance that the engineering stage will fail. Further, if the goal is to make a greater margin on offerings to existing customers, due to greater efficiencies rather than increased prices, then the market risks may be quite small.

Table 9.5 compares the results for the two case studies using the three strategic metrics—NPV, NOV, and SV. The efficiency case study is best addressed with NPV. In contrast, addressing the R&D case study with NPV results in undervaluation by $259 million. Thus, NOV is the metric of choice for this case study. The SV of this portfolio of two investments is $749 million, equaling the NPV for the efficiency investment plus the NOV for the R&D investment.

TABLE 9.5. Comparison of Methods

This analysis shows that both investments are economically justified. However, it is quite possible that the firm may not have the economic resources, or human resources, to pursue both investments. The choice, then, is between investing in getting better at what they are already doing and investing in doing new things. As elaborated in Chapter 8, each of these choices has associated uncertainties and risks. The specific nature of these uncertainties and risks is likely to dictate the final investment decision.

However, another perspective can help to resolve this. The firm could decide to proceed with the efficiency initiative while also proceeding to buy the option on the new offerings embodied in the R&D initiative. In other words, the firm need not at this point decide to proceed with the new market offerings. It only needs to assure that it has the capability to pursue these new offerings if they still make sense two years from now. Of course, this is the essence of option-based strategy and shows why the real options approach is so valuable.

9.5 INVESTMENT DECISION MAKING

This section focuses on two issues. First, the need to consider potential investments in the context of the portfolio of investments is discussed. The goal is to balance returns and risks across the portfolio of candidates. Second, this section focuses on 14 case studies of actual investment decisions where these decisions were made using the models and methods elaborated in this chapter.

9.5.1 Investment Portfolios

Typically, investment decisions are not made in the isolation of a single investment opportunity. Such decisions are usually made in the context of several alternatives. The decisions of importance concern which alternatives receive investments and which do not. This involves considering the portfolio of alternatives.

The portfolio of technology options can be portrayed as shown in Fig. 9.5. Return is expressed by NPV or NOV. The former is used for those investments where the lion’s share of the commitment occurs upstream and subsequent downstream “exercise” decisions involve small amounts compared to the upstream investments. NPV calculations are close enough in those cases.

Risk (or confidence) is expressed as the probability that returns are below (risk) or above (confidence) some desired level—zero being the common choice. Assessment of these metrics requires estimation of the probability distribution of returns, not just expected values. In some situations, this distribution can be derived analytically, but more often Monte Carlo analysis or equivalent is used to generate the needed measures.

The line connecting several of the potential investments (PA, PB, PH, and PZ) in Fig. 9.5 is termed the “efficient frontier.” Each potential investment on the efficient frontier is such that no other potential investment dominates it in terms of both return and confidence. In contrast, candidates interior (below and/or left) to the efficient frontier are all dominated by other candidates in terms of both metrics. Ideally, from an economic perspective at least, the candidates in which one chooses to invest—purchase options—should lie on the efficient frontier. Choices from the interior are usually justified by other, typically noneconomic attributes.

A primary purpose of a portfolio is risk diversification. Some investments will likely yield returns below their expected values, but it is very unlikely that all of them will—unless, of course, the underlying risks are correlated. For example, if the success of all the potential R&D investments depends on a common scientific breakthrough, then despite a large number of investments, risk has not been diversified. Thus, one usually designs investment portfolios to avoid correlated risks.

While this makes sense, it is not always feasible—or desirable—for R&D investments. Often multiple investments are made because of potential synergies among these investments in terms of technologies, markets, people, etc. Such synergies can be quite beneficial, but must be balanced against the likely correlated risks.

Further, it is essential to recognize that options are not like certificates that are issued on purchase. Considerable work is needed once “purchase” decisions are made. Options often emerge piecemeal and with varying grain size. Significant integration of the pieces may be needed before the value on which the investment decisions were based is actually available and viable.

9.5.2 Investment Case Studies

Table 9.6 summarizes the key metrics for 14 case studies of real investment decision making from a wide range of domains. These examples represent a total of $4.2 billion of NOV. Put differently, these enterprises—in both private and public sectors—made investments that “bought” options worth $4.2 billion more than they had to invest to gain these options.

TABLE 9.6. Example Options-Based Valuations of Technology Investments

The ways in which these investors bought options fall roughly in four categories. The nature of these categories provides insights into how options can be “purchased” as by-products of investments primarily intended for more near-term purposes.

9.5.2.1 Investing in R&D

In many cases, options are bought by investing in R&D to create “technology options” that may, or may not, be exercised in the future by business units. Ideally, the nature of the technology options of interest will be driven by the future aspirations of the business units. The characteristics and competitiveness of these aspirations are typically laced with uncertainties. These uncertainties are the basis for needing options, usually more options than will later be exercised as the business needs to hedge against alternative future market situations.

9.5.2.2 Running the Business

These examples are situations where the option was bought by continuing to operate an existing line of business, perhaps with modest current profitability but substantial future opportunities. Running the business, which may have resulted from exercising an earlier option, can lead, in effect, to purchasing an option for what the business can become in the future. Indeed, as the wireless LAN example illustrates, the NOV of the potential future business may be much larger than the NPV of the current business. Thus, SV may be dominated by NOV and may justify sustaining a small NPV business.

9.5.2.3 Acquiring Capacity

These examples involve the possibility of acquiring capacity to support market growth that is uncertain in terms of timing, magnitude, and potential profitability. Building or acquiring capacity that is eventually not needed could be an enormous waste of scarce resources. Purchasing an option on capacity may be a much better investment than actually investing in acquiring capacity, despite the fact that the option may not be exercised. The option may take the form of a contingent contract with an owner of capacity for the use of the capacity if needed. Another form could be a contingent contract to purchase capacity, perhaps at a price that includes a discount for the amount paid for the original option.

9.5.2.4 Acquiring Competitor

These examples involve the possibility of acquiring competitors to support growth in current and adjacent markets that is uncertain in terms of timing, magnitude, and potential profitability. Such acquisitions can provide market offerings, capacity, and customers. Interestingly, acquisitions can also provide new technology options. Thus, it is quite possible that purchasing this type of option may provide one or more instances of the other three types of options.

9.5.3 Interpretation of Results

Consider the example of semiconductor memory in the second row from the bottom of Table 9.6. For $109 million of R&D, this company “purchased” an option to deploy this technology in its markets four years later for an expected investment of approximately $1.7 billion. The expected profit was roughly $3.5 billion. The NOV of over $0.5 billion reflects the fact that they bought this option for much less than it was worth.

The option value of over $600 million (i.e., the R&D investment plus the NOV) represents roughly one-third of the net present difference between the expected profit from exercising the option and the investment required to exercise it (i.e., 600/(3500 − 1700)). This is due to considerable uncertainties in the >10-year time period when most of the profits would accrue.

In the second row from the top of Table 9.6, a government agency invested $420 million in R&D to “purchase” an option on unmanned air vehicle technology that, when deployed 10 years later for $72 million, would yield roughly $750 million of operating savings when compared to manned aircraft providing the same mission effects. The NOV of $137 million represents the value of this option in excess of what they invested.

The option value of roughly $560 million (i.e., the R&D investment plus the NOV) represents over two-thirds (i.e., 560/(750 − 72)) of the net present difference between the expected cost savings from exercising the option and the investment required to exercise it, despite the returns occurring in a similar >10-year time frame. Why is this ratio 1/3 for the semiconductor memory investment but 2/3 for the unmanned air vehicle technology investment?

The answer may be obtained by examining the quotient of expected profit (or cost savings) divided by the investment required to exercise the option. This ratio is quite different for these two examples. This quotient is roughly 2.0 (i.e., 3500/1700) for the semiconductor memory option and 10.0 (i.e., 750/72) for unmanned air vehicle technology investment. Thus, the likelihood of the option being “in the money” is significantly higher for the latter. This is why the option value is one-third of the net present difference for semiconductor memory and two-thirds for unmanned air vehicle technology.

9.6 VALUE-CENTERED R&D

Thinking in terms of options can have an enormous impact on how an enterprise approaches and formulates its overall strategy. This is particularly true for research and development. Listed below are 10 principles for value-centered R&D (Rouse and Boff, 2004). Several, but not all, of these principles relate directly to the real option model, methods, and tools discussed thus far in this chapter.

Principles 1 to 3 focus on characterizing value. These principles argue that the overarching purpose of the R&D function is to provide the right portfolio of technology options. Ideally, these options will both be aligned with business unit aspirations and have NOVs that far exceed the enterprise’s R&D budget.

Principles 4 and 5 concern assessing value. This involves understanding how the organization creates and deploys value. Identification of value streams and the networks they form often results in streamlining these streams and networks. The assessed values of the options flowing through these streams and networks can help to decide where the value is added—or not added—and how it can be enhanced.

Principles 6 to 10 emphasize managing value. The technical constructs of value and options are necessary for success, but not sufficient. Governance, structure, affiliation, champions, and incentives and rewards are key elements of success with value-centered R&D. Unfortunately, elaboration of these elements of success is beyond the scope of this book.

9.7 SUMMARY

This chapter has shown that it is pretty straightforward to formulate option pricing models and create associated option calculators. What is not straightforward is framing the options appropriately. There are some characteristics of well-framed options. First of all, “purchasing” an option typically involves a relatively small investment now to purchase the right to “exercise” the option later if the conditions at that time make sense. Exercising an option usually requires a significantly larger investment than purchasing the option. One purchases an option to hedge against the downside possibility that one may not choose to exercise the option at a later point when the true nature of the market opportunity is much clearer. Thus, one is making a two-stage (or multistage) investment decision. In contrast, if the lion’s share, or perhaps all, of the investment is up front, one is making a single-stage investment and the real options construct in this chapter will not yield economic valuations much different than traditional financial analyses discussed in Chapter 8.

Why would one make an up-front, albeit relatively small, investment in an option that one might not exercise? The primary motivation is to avoid making a very large investment in an attractive future that may not materialize. Perhaps people will not want to avail themselves of the functionality possible with the technology in which one has invested. One buys an option because one wants the right to the future if it emerges, but one does not want to commit to it now. Further, because options require much smaller initial investments, one can bet on several alternative futures rather than just one or two. This makes it much more likely that at least one of the bets will have been right.

A second characteristic of a well-framed option is that one can see how a small investment now to purchase an option can gain the right to later exercise the option and secure a large cash flow, large being relative to the costs of purchasing and exercising the option. Typically, this large cash flow will be a new cash flow because other firms usually already own existing large cash flows. Of course, competitors’ existing cash flows may be replaced by their securing the new cash flows. Consequently, one needs to make sure that one has some reasonable chance of winning this competition.

A third characteristic of interest is situations where existing owners of cash flows—the incumbents—are unlikely to be dominant competitors in the future. For example, incumbents may be unlikely to afford the scale necessary, or may not have the agility to compete in fast cycle time markets. Of course, there is also the possibility that the emerging markets will be so new that there will be no incumbents. This still raises the questions of why one will compete successfully.

With these three characteristics of well-formed options in mind, one needs to think about capabilities and markets that will yield the desired large cash flows. One approach to this is to start with particular ideas, capabilities, or technologies and ask where and how they might be transformed from being inventions to being market innovations. An alternative approach is to work backwards. What capabilities will people expect and be willing to pay for, even if they are not sure how to create these capabilities? If the current market vision becomes a reality, what capabilities are consumers and providers going to want provided in some way?

Next, one needs to think about what one would need to do to gain the option to provide these capabilities in the future. Perhaps R&D would be needed. One or more alliances might be the key. Possibly, implementing a new technology in an existing line of business would provide the competency to employ this technology in a new line of business in the future. Overall, the question is what relatively small investment one should make to purchase an option that will enable one to later make a much larger investment if market opportunities have evolved in a manner that makes exercising this option attractive.

Thinking in terms of real options not only results in different economic assessments, that is, different numbers, but can also lead to new ways of thinking about one’s enterprise. Companies with large portfolios of viable and attractive options have considerable leverage to take advantage of the future regardless of the substantial uncertainties and risks associated with the future. Their options provide the means to success no matter how the contingencies play out.

BIBLIOGRAPHY AND REFERENCES

Amram M, Kulatilaka N. Real options: managing strategic investment in an uncertain world. Boston: Harvard Business School Press; 1999.

Black F, Scholes M. The pricing of options and corporate liabilities. J Political Econ 1973;81:637–659.

Boer FP. Traps, pitfalls, and snares in the valuation of technology. Res Technol Manag 1998;41(5):45–54.

Boer FP. The valuation of technology: business and financial issues in R&D. New York: Wiley; 1999.

Dixit AK, Pindyck RS. Investment under uncertainty. Princeton, NJ: Princeton University Press; 1994.

Luenberger DG. Investment science. Oxford, UK: Oxford University Press; 1997.

Merton RC. Theory of rational option pricing. Bell J Econ Manag Sci 1973;4(1):141–183.

Pennock MJ, Rouse WB, Kollar DL. Transforming the acquisition enterprise: a framework for analysis and a case study of ship acquisition. Syst Eng 2007;10(2):99–117.

Rouse WB, Boff KR. Value-centered R&D organizations: ten principles for characterizing, assessing & managing value. Syst Eng 2004;7(2):167–185.

Rouse WB, Howard CW, Carns WE, Prendergast EJ. Technology investment advisor: an options-based approach to technology strategy. Inf Knowl Syst Manag 2000;2(1):63–81.

Smithson CW. Managing financial risk: a guide to derivative products, financial engineering, and value maximization. New York: McGraw-Hill; 1998.

Trigeorgis L. Real options: managerial flexibility and strategy in resource allocation. Cambridge, MA: MIT Press; 1996.

1Robert Merton and Myron Scholes won the Nobel Prize in Economic Sciences for this work in 1997. Fischer Black, who died in 1995, was not eligible to be included but was noted in the Nobel citation.