Valuation of intellectual property
A real option approach
Jow-Ran Chang
Department of Quantitative Finance, National Tsing Hua University, Taiwan
Mao-Wei Hung and Feng-Tse Tsai
College of Management, National Taiwan University, Taiwan
Abstract
Purpose – This paper aims to provide a new approach to evaluate intellectual property (IP) and uses a cautious view of how volatility impacts the economic value of IPs.
Design/methodology/approach – Real option is a useful tool for valuing investments under uncertainty and if it is applied to the valuation of IP with some modifications, it is also widely accepted. However, it is still debatable whether there is a constant rate-of-return. This paper incorporates a sensitivity variable to account for the volatility of the expected rate of return. Thus, rate-of-return can be a constant or increase with volatility.
Findings – First, it was found in the simple model that Vega may be negative when the option is deep in the money. Second, in the general model, the option can be seen as a sequence of options and under the constant rate-of-return shortfall setting, it resembles traditional financial options with positive Vega.
Originality/value – The scenario set-up allows the authors to explain why uncertainties of future cash flows drive firms to invest now instead of later.
Keywords Knowledge economy, Intellectual property, Intangible assets, Rate of return Paper type Research paper
Introduction
The network economy accelerates the acquirement of knowledge while the knowledge economy promotes the accumulation of knowledge. Knowing the value of knowledge-based assets is essential in order to evaluate a firm accurately. Knowledge-based assets are increasingly significant, especially for high-technology companies. Evidence shows that these firms spend a lot on research and development (R&D) and heavily emphasize intellectual property (IP) rights and patents. IPs provide firms with a wide array of growth opportunities and competitive edge.
IP refers to know-how, patents, copyrights, trademarks or designs, trade secrets, etc. IP is usually inseparable from the enterprise and can be traded in mergers and acquisitions (M&A) or transferred in bankruptcy situations. Besides, intellectual capital is a crucial asset in order for a company to remain competitive. For example, patents can legally exclude potential entrants from manufacturing and selling the company’s patented products. Therefore, investment projects in IP or trade and transfers of IP not only have to take future cash inflows into discretionary consideration, but also strategic decision making. Moreover, firms need to devote themselves to putting most of their IPs under the protection of law. Thus, intellectual property right” is a hot issue.
In the knowledge economy, more and more IPs follow hard to the heel of new technologies and valuation of IP becomes a critical issue. Managers have to tell their
The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at www.emeraldinsight.com/researchregister www.emeraldinsight.com/1469-1930.htm
Valuation of
intellectual
property
339
Journal of Intellectual Capital Vol. 6 No. 3, 2005 pp. 339-356
q Emerald Group Publishing Limited
1469-1930 DOI 10.1108/14691930510611094
bosses the value of the project for new technology and negotiators must show the fair market value of the IP if they are to deal with a technology transfer project in an exchange market. Besides, accountants need to enlist the true value of these intangible assets in the balance sheet even though they have no market value for such assets. Due to the intangibility of intellectual capital, it is hard to assess the “hidden” value of these assets. Although some valuation approaches can approximately measure the assets’ value, they usually leave out their latent value.
In traditional financial worlds, the book value of a firm mainly consists of tangible assets. However, in the wake of placing more attention on intangible assets such as IP, many researchers have been trying to find the fair value for these assets. In most books, the income approach, cost approach, market approach, discount cash flow (DCF) and real option method (ROM), which may be combined with game models are mentioned. All of them are adopted in reality, but none of them are satisfactory. However, the ROM surpasses others in considering uncertain situations in the future with well-defined models.
Real options are derived from financial options and are applied to pricing investment projects. The value of an investment project depends on its investment opportunities in the future and this opportunity is an option. Unlike financial options, real options must satisfy the characteristics of a project such as flexibility, irreversibility and infinite life. Flexibility means firms have the right, but not the obligation, to “exercise” the project. This is also known as the call-type option. Moreover, firms have the right but not the obligation to “abandon” the project and this is equivalent to a put-type option. Irreversibility means that when firms decide to invest, they lose future investment opportunities and pay a sunk cost. This concept is similar to exercising an option with a “strike price”. An interesting aspect of real option is that it has infinite life without “maturity”. Therefore, we have to modify the option-pricing model in certain ways. Real options hold most of the properties inherent in financial options, thus we can easily follow an option-pricing method to construct the valuation process.
Furthermore, we are more interested in the amount invested in R&D or intellectual capital investments when the future market is uncertain. If we use ROM and take investment opportunity as a financial option, higher uncertainty will heighten the option value and inhibit the investment. Following this, firms will eventually reduce R&D or intellectual capital outlays under an uncertain world. However, in the real world, we find that many firms do not cut their investments on intellectual capital, but instead raise their expenses on these investments. The rationale behind this phenomenon is that R&D investments may bring more cash flows when future price is more volatile, and can easily cover those initial outlay costs. Some firms increase investment for strategic concerns because they want to seize future markets. And some firms increase investments in uncertain circumstances as a bet for their profit targets. ROM
The option pricing theory (Black and Scholes, 1973; Merton, 1974) is a significant contribution to asset pricing theories. Binomial tree model proposed by Cox, Ross, and Rubinstein (CRR) is an auxiliary difference method in option valuation. CRR explains the structure of an option better than a formula and converges well with the Black and Scholes (1973) formula. ROM extends its applications to capital budgeting for valuing
JIC
6,3
investment projects as well as other intangible assets. Trigeorgis (1993) furthermore dealt with real options by allowing for debt financing which relaxes the assumption of an all-equity firm. The value of the project can be improved by this additional financial flexibility (i.e. equity holders have the option to default on debt payment for limited liability). Reifer (2003) introduces real option framework that permits software experts to value trade secrets when they face litigations. Differing from traditional NPV, ROM can capture the flexibility characteristic in investment projects and explain some undervalue situations[1] in traditional NPV approaches. Pengfei and Yimin (2000) introduce real option valuation in high-tech firms. Bouteiller (2001) suggests an extended framework for valuation of intangibles assets and intellectual capital. Faulkner (1996) applies ROM instead of DCF to R&D valuation and strategy and suggests that ROM can identify important sources of value missed by the traditional DCF. He points out that ROM is as vulnerable as DCF when uncertainty disappears, meaning future opportunities are also wiped away. Herath and Park (1999) evaluate R&D project with an options approach and indicate how valuations could be linked to a company’s stock price. Iversen and Kaloudis (2003) briefly survey some intangible valuation approaches and partly focus on patent valuation. They explore the relationship between intangibles valuation and innovation process.
However, ROM also has its limitations when applied to the real world. These difficulties are as follows:
. The estimation of volatility is difficult in practice (Sudarsanam et al., 2003) since
the underlying investment project or IP is not frequently traded in the secondary market and this results in their prices lacking the reliability of the market prices. Sometimes some trade is “under-table” and hence the trading prices are hard to acquire. One substitute solution is to find a security that is perfectly correlated with the underlying project. But this again brings about other problem. One example is the accuracy of ROM relying on the efficient capital market assumptions. Under these assumptions, substitute securities cannot play a perfect role in the underlying project. Pitkethly (1997) mentions that the variance of future returns would be different within the life of the patent since the survival of the patent will result in the company being more successful or profitable.
. There exists an inexact mapping of the assumptions or inputs between option
pricing theory and real option application. For example, does the growth rate of patent cash flows follow log normal distribution? Is it reasonable to set the cost of construction to be a constant or to relax it following a random process? Some of the difficulties can be solved in our study using the Monte Carlo simulation.
. Patents contain “adverse rights” which run counter to the notion of “having an
option”. Furthermore, the option value of a patent can be reduced or eliminated by a third party filing and contesting the claim.
Real options allow for flexibility to exist in investment decisions. They provide options to defer, time to build, alter operating scales, abandon, switch, grow, etc. Trigeorgis (1993) described how these real options are important in applications. Despite the problem of trading in the open market, real options are valued similarly to financial options. However, in order to solve the trading problem, it is possible to construct a dynamic portfolio by trading a “twin security” with perfectly correlated underlying asset (Pindyck, 1991). Another problem caused by the “not traded” aspect of the real
Valuation of
intellectual
property
341
options is that its growth rate may fall below the twin security’s equilibrium of expected rate of return. Thus the difference between them is seen as the “rate of return shortfall” in an investment project. If the underlying assets are commodities, it is also called “convenience yield”. Hence a dividend-like option pricing should be applied (McDonald and Siegel, 1984, 1985).
However this dividend-like rate is not as simple as the dividend rate in financial options. In spite of that, most authors introduce this shortfall rate[2] in the same way as the constant dividend rate which requires the use of a market equilibrium model (McDonald and Siegel 1985). Dixit and Pindyck (1994) state that “However, in practice, the convenience yield can vary (stochastically over time and/or in response to market-wide variables such as total storage)” Smit and Ankum (1993)[3] use the real option method and the game-theory approach to value an investment project under competition. They discover that the option to defer under competition may be eroded because the market may compel firms to invest earlier. On the other hand, competitive environments enhance the rate of return shortfall and further eat away the flexibility value. Miltersen (2000) applied Miltersen and Schwartz’s (1998) model to value natural resource investment under stochastic interest rates and convenience yields. He found that stochastic convenience yields would affect the value of the project. Sarkar (2000) illustrated that the idea of a negative uncertainty-investment relationship is not always correct and, in certain situations, an increase in uncertainty can have a positive impact on investment. Moreover, Davis (2002) discussed the impact of volatility on growth options[4]. He addressed that the rate of return of the underlying asset be a constant and the rate of return shortfall be a variable that varies with volatility. Hence he concluded that increasing volatility could damage the value of at or in the money growth options. The finding is insightful for explaining several specific industries and counter-intuitive in explaining the relationship between option value and volatility. Therefore, volatility will not just promote the value of options, but also the indirect trade-off effect of volatility on dividend-like rate will be destroyed.
In pricing the economic value of intellectual assets, it is important to estimate the volatility of the output price. As we know in financial options, increasing volatility will raise the value of options. Therefore, sellers of IP would manipulate the estimation of volatility to enhance the value, while buyers would tend to underestimate the volatility for a cheaper price. However, if the rate-of-return shortfall factor is taken into consideration, and correctly calculated for its relationship with volatility, then the value of IP will not proportionally increase with the volatility of the underlying price. This can help negotiators further and deeply find out not only the fluctuation of future price, but also opportunity costs that need to be taken into account.
The relationship between option value and volatility
Discussions about the relationship between option value and volatility are as follows:
. Positive relationship: most scholars hold this position including Trigeorgis
(1993), McDonald and Siegel (1985) and Pindyck (1991).
. Negative relationship: Jagannathan (1984), McDonald and Siegel (1984),
Kulatilaka and Perotti (1998), Sarkar (2000) and Davis (2002).
Here we propose a parameter called gamma to measure the sensitivity of expected rate of return of the underlying asset to output price volatility. Gamma may vary with
JIC
6,3
different kinds of latent factors under different markets or miscellaneous situations. It becomes a proxy for measuring the growth rate of demand affected by the change of the underlying volatility. For instance, when oil’s price volatility increases, gamma measures the extent of the drift of the demand. It may depend on an exponential decreasing economic rent in a competitive market or some specific industry with growth options. Gamma might be a constant or a linear function of volatility.
Our model coordinates all situations regarding the relation between option value and underlying volatility, and resolves many unreasonable assumptions about constant rate of drift demand or invariant rate of return shortfall when volatility changes in order to preserve the original properties of options. Since call restricts the downside loss, but does not restrict the upside potential, high volatility benefits and enhances the value of the option. We break the myth of this monotonic relationship in our paper. Hence, when valuing IP with ROM, manipulating the price of IP by changing the volatility of future cash flow need not be attained anymore.
Davis (2002) also provides that negative impacts of the volatility on option value will more likely happen when it is in or at the money. He challenges most authors by implying that the growth rate of demand will follow higher price volatility by proposing the empirical example of oil exploration. However, he claims that the growth rate of demand is independent of price volatility and sets it as a constant. This assertion is much different from most authors assuming that the level of changes in required rate of return is the same as the level of changes in expected rate of return when price volatility changes. I accommodate two extreme views and argue that the growth rate of demand in real investment project should not adhere to financial assets, since the project is not frequently tradable as financial products are. In some cases, the adjustment to the change of volatility will cause a shortfall in the rate of return. Therefore, we set the relative adjustment growth rate of demand as gamma, and relax the tricky assumption of fixed or synchronous adjustment growth rate. Furthermore, with this coefficient, rate of return shortfall may or may not be a function of volatility. Besides, we proposed that gamma plays a crucial role similar to that of the elasticity of demand and explains situations where constant or variable rate of return shortfall represents the dividend rate in financial options or convenience yield in commodity options. Higher volatility has a direct effect on option value increases but an indirect effect on decreases if the rate of return shortfall increases with volatility. This means that larger change in price seemingly raises the option value, but it also affects the growth rate of demand and indirectly induces more needs for rate of return. However, arriving at a perfect estimation of volatility for non-traded or infrequently traded assets is still an open question. This is also one problem we face when applying ROM to valuing intangible assets. In addition, many intellectual properties are developed in many stages, and the volatilities are different in accordance with the maturity of the product[5]. Herath and Park (2002) take such sequential-investment decisions as compound real options and use Monte Carlo to estimate the volatility parameter[6]. Our research only applies to single-stage models and may be extended to multi-stage models in future research.
Moreover, we apply three models – simple model, finite-horizon model[7] and general model – to discuss the relationship between option value and volatility. Davis (2002) has shown results of the negative relationship in the simple model. However, comparing his conclusion with the relationship in the general model, which allows the
Valuation of
intellectual
property
343
project to be shut down when the price falls below the level of variable cost[8], we find that the negative relationship in the general model is not as serious as it is in the simple model. This finding explains how options with more flexibility are hurt less by the indirect effect than options without flexibility.
Model I. Simple model
First, we propose a simple model in valuing an investment project and apply it to evaluate an IP. Suppose the present value of the IP is V, and the cash flow generated by IP is P[9] (or the output price of the IP) which follows the stochastic process as following:
dP ¼aPdt þsPdz; ð1Þ
wherea is the drift rate of P, sis the standard deviation of P, and dz is a Wiener process.
Assumption (spanning)[10]. Stochastic deviations in P can be spanned by existing assets (Pindyck, 1991).
Under this assumption, we can construct an asset or a dynamic portfolio of assets whose price x is perfectly correlated with P. And x follows the Brownian motion as follows:
dx ¼ mxdt þsxdz ð2Þ
mis the expected return of the asset and it satisfies capital asset pricing model (CAPM) representing m ¼ t þlrVms; where r is the risk-free rate, andlis the market price of
the risk, andrVmis the correlation between x and the value of market portfolio Vm.
It makes sense thata,m(risk-adjusted return). And the difference between them is denoted by d¼m2a (rate-of-return shortfall; opportunity cost of the option waiting for investing).
If the cost to introduce IP (similar to an investment cost) is I, then the net payoff can be written asP¼ V 2 I :
By NPV method, the investment is feasible ifP$ 0:
However, generally speaking, investment is irreversible and the expense will become sunk cost; therefore, the investment plan should function as an option to invest, and by exercising this option, he/she will lose the value to wait[11]. Different from financial options, the life of real options is generally infinite – perpetual options. Thus, we can represent the project value V as the following:
V ¼P
d ð3Þ
(see[12]).
Or P ¼dV indicates the dividend stream, i.e. the cash flow of the project. In other words, V is the function of P. We can write this as V ¼ V ðPÞ.
Let F ¼ FðPÞ be a contingent claim or the value of the option to invest.
Construct a dynamic hedge portfolio F¼ F 2 FPP, where FP ¼›F=›P: The
strategy of this portfolio is to hold option F and short ›F=›P unit of P. The short position will generate cash flow with payout rate dto attract rational investors to
JIC
6,3
hold this portfolio. Total return from this portfolio over infinite short period dt is dF 2 FPdP 2dPFPdt:
The portfolio is riskless during dt and earns the same as risk-free securities. This follows that dF¼ rFdt and by combining the above equations, we have:
dF 2 FPdP 2dPFPdt ¼ rðF 2 FPPÞdt: ð41Þ
By Ito’s lemma, we can write:
dF ¼ FPdP þ
1
2FPPðdPÞ
2 ð5Þ
ðdPÞ2¼s2P2dt:
Substituting these into the above equations, we have the differential equation with boundary conditions as follows:
1 2s 2P2F PPþ ðr 2dÞPFP2 rF ¼ 0 ð6Þ Boundary conditions (BC): (1) Fð0Þ ¼ 0. (2) FðP* Þ ¼ V ðP* Þ 2 I .
(3) FPðP* Þ ¼ VPðP* Þ (smooth pasting condition).
Where P*is the critical value or hurdle value of P satisfying the optimal investment
rule. This means that the best time to invest, (i.e. exercise the option) is when P $ P* . The final solution is as follows[13]:
FðPÞ ¼ aPb ð7Þ b¼1 22 r 2d s2 þ r 2d s2 2 1 2 2 þ2r s2 " #1=2
Using BC (2) and (3), we can solve:
P * ¼ b
b2 1dI ;
a ¼ðP *Þ
12b
bd :
Now we go on to consider the impact of price volatility (or demand uncertainty[14]) on this option value. This will be the comparative static analysis of›F=›s[15].
We have stated above that the rate-of-return shortfall in an investment project will indirectly affect the value of an option. For this reason, we will discuss the shortfall rate and indicate it as a function of volatilitydðsÞ ¼ mðsÞ 2aðsÞ[16].
In financial options,dis usually fixed under case m0ðsÞ ¼a0ðsÞ:
Valuation of
intellectual
property
345
However, in real options, especially growth options, it is still disputable whether to assume thatdis a constant. This is because the constant shortfall rate is obtained via the assumption that the drift (or growth rate) in demand for a firm’s output increases when demand uncertainty is increased, or via the assumption that bothmandaare constants. This is an unreasonable assumption and it is equivocal in oil industries. Evidences show that the rate-of-return shortfall increases when the price of oil has a higher volatility. Therefore, we definea0ðsÞ ¼gas the sensitivity of the drift of output price against volatility, which may change within different industries.
To see the restriction ofg, we recall that m ¼ r þlrVms; and assume that both the
market price of risk and the correlation between output price and market value are independent of price volatility. Then we have m0ðsÞ ¼lrVm:
We have deduced d. 0, so r þlrVms2aðsÞ . 0: Integrating a
0ðsÞ ¼g with
respect tosand dropping the constant term, we haveaðsÞ ¼kþgs:
Hence, the condition will satisfy the inequality relation thatg,lrVmþ ðr 2k=sÞ: . Case 1. m0ðsÞ ¼a0ðsÞ
In the case,d0ðsÞ ¼ 0; anddis a constant.
This is an unique situation under Case 1. m0ðsÞ ¼a0ðsÞ ¼ 0; we assume that
there is no systematic risk for underlying asset and investors are risk-neutral;
m¼ r (risk-free rate). Some authors (McDonald and Siegel, 1984; Teisberg, 1994) use this case and assume that market price of risk decreases when volatility increases.
. Case 2.a0ðsÞ ¼ 0; m0ðsÞ – 0; thend0ðsÞ ¼ m0ðsÞ . 0 (Davis, 2002). . Case 3 (general case). m0ðsÞ $a0ðsÞ,d0ðsÞ $ 0.
Comparative static analysis of›F/›s
. Case 1.›F=›s. 0 (only direct effect without indirect effect fromd). . Case 2.›F=›sis undetermined.
. Case 3.›F=›sis also undetermined.
Incorporating the indirect effect might offset the direct effect that increasing volatility will enhance the option value. We can observe in the equation thatdincreases withs
and enlarges the threshold or the critical value of cashflow P*. As a result, this will
cause the option to become less valuable. But largersalso generates higher probability for cashflow to overstride the threshold value, and for this reason, options will become more valuable.
II. Finite period model
Some IPs such as patents have finite lifetime, hence this finite-period-model is suitable for these kinds of IPs.
Assume that the value of the underlying asset follows Ito process as follows: dP
P ¼adt þsdz: The expected rate of return to hold this asset ism.a.
The payoff at maturity is maxð0; PT2 I Þ.
JIC
6,3
According to the B-S formula: FðP0; I ; tÞ ¼ P0N ðd1Þ 2 I e2rtN ðd2Þ d1¼ lnðP0=i Þ þ ½r þ ðs2=2Þt spffiffit d2¼ d12s ffiffi t p :
However, if we set I ¼ 0 in the above equation, we have FðP; 0; tÞ ¼ P: This implies:1 dt E dFF ¼ 1 dt E dPP
¼a, m, therefore the option earns a rate of return under the market equilibrium rate of return.Now let P * ¼ Pe2ðm2aÞtand by Ito’s lemma, we
have:
dP *
P * ¼ mdt þsdz: Letd¼m2a.
Then B-S formula becomes:
FðP0; I ; tÞ ¼ P0e2dtN ðd*1Þ 2 I e 2rtN ðd* 2Þ d*1 ¼ lnðP0=I Þ þ ½r 2dþ ðs 2=2Þt spffiffit d*2 ¼ d*12spffiffit: Checking again, we find:
1 dtE dF F ¼ 1 dt E dP P ¼a
reached the market equilibrium rate of return.
Financial options use a similar formula with dividend payment. Therefore, higher dividend rate will cause option value to decrease. So we may have a negative relationship between F andsin some region of P ifscan enhanced.
III. General model
If we extend the model to a more realistic one under the assumption (Pindyck, 1991) that:
. The project can be shut down without any cost if P is lower than variable cost c,
and restarted if P is above c.
. The project produces one unit of output in each period.
Valuation of
intellectual
property
347
Now the project is taken as a set of options (McDonald and Siegel, 1985) and its value becomes V ðPÞ ¼ A1Pb1 P , c A2Pb2þPd2cr P $ c 8 < : ; ð9Þ where b1¼ 1 22 r 2d s2 þ r 2d s2 2 1 2 2 þ2r s2 " #1=2 . 1; b2¼ 1 22 r 2d s2 2 r 2d s2 2 1 2 2 þ2r s2 " #1=2 , 0; A1¼ r 2b2ðr 2dÞ rdðb12b2Þ cð12b1Þ; A2¼ r 2b1ðr 2dÞ rdðb12b2Þ cð12b2Þ
and the option value of P is FðPÞ ¼ aPb1 P # P * V ðPÞ 2 I P . P * ( ð10Þ where a ¼b2A2 b1 ðP *Þðb22b1Þþ 1 db1 ðP *Þð12b1Þ
and P*satisfies the equation as following:
A2ðb12b2Þ b1 ðP *Þb2þðb12 1Þ db1 P * 2c r2 I ¼ 0: Numerical analysis
We will use three cases to compare the differences between constant rate-of-return shortfall volatility. The first case is similar to financial options that have constant dividend rate. The simple model functions as well as a financial option when the option is not deep in the money. However, when the option is deep in the money, the option value will decrease as volatility increases. This is because we assume the option is a perpetual option. We can also observe this phenomenon in the numerical analysis below.
JIC
6,3
Case 1. Assume thatd¼ 0:098 is a constant, then the option is at the money when the initial cashflow is at the 9.8 level. We found that when the cashflow is above the 14.3 level, the option value will be negatively correlated with volatility (see Figure 1). After further analysis, we found that the relation between the coefficientband s the coefficient a andsare as shown in Figure 2 and Figure 3.
Figure 1. P ¼ 20, I ¼ 100, r ¼ 0:05, r ¼ 0:8, l ¼ 0:3, g ¼ 0, d ¼ 0:098 Figure 2. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0
Valuation of
intellectual
property
349
The option value is FðPÞ ¼ aPb, therefore, its value is ambiguous under the tradeoff
between a andbwhen volatility increases. Especially when the cashflow is deep in the money, the option with lower volatility has lower a but largerbhence overtakes the option with higher volatility (see Figure 4).
This result is very different from financial options with finite horizon. Financial options always have positive Vega.
Case 2. Assumedincreases with the same velocity asa, (i.e.d0ðsÞ ¼a0ðsÞ . 0). In this case,d¼ 0:098 fors¼ 0:02 andd¼ 0:12 fors¼ 0:03. We see that option values with low volatility surpasses option value with higher volatility when the option is out-of-the-money (around initial cashflow on the level of 7.4). Higher increasing rate of
dwill make option value with larger volatility decrease even more. The extreme case of infinite dividend rate will make option value go to zero (see Figure 5).
Figure 3. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0 Figure 4. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0)
JIC
6,3
350
Case 3. Setg¼ 0:02 (the adjusted rate of expected growth rate of cashflow against volatility changes), i.e.d0ðsÞ 2a0ðsÞ ¼ 0:2: We found that two curves intersect at 8.8, which is around the level of the initial cashflow. Both option values are enhanced here (see Figure 6).
We continue to study the general model under the three cases.
Case 1. We use the third model to satisfy the classical economic theory that firms will shutdown if its cashflow is below its variable cost. We will consider a sequence of
Figure 6. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0:2 Figure 5. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0
Valuation of
intellectual
property
351
real options at different stages to help us find a more realistic option value. In this model, we assume that the variable cost is a constant. The results are different from the previously mentioned simple model and options behave more like traditional financial options with dividend rate (see Figure 7).
Case 2 (see Figure 8). Case 3 (see Figure 9).
Figure 8. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0, c ¼ 5 Figure 7. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0, d ¼ 0:098, c ¼ 5
JIC
6,3
352
Managerial implications
The purpose of discussing the relationship between option value and volatility comes from unveiling conventional views of positive Vega in financial options. A new mindset is required to evaluate intangible assets. Those who try to measure the true value of intangible assets should not be overly-optimistic or overly- pessimistic on the variation of future cash flow. First, they should not rely on subjective anticipation of future changes in cash flow. They have to do more economic analysis on the effect of these changes since the relationship between volatility and option value is not monotonic. Second, they should place more emphasis on increasing managerial flexibility and making the models more general or practical. Lastly, if these assets are easily exchanged or traded, one would have a better prediction of the fair value, and this in turn would raise his or her bargaining power.
Conclusion
We provide a cautious view of how volatility impacts the economic value of intellectual properties. First, we found in our simple model that Vega may be negative when the option is deep-in-the-money. If the rate-of-return shortfall is variable and increases with volatility, option value would have a negative relation with volatility. This condition is prevalent even in out-of-money situations. Second, in the general model, option can be seen as a sequence of options and under the constant rate-of-return shortfall setting, it looks like traditional financial options with positive Vega. And if rate-of-return shortfall changes with volatility, our results can still be that volatility curtails option value in certain ranges.
Advance research will focus on providing more general models suitable for the real world. Investments have to take many situations and factors into consideration when making decisions. In addition, relaxing the constant volatility restriction and applying
Figure 9. P ¼ 20, I ¼ 100, r ¼ 0:05, l ¼ 0:3, r ¼ 0:8, g ¼ 0:2, c ¼ 5
Valuation of
intellectual
property
353
it to an appropriate volatility curve setting can more explicitly explain the relationship between option value and volatility.
Notes
1. Expanded NPV ¼ static NPV of expected cash flows þ value of options from active management (Trigeorgis (1993).
2. Dixit and Pindyck (1994) analogize the dividend rate for the developed reserve is the net production revenue less the rate of depletion; one forgoes this by delaying development. 3. Also refer to Smit and Trigeorgis’ forthcoming article about the use of real options and game
theory.
4. An early investment (such as lease on oil reserves or undeveloped lands, R&D and etc.) will open up growth opportunities in the future.
5. Pitkethly (1997) addresses this problem in applying option pricing as “compounddedness”. 6. Trigeorgis (1996) and Martzoukos and Trigeorgis (2002) also discuss the complex
interactions.
7. This model (OPM with dividends) is to complete the situation for some products with finite life.
8. This corresponds to the situation that no investment under this level in the long run.
9. Here, we omit the suffix of Ptto P.
10. If spanning does not hold, dynamic programming can be used (Pindyck, 1991). 11. The Adjust Net Present Value (ANPV) is NPV with option value in some books.
12. V ¼ S1t¼0½P=ð1 þ dÞt ¼ P=d:
13. The calculation refers to Appendix. 14. We assume firms are price takers.
15. The expression of this differential result is complicated and available by Mathematica. 16. This formula is derived in the second model with finite period.
References
Black, F. and Scholes, M. (1973), “The pricing of options and corporate liabilities”, Journal of Political Economy, Vol. 81 No. 3, pp. 637-54.
Bouteiller, C. (2001), “The evaluation of intangibles: advocating for an option based approach”, working paper, Reims Management School, Reims.
Davis, G.A. (2002), “The impact of volatility on firms holding growth options”, The Engineering Economist, Vol. 47 No. 2, pp. 213-31.
Dixit, A. and Pindyck, R. (1994), Investment under Uncertainty, Princeton University Press, Princeton, NJ.
Faulkner, T.W. (1996), “Applying options thinking to R&D”, Research Technology Management, Vol. 39 No. 3, pp. 50-6.
Herath, H.S.B. and Park, C.S. (1999), “Economic analysis of R&D projects: an options approach”, The Engineering Economist, Vol. 44 No. 1, pp. 1-35.
Herath, H.S.B. and Park, C.S. (2002), “Multi-stage capital investment opportunities as compound real options”, The Engineering Economist, Vol. 47 No. 1, pp. 1-26.
Iversen, E.J. and Kaloudis, A. (2003), IP-Valuation as a Tool to Sustain Innovation, STEP report 17, NIFU STEP, Oslo.
JIC
6,3
Jagannathan, R. (1984), “Call options and the risk of underlying securities”, Journal of Financial Economics, Vol. 13 No. 3, pp. 425-34.
Kulatilaka, N. and Perotti, E.C. (1998), “Strategic growth options”, Management Science, Vol. 44 No. 8, pp. 1021-31.
McDonald, R.L. and Siegel, D.R. (1984), “Option pricing when the underlying asset earns a below-equilibrium rate of return: a note”, The Journal of Finance, Vol. 39 No. 1, pp. 261-5. McDonald, R.L. and Siegel, D.R. (1985), “Investment and the valuation of firms when there is an
option to shut down”, International Economic Review, Vol. 26 No. 2, pp. 331-49.
Martzoukos, S.H. and Trigeorgis, L. (2002), “Real (investment) options with multiple sources of rare events”, European Journal of Operational Research, Vol. 136 No. 3, pp. 696-706. Merton, R.C. (1974), “On the pricing of corporate debt: the risk structure of interest rates”, The
Journal of Finance, Vol. 29 No. 2, pp. 449-70.
Miltersen, K.R. (2000), “Valuation of natural resource investments with stochastic convenience yields and interest rates”, in Brennan, M.J. and Trigeorgis, L. (Eds), Project Flexibility, Agency, and Competition: New Development in the Theory and Application of Real Options, Oxford University Press, New York, NY, pp. 183-204.
Miltersen, K.R. and Schwartz, E. (1998), “Pricing of options on commodity futures with stochastic term structures of convenience yields and interest rates”, Journal of Financial and Quantitative Analysis, Vol. 33 No. 1, pp. 33-59.
Pengfei, H. and Yimin, H. (2000), “Real option valuation in high-tech firm”, Graduate Business School, Go¨teborg.
Pindyck, R.S. (1991), “Irreversibility, uncertainty, and investment”, Journal of Economic Literature, Vol. 29 No. 3, pp. 1110-48.
Pitkethly, R. (1997), “The valuation of patents: a review of patent valuation methods with consideration of option based methods and the potential for further research”, working paper WP21, Judge Institute, Cambridge, UK, pp. 1-30.
Reifer, D.J. (2003), “Use of real options theory to value software trade secrets”, Proceedings of the 2nd International Workshop on Economics-Driven Software Engineering Research, Limerick, Ireland, ACM Press, New York, NY.
Sarkar, S. (2000), “On the investment-uncertainty relationship in a real options model”, Journal of Economic Dynamics & Control, Vol. 24 No. 2, pp. 219-25.
Smit, H.T.J. and Ankum, L.A. (1993), “A real options and game-theoretic approach to corporate investment strategy under competition”, Financial Management, Vol. 22 No. 3, pp. 241-50. Sudarsanam, S., Sorwar, G. and Marr, B. (2003), “Valuation of intellectual capital and real option
models”, PMA Intellectual Capital Symposium, Cranfield University Cranfield, UK. Teisberg, E.O. (1994), “An option valuation analysis of investment choices by a regulated firm”,
Management Science, Vol. 40 No. 4, pp. 535-48.
Trigeorgis, L. (1996), Real Options: Managerial Flexibility and Strategy in Resource Allocation, MIT Press, Cambridge, MA.
Trigeorgis, L. (1993), “Real options and interactions with financial flexibility”, Financial Management, Vol. 22 No. 3, pp. 202-24.
Further reading
Smit, H.T.J. and Trigeorgis, L. (2004), “Real options: examples and principles of valuation and strategy”, in McCahery, J. and Renneboog, L. (Eds), Venture Capital Contracting and the Valuation of High Technology Firms, Oxford University Press, Oxford.
Valuation of
intellectual
property
355
Appendix
Equation (6) is of the Cauchy-Euler type. We can use a solution of the form F(P) ¼ Pb and
substitute it into Equation (6), to derive 1
2s
2bðb 2 1Þ þ ðr 2 dÞb 2 r ¼ 0
The left-hand side of the above equation is negative at b ¼ 0 and b ¼ 1: Hence one of the roots is negative and the other exceeds 1. We write down the general solution as the following:
FðPÞ ¼ a1Pb1þ a2Pb2; b1¼ 1 22 r 2 d s2 þ r 2 d s2 2 1 2 2 þ2r s2 " #1=2 . 1; b2¼ 1 22 r 2 d s2 2 r 2 d s2 2 1 2 2 þ2r s2 " #1=2 , 0: B.C. (1) implies a2¼ 0.B
