Research Policy 35 (2006) 343–354
`
`Application of real options analysis for pharmaceutical R&D
`project valuation—Empirical results from a survey
`∗
`
`, Ali Hassan
`Marcus Hartmann
`Technical University of Berlin, Institute of Chemistry, Research Group Chemical Economics, Germany
`
`Received 1 March 2005; accepted 1 December 2005
`Available online 28 February 2006
`
`Abstract
`
`Real options analysis was often recommended as an emerging valuation technique for high-risk investment projects. Former
`inter-sectoral surveys have drawn an ambivalent picture of real options usage in general. In addition, there is a lack of sector-specific
`investigations. In the following article the results of an in-depth analysis of collected empirical data regarding the application of this
`new tool in the pharmaceutical sector is presented by capturing the internal view from the pharmaceutical companies themselves
`and the external view from the health care departments of financial service firms. R&D stage specific modi of application, reasons
`for reluctance in the employment of real options and their assumed future prospects are elucidated.
`© 2006 Elsevier B.V. All rights reserved.
`
`Keywords: Real options; Pharmaceutical industry; R&D; Investment analysis; Project valuation
`
`1. Introduction
`
`“The value of R&D is almost all option value” pos-
`tulated Myers in 1984 who firstly recognized the anal-
`ogy between financial options and real world invest-
`ments. For this relationship, he coined the expression
`real option. This term describes the cognition that, based
`on the resemblance mentioned above (R&D) investments
`can be valued similar to financial options. The scien-
`tific basis for this task is provided by the research of
`Black/Scholes and Merton who were awarded the Nobel
`prize in 1997. Real options account for management flex-
`ibility which delivers a significant value contribution
`in the presence of uncertainty. Therefore, real options
`
`∗
`
`Corresponding author at: Technical University of Berlin, Insti-
`tute of Chemistry, Research Group Chemical Economics, Sekr. TC
`8, Strasse des 17. Juni 135, 10623 Berlin, Germany.
`E-mail address: Hartmann@chem.tu-berlin.de (M. Hartmann).
`
`0048-7333/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
`doi:10.1016/j.respol.2005.12.005
`
`analysis (ROA) was recommended several times to be
`more adequate than traditional Net Present Value (NPV)
`for judging R&D projects (e.g. Newton et al., 2004).
`In addition, following a real option’s perspective on
`R&D projects in R&D-intensive companies has a pos-
`itive impact on both their R&D performance and their
`financial performance (Kumaraswamy, 1998).
`These notable statements raise the question of the
`actual level of usage of ROA for valuation tasks inside
`the affected companies. However, due to the exclusive
`inter-sectoral nature of the surveys conducted so far,
`there is no exact picture available that tracks the con-
`crete situation in one particular real option branch.
`This paper aims at investigating the application of real
`options analysis in the pharmaceutical industry. Thereby,
`R&D projects as well as the assessment of whole compa-
`nies are focused. The study considers every R&D stage
`and the different project valuation methods applied there.
`The current and the expected usage of real options anal-
`ysis are determined. The data collection is performed
`
`

`
`344
`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`using a survey based on a written questionnaire. The
`main international pharmaceutical companies as well as
`the health care departments of financial service compa-
`nies have been addressed.
`The current article presents the first detailed empiri-
`cal data of real options analysis for pharmaceutical R&D
`and comprises the following sections: we begin with a
`short introduction on the current situation of the phar-
`maceutical industry and the particular features of the
`R&D process in this sector (Section 2). An overview
`of real options analysis is given in Section 3. The results
`from other surveys regarding real options usage are pre-
`sented in Section 4. Subsequently, the concept of the
`actual survey is explicated (Section 5), accompanied by
`the respective outcomes (Section 6). Finally, a critical
`discussion is undertaken in Section 7.
`
`2. Challenges of the pharmaceutical industry
`and the research and development process
`
`During the last decade, the pharmaceutical compa-
`nies delivered double digits growth rates on average. To
`sustain this path, at least four new drug launches (with
`annual sales of $ 350 million) per year are required for
`every of the large pharmaceutical companies (Bolten
`and DeGregorio, 2002). However, between 1996 and
`2001 only one new drug launch of this category was
`achieved on average (Bolten and DeGregorio, 2002), and
`this decline in productivity proceeds (Kola and Landis,
`2004). Despite of continuously raising R&D budgets,
`an increase in launched products could not be observed
`(Booth and Zemmel, 2004).
`This productivity crisis has several causes. In the
`following the most relevant ones will be mentioned.
`First, the diseases remaining without satisfying treat-
`ments (e.g. cancer and neuro-degenerative illnesses) are
`much more complex than the others, because the under-
`lying mechanisms are not completely understood, yet.
`Second, most of the traditional pharmaceutical compa-
`nies are not able to integrate the emerging knowledge,
`e.g. of the genome information into their R&D pro-
`cesses. The capabilities to develop innovative therapeu-
`tic approaches are established mainly in biotechnology
`firms. Third, the regulatory authorities are more cautious
`and have extended their safety requirements in response
`to observed risks of already marketed drug that con-
`tributes to rising R&D costs (see the recent case of cox-2
`inhibitors for treatment of arthritis).
`In addition, the competitive situation of the pharma-
`ceutical industry has worsened by the patent expiries of
`many blockbuster drugs and the immediate occurrence
`of more cost-effective generics that place a huge pres-
`
`sure on the pharmaceutical industry, because the generics
`are gaining rapidly significant market share resulting in
`revenue decreases for the originators (Grabowski and
`Vernon, 1996). Thereby, many pharmaceutical compa-
`nies are investing in their generic business to profit
`from the ongoing growth path of this product class
`(see Novartis and Pfizer). Furthermore, the efforts for
`cost-containment in public health care systems in com-
`bination with the political support of generics make the
`pharmaceutical business increasingly difficult.
`The above mentioned innovation gap resulting from
`a lack of a promising drug pipeline represents the major
`problem. On the one hand, solutions are being pursued in
`mega-mergers to realize synergies as well as to combine
`R&D efforts and sales forces (e.g. Sanofi and Aven-
`tis). On the other hand, almost all pharmaceutical firms
`depend more or less significantly on strategic alliances
`with biotechnology companies and on in-licensed tech-
`nology as well as therapeutic molecules that stem mostly
`from the biotechnology sector, too (Pavlou and Belsey,
`2005). Another approach is represented by extending the
`usage of a certain drug to further application areas. This
`attempt to receive additional approvals for further indi-
`cations is known as drug repositioning (Ashburn and
`Thor, 2004). Considering this difficult environment, the
`effective resource allocation to the most valuable R&D
`projects is, under these circumstances, one of the most
`challenging tasks of quantitative portfolio management.
`The basis for executing this task are appropriate and effi-
`cient project valuation methods.
`In addition to the above mentioned factors, that shape
`the overall business of the pharmaceutical industry, the
`development of an innovative new drug is associated with
`many uncertainties. Within 10–15 years a new active
`substance has to complete a regulatory fixed sequence
`of R&D stages. Cumulated costs for this task amount to
`approximately $ 900 million on average (Kaitin, 2003).
`The R&D process is marked by high attrition rates due to
`scientific failures. The so-called technical success prob-
`ability achieves only 8% for a new drug (Gilbert et al.,
`2003) and is especially low in the first R&D stages. In
`addition to technical risks, the potential drug candidates
`also face the market risk that results from the unpre-
`dictable commercial performance after market introduc-
`tion.
`The entire R&D process is a highly regulated sequen-
`tial procedure (see Fig. 1) starting with the so-called
`research stage that covers the biological validation of
`the drug target and the subsequent chemical optimisa-
`tion of the potential drug candidate. Moving forward to
`early development, pre-clinical phase mainly comprises
`animal testing. Before entering the clinical phases an
`
`

`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`345
`
`Fig. 1. The pharmaceutical R&D process, adapted from Mohr et al. (in press).
`
`investigational new drug application must be submit-
`ted to the regulatory authorities. Following a positive
`decision, the compound is administered to healthy vol-
`unteers in clinical phase I to gather information about
`safety and dosage. In clinical phase II, application to
`a small number of patients is done to obtain proof of
`the concept. The next step of late development is char-
`acterised by clinical phase III studies that include a
`larger number of patients to ensure statistical signifi-
`cance. After successful completion, a new drug appli-
`cation is submitted to the regulatory authorities to be
`eventually admitted for market launch of the product
`candidate.
`
`3. Real options analysis
`
`In general, two different modi of usage of real options
`analysis (ROA) can be distinguished. Firstly, a utilisation
`can take place in a conceptional manner (real options
`reasoning, ROR) meaning that emphasis is attributed to
`the innovative management philosophy rather than to
`new calculation methods. This “application as a concept”
`aims to provide a more holistic analysis of the project
`features from an option’s perspective.
`The second mode of usage is based on the first one.
`After identifying all relevant options, it is possible to
`employ the real option methodology for concrete valu-
`ation procedures (real options pricing, ROP). Here, two
`common techniques show practical importance. The first
`one refers to the famous Black/Scholes (B/S) equation
`(Black and Scholes, 1973; Merton, 1973) which offers
`an analytical (formula-based) and exact solution. The
`Geske model (Geske, 1979) provides an extension of
`B/S for the valuation of sequential options. The sec-
`ond but less common method is given by the so-called
`binominal lattice (Cox et al., 1979). Here, future cash
`flows are modelled in each time step by an up- or down-
`ward (binominal) movement whose extent is derived
`from the market volatility. Furthermore, binominal lat-
`tices offer widespread extension possibilities including
`
`the accountance for both the technical risks as well as
`market uncertainties.
`Furthermore, there are combinations of Expected Net
`Present Value (Kellogg and Charnes, 2000) and real
`options analysis referring to Smith and Nau (1995).
`These authors explained that under certain circum-
`stances decision analysis and real options can lead to
`identical results. Emphasis is attributed to decision trees
`and the intensive investigation of the value drivers often
`integrated in a comprehensive risk assessment. There-
`fore, this concept facilitates the immediate application
`of the important essence of the real options consider-
`ations without the need for a fundamental change of
`current valuation techniques. For this purpose, there are
`various approaches especially from consultancies, e.g.
`PricewaterhouseCoopers (Krolle and Oßwald, 2001) and
`Bioscience Valuation (Bode-Greul, 2000). In addition,
`suitable software tools for structuring and valuing R&D
`projects in this manner are commercially available. In
`the following, we consider these hybrid methods also as
`a conceptional real options approach.
`
`4. Evidence of practical application of real
`options analysis
`
`4.1. Inter-sectoral level
`
`Until the end of the old millennium several inter-
`national companies (e.g., Merck and Co., Boeing)
`reported the application of
`real options analysis.
`Referring to this data, some authors have already
`announced the “real options revolution” (Coy, 1999).
`Copeland and Antikarov expected in 2001 that the real
`options approach would convert into a standard method
`by the end of the current decade. In contrast to this
`overwhelming euphoria, actual inter-sectoral surveys
`uncovered a stagnate or even a decreasing dissemination
`of real options analysis. A survey of Bain and Co. in
`2000 revealed that only 9% out of 451 participants use
`ROA while observing an abandonment rate of 32%
`
`

`
`346
`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`in this particular year (Teach, 2003). Ryan found in a
`sample of 205 Fortune 1000 firms that only 11.4% use
`ROA as an auxiliary method compared to 96% for NPV
`(Ryan and Ryan, 2002).
`The majority of real option-related empirical sur-
`veys concentrated on the conceptional level, meaning
`the novel management approach. Busby and Pitts (1997)
`showed for companies represented on the FTSE Index
`100 that the higher management of the respective finance
`departments is aware of the presence of options inherent
`to the investment projects. Often, the embedded flex-
`ibility was mentioned as the crucial criterion for the
`investment. Valuation of these options was not exe-
`cuted with the real options pricing routines because
`the participants were not familiar with them. A further
`inter-sectoral study was conducted by Howell and J¨agle
`(1997), also in England. The aim was to compare the
`correspondence of the intuitive real options valuation
`with the respective results derived from real options the-
`ory. The outcomes show clear differences in a direction
`towards an over-valuation by the participants. These dif-
`ferences were the lowest in real options branches as oil
`and pharma. Vollrath (2001) surveyed the capital bud-
`geting approaches of the largest German companies and
`additional ones from real options branches including
`pharma. Real options are ranked at the lowest position.
`The knowledge of the real options approach constitutes
`30–35% depending on the company level in question.
`The abandonment was explained by the complexity of
`the approach that resulted in a black box problem.
`Another reason for the reluctance of ROA application
`is that sometimes RO are perceived and disgraced as a
`New Economy tool. However, this point is not valid if the
`application is done properly. In addition, implementation
`must be supplemented by organisational alterations that
`allow for and facilitate options thinking. Although, basic
`pricing (software) tools are available, now, it is always
`necessary to adjust them oneself significantly to the con-
`crete sector and task in question.
`
`4.2. Sector-specific usage: the case of the
`pharmaceutical industry
`
`Beyond the inter-sectoral evidence, the knowledge
`about the sector-specific application of ROA is extremely
`limited. In pharmaceutical R&D normally sufficient
`market and project data are available to make reli-
`able assumptions about the associated uncertainties for
`the drug candidate as an important input parameter
`for real options analysis. In addition, a scientific- and
`engineering-oriented corporate culture is given that may
`facilitate reference to complex methods (Teach, 2003).
`
`The implicit accountance for real options consid-
`erations in the pharmaceutical
`industry was shown
`recently by means of a statistical evaluation with respect
`to patents and the subsequent exploration of certain
`research directions between 1979 and 1995 (McGrath
`and Nerkar, 2004). Another implicit accountance for real
`options was publicized by Woerner and Grupp (2003)
`who revealed that the enterprise value (indicated by
`the share prices) of a sample of US biopharmaceutical
`companies can be considered as the value of a basket
`option on their R&D portfolios. Thereby, characteristic
`parameters of the risk neutral density function implied
`in observed share prices can be used as R&D return indi-
`cators to describe the perceived commercial potential of
`R&D by the capital markets. In spite of several pricing
`case studies published by academia so far (e.g. Kellogg
`and Charnes, 2000; Cassimon et al., 2004), empirical
`data of usage in daily routines are still lacking.
`Only Merck and Co. reported the application of real
`options pricing with B/S with respect to valuation of
`biotech investments (Nichols, 1994). Remer et al. (2001)
`revealed that the real options approach is merely known
`and subsequently not applied in European biotechnol-
`ogy companies. These findings were confirmed by a
`small survey conducted by Lun and Peske (2002) by
`interviewing selected German biotechnology companies
`regarding their methods for investment analysis. Here,
`real option-based approaches show only a marginal role
`that might be due to the low grade of maturity of this sec-
`tor in Germany resulting in the absence of specialised
`finance departments that apply sophisticated valuation
`tools regularly. Currently, industry experts propagate
`an integrative approach to project evaluation that also
`includes real options analysis (Jacob and Kwak, 2003).
`
`5. Survey conception
`
`The empirical results that will be presented in the
`next section were collected by a survey based on a
`written three-page questionnaire executed between
`February and October 2004. To obtain a holistic picture
`of the dissemination of real options analysis (ROA) in
`the pharmaceutical industry, we pursued a double-sided
`approach. On the one hand, we addressed companies
`from the pharmaceutical/biotech sector (in the follow-
`ing: pharmaceutical section) to capture the internal view
`of their R&D. On the other hand, we supplemented this
`by the external perspective of the health care divisions
`of investment banks, auditors and consultancies (in the
`following: capital market section). The first class concen-
`trates more on private risks defined by technical success
`rates, whereas the latter focuses also on market-related
`
`

`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`347
`
`risks. By combining these two aspects, it is intended to
`yield a good approximation for the overall situation.
`The survey range comprised leading international
`research-based pharmaceutical and biotech companies.
`They were selected following a top 50 company list
`ranked by sales in 2002 and 2003 (Sellers, 2003, 2004).
`The focus on the largest and most important com-
`panies was done due to the former findings of sur-
`veys with respect to smaller biotech companies that
`revealed a marginal usage of cutting edge project val-
`uation approaches such as Expected Net Present Value
`and real options. Therefore, more sophisticated methods
`on cutting edge might be more likely to find in larger
`firms which maintain distinct departments for portfolio
`analysis. In total, 28 answers (out of 56) from partici-
`pants stemming from the three most important markets
`Europe, USA and Japan (according to the headquar-
`ters location) were received (see Fig. 2A). The related
`sizes of the R&D departments and the company divisions
`covered by the answers are depicted in Fig. 2B and C,
`respectively. In the financial service section, we included
`national as well international companies by contacting
`the health care divisions, mainly of their German or
`European offices (distribution: see Fig. 2D). In this sur-
`vey part, we received 27 (out of 56) responses. Although
`the answers are not supposed to be significant for the
`whole pharmaceutical sector, they have a clear impor-
`tance for large international pharmaceutical companies
`by revealing important trends with respect to the usages
`of different methods for project valuation in general and
`real options in particular.
`
`6. Survey results
`
`6.1. Valuation methods
`
`The first question regarding content referred to the
`valuation methods used for different valuation tasks. In
`order to provide an overview of the answers of the par-
`ticipants, the usage of each method by the participants is
`related to the number of answers. The resulting percent-
`age was then categorised into four groups. A percentage
`that exceeded 50% was classified as a main method (most
`intensive grey background and white letters) used by the
`majority. Then, two kinds of auxiliary methods were
`defined. The first one, the so-called auxiliary method
`I (the second most intensive grey background), cov-
`ers the area between 50% and 26% usage. The second
`one, the so-called auxiliary method II (the third most
`intensive grey background), comprises the values rang-
`ing from lower than 26% down to 11%. The remaining
`niche methods (lightest grey background) are negligible.
`Methods without any mentioning are excluded and not
`represented.
`The results from the pharmaceutical section and
`the financial
`service companies are depicted in
`Tables 1 and 2, respectively. Except for the research
`phase, the survey clearly confirmed the assumed domi-
`nance of NPV-based valuation approaches in R&D in the
`pharmaceutical section. However, seemingly ROA also
`found its place in the method set with its peak usage lying
`in clinical phases. In contrast, ROA has absolutely no
`significance in the research stage. Here, while observing
`
`Fig. 2. (A) Regional distribution of the pharma participants according to location of headquarter. (B) Distribution of pharmaceutical R&D expen-
`ditures of the participants in 2003 (million Euro). (C) Pharma company divisions covered by the answers. (D) Distribution of the participants from
`the financial service section.
`
`

`
`348
`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`Table 1
`Evaluation methods in the pharmaceutical section (E)NPV: (Expected) Net Present Value, DCF: Discounted Cash Flow, RoE: Return on Equity,
`RoI: Return on Investment, EVA®: Economic Value Added
`
`Table 2
`Evaluation methods in the capital market service section (E)NPV: (Expected) Net Present Value, DCF: Discounted Cash Flow, RoE: Return on
`Equity, RoI: Return on Investment, EVA®: Economic Value Added
`
`

`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`349
`
`the lowest values for NPV, that keep their main meth-
`ods status in spite of this, scoring models almost reach
`classification as a main method. Risk analysis is mainly
`provided by decision trees (DT), scenario and sensitivity
`analysis (that is valid for all phases). In addition, Monte
`Carlo simulation (MCS) and payback period have some
`dissemination. Again, the research stage delivers a dif-
`ferent picture, described by a lower prevalence for all
`risk assessment approaches.
`Company valuation, especially for early biotechs
`(not publicly listed and without marketed products) and
`young biotechs (publicly listed and without marketed
`products), shares many features with R&D project val-
`uation due to the fact that their lead R&D compound
`accounts for the key value driver. However, even by
`considering mature biotech firms with marketed prod-
`ucts (“old” biotechs) from the pharmaceutical section’s
`perspective, ROA usage was not detected in this set-
`ting. One organisational reason may be given by the fact
`that our participants main task was R&D portfolio man-
`agement rather than company valuation, which is often
`executed by distinct M&A divisions together with invest-
`ment banks and consultancies.
`Switching to the capital market section, slight vari-
`ations occur. Here, the clear dominance of NPV-based
`approaches in R&D project valuation is observed, too.
`Conversely, there is a little usage of ROA in the research
`stage while its peak utilisation is situated in the pre-
`clinical phase and clinical phase I, declining towards
`market introduction. Other important valuation meth-
`ods are scoring models for research and pre-clinic
`projects, return on equity (RoE)/return on investment
`(RoI)/economic value added (EVA®) for clinical projects
`as well as multiples for clinical phase III and registra-
`tion. Risk assessment is not performed as often as was
`observed in the pharmaceutical section, but also mainly
`covers MCS, DT, scenario and sensitivity analysis.
`The capital market section additionally includes two
`further company valuation tasks, namely that of small-
`and medium-scale pharma as well as Big Pharma. Pre-
`dominately for all company valuations are NPV-based
`approaches and multiples. ROA always amounts to aux-
`iliary method II status. Only RoE, RoI and EVA® show
`some further importance for larger companies. Risk
`assessment concentrates more or less on scenario and
`sensitivity analysis.
`
`6.2. Personalised medicine
`
`After the completion of the deciperation of the human
`genome and the beginning of understanding the molec-
`ular basis of diseases, the establishment of a new era
`
`Fig. 3. Does personalised medicine require new project valuation
`methods?
`
`of drug development is on the rise. Tailor-made drugs
`(pharmacogenomics) will probably replace or at least
`transform the current blockbuster business model in the
`mid to long term. The application areas of these future
`pharmaceutical compounds will be limited to certain
`patient sub-populations resulting in low(er) net present
`values. Much of their value will be contributed by growth
`options (e.g. broadening the therapeutic indication or
`the mode of administration). This raises the question
`whether new valuation methods are required to cope with
`these fundamental changes.
`Interestingly, among the pharmaceutical companies a
`clear direction could not be observed. 43% answered the
`question with yes and the same amount with no. Indeed,
`it seems to represent an issue that future research has
`to deal with. The relatively high percentages of blank
`answers (15%) might be explained by low immediate
`importance of this issue that may render the involvement
`with a first example within the own company. In contrast,
`the capital community has a more clear feeling about this
`issue with a rejection rate of 67%. Only one third agrees
`with the hypothesis above. The majority of the combined
`sample does not follow this argumentation and considers
`the current tool set as sufficient. However, a remarkable
`amount sees the necessity for new methods (Fig. 3).
`
`6.3. Knowledge and modi of usage of the real
`options approach
`
`The first reported practical application of real options
`pricing of a pharmaceutical company stems from Merck
`and Co. (Nichols, 1994). Since then, several examples
`and case studies were presented by academia. In addi-
`tion, real options theory is taught in almost all MBA
`courses and standard textbooks on corporate finance
`cover this topic. However, does this mean that all com-
`panies are familiar with this new approach and that they
`have dealt intensively with its relevance or even use it?
`
`

`
`350
`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`Table 3
`Knowledge of real options analysis (ROA)
`
`Level of ROA
`knowledge
`
`Pharmaceutical
`section
`
`Capital market
`section
`
`Combined
`
`between the share of companies that have dealt with the
`theory and those few that have implemented real options
`analysis raises the next question regarding the reasons
`for this reluctance.
`
`No knowledge
`Known by name
`Contents known
`Application: real
`options reasoning
`Application: real
`options pricing
`
`14
`14
`43
`11
`
`18
`
`All numbers in percent.
`
`7
`33
`37
`11
`
`11
`
`11
`24
`40
`11
`
`15
`
`Table 3 answers this question. Within the pharmaceu-
`tical company section, 15% have no knowledge about
`real options analysis. All these answers stem from R&D
`functions associated with project management. The same
`percentage only knows them by name. On the other hand,
`44% have dealt at least with the underlying theory and its
`implications. Finally, only 26% have implemented real
`options analysis in their daily routines. 15% apply the
`instrumental real options approach and 11% the concep-
`tional one.
`In the capital market section the overall knowledge
`of real options analysis is slightly higher. Only 7% have
`not heard about it, so far. A 33% are familiar with at
`least the name. However, the discussion of the contents,
`reaching 37%, is less common than in the pharmaceutical
`section. Regarding ROA usage, there are no major dif-
`ferences between the two groups, although real options
`pricing appears to be somewhat more common in the
`pharmaceutical section.
`It can be concluded that overall level of knowledge
`is relatively high. However, the significant difference
`
`6.4. Obstacles for the usage of real options analysis
`
`To elucidate the hindrance reasons for the initial or
`further employment of real options analysis, a large vari-
`ety of potential reasons where given to the participants.
`Fig. 4 shows the outcome to this question. The topics
`are ranked by their number of enumeration yielded by
`the overall survey sample. To simply compare the differ-
`ences between the two subgroups, they are also included
`separately.
`The far most relevant aspects are provided by the
`assumed complexity of the real options approach and the
`lack of acceptance from decision-makers and customers.
`There is almost no difference between the pharmaceuti-
`cal and the financial service companies. Lack of trans-
`parency and lack of options pricing knowledge are also
`nearly equally distributed within the two subgroups with
`a small tendency for pharmaceutical firms to be more
`critical. Satisfaction with existing methods is higher for
`the pharmaceutical companies whereas the financial ser-
`vice firms are more concerned about the non-standard
`method character of real options pricing. Reliability and
`integration into existing models are further criteria that
`are raised more from the pharmaceutical side. On the
`other hand, the implementation expenditures are con-
`sidered as very high, preferably by the financial service
`companies. This might be explained by the consequent
`orientation to customers external to the company of these
`firms. In contrast, valuations inside pharmaceutical com-
`
`Fig. 4. Reasons for reluctant (further) usage of real option pricing.
`
`

`
`M. Hartmann, A. Hassan / Research Policy 35 (2006) 343–354
`
`351
`
`panies are exclusively used internally as decision support
`for the higher management.
`Especially the pharmaceutical section wants to see
`more case studies. The reason for this apparently lies
`in the heterogeneity of the applied real options pricing
`methods published so far. Furthermore, the quality of
`the case studies presented to date, are not satisfactory
`at all regarding the complex situation in the pharmaceu-
`tical environment, e.g., different scenarios with respec-
`tive distinct features and competition challenges to be
`modelled. Surprisingly, there is very little mention of
`concerns regarding the methodical point of view com-
`pared to organisational aspects. The scientific basis of
`real options pricing was put to question only three times
`as well as the difficult determination of the appropriate
`volatility (twice).
`
`6.5. Comparison of the real options approach to the
`NPV method
`
`Due to the fact that reliance was mentioned as an
`important problem of the real options calculation results,
`the levels of predictability of RO and NPV, respectively,
`were compared on a scale ranging from 0 (poor) to
`10 (excellent). Within the pharmaceutical section the
`preference for the NPV approach is obvious and clear:
`6.4 versus 5.4. In contrast, the financial service compa-
`nies see almost no differences regarding predictability
`between the two methods: 5.8 (NPV) versus 5.6 (ROP).
`Interestingly, the value for the NPV is clearly lower than
`reported in the pharmaceutical section and the variance
`of the answers for the RO value in the capital market
`section is 50% higher than that for NPV, whereas in the
`pharmaceutical section these values are almost equal.
`This indicates that a clear picture of the RO approach
`has not been formed, so far, especially within the finan-
`cial service companies.
`
`6.6. Real options pricing
`
`6.6.1. R&D project valuation
`In the entire Section 6, multiple answers were possi-
`ble. Here, the participants were questioned concerning
`the concrete real options pricing (ROP) methods that
`are used or intended to be used for valuation purposes.
`Fig. 5 shows that in the pharmaceutical section a higher
`tendency occurs for the application of Black/Scholes in
`project valuation as it is stated by the financial service
`companies. This represents an interesting point due to
`the fact that the B/S equation is not able to capture the
`technical risk of an R&D project directly. The more flex-
`ible lattice-based approaches are preferred by the capital
`
`Fig. 5. Methods of real options pricing for R&D projects.
`
`market section. The G