Measuring Value in New Biotech Firm An Empirical Modelling Approach

This study links two key disciplines in finance and science in a way which is representative of the many challenges in the development of the knowledge economy. The biotechnology sector is regarded as one of the major growth engine in the development of the knowledge economy and in order to sustain this, it is imperative that biotech firms continue to attract capital needed to finance business activity. An important aspect of raising capital is the ability to impute a fair value on the asset. Biotech firms are typically intellectual property intensive and derive its value from this source. However, the valuation of intangibles remains a contentious issue in finance as to the most suitable approach. In this study, the financial and economic issues pertaining to the valuation of new biotech firms are evaluated and addressed vis-à-vis the value drivers of these firms. The value drivers are tested for significance and hypotheses are formed on their financial, economic and management implications. The investigative process involves using quantitative methods, such as descriptive statistics, logit modelling and multivariate regression, to evaluate the valuation issues in the context of the challenges faced in valuing these potentially high-growth and intellectual-capital intensive firms.


Introduction
Innovation research in economics is extensive and increasing (Cohen and Levin 1989), stimulated primarily by the major role of innovation in the theory of economic growth.Contrary to the conventional economic theory of diminishing returns developed in the nineteenth century, Arthur (1996) advocates the law of increasing returns.According to Arthur, increasing returns 'are the tendency for that which is ahead to get further ahead, for that which loses advantage to lose further advantage.'Firms with these characteristics make it difficulty to forecast their future cash flows and growth rates.The high risks associated with innovative business ventures, especially growing start-ups in industries like biotechnology, raise serious questions about the choice of an appropriate discount rate and consequently value.This makes it imperative intangible value drivers like technology know-how, human capital, speed of growth, competition and customer loyalty need to be considered in the valuation process.
The valuation of a firm is important as entrepreneurs rely on it to raise capital to continue business activity.Valuation is an important step in the funding process in that it involves assigning a quantitative value, even for intangible asset like biotechnology, because money is the common denominator that brings together technologists and investors for negotiating transactions and decision support involving technology.Like all young companies, new biotech firms face difficulties in early-stage financing and can only rely on equity capital in exchange for ownership of part of the company rather than loans.The availability of public equity is also limited for early-stage biotech firms due to the lack of a performance track record to rely on for raising funds in public equity markets, posing a formidable challenge for valuing biotech without secondary data.
This paper investigates the contemporary financial and economic issues relating to value drivers of new firms in the Australian biotechnology (biotech) sector and their dynamics at early-stage development.Sometimes referred to as dedicated biotechnology firms or generally small and medium sized enterprises (SMEs) and are often founded by key scientists and mainly employ technical staff.They usually form part of a bigger network or cluster (Swann et al. 1998), for example within universities or technology parks, and 'co-produce' with other units within the network (Senker 1998).The investigative process involves using quantitative methods, such as descriptive statistics, logit modelling and multivariate regression, to evaluate the valuation issues in the context of the challenges faced in valuing these potentially high-growth and intellectual-capital intensive firms.The biotech sector value drivers are identified and used as basis to develop a valuation model for early-stage biotech firms.In this context, the methods for analysing technology initiatives and estimating their future value are addressed in detail.

The Biotechnology Firm
The value of a biotechnology firm company is typically driven by the value of intellectual capital residing in the firm and its ability to convert it into future cash flows.The intellectual capital of biotech firms is associated with the primary activity of conducting research to develop new drugs (R&D).Biotech firms are perceived to have a high-level performance environment and one that composes of components that are tightly bound to one another and calls for an integral processes or integral product architecture (Ulrich 1995).The individual components, which are closely related and highly coordinated, perform many functions similar to key value drivers that form the value-chain of a firm.This scenario conjures an integral environment which consists of collaborations, interactions and contingencies among the key factors, which would require an integrated and dynamic approach in order to conduct a comprehensive valuation of the biotech firm.Therefore, the ability to identify and strategically manage the key value drivers of a biotech venture is a key prerequisite for value maximizing and profit maximizing.
Diagram 1 Equity Returns of the Biotech Sector -1992Sector - to 2002 The returns for the Australian biotech sector are depicted in Diagram 1 show an upward trend from 1992 to 1997 and after which there is a progressive decline to 2002.Tilling (2000) provides some probable explanations for this decline: High risk of the industry makes capital more expensive; High cost and long lead time for regulatory processes to approve drugs; Low profitability as biotech has created few products and a lot of promise (a worldwide average of 80% of all publicly held biotech firms lose money); High cost of consolidations and the effects of increasing competition, and The view that the biotech industry has become a mature, emerging industry.The days of rapid growth in stock prices may be over.
The volatility of returns for the sector from 1992 to 2002 is 8.9% (standard deviation -σ) compared to the market of 14.57% (calculated from the S&P/ASX 200 index) in the same period.This shows that the biotech sector is less volatile than the general market in the period.The risk and return statistics for both the biotech sector and the general market as represented by the S&P/ASX 200 is presented in Table 1 below.

Literature
Modern financial management is largely concerned with the optimal matching of the uses and sources of corporate funds that will maximise the firm's market value (Chew 1997).The value of a financial asset is the present value of the cash flows expected from that asset.The idea of the rational investor and the efficient markets hypothesis underpin equity prices.
They reflect the true value of economic fundamentals and market efficiencies preventing attempts by investors to make excess profits.This value may be determined by capitalising either the dividends or the future earnings to which the stockholders are or maybe entitled and the stock today is calculated as the present value (PV) of an infinite stream of cash flows (Gordon and Shapiro 1956): Value of firm = P 0 = ∧ PV of expected future cash flows where: CF t = the net cash flows in period t; K = the discount factor; and ∧ PV = the current market price of the firm.This is also known as the discounted cash flow method (DCF) and it operates under the premise that the value of a firm is obtained by the sum of the present value of cash flows to be generated by the firm's existing assets and the present value of cash flows to be generated from future growth opportunities.For most startup biotech firms, the absence of positive cash flow poses a problem in adopting the DCF method for valuing a firm.Even if positive cash flows can be reasonably estimated in the absence of historical earnings, the extrapolation of growth on earnings estimates might be too subjective.The most commonly used DCF valuation approaches are those based on intrinsic financial attributes and earnings (Guatri 1994).The intrinsic method is based on present value of expected cash flows associated with specific strategic or management choices.The earnings method is generally characterised by the use of free cash flows for valuing the firm.The premise of this study is based on the limited historical earnings data for startup biotech firms, which necessitates an 'intrinsic approach' using econometrics to identify and specify value drivers for valuing biotech investment.
In order to address the integral nature of the biotech environment (see Section 2 above), this study adopts an integrated and holistic approach for valuing the biotech firm by incorporating and testing the relevant value drivers derived from the primary survey conducted in this research.The specification of the biotech valuation model is done using multivariate regression analysis, which helps to address the integrative and interactive elements of the key value drivers according the proposition by Ulrich (1995).This approach addresses the idiosyncrasies of the unique and dynamic biotech business environment, which is volatile, high risk, high growth and rapidly evolving.This paper builds on research done on biotech valuation by others researchers such as Kellogg and Charnes (2000) in real options, Myers and Howe (1997) in life cycle financial models and Arojarvi (2001).
In this study, the returns for the Australian biotech sector are used as the proxy for measuring the value of a biotech firm.There have been extensive studies conducted on the relationships between stock market returns and fundamental economic activities in the United States (Fama 1970(Fama , 1990(Fama , 1991)).By using live data, real economic are reflected in the testing and development of the biotech valuation model.Chen, Roll and Ross (1986), Chen (1991), Pearce and Roley (1988) and Fama (1991) have modelled the relation between asset prices and real economic activities using factors such as productivity, growth rate of gross national product, production rates, yield spread, inflation, unemployment and other real activity indicators.

Data and Methodology
The statistical technique of regression analysis is commonly used in economic analysis and this approach forms the basis of the quantitative analysis of this study.It involves three steps of model specification, estimation, and inference.Using this process, both primary and secondary data are used in a series of statistical triangulations and manipulations to assess and estimate the relationship between the biotech firm's value and economic factors that contribute to value creation.This paper builds on earlier research by Oh (2001) and Bose and Oh (2004) on intellectual capital valuation using OLS-regression analysis techniques.A logit valuation model for biotech firms is developed in Bose and Oh (2004) and the results from this model highlight several pervasive value drivers for a firm operating in the biotech sector (see Table 1).This study continues the development of the valuation process using the results from Bose and Oh (2004) to test and build a more comprehensive econometric valuation model for the biotech firm.
The empirical analysis of the high technology firms using financial econometric techniques (Mills 1999;Campbell, Lo, and MacKinlay 1997) is a relatively new area of research.The modelling techniques used in this paper incorporate both logistic modelling and multivariate regression techniques for modelling biotech valuation models.In addition, descriptive statistics and a variety of current econometric models for the analysing financial markets are used to evaluate the Australian biotech sector.
This study starts by capturing the biotech firm's value drivers specified in Bose and Oh (2004), which uses a triangulation of methods: literature review, case study, questionnaire survey and logit analysis to identify biotech value drivers.The biotech value drivers are first identified from literature and used in the case study interviews with biotech firms.The case study response is then analysed using NVIVO software and consolidated into a questionnaire survey, using a combination of uni-dimensional scaling technique (i.e.Likert scale) and sent to a sample of 100 Australian biotech firms of which 33 responded.The logit modelling approach is then used to evaluate the results from the questionnaire survey to statistically specify the biotech value drivers (see results Section 4.1 below).The limitation of such an approach for selecting the value-pervasive variables is that it could have a management bias in the data as they reflect management's perceptions of value and not that of the market.
In this paper, the value parameters specified in the logit analysis (Bose and Oh 2004) are use to estimate the biotech valuation model (BVM) using multivariate regression analysis.
This process involves using real market data to test the parameters specified in the logit model in order to estimate and specify the BVM.It is noted that the use of underlying value drivers as a means for estimating the value of intellectual-capital intensive assets is a challenging endeavour and is also a relatively new area of research (Helfert 2000).On this premise, the research carried out in this study is of a highly exploratory nature and the statistical tests done are subject to less stringent bounds compared to other more established research areas.The rest of this section lays out the relevant theory and statistical manipulations explaining the rationale and their implications.

Logit Models
The use of logit modelling is to address the unidimensional responses of the biotech questionnaire survey in determining the pervasive biotech value drivers (Bose and Oh 2004).Here, the inherent correlations between the variables are measured and scored according to their pervasiveness in value creation for the biotech firm (see below).In the survey, the series of questions asked are representative of possible value drivers, and they take on responses from "strongly disagree" to "strongly agree".By adopting the logit approach, we are able to quantify these responses and identify the pervasive value drivers through statistical specification.To describe the process we start with the matrix notation of the standard regression, which may be written as: (1) where: y = a vector containing observations on the dependent variable; X = is a matrix of independent variables; β = a k-vector of coefficients; and ε = is a disturbance or error term.
The operations of the model are applied on the data collected from the questionnaire survey to analyse for the pervasiveness of the specified set of value drivers.By doing so, the specified logit equations will help to explain whether the set of independent variables (i.e.value drivers) are statistically significance in collectively determining the value of firms in the biotech sector.The statistical tests conducted are to determine the pervasiveness of the value drivers on the value of a ε β + = X y biotech firm.The statistical manipulation starts with the assumption that there are m companies and n questions (on value drivers), and we give every response a score as in Table 2:

Strongly agree 4
Then the matrix is specified and the above can then be denoted as follows: (2) where: π j = the degree of agreement; The objective of this regression manipulation is to statistically specify the value drivers and determine the pervasiveness of these variables on the dependent variable according to the scores given by the respondents to the questionnaire.By giving each value driver a score between 0 to10, the scores of the value drivers are treated as the independent variable.The first step of the process is to observe from the graph for each sector for the degree of agreement accorded by the respondents to the value drivers (Diagram 2), the higher the degree of agreement to a question, the higher the score will be for the corresponding value driver.In the second step, the scores are statistically adjusted to find the best-fit logit model.In other words, between 3 to 5 of the most pervasive value driver are scored from a scale of 1-10, and the remaining value driver are given a score of zero and excluded.

Multifactor Regression BVM
From the results derived from the logit analysis in Section 3.1 above, the estimated parameters are used to specify and develop a linear multi-beta model, using the method of ordinary least squares (OLS) for estimating the value of a biotech firm.This relationship is consistent with the CAPM, which proves that there is a linear relationship among prices of assets in a general equilibrium, where the investors select assets to maximise the meanvariance utility.Traditional asset pricing methodologies, such as those of Breeden (1979), Ross (1976), Black (1972), Lintner (1965) and Sharpe (1964) show that the expected return on a financial asset is a linear function of its covariances with some systemic risk factors (as in the CAPM).This characteristic is assumed in the development of the BVM as specified below.The statistical tests and manipulations are conducted using Microsoft Excel software.The BVM model is developed incorporating significant pervasive variables and is specified and estimated for its ability to capture the essence of the fundamental economic and financial forces that contribute to the biotech firm's value, in a concise and readily testable form.Thus, if there are k factors, a general representation of the typical BVM can be written as: where: Biotechret t = the return of biotech firms in period t; α i = the expected value if each factor has a value of zero; (F 1 ) t , (F 2 ) t …(F k ) t = the values of factors 1, 2 … k with pervasive influence in period t; (β 1 ) & (β 2 ) = sensitivities of the biotech firm to the factors; (β k ) = the change in the value of the biotech firm per change in factor k; … = terms of the form (β k )(F k ) t with k going from 3 to k -1 in period t; and ε it = random error term.
The data for the logit analysis are consolidated from a questionnaire survey received from 33 biotechnology firms and the market data for the multivariate regression evaluation are from various sources as discussed below.

Sector Return
This time series data was sourced from the Standard and Poor's Industry Financial Data for Australia.The figures for each year show the average returns from companies listed on the Australia Stock Exchanges and estimates of returns of private firms and unlisted companies from the database of the industry association, i.e.Alltech Biotech.

Patents Issued
The database of IP Australia shows the total number of patents issued in those years in Australia for biotechnology products to Australian and international companies.

Management Qualifications
This information was sourced from the database of Alltech Biotech and contains information of tertiary level education of management and research staff of member firms.An estimate of the figures for non-member firms was added to arrive at the total.

Market Risk Premium
This time series data was sourced from by the Australian Bureau of Statistics.The calculations were done by deducting the Australian 10-year treasury bond rate from the market return.

R & D Expenditure
This data was estimated by Alltech Biotech.R&D expenditures by Australian government for biotech were not available prior to 1995 and R&D expenditure from government sources for the years 1995 -2002 did not agree with that of Alltech Biotech.Therefore, an average was calculated to arrive at the figures for the study period.

BVM Specifications
This section sets out the results from the logit and regression analyses conducted on the biotech sector.

Results Logit Analysis
The correlations between the identified independent variables (x) and dependent variable (y) in each sector and their scores have been plotted and the graphs are presents below:

5173, Diagram 2 Degree of Agreement of Biotech Value Drivers
In Table 3 below, the four most biotech pervasive value drivers from the logit analysis are presented.Government R&D support (R&D) 7 In using the logit model to specify the pervasive biotech value drivers, transformation by way of statistical manipulations between correlations and the assigned scores (see above) are conducted on the independent variables.This process identifies the pervasive biotech value drivers according to their distinctive scores and from the list, four of the most pervasive value drivers are selected (Table 3).Lehmann and Modest (1988), and Connor and Korajczyk (1988) find little sensitivity to increasing the number of factors beyond five and Fama and French (1995) find that stocks require only three factors.
The statistical rationale in the selection process for exclusion of the less pervasive independent variables is that, from the response of a value driver within the group of questions in the survey, the degree of agreement in the group of questions varies, if the degree of agreement for a value driver is highly different from the others, it is considered nonpervasive and excluded from the selection process.The use of the highest correlations as the basis for identifying the top 3-5 value drivers, which are incorporated on an integrative and collective basis for their value-generating power as a group of pervasive value drivers in biotech firms.This approach is considered appropriate as strong corporate performance is dependent on the right combination and effectiveness of various corporate and market factors (Ulrich 1995) as generally practised in equity value market analysis.Roll and Ross (1980) using factor analysis found that only three and possibly four factors explained the return generating process of US equities.Factor modelling potentially provides the benefit of reducing the variance of the abnormal return by explaining more of the variation in the normal return.This variance reduction is typically the greatest in cases where the sample firms have a common characteristic, in this case the biotech firms, where they are all members of one industry sector (Campbell et al. 1997).

Logit Specifications and Multivariate Regression Results
The BVM model building process involves the selection of significant explanatory factors for biotech firms' value.The objective is to construct the BVM, using regression analysis, with 3 to 5 factors that could parsimoniously estimate the value of biotech firms.The process includes descriptive statistics, regression estimation and hypothesis testing, to confirm and select only those factors that make a significant contribution to the value creation of the biotech firm.If there are no significant relations between the identified factors and the value of the biotech firm, we can conclude that these factors are non-pervasive to value and do not reflect the real value creating activities in the biotech sector.
The regression manipulations and specification for the BVM model are conducted using the data derived in Section 4.1.The specified BVM model and the statistical significance of the pervasive value drivers are tested and discussed in this section.The preliminary BVM based on the value driver parameters specified in the logit model can be represented by the following equation ( 4): Biotechret t = the return of biotech firms in period t; P t = Patents issued in biotechnology sector in period t; MQ t = Quality of management in period t; MRP t = Market risk premium in period t; R&D t = Research and development intensity in period t; (β 1 ), (β 2 ), (β 3 ) & (β 4 ) = sensitivities of biotech firm to the factors, and ε it = random error term.
The relationships between the logit specified value drivers and the value of the biotech firms can be presumed to take on the following relationships: The conjectures in Table 4 depict the possible relationships between the variables, which may or may not be confirmed by the statistics in the final evaluation of the BVM model.They represent the relationships that exist according to common perceptions, established theory and intuitions.Patents issued have implications for both research and development intensity and competition.It is positive for biotech valuation if R&D intensity helps biotech firms develop more marketable products and hence revenue.It may be negative for biotech valuation if more patents mean higher level of competition for market share in the sector, which ultimately affects profitability.It could also imply more competition for scarce capital causing cost of capital to increase or project abandonment by those firms unable to secure funding.A high level of professional in management is intuitively good for the biotech valuation as it could enhance performance and profits.Risk has a negative connotation and is considered bad for business.Risk can manifest itself in many forms for a biotech firm and they include the R&D failures, cash flow or funding problems and competition.R&D intensity in the form of government funding and infrastructure supports that encourage R&D activities is beneficial to the biotech firm and should improve a biotech firm's market value.
In conducting the t-statistics tests, a more liberal 20% significance level is used due to the small sample size (n = 11 yearly observations) and the fact that the valuation of startup biotech firms are a rather new market phenomenon with a paucity of information or data about value drivers and their influence on biotech firm value.Under these circumstances, it is deemed prudent that a 20% significance level, for hypothesis testing against a two-sided alternative, is appropriate for the exploratory nature of this research.It is expected that a more stringent significance level can be applied for future research as more definitive relationships and parameters are established between biotech firm value and real economic factors.It is also a general view that "different researchers prefer different significant levels depending on the particular application and underlying agenda and there is no 'correct' significance level" Wooldridge (2000).The p-values for the t-statistics will also be computed for those factors that are considered pervasive to e-commerce returns to ascertain their degree of influence even if their β F coefficients are statistically insignificant.The evaluation of those factors judged pervasive using the p-value, we attempt to tease out further characteristics of these explanatory variables.The correlations between the variables in Equation 4 are presented in Table 5.All the explanatory variables have a negative correlation with biotechret.This suggests that there is an inverse relationship between the dependent variable and individual explanatory variable.The explanatory variables or value drivers have relatively high correlations between them, from a high of 0.9645 for R&D/MQ to a low of 0.5125 between P/MRP.Only the correlations between biotechret and P and MRP correspond to that extrapolated in Table 4.
The estimated BVM model developed from the regression analysis is represented by Equation ( 5).The choice of value drivers in the model (i.e. the explanatory variables in the BVM) is conducted using stepwise regression, which is a combination of the forward selection and backward elimination methods by running models and selecting the variables with the largest F or t-statistic.
Biotechret t = 144.80-0.0055(P t ) + 0.7465(MQ t ) -15.0301(MRP t ) -0.3607(R&D t ) (5) (1.4082) (-1.5150) (1.4510) (-1.6196) (-1.1874)R 2 = 0.86 Equation ( 5) above summarises the important statistics of the estimated regression BVM equation for the Biotech sector with the coefficient of determination or R 2 of 86%.This suggests that in the regression equation the explanatory variables explain 86% of the dependent variable as represented by R 2 in the study period.This is considered high in terms of explanatory power of the value drivers on biotech return.The statistical significance of the coefficient of each value driver for statistical inference at the 20% level of significance is shown in Table 6 below.Though the R&D factor as a value driver is statistically insignificant at the 20% level in terms of t-statistic (Table 5), its p-value is 0.2799, hence we would observe only 29.97% of the t-statistics in all random samples for this factors when the null hypothesis is true.This is relatively strong evidence against H 0 as indicated by the p-value for this factor signifies that it does have a certain degree of pervasive influence on value creation for the biotech firm despite its insignificant t-statistics.
Table 6 shows that P, MQ and MRP are significant and are accepted as pervasive value drivers for biotech valuation in the hypothesis tests.As such, they are considered explanatory factors for biotech valuation.This implies that the accepted value drivers are statistically significant and have inferential implications about the population values.The revised BVM for biotech firms incorporating the three value drivers is present as follows: FirmsValue Biotech = 254.81-0.0018(P) + 0.1531 (MQ) -24.3700(MRP) (6) (5.5535) (0.9353) (1.2161) (4.8097) R 2 = 0.82  7) above represents the final BVM for the biotech sector value and it summarises the important value driver in the biotech industry with a coefficient of determination or R 2 of 78%.This is considered high in terms of explanatory power of the value drivers on biotech return as this suggests that the value driver, MRP, explains 78% of the return of the biotech firm in the study period.The critical value for a two-sided hypothesis test at 9 degrees of freedom is 1.383.The t-statistics of the revised BVM (as represented by Equation 7) is -5.5875, which therefore rejects the null hypothesis.This tstatistic also satisfies the test for 1% significance level at the 1% critical value of 3.250.Equation ( 7) represents the final BVM model for firms in the biotech sector, which specifies and implies that risk is the key pervasive value driver to the biotech firm valuation.

The BVMs and their Value Drivers -Implications for Biotechnology Valuation
The problem of deciding which value drivers or independent variables to include in the multiple regression equation is closely linked to the decision of how to define the best model that explains the dependent variable, which is the value of the biotech firm, using the smallest possible set of independent variables.Ostensibly, the estimated 4-factor BVM model (Equation 5) would best represent the longterm relationship between the biotech firm's value and the specified value drivers or economic variables.However, after conducting the t-statistics hypothesis tests on 4-factor BVM model (Table 6), we reject "R&D Expenditures" as being a pervasive factor for biotech valuation, leaving only P, MQ and MRP, implying a 3-factor BVM model as depicted by Equation 6. Subsequent t-tests on this 3-factor BVM model revealed that only one factor (as in Equation 7 above), MRP -market risk premium, is statistically significant in driving value for the new biotech firm in the study period.

Economic Implications of the Value Drivers
The inferred economic rationale of each of the value drivers tested in the BVM development process is discussed below.

Patents Issued (P)
Patents do not only encourage biotech firms to spend on R&D and innovations as it provides a monopoly to benefit from the intellectual assets for a limited time, but also provide a basis of valuation (Leuhrmann 1997).The proxy for measuring the "uniqueness of innovation" in this research is the number of patents issued.According to Levy (1998), the uniqueness of an innovation is a major determinant of value, because it would have intrinsic market appeal and are genuinely unique as distinct from those that are merely extensions or improvements (Kuratko and Hodgetts 1998).The ability to develop new drugs, as reflected in the number of newly registered patents, to cure chronic diseases conjures greater value than mere improvements to existing medications, whose benefits may only be marginal.The value implication of patents is in the challenges posed by the biotech industry in calculating the value of intellectual assets, primarily because of their intangibility.And the very reason why firms invest in intellectual assets is to gain rewards from their use in the knowledge economy (Hovey 2002).

Quality of Management (MQ)
Other than be able to discover new pharmaceutical solutions or adopt effective innovations, the successful biotechnology firm needs to understand and manage all aspects of a firm's operations (Ulrich 1995) and be able to find a home for the newly discovered products and services in the marketplace (Day 1999).Key management factors such as 'access to skilled personnel' (Gustafsson 2000) and 'centers of scientific excellence' (Senker 1998) have been identified as an important value drivers in the biotech sector.Therefore, management must be equally proficient in controlling the critical success factors, costs of development, securing sources of capital, managing the marketing effort, financial management, and every other aspect of management that will allow firms to survive and provide an above average return on capital for investors.
The key to effective management in the biotechnology sector also lies in the effective control or management of costs and risks (Weil and Cangemi 1983) inherent in developing new drugs.Further, as the industry enters a new and more mature phase, and it becomes evident that previous, unrealistically high returns will no longer be available (Tilling 2000), management will have to formulate strategies of converting laboratories into businesses that can fight for market share and profit.In other words, the industry has begun to acquire the characteristics of a mature industry with its inherent problems of competitive pressures and changing investor expectations of risk and return.Primarily, there must be the ability of firms to generate and commercialise a steady flow of technologies to treat known and emerging diseases.

Market Risk Premium (MRP)
This factor measures the additional return over the risk free asset to compensate investor for assuming the average market risk.The market risk is that part of an asset's risk that cannot be diversified away and is also known as systematic risk, which contributes to the riskiness of an investment portfolio.
Like firms in other industries, major risks for nascent biotech firms can arise from failure of R&D, cash flow liquidity (Arojarvi 2001), competitors (Levy 1998), breaches of patent laws (Hovey 2002), technological uncertainties (Senker 1998) and safety fears (Razgaitis 1999).There is a systematic risk component associated with the cash flows of technology-intensive ventures while the technical risks are idiosyncratic (Oh 2001;Berk, Green and Naik 1998).In this research, the market risk premium is used as a proxy for the biotech sector systematic risk factor.The MRP here is assumed to be representative of the market risk inherent in the biotech firm, be it funding, competition or other classes of risk.For biotech firms operating in a high market risk premium environment, it is imperative that they attempt to reduce the variability of cash flows, especially in start-ups, to minimise its cost of capital thus maximising firm value.By reducing uncertainty, biotech firms are able to create a steady state business that would help to reduce underinvestment and overinvestment problems that firms normally face when making investment decisions.Those biotech firms facing acute capital rationing or high costs in delaying or altering their investment strategies tend to benefit most from minimising risk through active risk management and thus create firm value (Oh, forthcoming).
The relevant and definitive risks affecting the valuation of individual biotech ventures, both systematic and unsystematic, need to be determined, measured and fully addressed in the evaluation process in the form of risk premia earned for firm external factors (Oh 2001), during development (Berk et al. 1998) or human capital Darby et al. (1999).

R&D Expenditures
The government's legal, R&D and infrastructure support is regarded as an important value driver, primarily because it reduces the costs and risks of developing intellectual assets (Westland 2002).In many countries, there are active lobbies that seek to maximise the availability of government support for the development of intellectual capital (Razgaitis 1999).
The research and development cycle is probably the most critical phase in the entire life of a biotech product.This cycle determines to a large extent the market success or failure of a product (Chase and Aquilano 1985).However, in the dynamic business environment of the biotech sector, there is an established need for shorter turnaround time.That is, for any high-technology, market-driven innovative products, there is a window of opportunity for market exploitation that is constantly shrinking in length as competition brings new products, and more frequently (Levy 1998).In these circumstances, governments' help in the combined form of infrastructure support, provision of cash grants and sources of cheap development finance, and tax support, will not only help firms in achieving quicker turnaround times, but also reduce the alarming cash burn rate, which often send firms into bankruptcy (Meyer & Utterback 1995) The BVMs and their Explanatory Power The multivariate regression models of Equation ( 5) and ( 6) depict the interactions of 4 and 3 explanatory variables, respectively, acting on the dependent variable, biotechret.
In the four-variable BVM of Equation ( 5), these explanatory variables are the value drivers P, MQ, MRP and R&D, which with the exception of MQ have a negative correlation with the dependent variable.In Equation ( 6), these explanatory variables are the value drivers P, MQ and MRP, where only MQ maintains a positive correlation with biotechret.The implications from the research findings from a strategic management perspective would be to practise prudence in monitoring and evaluating the changes to these pervasive value-drivers in strategic decision-making.
All the specified value drivers in our regression outcome have one common link in risk implication.For the biotech firm, the higher the risk associated with revenue the greater the impact it would have on current market value.The final BVM (Equation 7) specifies the prime value driver as the MRP or the market risk premium, which is proxy for the inherent risk of a biotech firm operating in that industry.This does not undermine the significance and quality of the other value drivers but does highlight where management should focus more attention and resources.It is conceivable that value drivers will change over time due to changes in market conditions.As discussed earlier, the biotech return is used as the proxy for biotech corporate value and the specified negative correlation for MRP is consistent with the relationship between risk and return, i.e. the higher the risk the higher the discount factor and lower the value.
Introducing new biotech technologies into the market has its fair share of risk, and may even be considered extremely risky in certain circumstances.Managing the project life cycle from beginning to successful commercialisation of the new technology entails difficult decisions that involve risks and determine the future course of the program as well as the firm's future revenue and profitability.This means that it is far more difficult to estimate cash flows, growth and discount rates for new biotech than in more traditional and stable firms.The concept of risk in finance underlies the consequences of undesirable outcomes and their implications to individual investors or firms.The need to understand the nature and source of potential risk is to make effective management decisions to create value for the firm.The value of the firm is represented by its market value and it is affected by the amount of uncertainty (i.e.risk) in its future cash flows.In this context, risk can be viewed from two perspectives, firstly the firm-specific or idiosyncratic risk and, secondly the portion of total risk that cannot be diversified away -the market or systematic risk.
From our research, it appears that risk is the dominant factor and major concern of management.The ability to eliminate risk, or even avoid it, is very nearly impossible, particularly so in intellectual capital investments because of their characteristic of intangibility.From a valuation perspective, it is important to understand, assess, and incorporate risk into the investment evaluation process as accurately as possible.In respect of technology valuation, four different categories of operating risk (Razgaitis 1999) should be observed.The inference for management is that relevant risks affecting the valuation of biotech investment need to be determined and measured in the evaluation process considering the risk premia earned for these risk factors such as the NASDAQ composite index, level of consumer confidence and foreign exchange rates (Oh 2001), during technology development (Berk et al. 1998) or the firm's human capital Darby et al. (1999).Once the relevant risks faced by biotech firms have been identified and quantified, they can be assessed and estimates derived with respect to price sensitivity, development costs, potential market size, competition and other factors.

Conclusion
This study attempts to identify and measure the pervasive factors that contribute to driving the value of new biotech firms.Statistical manipulations such as logit and multivariate analyses are conducted on the relevant value-driving factors to identify and estimate the pervasive factors with the use of regression analysis to model the relationship between value of the biotech firm and the pervasive factors.The pervasive factors are then statistically tested for significance and the economic rationale and implications of their value characteristics vis-àvis the broader biotech sector are explained.

Major Findings and Implications for Biotech Valuation
The final BVM model (Equation 7) suggests that the fundamental and crucial value driver for the biotech sector firm is the level of risk premium in the market.In our study this is represented by MRP and is the difference between the market return and the risk-free rate.A biotech firm's ability to manage risk affects its corporate value or share price and this is proposition is consistent with studies conducted by Cooper et al. (1997), Gustafsson (2000) and Oh (2001).When evaluating investment in new biotech firms, the price is normally justified from a fundamental perspective but the significant challenge to the investor is be able to estimate risk from further development, which may lead to success or failure.So risk management can create, sustain or destroy shareholder value and how a well company manages its risks eventually decides its worth.This requires a good risk management process to allow the firm to exploit opportunities for future growth while protecting the value already created.Through strategic risk management the value drivers that are considered vital to the success of the venture are protected and the firm's value is enhanced.
In terms of risk and return relationship, the biotech sector differs in performance to the general market as shown in Table 1.The average return of the sector was relatively higher than the market return but subject to lower risk that that exhibited by the market in the study period.This is an abnormality that could be attributed to the high unsystematic characteristics on many high technology ventures (see Islam and Oh 2003 on systematic and unsystematic risk characteristics).

Limitations
The time series data available in for conducting this study have been limited in that the data was for 11 years only, not all data used were from the same source and some had to be estimated using proxy data.The evaluations carried out in this study had not taken into consideration the lead, coincidental or lag effects of the variables.The exploratory nature of this research means that this aspect of economic analysis has been overlooked and should be addressed for a more definitive research outcome.This study addresses biotech value from the perspective of the organisation rather than the shareholders.The purpose of this study is to tease out the organisational value drivers and shareholders' interests are presumed to be represented by maximising value and maximising return.The situation of information asymmetry and vagueness of early-stage innovation in new biotech firms gives the approach adopted in this research a degree of relevance in that it would be relatively more difficult for outside investors to fully appreciate the value aspects of the venture.The interactive influence of explanatory variables on biotech return may not have been fully understood and appreciated in this research such as the influence of each explanatory variable A on biotech return/explanatory variable B correlation.A more advanced study using "partial correlation" would provide a better insight into this aspect of the interactive relationships of the explanatory variables.
These limitations provide the opportunities for future research and improvement of the BVM.
of agreement to a question of j and the degree of agreement can be seen as a proportion which is a Bernoulli variable.Its distribution is specified by probabilities P = (Y = 1) = π for success and P (Y = 0) = (1-π) for failure.Therefore we can now use logit model and