Risk taking in research and development activities and patenting after gain or loss experiences

Table of Content:

1 Introduction

2 Motivation

3 Definitions

4 Explanations for Entrepreneurial Risk taking
4.1 Agency Theory
4.2 Expected Utility Theory
4.3 Risk Return Paradox
4.4 A Crisis as Incentive Constraint for Innovativeness
4.5 Prospect Theory
4.5.1 Reference Points and Aspiration Levels
4.6 Mental Accounting
4.7 Starting Points´ Effects on Risk taking Behavior

5 Questionnaire Experiment
5.1 Mode of operation of the Questionnaire
5.1.1 Part A
5.1.2 Part B: Certainty Equivalent
5.1.3 Part C:The socio-demographic Part
5.2 Evaluation of the Data
5.3 Multinomial Logistical Regression
5.4 Results

6 Conclusion

7 Table of Literature

8 Appendix

1 Introduction

This seminar paper deals with the descriptive decision theory, which tries to show the actual decision behavior especially of ENTREPRENEURS. The question I particular examine is whether research and development activities and patenting is influenced by prior gain or loss experiences of entrepreneurs. This paper objective is to attain an empirically confirmation for this hypothesis. To make this possible I will turn out the evolutionary steps of strategic choice behavior models and show in an experimental study that prior gain or loss experiences have a considerable and significant impact on entrepreneurial risk taking and, as I will show later, as well on research and development activities.

2 Motivation

To examine entrepreneurial decision behavior under risk and after personal gain or loss experiences, we should take into account that research and development activities, in sense of developing a new product or product idea and the following placing of the innovation within the market, is obviously quite more risky then a `simple´ product-innovation. Empirical studies show that success of innovative product ideas is indicated by exceptionally low success probabilities about 3 - 5 %. (Perlitz / Löbler 1985, p 426) This gives an idea about the fact that research and development activities involve very high risks. If an Entrepreneur further chooses to patent the Innovation or the new product it produces additional costs and thereby increases the risk, because the potential loss if the venture fails is increasing as well. However, patenting is sometimes necessary to protect innovations against failure or to enlarge the possible outcome. But if entrepreneurs only accept increasing risks if there is an increasing expected gain as well, it raises a problem. The R & D activities involve high risks which reduce the expected profit and a small expected profit is not a very good reason for taking big risks. So what is the explanation for this phenomenon?

3 Definitions

To answer this question scientifically correct some conceptualities should be

defined or evaluated in front of the further examination:

Institute for Entrepreneurship / Innovation management Seminar Group C

- Innovation: The word innovation has his origin within the latin language and means renewal or creation of something new (see Gabler, Business Lexica 2000, p 1898). Knight stated: `Innovation is adoption of a chance which is new to an organization and to its relevant environment.´ (see Knight 1967, p 478). A stronger definition is that a new solution is only an Innovation if there are some advantages to the prior solution. (see Geiger / Heyn 1975, p 1604) Innovation means enforcement and execution of an idea up to the market ability. (Perlitz / Löbler 1985, p 426) They classify Innovations into three parts:

- (1) Product innovation: is a new or an improved Product, involving the highest risk;
- (2) Process innovation: is a new or an improved production method , involving less risk then (1) and
- (3) Social innovation is an amendment within the humanism and means the change of norms or changes within the interaction process of individuals. (Perlitz / Löbler 1982, p 427)

Therefore we can focus the further examination on R & D activities and patenting of organizations since it is shown that an innovation is per definition the result of R & D activities. The equal is valid for Patenting. It means if we analyze the innovativeness of enterprises we analyze R & D activities as well (under the assumption that there is a significant positive correlation between them).

- Risk: Risk is something which endangers the process of target level setting and target attainment and has a negative influence on entrepreneurial decision making. (see Gabler, Business Lexica 2000, pp 3279-3280) Risk is a concept that captures the uncertainty or probability distribution associated with the outcome and it is relevant before the occasion (i.e., ex-ante) and it essentially must be measured in actual empirical work after the occasion (i.e., ex-post), because only one of the possible outcomes is the result. Risk is measured either by nonlinearities in the revealed utility for money or by the variance of the probability distribution of possible gains and losses associated with a particular alternative (Pratt 1964; Arrow 1965). Risk is often defined as variance in returns over time. The most common kind of measure is the return on Equity, but the return on total capital is also often used. (Bowman 1982, p 34) In classical decision theory, risk is perceived as reflecting variation in the distribution of possible outcomes, their likelihoods, and their subjective values, but risk can also be seen as a danger or hazard, which means that risk is only associated with negative outcomes. (March / Shapira 1987, p 1404) A more general definition is that risk and uncertainty are synonyms, in which uncertainty is the incompleteness of information (Perlitz / Löbler 1985, p 426).

In the further examination I will concentrate on the variance of the expected outcomes under the assumption that enlarging the variance leads to higher perceived risk. We use this definition for the design of our questionnaire experiment.

4 Explanations for Entrepreneurial Risk taking

4.1 Agency Theory

Moral Hazard Problems always arise if the decision maker deals with other people’s money. It means that the decision maker has not the same incentive structure like the owner of the money. If the venture fails and the money got lost the decision maker may lose his job and in follow his wage, but this is mostly not equal to the money he ´burned`. Some interesting Principal Agent relationships are manager - shareholder, entrepreneur - employee, enterprise - investors, manager - bank, and so on. Translated to our examination context does it mean, that managers which are not liable with their own money, act maybe different in risk taking to those who `gamble´ with their own money. In the opposite they don’t have the same incentives to enlarge the profit of the company like e.g. shareholder, because he has to work harder and don get the money he additionally `produce´. Moral Hazard is a problem which complicates the examination of risk taking behavior.

4.2 Expected Utility Theory

The most important normative model for decision behavior under risk, developed by Neumann and Morgenstern (see Eisenführ/ Weber, 1999, p. 220), is the expected utility theory (EUT), which is said to be able to characterize the optimal behavior of individuals (see Thaler, 1985, p. 200). The normative EUT describe how people should behave according to rationality axioms and their preference orders. One of the key assumptions of the EUT is that decision maker are risk- averse. (see Fiegenbaum 1990, p 189) The proposition of the theory is that a utility value is calculated for each alternative choice. This means that the expected utility of an arbitrary alternative Ai arises weightedly by transformation of her results Xij by using values with the probabilities pj with which it comes to the environmental states Sj which are adapted if it comes to the possible environmental states Sj. Regarding to this statement people choose the best Alternative A which fulfills the condition Max E (U (Ai)) or in other words it is assumed that an individual choose the alternative A out of a set of Alternatives Ai (i=1,..,m) which maximizes its utility for the same level of risk (Fiegenbaum / Thomas, 2004, p. 85). The necessary rationality Axioms are completeness, transitivity, continuity, independency (Substitution axiom) and the reduction of composited lotteries. (Herstein / Milnor, 1953) Later it is shown that the preferences of decision makers systematically infringe the rationality axioms. (see Kahneman / Tversky, 1979, p 265) The most relevant paradoxes which avoid the rationality axioms are the Endowment Effect (Thaler 1980), the Allais Paradox (Allais, 1953), the Elsberg Paradox (Elsberg, 1961), the Probability Bias, the Intransitivity of Preferences, the Preference Reversals and the Reflection Effect (see Kahneman / Tversky, 1979). Due to these facts EUT is not able to describe the individual choice behavior under risk (see Kahneman / Tversky, 1992, p. 297) and it as well not suitable to provide a good explanation for our phenomenon.

4.3 Risk Return Paradox

This part deals with the relationship of financial risk and return in Organizations. In general one should assume that the relation between the mean yield (ROI) of an enterprise and the business risk (variance) should be positive! It means that additional risks are only acceptable in connection with increasing chances of winning. This type of risk averse behavior is obviously expected of companies, managers and investors (see Bowman, 1982, p 33)

Earlier statements claim that entrepreneurs and manager weigh risk and return together (see Conrad / Plotkin 1968, p 90). Amour and Teece state that economic theory suggest the rate of return associated with a particular asset is a function of the risk inherent in the asset, and (assuming risk aversion) the higher the risk, the higher the expected return. (see Amour / Teece 1978) This shows that business administration as well as economic theory and literature maintain that there is a positive relationship between risk and return. Since the work of Bowman (1980), the assumption of generally positive risk-return relationships has been challenged (see Lehner M., 2000, p 63). Bowman shows in an empirical study, contrary to the earlier cited findings, that business risk and the expected return are negatively correlated across companies and within industries (see Bowman, 1982, p 33). The results of both experiments support the statement, that there is a negative correlation paradox, in which the industries show no significant correlation and the companies within industries show a significant negative correlation (see Bowman, 1980, pp 23-24). The essence of his findings is that high average-profit companies tended to take lower risks (Bowman, 1980, p 19) and that low profit companies (troubled companies) tended to act risk prone. (see Bowman, 1982, p 33). Another study support these results and shows that if a target value is established at the industry level, risk and return are negative correlated for below target level firms (risk seeking) and positive correlated for above target level firms (risk averse). (see Fiegenbaum / Thomas 1988, p 93-97) A further study offered as a possible explanation for high risk taking that, because of the diminishing sensitivity of the prospect theory value function for larger amounts (losses), people tend to “gamble with bankruptcy” (see Thaler / Johnson 1990, p 643) If we take these results and combine it with our assumptions we can derive that higher average profit companies tend to be less innovative, in sense of less R & D activities and that troubled firms have a higher willingness to innovate and take higher risks although the expected return is small.

In combination with other studies who dealt with risk and return relationship [e.g., March (1988), Bettis (1981), Singh (1986)] we can conclude that the characteristics of the firms´ environment, strategy and implementation processes have a significant and important impact on the firms risk level and its associated rate of return (see Fiegenbaum 1990, p 189). These findings support the basic propositions of prospect theory. (see Fiegenbaum / Thomas 1988, p 85) Bowman state, that a potential explanation for the negative correlation paradox could be, that within an industry good management will bring out both, higher returns and lower financial risk. (see Bowman, 1982, p 35) And manager state about themselves that they take only well calculated risks and they are able to reduce the risk by using skills to control the dangers. (see March / Shapira 1987, p 1410)

4.4 A Crisis as Incentive Constraint for Innovativeness

Other authors discussed the statement that an economical crisis is a necessary incentive constraint for the willingness to innovate and thus as well for the risk prone behavior of entrepreneurs.

Schumpeter approves this declaration and argued that a crisis is the triggering moment for innovations. (see Schumpeter 1947, p 149) This hypothesis is also supported in a study done by Perlitz and Löbler. They show that a crisis has a considerable initial effect for process- and product innovations. Furthermore they argue that the classical risk return paradox is not able to explain risk taking behavior of enterprises or entrepreneurs, because it takes only the chances of success into account and this is not enough to make clear why high risk involving innovations are realized. Within the carried experiment the individuals (230 managers) are confronted with two possible environmental states. The first one was a chance situation and the second one economical crisis. It is shown that most of the individuals within the situation of chance prefer the safe option and within the economical crisis the risky innovation instead of the safe option. In addition they could proof that people prefer the less risky process innovation (with 25 % probability to lose) toward the quite more risky product innovation (with 75 % probability to lose). They argue that these results stay in harmony with the statement of Schumpeter and relativize the classical risk return approach. Their results make clear that the incentive system of managers or entrepreneurs has to be seen in relation to their environmental background which significantly influences their willingness to innovate. (see Perlitz / Löbler 1985, pp 424-432) Out of these statements we can derive that R & D activities and patenting depends on environmental states and therefore assume that in case of a crisis the R & D activities increases. The reason is, according to the foregoing study, that managers or entrepreneurs see the innovation as a chance or a `straw´ to overcome the crisis and in follow to break even. Due to these facts their willingness to innovate increase and therefore they will take higher risks as they would normally accept. These assumptions could be a possible explanation for our phenomenon as well.

The next model considers both theoretical approaches through replacing of the utility function with the psychologically richer value function, (see Thaler 1985, p 201) which is able to illustrate entrepreneurial decision behavior under risk.

4.5 Prospect Theory

Kahneman and Tversky offered a new alternative model called prospect theory as a critique of the expected utility theory. It is a model of mathematical psychology which takes perception phenomena in context-depending decisions into account. Therefore this model is able to represent real decision behavior. They state that if people choose among risky prospects, they underweight merely probable outcomes in comparison with certain outcomes. This tendency, the so-called certainty effect, contributes to risk aversion in choices involving sure gains and in the opposite to risk seeking behavior in choices involving sure losses. (see Kahneman / Tversky, 1979, p 263) If people choose among a sequence of choices, they ignore the final expected outcomes and assign value to gains or losses. This leads to a value function which is normally concave for gains and convex for losses, in which the convex part for losses is generally steeper than the concave part for gains. This is identical to a leaning S-shaped curve, with either end of diminishing incremental importance or diminishing sensitivity for higher values.

[...]