First Part: Managerial Perspectives on Risk and Risk Taking.

Topic 1 :Risk and Risk Taking

What is the intiutive meaning of "risk"?

Intuitively, risk is a hazard or a danger, an uncomfortably high possibility that something bad will happen. It is not expressed in terms of probabilities or expected values but in terms of words. For some it is how much money you have invested in an idea (or stand to lose if it fails).

Managers think that risk is not the same as gambling. They don't consider probabilities and often ignore unlikely events.

How is that meaning translated into definitions of risk in the literature on decision making and risk taking?

In decision making risk is seen as the variability in int he distribution of possible outcomes, likelihoods, and subjective values. Or the non-linearity in the perceived utility of money. Or the variance in distribution of possible gains or losses associated with a particular alternative. Or as the expected value or return of an alternative.

How does it happen that risk is sometimes defined as non linearity in the utility function of money, sometimes as variability in the outcome distribution conditional on a choice, and sometimes as the likelihood of a specific adverse outcome?

Partly it depends on which field is making the interpretation of risk. Statisticians may view it as variance in the conditional probability. Economists, who assume people calculate and use utility functions assume people may change their perceived utility of an option if too much money is invested in it. Crisis management or actuaries might focus on the likelihood of a specific outcome.

What difference does it make?

Not much in a sense (it's all based on the same thing). The ability to calculate outcome values and probabilities may be easier in some situations than calculating utility functions. In others where you are only focusing on one specific outcome, the likelihood may be more important. Other times the magnitude of the outcome may be critical.

What is the role of risk taking in learning?

Risk taking increases the probability that one will find something of great value, but at a search cost. While the expected result might be negative, it creates the possiblity of large gain in skill or capacity or expertise or quality. Those who do not risk may not learn as fast.

In particular, what is the connection between risk taking and exploration?

Taking more risk means that one samples from an alternative pool of higher variance. While there is an increased chance to find nothing in an explorative search, there is also a probability to find something new and innovative.

Topic 3: Risk Preferences:

What are risk preferences?

Risk preferences are individual tendencies to evaluate variation in expected outcomes. Risk-averse people are willing to sacrifice soem expected value to reduce variation. Risk-seeking people are willing to sacrifice some expected value to increase variation.

What is a "trait" conception of risk preferences?

Standard theories assume that risk-preferences are stable and a factor of personality.

A context-dependent conception?

Researchers have found risk taking is more situational, depending on how the risk was framed and what is being risked (e.g., business vs personal money).

Most people are seen as being risk-averse.

What are the predictions of prospect theory with respect to risk taking?

Decision makers are more risk averse when returns are expected to be above target than when they are expected to above target.

People are also more likely to take risk far above a target (high slack) and take little risk when they are near the survival point.

Of theories of innovation and creativity?

In technologically mature worlds, success will tend to breed failure. There are few alternatives left to evaluate, so excessive search will be unproductive. However, in technically young worlds increased search will potentially uncover innovations that allow the organization to continue its success.

Increased success creates slack which is used to search for innovation.

The rich will get richer if slack search produced more returns than failure-induced search.

Of ideas about threat/rigidity?

If an organization must make decisions that threaten its survival, it tends to become more rigid and risk averse.

Topic 4: Exploration and Reliability

What is the relation between exploration and reliability?

Successful exploration produces slack which reduces reliability. Slack decreases in bad times and increases reliability (through less exploration and more exploitation).

In general, reliability tends to increase education and experience.

What is the role of ignorance and foolishness?

Ignorance is a source of variablity. Lack of communication, trust, structure, etc. can increase risk unconsciously. Knowledge increases mean performance and the reliability of the outcome, which reduces risk taking.

Foolishness usually occurs during high slack times. It can either reduce efficiency and performance or produce innovations that cause continued success.

What are the organizational factors that influence risk taking and exploration?

Success-induced Bias

People are more likely to attribute success to ability and failure to luck. Success is deserved and failure is a manifestation of risk. Sucessful people tend to underestimate risk, and unsuccessful people overestimate risk.

Successful organizations build a "can do" attitude that leads people to underestimate risk. People who are by intention risk-averse may actually be risk-seeking in behavior.

Failure-Induced Bias

People who fail overestimate risk, take less chances, and thus reap less gains, causing a self-perpetuating loop.

Failure to Estimate Extreme Probabilities

Organizations who have never experienced a failure will underestimate its occurance and "relax its guard" , causing large motivational impact when it happens. Fortunately, this will increase vigilance and overestimate risk, causing it to become more balanced.

But organizations who experience failure (e.g. R&D) will overestimate the chance of failure, reduce search, and not find a discovery, further reducing expectations and search and making things ever worse.

Second Part

Consider the discussion found in James G. March, A Primer on Decision Making: How Decisions Happen (New York: Free Press, 1994), pp. 49-52 In particular, specify precisely the model on page 51 and show how the predictions sketched there follow from the assumptions.

A. Assume a hierarchical organization, where reputation is accumulated over a series of performances and promotions based on performance. When there is a vacancy, the person with the highest reputation is promoted.

B. When reputation is earned over many samples (at the same level) promotion is primarily based on ability. Random fluctuations are smoothed out on the average.

C. When reputation is based on only a few samples, random luck plays a bigger part in getting the promotion. If the hierarchy is steep (few slots in the above level), promotion is more based on luck.

What are the implications?

In part C, The result is a slight increase in ability as one moves up but a much higher risk preference (becaused it was the risky people who got lucky that were promoted).

D. Thus the procedure that appears to promote people on the basis of their abilities actually moves them ahead on the basis of the amount of risk they take.

This may cause in competitive hierarchies more and more risk taking as people try to get ahead. People with high ability may then favor organizations with high sample rates to get their just reward, while people with low ability may prefer the opposite.

The limitations?

In most situations people gain experience, knowledge and skills at the higher level, which should increase average performance, increase relability, and reduce risk taking.

Also, organizations can "pool" experience across people to increase average performance and ability via rules and procedures to ensure such knowledge is transferred to newer members.

Limitations of Increasing Reliability

But this assumption caused the average risk taking to fall as the organization ages, which could hurt the organization in competitive situations where some risk taking is necessary, especially when the number of organizations is large.