A research stream known as prospect theory describes how decision biases lead to results that differ from those predicted by classical utility theory (Kahneman and Tversky, 1979). Prospect theory hypothesizes that individuals will experience potential losses more intensely than potential gains, and will be more risk-seeking in loss situations, while more risk-avoiding in gain situations. This study includes 948 participants from the PRC and 318 students from the USA. All of our attempts to replicate these findings in the Peoples' Republic of China have revealed a different pattern. Chinese subjects consistently demonstrated risk-seeking preferences, both in gain and loss situations.
The impetus for this study came somewhat serendipitously. In 1999 the Project Management Institute (PMI) requested the first author to discuss project management practices with groups of upper-and middle-level government and business managers in the Peoples' Republic of China (PRC). The Chinese were interested in bringing PMFs project management professional (PMP) certification to China. The first author conducted a prospect theory exercise during several discussions in the context of describing western management communication ideas. Results did not conform to predictions. Prospect theory didn't seem to work in China, although the first author had always found strong support for the theory when using this exercise among experienced business people in the USA. Attempts during several subsequent trips (20+) to the PRC yielded similar findings. Prospect theory hypothesizes that individuals will experience potential losses more intensely than potential gains, and will therefore be more risk-seeking in loss situations and more risk-avoiding in gain situations. In China risk-seeking seemed to dominate in both gain and loss situations.
Why didn't Prospect Theory seem to work in China? Was this an artifact of the informal exercise presentation or do the Chinese have different dispositions toward risk-taking? The first author then contacted the second author to more formally study this issue and to gain a Chinese perspective when conducting this research.
We designed Phase One of our study to test prospect theory in a more formal and controlled setting using a sample of 300 Dongbei University of Finance and Economics (DUFE) students in Dalian, PRC. As shown in the third section of this paper, Phase One results confirmed risk-taking preferences in both gain and loss situations, contrary to prospect theory predictions, but consistent with the initial informal findings. Given these unusual findings, Phase Two of this study attempted to replicate these findings and to more comprehensively test possible explanations. A separate sample of 606 PRC students and 318 students attending the University of Scranton (USA) was studied. These Phase Two results are also shown in the fourth section of this paper. The final section of this paper discusses the finding of both phases and draws some conclusions, not only about this study, but about cross-cultural research in general. Avenues for future research are offered.
THEORETICAL BACKGROUND AND LITERATURE REVIEW
Kahneman and Tverksy's (1979) seminal work on prospect theory proposed an important alternative to the classical expected utility model (Von Neumann & Morganstern, 1947). Prospect theory is descriptive. It attempts to describe how people make choices under uncertainty. Some important aspects of this theory and how they differ from expected utility theory, as suggested by Thaler (1980) and Tversky & Kahneman (1992), include:
1) Decisions are made based upon a value function not a utility function;
2) The value of a prospect is measured in terms of change in wealth (gains or losses) not in terms of wealth position;
3) Gains and losses are conceptualized with respect to a reference point;
4) Losses are experienced more intensely than gains (the slope of the value function is typically steeper for losses than for gains); and
5) There are diminishing returns to value (the value line is concave for gains and convex for losses). …