Single children 'little emperors'

A study recently published in Science by four Australian scholars (myself, Professors L. Cameron and L. Gangadharan of Monash University and Associate Professor N. Erkal of University of Melbourne) suggests that individuals who grew up as single children as a result of the strict implementation of the family planning policy in 1979 are, on average, less trusting, less trustworthy, more risk-averse, less competitive, more pessimistic and less conscientious.

The research is based on data collected from economics experiments involving 421 people born in Beijing just before and just after the introduction of the family planning policy, under which most urban families are required to have only one child. The samples spread evenly across the (pre-policy) 1975, 1978, and (post-policy) 1980 and 1983 birth cohorts, with gender balance within the groups. Surveys to elicit personality traits were also used.

The behavioral consequence of not having siblings has been of interest in developmental psychology for many decades. It is believed that pro-social development is shaped by parents as well as through social interaction with peers, including siblings. The difference in relationships with parents and lack of interaction with siblings have been identified in psychology as two reasons for a single child to develop differently from those having siblings.

However, if parents give birth to single children differ in their personality and behavioral traits from parents who choose to have more than one child, such differences may be inherited by their children. Thus, behavioral difference observed from simply comparing children with and without siblings would be a combination of the effects of growing up with or without siblings as well as the differences in personality traits children inherit from their parents.

To tease out the inherited behavioral differences, we used an econometric method called "instrumental variable" approach. It allowed us to use the family planning policy as an exogenous shock to individuals' fertility to identify a group of people who would otherwise have grown up with siblings but because of the family planning policy did not.

This is important because it allows us to compare the two groups - whose parents have almost the same personality and behavioral traits - and attribute their behavioral differences to the sole fact that one grew up as single children while other did so with siblings. The average difference in the behavior of the two groups tells us the effect of growing up as single children as a result of the family planning policy.

The four experiments (the Dictator Game; the Trust Game; the Risk Game; and the Competition Game) we conducted were standard games from economics literature. For example, in the "Competition Game", which investigates participants' competitiveness, participants were asked to add up as many sets of five two-digit numbers as possible in five minutes. The numbers were randomly generated and presented in rows, and the participants were asked to choose between two payment schemes.

Option 1 was a piece-rate that paid 5 yuan for every sum correctly completed. In Option 2, payments were determined in a competitive way. Each participant was randomly and anonymously paired with someone else in the room. He/she was paid 10 yuan for every sum correctly completed if he/she had completed more sums than the person with whom he/she was paired, 5 yuan if both participants completed the same number of correct sums, and 0 yuan if he/she lost the competition. In the competition game, many fewer single children par-ticipants chose to compete than those born before the policy (44.2 percent versus 51.8 percent).

Our experimental results show that people who are single children because of the family planning policy are likely to be less trusting, less trustworthy, more risk-averse, and less competitive. The post-experiment survey also shows that they are less likely to be optimistic and conscientious.

Some may challenge the findings, arguing that 421 observations cannot represent the cohorts of Beijing residents we are trying to represent and the findings from Beijing cannot be extended to other cities.

In statistics, the sample representativeness comes from randomness of the sampling. Conditional on random sampling the smaller the sample, the less likely it is to get statistically significant results. But the fact that we could get statistically meaningful results with 400 samples suggests that the results will be more significant with a larger sample size. We have described our sampling method in the Science paper in great detail, and we believe that our sampling strategy will ensure skeptics that our sample selection was indeed random.

We chose Beijing for our study because it is a city where the family planning policy was (and is still being) enforced most strictly. Thus, use Beijing allows us to get a cleaner "one-child policy" effect. We understand that Beijing is much richer than other cities. In the study, we did controls for education levels, which are often used as proxy for family income, and current family income levels, but they did not affect our results.

We also would like to stress that although relative to the pre-family planning policy cohorts the post cohorts are significantly less likely to have pro-social behavior, less risk-loving and less competitive, this does not necessarily imply that Chinese on average are less pro-social, less risk-loving and less competitive.

After all, currently the size of rural hukou (house registration) population, the majority of who didn't grow up as single children, is twice as large as the urban population. Among the group born after 1980, the urban hukou population is only one-fourth of the rural population. Our finding is confined to people with urban hukou, which may not be regarded as "average" Chinese. Furthermore, even compare urban post family planning policy generation with their counterparts in other countries, we still cannot conclude that Chinese are relatively less pro-social, more risk-averse and less competitive.

In fact, there is evidence to suggest that relative to people in other countries, Chinese on average are more trusting. For example, based on the 2007 World Value Survey, an article in The Wall Street Journal said 49 percent of the surveyed Chinese citizens believed "most people can be trusted" compared with 39 percent in the US.

Traditional Chinese culture is very altruistic and family oriented. It has always regarded hard work as a virtue. Traditions going back to thousands of year are not easily wiped out. However, the fact that our study has revealed that the strict family planning policy has significantly changed pro-social behavior of China's younger generation is alarming. We sincerely hope that the Chinese government will take into account our findings when considering whether to relax the family planning policy. Otherwise, overtime, it may have significant ramifications for Chinese society.

The author is a professor of economics at Australian National University, Canberra. The article is based on her essay and interview with China Daily's Zhu Ping.