EJBO - Electronic Journal of Business Ethics and 
Organization Studies

Vol. 12, No. 2
ISSN 1239-2685
Publisher: Business and Organization Ethics Network (BON)
Publishing date: 2007-11-12

ARCHIVES (2004-)
ARCHIVES (1996-2004)

Moral choice within the financial sector of the Shenzhen Special Economic Zone, People's Republic of China

By: Gordon Woodbine [biography]


The following paper examines the classical agency assumptions as they might apply within a little researched region, a Special Economic Zone within the People's Republic of China. It was found that the predictive strength of the theory was again proved within the context of an experimental design model, which involved 400 employees within ten large financial sector organizations.

When faced with a significant moral hazard and the possession of private information, financial sector employees displayed a preference not to advise management about the existence of unethical practices within the workplace. An analysis of the data demographics identified female employees within the 25 to 30-year age bracket, who do not have supervisory responsibilities, as the group most at risk in this situation. It is apparent that levels of adversity to risk change depending on an employee's personal demographics


An agency relationship exists when one party, the principal delegates actions and decision making to another party, the agent. This arrangement is extremely common in all areas of accounting, finance and economics. Agency theory provides the framework for studying these kinds of relationships and extends and formalizes ideas on conflict of interest between organizational stakeholders (Jensen and Meckling, 1976; Fama and Jensen, 1983). In theory, the principal assumes the agent will do exactly what is required, according to the terms of an agreement or contract. The principal operates through other people to achieve some goal, which they are incapable or indisposed to complete themselves. Economists would define the goal as profit maximization. In return, the agent is compensated for his or her effort, called an agent's fee or salary (where the agent is an employee).

Economists also argue that the agent is an "unconstrained self-interest maximizer" (Chi, 1989) who will take any opportunity to put his or her own interests before that of the principal. This phenomenon is called the "agency problem" (Eisenhardt, 1989). The problem manifests itself in many forms, ranging from laziness and lack of attention to a task through to corruption of the worst kind (e.g. bribery, theft etc.). Eisenhardt and others claim that the agency problem is created for two reasons:

    (a)The agent's activities are generally unobservable to the principal, inducing the agent to shirk. This is termed a moral hazard.
    (b)The agent has access to private information that affects his or her position and which may not be in the agent's interest to divulge. This places the parties in a position of information asymmetry.

Private information can vary from something as insignificant as knowledge by the agent that he or she can complete a job in fifty minutes, but be paid for sixty minutes' work, through to something serious, such as a prior knowledge that he or she did not possess a formal qualification to undertake a task, of which the principal was ignorant.

Objective of research

The purpose of this research is to examine the effects of the agency problem within a particular business environment - the financial sector within the Shenzhen Special Economic Zone of China. This sector was selected for two reasons. Firstly, this region of China is one of the fastest growing areas, and an area that the Central government in Beijing has specially favoured as part of its openness policy. Secondly, the financial sector is itself experiencing substantial reform. Many changes are in process as this important sector prepares for the major effects that arise from China's entry to the World Trade Organization.

The assumption economists make about agents being motivated only by self-interest appears extremely pessimistic as it describes people as opportunistic. There appears to be no room in their theory for morality (Bowie and Freeman, 1992). The fact that most principal-agent relationships are successful would imply that there must be room for trust and honesty and that agents have other interests apart from a desire simply to "get rich quickly". It is simplistic to assume that one's predilection to be unethical is constrained only by such mundane issues as the size of the remunerative package or whether the principal is able to check on the quality of work performed and impose appropriate penalties for inefficient or irregular performance. Professionals in particular, ought to operate without consideration of these issues.

As a predictive model, however, agency theory has demonstrated time and again that self-interest (on the part of the agent) is a major consideration (Bohren, 1998). This study examined the extent to which self-interest exerts itself within an agency context in a selection of financial sector organizations operating in Shenzhen. A total of 400 workers were included in the survey of ten different institutions. Table 1 summarizes the demographics for this sample.

Table 1 provides an outline of the various characteristics of the financial sector operatives included in the sample. In some respects they are quite similar in that as employees of mostly state owned institutions they receive wages and salaries. This is seen as the principal form of compensation bonding workers to their employers (the state), one that hopefully encourages them (as agents) to act in the interests of the principal. To add to this various forms of monitoring is in place, which works to discourage unethical, inefficient and illegal actions. This would tend to vary depending on the circumstances. There will also be differences in the "risk averseness" of workers, depending on their age, education status, wealth, employment status etc..

Table 1
Demographic data for sub-sectors

(Total 400; 100%)State banksCommerc-
ial banks

    Male (165; 44)
    Female (209; 56)
(n*  %)

84  51
79  49
(n*  %)

41  50
41  50
(n*  %)

15  27
41  73
(n*  %)

24  33
48  67
Level of education
    High school (63; 17)
    Diploma (118; 31)
    University degree (148; 39)
    Masters degree (46; 12)
    Doctorate (6; 1)

15  9
59  34
75  46
13  8

3  4
14  17
38  46
23  28
5  6

9  16
21  38
18  32
7  12
1  2

36  45
24  30
17  21
3  4
Employment Status
    Clerk (261; 69)
    Supervisor (64; 17)
    Department manager (49; 13)
    Senior manager (6; 1)

102  63
28  17
28  17
3  2

50  60
22  27
10  12
1  1

47  82
5  9
5  9

62  79
9  11
6  8
2  2

Age (29.8 years)
Business experience (9.0)
Loans experience (3.4)
Securities experience (3,6)
* n = Sample size

A field experiment was devised to test the extent to which the agency problem was likely to affect various groups of workers. All 400 participants were asked to listen to a hypothetical business story. Two versions of the story were prepared and part of the sample was given one of the versions to listen to. Both versions of the story involved a bank officer who was in a quandary about whether to advise his employer about unethical activities being conducted by a group of his fellow workmates. In one version (referred to as Version 2 below) the position was made even more difficult by the fact that the bank officer had withheld information from his employers about his past that might be discovered if an investigation was to occur after he disclosed the unethical activities. His future promotion and job prospects could be affected as a result.

The experiment was based on similar experimental research conducted by Harrell and Harrison (1994) and Rutledge and Karim (1999) who examined the behaviour of managers in the United States who were faced an agency issue that involved moral hazard and adverse selection criteria. In both instances the researchers discovered that managers faced with both elements of the agency problem were more likely to act in their own interests, rather than the interests of their employer.

In this present Chinese survey respondents were asked to make a moral choice by indicating the likelihood of their informing management about the unethical activities described above (assuming they were the bank officer described in the story). Respondents displayed a preference by placing a mark along a continuum, ranging from "definitely not advise management" to "definitely advise management". A copy of the relevant section of the questionnaire is contained in Appendix 1

The principal hypothesis (H11) of the survey can be stated as follows: Financial sector employees, faced with strong adverse selection criteria, are more likely to act in their own interests compared to others whose agency conditions are less demanding. Adverse selection criteria refer to both the incentive to shirk (avoid obligations) and the possession of private information.

Additional tests were also undertaken to determine the extent to which moral choice was influenced by certain demographic factors, such as the age, gender, business experience, educational level and employment status. Each of these variables was examined separately and the general null hypothesis (H20) worded to state: In situations involving adverse selection criteria, the likelihood of a financial sector employee advising management of unethical activities witnessed in the workplace, is unaffected by differences in demographic factors.

Results of agency experiment

The results of the univariate statistical analysis of the data confirmed that, for the sample as a whole, employees indicated they were more likely to withhold information from management about unethical activity in situations where their reputations and future employment prospects were likely to be affected. Table 2 confirms this outcome. This outcome lent support for the major hypothesis (H11) that employees displayed greater self- interest orientations in situations where adverse selection criteria are present.

It is important to note that the demographic information relating to each of the control and treatment groups (i.e., those who were given Version 1 and 2 of the story respectively) was not significantly different, except for differences in age. The statistical analysis was repeated after adjustments for age differences were made and the results of the test were still the same. It can be noticed that the hypothesis proved correct for all four types of financial sector entities, except for employees in the state banks.

Table 2
Effects of agency experiment on moral choice

Version of storyTotal

Version 1 (Some adverse
selection criteria present)

Version 2 (Strong adverse
selection criteria present)










* Differences statistically significant for p<0.01
** Differences statistically significant for p<0.05
*** Differences statistically significant for p<0.1

A univariate independent t-test analysis was completed for each of the major demographics to determine whether differences in mean scores between the two groups (i.e., story version 1 and version 2) might better explain the significant change in moral choice response scores. Univariate statistical tests of the relevant categories are summarized in Table 3.

Table 3
Univariate tests of personal demographics

Control group
(limited moral hazard)
Treatment group
(moral hazard and
private information)

Age groups

25 years and under
26 years to 30
31 years to 35
36 years to 40
41 years to 45
Over 45 years
moral choice score (sample

7.19 (52)
7.77 (85)
7.39 (23)
6.80 (20)
10 (1)
moral choice score (sample

7.55 (19)
6.33 (79)*
6.54 (34)
7.02 (26)
5.88 (8)
10 (1)


7.31 (71)
7.54 (109)

6.68 (85)
6.58 (89)**
Business experience

4 years or less
8 years or less
Over 8 years

7.32 (51)
7.51 (48)
7.32 (57)

6.26 (39)
6.83 (54)
6.81 (68)
Educational level

No tertiary qualifications
Tertiary qualifications

7.71 (94)
7.15 (95)

6.71 (76)**
6.44 (98)***
Employment status

Clerical officer or
Manager or supervisor

7.53 (134)
7.45 (55)

6.24 (111)*
7.26 (62)
* Significant for p<0.01 ** Significant for p<0.05 *** Significant for p< 0.1

Gender statistics
There was little difference between gender moral choice response scores for Version 1 of the story, however, the response scores for females dropped significantly for Version 2. This indicates that females tended to become significantly less likely to inform management about unethical work practices if their reputations and careers were likely to be affected.

Age differences
Employee ages ranged from under 20 years to over 50. Surprisingly few employees in the sample were over 45 years of age (i.e., 3% of total sample). This might reflect that the financial industry itself is relatively young in Shenzhen rather than any industry policy only to employ young people.

The sample was divided into six sub-groups according to age. The largest sub-group was the 25-30 year old employees containing 45% of the total sample. It was also this group that appeared most vulnerable in terms of the agency problem. Employees in this group were much less likely than employees in other age groups to advise management about unethical practices if their reputation and career prospects will be affected.

Education level
For this analysis the sample was divided into two groups: those with no tertiary education and those holding an undergraduate degree or better. The sample was fairly equally divided Mean scores for employees holding tertiary qualifications was somewhat lower although the differences for both the control and treatment groups were not statistically significant (p < 0.05). Both groups displayed a tendency to become less inclined to advise management about unethical practices if their reputations and careers were likely to be affected. The effect of the agency problem on employees with no tertiary qualification was somewhat greater than for the other tertiary educated employees.

Employment status
Two-thirds or 67% of the sample consisted of clerks with no supervisory duties. The balance included supervisors, departmental and senior managers. It was clearly evident from the analysis of the two independent samples that clerks were very much more likely not to advise management about unethical practices if their reputations and careers were likely to be affected.

Discussion and conclusions

The investigation of the sample of financial sector employees working in the Shenzhen Special Economic Zone clearly demonstrates that, like employees in other developed nations, they are likely to be affected by agency issues. The impact of the moral hazard and adverse selection criteria described in this study is significant and the major hypothesis (H11) was supported. Financial sector employees appear less likely to advise management about unethical work practices if such action will affect their reputation and career prospects.

It should also be noted that the impact of the presence of a moral hazard (i.e., incentive to shirk or avoid responsibility) in even the control group (Version 1) was somewhat significant in that, on average, respondents indicated they were less than definite (mean score 7.43) about their intentions of advising management of the existence of unethical work practices.

The hypothesis (H11) was satisfied in three of the four financial sector groups. Only the employees of state banks displayed no significant change in preference about advising management of the unethical activities of colleagues. Four individual sub-samples made up this group and their combined demographic structure was similar to that of commercial banks. However, two of the sub-samples (i.e., a suburban branch included in the control group and a head office branch included in the treatment group) displayed moral choice scores that were at deviance to the observed trend.

A further analysis of the personal demographics associated with the two state bank sub-samples provided further evidence of the importance of the variable, adversity to risk in relation to moral choice. The control group sub-sample was significantly younger (by six years) than the treatment group sub-sample. The average age of the latter was relatively high (34 years) and consisted of mainly male officers with greater tenure and higher employment status. These risk related factors probably had a significant bearing on the relatively conservative choices these officers made, even when adverse selection criteria was present. It is apparent that the outcome to support the first hypothesis (H11) would have also been supported had sufficient allowance been made for these differential issues.

Following the analysis of available personal demographics it was evident that the second null hypothesis (H20) was rejected on a number of occasions. In summary, it can be suggested that financial sector employees between the age of 25 and 30 years, who had not completed a university education, were much more vulnerable to the agency issues, especially those involving adverse selection criteria. Again, females in this category were even more likely to be affected.

Now that it has been possible to identify the subgroups of financial sector employees most vulnerable to the agency problem described in this research, the question that arises is whether the same subgroups are themselves likely to actively participate in unethical activities in the workplace. Because of the nature of the demographics associated with the subgroups, it is fairly certain that a substantial component of financial sector employees fit the character of being particularly averse to risk.

The results of the field experiment cannot be used to make direct predictions about the moral behaviour of individual financial sector employees (numerous other personal and contextual factors need to be considered). However, it can be argued that placing respondents in a passive third party role (as described in the field experiment) avoids confrontation and is more likely to encourage employees to provide objective and realistic assessments.

The research does support the reality that the passive acceptance of unethical work activities is often accepted in the workplace. This position is hardly conducive to the creation of an acceptable work climate especially if practiced by senior management.

In many instances reported in the Chinese newspapers, senior officers are involved in corrupt activities. Cases involving junior staff are rarely reported because they are usually handled in-house by individual institutions or are comparatively less newsworthy. One should take note of statistics recently published by the Beijing Haidian District Procuratorate Office (reported in a recent edition of the China Youth Daily) (Li, 2001), which indicated that the number of white-collar crimes committed by financial sector officers under 30 years of age had increased significantly. Officers in this age group are more educated and apply sophisticated methods to defraud and embezzle. Such reports provide support for the findings in this particular survey.

What can be done to reduce the incidence of unethical activity in financial sector organizations? Researchers (e.g., Snoeyenbos and Jewell, 1983) have advocated that institutions be strongly involved in establishing administrative mechanisms for managing the problem. They suggested three such mechanisms be introduced.

    1. Written codes of ethics that are made available to all staff.
    2. High ranking ethics committees that administer a program that involves providing various services to management and workers (e.g. training, reporting and investigative services, awareness campaigns, employment selection policy)
    3. Extensive training in business ethics

In the ten different institutions included in this survey, senior administrators were interviewed and asked whether any of the above mechanisms had been established. Six of the institutions reported having written codes of ethics, only two were aware that of the existence of high ranking ethics committees although eight reported some form of ethics training. Ethics training included periodic lectures and the issue of written instructions.

The author believes that financial sector management needs to be vigilant in its efforts to encourage employees to become more ethically aware. It is important that the above mechanisms are essential as part of a program aimed at internalizing ethical beliefs. To this extent it is advocated that the institutions strive to establish appropriate ethical climates at all levels of management. The mechanisms listed above are therefore necessary, but they are not sufficient in themselves. As Fritzsche (2000) and other organizational researchers believe, an ethical climate that advocates professionalism will promote good standards. However, it is essential that junior employees witness these ethical standards being actively practiced by their supervisors and managers; otherwise no ethics policy will be successful.


Bohren, O., (1998), "The agent's ethics in the principal-agent model", Journal of Business Ethics, 17(7), pp. 745-755.

Chi, S.K., (1989), "Ethics and agency theory", Unpublished doctoral research paper, (Department of Accounting, Washington University, Washington).

Eisenhardt, K.M., (1989), "Agency theory: An assessment and review", Academy of Management Review, 14(1), pp. 57-74.

Fama, E.F., and Jensen, M.C. (1983), "Separation of ownership and control", Journal of Law and Economics, 26, pp. 327-349.

Fritzsche, D.J., (2000), "Ethical Climates and the ethical dimension of decision making", Journal of Business Ethics, 24(2), pp. 125-140.

Harrell, A. and P. Harrison, (1994), "An incentive to shirk, privately held information, and managers' project evaluation decisions", Accounting, Organization and Society, 18 (7), pp.569-577.

Jensen, M. and Meckling, W. (1976), "Theory of the firm: Managerial behaviour, agency costs and ownership structure", Journal of Financial Economics, 3, pp. 305-360.

Li, C., (2001), "Young, gifted and corrupt - a rising cadre type", South China Sunday Morning Post, June 24.

Rutledge, R.W. and K.E. Karim, (1999), "The influence of self-interest and ethical consideration on managers' evaluation judgments", Accounting, Organizations and Society, (24):173-184.

Snoeyenbos, M. and Jewell, D. (1983), "Morals, management and codes". In Snoeyenbos, M., R. Almeder and J. Humber (eds), Business Ethics, 179-197. Buffalo NY: Prometheus Books

Appendix 1

Section of research questionnaire

Dilemmas in the workplace

The following space is to be used to record your responses to the short story you have just listened to concerning a dilemma in the workplace. You are required to put yourself in the position of Mr. Gong and indicate how you might respond to the question raised at the end of the story.

Insider trading - Whether to tell management

If you were Mr. Gong would you advise management about your work mates' unethical activities?
(Please circle one of the numbers on the scale below)

say nothing
say nothing
say nothing

1       2       34       5       67       8       9       10