DETERMINANTS INFLUENCING THE INTENTION TO CAUSE THE MORAL HAZARD OF VIETNAM COMMERCIAL BANKS’ STAFF

Anh THE KHUC^1,Linh HOAI DO²*, Xuan THANH NGO³

¹ Lecturer and Researcher, School of Banking and Finance, National Economics University, 207 Giai Phong Road, Hai Ba Trung District, Hanoi City, 100000, Vietnam.

² Lecturer and Researcher, Deputy Head, Department of Banking Management, School of Banking and Finance, National Economics University, 207 Giai Phong Road, Hai Ba Trung District, Hanoi City, 100000, Vietnam.

³ Lecturer and Researcher, School of Banking and Finance, National Economics University, 207 Giai Phong Road, Hai Ba Trung District, Hanoi City, 100000, Vietnam.

*Corresponding Author

E-mail: [email protected]; [email protected]

ABSTRACT

Vietnamese commercial banks are complying with Basel 2 regulations, particularly those pertaining to pillar 2. This article concentrates on moral hazard and the elements that influence bank employees' intentions to produce a moral hazard. We distributed a questionnaire to persons now employed in the banking industry, combining the Theory of Planned Behavior (TPB) was produced by Ajzen (1991) with the attribution theory was created by Weiner (1985). The structural equation modeling (SEM) results indicate that the decision manager and attitude toward the behavior have a detrimental effect. By contrast, positive affect is governed by subjective norms and perceived behavior. Additionally, the research discovers that moral hazard motivation (working motivation) has a detrimental effect on the decision manager and is influenced by distinctiveness and consensus. The research findings suggest that Vietnamese commercial banks should establish work standards for employees to strengthen the job's uniqueness and establish a business culture that discourages deviant behavior.

Keywords: Attribution Theory, Commercial Bank, Moral Hazard, TPB

INTRODUCTION

Commercial banking is unlike other businesses in that moral hazards can emerge at any level of the organization, from management to workers (Anginer & Demirgüç-Kunt, 2018). A moral hazard in banking activities damages the bank and its customers financially and can result in an explosion of risks and the collapse of the entire system.

Pillar 2 of Basel's risk management principles cover the causes of operational hazards caused by human acts such as fraud, forgery, theft, unauthorized activity, inefficient regulation, or ineffective human resource management. This is essentially a form of moral hazard (Girling, 2022). From Basel's perspective, moral hazard management is not a different topic but rather a few minutes of reflection on operational risk management. This is incompatible with the banking business context, where moral hazard manifests itself in various ways. Basel 2 and Basel 3 are crucial for transitional economies like Vietnam to participate in the international financial sector. Vietnam's commercial banking system faces numerous market risks as it pursues expansion objectives (Nguyen & Pham, 2020), implying that moral hazards would arise. To manage operational risk in accordance with Basel 2 and Basel 3, one factor to consider is moral hazard management. This essay focuses on that element and examines it through the lens of the desire to create moral hazard among the staff of Vietnam's commercial banks.

This study employs two novel theories to analyze behavioral intention. Specifically, Ajzen (1991) produced the Theory of Planned Behavior (TPB), while Weiner (1985) created the attribution theory (1985). Apart from the introduction, we will present a literature review and research hypotheses in part 2, research methodologies and models in part 3, and research findings and discussion in part 4. Additionally, in part 4, the research team explores the policy implications necessary to mitigate the consequences of bank staff's moral hazard intents and the policy implications necessary to perform better than Basel pillar 2.

Literature Review

The intention to cause the moral hazard to commercial bank’s staff

The moral hazard of banks is examined from a variety of perspectives, including the idea that banks will spend more in sectors with higher risks when protected by deposit insurance (Anginer & Demirgüç-Kunt, 2018), or commit acts of violation of the law, hence causing adverse market consequences to maximize profits (Cheng, Phung, Hsiao, Shen, & Chen, 2018). With a broader view of the market as a whole, Girling (2022) contends that banks take moral hazard in pursuit of profit, knowing that their activities may jeopardize the profits of other organizations and persons in the market. The analysis in this paper is based on Girling (2022).

According to Ajzen (1991), an individual's expectations of their behavior and the degree to which those expectations are justified is referred to as the intention to conduct the behavior. Concerning the intention to create moral hazard in the financial sector, Awang, Rahman, and Ismail (2019) take a similar stance, arguing that the intention to perform the behavior is a matter that an individual can assess the risks involved but still perform due to the perceived benefits; or that individuals may make risky decisions due to the self-interest issue (Kashif & Khattak, 2017). According to this study, the intention to create moral hazard is the degree to which an individual feels that the risk outweighs the rewards, prompting them to take activities that may damage the bank or other financial market objects.

To cause a moral hazard in the financial sector, Awang et al. (2019) take a similar stance, arguing that the intention to perform a behavior is a matter of an individual being aware of the risks they may face but still engaging in risky behavior for the sake of great benefit; or individuals may engage in risky behavior due to the self-interest issue (Kashif & Khattak, 2017). According to this study, the intention to create moral hazard is the degree to which an individual feels that the risk outweighs the rewards, prompting them to take activities that may damage the bank or other financial market objects.

Research hypotheses

The term "distinctiveness" refers to an individual's ability to demonstrate the same behavior in various situations or scenarios. McKillop and Wilson (2015), Jones, Napiersky, and Lyubovnikova (2019), Navarro Cid, Rueff-Lopes, and Laurenceau (2022) assumed that the individual's behavior could be evaluated in frequency in the examined environment. When an individual's behavior consistently occurs in unfairly competitive circumstances (without enough supervision), it has a beneficial influence on the moral hazard motive. As a result, we assume the following:

H1: Distinctiveness affects the moral hazard motivation of the bank's employees (working motivation).

Consensus is a term used in attribution theory to describe the relationship between the behavior of an individual understudy and the behavior of other individuals in the same context, setting (Heider, 2013; Weiner, 1985). When people behave similarly in comparable situations, we might claim that their behavior reveals their identity. If others act equally, the consensus is high in this scenario, and vice versa. Heron (2020) demonstrated that when there is a high degree of coherence, the source of the behavior is irrelevant. On the other hand, if the behavior has a low level of consensus, the cause for the behavior is internal. Therefore, we form the following hypothesis:

H2: Consensus affects the bank's staff's moral hazard motivation (work motivation).

Individuals who exhibit the same behavior at different periods are consistent. According to Kelley (1973), consistency refers to an individual's frequency of performing the same behavior. Suppose an individual repeats an action numerous times over an extended period. It can be inferred that personal motivations are the primary motivator for that individual to conduct this behavior (Tamborini, Prabhu, Lewis, Grizzard, & Eden, 2016). At the same time, Berg and Koziol (2017); García-Arroyo and Segovia (2021) argue that consistency has the most significant influence on the attribution of an event's cause. As a result, we hypothesized:

H3: Consistency affects the bank's employees' moral hazard incentive (work motivation).

In attribution theory, Kelley (1973) argues that the motivation of employees to work is highly dependent on the Manager's decision - for example, career orientation or labor position. Yoo, Chen, and Frankwick (2020) demonstrate that the working motivation of bank employees is strongly related to the Manager's orientation. The findings indicate that when employees have specific motivations, it affects managers' decision to establish a bank's culture. As a result, we assume the following:

H4: Bank employees' moral hazard motivation (work motivation) affects the Manager's choice.

The Manager's choice is reflected in the organization's system of regulations, objectives, and plans (Narteh & Odoom, 2015). To complete the superior's objectives and responsibilities, an ideal manager must continually maintain the safety and security of his colleagues, minimizing the conditions for the formation of moral hazard behavior. Ahmad and Gao (2018); Kia, Halvorsen, and Bartram (2019); Ahmed, Chaojun, Hongjuan, and Mahmood (2022); Niu and Liu (2022) have demonstrated that managers' decisions directly affect the thinking and behavior of personnel who work for them. Staff subjected to stricter controls and inspections are less likely to participate in unethical behavior. As a result, we assume:

H5: The Manager's action affects the bank's employees' desire to create moral hazard.

Fishbein and Ajzen (1975) define subjective norm as what individuals believe others think or judge when that individual engages in a particular behavior. The term "subjective norms in banking" refers to the standards, basic rules, conventions, and ethics within the banking profession. Awang et al. (2019) demonstrate that the subjective norm is related to the bank's staff's misconduct judgment. As a result, we form the hypothesis as below:

H6: Subjective norms affect the intention of bank employees to create moral hazard.

The term "perceived behavioral control" relates to an individual's perception of how easy or difficult it is to do the desired behavior. Additionally, it reflects the individual's prior experiences or judgments about the issue at hand while executing the act (Ajzen, 1991). Kashif and Khattak (2017) investigate the relationship between employee ethics, "love money," and an employee's ethics; the findings indicate a strong correlation between staff perceived behavioral control toward ethics and ethical and legal infractions. As a result, we assume that:

H7: Perceived behavioral control affects the desire of bank employees to create moral hazard.

Attitude toward a behavior measures an individual's good or negative feelings about the behavior's performance or intention to conduct it (Ajzen, 1991). Although there are some divergent views on this factor, Bauer, Bernroider, and Chudzikowski (2017) discovered that an individual's attitude toward compliance with information security policies is a critical variable that can help close the gap between legal and unethical information security behavior raised in the study. This is also the conclusion of Saleem, Bhutta, Nauman, and Zahra (2019); Van Greuning and Bratanovic (2020); Pop, Lotrean, Buzoianu, Suciu, and Florea (2022). Thus, we hypothesized:

H8: Attitudes about behavior affect the intention of bank employees to engage in moral hazard.

MATERIALS AND METHODS

Research method

Marketing refers to research on consumers' behavioral intentions to purchase products, services, and items. However, our understanding of the purpose of creating moral hazards in the banking business is limited. As a result, we conduct qualitative and quantitative research in our analysis.

We conducted in-depth interviews with industry experts as part of our qualitative research: Four individuals studied the theory, while twelve worked in Vietnamese commercial banks. We conducted direct and indirect interviews in response to the Covid19 epidemic's impact. In-person interviews took place in private homes, cafes, and businesses; indirect interviews took place via Zoom platforms or Microsoft Teams. Each interview was between 25 and 40 minutes long. The qualitative research findings indicate that experts concur with the conclusions we propose, in which the component "managerial decision" is emphasized as a critical factor. Qualitative research also enables us to adapt the questionnaire's questions to the Vietnamese context, as they aim to create moral hazard is a touchy subject.

After editing the questionnaire in response to expert feedback, we distributed it to 1,000 bank employees, but 312 respondents did not answer. Of the 668 observations we received, 124 provided an incomplete or identical response. Thus, there are a total of 564 relevant observations.

Data processing methods: We employ SPSS25 and AMOS20 software for fundamental technical analysis (statistics, Cronbach's Alpha, EFA, CFA, and SEM). Because the model is illustrated in Figure 1 and there is just one element affecting the Manager's decision, we divide it into two models when we process data using structural equation modeling.

Research model

Based on the results of the research review and in-depth interviews, we propose the following research model:

Figure 1. Proposed research model

The model's variables are derived from past research and fine-tuned through qualitative analysis. Where Distinctness (DI) comprises four components, Consensus (CS) consists of three items, while Consistency (CO) consists of four. Three items comprise the moral hazard motivation (working motivation) of bank employees (IA), five items comprise the decision of the Manager (DOM), five items comprise the subjective norm (SN), three items contain the perceived behavioral control (PBC), and four items comprise the attitude toward behavior (ATB). Intention to create moral hazard among bank employees (DOB) consists of three components.

Validity

Descriptive statistics of the sample

Out of 564 satisfactory observations, the general statistical results are shown in Table 1.

Table 1. Statistics of the sample

Characteristics		Frequency	Ratio
Gender	Male	327	57.98%
Gender	Female	237	42.02%
Age	Under 30	71	12.59%
	From 31 to 40	197	34.93%
	From 41 to 50	170	30.14%
	Over 50	126	22.34%
Workplace	State-owned banks	254	45.04%
Workplace	Joint-stock commercial banks	310	54.96%
Working time	Less than one year	26	4.61%
	From 1 to 3 year	145	25.71%
	From 3 to 5 year	136	24.11%
	From 5 to 10 year	138	24.47%
	More than ten year	119	21.10%

(Source: Authors’ calculation)

Male respondents made for 57.98 percent of the total, significantly more than female respondents. This demonstrates that men appear to take more risks than women in Vietnam. Individuals under the age of 30 represent a relatively small part (12.59 percent) of the total age group, which equates to a short working duration (only 4.61 percent under one year, then 25.71 percent from 1 to 3 years old). We discovered that the age range of 31 to 50 years old accounts for the most significant amount, equating to more than three years of labor. In Vietnam, state-owned commercial banks account for a sizable proportion (over 40%), accounting for more than half of all respondents. The research sample can be regarded to be representative.

The result

Due to the model's peculiarities, we split the testing into two models, with model 1 handling hypotheses H1–H4 and model 2 handling the remaining hypotheses.

The result of the first model

In the first model, the exploratory factor analysis results using the observed variables "decision making" were favorable, as indicated by the coefficient KMO = 0.853, Sig = 0.000, demonstrating the high reliability of exploratory factor analysis. This factor's average variance extracted, and convergent validity Eigenvalues are 62.581 percent (more than 50%) and 3.129 percent (greater than 1), respectively. As a result, this component assures that the original survey data can be represented. The independent variable's item comprises three components: Distinctiveness, Consensus, and Consistency with 11 items, all of which are satisfactory criteria. Both the intermediate variable's object "moral hazard motivation (working motivation) of bank staff" (3 items) and the dependent variable's item "management choice" (5 items) converged and performed well on the items.

The CFA analysis of the model fit metrics revealed that the Chi-square/df value is 2, the TLI, CFI, and GFI are all greater than 0.9, the RMSEA coefficient is 0.05, and the PCLOSE coefficient is greater than 0.05, indicating that the model is market-fit. The observed variables representing the factors all have P-values equal to *** (equivalent to 0.000). As a result, the observed variables are validated to be a good representation of the CFA model's parts (Figure 2).

Diagram

Description automatically generated

Figure 2. CFA results of the first model

(Source: Authors’ calculation)

The fit criteria for the model 1 indicate that the Chi-square/df = 2,399 3, the TLI = 0.959, the CFI = 0.966, and the GFI = 0.940 are all greater than 0.9, and the RMSEA = 0.050 0.050, and the coefficient PCLOSE = 0.499 > 0.050 are all greater than 0.499. Thus, the model is consistent with market data (Figure 3).

The independent variables Distinctiveness (DI), Consistency (CS), and the intermediate variable moral hazard motivation (working motivation) of the bank's staff (IA) all have a P-value less than 0.05 (with a 95% confidence interval), indicating that the relationship and influence of these variables on the dependent variable Manager’s decision can be confirmed (DOM). Specifically, the regression coefficients for the independent and intermediate variables reveal that the variables Distinctiveness and Consistency have the same effect on the median variable. The moral hazard motivation (working motivation) of bank employees is about equal, with absolute values of 0.149 and 0.147. However, because the regression coefficient for the factor Distinctiveness is negative, it will have a negative effect on the variable Worker's motivation. The regression coefficient for the intermediate variable worker's motivation is -0.114, indicating an antagonistic relationship with the dependent variable Manager's decision.

However, the P-value of the independent variable Consensus (CO) has a value of 0.696, more excellent than 0.05 (with 95% confidence), indicating that this independent variable has no effect significant impact on the intermediate variable Worker's motivation.

Diagram, schematic

Description automatically generated

Figure 3. SEM results of the first model

(Source: Authors’ calculation)

The result of the last model

The CFA analysis of the model fit metrics revealed that the Chi-square/df value is 2, the TLI, CFI, and GFI are all greater than 0.9, the RMSEA coefficient is 0.05, and the PCLOSE coefficient is greater than 0.05, indicating that the model is market-fit. The observed variables that represent the factors all have P-values equal to ***. (that is, equal to 0.000). As a result, it is validated that the observed variables accurately represent the elements in the CFA model (Figure 4).

Diagram

Description automatically generated

Figure 4. CFA results of the last model

(Source: Authors’ calculation)