Effects of Organizational Culture and Pay Levels on Employee Retention: Focused on Generational Difference

Abstract

Research on generational differences in the workplace is important as it leads to more effective recruitment, retention, and employee wellbeing in terms of human resource management. This study investigates whether there are differences in perceptions of turnover intention, pay level, and organizational culture between different generational groups. First, this study identifies the relationship between pay level and turnover intention. Second, it confirms the moderating effect of organizational culture on the relationship between pay level and turnover intention. Finally, a three-way interaction analysis was conducted to determine whether these relationships differed between Baby Boomers, Gen Xers, and Millennials. As a result of the empirical analysis, the millennial generation showed higher turnover intention than other generations. It was confirmed that the effect of hierarchical culture was greater than that of pay level on the decrease in turnover intention in the millennial generation than in other generations. In other words, it was found that as the hierarchical culture weakened, the turnover intention of Millennials decreased more significantly than that of other generations. Based on these findings, we propose recommendations for understanding Millennials in the workplace and managing generational diversity.

1. Introduction

Diversity in age, race, gender, religion, and ethnicity in the Korean labor market is increasing. As the diversity of the workforce increases, diversity management is emerging as a major issue in organizations (Vinarski-Peretz et al., 2024; Yang & Konrad, 2011). Age diversity can be considered as inter-generational diversity in the workplace. Inter-generational issues are universal problems that occur in any era and society and are important in organizations as well. Given that each generation has different learning and social experiences depending on their age, there always exists a generational gap.

When the pace of change is fast, the gap between the generations is large and conflicts between them increase (Lyons & Kuron, 2014). This phenomenon can be observed by examining the growing issue of the generational differences in organizational management, especially for the millennial generation, which has been described as having a large gap with the previous generation. The same generation shares and experiences similar historical, social, and cultural events that have influenced the formation of their attitudes and values (Twenge, 2010). Twenge (2010) has argued that knowing and understanding generational differences is important for human resource management because each generational group has values and characteristics that directly affect attitudes and behaviors. In an organization, discriminatory attitudes and behaviors between generations can predict motivation to perform tasks. Accordingly, studies on differences between generations within an organization can lead to more effective recruitment, retention, succession management, communication, employee involvement, and conflict resolution (Dencker et al., 2008).

Therefore, this study examines how work attitudes differ between generations within an organization and suggests organizational management solutions accordingly. This study has several objectives. First, we identify the relationship between pay level and turnover intention. Second, we evaluate the moderating effect of organizational culture on the relationship between pay level and turnover intention. Finally, we examine whether this research model applies differently to each generation, consisting of Baby Boomers, Generation X, and Millennials. In doing so, we identify generational characteristics of Korean employees and suggest implications for human resource management.

The contributions of this study are as follows. First, most of the existing research on generational gaps in the workplace has been conducted by analyzing a single company, making it difficult to generalize the research findings. This study analyzes data from 9053 employees in 510 companies, thus contributing to the generalization of research findings on generation gaps. Second, this study examines the interaction effect of organizational culture, focusing on the differences between generations, using research models of previous studies examining the relationship between pay level and turnover intention. Most previous studies examining the relationship between employees’ attitudes towards organizational culture and performance have assumed that employees are a homogeneous group. Therefore, our findings on the differences in perceptions of organizational culture across generational groups will contribute to the field of organizational culture research.

2. Theoretical Background and Hypotheses

2.1. Pay Level and Turnover Intention

Pay level is one of the factors that motivate employees in the workplace and has a significant influence on their job performance (Lawler, 1981). Many studies have shown that a high pay level reduces the turnover rate of employees (e.g., Porter & Steers, 1973). The organization pays wages in return for labor performed by the employees. If the employees feel that the pay level is not fair compared with that of others, employees will be dissatisfied with the organization and pay level, which may increase their turnover intention.

In the turnover model proposed by March and Simon (March & Simon, 1958), employees compare the incentives they receive from the organization with their contributions to the organization. In this process, employees decide to leave the organization when they determine that the incentives provided by the organization do not meet their expectations (Murray & Holmes, 2021; Obeng & Atan, 2024). When a worker decides to leave, they consider the costs and benefits. In the process, if a worker decides that it is better to continue working than to leave, the worker will not leave the organization (Mueller & Price, 1990). In the cost–benefit analysis, if a wage higher than what the current organization is offering can be obtained through a turnover, the worker’s wage satisfaction is the primary motivating factor for moving.

Studies examining the relationship between wage satisfaction and turnover have found that the higher the dissatisfaction with wages, the higher the turnover (Price, 1989; Xiong et al., 2018). A higher salary level leads to positive job satisfaction, and higher job satisfaction reduces employees’ intention to leave (Y. Li et al., 2020). Additionally, R. Li and Yao (2022) have found that pay satisfaction is moderately correlated with teachers’ intention to leave. Jolly et al. (2021) found that perceived organizational support mediates the negative relationship between pay satisfaction and turnover intentions in the hospitality industry. As such, previous studies have shown that the higher the pay level, the lower the turnover intention.

Hypothesis 1-1:

Pay level will negatively affect turnover intention.

2.2. The Interaction Effect of Hierarchical Culture on the Relationship Between Pay Level and Turnover Intention

Organizational culture consists of artificial creations, values, and fundamental beliefs (Schein, 1990) shared by organizational members and existing at multiple levels such as groups and organizations (Detert et al., 2000). Organizational culture influences individual attitudes and behaviors in an organization, and many studies have shown the relationship between organizational culture and organizational performance (Hartnell et al., 2011; Smircich, 1983). Research on inter-generational conflict within an organization points to organizational culture as a factor that deepens inter-generational conflict (Choi et al., 2010). Confucianism heavily influences Korea’s corporate culture, which consists of authoritarianism, hierarchy, seniority-centrism, and collectivism (S. Park & Park, 2018). Therefore, in this study, the effect of hierarchical culture in an organization, which is the central corporate culture of many Korean companies, on the organization’s employees is examined by focusing on the differences among generations.

Hierarchical culture emphasizes stability and control within the organization and is centered on regulation by rules and maintenance of order. In an organization centered on a hierarchical culture, consistency, predictability, efficiency, and accuracy are essential in the working process (Denison & Spreitzer, 1991). In a hierarchical culture, clear communication, task standardization, and customization are valuable, and rules and procedures are emphasized with little discretion to maintain efficiency. Hierarchical culture is characterized by the pursuit of organizational stability and balance through structural features, such as formalization and centralization, while performing routine and standardized jobs (Quinn, 2011).

Hierarchical organizational culture affects the relationship between pay levels and turnover intentions. For example, even if employees are dissatisfied with the low pay level and search for other organizations that offer higher wages as an alternative, if the organizational culture such as the work process and communication method fits the individual well, the intention to leave the current job will be lowered. Conversely, an organizational culture that does not suit the individual well is expected to increase the intention to leave the current organization even if the organization provides a relatively high level of pay. Holzwarth et al. (2020) showed the results that organizational communication affects turnover intention through organizational commitment. It was argued that turnover intention increased in organizations with low levels of vertical and horizontal communication. J. S. Park and Kim (2009) have also argued that a positive organizational culture lowers turnover intention by increasing employees’ job satisfaction and organizational commitment. Lee and Jang (2020) found that the organizational culture also indirectly affected turnover intention through job stress and fatigue of clinical nurses. Therefore, in organizations with a strong hierarchical culture, turnover intention is higher than in organizations without one (Chapman et al., 2018). In this case, the effect of pay level on turnover intention is stronger than that in organizations that do not have a strong hierarchical structure.

Hypothesis 1-2:

Hierarchical culture will moderate the negative relationship between pay level and turnover intention. Therefore, the hierarchical culture will strengthen the negative relationship between pay level and turnover intention.

2.3. Generations in the Workplace

A generation is defined in various ways. According to Mannheim (1952), a generation refers to a group of people who are born and grow up during a specific period, simultaneously experience significant social and historical events, and share a similar worldview. Schaie (1965) has defined generation as a concept that includes people born during the same period. According to this definition, individuals belonging to the same generation create personalities, values, and attitudes toward organizations because of similar social situations (Wey Smola & Sutton, 2002). Consequently, the meaning of work, preferred climate in the organization, and style of working are unique for each generation (Kupperschmidt, 2000). In a study on a Korean sample, J.-H. Park (2001) defined a generation as, “A group that shares the historical and cultural experiences of a society at a similar life cycle stage, and sharing these experiences leads to similar value and behavior patterns”. The same generation shares similar identities because they experience similar economic situations, social contexts, and historical events (J.-H. Park, 2001). In this study, we divide the generations into Baby Boomers, Generation X, and Millennials, and examine the interaction effect of organizational culture on the relationship between pay level and turnover intention, focusing on the differences across groups by generation.

Recent research highlights the differences in responses to human resource (HR) systems and organizational culture among different generations of Korean workers. Cho et al. (2022) demonstrate that the effects of human resource management on individual attitudes and behaviors differ across generations. Unlike earlier generations, the MZ generation shows that motivation-enhancing HR practices significantly affect turnover intention and organizational loyalty. Yoon and Jung (2023) found that clan culture’s impact on organizational commitment is less significant for the MZ generation compared to previous generations. However, when the perceived importance of human resources acts as a mediator, the relationship between clan culture and organizational commitment becomes stronger for the MZ generation.

2.3.1. Millennials

Millennials were born in the 1980s and the 1990s, excelling in information and communication technology, experiencing the global financial crisis in 2008, and starting a social life in an employment environment that deteriorated during the economic downturn (Magni & Manzoni, 2020). Millennials, who grew up seeing their parents’ generation’s loyalty to the organization and devotion to their family, have the mindset that work is not everything in life. Millennials value the quality of life of individuals, such as leisure and freedom (Wey Smola & Sutton, 2002). Unlike other generations, Millennials have the characteristics of living in a nuclear family and undergoing various experiences related to technological development based on economic stability (Fishman, 2016). For Millennials, it is important to enjoy work, find meaning through pleasure, value leisure time, have time autonomy, and protect their personal life at work. Millennials, who prefer to do new things at work or things that are more valuable and challenging to themselves, show high values for enjoyment, individualism, diversity, work–life balance, and self-development (Lyons & Kuron, 2014). Ju and Jang (2019) have shown that the younger generation has a more robust demand for organizational change, whereas the older generation tends to maintain the current situation instead of organizational change in the Korean sample.

Job satisfaction factors that significantly affect the turnover intention of Millennials are work, communication, and pay for performance (Ju & Jang, 2019). The preferred organizational climate for Millennials is that they communicate and work in a flexible, informal, and intimate environment, instead of taking orders based on formal hierarchies (Magni & Manzoni, 2020). Moreno et al. (2022) analyzed the predictors of the turnover intention of millennial employees; compared to Baby Boomers and Generation X, supportive organizational culture appeared to have a more significant impact on turnover intention. In an organizational culture that emphasizes control and strong top-down communication, the effect of hierarchical culture on turnover intention is expected to be higher for Millennials than for other generations.

Hypothesis 2:

The moderating effect of hierarchical culture on the negative relationship between pay level and turnover intention will be moderated by generations.

Hypothesis 2-1:

The moderating effect that hierarchical culture weakens the negative relationship between pay level and turnover intention will be strong in the millennial generation.

2.3.2. Generation X

Generation X is from the era in which Internet use began, that is, the 1990s. This generation experienced a rapid social change in the information age amidst the emphasis on globalization and International Monetary Fund (IMF) crisis.1 Ideological conflict had come to an end worldwide, and Korea experienced an uprising in June 1987 as a beneficiary of industrialization (S. Park & Park, 2018). Generation X has a strong sense of individualism and de-authoritarianism (Fishman, 2016). Compared with Baby Boomers, Generation X is less loyal to organizations, values work–life balance over social status and job security, prefers individual growth opportunities in organizations, and tends to focus on careers instead of organizations (Lyons & Kuron, 2014). Accordingly, in an organizational culture, organization-oriented control and order are emphasized by Generation X, who believe that an individual’s career is more important than loyalty to the organization. Furthermore, if organizational communication is downward, then the effect of hierarchical culture on turnover intention is expected to be higher than that for the baby boomer generation.

Hypothesis 2-2:

The moderating effect that hierarchical culture weakens the negative relationship between pay level and turnover intention will be strengthened in Generation X.

2.3.3. Baby Boomer Generation

The baby boomer generation was born after the Korean War in 1950 and experienced rapid economic growth since 1960, an oil shock in the early 1980s, and the IMF economic crisis in the late 1990s. They have maintained a stable economic status through active economic activities, along with rapid economic growth (S. Park & Park, 2018). This generation politically experienced ideological conflicts, restoration, and democratization, getting through the era as a leader in industrialization and out of absolute poverty. To this end, this generation is characterized by long working hours, poor working conditions at companies, and self-sacrificing for childcare and parenting at home (S. Park & Park, 2018). In an organization, the baby boomer generation is a group with high loyalty and commitment to the organization (Lyons & Kuron, 2014). The baby boomer generation, which occupies the upper ranks of organizations, emphasizes legal authority based on modern values, and is characterized by being achievement oriented (Lyons & Kuron, 2014). In the case of the baby boomer generation, it can be expected that the moderating effect of hierarchical culture will not be significant compared with that of the other younger generations. This is because Baby Boomers emphasize order and hierarchy and are accustomed to top-down communication.

Hypothesis 2-3:

The moderating effect that hierarchical culture weakens the negative relationship between pay level and turnover intention will be weakened in the baby boomer generation.

The research model discussed is shown in Figure 1.

3. Materials and Methods

3.1. Sample and Procedures

This study uses employee data from the 2019 Human Capital Corporate Panel (HCCP) of the Korea Research Institute for Vocational Education and Training. The HCCP started the survey in 2005. It was conducted in such a way that the surveys of the company and its employees matched once every two years. One benefit of using this panel is that it simultaneously surveys a company and its employees. Another advantage is that this panel allows for a longitudinal analysis because a follow-up survey of the same company was conducted. However, in the case of employees, it is difficult to follow up with the same worker because of factors such as turnover and position change; therefore, only a cross-sectional analysis is possible. Therefore, this study only includes data from employees from a 2019 survey in the analysis. In total, 9053 employees from 510 companies were included in the final analysis. The characteristics of the data are shown in

Table 1. Data sample characteristics.

N = 9053		Frequency	Percent (%)	N = 9053		Frequency	Percent (%)
Industry	Manufacturing	7256	80.20	Job level	Staff	2355	26.00
	Finance	501	5.50		Administrative manager	1166	12.90
	Non-financial sector	1296	14.30		Assistant manager	1971	21.80
Firm size *	Fewer than 300	5215	57.60		Manager	1551	17.10
	300–999	2910	32.10		Deputy general manager	1083	12.00
	More than 1000	928	10.30		Department manager	655	7.20
Gender	Female	2531	28.00		Executive manager	91	1.00
	Male	6521	72.00
Marital status	Not married	3954	43.70		Production manager	181	2.00
	Married	5097	56.30
Employment	Temporary employment	121	1.30	Education	Graduate from junior high school	41	0.50
	Permanent employment	8932	98.70		Graduate from high school	430	4.70
Unionization	Nonunion	8059	89.00		Graduate from vocational high school	1061	11.70
	Joined a union	994	11.00		Graduate from junior college	1622	17.90
Generations	Baby boomers	117	1.30		Bachelor’s degree	5404	59.70
	Generation X	3112	34.40		Master’s degrees	452	5.00
	Millennials	5818	64.30		Doctorate degrees	42	0.50

* Firm size = Number of employees.

.

3.2. Measures

This study examines the effects of pay level, hierarchical culture, and generational gap on turnover intention. To analyze the research model, the following control variables were used: industry of the company, firm size, employment period, job level, gender, marital status, educational background, employment status, union membership, and organizational culture.

3.2.1. Turnover Intentions

The measurement of turnover intention consisted of two items measured on a five-point Likert scale. The items are “I am considering moving to a company that offers better job conditions than the company I am currently working for” and “If I decide to leave this company, I will lose too many things in my life”. The reliability coefficient (Cronbach’s α) was 0.524.

3.2.2. Pay Level and Hierarchical Culture

For pay level, the value obtained by converting the employee’s annual income to the log was used. Hierarchical culture consists of three items measured on a five-point Likert scale. The reliability coefficient was 0.654. The survey items were as follows: “The company values formal procedures, rules, and policies”, “The company has a top-down communication and information flow”, and “The company has an organizational climate that emphasizes a sense of hierarchy”.

3.2.3. Generations

Generations are used to classify groups based on the year of birth of the employee. Researchers applied the base year for classifying generations in various ways. Millennials are mainly grouped as people born between 1981 and 1996 and the children of the baby boomer generation (Ng et al., 2010). However, some studies classify people born after 1980 as Millennials (Angeline, 2011), while others define people born between 1982 and 2004 as Millennials (Valenti, 2019; Walden et al., 2017). Generally, the classification by generation consists of three groups. The group of people born between World War II and 1960 is classified as the baby boomer generation. The group of people born between 1960 and 1980 is classified as Generation X. Finally, those born after the 1980 are classified as Millennials (Benson & Brown, 2011; Lyons & Kuron, 2014). In this study, the group of employees born before 1960 is defined as the baby boomer generation, referring to the generation classification of previous studies. The group of employees born between 1961 and 1980 was defined as Generation X. Finally, the group of employees born after 1981 was defined as the millennial generation. Among the respondents, the baby boomer generation accounted for 1.3% with 117 respondents, Generation X accounted for 34.4% with 3112 respondents, and the millennial generation accounted for 64.3%, with 5818 respondents.

3.2.4. Controls

Control variables were selected because they could affect turnover intention and independent variables of this study. The demographic and firm characteristic variables were controlled. Specifically, variables such as the industry and size of the firm to which the employees belonged, employees’ tenure, job level, marital status, education, employment status, and labor union membership were controlled. Finally, hierarchical, innovation, market, and clan cultures were included as control variables in the analysis of the research model. The reliability coefficients were 0.861, 0.770, and 0.875 for innovation, market, and clan cultures, respectively.

3.3. Analytic Strategy

Three-Way Interaction Analysis

This study examines the three-way interaction effect of pay level, hierarchical culture, and generations on turnover intention. A regression analysis model is used to analyze the three-way interaction effect, and the results were confirmed using a bootstrapping method with 5000 samples at the <0.10 level.

The regression equations used to analyze the three-way interaction effect of pay level, generation, and hierarchical culture on turnover intention are presented below. Equation (1) is used to analyze the two-way interaction effect of pay level and hierarchical culture on turnover intention. Equations (2) and (3) are used to analyze the three-way interaction effect of pay level, hierarchical culture, and generations on turnover intention. Generally, the equation for analyzing the three-way interaction effect is given by Equation (2). Given that this research model divides generations into three types, namely, Baby Boomers, Generation X, and Millennials, the generation variable was converted into a dummy variable and entered into the model, as shown in Equation (3) (Dawson, 2014).

$$\mathrm{Y}=b_0+b_{1}X+b_{2}Z+b_{3}XZ+b_{4}C+\epsilon$$

(1)

$$\mathrm{Y}=b_0+b_{1}X+b_{2}Z+b_{3}W+b_{4}XZ+b_{5}XW+b_{6}WZ+b_{7}XZW+b_{8}C+\epsilon$$

(2)

$$\begin{array}{c}\mathrm{Y}=b_0+b_{1}X+b_{2}Z+b_3{W}_1+b_4{W}_2+b_{5}XZ+b_{6}X{W}_1+b_{7}X{W}_2+b_8{W}_{1}Z+b_9{W}_{2}Z+b_{10}XZ{W}_1+b_{11}XZ{W}_2+\hfill \\ b_{12}C+\epsilon \hfill \end{array}$$

(3)

Y = turnover intention, X = pay level, Z = hierarchical culture, $W_1$ = generation X, $W_2$ = Millennials, C = control variables.

4. Results

4.1. Descriptive Statistics and Correlations

Table 2. Characteristics of generational groups.

N = 9047		Baby Boomers		Generation X		Millennials
		Frequency	Percent (%)	Frequency	Percent (%)	Frequency	Percent (%)
Firm size	Fewer than 300	96	82.10%	1837	59.00%	3279	56.40%
	300–999	15	12.80%	991	31.80%	1901	32.70%
	More than 1000	6	5.10%	284	9.10%	638	11.00%
Industry	Manufacturing	111	94.90%	2512	80.70%	4630	79.60%
	Finance	0	0.00%	142	4.60%	356	6.10%
	Non-financial sector	6	5.10%	458	14.70%	832	14.30%
Length of service	Mean	5.064	-	5.071	-	4.104	-
Gender	Female	21	17.90%	513	16.50%	1995	34.30%
	Male	96	82.10%	2599	83.50%	3823	65.70%
Marital status	Not married	3	2.60%	382	12.30%	3569	61.40%
	Married	114	97.40%	2730	87.70%	2248	38.60%
Employment	Temporary employment	5	4.30%	7	0.20%	109	1.90%
	Permanent employment	112	95.70%	3105	99.80%	5709	98.10%
Unionization	Nonunion	98	83.80%	2734	87.90%	5223	89.80%
	Joined a union	19	16.20%	378	12.10%	595	10.20%
Job level	Staff	46	39.30%	339	10.90%	1969	33.80%
	Administrative manager	5	4.30%	163	5.20%	997	17.10%
	Assistant manager	4	3.40%	233	7.50%	1732	29.80%
	Manager	7	6.00%	645	20.70%	898	15.40%
	Deputy general manager	8	6.80%	895	28.80%	179	3.10%
	Department manager	13	11.10%	628	20.20%	14	0.20%
	Executive manager	17	14.50%	73	2.30%	1	0.00%
	Production manager	17	14.50%	136	4.40%	28	0.50%
Pay level	Mean	8.288	-	8.489	-	8.169	-
Turnover intention	Mean	2.71	-	2.87	-	3.26	-
Sum	-	117	1.30%	3112	34.40%	5818	64.30%

presents the characteristics of each generation group. It was confirmed that, compared to other generations, the baby boomer generation group had a large number of employees belonging to firms with less than 299 employees in the manufacturing industry. This seems to be the result of the company’s practice of re-employing some of its retired employees as contract workers.

The pay level of the baby boomer generation was higher than that of Millennials and lower than that of Generation X, and their turnover intention was the lowest among the generation groups. The following two reasons can be considered to explain the lower pay level of the baby boomer generation, compared to that of Generation X. First, the size of companies where baby boomer generation employees work is relatively small compared to that of the companies where employees from other generations work. Second, given Korea’s peak wage system, the wages of employees aged 50 and older is less. In the case of re-employed workers after retirement, their wages also drop to around 27.00% of their pre-retirement wages on average. Therefore, the pay level of employees in the baby boomer generation is lower than that of employees in Generation X.

The size and industry of firms where Generation X are mainly employed is similar to those of the companies in which the Millennials are employed. In terms of gender, Generation X had a higher proportion of males than Millennials, similar to the baby boomer generation, and a higher proportion of employees with managerial positions. The turnover intention of Generation X was found to be in the middle of the baby boomer and millennial generations, but the pay level was the highest among the generation groups.

Finally, because the millennial generation is the youngest, there were many employees at the staff-level, and the proportion of women was higher than that of other generations. It can be seen that the millennial generation has the lowest pay level and highest turnover intention among the generation groups.

Before hypothesis-testing, descriptive statistics and correlations between variables were analyzed.

Table 3. Descriptive statistics and correlations.

		M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1	Manufacturing	0.80	0.40
2	Firm size 1	1.53	0.67	−0.19 **
3	Length of service 2	4.45	0.85	0.02	0.12 **
4	Gender 3	0.72	0.45	0.11 **	0.01	0.14 **
5	Marital status 4	0.56	0.50	0.04 **	0.01	0.50 **	0.18 **
6	Graduate from high school	0.16	0.37	0.11 **	−0.12 **	0.09 **	−0.08 **	0.04 **
7	Bachelor’s degree	0.60	0.49	−0.058 **	0.13 **	−0.11 **	0.14 **	−0.06 **	−0.54 **
8	Employment 5	0.99	0.11	0.06 **	−0.03 **	0.13 **	0.10 **	0.06 **	0.00	0.04 **
9	Unionization 6	0.11	0.31	−0.14 **	0.16 **	0.12 **	0.02 *	0.04 **	0.11 **	−0.08 **	0.04 **
10	Turnover intention	3.12	0.85	0.00	−0.06 **	−0.23 **	−0.14 **	−0.18 **	−0.03 **	0.04 **	−0.03 **	−0.06 **
11	Baby boomers	0.01	0.11	0.04 **	−0.05 **	0.08 **	0.03 *	0.10 **	0.12 **	−0.11 **	−0.03 **	0.02	−0.06 **
12	Generation X	0.34	0.48	0.01	−0.03 **	0.53 **	0.19 **	0.46 **	0.16 **	−0.17 **	0.07 **	0.03 **	−0.21 **	−0.08 **
13	Millennials	0.64	0.48	−0.02	0.04 **	−0.54 **	−0.19 **	−0.48 **	−0.19 **	0.19 **	−0.06 **	−0.03 **	0.22 **	−0.15 **	−0.97 **
14	Pay level	8.30	0.34	−0.06 **	0.09 **	0.47 **	0.20 **	0.36 **	−0.18 **	0.13 **	0.06 **	−0.10 **	−0.26 **	0	0.46 **	−0.46 **
15	Hierarchical culture	3.49	0.66	−0.04 **	0.13 **	0.04 **	0.03 *	0.04 **	−0.10 **	0.08 **	−0.02	0.03 *	−0.09 **	0	0	0	0.07 **

** p < 0.01, * p < 0.05, N = 9053. ¹ Firm size = Number of employees, ² Length of service = log(tenure), ³ Gender: 0 = Female, 1 = Male, ⁴ Marital status: 0 = Not married, 1 = Married, ⁵ Employment: 0 = Temporary employment, 1 = Permanent employment, ⁶ Unionization 0 = Refuse to Joined a union, 1 = Joined a union.

shows the descriptive statistics and correlation analysis results. Firm size, education level, and turnover intention were positively correlated with generation. Length of service, gender, marital status, employment, and labor union membership were found to have a negative correlation with generation. Therefore, it can be seen that the larger the size of the firm, the higher the proportion of employees in the younger generation. The Millennials have a higher level of education than other generations, but they have a lower pay level and a lower percentage of regular employees.

4.2. Confirmatory Factor Analysis

Confirmatory factor analyses (CFA) was conducted to examine the distinctiveness of the dependent and independent variables (Martínez-Navalón et al., 2020; Reyes-Menendez et al., 2019). The analysis was conducted using M-Plus software. We estimated a one-factor model with turnover intention and organizational culture as the factors. We also estimated a two-factor model, in which turnover intention and organizational culture were divided into individual factors. We finally estimated our theoretical model, in which organizational culture is divided into four factors.

Considering the comparative fit index (CFI), Bentler (Bentler, 1990) has proposed that a model with a value of 0.90 or higher has an appropriate fit while a model with a value of 0.95 or higher has a good fit. Browne and Cudeck (Browne & Cudeck, 1992) have suggested that the root mean square error of approximation (RMSEA) with a value of 0.05 or less is considered a good fit; if this value is between 0.05 and 0.08, then it is regarded as a fair fit. Hu and Bentler (Hu & Bentler, 1999) have reported that if the standard root mean square residual (SRMR) value is 0.08 or less, the model fit is appropriate (Kim & Lee, 2022).

The results indicate that the research model fits the data well. The RMSEA = 0.073, CFI = 0.954, normed fit index (NFI) = 0.928, SRMR = 0.049. The research model was also superior to the alternative models (a one-factor model equating all variables [$∆X^2$= 11,452, df = 18, p = 0] and a two-factor model equating the four facets of organizational culture [$∆X^2$ = 10,903, df = 17, p = 0]. The fit indices of the one-factor model are as follows: RMSEA = 0.143, CFI = 0.768, NFI = 0.726, SRMR = 0.088. The fit indices of the two-factor model are as follows: RMSEA = 0.141, CFI = 0.777, NFI = 0.733, SRMR = 0.085. These results of absolute fit indices indicate that the one- and two-factor models have a poor model fit (Bentler, 1990; Browne & Cudeck, 1992; Hu & Bentler, 1999). Based on the results of relative and absolute fit indices, the research model was selected as the final model. These results support the convergent validity of our research model. These results show that one common factor does not explain most of the covariance between measures. This indicates that the possibility of common method bias is minimized (Podsakoff et al., 2003).

4.3. Hypothesis Testing

Table 4. Results of regression analyses for turnover intention.

Turnover Intention	Model 1	Model 2	Model 3	Model 4	Model 5
	β	β	β	β	β
Constant	3.727 ***	7.209 ***	5.061 ***	7.037 ***	−8.982
Manufacturing	0.147 ***	0.031	0.030	0.035	0.033
Non-financial sector	0.216 ***	0.097 *	0.094 *	0.104 *	0.101 *
Firm size: fewer than 300	0.116 ***	0.034	0.033	0.040	0.040
Firm size: 300–999	0.089 **	0.020	0.019	0.022	0.022
Log (length of service)	−0.263 ***	−0.236 ***	−0.236 ***	−0.214 ***	−0.216 ***
Staff	0.427 ***	0.065	0.064	0.010	−0.009
Administrative manager	0.360 ***	0.033	0.029	−0.029	−0.056
Assistant manager	0.386 ***	0.108	0.103	0.037	0.004
Manager	0.340 ***	0.093	0.088	0.038	0.001
Deputy general manager	0.159 +	0.045	0.042	0.023	0.000
Production manager	0.024	−0.159	−0.165 +	−0.193 +	−0.217 *
Gender 1	−0.125 ***	−0.093 ***	−0.092 ***	−0.085 ***	−0.083 ***
Marital status 2	−0.048 *	−0.064 **	−0.063 **	−0.049 *	−0.054 **
Graduate from high school	0.170	0.184	0.188	0.118	0.174
Graduate from junior college	0.200	0.225 +	0.229 +	0.143	0.203
Bachelor’s degree	0.225	0.242 +	0.244 +	0.152	0.210
Master’s degree	0.071	0.167	0.172	0.082	0.141
Employment 3	−0.026	−0.062	−0.062	−0.071	−0.065
Unionization 4	−0.058 +	−0.108 ***	−0.108 ***	−0.107 ***	−0.113 ***
Innovative culture		−0.267 ***	−0.267 ***	−0.267 ***	−0.266 ***
Clan culture		−0.246 ***	−0.247 ***	−0.248 ***	−0.249 ***
Market culture		−0.041 *	−0.042 *	−0.040 *	−0.039 *
Pay level (X)		−0.216 ***	0.045	−0.213 ***	1.70 +
$\mathrm{Generation} \mathrm{X} (W_1$)				0.114	15.752 +
$\mathrm{Millennials} (W_2$)				0.205 **	15.333 +
Hierarchical culture (Z)		0.123 ***	0.729 *	0.123 ***	5.031 +
$\mathrm{X}\times W_1$					−1.853 +
$\mathrm{X}\times W_2$					−1.817 +
$\mathrm{X}\times \mathrm{Z}$			−0.073 +		−0.586 *
$\mathrm{Z}\times W_1$					−4.746 *
$\mathrm{Z}\times W_2$					−4.830 *
$\mathrm{X}\times \mathrm{Z}\times W_1$					0.562 *
$\mathrm{X}\times \mathrm{Z}\times W_2$					0.580 *
F	47.97 ***	132.14 ***	127.13 ***	122.85 ***	97.350 ***
$R^2$	0.113	0.305	0.308	0.309	0.310

*** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.10, β = unstandardized regression coefficients, ¹ Gender: 0 = Female, 1 = Male, ² Marital status: 0 = Not married, 1 = Married, ³ Employment: 0 = Temporary employment, 1 = Permanent employment, ⁴ Unionization: 0 = Refuse to Joined a union, 1 = Joined a union.

presents the results of analyzing the two-way interaction effect of pay level and hierarchical culture and the three-way interaction effect of pay level, hierarchical culture, and generations on turnover intention.

First, Model 1 reports the effects of control variables on turnover intention. Turnover intention was higher in manufacturing companies than in the financial or service industries, and small companies had a higher turnover intention than large-scale companies. Turnover intention was found to be higher for employees with short tenure, lower job positions, women, and those who were unmarried than for those who were married. Model 2 reports the effects of pay level and hierarchical culture on turnover intention. The analysis confirmed that pay level (β = −0.216, p < 0.001) and hierarchical culture (β = 0.123, p < 0.001) were negatively and positively associated with turnover intention, respectively. Hypothesis 1-1 proposed that pay levels negatively affect turnover intention. As expected, pay levels are negatively related to turnover intention. Thus, Hypothesis 1-1 is supported. Results showed that the higher the pay level, the lower the turnover intention, and the stronger the hierarchical culture, the higher the turnover intention.

Model 3 examined the two-way interaction effect of pay level and hierarchical culture on turnover intention. As expected, the effect of the negative relationship between pay level and turnover intention was weakened by hierarchical culture (β = −0.073, p < 0.10). However, because the interaction effect was weak, a slope difference test was conducted to examine the actual degree of the interaction effect (Dawson & Richter, 2006). The results showed that both the upper and lower slope differences were significant (p = 0.003). Hypothesis 1-2 predicted that hierarchical culture moderates the negative relationship between pay levels and turnover intention. Thus, Hypothesis 1-2 is supported. Figure 2 illustrates the interactive effect of pay levels and hierarchical culture on turnover intention. Thus, it is confirmed that the negative relationship between pay level and turnover intention becomes stronger when hierarchical culture is high.

Models 4 and 5 report the results of the three-way interaction effect of pay level, hierarchical culture, and generation on turnover intention. Similar to Model 2, in Model 4, pay level was negatively associated with turnover intention (β = −0.213, p < 0.001), and hierarchical culture was positively associated with turnover intention (β = 0.123, p < 0.001). In Model 4, a generation variable was added; hence, the turnover intention of Millennials was higher than that of other generations (β = 0.205, p < 0.01). Finally, as predicted by the hypothesis, in Model 5, generational differences were shown to positively moderate the interaction effects of hierarchical culture on pay levels (β = 0.562, p < 0.05, β = 0.580, p < 0.05). Hypothesis 2-1 predicts that the moderating effect of hierarchical culture weakens the negative relationship between pay level and turnover intention for the millennial generation. The research results support Hypothesis 2-1. Hypothesis 2-2 was supported because it predicted that the moderating effect of hierarchical culture would weaken the negative relationship between pay level and turnover intention for Generation X. Finally, Hypothesis 2-3 predicted that the moderating effect of hierarchical culture would weaken the negative relationship between pay level and turnover intention for the baby boomer generation. Consequently, Hypothesis 2, which predicted that the negative moderating effect of hierarchical culture on the negative relationship between pay level and turnover intention would be moderated by generations, was supported.

Figure 3 shows the three-way interaction effect by generation. In the baby boomer generation, the negative effect of pay level on turnover intention was not significant when the hierarchical culture was low; however, its negative effect of pay level on turnover intention was highest among the generation groups when the hierarchical culture was high. Therefore, the baby boomer generation has a low turnover intention if the pay level is high in an organization with a high hierarchical culture. In other words, even in organizations with a strong hierarchical culture, the baby boomer generation has a low turnover intention if the pay level is high. In this case, it is important to raise the pay level to lower the turnover intention of Baby Boomers.

Subsequently, in the millennial generation, the effect of pay level on turnover intention was relatively weak compared to other generations; however, the effect of hierarchical culture on turnover intention was the strongest among generations. These findings suggest that weakening the hierarchical culture is more important than raising the pay level to lower the turnover intention of the millennial generation.

Finally, Generation X shows the effect of pay level and hierarchical culture on turnover intention at the intermediate level of Baby Boomers and Millennials. For Generation X, the effect of pay level on turnover intention was lower than that of Baby Boomers but higher than that of Millennials. The effect of hierarchical culture on lowering turnover intention was higher than that of Baby Boomers but lower than that of Millennials. The negative relationship between hierarchical culture and turnover intention was higher than that of Baby Boomers but lower than that of Millennials. These results indicate that the turnover intention perceived by Generation X is influenced by pay level and hierarchical culture. Across all generations, organizations with a weak hierarchical culture showed lower turnover intentions than organizations with a strong hierarchical culture. Therefore, the effect of reducing turnover intention according to pay level was highest in Generation X. In the case of organizations with a strong hierarchical culture, when the pay level was low, the turnover intention of Generation X was the lowest among all generations. However, as the pay level gradually increased, there was a reversal phenomenon between generations regarding turnover intention; accordingly, the turnover intention of the Baby Boomer generation was the lowest among the generations.

5. Discussion

This study confirms the effect of organizational culture on the relationship between pay levels and turnover intention, focusing on inter-generational differences. Particularly, it examined whether the effect of hierarchical culture on turnover intention in Korean companies influenced by Confucianism (Choi et al., 2010; Ju & Jang, 2019) differs for each generational group.

First, Millennials manage their careers and networks through connections with various media based on excellent information technology. They experienced a global financial crisis in 2008 and entered the labor market. Therefore, they have a negative attitude towards the strong organizational commitment of their parents’ generation and put more emphasis on their personal life than on work or organization (Lyons & Kuron, 2014). Given that Millennials prefer to do meaningful work, protect their personal lives, and welcome challenges, they do not emphasize rules and procedures for stability and control. Therefore, it can be predicted that organizations with rigid communication and a hierarchical culture of command and discipline will have a relatively high level of turnover intention. Furthermore, it can also be predicted that weakening the hierarchical culture is more effective than providing a high level of compensation to lower the turnover intention of Millennials. The empirical analysis confirmed that Millennials had higher turnover intentions than other generations and that the decrease in turnover intention in organizations with a weak hierarchical culture was greater than the decrease in turnover intention caused by higher wages.

Second, the baby boomer generation was the leader of Korea’s economic development in the 1970s and 1980s. They are an achievement-oriented generation who are hardworking and sincere, have a high level of loyalty and commitment to the organization, and value promotion to a higher position in the organization more than leisure in their personal life (S. Park & Park, 2018). The pay level is important in terms of monetary compensation, but there is sometimes also the value of recognition and high status in the organization. Therefore, it can be predicted that the effect of pay level on turnover intention will be stronger than that of hierarchical culture on Baby Boomers. Examining the analysis results of the study, it can be seen that the turnover intention of the baby boomer generation was lower than that of other generations. In particular, organizations with a weak hierarchical culture showed a low level of turnover intention, regardless of the pay level. However, in organizations with a strong hierarchical culture, the level of turnover intention at low pay levels is high enough to be similar to that of Millennials. Therefore, as the pay level increases, the turnover intention decreases sharply, to the lowest level of any generational group.

Finally, Generation X in Korea has experienced rapid globalization due to the development of the Internet and the IMF crisis. They have experienced an unstable and flexible labor market. They are less loyal to the organizations, more individualistic, and place more emphasis on managing their careers rather than organizations than Baby Boomers. However, this generation does not pursue work–life balance as much as Millennials and is not as loyal and committed to their organization as Baby Boomers. Specifically, it is the generation with the highest level of family-career conflict due to its high level of career immersion (Beutell, 2013; Lyons & Kuron, 2014). Therefore, Generation X shows an intermediate level of turnover intention between the baby boomer and millennial generations. It is expected that the positive effect of hierarchical culture on turnover intention and the negative effect of pay level on turnover intention will be similar. In our research, we found that Generation X showed an intermediate level of turnover intention between Baby Boomers and Millennials in both cases where the hierarchical culture was weak and strong. When the hierarchical culture is strong and pay level is low, Gen Xers have the lowest turnover intention of any generational group. When the hierarchical culture is weak, they have the highest decrease in turnover intention of the generational groups as the pay level increases.

5.1. Research Contributions

This study makes academic contributions in the following four areas. First, this study uses data from about 510 Korean companies to expand the generalization of research results on generational differences within organizations. It has been difficult to generalize the research findings because most of the research on generational differences is qualitative, such as an interview or a study conducted with a single company. Second, this study examines the moderating effect of organizational culture on the relationship between wage level and turnover intention, verified by dividing the generational group. This result is different from existing research on organizational culture, which assumes that the group of employees is homogeneous. Research in the field of organizational culture often assumes that employees are a homogeneous group. However, the findings of this study indicate that employees should be perceived as a heterogeneous group, particularly when considering generational differences. These results could enhance our understanding of employee diversity within the context of organizational culture research. Third, it is important to examine who is changing jobs in turnover-related studies. This study examines the factors that influence turnover intention by generation. Therefore, this study contributes to the literature on the reasons for job turnover among the older and younger generations. The finding of this study that perceptions of reward are heterogeneous across generational groups indicates a necessity for further research on the moderating effects of generational groups in compensation research. Finally, while most studies have examined the differences between generations and the level of difference between groups, this study makes an empirical contribution because it verifies the effect of antecedent factors by examining the interaction effect of generations.

5.2. Practical Implications

The results of this study can provide implications for managers of companies in which generational differences should be considered as antecedent factors influencing turnover intention. Companies’ managers should consider a differentiated management method for different generational groups by classifying worker groups. Efforts are needed to reduce the negative impact of differences in values, beliefs, and communication methods among generations on productivity and job attitudes.

The dynamics of employee retention across different generations highlight the unique turnover intentions exhibited by Baby Boomers compared to Generation X and Millennials. The result of this study indicates that Baby Boomers typically demonstrate lower turnover intentions, suggesting that organizations might benefit from strategically allocating resources. By focusing on reducing turnover intentions among Generation X and Millennials rather than Baby Boomers, management could enhance overall retention rates.

Unlike Baby Boomers, it is more important to consider various factors such as job characteristics, communication, and organizational culture than monetary compensation to reduce the turnover of Millennials. Compared to other generations, Millennials move frequently between employers and have higher turnover intentions. Therefore, to lower this generation’s turnover intention, it is more important to weaken the hierarchical culture rather than increase financial rewards.

In the case of Generation X, because both the pay level and hierarchical culture affect their turnover intention, it is necessary to maintain an appropriate pay level and simultaneously weaken the hierarchical culture.

Overall, recognizing and adapting to the diverse needs and preferences of different generational groups can play a significant role in enhancing employee retention and overall organizational effectiveness.

5.3. Limitations and Future Research

The limitations of this study are as follows. First, there is no academic consensus on the classification of generations. There are several ways to classify generations, such as age, cohort, and time period, but there is no clear agreement yet. This classification of generations needs to be performed differently based on nationality and taking into account national differences (Lyons & Kuron, 2014). However, given the lack of research on generational differences in Korea, this study did not classify the generations to fully account for the effects of country. In future research, it is necessary to classify generations according to major events and cultures in Korea. In order to generalize the findings on generational differences, it would be interesting to compare generational characteristics across countries.

Second, a difficulty in generational studies is controlling for the effects of life cycle, age, and organizational context on generational differences (Lyons & Kuron, 2014). Even though this study used a number of control variables to control for effects other than the generational effect, such as industry, educational background, position, marital status, and size of the firm in which workers are employed, it was not possible to free ourselves from the problem of controlling for the effects of differences in life cycle and age-related behavioral changes. Future studies will need to use longitudinal cohort studies to control for this effect.

However, even in a cross-sectional study such as this one, examining the effect of generational differences on work attitudes in the current organizational situations can contribute to the improvement of organizational performance through the rapid change in human resource management for Millennials. In a follow-up study, it is necessary to study the same-generational groups over a longer period of time to examine the impact of generational differences on the organization.

Third, the group size of the baby boomer generation was smaller than that of the other groups, and the reliability coefficients for turnover and hierarchical culture were relatively low. This limitation may be due to the characteristics of the panel data; therefore, caution is required in the interpretation of the results. Future studies will need to investigate generational differences in perceptions using firm-level variables, such as the human resource management system.

Fourth, the Cronbach’s α values for turnover intention and hierarchical culture were 0.524 and 0.654, respectively, both of which are below the generally accepted threshold of 0.7. These low values can be attributed to the relatively small number of items (2–3) in the scale, as well as potential respondent errors in answering the reverse translation questions. Hair et al. (2014) note that Cronbach’s α is sensitive to the number of items and often underestimates internal consistency reliability. Bonett and Wright (2015) emphasize that low Cronbach’s α values frequently arise from violations of equal variances assumptions, poorly designed surveys, or small sample sizes. They propose a more flexible method for estimating confidence intervals that does not assume equal variances and covariances among items, thereby enhancing its applicability in real-world research. Since the data used in this study is panel data, there are inherent limitations related to the number of items in the scale and potential respondent errors. Future studies should focus on selecting an appropriate number of items for the scales, increasing the sample size, and implementing well-designed surveys to prevent low Cronbach’s α values. These enhancements are expected to enhance the reliability of the findings and ensure the robustness and credibility of the results.

Finally, we used employees’ self-reported responses as our research data. In this case, common method bias may occur for reasons such as social desirability, consistency motive, and leniency bias (Podsakoff et al., 2003). We conducted confirmatory factor analysis and exploratory factor analysis for Harman’s single-factor test, which is the most commonly used method for dealing with common method variance (Podsakoff et al., 2003). We found that the probability of common method bias occurring was low. In future studies, efforts should be made to use different data sources to reduce the probability of common method bias.