Differentials in Information Technology Professional Category and Turnover Propensity: A Study

Introduction

The employee turnover phenomenon can be viewed in two distinctively different viewpoints—one is from the employers’ point of view and the other is from the employees’ point of view. It is beyond doubt that the departure of any potential employee from the company would not only be a great loss to the company but its replacement would also entail high cost both in terms of money and time. This departure of the high-valued employees has also a significant negative impact on company’s competitive advantage. On the other hand, a rational employee would always try to optimize the price of his or her potential capabilities as well as his or her professional position. Therefore, employee turnover phenomenon needs to be examined by considering both the viewpoints.

Human resource is considered as intangible capital (Leslie, 2003) with distinctive functional capabilities that control and augment both physical capital and other resources. This intellectual capital has become the obvious concern of this century which in turn diffused to develop the hypercompetitive market rivalries in the present world markets. Success in the present dynamic, competitive markets depends more on innovation, speed and adaptability which are largely derived from a firm’s own employees and the way they are managed (Pfeffer, 1994). Various scholars (Becker, 1975; Grant, 1996; Lawler, 1996; Levine, 1995; O’Reilly & Pfeffer, 2000; Wernerfelt, 1984) advocated that in order to gain competitive advantage, the firms need to adopt management practices with high involvement of human resources. These arguments are the genesis for the development of today’s strategic human resource management (SHRM). The strategic literature has focused heavily on the role of firm-level resources as a source of rent (Amit & Schoemaker, 1993; Barney, 1986; Peteraf, 1993). Nevertheless, the knowledge management plays important part in organization’s rent generation, that is, if a firm can deploy knowledge resources more efficiently and effectively than its rivals, it may achieve a substantial advantage (Coff, 2003). The presumption of bounded rationality (i.e., there is a limit on managerial cognitive ability) is at the core of the knowledge management literature. The portion of economics that intersects with strategic management (e.g., transaction cost economics, agency theory and human capital theory) has also assumed bounded rationality. Even game theoretic analyses explore the issues of imperfect and asymmetric information.

However, the major problem imminent to the management group of the organizations is turnover of human resources. Information technology (IT) sector is one of the most important as well as highly labour-intensive service sectors. This study is an endeavour to explore the factors behind the turnover of the high-valued IT employees which become a constant threat to the concerned companies towards achieving their desired goals under the Indian socio-economic market structure.

Literature Review

Academically different schools have discussed employee turnover from various perspectives and have identified various causes thereof. There are two types of employee turnover—voluntary (Gupta & Jenkins, 1991; Mowday, Porter & Steers, 1982; Saiyadain & Ahmad, 1997) and involuntary (Shaw, Delery, Jenkins & Gupta, 1998). Most of the researchers have focused on voluntary employee turnover since much of the turnover is voluntary and subject to control by the managers (Morrell, Loan-Clarke & Wilkinson, 2001; Price, 2001). Employees’ voluntary turnover depends on the demand for their intellectual capital as well as the availability of alternative job opportunities in the market. So, the rate of employees’ voluntary turnover varies significantly across different sectors of an economy. Organizations are much concerned about retaining their key employees (i.e., high-performer employees and those have important contribution to stimulate the organization’s innovative behaviour).

During the last few decades, scholars from different disciplines have been trying to identify the causal factors behind employees’ voluntary turnover and bring forth various models, theories and vast empirical case studies on employee turnover. Most of the turnover models established that job satisfaction plays a key role in the turnover process. Organizational psychologists developed various approaches, such as job characteristics approach, social information processing approach and dispositional approach to establish theoretically the influence of employees’ level of job satisfaction on other important outcomes (such as job performance, absenteeism and employee turnover). In this regard, Lee and Mitchell (1994) argued that the employee turnover process models are useful but they have ignored some basic properties of human decision-making processes. They developed an Unfolding Model of the turnover process and argued that some kind of ‘shocks’ in the system are responsible for making the employees to evaluate their job or job situation. Job offered by other organizations may even be considered as a ‘shock’ to the system and it forces the employee to think consciously about his or her job situation and compare it to the outside job offer. In such a situation, it is also possible that the employee may be reasonably happy in his or her job but ultimately leave the organization simply be-cause the offered job is a better one. Mitchell, Holtom, Lee, Sablynski and Erez (2001) developed Job Embeddedness employee turnover model combining the forces that keep a person from changing his or her employer. From individual-level decision-making point of view, various studies tried to explain the employees’ turnover intent. In fact, constructs have been developed, such as optimal turnover (when poorly performing employees decide to leave an organization) and dysfunctional turnover (can be viewed as turnover translate into increased costs associated with recruit and training of new employees and also may tarnish the image of the organization). In addition to psychological theories, there are economic theories (e.g., human capital theory, search theory and matching theory) and social theories (social exchange theory) which appeared to be very much pertinent to explain the employee turnover phenomenon.

Various empirical studies suggested that job satisfaction and organizational commitment have consistent and negative relationships with turnover (Jaros, 1997). In this regard, Irving, Coleman and Cooper (1997) introduced a new attitude construct to provide a new measure of occupational commitment, whereas Shore and Tetrick (1991) developed a different measure of perceived organizational support. It is argued that justice perception (Aquino, Griffeth, Allen & Hom, 1997) and burnout (Wright & Cropanzano, 1998) influence the attitudes and that in turn affect the employee turnover. In fact, the traditional attitude measure suggested that negative attitudes combined with job search predict employees’ leaving an organization (Blau, 1993). Studies also reported evidence, such as work overload, role ambiguity, role conflict and job stress in determining the turnover decisions (Bostrom, 1981; Goldstein & Rockart, 1984; Ivancevich, Napier & Wetherbe, 1983; Li & Shani, 1991; Sethi, Barrier & King, 1999; Weiss, 1983). Sharma and Bajpai (2014) argued that employee teamwork is a key driver of organization which has a significant impact on job satisfaction of employees. They also argued that salary satisfaction is being acted as the most important catalyst for enhancing the level of job satisfaction of the employees.

Pfeffer and Moore (1980), Pfeffer (1983, 1985), Wagner, Pfeffer and O’Reiily (1984) and various other scholars have observed the demographic constitution of an individual that influences several important behavioural pattern (job tenure, communication with the firm, job transfer, promotion and turnover) of the employees. Both ‘age’ and ‘job tenure’ of employees are appeared to be the important influential factors in determining voluntary employee turnover (Bluedorn, 1982; Price & Mueller, 1986) and are found to be negatively related with turnover (Cotton & Tuttle, 1986; Griffeth, Horn & Gaertner, 2000). Studies have also carried out to establish the relationship between gender and turnover and it has appeared that Mangione’s (1973) study found no significant relationship; Stumpf and Dawley (1981) observed that men were more likely to quit the job but the studies by Hom and Griffeth (1995) and Cotton and Tuttle (1986) found that women are more likely to leave the organization than men.

Research Objectives

Voluntary turnover of the talented employees has attracted much attention among academic and practitioners because they often comprise the organization’s core human capital and their turnover affected significantly on organization’s competitive advantages (Houkeslnge, 2001; Shaw, 1999). Employees’ socio-economic environment is the manifestation of their attitude towards life and work and the prevailing job opportunities add a positive effect on their propensity to change organization. Information technology sector is the leading service sector in the world as well as in India and employee turnover appears to be the highest in this sector. For achieving or maintaining the market competitive advantage, it becomes necessary to attract as well as to retain the talented employees. The objectives of this study are (i) to unveil the causal factors behind the IT employee turnover under the Indian socio-economic and market environments; (ii) to examine whether there are any differential reasons for leaving an organization between the managerial and technical categories of employees; and (iii) to examine the role of employees’ attitude towards life and work on employees’ turnover propensity.

Theoretical Framework

The employee turnover phenomenon is the consequence of various impulsive factors. We classify these factors into exogenous pull and endogenous push factors which insist or compel an employee to leave an organization voluntarily. For the sake of simplicity, we assume that the goal of an employee is to optimize the professional achievement and the employee will always accept any available better offer for upgrading his or her professional career. It is also assumed that the alternative job opportunities are available in the market.

Let, Q_t implies employee’s voluntary decision to leave an organization at time t, and P t l and P t h are the respective impulsive pull ¹ factors and push ² factors at time t. Then,

Q t = Q ( P t l , P t h ) ; d Q t d P t l > 0 , d Q t d P t h > 0

(1)

Again, let an employee is confronting with X₁, X₂, …, X_n, exogenous pull factors and Y₁, Y₂, …, Y_m, endogenous push factors. That is,

P t l = f ( X 1 , X 2 … … X n ) ; d P t l d X t > 0 , d P t l d X 2 > 0 , … … … . . d P t l d X n > 0

(2)

P t h = f ( Y 1 , Y 2 … … Y n ) ; d P t h d Y t > 0 , d P t h d Y 2 > 0 , … … … . . d P t h d Y m > 0

(3)

Thus, the voluntary decision of employees to quit (Q_t) an organization depends on a number of factor impulsions and the impact of these factor impulsions varies from employee to employee. If we assume a linear relationship between Q_t and its predictor variables, then the required equation will be of the following form:

Q t = α + β i ∑ ​ n i ​ ​ = ​ ​ 1 X t i + γ j ∑ ​ m j ​ ​ = ​ ​ 1 Y t j + ε t

(4)

where ε_t is the residual error at time t.

However, the outcome of Q_t is reflected only when the decision of the employee has been turned up, that is, either the employee quits or stays in the organization. Then, the dependent variable Q_t becomes a dichotomy. If we assign values 0 and 1 to employee’s staying in and quitting the organization, respectively, then the coefficient of each independent predictor will show their respective contribution to the variation of Q_t. From the relevant independent predictors and coefficient, our objective becomes not to find a numerical value of Q_t as in linear regression but the probability (θ) that it is 1 rather than 0. Then, the outcome will not be a prediction of a Q_t value but a probability value which can be any value between 0 and 1. Normalize the distribution by log transformation and this log transformation of the θ values to a log distribution enables us to formulate a normal regression equation. The log distribution (or logistic transformation) of θ is the log (to base e) of the odds ratio that the dependent variable is 1 and is defined as

log [ θ 1 − θ ] = ln [ θ 1 − θ ] ,

where θ ranges between 0 and 1.

Hence, the required equation becomes

I n [ θ 1 − θ ] = α + β i ∑ ​ n i ​ = ​ 1 X t i + γ j ∑ ​ m j ​ = ​ 1 Y t j + ε t

(5)

where P(Q_t = 1) = θ and P(Q_t = 0) = (1 − θ)

Data and Research Methodology

Primary information regarding the causal factors of employee turnover in the IT sector is collected by circulating a preordained questionnaire among 460 employees working in 17 different reputed IT firms in Kolkata. The questionnaire contains multidimensional questions pertaining to capture the behavioural patterns of the IT/Information Technology enabled Service (ITeS) employees to leave a company under the influences of different exogenous and endogenous factors. Out of 460 respondents, 420 have changed at least one company and out of them 214 respondents have changed three or more companies.

In the present study, we have considered only those employees who have at least changed one company at the time of survey. The six plausible causal factors (e.g., ‘higher salary’ [HS], ‘higher portfolio’ [HP], ‘higher company brand name’ [HCBN], ‘others’ [OTH]; employee’s age [AG] and employee’s attitude [ATT] towards life and work) are considered to be influential for the Indian IT/ITeS professionals to leave their jobs voluntarily. The respondents are asked to rank the first four motivating factors according to their rationale of leaving the preceding companies. Here, the implicit assumption is that any decision of the employees depends on their attitudes towards life and work. Therefore, to capture the employee’s attitude, we categorize our respondents into two groups in accordance with respondent’s own assessment—one group who have given ‘Highest Priority to Work-Life’ (HPWL) and the other group who have given ‘Highest Priority to Social-Life’ (HPSL). We developed a theoretical framework on employee turnover and on that basis we built up the logistic regression models. In order to capture the differential reasons between managerial and technical category of employees, we have constructed two separate logistic regression models, one for the managerial category and the other for the technical category of employees and have done a comparative study between them. Among our sample respondents, there are 113 managerial and 307 technical categories of employees who have changed at least one company. When the outliers are excluded, the numbers of managerial and technical samples become 102 and 287, respectively. A descriptive statistics has been presented for the 420 sample respondents (of which 113 are managerial category and 307 technical category employees) who have at least changed one company at the time of survey with respect to their gender, number of company changes and their experience in IT by occupational category and age group.

Some Empirical Observations

There are 113 managerial (males = 87 and females = 26) and 307 technical (males = 214 and females = 97) categories of employees in the 420 sample respondents who have at least changed one company at the time of survey (Table A1). In both the categories, employees of the age group of 30–40 is greater in number (61 out of 113 and 177 out of 307 in the managerial and technical category, respectively) compared to other two age groups. The employees who have changed three or more companies are observed to be more in the age group of 30–40 compared to other two groups. The average experience in IT of the managerial category employees (8.4 years) is only 1 year greater than that of the average experience of the technical category of employees (7.3 years).

Respondents were asked to rank the six given plausible reasons (‘higher salary’, ‘higher portfolio’, ‘higher company brand name’, ‘scope of foreign assignment’, ‘breach of commitment’ and ‘others’) for leaving their preceding company. The respondents have given rank-1 to rank-6 in accordance with their most to least important reason for leaving the preceding company. The distribution of rank-1 will show the distribution of prime reasons for leaving the company. It has been observed that the attraction of ‘higher salary’ is the prime reason for most of the managerial and technical category of employees (see Table A2). It is also observed that the younger employees are mostly attracted by the ‘higher salary’ offers. In other words, the ‘higher salary’ attraction, in general, enhances the employees’ propensity to leave the company and mostly affects the younger employees irrespective of their professional category. The second highest prime reason for attrition appears to be ‘higher company brand name’ followed by ‘higher portfolio’ ‘others’, ‘breach of commitment’ and ‘scope of foreign assignment’ (see Table A3). One important reason to be noted here is that the IT employees least bother for the ‘scope of foreign assignment’ and they are least affected by the employers’ ‘breach of commitment’ and thus the effect of these two reasons is least to employees’ propensity to change the company.

Model

Dependent Variable (Y)

Employee’s propensity to change a company is the dependent variable of the model. We define employee’s propensity to change companies as follows:

E m p l o y e e ’ s p r o p e n s i t y t o c h a n g e c o m p a n y = E m p l o y e e ’ s I T exp ​ e r i e n c e ( i n y e a r s ) E m p l o y e e ’ s n u m b e r o f c o m p a n y c h a n g e s

This ratio is the average time that an employee has remained in one job or in other words, this ratio is employee’s average propensity to change a company. Higher value of the above ratio indicates lower propensity to change and vice versa. We classified our respondents into two groups—high-propensity group and low-propensity group. The median value of this employee’s propensity is taken as a cut-off value. The employees having median value of propensity to change or the less than the median value are assigned 1 (e.g., high-propensity group) and values above the median value are assigned zero (e.g., low-propensity group). We therefore make our dependent variable a dichotomous one by putting 0 for those employees who have low-propensity to change company and l for those who have high-propensity to change company.

Dependent variable (Y) becomes a dichotomous variable and becomes Y = ln ( p ^ 1 − p ^ )

Therefore, we fit a linear logistic regression model which is of the following equation form:

I n ( p ^ 1 − p ^ ) = β 0 + β 1 X 1 + β 2 X 2 + β 3 X 3 + β 4 X 4 + β 5 X 5 + β 6 X 6

Here, p ^ = probability (Y = 1) implies the probability of an employee to quit the company and ( 1 − p ^ ) = probability (Y = 0) implies the probability of an employee to stay in the company.

Results of the Logistic Regressions

Concluding Remarks

The production of the organizations in the IT sector is very much labour intensive. Any loss of potential human resource of an individual organization would undoubtedly be a great loss to the organization. However, when inefficient employees leave the organization and are replaced by comparatively efficient ones, then it would certainly be beneficial to the organization. Therefore, the problem of employee turnover, in true sense, is the problem of voluntary turnover of high potential employees.

Among the six plausible factors, ‘higher salary’ appears to be the prime reason of most of the IT employees for leaving an organization. Next to salary, it is the ‘higher portfolio’ followed by attraction of ‘higher company brand name’—all are in the array of pull factors. This behavioural pattern persists uniformly among all the IT employees across gender and ages. One distinctive feature is that the propensity to change company is much higher among younger IT employees reflecting their zeal to reach at the top of the professional ladder within the shortest possible time. There are distinctively different strengths of the causal factors that are observed between managerial and technical category of employees. The strength of HS attraction appears to be the highest among other causal factors to the managerial category of employees, whereas, in case of technical category of employees, it is the attraction of HP. Employees’ attitude, especially for the managerial category of employees, appears to be one of the important predominant factors behind their propensity to change organizations.

Most of the employee turnover model established that job satisfaction plays a key role in the employee turnover phenomenon. Job satisfaction is a positive stable emotional state of mind resulting from the appraisal of various job-related issues, such as salary, portfolio, company’s brand name, working condition, etc. All these job-related issues are highly related to socio-economic well-being of an employee. An individual employee can attain the stable state by virtue of a single factor or a combination of two or more factors. The choice of factors towards attaining this stable state depends on the basic attitude of the employee. Therefore, one can intuitively say that job satisfiers (the prime factor[s] of job satisfaction) may differ employee to employee in accordance with their attitude towards life and work. This necessitates studying of attitudes of highly potential employees to facilitate employee retention policy.

In this connection, one may also argue that the employee turnover phenomenon is encouraged by the organization’s recruitment policy. Every organization tries to recruit the employees who are well qualified and experienced and thereby indirectly encouraging the employee turnover because recruitment of experienced employees enhances the employee turnover of employee’s ex-organizations.

Considering only six plausible causal factors is the limitation of the present study. However, the present study tries to initiate a new way of thinking by classifying the causal factors into push and pull factors, focused on some social–economic dimensions of employee turnover behaviour across age group and gender and professional category and that would be helpful to employee retention policy formulation as well as for the future research.

Organizational Policy Implications

It would be desirable that the HR practitioners should acknowledge the differences between individual goals of employees and the overall organizational objectives. Any employee retention strategy should therefore aim to satisfy both the objectives, that is, to converge these two apparently diverse objectives as far as possible. Therefore, at the outset, it would be beneficial as well as necessary to know exactly where the organization and the employee stand in relation to each other’s expectations. Keeping in mind the nature and the characteristics of the turnover phenomenon that revealed in our study on IT professionals, we are suggesting a few retention strategies as organizational HR policies which we hope will have a greater implication on talent management.