Rewards, autonomous motivation and turnover intention: Results from a non-Western cultural context

Abstract The purpose of this study is to examine the influence of two reward types (i.e., monetary reward and non-monetary rewards, such as competence development, autonomy support, and recognition) on autonomous motivation and further explore whether autonomous motivation plays a mediating role in the relationships between rewards and turnover intention. The study used a survey data from 100 employees working in public sector banks in Pakistan. The hypothesized relationships were assessed using partial least squares structural equation modelling technique. The results revealed that monetary reward and competence development were positively related to autonomous motivation, which in turn had a negative association with turnover intention. The indirect effects of rewards on turnover intention were only supported for monetary reward and competence development, as there was no significant link from autonomy support and recognition to autonomous motivation. We discuss implications for research and practice.


PUBLIC INTEREST STATEMENT
Human capital is recognized to be a critical resource of firm performance, but to capitalize on its human resources organizations need to manage the issue of employee turnover.Earlier research supports the impact of rewards on turnover intentions, however, majority of studies within this area have paid less attention to the role of rewards in intrinsically motivating employees and subsequently reducing their quit intentions.Moreover, much of such evidence comes from a Western cultural context.The current study contributes to a better understanding of the efficacy of rewards by simultaneously examining the role of two reward types (monetary and non-monetary) in shaping autonomous motivation, and in turn, turnover intention, and testing the study assumptions in a non-western cultural context.Our findings suggest that a fair and noninstrumental compensation package, and competence and skill development practices might be a very efficient vehicle to foster employee self-motivation, and in turn, employee retention.

Introduction
The importance of rewards in managing employee turnover has received considerable attention in the management literature (De Gieter & Hofmans, 2015;Tymon et al. 2011;Kuvaas, Buch, Gagne, Dysvik, & Forest, 2016).Although the impact of rewards on turnover intention has been extensively examined, the bulk of research within this area has focused on either monetary or non-monetary aspects of rewards (Gillet, Gagné, Sauvagère, & Fouquereau, 2013;Kim & Fernandez, 2017) and has failed to consider motivation as an intermediary mechanism (De Gieter, De Cooman, Hofmans, Pepermans, & Jegers, 2012;Weng & McElroy, 2012).Research on the role of autonomous motivation in the relationship between monetary rewards and turnover intention is even more scarce (Gerhart & Fang, 2015).Nonetheless, there are quite a few studies in the work domain that have examined motivation as a mechanism between rewards and turnover intention, but even these studies have focused on one type of reward (monetary or non-monetary), and their findings on the link between compensation and autonomous motivation are inconsistent.For example, Gillet et al. (2013) used autonomous motivation as a mechanism between rewards and turnover intention, but the authors included only non-monetary aspects of rewards.Kuvaas et al. (2016) examined the effects of monetary compensation on turnover intention with intrinsic and extrinsic motivation as mediators, but these authors found a negative relationship between annual performance pay and autonomous motivation.Olafsen, Halvari, Forest, and Deci (2015) proposed a positive association between pay and autonomous motivation, but their findings did not support their hypothesis.The association between the amount of performance pay and intrinsic motivation led to a positive relationship in the study by Kuvaas, Buch, and Dysvik (2018).Thus, the predictive ability of financial rewards in influencing autonomous motivation is yet to be established and further exploration is needed regarding whether monetary aspects of rewards have an incremental predictive validity over non-monetary rewards in the explanation of autonomous motivation.
Traditionally, giving money in exchange for work has been assumed as less conducive for employee psychological need fulfillment (Deci, Koestner, & Ryan, 1999), which led to a reduced attention by researchers in examining the role of autonomous motivation in the link between monetary rewards and employee outcomes.However, the recent understanding suggests that financial compensation is not necessarily bad for motivational quality, and it can even contribute to autonomous motivation through an informing effect and satisfaction of competency and autonomy needs (Gagné & Forest, 2008).Many recent studies support this notion by arguing that monetary rewards may lead to autonomous motivation if monetary incentives elicit justice perceptions (Olafsen et al., 2015;Ryan & Deci, 2017), and when rewards are less contingent on performance (Balkins et al. 2015) and have an informing rather than controlling effect (Kuvaas et al., 2018;Thibault Landry et al., 2017).These authors argue that rewards allocated in this manner nurture feelings of competence and autonomy, which, in turn, support higher motivational quality.
Most previous rewards research also comes from a Western cultural context.However, there are a few exceptions.For example, Chiang and Birtch (2012) conducted a comparative study of Finland and Hong Kong, to examine the performance implications of monetary and non-monetary rewards.Although these authors did not use motivation as a mediating mechanism, cultural differences played an important role in their study.Jang, Shen, Allen, and Zhang (2018) adopted a cross-cultural perspective examining how turnover intentions are determined by certain job resources such as job control and participation in decisions.The findings revealed that these relationships vary as a function of cultural dimensions of collectivism and uncertainty avoidance (UA).This suggests that the relationship patterns between incentives and employee motivation, and turnover intention may vary across cultures (c.f., Chiang & Birtch, 2007).This is because cultural characteristics have been posited to affect employees' reactions to certain stimuli within a culture due to the influence of the societal level cultural values on individuals' cognitive structures and personal values (Mustafa & Lines, 2013, 2012;Peterson & Barreto, 2014).
The present study contributes to the existing literature by simultaneously examining the role of two reward types in shaping autonomous motivation, and in turn, turnover intention.By testing a model that incorporates both monetary and non-monetary rewards, our study explores whether monetary compensation has an incremental predictive validity over non-monetary rewards in explaining autonomous motivation.As we test our assumptions among employees of public sector banks in a non-Western cultural context, our findings will contribute to a better understanding of cultural specificity versus generalizability of employee motivational reactions in the face of rewards.The study will specifically offer some interesting insights regarding the efficacy of rewards in a comparatively under-researched country, Pakistan, with particular relevance to the public sector banks in the country.

Theoretical background and hypotheses
Rewards are categorized as monetary and non-monetary rewards.Monetary rewards include financial compensation, such as base pay, performance pay, and other financial incentives, such as commission, bonus, etc.Among non-monetary rewards, empowerment, competency development, and employee recognition are the core categories (Armstrong & Murlis, 2007).An effective reward system is considered essential for motivating and retaining employees (Singh, 2003).Motivation, which is assumed to act as the primary mechanism to explain the effects of rewards on turnover intentions (Gerhart & Fang, 2015), is distinguished as autonomous motivation and controlled motivation (Deci & Ryan, 2000).Autonomous motivation represents engaging in an activity with complete free will and choice, while controlled motivation denotes that a person behaves in response to an externally produced inducement (Deci & Ryan, 2008;Ryan & Deci, 2000).Past research indicates that, on average, autonomous motivation leads to more positive work outcomes as compared to controlled motivation (Cerasoli, Nicklin, & Ford, 2014;Deci & Ryan, 2008).Nonetheless, there have been rare attempts to examine the influence of the monetary and non-monetary aspects of rewards on autonomous motivation.Monetary rewards have generally been linked to controlled motivation, and the effects of non-monetary rewards (e.g., autonomy support) have mainly been examined in the extent to which employees have a self-motivation to perform their work.According to recent assertions, irrespective of the category, rewards can raise autonomous motivation if organizations convey the message of their focus on competence and capability through rewards, and such a message is stronger when it is conveyed through various sources, such as contingent pay based on assessments of competence, informational feedback, and acknowledgment of the individual (Sanders et al., 2018).

Monetary reward and autonomous motivation
Using an self-determination theory (SDT) perspective, it is generally argued that giving money in exchange for work is transactional, and thus, does not address employees' autonomy, competence, and relatedness needs.Therefore, motivational quality tends to be lower when compensation is used as the primary driver for motivating employees at work (Kuvaas et al., 2016;Kuvaas, Dysvik, & Buch, 2014).However, many recent studies contend that compensation can contribute to motivational quality if the way in which the level of pay is determined is perceived to be fair and just (Gagné, Bérubé, & Donia, 2007;Manganelli, Thibault-Landry, Forest, & Carpentier, 2018;Olafsen et al., 2015) and rewards are delivered in a manner that highlights the competence of recipients (Ryan & Deci, 2017) and recognizes volitional behavior (Thibault Landry et al., 2017).Rewards can contribute to the feelings of competence and recognition of volitional behavior, for example, by offering monetary incentives in a way that employees are not aware of the amount, form, and timing of the incentive, and by allowing employees more discretion in selecting meaningful performance outcomes, and the means of attaining them (Balkins et al. 2015).Rewards allocated in this manner may contribute to the satisfaction of competence and autonomy needs, which, in turn, may lead to valuable employee outcomes, such as increased autonomous motivation (Manganelli et al., 2018;Olafsen et al., 2015;Thibault Landry et al., 2017).This implies that monetary compensation has positive effects on autonomous motivation when it is less linked to the achievement of targets (Thibault Landry et al. 2018) and is offered on an ex-post basis using generalized and broad performance measures (Balkin et al., 2015).
The positive effects of compensation on higher motivational quality supports the assumptions of cognitive evaluation theory (CET) (Shalley & Perry-Smith, 2001).According to CET, rewards may have a controlling or informational effect, and incentives that have a controlling effect undermine, while those with an informing effect boost intrinsic motivation (Kuvaas et al., 2018).In the compensation context, rewards that are less contingent on particular performance levels may have an informing effect, and thus, may benefit autonomous motivation (10 Kuvaas et al., 2018).This suggests that even extrinsic rewards that offer informational feedback about performance may have positive implications for autonomous motivation (Amabile & Pillemer, 2012).
The assertion that compensation may not necessarily be bad for autonomous motivation is also consistent with the social exchange theory.If the allocated reward is low on instrumentality (it is not tied to short-term performance but portrays a broad range of future behaviors and expectations, and its level reflects long-term diffuse exchanges in the past), then it is a gesture of an employee's worth to the organization, and thus, may foster a social exchange relationship between the two (Kuvaas et al., 2016).Based on the above, we suggest the following: H1.Monetary reward is positively related to employee autonomous work motivation.

Autonomy support
Autonomy support, such as offering opportunities to experience choice and self-organize, are posited to have a positive impact on the healthy functioning of individuals (Ryan & Deci, 2000).Autonomy supportive interpersonal environments have been found to encourage more autonomous motivation in different contexts (Gillet et al., 2013;Koponen, Simonsen, & Suominen, 2017;Kuvaas, 2009;Muraven, Gagné, & Rosman, 2008;Nie, Chua, Yeung, Ryan, & Chan, 2015;Slemp, Kern, Patrick, & Ryan, 2018).This is because situations that are autonomy supportive are conducive for the satisfaction of basic psychological needs (Deci & Ryan, 2000).Earlier research suggests that situations that support greater autonomy nurture autonomous motivation because individuals show more endorsement and commitment to a particular course of action when they freely choose that course of action based on its congruence to their needs and desires (Deci, Connell, & Ryan, 1989).Likewise, previous evidence shows that employees' feelings of self-worth (Elloy & Randolph, 1997;Shamir, House, & Arthur, 1993) and their sense of competence (Conger & Kanungo, 1988;Mustafa, Glavee-Geo, Gronhaug, & Saber Almazrouei, 2019) is raised when they experience opportunities to exercise self-direction and selfcontrol.Moreover, it has been argued that decentralized structural conditions that offer greater empowerment and autonomy to employees are likely to foster organizational justice perceptions (Schminke, Cropanzano, & Rupp, 2002), and perceptions of justice and fairness have been posited to improve autonomous motivation through psychological need satisfaction (Olafsen et al., 2015).Consequently, we suggest the following: H2.Autonomy support is positively related to employee autonomous work motivation.

Competence development
Organizations offer their employees the opportunity to increase their competence through developmental opportunities, such as job rotation, training, and further education (Jamison & O'Mara, 1991;Pfeffer, 1998).Employees generally value competence development practices (Boselie, Dietz, & Boon, 2005), and those who perceive high developmental prospects show positive performance outcomes and have a higher inclination to stay with the current organization (Dysvik & Kuvaas, 2008;Kraimer, Seibert, Wayne, Liden, & Bravo, 2011).Growth and development practices convey a message to employees that their employability is cared for, and that their contribution is highly valued by their organization (Lee & Bruvold, 2003).Moreover, the development opportunities imply that the organization trusts the current abilities of its employees and wants them to develop further.This suggests that development practices reflect an organization's focus on nurturing employee competence and capabilities and their worth and belongingness to the organization.Thus, development practices may lead to the satisfaction of relational and competence needs that, in turn, boost autonomous motivation (Gagné & Deci, 2005;Richer, Blanchard, & Vallerand, 2002;Thibault Landry et al., 2017).Consequently, we suggest the following: H3. Competence development is positively related to employee autonomous work motivation.

Recognition
Organizations also use recognition practices (e.g., respecting one's perspective, appreciation letters, award ceremonies, and recognition plaques) to motivate employees.Recognition is argued to be a constructive response to an employee's contribution that is reflected by his or her engagement and commitment to work.Recognition also represents an evaluation and celebration of an employee's professional endeavors and results produced by him or her and appreciated by the organization (Brun & Dugas, 2008).Previous research suggests that acknowledging employees' effort and good work has beneficial effects on their psychological outcomes, such as morale and self-esteem (Rosen & Berger, 1991), which may act as a source of intrinsic motivation (e.g., Sheldon, Elliot, Kim, & Kasser, 2001).Moreover, employee recognition has also been argued to be a key factor in building meaningfulness of work (Grawitch, Gottschalk, & Munz, 2006), which fosters intrinsic work motivation (Hackman & Oldham, 1976).It has been further posited that the acknowledgment of an individual (e.g., praise and recognition) acts as a channel that underpins an organization's focus on competence and capability (Sanders et al., 2018).Likewise, respecting employees' feelings and perspective signals that organizations acknowledge and recognize the abilities of their employees; thus, allowing them to believe in their prowess and competence (Hirst, Van Knippenberg, Chen, & Sacramento, 2011).Therefore, we propose the following: H4.Recognition is positively related to employee autonomous work motivation.

Autonomous motivation and turnover intention
Previous research showed that employees are less inclined to quit their jobs if their autonomy is supported (Gagné, 2003).Empirical evidence suggests that autonomous motivation is negatively associated with turnover intentions (Dysvik & Kuvaas, 2008, 2010;Kuvaas et al., 2016).The importance of autonomous motivation lies in one's behaving in accordance with one's choice and free will and engaging in an activity without an externally induced pressure (Deci & Ryan, 2008).Thus, it may be reasonable to expect that employees who feel a sense of volition and choice in their jobs are less likely to leave the organization or to seek alternative employment.Thus, we suggest the following: H5.Autonomous work motivation is negatively related to turnover intention.

Autonomous motivation as a mediator
High autonomous motivation is believed to have the potential to reduce turnover intention which is evident from the negative effects of autonomous motivation on turnover intention in a wide variety of settings (Dysvik & Kuvaas, 2008, 2010;Kuvaas et al., 2016;Richer et al., 2002).Previous studies also showed that monetary rewards (Kuvaas et al., 2018) and non-monetary rewards, such as autonomy support (Gillet et al., 2013;Pelletier, Fortier, Vallerand, & Briere, 2001), promote autonomous motivation.Besides, several researchers have shown that the relationship between monetary and non-monetary rewards and turnover intention is negative.For example, research showed that the level of intention to stay in an organization is high when employees receive recognition from their organizations (AbuAlRub & AL-Zaru, 2008;Bhatnagar, 2014).Likewise, previous research showed that when employees perceive high levels of developmental support, the organization benefits in terms of lower turnover (Kraimer et al., 2011;Nerstad, Dysvik, Kuvaas, & Buch, 2018).The decreased turnover intention may represent ways by which employees can recompense their organization for its development practices (Allen, Shore, & Griffeth, 2003) and its support and care for employees (Lee & Bruvold, 2003;Wayne, Shore, & Liden, 1997).As far as monetary rewards are concerned, the compensation level is assumed to have a sorting effect, such that those with a high compensation level are likely to have a lower turnover intention (Gerhart & Rynes, 2003;Kuvaas et al., 2016).
From the above, we can infer that when rewards signal employee worth to the organization, recognize volitional behavior and acknowledge effort and performance in a way that may enhance feelings of competence, then employees' motivation toward work will emanate from their integrated values and interests, which, in turn, will decrease their turnover intentions.Thus, we hypothesize the following: H6a.The relationship between monetary reward and turnover intention is mediated by autonomous motivation.
H6b.The relationship between non-monetary rewards (autonomy support, competence development, and recognition) and turnover intention is mediated by autonomous motivation.

Sample and data collection
The participants in this study were Pakistani employees working for four different public sector banks.The participants were full-time employees, mainly working in two cities, Rawalpindi and Islamabad.Overall, 120 survey questionnaires were distributed via e-mail, inviting potential participants to complete the online survey.A total of 100 participants completed the survey, resulting in a very satisfying response rate of 83.3%.The final sample included 79% male and 21% female participants.On average, participants were 31.6 years old (minimum = 23 years; maximum = 52 years).62.6% of the participants held an MBA degree, while 37.4% had a bachelor's degree in business.The majority of the participants (86.9%) had tenure of between one and ten years at their current organization, 11.1% had worked there for between 11 and 20 years, and only 2% had work experience of more than 21 years.The sample consisted of 29.3% managers, 12.1% officers in different grades, 11.1% trade officers, and 47.5% employees in other categories such as those responsible for clerical work, accounting, data entry, and cash management etc.

Measures
All the constructs were measured using previously validated scales (see appendix A).To measure autonomous motivation, the scale developed by 64 Kuvaas (2006) was used.Monetary and nonmonetary rewards (autonomy support, competence development, recognition) were assessed following Tremblay, Rondeau, and Lemelin (1998) and Paré and Tremblay (2007).Turnover intention was tapped following Alexandrov, Babakus, and Yavas (2007) and Singh, Verbeke, and Rhoads (1996).We measured all the items on a 5-point scale where 1 and 5 stood for strongly disagree and strongly agree respectively.
We controlled for the effects of age, gender, tenure, education, and job type.

Results
The analysis was conducted using SmartPLS software.SmartPLS is a partial least squares path modeling technique that simultaneously tests the measurement (the relationship between indicators and their latent constructs) and the structural model (the relationship between constructs).PLS is very useful for model estimation when the sample size is small, and when the model is complex (Hair, Hult, Ringle, & Sarstedt, 2014).

Assessment of the measurement model
The measurement model attempted to confirm whether the manifest variables correctly capture the theoretical constructs.We assessed the measurement model with respect to individual item reliability, internal consistency and discriminant validity.For accepting item loadings, we used the minimum level of 0.05 (Barclay et al., 1995).Two items from competence development (CD1 and CD2) and one item each from turnover intention (T1) and autonomous motivation (AM5) were deleted from subsequent analysis for showing poor loadings.The loadings for the rest of the indicators exceeded 0.630, suggesting an adequate correlation between the indicators and their respective constructs (Wetzels, Odekerken-Schröder, & Van Oppen, 2009).Moreover, all the composite reliability (CR) ratios were above 0.7, which indicates adequate internal consistency of the measures.Fornell and Larcker's (1981) criterion was used to assess convergent and discriminant validity.Convergent validity is confirmed if the average variance extracted (AVE) exceeds 0.50.The AVE for all constructs was above 0.5, which establishes the convergent validity of the latent constructs.The results also suggest the existence of discriminant validity among the constructs.The discriminant validity is confirmed if the square root of AVE (diagonal elements) is higher than the latent variable's correlation with other constructs (off-diagonal values in the corresponding rows and columns).

Common method variance
We collected data from the same respondents using a one-time survey that could potentially lead to the occurrence of common method variance.To assess the presence of any such concerns, we used Harman's (1976) one-factor test which assumes that the common method variance might exist if the unrotated factor solution results in a single factor or one factor explains most of the variance in the variables (Podsakoff & Organ, 1986).An exploratory factor analysis showed that the largest factor explained only 28.4% of the variance, which indicates the absence of any common method variance related issues in our data.

Assessment of the structural model
The significance of the path coefficients was assessed with bootstrap analysis in SmartPLS3.Figure 1 shows the path estimates of the model's structural main direct effects between the latent variables.Table 3 presents path coefficients, t-values, effect size and variance inflation factor (VIF) scores.The results of the estimation of the inner model revealed that it explained 46.5% of the autonomous motivation variance and 18.6% of turnover intentions.Of the 46.5% variance in autonomous motivation, 13,1% was contributed by monetary reward alone, which suggests the incremental validity of monetary reward above non-monetary rewards in explaining autonomous motivation.Further, the path coefficients showed that only two of our four propositions concerning the influence of rewards on autonomous motivation were supported.In support of hypothesis 1, we found a significant positive relationship between monetary reward and autonomous motivation (β = 0.417; p < 0.001).Contrary to our expectation, the estimation of the structural model offered no support for a significant relationship between autonomy support and autonomous motivation (H2; β = 0.177; p = 0.098).The results of the analysis showed support for the positive effect of competence development on autonomous motivation (H3; β = 0.241, p < 0.05), while the effect of recognition on autonomous motivation was not supported (H4; β = 0.065; p = 0.617).The study also proposes that autonomous motivation influences turnover intention (H5).The data support this relationship (β = -0.396,p < 0.001).
The study proposes that monetary and non-monetary rewards will have an indirect effect on turnover intention via autonomous motivation (H6a and H6b).The results show that monetary reward has a significant link with autonomous motivation and autonomous motivation, in turn, is a significant predictor of turnover intention.Likewise, competence development influences turnover intention indirectly, as competence development is a significant predictor of autonomous motivation, which in turn has a significant negative association with turnover intention.The bootstrapping estimations supported the indirect influence of monetary reward and competence development on turnover intention respectively (β = -0.177;p < 0.01; β = -0.102;p < 0.05).The mediation effects for relationships from autonomy support and recognition to turnover intention were not supported, as no direct link from these two factors to autonomous motivation is present.Thus, the data support H6a, but partly support H6b.Table 4 shows results of significant indirect effects.

Discussion
This study examined the impact of monetary and non-monetary rewards on autonomous motivation and further explored whether autonomous motivation plays a mediating role in the relationships between rewards and turnover intention.Results revealed that monetary compensation and competence development significantly enhance autonomous motivation, which, in turn, reduces turnover intention.Our findings underscore the importance of monetary incentives (in addition to non-monetary incentives) to improve autonomous motivation and to encourage employee retention.The study further indicates that individuals' motivational reactions in the face of rewards may differ across cultures.
The positive impact of monetary reward on autonomous motivation is consistent with recent assertions that compensation is not necessarily detrimental to the motivational quality if it is low on instrumentality and when it is perceived to be fair (Kuvaas et al., 2016;Olafsen et al., 2015).One reason for our finding a positive relationship between monetary reward and autonomous motivation might be that our scale mainly captured responses about an equitable base compensation.Base compensation has been argued to be low on instrumentality because the contribution expected in return for base pay is often relatively vague and diffuse (Kuvaas et al., 2016), and perceptions of justice associated with the allocation of monetary compensation fulfill employees' psychological need satisfaction, leading to higher autonomous motivation (Olafsen et al., 2015).
The present results may also be viewed as a reflection of the cultural values of Pakistani society.The national culture of Pakistan is characterized by high power distance (PD), UA, ingroup collectivist (Hofstede, 1980a), and masculine values (Shamim & Abbasi, 2012).It has been argued that individuals' cognitive structures and personal values are partly shaped by the societal values (Peterson & Barreto, 2014).The reason why an individual's values become partly similar to the overarching values in his or society is because those values are internalized as his or her own during early socialization processes (Jang et al., 2018;Mustafa & Lines, 2013).Earlier studies posit that values that are deeply internalized become part of one's inner self (Mustafa & Lines, 2012;Ryan & Deci, 2000) and individuals are intrinsically motivated to fulfill them (Ryan & Deci, 2000;Wan, Chiu, Peng, & Tam, 2007).Thus, the pattern of relationships we found in this study may, in part, stem from the cultural characteristics of Pakistani society.For example, the weaker association between autonomy support and autonomous motivation may stem from society's hierarchical values.Individuals in hierarchical societies value monetary incentives and view financial success as a means to greater employment security and status (Chiang & Birtch, 2012).Besides, the stronger preference for financial rewards may be linked to the increased masculinity.Masculine societies emphasize achievement and material success (Hofstede, 2001).In such cultures, financial incentives are likely to be highly attractive (Johnson & Droege, 2004), because affluence and admiration is highly valued, and economic gains are instrumental in achieving them (Hofstede, 1980b).Our results are also consistent with previous research that shows that individuals high on ingroup collectivist values demonstrate a greater preference for monetary benefits (e.g., Kickul, Lester, & Belgio, 2004).For example, Chiang and Birtch (2012) found that employees in Hong Kong (high on ingroup-collectivism) demonstrated a higher financial reward orientation than Finnish employees (low on ingroup-collectivism).
Why respondents in this study reacted more positively to competence development may also reflect the cultural emphasis in the country.Pakistan is a country where masculine characteristics, such as assertiveness, accomplishment, and ambition, are highly valued.Masculine employees prefer performance dimensions that emphasize personal achievement and accomplishment (Beer & Katz, 1998).This suggests that employees in a society with such cultural characteristics may react positively to motivators that offer opportunities to gain prowess and competence.The society's hierarchical values offer a compelling explanation for the weaker association between autonomy support and autonomous motivation.High PD cultures place a strong emphasis on authority and structure, which is manifested in complying with the decisions of superiors and showing reluctance to take initiatives and to accept additional responsibilities (Palich, Horn, & Griffeth, 1995).In such cultures, practices that provide greater autonomy and control to subordinates may receive less support because shifting superiors' power to followers so that they plan and schedule their own work is not in harmony with the emphasis placed on hierarchy and structure (Peretz & Fried, 2009).This is consistent with previous evidence that reveals that people in hierarchical societies are less receptive to delegation and participative leadership (Elenkov, 1998;House, Hanges, Javidan, Dorfman, & Gupta, 2004).
The weaker preference for recognition may, in part, be a reflection of low cultural femininity and UA values.Individuals in feminine cultures, as opposed to their counterparts in societies high on masculinity, tend to be more receptive to non-monetary incentives such as social recognition because of their relational and nurturing orientation (Cohen & Keren, 2008).In high UA cultures, annual salary increases, and other emoluments play a more crucial role in motivating employees than individuals in low UA cultures (Chiang & Birtch, 2012).

Practical implications
Our results should help managers better understand the interrelationships between rewards, autonomous motivation, and turnover intention.Our findings suggest that the amount of monetary compensation, especially the level of base pay and the annual salary increase that employees perceive as fair might serve as a strong indicator of how an organization values and owns it employees and may enhance autonomous motivation by boosting the feelings of relatedness.Among non-monetary rewards, competence development opportunities (e.g., on and off-thejob training, certification programs, and other formal and informal learning and growth opportunities) might be more appealing and more beneficial for increasing autonomous motivation and reducing turnover.Thus, in the current setting, a fair and non-instrumental compensation package may be a key building block to improving autonomous motivation, which determines inclination to leave the organization.Moreover, competence and skill development practices might be a very efficient vehicle to foster employee self-motivation, and in turn, employee retention.

Limitations and suggestions for future research
Our study is not without limitations.First, our sample comprised employees of public sector banks in a single country that could be extended to other occupational settings within Pakistan as well as in other countries to test the generalizability of our findings.Second, we used a single scale to capture monetary rewards, which mainly covered employee perceptions of base pay.Future studies should use different scales to capture base and performance contingent pay, to examine the unique effects of each category of monetary compensation on outcome variables.Moreover, we examined direct relationships between rewards and autonomous motivation without including need satisfaction and incentive effects in the model, which should be addressed in future studies.Further, we asserted that the pattern of relationships we found in this study might be a reflection of the cultural characteristics of Pakistan, but we did not include cultural variables in the study; future studies should examine whether certain cultural value dimensions have contingency effects in the relationship between rewards and autonomous motivation.Lastly, given the lack of research on how different rewards interact with each other to influence outcomes, future studies should explore whether rewards operate independently to affect autonomous motivation and turnover intention, or whether these outcomes are products of parallel effects/trade-offs between different rewards.You are free to: Sharecopy and redistribute the material in any medium or format.Adaptremix, transform, and build upon the material for any purpose, even commercially.The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms: Attribution -You must give appropriate credit, provide a link to the license, and indicate if changes were made.You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

No additional restrictions
You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.

Figure 1 .
Figure 1.Structural model estimation.*** p < 0.001, ** p < 0.01 (twotailed), ns Not Significant in our work unit have great latitude for the organization of their work (e.g., work schedules) AS2.Employees in my work unit have much autonomy in project management AS3.In my work unit, employees have considerable freedom regarding the way they carry out their work Recognition R1.In my work unit, employees' suggestions are seriously taken into consideration R2.In my work unit, supervisors tangibly recognize employees' efforts in different ways (e.g., sports events; dinners at restaurants) R3.In my work unit, employees receive written recognition from their supervisors.R4.In my work unit, supervisors regularly congratulate employees in recognition of their efforts.Monetary reward MR1.I estimate my salary as being fair internally MR2.My salary is fair in comparison with what is offered for a similar job elsewhere MR3.My compensation level adequately reflects the level of my responsibility in the organization MR4.The pay increases and/or bonuses I received in the last 2 years adequately reflect my recent performance evaluations Autonomous motivation AM1.The tasks that I do at work are enjoyable AM2.My job is so interesting that it is a motivation in itself AM3.The tasks that I do at work are themselves representing a driving power in my job AM4.My job is meaningful (IM) AM5.I feel lucky being paid for a job that I like very much Turnover intentions TI1.I frequently think about leaving my current organization.TI2.I often think about searching for a job in another organization.TI3.It is most likely that I will quit this job in near future.

©
2019 The Author(s).This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.
Cogent Business & Management (ISSN: 2331-1975) is published by Cogent OA, part of Taylor & Francis Group.Publishing with Cogent OA ensures: • Immediate, universal access to your article on publication • High visibility and discoverability via the Cogent OA website as well as Taylor & Francis Online • Download and citation statistics for your article • Rapid online publication • Input from, and dialog with, expert editors and editorial boards • Retention of full copyright of your article • Guaranteed legacy preservation of your article • Discounts and waivers for authors in developing regions Submit your manuscript to a Cogent OA journal at www.CogentOA.com

Table 1
shows loadings, and CR and AVE values.Discriminant validity coefficients are presented in Table2.
CR, composite reliability; AVE, average variance extracted

Table 2
Bold numbers on the diagonal show the square root of the AVE.Numbers below the diagonal represent the construct correlations.