Does Employee Pay Variation Increase Government Performance? Evidence From a Cross-National Analysis

Abstract

Pay variation across positions, functions, and ranks can affect government performance by influencing the ability of the government to recruit and incentivize civil servants, but this proposition has not been systematically examined. Taking advantage of a new panel dataset, we develop and test the theoretical linkage between pay variation of civil servants and government performance. Our findings show a contingency-based relationship between pay variation and government performance. On average, neither total pay variation nor vertical pay variation is significantly related to government performance measured by the World Bank’s Worldwide Governance Indicators. However, total pay variation is consistently and negatively correlated with government performance in low-income countries. The findings suggest the importance of accounting for national contexts in implementing administrative reforms and are a cautionary lesson about applying theories based on research on private firms to the public sector.

Keywords

pay variation government performance pay for performance cross-national analysis human resource management

Introduction

As most public compensation is funded by tax-supported public budgets, pay compression—salaries constrained rigidly across ranks and positions at the same ranks—makes it difficult to recruit and retain high-skill professionals and executives. Increasing pay spread across ranks and positions at the same rank could improve government performance (World Bank, 2014). Increasing pay variation enables the government to give high-performing individuals competitive salaries, an important factor for prospects’ decisions to work for the government (Liu & Tang, 2011; Schanzenbach, 2015; U.S Merit System Protection Board, 2008). Increasing pay variation can also increase the expected reward for promotion and performance and enhance the motivational effects of performance-based promotion and pay. The favorable sorting and motivational effect of pay variation can eventually improve government performance. Still, pay variation can breed a sense of pay inequity, dampen morale, and reduce collaboration and performance (Cowherd & Levine, 1992; Pfeffer & Langton, 1993). Ambiguity in performance evaluation and public employees’ limited control over measured outcomes in the public sector weakens the links among effort, performance, and pay (Pearce & Perry, 1983), compounding the challenge of perceiving pay variation as fair.

Despite its salience to talent acquisition, employee motivation, and organization performance, scholars have rarely examined pay variation (Borjas, 2002; Katz & Krueger, 1991; Keppeler & Papenfuβ, 2022; Schanzenbach, 2015). The scarcity of research warrants attention because pay variation may significantly affect the effectiveness of reforms and overall government performance. To fill the gap, we utilize a new cross-national panel data to examine the relationship between pay variation and government performance around the world. The larger number of observations over multiple years enables us to examine the relationship for countries at different stages of development. Specifically, we study two primary questions:

(1) Whether pay variation affects organization performance in the public sector?

(2) Does the pay variation-organization performance linkage differ across nations?

While we do not identify a universal relationship between total pay variation and government effectiveness across countries, we do find consistent evidence for their negative correlation in low-income countries. Drawing on previous research on pay variation (Conroy et al., 2014; Gupta et al., 2012), we argue that pay variation that can be justified by employees’ input should be differentiated from unjustified variation because the difference can reconcile competing expectations about the effects of pay variation on organizational performance (Trevor et al., 2012). We also find evidence in support of the claim. Justified pay variation did not reduce government effectiveness while unjustified one did. The finding can be attributed to the possibility that justified pay variation is perceived as fair and embodies procedural justice.

While prior studies have examined the relationship between horizontal pay variation, such as performance-related pay for agency heads and performance (Binderkrantz & Christensen, 2012; Papenfuß & Keppeler, 2020), few have examined how total pay variation and vertical variation affect government performance. This article is among the first to systematically examine how pay variation influences organizational performance in the public sector. Studying this relationship in the public sector is critical because findings in the private sector may not be generalized to the public sector. Research in the private sector generally holds that justified pay variation can improve organizational performance by attracting talents and motivating employees (Conroy et al., 2014; Gupta et al., 2012). Our study does not find significant positive correlations between pay variation and government effectiveness. We attributed the insignificant pay variation-government performance relationship to ambiguity in performance evaluation, relatively low valence of financial incentives to public servants, public employees’ limited control over measured outcomes due to extensive external controls and collaborative nature of work, and often limited amount of funding for rewarding civil servants. These factors can weaken the links among effort, performance, and pay and undermine the motivational effect of pay variation (Pearce & Perry, 1983). The finding underlines the importance of accounting for the differences between public and private sectors in public management research. It also serves as an illuminating example of how the differences between the public and private sectors affect managerial practices, in particular in the area of human resource management (HRM).

Moreover, we hold that favorable institutional conditions including merit-based recruitment and rule of law are precondition for effective sorting and the motivating effects of pay variation. Without these conditions, increasing pay variation can be seen as a tool of patronage and feeds the sense of pay inequity, undermining morale, and eventually government effectiveness. We indeed find that increasing pay variation is associated with lower government effectiveness in low-income countries where the favorable institutional conditions are likely to be lacking (Nistoskaya et al., 2021). The finding echoes the need for studying contextual conditions for possibly heterogeneous effects of pay variation (Pfeffer & Langton, 1993) This article paper establishes economic development level as a moderating condition for the effect of pay variation on government performance and contributes to understanding how contexts matter in public human resource management research. Advancing the contingency-based understanding of administrative reforms will further the shift in governance from best practice to best fit and the development of middle-range theory on administrative reforms considered the “bedrock for global public administration knowledge” (Perry, 2016, p. 533).

Literature Review and Theoretical Hypotheses

Employees expect decent income and pay increments with promotion or performance. Public employees also rate pay as an important reason for joining and staying in government in both developed and developing countries (Ang, 2020; U.S Merit System Protection Board, 2008). However, the variation of pay across employees is also much smaller than that of the private sector. Some countries have introduced performance-related pay and increased pay variation in the public sector (World Bank, 2014). Due to data limitations, however, few studies examine the relationship between pay variation and government effectiveness.

The effects of pay variation on performance have been shown to be mixed in the private sector (Conroy et al., 2014; Gupta et al., 2012; Rouen, 2020). Given the relative paucity of theoretical building for the relationship between pay variation and organizational performance in the public sector, we review and draw on the literature in the private sector to develop hypotheses. To better understand the mechanisms through which pay variation influences performance, it is helpful to know three types of pay variation: (1) horizontal pay variation, the variation in pay among people holding similar jobs; (2) vertical pay variation, the variation in pay between people in different positions of organizational hierarchy; and (3) overall pay variation, the variation within and across jobs.

First, tournament theory is often cited to explain how vertical pay variation can motivate employees to improve performance. In the tournament, employees compete for a prize that is awarded based on their relative rank in the contest (Lazear & Rosen, 1981). The prize can include promotions, pay raises, prestige, and privileges. The reward for winning the tournament motivates participants to exert more effort and improve their performance. The amount of increased effort rises with the difference between the payoff in winning and losing the tournament (Connelly et al., 2014). In addition, participants need to believe that they can win the tournament with extra effort. Otherwise, they will lose hope and stop working hard to win the tournament. The promotion tournament and the pay differential among members in the organizational hierarchy have been the focus of studies in the private sector. The studies have yielded mixed evidence (Connelly et al., 2014; Gupta et al., 2012). As we will explained later, differentiating pay variation that can be justified by employees’ inputs from the unjustified variation plays an important role in reconciling the mixed effects.

Expectancy theory is often used to explain how pay variation, especially the horizontal one, can enhance both employee motivation and organizational performance. Expectancy theory holds that individuals can be motivated toward goals if (1) expectancy that efforts lead to goal attainment is strengthened, (2) expectation of a reward for meeting performance goals is enhanced (instrumentality), and (3) the values individuals place on rewards is increased (valence; Vroom, 1964) By increasing the expected reward for improving performance (valence), expectancy theory predicts that pay variation will motivate public employees to exert more effort to improve their performance (Vroom, 1964). In addition, pay variation is often accompanied by performance management measures such as performance-related pay that enhances the instrumentality that performance can lead to higher pay (Conroy et al., 2014).

Equity theory often serves as a theoretical foundation for why pay variation is associated with lower performance. Equity theory holds that people tend to use a ratio of the perceived value of their input (e.g., effort and ability) to the compensation for their labor to assess whether they have received fair compensation (Adams, 1965). Researchers later argue that differences in pay primarily determine the perception of pay equity because people tend to have an upward bias in evaluating their own input (Cowherd & Levine, 1992). Both others’ salaries in similar positions (Akerlof & Yellen, 1990) and relative managers’ wages in comparison with entry-level employees in the same organization (Cowherd & Levine 1992) will moderate how people perceive the equity of pay (Trevor & Wazeter, 2006). When employees’ pay falls short of their expectations of fair pay, they may feel a sense of inequity. The sense of inequity decreases employee satisfaction, causes counterproductive behaviors like reduced collaboration and shirking, and eventually undermines performance (Conroy et al., 2014; Cowherd & Levine, 1992; Pfeffer & Langton, 1993).

To reconcile competing predictions about the effects of pay variation on performance, pay variation that can be justified by demographic and performance factors needs to be separated from pay dispersion that cannot be justified by these factors (Conroy et al., 2014; Trevor et al., 2012). The motivational effect of pay variation in both tournament theory and expectancy theory assumes that higher pay is the result of better performance whether it is measured in comparison with others competitors (tournament theory) or through goal attainment (expectancy theory). Studies in the private sector have proved that justified pay variation can enhance performance (Conroy et al., 2014). The justified pay variation for people holding similar jobs and for people at different ranks can mitigate the risk of pay variation being considered unfair by employees. The equity theory holds that people consider their inputs in comparison with others to evaluate whether they receive fair compensation. If pay variation is based on people’s inputs ranging from their education, experiences, and previous performance, employees may still consider pay variation to be just. By contrast, unjustified pay variation can reinforce the perception of pay unfairness and undermine performance (Rouen, 2020).

Justified pay variation can attract and retain talents and the boost to human capital, in turn, will lead to better organizational performance (Trevor et al., 2012). Giving high performance relatively high salary will attract talents who are wary of getting under-compensated for their skills and knowledge (Lazear, 2000), and also enhance high performers’ satisfaction with pay (Trevor & Wazeter, 2006) and reduce their intention to leave. In addition, low-performers are likely to exit organizations that tie pay variation to performance (Trevor et al., 2012). Together, justified pay variation can improve organizational performance through favorable sorting effects.

Research on pay variation also noted that pay variation is more likely to improve performance if three conditions are met (Gupta et al., 2012). First, performance can be readily measured and attributed to individuals or units so that employees’ increased efforts can be expected to improve measured performance (Lazear, 2000; Trevor et al., 2012). Second, a strong link between performance and pay needs to be established. Pay variation unjustified by employees’ inputs such as education and performance either have null effects or can even reduce organizational performance (Rouen, 2020; Trevor et al., 2012). Third, employees need to highly value extrinsic motivation such as pay and promotion, the rewards for winning the tournament, and attaining goals (Connelly et al., 2014; Gupta et al., 2012). Appealing to extrinsic motivation for intrinsically motivated employees can be counterproductive (Weibel et al., 2010).

The review of literature in the private sector also suggests both opportunities and obstacles to improving public sector performance through increased pay variation. First, a high level of pay variation helps attract and maintain high-skilled employees and improve the quality of government employees (Borjas, 2002; Katz & Krueger, 1991; Schanzenbach, 2015). Flexible pay allows the government to give high-performing individuals competitive salaries, which are essential in a meritocracy. Furthermore, pay variation may increase the appeal of a long-term career in the government by increasing the prize of promotion within the organizational hierarchy (Ang, 2016). The expectation will excite job seekers and attract more high-skilled employees to join and remain in the public sector (Evans & Rauch, 1999).

Second, pay variation creates opportunities for employees to get larger rewards by meeting performance targets, therefore providing an incentive for achieving better management and performance. Pay variation is often accompanied by the implementation of performance-related pay (World Bank, 2014). So, the pay variation in combination with performance-related pay enhances both the instrumentality and valence of pay system, thereby driving public employees to spend more effort on improving performance.

In the public sector, at least three challenges exist for improving government performance through increased pay variation. First, diverse and ambiguous criteria for performance undermine the motivational effects of pay variation by weakening the links among effort, performance, and pay. Compared with private organizations, goals for public organizations are vague (Allison, 2004; Chun & Rainey, 2005). In addition, government tends to be under constant scrutiny from more diverse stakeholders who apply different criteria or weights for the same factors (Amirkhanyan et al., 2014). In addition, performance in the public sector is subject to more external influences and harder to be directly attributed to individual effort. The ambiguity in performance evaluation and limited control over measured outcomes to performance breeds distrust in performance appraisal systems especially after it is explicitly linked with pay (Pearce & Perry, 1983). The distrust in the performance appraisal system will raise questions of whether public employees’ increased efforts can attain goals (expectancy) and lead to pay increase (instrumentality), thereby undermining the motivational effect of pay dispersion.

Second, research in the private sector has shown that pay dispersion is associated with reduced collaboration (Conroy et al., 2014; Pfeffer & Langton, 1993). The reduced collaboration can be more detrimental to work in the public sector because public work tends to be more interdependent and requires more collaboration (Allison, 2004). Third, while mixed evidence exists on whether public employees highly value pay (Houston, 2000; Pearce & Perry, 1983; U.S Merit System Protection Board, 2008), the government is often constrained by laws and other environmental factors in its ability to offer sufficiently large pay increase (Perry et al., 2009). So, the pay increase associated with promotion and goal attainment may not be salient enough to be motivating.

Differentiating pay variation that can be justified by employees’ inputs including education, experience, and previous performance from unjustified variation is used to reconcile competing predictions in the private sector (Conroy et al., 2014; Rouen, 2020; Trevor et al., 2012). Justified pay variation has considered employees’ inputs and will be likely to be considered equal. Because of the aforementioned challenges in public sector, we expect justified pay variation to be as important, if not more important than in public sector, for moderating the effect of pay variation on performance. Unjustified pay variation in public sector is more likely to be caught by diverse stakeholders who constantly scrutinize government operation. The extensive external control in the public sector increases the possibility that exposure of unfair pay practice causes lengthy investigations and lawsuits. The unfair practices highlighted by the discussion and investigations of unfair pay practice will be especially demoralizing to public employees who tend to consider fairness part of their self-concept and an important factor shaping their commitment to the organization (Quratulain et al., 2019).

H1: Higher pay variation that can be justified by employees’ inputs is associated with increased government effectiveness.

The favorable sorting and motivating effects of pay variation need to be complemented by other favorable conditions to be effective. For instance, merit-based recruitment is necessary for pay variation to attract and retain talents. Pay variation according to positions and performance incentives talents to join and stay in the public sector if talents can believe that their skills and knowledge can give them a competitive advantage in accomplishing goals and gaining rewards (Dal Bo et al., 2013; World Bank, 2014). The positive sorting effect depends on merit-based recruitment in two ways. First, without merit-based recruitment, talents may not be selected to join public service in the first place. Second, talents need merit-based promotion/selection for higher positions to remain in the public sector because merit-based promotion convinces them of the possibility of getting more rewards by delivering good performance. However, it should be noted that the factors that attract people join an employer can be different from the factors that drive their attrition.

Corruption can also moderate the motivating effects of pay variation in two ways. Corruption is the abuse of power for private gain (Transparency International, 2022) and serves as an alternative for public employees to gain rewards from their job in the public sector. Hence, rampant corruption can weaken the perceived link between performance and rewards, the foundation for motivational effect in both tournament and expectancy theory (Connelly et al., 2014). In addition, increased reward from corruption reduces the relative valence of reward from keeping and excelling in one’s formal job in the public sector (Olken & Pande, 2012).

Rule of law can check patronage and abuse of power and also moderates the motivating effect of pay variation. Without effective checks on power, public employees tend to use power for patronage and personal gains instead of for performance improvement (Schuster et al., 2020). When patronage and corruption are prevalent in the public sector, increasing pay variation raises the reward for patronage in hiring and promotion, dampening any potentially beneficial effects of pay variation on bureaucrats’ quality and motivation. Furthermore, the combination of patronage and high pay variation can lead to pay inequity where well-connected public employees with mediocre performance and credentials earn a wage premium, deflating other employees’ commitment and decreasing their productivity (Breza et al., 2018).

In developing countries, the government usually lacks the resources to provide public services and enforce laws. As a result, informal rules usually govern not only business transactions but also government operations (Schick, 1998). The pervasive informality leads to widespread evasion of civil service rules, a norm of dishonesty, and corruption (Gächter & Schulz, 2016; Oliveros & Schuster, 2017). By contrast, formal rules govern both private and public sectors in developed countries because the rule of law facilitates market activities and is in interests of both private companies and government (Ang, 2016). With the established rule of law, public employees face stronger monitoring and sanction systems and, therefore, are incentivized to use power to advance public rather than private interests (Schuster et al., 2020).

Researchers indeed find that the institutional and governing conditions are generally most favorable in developed countries (Dahlström et al., 2012; Nistoskaya et al., 2021), a finding supported in our analysis of datasets from the World Bank and Quality of Government Institute (see Table 1). Because of the gaps in governing institutions, implementing administrative reforms that are shown to work in developed countries, such as granting managers greater autonomy, can backfire, and undermine government performance in developing countries (Hanushek et al., 2013; Schick, 1998; Schuster et al., 2020). Hence, we put forward the second hypothesis.

Table 1.

Descriptive Statistics of Key Governance Indicators by Income Groups.

Variable	N	M	SD	Minimum	Median	Maximum
	Low-income
Rule of law	550	−0.846	0.499	−2.009	−0.827	0.330
Corruption	549	−0.809	0.433	−1.722	−0.812	0.988
Meritocratic recruitment	342	−0.431	0.804	−2.090	−0.427	1.642
	Lower-middle income
Rule of law	718	−0.566	0.472	−1.780	−0.630	0.782
Corruption	718	−0.530	0.509	−1.440	−0.568	1.647
Meritocratic recruitment	495	−0.424	0.752	−2.090	−0.301	1.642
	Upper-middle income
Rule of law	563	−0.120	0.607	−2.322	−0.154	1.356
Corruption	556	−0.179	0.593	−1.471	−0.272	1.592
Meritocratic recruitment	474	−0.128	0.821	−1.502	−0.153	1.569
	High income
Rule of law	545	1.198	0.652	−1.916	1.287	2.100
Corruption	545	1.156	0.808	−1.400	1.272	2.470
Meritocratic recruitment	513	0.810	0.962	−1.113	0.851	2.443

Note. The higher the value, the better the outcome is.

H2: Higher pay variation is more likely to be associated with higher government performance in developed countries than in developing countries.

Methods

The hypotheses were tested with a large panel data set across countries and over multiple years. We next define the dependent and independent variables, data sources, and analytic strategy.

Data and Measurement

Government effectiveness

The dependent variable is the government effectiveness indicator from World Bank’s Worldwide Governance Indicators (WGI) from 2000 to 2018, the period when the data of pay variation is available. All the other variables cover the same period from 2000 to 2018 unless otherwise specified. The indicator captures the quality of formulating, implementing public policy and service delivery, public employees’ competence, independence from political pressure, and reliability of government policy. It has been used in many studies examining government effectiveness (Garcia-Sanchez et al., 2013; Lee & Whitford, 2009). The government effectiveness was standardized with a mean of zero and standard deviation of one for each period. The authors of WGI have also addressed various concerns of using the indicators for cross-country comparison over time (Kaufman et al., 2007). Government effectiveness was not collected for 2001. To fill the missing value in 2001, we used the average for a country between 2000 and 2002.

Total pay variation

Pay variation is measured by the ratio of public employees’ 90th percentile weekly wage to 10th percentile weekly wage for public employees from Worldwide Bureaucracy Indicators (WWBI) between 2000 and 2018 (Bureaucracy Lab, 2020). The wage does not include bonuses or other payments. This indicator measures overall pay variation, which is commonly used in private sector studies (Gupta et al., 2012) and mitigates difficulties in finding the right reference groups in constructing indicators for horizontal or vertical pay variability (Rouen, 2020).

This definition remains agnostic about the source of the variation or whether it is justified. A large ratio can be the result of large vertical pay and small horizontal pay, or the result of small vertical pay or large horizontal pay (Gupta et al., 2012). Our understanding is that vertical pay variation usually refers to the pay variation across ranks while horizontal pay variation refers to the pay variation among employees in similar positions and/or performing similar tasks. The ratio of 90th percentile weekly average wage to 10th percentile weekly average wage in the public sector is the result of both vertical and horizontal pay variation. So, it is a good measure of total pay variation.

WWBI is based on microdata from high-quality household surveys frequently supported and managed by international organizations such as World Bank (Baig et al., 2021). Based on representative household/labor surveys, WWBI contains measures of demographic characteristics, compensations, and wages bills in the public sector of 132 countries from 2000 to 2018. The median and mean numbers of country-year observations in our data is 4 and 6, respectively.

Several measures are taken to ensure the quality of data in WWBI (Bureaucracy Lab, 2020). First, only household/labor surveys with no more than 40% of missing values in key variables of employment demographic characteristics and compensation are utilized in WWBI. The 40% cut-off point was chosen based on the experience of the team creating WWBI (Bureaucracy Lab, 2020) and previous studies (Dong & Peng, 2013). Second, surveys are also excluded if they contain either fewer than 5% of paid employees or fewer than 200 total observations of public paid employees (whichever is smaller). Third, only surveys containing representative samples in both urban and rural areas are included. To make measures compatible across surveys, WWBI broadly defined the public sector, including not only the general government but also other entities, like corporations, controlled by the state. The WWBI’s 63,282 country-year data points across 92 indicators were built from over 53 million unique worker-level observations from 846 surveys. Such a large number of representative observations increases the reliability of measures of pay variation.

Vertical pay variation

To capture pay variation across ranks, we use the ratio of senior public manager’s weekly average salary to the clerk’s weekly average salary in WWBI. Managers refer to those who direct, coordinate, and evaluate operations of public organizations. The classification includes senior officials, elected legislators, and managers of any kind. WWBI defines clerks as any employees who are primarily responsible for recording, organizing, and retrieving information and performing other administrative duties. The category contains general and keyboard clerks, customer services, and other clerical support workers. Using senior managers and clerks as the reference groups are also consistent with previous research on vertical pay variation (Cowherd & Levine, 1992).

Justified and unjustified pay variation

To distinguish between pay variation justified by employees’ inputs from the one that cannot be justified, we draw on previous models predicting explained pay variation at the individual level (Trevor et al., 2012; Trevor & Wazeter, 2006) and selected corresponding variables at the national level. We chose the share of public employees with tertiary education, the average age of public employees with salaries, performance-based management of pay and promotion, merit-based recruitment, and percentage differences in public sector wages compared to private sector wages in local currency, their squared terms, and yearly dummies to predict both the total pay variation and vertical pay variation.

Both educational attainment and experience have been shown to affect individuals’ salary level (Trevor et al., 2012; Trevor & Wazeter, 2006). At the aggregate level, a smaller share of public employees with tertiary education may increase the wage premium received by public employees with tertiary education degrees and increase the pay dispersion. Similarly, a relatively young workforce may increase the return on experience and increase the pay for experienced employees and pay variation. Performance-based management of pay and promotion and merit-based recruitment establish performance as a valid reason for candidate selection, promotion, and increased pay and, thereby, possibly increasing the pay variation (World Bank, 2014). The average wage level can also predict the pay variation because high average pay and high pay variation are often used together to attract and retain talents (Gerhart & Newman, 2019). So, we included the pay differential between the public and private sectors as a measure of the average pay level in the public sector.

Previous studies often used Ordinary Least Squares (OLS) with clustered standard error to generate the predicted wage as the measure of explained/justified wage (Trevor et al., 2012). Lasso regression is sometimes recommended to ensure coefficient stability in new samples (James et al., 2013). To select the best model, we randomly assigned 80% of countries to the training set and 20% of countries to the test set. Then we ran OLS, lasso, and ridge regression and record the mean sum of errors (MSE) for each model in the testing set. OLS was shown to have the lowest MSE.

As we have panel data and need to account for country-specific factors influencing pay variation, we decided to use random effect models over OLS and fixed effects. We prefer random effect over fixed effect because our relatively short and unbalanced panel data makes it appropriate to pool estimates from both OLS and fixed effects in calculating the coefficients (Wooldridge, 2010). Researchers (Trevor et al., 2012) also noted that we should not include performance management and average pay level as controls in the main model for government effectiveness because doing so will remove the influence of these factors on pay variation and leads to coefficients for only unjustified pay variation.

Controls

We controlled for several previous explanations linking certain features of bureaucracy with government effectiveness.

Our model took into account the level of socioeconomic development in a country because the development can influence both the demand and supply for government effectiveness (La Porta et al., 1999). A country with higher socioeconomic development is more likely to have the resources whether it is human capital or financial capital to enhance government capacity. In addition, a higher level of economic development requires government to facilitate economic activities. Moreover, the educated public tend to have higher expectations for government performance and are ready to hold government accountable to higher standards. To operationalize relevant factors, we included a log transformation of country GDP per capita at the present value of dollar and literacy rate of people aged 15 or above from World Development Indicators. As the proportion of missing value for literacy rate is large (70%) and the between-country variation in literacy rate was seven times the within-country variation in the studied period (2000–2018), we used the country-specific average literacy rate in the studied period as the control in the models.

We controlled for freedom of political participation because the public enforces the accountability of government to performance through political participation. To account for the possible nonlinear relationship between political participation and government performance (Treisman, 2007), a quadratic term of political participation was also included. Although previous studies also included press freedom in the model, initial analysis showed that the Freedom House civic liberty measure, which includes press freedom (Freedom House, 2018), is highly correlated with the index of political participation (r = .938). Given that press freedom eventually needs to translate to political and administrative changes through political participation (Yanguas & Bukenya, 2016), we chose political participation as the control variable.

Gender diversity is also held to improve government effectiveness because female employees can bring new perspectives and creative solutions in public organizations traditionally dominated by men (Garcia-Sanchez et al., 2013). For this reason, we also included the share of females in total public sector employment from WWBI.

Moreover, meritocratic recruitment is also important for government effectiveness because it creates a different chain of accountability for bureaucrats and, thereby, separates bureaucrats’ interest from politicians (Dahlström et al., 2012). To account for meritocratic recruitment, we generated average scores from answers to specific questions in two waves of Quality of Government Expert Survey (2008–2012 and 2014p; Dahlström et al., 2015). We took the average because the time interval between the two surveys is small and correlations between the same measures in the two periods were high (above .8).

Table 2.

Operationalization of Variables.

Variables	Measures	Sources
Government performance	Government effectiveness index	Worldwide Governance Indicators
Total pay variation	The ratios of the 90th percentile wage to the 10th percentile wage in the public sector	Worldwide Bureaucracy Indicators
Vertical pay variation	The ratio of senior public manager’s weekly average salary to the clerk’s weekly average salary	Worldwide Bureaucracy Indicators
Public sector wage premium	The percentage differences in public sector wages compared to private sector wages (in local currency units) controlling for education, age, gender, and location and the difference between and the country average in a specific period and the country overall average	Worldwide Bureaucracy Indicators
Economic development	Log transformation of GDP per capital in present value.	World Development Indicators
Educational attainment	Share of population aged over 25 with at least a bachelor’s degree	World Development Indicators
Political participation	A measure of political participation, representation of minority interests, and actual control exerted by the party winning the election	Freedom House
Female representation in bureaucracy	Female share of total public employment	Worldwide Bureaucracy Indicators
Meritocratic recruitment	The average response to the degree to which candidates’ qualifications and connections play in hiring public employees	Quality of Government’s Expert Surveys.
Performance-based personnel management	The average response to the degree to which salary is linked to performance and misconducts being punished.	Quality of Government’s expert surveys.
Public service motivation	The average expert’s agreement with the statement that public sector employees strive to help citizens	Quality of Government’s expert surveys

Previous studies also used professionalism which includes meritocratic recruitment as a control, but studies often show that it was meritocratic recruitment that drives the positive correlation between professionalism and government performance (Dahlström et al., 2012; Rauch & Evans, 2000). For this reason, we only controlled for meritocratic recruitment. Table 1 contains operationalizations for the variables. Appendix A1 contains summary statistics for the variables. The between-country variation in variables for governance and pay was much larger than the within-country variation. Table 2 specifies the operationalization of our key variables.

Estimation Strategies

We used Stata 16 to conduct empirical analysis. As this is a panel dataset, we need to control for the country-specific variation. We prefer the random effect model over the fixed effect model to control for country-specific variation for two reasons. First, because of our unbalanced and short panel dataset, it makes sense for us to use random effects to pool estimates from Ordinary least squares and fixed-effects in calculating coefficients (Wooldridge, 2010). Second, we are also interested in learning the association of merit-based recruitment and public service motivation with government effectiveness to build conversations with previous literature (Dahlström et al., 2012; Ritz et al., 2016). Due to data constraints, we only have country-specific average for these variables. So, random-effects model is a more appropriate model choice.

A shortcoming of a random-effects model is the stringent assumption that the country-specific effect does not correlate with any of the independent variables. To relax the assumption, we ran a correlated-random-effects (CRE) model as our primary model. CRE model allows the correlation between country-specific effect and observed variables by including the country-specific average of independent variables in the model (Wooldridge, 2010). If the correlation between country-specific error and independent variables is driven by country-specific average values of independent variables, then CRE mitigates the concern of violating the assumption for the random-effects model. In CRE, the estimate for pay variation will be similar to the one in fixed effect. The interpretation is also similar in the sense that the coefficient represents the effect of within-country variation after controlling for a country-specific average level of pay variation.

A multicollinearity check indicated that log GDP per capita correlated with income group dummies and had a high variance inflation factor (VIF = 13.91). To reduce the correlation, we replaced the log GDP per capita with the difference between log GDP per capita and the log average GDP per capita for the income group in the specific year. After the transformation, the VIF for the new variable dropped to 5.51. In the Appendix, we also included results from fixed-effects model and CREs controlling for an imperfect measure of public service motivation to demonstrate the robustness of our results.

Results

The results suggest that there is no universal relationship between pay variation and government effectiveness, and H1 is not supported. We first presented results with pay variation and time dummies and then the results with other controls in Table 3. We consistently found that neither total pay variation nor vertical pay variation was significantly associated with government effectiveness at 5% significance level. These insignificant findings are perhaps not surprising because positive significant associations of justified pay variation may cancel out the negative associations of unjustified pay variation (Trevor et al., 2012).

Table 3.

Random Correlated Model Estimates for Total Pay Variation.

Variables	Total	Justified	Unjustified	Total	Justified	Unjustified
Total pay variation	−0.0327 (0.0213)			−0.0368* (0.0208)
Average pay variation	−0.157* (0.0855)			0.155** (0.0672)
Justified pay variation		−0.00526 (0.0255)			−0.00498 (0.0346)
Average justified pay variation		−0.324*** (0.0988)			0.00951 (0.0820)
Unjustified pay variation			−0.0371 (0.0269)			−0.0380 (0.0233)
Average unjustified pay variation			−0.0309 (0.0461)			0.0518* (0.0300)
Centered GDP per capita				0.0540 (0.0331)	0.0595 (0.0362)	0.0579 (0.0357)
Average log GDP per capita				0.505*** (0.115)	0.290** (0.121)	0.365*** (0.137)
Political participation				−0.0411 (0.0477)	−0.0397 (0.0505)	−0.0452 (0.0486)
Participation²				0.0409 (0.0371)	0.0433 (0.0393)	0.0395 (0.0394)
Average participation				−0.178** (0.0824)	−0.208** (0.0900)	−0.181* (0.0931)
Gov_female%				−0.0229 (0.0557)	−0.0284 (0.0597)	−0.0249 (0.0595)
Average female%				0.0572 (0.0849)	0.141 (0.109)	0.152 (0.107)
Meritocratic recruitment				0.200*** (0.0644)	0.254*** (0.0645)	0.236*** (0.0680)
Average literacy rate				−0.0823 (0.0922)	−0.00343 (0.0923)	−0.0467 (0.0948)
Constant	−0.0194 (0.0889)	1.958*** (0.559)	0.152 (0.106)	0.0803 (0.119)	0.000349 (0.464)	0.0682 (0.125)
Observations	678	574	574	486	462	462
Number of countries	100	73	73	68	61	61

Note. All the models include the year dummies for the period from 2001 to 2018 and the standard error was clustered at country level.

p = .1. **p = .05. ***p = .01 (two-sided tests).

Contradicting H1, we found that justified pay variation was not significantly associated with government effectiveness at 5% significance level. In power analysis (details can be requested upon requests), we found that the sample size is sufficient to detect an increase in expected R square change as small as 0.004 for a model with power at 0.8. The insignificant finding remained robust in fixed-effects- models (see Tables A2–A4 in Appendix A) and CREs controlling for public service motivation (Tables A5 and A6 in Appendix A). Although justified pay variation was often found to be positively associated with organizational performance, our results indicate that this relationship did not hold in the public sector. As we will expand in the discussion section, we attribute the insignificant relationship to the weak links among effort, performance, and pay in the public sector and the correspondingly weak motivational effect of pay variation. The average pay variation had mixed relationships with government effectiveness. As we are primarily interested in whether changing pay variation for a country over time can improve its government effectiveness, the between-country comparison is not of primary interest (Table 4).

Table 4.

Random Correlated Model Estimates for Vertical Pay Variation.

Variables	Total	Justified	Unjustified	Total	Justified	Unjustified
Vertical pay variation	−0.00520 (0.0169)			0.00450 (0.0221)
Average vertical pay variation	0.233** (0.0926)			−0.0420 (0.0580)
Justified vertical pay variation		0.0105 (0.0231)			0.0862* (0.0506)
Average justified vertical pay variation		1.924 (1.291)			−1.590** (0.695)
Unjustified vertical pay variation			−0.0143 (0.0188)			−0.00447 (0.0231)
Average unjustified vertical pay variation			0.250 (0.206)			−0.00885 (0.112)
Centered GDP per capita				0.0731* (0.0386)	0.0859** (0.0404)	0.0741* (0.0414)
Average Log GDP per capita				0.308** (0.126)	0.203* (0.118)	0.261** (0.129)
Political participation				−0.0284 (0.0525)	−0.0199 (0.0588)	−0.0389 (0.0569)
Participation²				0.0498 (0.0367)	0.0534 (0.0369)	0.0405 (0.0402)
Average participation				−0.225** (0.0935)	−0.304*** (0.108)	−0.264** (0.104)
Gov_female%				−0.0229 (0.0601)	0.0231 (0.0620)	−0.00652 (0.0636)
Average female%				0.142 (0.100)	0.193* (0.109)	0.152 (0.109)
Meritocratic recruitment				0.211*** (0.0633)	0.166** (0.0725)	0.203*** (0.0694)
Average literacy rate				−0.00572 (0.120)	−0.00530 (0.112)	−0.0257 (0.123)
Constant	0.0578 (0.0922)	−3.542 (2.455)	0.189* (0.107)	0.0432 (0.134)	3.008** (1.337)	0.0243 (0.146)
Observations	596	511	511	421	401	401
Number of cc	90	70	70	61	56	56

Note. All the models include the year dummies for the period from 2001 to 2018 and the standard error was clustered at country level.

p = .1. **p = .05. ***p = .01 (two-sided tests).

For H2, we found mixed evidence (see Table 5). In support of H2, we found a consistently negative correlation between pay variation and government effectiveness in low-income countries. A one standard deviation increase in the pay variation was associated with a 0.066 standard deviation decrease in government effectiveness. In addition, the negative marginal effect of pay variation drops significantly in lower-middle-income countries. On the other hand, the average marginal effect did not keep increasing and turn positive for upper-income or high-income countries (see Figure 1). For countries in lower-middle-income groups, the interaction between pay variation and income group was positive, but it was not sufficiently large to offset the negative coefficient of total pay variation, leading to an insignificant marginal effect. The same goes for countries in upper-middle-income countries. Although the average marginal effect declined for high-income countries, the effect was still insignificant due to high estimation uncertainty (high standard error).

Table 5.

Correlated Random Effect Model Estimates With Interaction Terms Between Income Group and Pay Variation.

Variables	Total	Total justified	Total unjustified	Vertical	Vertical justified	Vertical unjustified
Total pay variation	−0.0612** (0.0305)
Average pay variation	0.142** (0.0641)
Lower-middle income	−0.162*** (0.0428)	−0.199*** (0.0623)	−0.193*** (0.0588)	−0.184*** (0.0534)	−0.123 (0.0959)	−0.222*** (0.0610)
Upper-middle income	−0.0894 (0.0755)	−0.128 (0.0826)	−0.135 (0.0883)	−0.116 (0.0994)	−0.0634 (0.130)	−0.167 (0.106)
High income	−0.0683 (0.0890)	−0.121 (0.103)	−0.106 (0.102)	−0.0213 (0.118)	−0.0458 (0.147)	−0.0835 (0.126)
Lower-middleincome × pay variation	0.0627* (0.0336)
Upper-middle income × pay variation	0.0274 (0.0483)
High income × pay variation	−0.0201 (0.0724)
Total justified pay variation		0.0342 (0.0517)
Average justified pay variation		−0.000189 (0.0796)
Lower-middle income × justified pay variation		0.00903 (0.0557)
Upper-middle income × justified pay variation		−0.0492 (0.0564)
High income × justified pay variation		−0.0904 (0.0652)
Total unjustified pay variation			−0.0838** (0.0330)
Average unjustified pay variation			0.0543* (0.0281)
Lower-middle income × unjustified pay variation			0.0728** (0.0337)
Upper-middle income × unjustified pay variation			0.0653 (0.0493)
High income × unjustified pay variation			0.0261 (0.0607)
Vertical pay variation				0.0479 (0.0395)
Average vertical pay variation				−0.0471 (0.0579)
Lower-middle income × vertical pay variation				−0.0115 (0.0496)
Upper-middle income × vertical pay variation				−0.0238 (0.0453)
High income × vertical pay variation				−0.101** (0.0489)
Justified vertical pay variation					−0.00583 (0.0688)
Average justified vertical pay variation					−1.312* (0.708)
Lower-middle income × vertical justified pay variation					0.0965 (0.0601)
Upper-middle income × vertical justified pay variation					0.0928 (0.0583)
High income × vertical justified pay variation					0.104 (0.0695)
Unjustified vertical pay variation						0.0169 (0.0377)
Average unjustified vertical pay variation						−0.0183 (0.114)
Lower-middle income × vertical unjustified pay variation						0.0202 (0.0486)
Upper-middle income × vertical unjustified pay variation						0.00432 (0.0462)
High income × vertical unjustified pay variation						−0.0808* (0.0470)
Centered GDP per capita	0.0431 (0.0369)	0.0581 (0.0404)	0.0419 (0.0389)	0.0423 (0.0403)	0.0832* (0.0464)	0.0506 (0.0426)
Average log GDP per capita	0.483*** (0.126)	0.308** (0.129)	0.352** (0.146)	0.280** (0.130)	0.240* (0.135)	0.251* (0.133)
Political participation	−0.0266 (0.0465)	−0.0178 (0.0532)	−0.0392 (0.0457)	−0.0295 (0.0537)	−0.0279 (0.0586)	−0.0450 (0.0578)
Participation²	0.0399 (0.0335)	0.0411 (0.0352)	0.0419 (0.0319)	0.0518 (0.0327)	0.0389 (0.0338)	0.0433 (0.0339)
Average participation	−0.195** (0.0824)	−0.235*** (0.0902)	−0.183** (0.0919)	−0.211** (0.0936)	−0.270** (0.109)	−0.242** (0.106)
Gov_female%	−0.0295 (0.0536)	−0.0360 (0.0509)	−0.0231 (0.0598)	0.000803 (0.0637)	0.0278 (0.0584)	0.0181 (0.0683)
Average female%	0.0527 (0.0843)	0.135 (0.107)	0.137 (0.111)	0.102 (0.105)	0.161 (0.116)	0.117 (0.112)
Meritocratic recruitment	0.180*** (0.0666)	0.228*** (0.0637)	0.214*** (0.0683)	0.187*** (0.0658)	0.170** (0.0733)	0.167** (0.0724)
Average literacy rate	−0.0416 (0.0932)	0.0414 (0.0892)	0.0146 (0.0892)	0.0656 (0.120)	0.0308 (0.111)	0.0494 (0.124)
Constant	0.130 (0.135)	0.119 (0.466)	0.159 (0.140)	0.0871 (0.135)	2.543* (1.373)	0.115 (0.148)
Observations	486	462	462	421	401	401
Number of countries	68	61	61	61	56	56

Note. All the models include the year dummies for the period from 2001 to 2018 and the standard error was clustered at country level.

p = .1. **p = .05. ***p = .01 (two-sided tests).

Figure 1.

The average marginal effect of total pay variation across income groups.

Moreover, the association with government effectiveness diverges between total justified and total unjustified pay variation. Justified pay variation was not significantly associated with any government effectiveness at 5% significance level (see Figure 2). By contrast, unjustified pay variation was negatively and significantly associated with government effectiveness (p = .011; see Table 5 and Figure 3). One standard deviation increase in the total unjustified pay variation was associated with a 0.090 standard deviation decrease in government effectiveness. The result suggests that the negative association of total pay variation can be mainly attributed to the negative association of unjustified pay variation. The marginal effects of unjustified pay variation were also insignificant in other income group countries. Our findings also suggest that vertical pay variation was not significantly associated with government effectiveness. The results remained robust in fixed effects models (see Tables A2–A4 in Appendix A) and CRE controlling for public service motivation (see Tables A5 and A6 in Appendix A).

Figure 2.

The average marginal effect of justified pay variation across income groups.

Figure 3.

The average marginal effect of unjustified pay variation across income groups.

Discussion

In this article, we theoretically develop and empirically test the linkage between pay variation of public employees and government performance by using a recently developed worldwide panel dataset. While pay variation is of particular importance to government talent acquisition and performance improvement, few studies systematically test this proposition. To the best of our knowledge, we report one of the first quantitative findings exploring the impacts of pay variation on government performance and contribute to ongoing research about how the differences between the private and public sectors and national contexts shape managerial practices in the public sector and the danger of applying theories based on research on private firms to public sectors.

First, we found that neither overall pay variation not vertical pay variation, on average, significantly affect government performance. One explanation for this result is that positive motivating effects of justified pay variation are offset by negative effects of unjustified pay variation, particularly those associated with perceived inequities in pay variation. Although justified pay variation increases the prize of promotion and meeting performance targets, the unjustified may also make public employees feel deprived in comparison with others, which eventually leads to behaviors undermining performance. This explanation is also consistent with private sector studies aiming to reconcile the mixed effects of overall pay variation on performance (Conroy et al., 2014; Gupta et al., 2012; Rouen, 2020).

The results for H1 also suggest that differences between the private and public sectors may shape the relationship between pay variation and organizational performance in the public sector. Recent studies on the topic in the private sector point to the importance of differentiating justified and unjustified pay variation in reconciling competing effects of pay variation on organizational performance (Conroy et al., 2014; Gupta et al., 2012). Research in the private sector shows that increasing justified pay variation can improve organizational performance (Rouen, 2020; Trevor et al., 2012). However, the results for H1 show that this relationship does not hold in the public sector. The insignificant relationship may reflect various differences between the public and private sectors. Performance criteria in the public sector are more diverse and ambiguous, and performance in public organizations is subject to external political influences beyond public employees’ control (Allison, 2004; Chun & Rainey, 2005). The ambiguity in performance evaluation and employees’ limited control over measured outcomes weaken the link between increased effort and performance (Pearce & Perry, 1983) Moreover, the political and legal constraints undermine the connection between performance and pay (Heinrich, 2007). Furthermore, the amount of available financial resources in public sector is limited (Perry et al., 2009), so the prize of getting a promotion or meeting a performance target may not be attractive enough. Meanwhile, public employees are found to hold motivations and values different from their counterparts in private firms. (Rainey & Bozeman, 2000). That might relate to their stronger preferences for equity in pay variation and may explain why pay variation does not work. Hence, characteristics of the public sector weaken the expectancy, instrumentality, and valence of pay variation, thereby reducing its motivational effect to insignificance.

We found partial support for H2. Although total pay variation was not significantly associated with government effectiveness in high-income countries, it was consistently and negatively associated with lower government performance in low-income countries. Differences in institutional environments between developed and developing countries (Schick, 1998; Schuster et al., 2020) can account for the contrast. As is analyzed in the hypothesis development section, the potential of pay variation in attracting talents and motivating employees depends on favorable institutional conditions such as rule of law and merit-based recruitment. Without merit-based recruitment, talents cannot join and remain in the public sector despite large pay variations. Without rule of law, public employees may be incentivized to abuse power for personal gains rather than work hard to improve government performance.

These conditions are more likely to be present in developed countries for at least two reasons. First, government is more likely to have more resources to build capacity and enforce rules in developed countries (Schick, 1998). Second, the interest of private companies and government are more aligned with strengthening rule of law in developed countries because the rule of laws promotes the development of market economy by providing stability and predictability (Ang, 2016). The institutional environments are indeed more favorable in developed countries versus developing countries (Dahlström et al., 2012; Nistoskaya et al., 2021).

In developing countries, a weak rule of law and connection-based recruitment may allow unprincipled public employees to engage in corruption and patronage (Schuster et al., 2020). A competitive salary may be offered to less qualified employees. So, increasing total pay variation may be more likely to increase the unjustified pay variation and breed a sense of pay inequity. Perceived pay inequity can demoralize employees and lead to a decline in their productivity (Breza et al., 2018). The significantly negative association of unjustified total pay variation with government effectiveness supports this explanation.

The results suggest that administrative reforms like pay variation should take national and sector-specific contexts into account, otherwise the reforms may not succeed and could even undermine government effectiveness. Implementing advanced reforms like increasing pay variation will only yield benefits if favorable governing conditions are in places (Schick, 1998). In the context of patronage hiring practices and weak rule of law, increasing pay variation is less likely to attain the goal of improving government performance. Some countries might adopt pay variation and other administrative reform agendas due to pressure for gaining legitimacy among donors and international organizations, but implementing reforms can be counterproductive due to misalignments between reforms and contexts. It is thus crucial to self-diagnose the fit and relevance of pay variation before exploring its application in the public sector. It may also be helpful to pilot different versions of pay variation in specific sectors and regions before large-scale policy implementation. This method enables policy adjustment according to local conditions and reduces the risk of policy failure (Ang, 2016).

The null effect or detrimental effects of pay variations on government effectiveness is also consistent with previous studies on the effectiveness of using pay as levers to improve government performance (Perry et al., 2009). Given the general ineffectiveness of pay as a lever to improve performance in public sector, a natural question arises of what else can be done. Building merit-based recruitment and promotion can be an area of continuous reforms in developing countries. Consistent with previous research (Dahlström et al., 2012), we also found a positive association of merit-based recruitment with government effectiveness. For developed countries, the impact of public service motivation (PSM) on organizational performance can be a promising area for future studies. PSM has been associated with outcomes such as organizational citizenship behavior (Harari et al., 2017), which may enhance organizational performance. We also see a consistently positive but insignificant relationship of PSM with government effectiveness in this study. We noted that our measure has limited concept validity and the measurement error may attenuate the relationship between PSM and government effectiveness. Future studies can explore whether public service motivation assessed through more robust measures is indeed associated with better government performance around the world.

While our results provide new evidence about the influence of pay variation, this study is limited in ways that call for future research to deepen our understanding of the effects of pay variation in public sector human resource management. First, the mechanisms by which pay variation affects government performance could be explored in future research. This would help to open the black box through which pay variation influences government performance. A key to opening the black box would be detailed measures of pay variation. The measure used in the present study uses overall pay variation and vertical pay variation. It has not explored horizontal pay variation affects organizational performance (Gupta et al., 2012). Future studies should measure all dimensions of pay variation and examine which dimensions influence government effectiveness and the processes and mechanisms by which they work. In addition, research also show that justified pay variation can also lead to trade-off between short-term and long-term performance in private sector (Connelly et al., 2013), it is also worth examining this phenomenon in public sector.

Second, we call for future studies to generate experimental evidence for causal relationships, which addresses the limitations of archival data used in this study. We only corroborate the correlation between pay variation and organization performance, and future research should causally estimate such a link through experiment designs, particularly field experiments. Scholars could partner with central governments or international organizations like the United Nations and World Bank to conduct field experiments when they consider pay reforms to redesign HR systems.

Moreover, it is important to delve deeply into subnational, and local levels to explore how pay variation and its moderating conditions work. Our study assumes that public pay systems in national and subnational governments are unified. The assumption is more likely to hold in small and homogeneous countries than in large and diverse countries. Similarly, significant within-country variation may exist in organizational practices, cultural norms, and rule of law. The cross-country analysis may overlook within-country variations that can further enrich our understanding about the contingency-based relationship between pay variation and government effectiveness.

Future studies may also benefit from getting access to individual-level data. To distinguish between justified and unjustified pay variation, we used the aggregated public employees’ characteristics and overall managerial practices to predict pay variation. Aggregation inevitably adds more noise to the prediction. Getting access to individual-level data will improve the precision of estimating which part of pay can be justified by employees’ inputs. Despite this limitation, this study has tapped into the best available dataset. Given the importance of the topic, this study marks only a starting point for examining this important relationship.

Conclusion

Overall, the findings echo important themes and advance collective understanding in public administration. The nuanced finding of the general null effect between pay variation and government effectiveness and significant effects in low-income countries echoes the call for advancing contingency-based theory building in public administration (Perry, 2016). We argue that the relationship between pay variation and government effectiveness is moderated by country development level because socioeconomic development provides both the resources and incentives to build and maintain these favorable institutional conditions; these conditions are necessary for pay variation to realize its potential in improving government effectiveness. Our results partially support this argument and illuminate the importance of recognizing national differences in advancing public management research. We also note that the differences between the public and private sectors may prevent justified pay variation from having significantly positive effects on government effectiveness. Hence, our study illuminates the importance of recognizing national contexts and highlights the importance of taking account of sectoral nuances in civil service pay design and advancing public management research.

Supplemental Material

sj-docx-1-rop-10.1177_0734371X221141988 – Supplemental material for Does Employee Pay Variation Increase Government Performance? Evidence From a Cross-National Analysis

Supplemental material, sj-docx-1-rop-10.1177_0734371X221141988 for Does Employee Pay Variation Increase Government Performance? Evidence From a Cross-National Analysis by Xu Han, Liang Ma and James L. Perry in Review of Public Personnel Administration

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the National Natural Sciences Foundation of China (Grant/Award Number: 72274203) for financial support.

ORCID iDs

Xu Han

Liang Ma

James L. Perry

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Xu Han is Research Scientist II at Amazon. He received his Ph.D. from the School of Public Policy, University of Maryland. He has contributed to the World Bank Worldwide Bureaucracy Indicators (WWBI). His research interests include performance management and human resource management.

Liang Ma is Professor at the School of Public Administration and Policy, Renmin University of China, China. He is Co-editor of International Journal of Public Administration. His research interests include performance management, innovation, and digital governance.

James L. Perry is Distinguished Professor Emeritus and Chancellor’s Professor of Public and Environmental Affairs Emeritus at the Paul H. O’Neill School of Public and Environmental Affairs, Indiana University, Bloomington. He is Co-editor of Asia Pacific Journal of Public Administration. His recent books include Public Service and Good Governance for the Twenty-first Century (Univ. of Pennsylvania Press, 2020) and Managing Organizations to Sustain Passion for Public Service (Cambridge University Press, 2021).

References

Allison

G. T.

(2004). Public and private management: Are they fundamentally alike in all unimportant respects? In Shafritz

J. M.

Hyde

A. C.

Parkes

S. J.

(Eds.), Classics of public administration (pp. 396–413). Wadsworth/Thomson Learning. https://doi.org/10.2307/2212271

Adams

J. S.

(1965). Inequity In Social Exchange. In Berkowitz

(Ed.), Advances in Experimental Social Psychology (Vol. 2, pp. 267–299). Academic Press. https://doi.org/https://doi.org/10.1016/S0065-2601(08)60108-2

Akerlof

G. A.

Yellen

J. L.

(1990). The Fair Wage-Effort Hypothesis and Unemployment*. The Quarterly Journal of Economics, 105(2), 255–283. https://doi.org/10.2307/2937787

Amirkhanyan

A. A.

Kim

H. J.

Lambright

K. T.

(2014). The performance puzzle: Understanding the factors influencing alternative dimensions and views of performance. Journal of Public Administration Research and Theory, 24(1), 1–34. https://doi.org/10.1093/jopart/mut021

Ang

Y. Y.

(2016). How China escaped the poverty trap. Cornell University Press.

Ang

Y. Y.

(2020). China’s gilded age: The paradox of economic boom and vast corruption. University of Cambridge.

Baig

F. A.

Han

Hasnain

Rogger

(2021). Introducing the worldwide bureaucracy indicators: A new global dataset on public sector employment and compensation. Public Administration Review, 81(3), 564–571. https://doi.org/10.1111/puar.13355

Binderkrantz

A. S.

Christensen

J. G.

(2012). Agency performance and executive pay in government: An empirical test. Journal of Public Administration Research and Theory, 22(1), 31–54. https://doi.org/10.1093/JOPART/MUR039

Borjas

G. J.

(2002). The wage structure and the sorting of workers into the public sector. NBER Working Paper Series, 9313, 1–45.

10.

Breza

Kaur

Shamdasani

(2018). The morale effects of pay inequality. Quarterly Journal of Economics, 133(2), 611–663. https://doi.org/10.1093/qje/qjx041

11.

Bureaucracy Lab. (2020). Worldwide bureaucracy indicators codebook and explanatory note. Author.

12.

Chun

Y. H.

Rainey

H. G.

(2005). Goal ambiguity in U.S. federal agencies. Journal of Public Administration Research and Theory, 15(1), 1–30. https://doi.org/10.1093/jopart/mui001

13.

Connelly

B. L.

Haynes

K. T.

Tihanyi

Gamache

D. L.

Devers

C. E.

(2013). Minding the gap: Antecedents and consequences of top management-to-worker pay dispersion. Journal of Management, 42(4), 862–885. https://doi.org/10.1177/0149206313503015

14.

Connelly

B. L.

Tihanyi

Crook

T. R.

Gangloff

K. A.

(2014). Tournament theory: Thirty years of contests and competitions. Journal of Management, 40(1), 16–47. https://doi.org/10.1177/0149206313498902

15.

Conroy

S. A.

Gupta

Shaw

J. D.

Park

T. Y.

(2014). A multilevel approach to the effects of pay variation. Research in Personnel and Human Resources Management, 32, 1–64. https://doi.org/10.1108/S0742-730120140000032001

16.

Cowherd

D. M.

Levine

(1992). Product quality and pay equity between lower-level employees and top management: An investigation of distributive justice theory. Administrative Science Quarterly, 37(2), 302–320.

17.

Dahlström

Lapuente

Teorell

(2012). The merit of meritocratization: Politics, bureaucracy, and the institutional deterrents of corruption. Political Research Quarterly, 65(3), 656–668. https://doi.org/10.1177/1065912911408109

18.

Dahlström

Teorell

Dahlberg

Hartmann

Lindberg

Nistotskaya

(2015). The QoG expert survey dataset II. The Quality of Government Institute.

19.

Dal Bo

Finan

Rossi

M. A.

(2013). Strengthening state capabilities: The role of financial incentives in the call to public service. Quarterly Journal of Economics, 128(3), 1169–1218. https://doi.org/10.1093/qje/qjt008

20.

Dong

Peng

C. Y. J.

(2013). Principled missing data methods for researchers. SpringerPlus, 2(1), 1–17. https://doi.org/10.1186/2193-1801-2-222/TABLES/3

21.

Evans

Rauch

J. E.

(1999). Bureaucracy and growth: A cross-national analysis of the effects of “Weberian” state structures on economic growth. American Sociological Review, 64(5), 748–765.

22.

Gächter

Schulz

J. F.

(2016). Intrinsic honesty and the prevalence of rule violations across societies. Nature, 531(7595), 496–499. https://doi.org/10.1038/nature17160

23.

Garcia-Sanchez

I. M.

Cuadrado-Ballesteros

Frias-Aceituno

(2013). Determinants of Government Effectiveness. International Journal of Public Administration, 36(8), 567–577. https://doi.org/10.1080/01900692.2013.772630

24.

Gerhart

B. A.

Newman

J. M.

(2019). Compensation (13th ed.). McGraw-Hill Education.

25.

Gupta

Conroy

S. A.

Delery

J. E.

(2012). The many faces of pay variation. Human Resource Management Review, 22(2), 100–115. https://doi.org/10.1016/j.hrmr.2011.12.001

26.

Hanushek

E. A.

Link

Woessmann

(2013). Does school autonomy make sense everywhere? Panel estimates from PISA. Journal of Development Economics, 104, 212–232. https://doi.org/10.1016/j.jdeveco.2012.08.002

27.

Harari

M. B.

Herst

D. E. L.

Parola

H. R.

Carmona

B. P.

(2017). Organizational correlates of public service motivation: A meta-analysis of two decades of empirical research. Journal of Public Administration Research and Theory, 27(1), 68–84. https://doi.org/10.1093/JOPART/MUW056

28.

Heinrich

C. J.

(2007). False or fitting recognition? The use of high performance bonuses in motivating organizational achievements. Journal of Policy Analysis and Management, 26(2), 281–304. https://doi.org/10.1002/pam.20244

29.

Houston

D. J.

(2000). Public-service motivation: A multivariate test. Journal of Public Administration Research and Theory, 10(4), 713–728. https://doi.org/10.1093/oxfordjournals.jpart.a024288

30.

James

Witten

Hastie

Tibshirani

(2013). An introduction to statistical learning: With applications in R. Springer.

31.

Katz

L. F.

Krueger

(1991). Changes in the structure of wages in the public and private sectors (NBER Working Papers Series). NBER.

32.

Kaufman

Kraay

Mastruzzi

(2007). The worldwide governance indicators project: Answering the critics. World Bank.

33.

Keppeler

Papenfuβ

(2022). Understanding vertical pay dispersion in the public sector: The role of publicness for manager-to-worker pay ratios and interdisciplinary agenda for future research. Public Management Review, 24, 1846–1871. https://doi.org/10.1080/14719037.2021.1942531

34.

Lee

S. Y.

Whitford

A. B.

(2009). Government Effectiveness in Comparative Perspective. Journal of Comparative Policy Analysis: Research and Practice, 11(2), 249–281. https://doi.org/10.1080/13876980902888111

35.

La Porta

Lopez-de-Silanes

Shleifer

Vishny

(1999). Quality of government. The Journal of Law, Economics, and Organization, 2015, 690–694. https://doi.org/10.1016/B978-0-08-097086-8.75048-3

36.

Lazear

(2000). Performance pay and productivity. American Economic Review, 90(5), 1346–1361. https://doi.org/10.1257/aer.90.5.1346

37.

Lazear

Rosen

(1981). Rank-order tournaments as optimum labor contracts. The Journal of Political Economy, 89(5), 841–864.

38.

Liu

Tang

T. L.-P.

(2011). Does the love of money moderate the relationship between public service motivation and job satisfaction? The case of Chinese professionals in the public sector. Public Administration Review, 71(5), 718–727. https://doi.org/10.1111/j.1540-6210.2011.02411.x

39.

Nistoskaya

Dahlberg

Dahlstrom

Sundstrom

Axelsson

Mert Dali

Alvarado Pachon

(2021). The quality of government expert survey 2020 (Vol. 2020, Issue March). The Quality of Government Institute.

40.

Oliveros

Schuster

(2017). Merit, tenure, and bureaucratic behavior: Evidence from a conjoint experiment in the Dominican Republic. Comparative Political Studies, 51(6), 759–792. https://doi.org/10.1177/0010414017710268

41.

Olken

B. A.

Pande

(2012). Corruption in developing countries. Annual Review of Economics, 4, 479–509. https://doi.org/10.1146/annurev-economics-080511-110917

42.

Papenfuß

Keppeler

(2020). Does performance-related pay and public service motivation research treat state-owned enterprises like a neglected Cinderella? A systematic literature review and agenda for future research on performance effects. Public Management Review, 22(7), 1119–1145. https://doi.org/10.1080/14719037.2020.1740300

43.

Pearce

J. L.

Perry

J. L.

(1983). Federal merit pay: A longitudinal analysis. Public Administration Review, 43(4), 315. https://doi.org/10.2307/975834

44.

Perry

J. L.

(2016). Building global public administration knowledge. Public Administration Review, 76(4), 533–534. https://doi.org/10.1111/puar.12588

45.

Perry

J. L.

Engbers

T. A.

Jun

S. Y.

(2009). Back to the future? Performance-related pay, empirical research, and the perils of persistence. Public Administration Review, 69(1), 39–51. https://doi.org/10.1111/j.1540-6210.2008.01939_2.x

46.

Pfeffer

Langton

(1993). The effect of wage dispersion on satisfaction, productivity, and working collaboratively: Evidence from college and university faculty. Administrative Science Quarterly, 38(3), 382–407.

47.

Freedom House (2018). Freedom in the World 2018: The Annual Survey of Political Rights and Civil Liberties. Rowman & Littlefield Publishers.

48.

Quratulain

Khan

A. K.

Sabharwal

(2019). Procedural fairness, public service motives, and employee work outcomes: Evidence from Pakistani public service organizations. Review of Public Personnel Administration, 39(2), 276–299. https://doi.org/10.1177/0734371X17718029/ASSET/IMAGES/LARGE/10.1177_0734371X17718029-FIG2.JPEG

49.

Rainey

H. G.

Bozeman

(2000). Comparing public and private organizations: Empirical research and the power of the a priori. Journal of Public Administration Research and Theory, 10(2), 447–469. https://doi.org/10.1093/OXFORDJOURNALS.JPART.A024276

50.

Rauch

J. E.

Evans

P. B.

(2000). Bureaucratic structure and bureaucratic performance in less developed countries. Journal of Public Economics, 75(1), 49–71. https://doi.org/10.1016/S0047-2727(99)00044-4

51.

Ritz

Brewer

G. A.

Neumann

(2016). Public service motivation: A systematic literature review and outlook. Public Administration Review, 76(3), 414–426. https://doi.org/10.1111/puar.12505

52.

Rouen

(2020). Rethinking measurement of pay disparity and its relation to firm performance. The Accounting Review, 95, 343–378. https://doi.org/10.2139/ssrn.2867041

53.

Schanzenbach

(2015). Explaining the public-sector pay gap: The role of skill and college major. Journal of Human Capital, 9(1), 1–44. https://doi.org/10.1086/680596

54.

Schick

(1998). Why most developing countries should not try New Zealand’s reforms. World Bank Research Observer, 13(1), 123–131. https://doi.org/10.1093/wbro/13.1.123

55.

Schuster

Meyer-Sahling

J. H.

Mikkelsen

K. S.

(2020). (Un)principled principals, (un)principled agents: The differential effects of managerial civil service reforms on corruption in developing and OECD countries. Governance, 33, 829–848. https://doi.org/10.1111/gove.12461

56.

Transparency International. (2022). What is corruption? Author. https://www.transparency.org/en/what-is-corruption

57.

Treisman

(2007). What have we learned about the causes of corruption from ten years of cross-national empirical research? Annual Review of Political Science, 10, 211–244. https://doi.org/10.1146/annurev.polisci.10.081205.095418

58.

Trevor

C. O.

Reilly

Gerhart

(2012). Reconsidering pay dispersion’ s effect on the performance of interdependent work: Reconciling sorting and pay inequality. Academy of Management Journal, 55(3), 585–610.

59.

Trevor

C. O.

Wazeter

D. L.

(2006). A contingent view of reactions to objective pay conditions: Interdependence among pay structure characteristics and pay relative to internal and external referents. Journal of Applied Psychology, 91(6), 1260–1275. https://doi.org/10.1037/0021-9010.91.6.1260

60.

U.S Merit System Protection Board. (2008). Attracting the next generation to mining. Author.

61.

Vroom

V. H.

(1964). Work and Motivation. John Wiley & Sons.

62.

Weibel

Rost

Osterloh

(2010). Pay for Performance in the Public Sector–Benefits and (Hidden) Costs. Journal of Public Administration Research and Theory, 20(2), 387–412. https://doi.org/10.1093/jopart/mup009

63.

Wooldridge

J. M.

(2010). Econometric analysis of cross section and panel data (2ed ed.). MIT Press.

64.

World Bank. (2014). Pay flexibility and government performance: A multicountry study (Issue June). Author.

65.

Yanguas

Bukenya

(2016). ‘New’ approaches confront ‘old’ challenges in African public sector reform. Third World Quarterly, 37(1), 136–152. https://doi.org/10.1080/01436597.2015.1086635

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.24 MB