Introduction
Research on public administration shows that public employees working in service public institutions are more likely to commit to values that represent their desire to benefit the public’s best interest (Perry & Wise, 1990). Other studies even have shown these employees demonstrate more altruistic (Rainey, 1997) and democratic attitudes (Blair & Garand, 1995; Conway, 2000) than employees from the private sector.
Abundant research exists about the motivations that drive public employees’ desire to perform public and social service (Agarwal & Sajid, 2017; Christensen et al., 2017; Ertas, 2016). However, scarce research focuses on bureaucrats’ attitudes on gender issues to show their commitment to the public value.
The analysis of public employees’ attitudes matters because attitudes are important predictors of behavioral intention (Ajzen, 1991). Therefore, having public employees who believe in the importance of gender equality is a potential sign they will more likely deliver services or goods fairly to women. Considering gender equality is a critical dimension of democratic consolidation (Ionescu, 2018), public employees are expected to be more committed to public value by exhibiting attitudes that promote gender equality. Public value is defined in this article as the collective purpose and aspirations of society (Moore, 1995) that corresponds to values such as transparency, equity, among others. On the other hand, we define attitude as “a relatively enduring organization of beliefs, feelings, and behavioral tendencies towards socially significant objects, groups, events or symbols” (Hogg & Vaughan, 2005, p. 150).
In addition to the commitment to public value, the contexts in which public employees operate also should matter because laws, culture, social norms, and formal and informal institutions can moderate public employees’ attitudes toward gender equality. Our primary research question is: Do bureaucrats tend to favor more egalitarian gender attitudes than other citizens? We also want to analyze the interaction between the public employee as an individual and context. So, a secondary research question is: Does the context matter?
The article shows that bureaucrats’ attitudes on gender equality play a significant role in expanding the scope of the literature on public value in public administration. Using data from the most recent wave of the World Values Survey – WVS (2017–2020) covering 77 countries – this study compares gender attitudes of public-sector workers with those of non-public sector officials. The assessed attitudes include female participation in labor market, politics, and other areas. The study offers regional analysis.
We also analyze how those attitudes are conditional on structural opportunities for women in countries using the Gender Inequality Index (GII) with UN data for 2019. Finally, we study the effect of individual characteristics, such as education, religion, and age. Findings show public employees tend to have attitudes favoring gender equality. However, that effect is contingent on the gender inequality context in which public servants operate.
This study aims to make several contributions to research on bureaucratic behavior and gender equality. First, previous studies have focused on citizens’ perceptions of gender equality (Epstein, 2007; Fortin, 2005; Huddy & Terkildsen, 1993; Inglehart & Norris, 2003; Iversen & Rosenbluth, 2010; Pérez & Tavits, 2017; Sanbonmatsu, 2002). We extend existing research by understanding how public employees can internalize and promote values that affect gender inclusion in democratic contexts. In doing so, this work enriches the literature on public value in public administration and gender studies. Second, comparing public employees’ perceptions toward gender equality with those of non-public employees helps us understand the distinctiveness of public organizations and their workers, as well as their implications for governance and democracy (Bullock et al., 2015). In doing so, our study contributes to organizational literature by uncovering the distinctiveness of public employees. Third, the influence of the context in which public value is created and affected should be accounted for by explaining public employees’ perceptions toward gender equality and the effect of individual attributes. In this sense, our study contributes by predicting the direct and indirect effects of gender inequality context on bureaucrats’ perceptions of gender equality. Fourth, studies on citizens’ perceptions of gender equality have been conducted, but cross-country empirical studies are scarce on antecedents of public servants’ perceptions of gender equality. Our analysis helps advance the explanation of perceptions of gender equality across 77 countries. Finally, this study focuses on an understudied public value: equality. Gender equality as a public value matters for the design and implementation of public policies to guarantee inclusiveness.
Drivers of Gender Inequality
Many women still lag behind men in terms of equity across different countries and areas (Blau et al., 2006; OECD, 2012; Pérez & Tavits, 2019). In politics, for example, women are still underrepresented, although they have made progress (Arriola & Johnson, 2014; Hinojosa, 2012; O’Brien, 2015; O’Brien & Rickne, 2016). Development opportunities in economic (e.g., Fortin, 2005; Schwab et al., 2017) and educational (Bertocchi & Bozzano, 2020) and digital areas (e.g., Herbert, 2017) also are skewed against women in developed and underdeveloped societies.
Several factors seem to contribute to this gender gap. Although these drivers may vary across dimensions, common drivers can be identified. Some suggest patriarchal attitudes and beliefs promote and maintain gender inequality (Epstein, 2007; Inglehart & Norris, 2003; Iversen & Rosenbluth, 2010). Others argue asymmetrical attitudes toward women affect their political representation (Huddy & Terkildsen, 1993; Sanbonmatsu, 2002) and economic opportunities (Fortin, 2005). Nevertheless, Pérez & Tavits (2017) insist the existing research fails to explain how these attitudes develop in the first place and why they persist.
In response to this failure, Pérez (2016) & Pérez & Tavits (2017, 2019) believe the gender gap relates to the language one speaks. Their reasoning centers on differences in gender encoding between genderless and gendered languages. For example, Estonian or Finnish speakers do not designate the gender of objects. Words “he” and “she” are the same in these tongues. In contrast, in gendered languages, like Spanish and Russian, speakers must differentiate genders by assigning a definite article, such as “la” for female and “el” for male objects. Based on these differences, Pérez & Tavits (2019) hypothesize that speakers of genderless tongues express more liberalized attitudes toward gender equality. Using an experiment whose subjects were 1,200 Estonian/Russian bilinguals, Pérez & Tavits (2019) found support for this proposition. In response, they conclude language may have significant effects on mass opinion about gender equality.
Another line of research has centered on the link between democracy and gender equality (Kostenko et al., 2016, p. @349775). However, the direction of this relationship seems unclear for some researchers. In Freedom Rising, Welzel’s (2013) study shows that gender equality predicts democracy. Specifically, Welzel contends attitudes toward female reproductive freedoms (e.g., abortion, divorce, and pre-marital sex) strongly predict democracy worldwide. Based on considerable evidence, Welzel tests various explanations for the rise of freedom, providing convincing support for his theory of emancipation. Welzel contends the desire for emancipation (e.g., free choice and equal opportunities) is the origin of human empowerment. This human empowerment, according to Welzel, is the basis of democracy by vitalizing civil society, feeding humanitarian norms, enhancing happiness, and redirecting modern civilization toward sustainable development. Therefore, if people support women’s empowerment (captured with support of gender equality questions), they also are more likely to support liberal democracy.
Inglehart & Norris (2003) also explore the precursors of women empowerment which they term “the rising tide of gender equality.” They note that gender equality support is systematically rising worldwide, except in the regions with Islamic cultural legacy. To identify drivers of this rising tide, Inglehart & Norris (2003) offer two theses. The first one suggests “culture matters” (2003, p. 8) because traditions, norms, and customs affect world views. Moreover, modernized shifts in attitudes affect gender equality (p. 149). The second thesis suggests economic growth leads to predictable social changes, including gender relations equalizations (p. 10, 11). In sum, modernization restructures cultural values, disrupting traditional gender roles by offering women previously denied opportunities (p. 10, 11). Besides religious attitudes and values, other sources of gender equality exist, such as factors related to the cultural-institutional context (Mencarini, 2014).
According to Inglehart & Norris, the mechanism of the “rising tide of gender equality” consists of two phases. First, industrialization brings women into the labor market, declining fertility and raising educational levels. In the second phase (in postindustrial society), women get access to professional and managerial positions and gain political influence. Inglehart & Norris use the WVS data to demonstrate that, on the societal level, richer, politically stable, and technologically advanced societies are becoming more supportive of gender equality due to generational replacement. On an individual level, younger, less religious, better educated people, especially women, tend to support gender equality in all possible domains, including female rights. Their findings, in line with Welzel’s (2013), support the proposition that attitudes towards gender equality strongly predict democratic aspirations.
Kostenko et al. (2016) analyzed the relationship between support of democracy and support for gender equality in the seven countries covered by the first wave of the Arab Barometer project. In this case, the relationship reverses because preferences for democracy and degree of religiosity are predictors of gender equality support after controlling for age, gender, and education. In their study, the hypothesis about emancipating effects of education receives empirical evidence. More religious people (those who read the Quran more often) are found to be less gender egalitarian. However, their hypotheses about younger people’s liberal attitudes and correlation of gender egalitarianism with support for democracy fail to receive empirical support. This counterintuitive result relates to the effect of age. Younger people tend to be both better educated and more conservative in the Arab world.
While the above studies examine attitudes toward gender egalitarians by focusing on citizens in general, exploring public employees’ attitudes toward gender equality becomes more relevant for the following reasons. First, public servants carry out the implementation of countries’ legislation regardless of level of government. Therefore, understanding their attitudes toward gender egalitarianism sheds light on their neutrality in service delivery and law enforcement. Second, given the previously reported relationship between citizens’ attitudes toward gender equality and democracy, it is expected this relationship also applies to public servants especially in democratic settings. Public servants are expected to embrace democratic ideals of equal rights and opportunities. Moreover, as women gain more political and bureaucratic empowerment, examining public servants’ attitudes toward gender egalitarianism becomes important. If public servants’ views of gender egalitarianism fall short, women’s political and bureaucratic representation may confront additional challenges.
At the macro level, Rubery (2013) compares the contribution of the public sector to gender equality in 14 European countries. Rubery (2013) argues the public sector contributes to gender equality through employment, which is done in multidimensional ways. For instance, the public sector offers access to higher-level and professional jobs, good pay, and pensions.
Due Billing (2005) acknowledges the “ongoing debate about whether bureaucratic practices (rules, procedures, legislations) are effective at tackling gender inequalities – or whether non-bureaucratic approaches (e.g., influences on the culture, values) are preferable from the perspective of gender equality” (p. 264). Given the negative connotation that bureaucracies have, Billing contends “we should not take for granted that all bureaucracies are essentially bad,” as considerable variations exist across bureaucracies. This variation calls for empirical studies that assess public servants/bureaucrats’ views of gender equality. Regarding this point, it is important to emphasize an analysis of public employees’ perceptions, compared to perceptions of other employees, matters because organizational theories usually have “downplayed the importance of public sector distinctiveness” (Bullock et al., 2015; Daft, 2012). Considering this interest, the article analyzes the distinction between public employees’ attitudes against non-public employees’ attitudes.
Commitment to Public Value and The Importance of Context in Explaining Gender Equality Attitudes
Public value, understood as the collective purpose and aspirations of society as a whole (Moore, 1995), is a transversal element for public administration, as it invites reflection on how to promote well-being in societies (Forero et al., 2023). Likewise, public value is a dynamic measure, as it reflects the desires of citizens and materializes the mission of public organizations, represented in turn by democratic institutions (Moore, 1995; Sanabria & Guzmán, 2021). This is manifested with specific values, such as transparency, equity, justice, integrity, among others (Covey, 1990). However, there is a need to apply the concept of public value since it is “relatively vague and fuzzy and has been rarely operationalized in concrete terms” (Ju et al., 2019).
By being involved in designing and implementing public policies that reflect the mission of public organizations, bureaucrats have internalized public value as part of their intrinsic commitment to the notion of the public. Different research in the field of public administration has shown that, indeed, public servants are more committed to public value, internalizing values such as equity and accountability (Kim et al., 2013; Weske et al., 2020).
Although the literature has advanced in categorizing public values (Andersen et al., 2012; Jørgensen & Bozeman, 2007; Kernaghan, 2003), it is still necessary to test whether bureaucrats’ internalization of public value is effectively exhibited for other values such as equality. Indeed, research work on public value creation has analyzed traditional public values, such as efficiency and effectiveness (Barrutia et al., 2022; Lapuente & Van de Walle, 2020). On the contrary, other values related to democracy such as equality are understudied. These two elements of the analysis of public value, such as studying concept in specific contexts and non-traditional public values like equality, are part of the contributions of this article.
According to the U.S. Executive Order 12674, the basic obligations of public service and foundation for ethical behavior imply that public servants “shall adhere to all laws and regulations that provide equal opportunities for all Americans regardless of race, religion, color, sex (including pregnancy, gender identity, and sexual orientation), parental status, national origin, age, disability, family medical history or genetic information, political affiliation, and military service (US Department of Interior, 2024). This expected impartiality in public servants’ treatment of citizens has encouraged scholars to investigate the importance of public servants’ fair treatment of citizens. Some of this scholarship reports that public servants’ fairness, which assumes equal treatment, contributes to citizens’ (a) trust and support for government (Grimes, 2017; Marien & Werner, 2019), (b) compliance (Esaiasson, 2010; Mazerolle et al., 2013a), (c) satisfaction with government (Erlingsson et al., 2014), (d) service quality (Suzuki & Demircioglu, 2021), and (e) organizational legitimacy among other outcomes (Mazerolle et al., 2013b).
Under the premise that public employees are more committed to public value, the article analyzes how being a public employee has a positive impact on gender equality as a fundamental public value of democratic consolidation. Precisely, the discussion about public value and its manifestations in public values represents further efforts to ensure the connection between public employees and democracy (Johnson, 2014).
Consequently, our first hypothesis states:
-
Hypothesis 1: Public employees tend to report more positive attitudes of gender equality than non-public employees.
However, the individual behavior of bureaucrats does not occur in a vacuum. The contexts where public employees live provide situational opportunities and constraints affecting their attitudes and behavior (Johns, 2006). For this article, the context in which public employees perform their responsibilities matters considerably in shaping their views and attitudes toward gender equality. As Billing posits it, “[w]hat has developed historically and has caused many bureaucracies to be gendered is not necessarily a result of the ‘laws’ of bureaucracy but rather of some influences from the ‘surrounding’ society and the prevalent social view of men and women’s roles and capacities” (p. 263).
The context, for instance, may be influenced by adopting legal instruments that seek to achieve gender and/or race neutrality. One example is the implementation of equal opportunity and affirmative action policies across governmental bureaucracies, such as equal employment opportunity laws implemented in different countries. However, as Kelly & Newman (2001) found, these laws have failed to uniformly advance interests of men, women, and minorities. To amend that, Kelly and Newman propose a new conceptual framework that moves beyond conventional demographic (passive) representation to considering substantive (active) representation.
Moreover, the social, cultural, and family context in which public servants socialize should moderate their views of gender neutrality and/or egalitarianism. Thus, religion, customs, practices, and norms that reinforce gender inequality should moderate a public servant’s attitudes toward gender neutrality. The article advances understanding of the context of gender inequality as a complex and multidimensional construct where political, social and economic opportunities are involved. Consequently, we use the GII as the measure of gender inequality that reflects inequality in achievements between women and men in three dimensions – reproductive health, empowerment, and the labor market. As the GII builds on the Human Development Index, its design assumption is the experience of gender equality reflected in several human development outcomes related to women’s health, political empowerment, and standard of living. Following the importance of understanding the context in bureaucrats’ behavior analysis, we incorporate the context of gender inequality as a potential moderator of the effect of bureaucrats’ commitment with public value. Therefore, our second hypothesis states:
Data and Methods
The article uses data from the seventh wave of the WVS 2017–2020 (Haerpfer et al., 2020). The WVS, a project started in 1981, is conducted by a network of social scientists from leading universities worldwide. Overall, the WVS project investigates political and sociocultural change based on nationally representative surveys conducted in the seventh wave in 120 countries, which represent 94.5% of the world’s population. Our analysis is based on data from 77 countries with 124,854 participants who completed the survey in 2021. The cross-national nature of WVS is its main advantage in analyzing human values, including attitudes about organizational commitment. The survey includes in its questionnaire categories on individual attributes, permitting scholars to test the micro-level assumptions on different issues. The questions included in the WVS dataset have construct validity.1 The survey was applied between 2017 and 2020.
Additionally, we use country-level data from the GII in 2019. GII is a composite measure of gender inequality that reflects gender-based disadvantages in three dimensions – reproductive health, empowerment, and the labor market. The measure ranges from 0, where women and men fare equally, to 1, where one gender fares as poorly as possible in all measured dimensions (UNDP, 2021). Then, a low GII value indicates low inequality between women and men, and vice-versa. Fig. 1 shows the country values of GII around the world, indicating that several nations, mainly in Africa and Asia, have higher levels of gender inequality compared to countries in Europe and North America.
For the analysis of the hypotheses, the dependent variable is democratic attitudes related to gender attitudes, which is measured by the following question: “Many things are desirable, but not all of them are essential characteristics of democracy. Please tell me for each of the following things how essential you think it is as a characteristic of democracy. Use this scale where 1 means ‘not at all an essential characteristic of democracy’ and 10 means definitely ‘an essential characteristic of democracy’: Women have the same rights as men.”
To measure consistency, we use another dependent variable as a robustness check related to a measure of gender attitudes in general. The question is: “How would you feel about the following statements? Do you agree or disagree with them? When jobs are scarce, men should have more right to a job than women.” See Table 1 for descriptive statistics.
The main independent variable for testing hypothesis 1 is whether an individual works in the public sector, the private sectoror non-profit organizations. To measure the effects depending on the GII , we use data from 2019.
The study controls for several individual and contextual factors. The controls included in the model are education, woman, age, size of town, ideology (1–10 points scale from left to right), and attitudes toward abortion, as a measure of liberal ideas. We use fixed effects by region in the world (Americas, Asia Pacific, Europe Central Asia, Middle East & North Africa, Western Europe and South Saharan Africa). These regions exhibit attributes that are similar in some (or most) features, such as regime type, participation levels, traditions, norms, expectations, informal and formal rules (Mendez & Avellaneda, 2022). In the cultural map that is produced by WVS Association in each wave, results show that countries within the regions have similar patterns in terms of their values (WVS, 2023). Therefore, regional analysis captures the historical background and its influence on cultural and demographic characteristics.
The number of public employees corresponds to 30% of the sample. The percentage of public employees, private employees and non-profit employees by sex and the mean or percentage of the average responses for the dependent variables for each group are included in Table 2. Regarding the GII, countries in the sample had an average score of 0.34 with a range between 0.025 and 0.795. The country with a high Gender Inequality score is Yemen with 0.795, and the country with a small Gender Inequality score is Switzerland with 0.025.
Results
As mentioned, we use individual-level data from the seventh wave of the WVS (2017–2020) in 77 countries, and country-level data from the GII. In terms of modeling strategies, the article uses ordinary least squares (OLS) regression models, as a type of linear regression model. The first round of models includes independent variables and controls, and a second round of analysis includes an interaction between being a public employee and the value of GII for that country.2 This interaction shows the way public employee behavior is moderated by the context and its effects on gender attitudes.
Table 3 shows the influence of being a public servant on attitudes favoring gender equality. These OLS regression models include data for all countries where WVS was applied.
The reason for using an OLS model specification is that the dependent variable, which is the level of agreement with the statement of gender equality as an essential characteristic of democracy, is treated as continuous. This question ranks from 1, that is “It is not an essential characteristic of democracy” to 10, that is, “It is an essential characteristic of democracy.” Findings in model 1 (without interaction) point out that being a public servant is positively correlated with attitudes that women equality is an essential characteristic of democracy (with a 90% level of confidence). On the other hand, living in a country with a high level of gender inequality (measured through the GII) has a negative effect on perceiving women’s equality as an essential characteristic of democracy.
Model 2 introduces results with the interaction between public employee and GII. Results show the positive connection between being a public employee and gender equality attitudes is contingent on the country’s gender inequality context. Being a public employee in a country with a higher GII negatively correlates with the attitudes that women’s rights are essential for democracy. Results also show that older citizens and people with more education have gender attitudes favoring equality. Similarly, people who believe that abortion is justifiable tend to have more gender equality attitudes.
In order to test the robustness of findings, we use another dependent variable correlated with gender attitudes. This variable corresponds to the level of agreement or disagreement with the statement, “When jobs are scarce, men should have more right to a job than women.” Although it is not explicitly connected to democracy, the dependent variable allows us to test the significance of being public servants to explain, generally speaking, egalitarian gender attitudes. We model this variable as an ordinal variable (five levels of agreement where “1” is disagree strongly and “5” is agree strongly). Although this variable is ordinal, it is also possible to interpret the average marginal effects of OLS with an ordinal dependent variable. For consistency purposes, we also run the linear probability models of the second dependent variable with a logistic regression (based on transforming the ordinal variable into a dummy variable), and it shows consistent results. The interpretation of the average marginal effects shows, in general, bureaucrats tend to disagree with the idea that men should have more rights to jobs than women. Table 4 in the appendix shows the coefficients of the models.
When we include the interaction between being a public employee and GII, results seem to be consistent with findings of the dependent variable related to gender equality attitudes as a component of democracy. The main finding is the negative coefficient of being a public employee on gender equality changes to positive when we consider a country’s context. In other words, in countries with higher levels of GII, being a public employee is correlated with a higher level of agreement with the statement that favors men over women when jobs are scarce. Full models with coefficients and errors are in the appendix.
As additional robustness check, we employ the Variance Inflation Factor (VIF) to test whether models have severe multicollinearity problems. The VIF test results indicate no concerns with severe multicollinearity in the models. The average VIF ranges between 4.57 and 4.61 for the OLS regression models with and without interactions. No variables of interest yield greater than the VIF score of 3. Although some region controls have VIF scores higher than 10, the collinearity in control variables is not a problem, as long as variables of interest are not collinear. We conclude models do not suffer from severe collinearity in variables of interest.
Discussion and Conclusion
This study concludes that public employees’ attitudes are aligned with attitudes that gender equality is an essential component of democracy. Bureaucrats expressed higher levels of gender equalitarian attitudes among bureaucrats compared to non-public employees. However, the positive effect of being in public service is conditional on the country where the public employee lives. This shows contextual restrictions impeding access of women to political, social, and economic life do affect the way public employees perceive them. Moreover, our results not only demonstrate a country’s gender conditions affect bureaucrats’ attitudes about gender equality as part of democracy, but also job scarcity.
We are aware of the potential limitations of an analysis that only relies on self-reported data and that is why we include the GII as a reliable measure to capture gender inequality country conditions. However, the GII faces data limitations (similar to the WVS in countries where it was not applied). Another limitation is the use of national parliamentary representation, excluding other types of representation in government. This limitation calls for future studies covering other analyses of different gender equality indicators in the public sector, such as gender quotas, to explore the consistency of results using other indicators besides GII.
Another limitation of our study is the interpretation of results between regions. In the article, we show differences exist between geographical regions, but further analysis should investigate the underlying causes of these differences. Case studies focusing on a single region are a valid and helpful methodological strategy to identify the consequences of being a public servant for civic engagement. While we know that cross-country analysis offers a broad picture of factors affecting participation, more research is needed to isolate contextual and individual variables as predictors of higher levels of participation.
Despite these limitations, this paper contributes to the literature on public administration and gender studies by showing that public employees have a stronger sense not only of public interest in general, but also democratic values such as gender equality. Public employees play a major role in promoting and implementing policies encouraging gender equality as part of democracy.
The article contributes to the understanding of public value in specific contexts, which is still lacking in literature on public administration. Likewise, the paper studies equality as a non-traditional public value that matters for the design and implementation of public policies more diverse and inclusive.
The analysis of public employees’ attitudes matters because attitudes are important predictors of behavioral intention (Ajzen, 1991). Therefore, having public employees that believe in the importance of gender equality is a potential sign they will be more likely to deliver services or goods in a fair way to women.
In terms of practical implications, recognizing that public employees have attitudes favoring gender equality can be an important and valuable asset for public managers and policy-makers. Leaders in the public sector, for instance, should promote the participation of bureaucrats during different stages of policy cycles, encouraging them to be guarantors of including gender and other differential focuses. Since public employees already have internalized attitudes favoring equality, they can be allies in defending diversity and inclusion.
