Impact of corruption on public value creation: does national developmental and technological level matter? Open Access

Corruption has become a serious concern in modern governing systems, with widespread implications for public value creation (PVC) (Bozeman et al., 2018). The Corruption Perceptions Index 2022 report (TI, 2022) has indicated the rising trends of corrupt practices globally and the negative effects on public values, including peace, social cohesion, and economic inequality. The preamble of the Sustainable Development Goals (SDGs) states that corruption, bribery, tax evasion, and related illicit financial flows deprive developing countries of an estimated 1.26 trillion USD per annum, resulting in severe negative impacts on various welfare indices, including public health, education, environmental sustainability, gender balance, and poverty alleviation (UNDP, 2023).

There are many studies on corruption (Khan, 1996; Bozeman et al., 2018) and public value creation (Moore, 1995, 2013). However, scholarly research examining the relationship between corruption and PVC is limited. The few available studies (Bozeman et al., 2018) are primarily country-specific and relatively narrow in focus. The current study aims to address this gap by investigating many countries from around the globe. While there is a large body of literature on corruption and its implications for economic growth and development, there is no consensus on a precise definition of corruption. Additionally, there are different types of corruption, each with diverse implications (Khan, 1996). In this study, corruption is defined as the exercise of public power for private gains, manifesting as bribery, nepotism, favouritism, embezzlement, and so forth (Bozeman et al., 2018).

The public value framework has received increased attention from scholars over the last two decades. Since the publication of the seminal book by Moore (1995), there has been a proliferation of literature on public value and value creation. Nevertheless, controversies remain about the exact meaning of public value. Debates and discussions primarily occur at the theoretical level (Osborne et al., 2022; O’Flynn, 2021), although some empirical studies have made inroads to the scientific literature (Meynhardt and Jasinenko, 2020; Thøgersen et al., 2021).

In the simplest terms, the public value framework emphasizes the outcomes of the activities of public managers. Public value relates to citizens’ collective expectations with respect to government and public services (Moore, 1995). It raises an intriguing question – is the arbiter of value an individual or a collective public body? Moore (2013, p. 61) further clarifies that: “the most purely public value would be the collective’s valuing of aggregate social condition against some standard of the common good or the just”. Most of the PVC literature clarifies the conceptual and theoretical underpinnings of this concept. Bozeman et al. (2018, p. 13) argues that the public value framework relies on “normative consensus about (a) the rights, benefits, and prerogatives to which citizens should (and should not) be entitled; (b) the obligations of citizens to society, the state, and one another; and (c) the principles on which governments and policies should be based”. Nabatchi (2018) distinguished between public value and public values. “Public value refers to an appraisal of what is created and sustained by government on behalf of the public” (Nabatchi, 2018, p. 60), whereas public values imply normative consensus based on “emotional and cognitive assessments of individual persons” (Fukomoto and Bozeman, 2019, p. 636).

Some authors also define public values in terms of both processes and outcomes (Sancino et al., 2018). Thus, the outcomes are only possible through a value chain in which public managers alone or in collaboration with citizens, non-profit organizations, and/or the private sector are deeply engaged in the operating environment with activities, processes, procedures, and programs (Moore, 1995; Jaspers and Steen, 2022).

Many scholars have offered arguments in favour of different conceptual dimensions of PVC. Moore (2013) developed a set of items to measure public values of police services. Benington (2011) argued that public value assessment must go beyond market and economic considerations. Social, political, cultural, and environmental dimensions of value should be included in the public value assessment. Based on data compiled from Danish government surveys, Faulkner and Kaufman (2018) identified four components of the assessment of public values – outcome achievement, trust and legitimacy, quality of service delivery, and efficiency. The authors found these four components applicable across a range of policy and national contexts. However, the debate is still inconclusive. Given the lack of conceptual clarity, there is a gap in the literature regarding how to empirically examine PVC. The current study attempts to address this gap by empirically examining the impact of corruption on PVC in a cross-country setting. The current study draws upon the work of Meynhardt and Jasinenko (2020) who identified four dimensions of public value including moral/ethical (in terms of equality, fairness, and ethicality), hedonistic-aesthetical (relating to happiness, joy, and relaxation), utilitarian-instrumental (including utility, efficiency, and effectivity), and political/social aspects (such as social cohesion and solidarity). Though there are some studies that examine the relationship between corruption and PVC, most of the literature takes a simplistic approach in measuring PVC, and thus fails to incorporate the multi-dimensional nature of PVC. Therefore, the current study contributes to the literature by elucidating the PVC-corruption nexus in order to generate renewed interest in the subject. The current study is the first one to adopt a distinctive approach in linking corruption with PVC considering its salient dimensions which were ignored in previous literature.

Limited literature empirically examines the impact of corruption on PVC. Most researchers have focused on indirect measures of PVC and investigated how corruption affects these aspects of PVC. This aligns with Meynhardt and Jasinenko (2020), who suggested four dimensions of public value, as noted earlier. Prevailing levels of corruption in a country can adversely affect the different dimensions of public value. Corruption in public organizations can cause delays in service provision, pilferage and larceny, irresponsible conduct of officials, bureaucratic intemperance, as well as patronage and clientelism (Zafarullah and Siddiquee, 2001). Park and Blenkinsopp (2011) found that corruption reduces citizen satisfaction and highlighted the role of transparency and trust in curtailing the effects of corruption on citizen satisfaction. According to Boudreaux et al. (2018), corruption undermines trust in public institutions, compromises service quality, and diverts resources from their intended purposes, ultimately diminishing public value. Furthermore, lack of transparency erodes public values.

High levels of public corruption result in higher aggregate levels of state and local debt and potentially threaten the fiscal health of the government constraining the resource availability to support other service demands (Liu et al., 2017). Cooray et al. (2017) also indicated that an increase in corruption results in increased public debt and the presence of a larger shadow economy compounds the negative effects of corruption on public debt. Corruption can also reduce PVC through other means. For example, Sulemana et al. (2017) revealed that corruption undermines the well-being of both bribe victims and recipients. High corruption levels shape the perception of unfair treatment which may result in citizens’ less cooperative behaviour with public officials (Marien and Werner, 2019). Deep corruption, characterized by actions that undermine core public values, can have devastating effects on governance and society. These corrupt actions, often legally sanctioned, can lead to public values failures such as systemic discrimination and unequal access to essential services like education (Douglas and Meijer, 2016; Bozeman et al., 2018). In light of the above literature, the first investigated hypothesis is:

Corruption is more prevalent in economically less developed countries than economically more developed countries (Park and Kim, 2020). This implies that corruption may more seriously impede effective functioning of public services in developing countries depriving citizens of basic services. Though some developing countries have adopted e-government for delivering digital services, many fail to generate public value due to low levels of trust, lack of awareness, and need of effective anti-corruption strategies (Valle-Cruz, 2019).

Fitzsimons (2009) noticed a significant increase in corruption despite public sector reforms in transitional economies, while initiatives taken by developed country governments show reduction in corruption (Bertot et al., 2010; Stamati et al., 2015). Stamati et al. (2015) discussed the role of social media in the Greek public sector, which includes initiatives such as publishing all governmental decisions on the internet to ensure transparency and openness in order to reduce corruption. Bertot et al. (2010) mentioned the anti-corruption e-government initiative called the Seoul Metropolitan Government's Online Procedures Enhancement (OPEN) system for processing citizens’ applications for government services. This system helped in increasing transparency and curbing corruption. Based on the assessed literature, the following hypothesis is proposed:

Employing information technology significantly reduces corruption in the public sector (Pathak et al., 2009) and enhance PVC (Valle-Cruz, 2019). One recent study by Park and Kim (2020) reported that e-government significantly reduces corruption. E-government encompasses operations such as electronic service delivery, basic government operations, and technology-enabled citizen participation and engagement. E-government can generate public value in terms of client satisfaction and service quality, improved trust, communication, engagement (MacLean and Titah, 2022), e-service functionalities, equity, citizens’ self-development, efficiency, and openness of public organizations (Deng et al., 2018).

Technologically less developed countries are less likely to offer innovative public services and implement smart technologies in government to create public value (Criado and Gil-Garcia, 2019). As a result, developing countries face challenges in embracing technology, such as the adoption of e-government, compared to developed countries. These challenges include inadequate infrastructure, a large digital divide, and lack of IT skills (Twizeyimana and Andersson, 2019). Regulatory inconsistencies and higher levels of corruption lead to low infrastructure investment; thus, there is a pressing need for investment in technological innovations and large-scale investments (Armijo and Rhodes, 2017). Locatelli et al. (2017) contended that public officials are key players in executing megaprojects, which are more likely to be affected by corruption.

Based on the above literature, this study posits that technologically weaker countries are less likely to curb corruption and create values for citizens in their public services in comparison to technologically stronger countries. Therefore, the third hypothesis set forth is:

Data from the Worldwide Governance Indicators database developed by the World Bank, UN E-Government Knowledgebase developed by United Nations’ Department of Economic and Social Affairs (Division for Public Institutions and Digital Government), and the Human Development Index (HDI) developed by the United Nations Development Program (UNDP) constitute the basis for the empirical analysis in this study. Details on the applied indictors are presented below.

The World Bank’s country level governance indicators are estimated for over 200 countries and territories covering the following six dimensions: voice and accountability; political stability and absence of violence/terrorism; government effectiveness; regulatory quality; rule of law; control of corruption. The indictors are based on survey answers on views of enterprises, citizens, and experts in both developed and developing countries through 30+ data sources including survey institutes, think tanks, non-governmental organizations, international organizations, and private sector firms. A detailed description of variable definitions and aggregation procedures is found in the World Bank (2023). The average of three indicators including voice and accountability, political stability and absence of violence/terrorism, and government effectiveness are used to represent public value creation (PVC) in 2021. Table 1 shows the four public value dimensions as suggested by Meynhardt and Jasinenko (2020), item examples, and suggested proxies for these dimensions. For example, the moral/ethical dimension is measured through the voice and accountability indicator, Hedonistic-aesthetical and Utilitarian-instrumental dimensions are measured with the government effectiveness indicator, and the Political-social dimension is assessed through the political stability and absence of violence indicator. Specific definitional details of these indicators are shown in Table 1.

UN’s E-Government Survey cover 193 countries and is based on over twenty decades of longitudinal research including combination of primary data (collected by the UN Department of Economic and Social Affairs) and secondary data from other UN agencies (UN, 2023). To represent the technological level of a country in 2022, The E-Government Development Index, which is an average of the three sub-indices including online service index, human capital index, and telecommunication infrastructure index, is used here.

The developmental stage of a country is measured by UNDP’s human development index (HDI) for 2021, which equally weighs three sub-dimensions within health (long and healthy life), education (knowledge), and material living standards (a decent standard of living). More specifically, the health dimension is the life expectancy at birth, the education dimension is mean of years of schooling for 25+ years old adults and expected years of schooling for children of school entering age, and the standard of living dimension is gross national income per capita. The HDI information is estimated for 191 countries. Data sources behind the index include survey data as well as macroeconomic data from national accounts. UNDP (2023) includes further details about the HDI.

Since the three different data sources cover a different number of countries, it was not possible to include all the countries of the world. Nevertheless, this study ended up with information for 187 countries. This includes populous countries like China, India, USA, Indonesia, and Pakistan, as well as much smaller countries like Brunei, Malta, Maldives, Micronesia, and Luxembourg.

The above applied indices representing levels of public value creation, corruption, development, and technology, all have different scales. A linear transformation is therefore made to make sure each of the three indices varies between 0 (worst rank among countries) and 100 (best rank among countries). Additionally, the top one-third of countries are defined as having high development respectively being technologically advanced. Summary statistics are presented in Table 2 and the variables’ distributions are presented in Figure 1.

Table 2 and Figure 1 show that all four variables are concentrated around levels 50-60 with a range of 0-100. Generally, the best performing countries within the four indicators are from the Global North (often European countries), while the worst performing countries are frequently from the Global South (often African countries) (Table 3). The variables in Figure 1 and Table 2 are utilized one by one in the analysis of the PVC-corruption relationship.

The three hypotheses stated above are investigated through descriptive and statistical measures. The first hypothesis handles the issue of a possible relationship between PVC and corruption, which is approached in Figure 2, where each of the 187 countries are represented by their level of corruption on the horizontal axis and their level of PVC on the vertical axis. There is a clear negative relationship between PVC and corruption indicating that higher levels of corruption are seen together with lower levels of PVC. Focusing on the linear relationship, it is seen that a 1-point increase in corruption decreases PVC by 0.891 point (Table 4). This effect is remarkably high as it implies that PVC reduction is around 30 points when corruption is reduced from the third quartile (75.9 in Table 2) to the first quartile (42.2 in Table 2) – a 30 points reduction represents 53 percent of average PVC (56.5 in Table 2). It is also seen from Table 4 that the corruption squared term (-0.00556) is highly significant implying that the negative effect of corruption is even higher for high corruption countries compared to low corruption countries, e.g., for countries with first quartile corruption the effect is -0.790 (a 1-point increase corruption decreases PVC by 0.790) and for third quartile corruption countries the effect is -1.164 (a 1-point increase corruption decreases PVC by 1.164). To sum up, it can be concluded that hypothesis H₁ cannot be rejected, e.g., there is evidence that higher levels of corruption is associated with lower levels of PVC.

Figure 3 illustrates the relationship between PVC and corruption among developed and underdeveloped nations. The pattern emerging is that PVC in highly developed countries is less sensitive to corruption compared to PVC in less developed countries. More specifically, Table 4 shows that a 1-point increase in corruption decreases PVC by 0.714 point in the countries with high development, while a 1-point increase in corruption in less developed countries leads to a PVC reduction of 0.987 (= - 0.714 - 0.273). The higher sensitivity magnitude (for countries with low development) of -0.273 is statistically significant. Table 4 also shows that the development-corruption interaction term is significant, which implies that the negative effect of corruption is even higher for less developed countries compared to highly developed countries. For instance, (evaluated at the average corruption level) for countries with first quartile development level (36.6 in Table 2) the effect is -0.909 (a 1-point increase corruption decreases PVC by 0.909) and for third quartile development (77.5 in Table 2) countries the effect is -0.716 (a 1-point increase corruption decreases PVC by 0.716). Hypothesis H₂ can therefore not be rejected since it is seen that in countries with low development the devastating effect of corruption on PVC is even more severe compared to high development countries.

The testing of the last hypothesis is illustrated by Figure 4, which shows that countries that are not technologically advanced have a steeper PVC-corruption slope compared to technologically more advanced countries, e.g., PVC is more sensitive to corruption in the latter countries compared to the former. A 1-point increase in corruption thus decreases PVC by 0.974 point in technologically less advanced countries compared to a PVC reduction of 0.742 in technologically more advanced countries, which is a statistically significant difference. Also, generally, the higher the technological level the lower is the negative impact of corruption on PVC (the parameter 0.00557 is statistically highly significant). Since it clearly shows that technologically less advanced countries are more affected by a given increase in corruption than technologically advanced countries, hypothesis H₃ cannot be rejected.

The findings reveal that the negative relation between corruption and PVC is observed in both developed and developing countries. These findings are consistent with existing studies that examine the link between corruption and other relevant aspects of PVC (Cooray et al., 2017; Marien and Werner, 2019; Park and Blenkinsopp, 2011; Sulemana et al., 2017; Zafarullah and Siddiquee, 2001). Though the impact of corruption is more severe on PVC in developing nations, the prevalence of corruption in developed countries also hampers PVC. In the first decade of the new millennium, the World Bank became optimistic about the drastic reduction of poverty, continued economic growth, and the improvement of social conditions in developing countries although many of these countries have had bad governance records including pervasive corruption (Asadullah et al., 2014). However, contemporary evidence shows that bad governance including corruption is still a serious impediment to economic growth and sustainable development (World Bank, 2020). It is pertinent to note that the inverse relationship between corruption and PVC remains even though different approaches have been employed here. Technology and development levels are employed in order to conduct sensitivity analysis, which reveals statistically robust findings indicating a strong negative relation between corruption and PVC. With the advent of smart technologies and governments’ drive to adopt digital transformation, many technologically less advanced nations have opted to use smart technologies in public service delivery (Deng et al., 2018). Though it can be considered a vital step towards enhancing PVC, the reduction in corruption remains a significant factor to leverage technology as an enabler to enhance PVC. The adoption of digitalization also poses challenges with regards to new venues of corruption such as cybercrime (Schia, 2018).

A lack of consensus on public value definition and low availability of relevant national level data has largely resulted in theory building research which has further produced fragmented results in terms of different conceptualizations of PVC (Osborne et al., 2022; O’Flynn, 2021). Consequently, PVC is defined in various ways and often presents a limited consensus view of public value creation. This indicates the need to take a multidimensional view on PVC to achieve a more representative presentation of PVC. Therefore, this paper offers the first step towards considering the four dimensions of PVC as proposed by Meynhardt and Jasinenko (2020) to define PVC in a comprehensive manner and suggest empirical measures of PVC dimensions for conducting studies involving multiple countries.

Governments need to take a multidimensional view of public value to optimize value creation for the public. Rather than focusing narrowly on a limited number of dimensions of public service delivery, policy needs to be strengthened in terms of various dimensions of public value such as equality, fairness, ethicality, happiness, and social cohesion and solidarity. van der Wal et al. (2016) showed how Victorian public bodies were able to enhance value through corruption prevention programs, training, and effective leadership. Enaifoghe (2023) explored how the local government system in South Africa instilled a professional service culture by promoting laws, a code of conduct, and culture. Public service managers need to ensure that government departments deliver public value in practice and provide tangible benefits to citizens through transparency and efficient service delivery. This would show real steps taken by the governments to reduce corruption levels.

Improvements in telecommunication infrastructure enhance the e-government capability, which eventually helps control corruption (Silal et al., 2023). Luna et al. (2024) examined citizens’ perceptions of functionalities of public services in Mexican cities and revealed that the impact of value creation is more profound at the societal level if there is a more transparent and accountable government, equitable society, and stronger democracy. These examples show practical implications for public service managers to enhance PVC and tackle corruption challenges.

Corruption has seriously hampered the achievement of public values both in terms of development and technological advancement. The findings reveal that return from reducing corruption is higher in more corrupt societies, less developed societies, and technologically less advanced nations. This presents an encouraging scenario for policy development in developing nations since a given decrease in corruption contributes more positively to PVC compared to developed nations. This indeed is an interesting finding which underlines the potentially large benefits of controlling corruption in more corrupt nations. However, policy needs to clearly outline objectives with relevant incentives to achieve an enhanced level of PVC. Co-creation of public value through citizens’ involvement is a key factor so that they can provide feedback and identify roadblocks in paving the way for improved PVC (Mendez et al., 2024; Osborne et al., 2022). However, Waheduzzaman (2024) contended that e-participation may not result in corruption control and suggested that co-creation through e-participation requires enhancement and re-evaluation of e-participation mechanisms and anti-corruption strategies.

To conclude, building upon the theoretical PVC framework of Meynhardt and Jasinenko (2020), PVC was operationalized, and indicators related to corruption, development, and technology were included in order to estimate the direction and magnitude of the PVC-corruption nexus and its moderators. National indicators from 187 countries are utilized. A strong, consistent, and statistically significant relationship between corruption and PVC was estimated. It is found that a 1-point increase in the level of corruption is associated with an average 0.9-point reduction in the level of public value creation. This sensitivity of PVC to corruption is significantly higher in high corruption countries, and those with a low technological level. It is found that lower levels of development reinforce the PVC-corruption relationship, which is also the case with lower levels of technological advancement. The adverse effect of corruption on PVC is consequently robust to different specifications and types of countries. Avenues for future studies on the PVC-corruption relationship could experiment with alternative weightings of the PVC sub-dimensions, application of data from various years and business cycles, and different specifications of the statistical models. Public policy can play a crucial role in enhancing PVC through equity and transparency, corruption prevention programs, training, effective leadership, and a professional service culture.

Due to the abundance of theoretical discourses on PVC, there is a need to shift the focus from theory building research to theory testing research. It is timely and relevant to converge the discussion to reach consensus on defining the scope of PVC and moving towards an empirical examination of PVC. This study solely explores the moderating effects of technological and developmental levels on the relationship between corruption and PVC, whereas various other moderating factors could also influence this relationship. Therefore, future research may consider other moderating factors such as socio-economic variables, cultural factors, regime type, country size, and governance structures that could influence PVC-corruption dynamics. Moreover, due to lack of some data for cross-country comparative studies, it is suggested that leading institutions such as the World Bank, UN, and UNDP to include specific indicators which could clearly capture the notion of PVC in their annual surveys. This will serve as a springboard to guide future comparative empirical studies involving multiple countries, as well as more replication studies in the field. This is also needed because researchers have to rely on indirect measures of PVC in order to conduct country level empirical studies. Given the lack of PVC measures, future research should be focused on developing and validating a PVC measure that can be employed in empirical studies in addition to relying on indirect measures of PVC and relevant indicators developed by leading institutions.

Another potential area of future research relates to incorporation of vital inputs from policymakers and citizenry. In this regard, the notions of co-design, co-creation, and co-production of public value can be further explored in public service delivery. Moreover, as governments face varied challenges in achieving public value across the globe, researchers can also explore these challenges and develop policy suggestions.

Limitations to data and statistical modelling are highlighted. Data wise, alternative definitions of PVC might be considered, e.g., instead of a raw average of relevant dimensions, one could weigh the PVC sub-dimensions differently. The relationship between PVC and corruption seems very stable, nevertheless, one could apply different time periods to possibly give new insights into the relationship. Instead of looking at levels of the measures, one might also scrutinize the relationship between the change in PVC and the change in corruption. Moreover, transformations or dichotomizations of variables might provide further understanding of the presented relationships, and additional right hand side variables could be developed and included. Alternative statistical modelling approaches could be considered including the generalized method of moments (GMM), which is particularly useful when dealing with potential endogeneity and/or heteroskedasticity in the data that are common in aggregate-level analyses.