PublicationsThe U4 Blog

Since at least the mid-twentieth century, researchers have emphasised the social dimension of corruption. In his classic 1955 study of the Lucania region in southern Italy, for example, Banfield established the concept of ‘amoral familism’ to describe the importance of kinship obligations in shaping the rhythm of social interaction. More recently, an ethnographic school has argued that the ‘larger fabric of everyday social practices’ shapes patterns of corruption (see, for example, Olivier de Sardan Researching the practical norms of real governance in Africa; and Olivier de Sardan and Blundo Everyday corruption and the state). Scholars such as John Warburton and Davide Torsello continue to advocate for the study of corruption as a social process.

Developments in social norms theory and the growth of studies on corruption

While thinking on social norms has developed across a range of disciplines, scholars have more recently tied together these strands to present a more consolidated theory of social norms. Scholar Cristina Bicchieri has been influential in this and provides a ‘unified architecture’ of social norms (see, Norms in the wild: How to diagnose, measure, and change social norms,and The grammar of society: The nature and dynamics of social norms). This theory pays particular attention to the different types of norms; the links between norms and specific networks or reference groups that people identify closely with and care about; the role of social sanctions; and the varied strength of norms, as elaborated by Cislaghi and Heise in their 2018 article Four avenues of normative influence: A research agenda for health promotion in low and mid-income countries.

The number of studies explicitly tying this social norms framework to the study of corrupt practices has grown in the past decade.

The growing popularity of the social norms framework stems from the recognition that it can help address some of the puzzles presented by corruption. The social norms viewpoint reflects the view that people rarely make corrupt decisions on their own, but instead pay close attention to social information before deciding how to act.

What has emerged in the field of corruption is not necessarily a coherent research stream or recognised school, but rather a collection of studies across various disciplines that have empirically interrogated the relationship between social norms and corruption.

Qualitative in-depth analysis of social norms of corruption

Studies that used qualitative methods often focus on a particular community or region to address the question: What kind of social norms may matter for corruption – and how? Many of these studies affirm the importance of ‘in-group’ obligations in generating norms that can indirectly facilitate corruption. Urinboyev and Svensson’s (2018) fieldwork in a village in the Fergana valley of Uzbekistan shows how ‘illegal’ practices reflect not individual self-aggrandisement, but also social norms that have been generated through a traditional system of mutual aid that aims to ensure basic welfare for all residents. In areas where the state has scant presence and provide few basic services for citizens, this system of self-help is maintained by mutual expectations, meaning that each villager can expect certain behaviours from other residents. Within a limited village context, these social norms may be considered pro-social. However, once applied to the context of public office, such norms can lead to corruption as villagers put pressure on public officials who may be their kin.

Most of the recent studies using qualitative methods emphasise that injunctive norms related to corruption are rarely internalised as a straightforward guide to appropriate behaviour, but rather are maintained in conjunction with social sanctions. In Nigeria, for example, if a public official shuns corruption, he or she may be cast out as a pariah. In Uzbekistan, the close-knit social structure of ‘kin communities’ means that public officials who fail to deliver expected benefits face social sanctions such as gossip, ridicule, loss of respect and reputation, humiliation and even exclusion from life-cycle rituals. Public officials who try to keep the public office separate from the private sphere are maligned, while those who provide patronage for the village are afforded respect and social status.

Quantifying the impact of social norms on corruption

Macro-Level evidence

One of the first empirical studies testing the relationship between social norms and corruption was done by Fisman and Miguel (2007). The study draws observational data documenting the parking violations of United Nations diplomats stationed in New York City. At the time, diplomats posted to that city enjoyed immunity and could even park in prohibited locations with no legal repercussions. Some abused this privilege by parking in the middle of Fifth Avenue. The authors correlated the observed parking violations with the scores of the diplomats’ home countries on the Corruption Perceptions Index (CPI). The results showed that diplomats from countries with higher corruption levels committed more parking violations. The results provided early quantitative evidence that injunctive norms might be an important factor in explaining corrupt practices, especially in the absence of effective punishment.

Using the same data set, Kapoor and Ravi (2012) challenge the interpretation that injunctive norms drive this effect. They show that when ‘government effectiveness’ is included in the regression analysis, namely the correlation between home-country CPI and parking violations disappears. According to Kapoor and Ravi, it is the quality of a country’s institutions, not its cultural norms, that leads to the persistence of corruption.

Micro-level evidence

Scientists from various disciplines have tried to model a situation that reflects the basic components of a corrupt practice. Such experimental research aiming to uncover the influence of social norms on corruption has mostly studied bribery. In one of the most influential studies on the subject, Barr and Serra (2010) let foreign students at a British university play a bribery game. Their results reveal that students coming from countries with high levels of (perceived) corruption as measured by the CPI were more likely to bribe in the game. The tendency to bribe decreased with the length of time that the students had stayed in the United Kingdom. The results suggest that although injunctive norms seem to have an influence on willingness to bribe, people also recognise the ‘rules of the game’ and adjust to their environment.

These findings reveal that understanding the interplay of social norms and corruption requires specification. It appears that there is no single monolithic corruption norm that exists in a given society. Instead, people’s perceptions about the acceptability of corruption might differ according to their own background, the way corruption is framed, and the size of the bribe, while their willingness to engage in corruption themselves appears to be largely shaped by their belief about the frequency of corruption, in other words, by the perceived descriptive norms. Understanding the social normative forces that sustain corruption thus requires a closer look – one that differentiates between different types of corruption.

Empirical work has started to do just that. Hoffmann and Patel (2017) conducted a large survey into social norms of corruption in Nigeria. Surveying 4,200 respondents, they assessed perceived descriptive and injunctive norms across various regions, accounting for demographic background and distinguishing between different corruption types – bribery, extortion, embezzlement, and nepotism – across different sectors. The results further corroborate the view that social norms are heterogeneous. For starters, respondents perceived that bribing a police officer is less acceptable than bribing a nurse. Respondents also believed that women engage in corruption less frequently than men. At the same time, however, respondents condemned female corrupt actors more harshly than male. These results suggest that understanding the social normative forces that sustain corruption – and eventually contributing to its reduction – requires a nuanced analysis of the interplay of injunctive and descriptive norms (see, Jackson and Köbis' Anti-corruption through a social norms lens).

Where next?

Recent empirical research on social norms of corruption depicts a promising area of study that has gone through a growth spurt in the last few years. As a result, the corruption field is catching up with other fields to which social norms theory has been effectively applied, including health, school bullying, and sexual health. But gaps still remain.

It is important to differentiate between various corrupt practices. In our review, we identify a strong focus on bribery and a lack of research on ‘grander’ forms of corruption, such as embezzlement. We hope that future research combining qualitative and quantitative methods and making more refined distinctions between types of corruption can contribute to a more nuanced understanding of the social norms of corruption.

Several studies indicate that people often bribe even though they disapprove of it when they think that others are doing it too (see, for example, Köbis et al. 2015 ‘Who doesn’t?’—The impact of descriptive norms on corruption, and Abbink et al. 2018 The effect of social norms on bribe offers). There are encouraging signs that reducing these beliefs about the behaviour of others can help reduce corrupt practices (outlined in a recent article by Köbis et al. on Social norms of corruption in the field: Social nudges on posters can help to reduce bribery). Future research could shed light on the best ways to achieve lasting change in perceptions around social norms to fuel reductions in corruption. Open issues include, for example, whether and how narratives around corruption promoted by various (free) media (for example radio, TV, social media) shape both injunctive and descriptive norms.

In emphasising social norms, researchers could easily slip into an assumption that all corrupt practices can be explained by analysing the social norms in the society. That does not reflect reality, however, as other factors besides normative pressures clearly influence corrupt practices. As many of the studies confirmed, social norms do not exist in a vacuum; rather, the influence of social norms on corruption is nearly always relative. Finding out how social norms relate to other factors that drive corruption marks an important next step for this line of corruption research. Yet few frameworks exist to push this agenda forward. A corruption-specific conceptual framework that researchers can use to configure the right balance between social norms and broader institutional and structural factors would be very fruitful. In this regard, corruption scholars could build upon the conceptual work of Cislaghi and Heise (2018), who have mapped how social norms may relate to other factors within the health sector.

Finally, there is a need for more research on how social norms can be used as a vehicle for anti-corruption reforms. Little empirical research has examined social norms in the agencies responsible for anti-corruption enforcement, such as anti-corruption commissions or courts. An empirical turn to anti-corruption research could help explain why countervailing norms of public integrity are so difficult to embed in organisations – and could point to approaches that might be more successful. Recent studies have started to uncover that potential, initiating an inquiry that researchers and practitioners interested in enhancing our understanding of the deeper roots of corruption would do well to take forward.


    All views in this text are the author(s)’, and may differ from the U4 partner agencies’ policies.

    This work is licenced under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International licence (CC BY-NC-ND 4.0)