Baez CamargoIntegrity SystemsSubnational & Sectoralred and black heads in profile with opposite color people pointing toward exclamation mark in brain area

In recent years, increased interest in understanding how social norms may play a role in fueling and perpetuating corruption has sparked interesting discussions about what the implications may be for anti-corruption practice. (See, for instance, “Anti-corruption through a social norms lens,” “What Anti-Corruption Practitioners Should Read About Social Norms,“ and “Collective Action on Corruption in Nigeria: A Social Norms Approach to Connecting Society and Institutions.”)

This literature has done a lot to clarify why social norms are highly relevant to understand in cases where certain corruption practices have become widespread and entrenched. However, the discussion is far from settled, not least because of the difficulty defining what constitutes a social norm.  Here I adopt the definition proposed by Church and Chigas (2019): the mutual expectations held by members of a group about the right way to behave in a particular situation. Therefore, if most people believe acts of corruption are accepted and expected by those around them, it is not difficult to see why corruption can become incredibly difficult to eradicate.

In one of our previous research projects at the Basel Institute, we found evidence on how expectations of reciprocity and solidarity are associated with practices of bribery and favouritism in the health sectors of several East African countries. Among other things, we found that users of public health facilities in countries like Tanzania often offer unsolicited bribes and “gifts” in order to create a relationship with the healthcare provider. The expectation is that having a “provider friend” is an effective way to gain access to treatment in contexts where patients of public health facilities experience significant queues and wait times, and where essential medicines and medical supplies are in short supply.

How can we diagnose whether there is a social norm that is underpinning these observed behaviours? In order to discuss how to identify social norms, we need to first differentiate between four interrelated, yet distinct, concepts:

Concept Example
Personal attitudes: What the individual believes are fair, appropriate behaviours to adopt in certain circumstances A health worker may believe that it is wrong to accept any gift from a patient.
Descriptive social norms: Shared beliefs about what others do in certain circumstances A health worker believes that most health workers in the community routinely accept gifts from patients.
Injunctive social norms: Beliefs about what behaviours others approve or disapprove of in certain circumstances A health worker believes most other health workers think that it is fine to accept gifts from patients
Behaviours: What people actually do when confronted with certain circumstances A health worker accepts/rejects the gifts that users offer.

Thus, social norms are different from personal preferences and from actual observable behaviours, which makes them tricky to measure. The trick to identifying a social norm is whether there is an informal enforcement component of the norm. In other words, we know that we are confronted with a social norm when those who adhere to the prescribed behaviour are in any way rewarded (e.g., reputation, friendships, smiles) and those who break with the norm are somehow punished (e.g., social isolation, gossiping, frowns).

An important message that the social norms research brings to the fore is that anti-corruption measures that exclusively target individual incentives and behaviours are unlikely to yield the desired outcomes in contexts where collective beliefs generate social pressures that overpower individual preferences. Moreover, factors associated to sociality often are learned through processes of enculturation that are of a more intuitive (even unconscious) nature and not a result of deliberate cost-benefit calculations, which suggests the merits of informing anti-corruption practice (as well as policymaking more generally) with insights from behavioural science. (See the World Bank’s “World Development Report 2015: Mind, Society, and Behavior,” OECD’s “Behavioural Insights,” and the Institute of Development Studies’ “Using behavioural insights to address complex development challenges,” as well as the work of our project collaborators at the Behavioural Insights Team.)

In our research findings from Tanzania, we found a generalized perception by users and providers of health services that bribery and gift-giving is widespread, expected, and accepted in public health facilities. Furthermore, the findings indicate that those who refuse to engage (such as providers who reject gifts) suffer social costs, such as gossiping or bad-mouthing, whereas those who accept gifts are recommended to other users. Thus, we found out that the exchange of gifts and bribes at public health facilities in Tanzania is a social norm.

Next comes the proverbial “so what?” question: What are the implications of such findings for anti-corruption practice? More concretely, what type of intervention might be suitable to address this kind of petty corruption in the delivery of health services? Our GI-ACE project, “Addressing Bribery in the Tanzanian Health Sector: A Behavioural Approach,” aims to provide evidence to respond to these questions.

The main goal of the project is to design and test an intervention to tackle the social norms of bribery and gift-giving in health facilities. While we are still in the process of finalising the details of the intervention, the plan is to work with both providers and users of health services. In the case of users, the intervention will seek to work with trusted community-based social networks to promote the dissemination of anti-corruption messages. (Stay tuned for more on the rationale and methods for working with social networks in our next blog installment by my co-investigator Tobias Stark.)

To rigorously assess the impact of our intervention, we are developing research instruments that will help us measure the four concepts delineated above: personal attitudes; perceived descriptive social norms; perceived injunctive social norms; and behaviours among users of public health services regarding bribery and gift-giving. Measuring all four is extremely important in order to produce evidence that can inform future anti-corruption interventions to address social norms of corruption.

For instance, one possibility could be that personal attitudes differ from the social norm, meaning that most people might prefer NOT to engage in bribery and gift-giving, but may do so, nonetheless, because they feel it is socially accepted and expected. Another possibility is that social norms are aligned with personal attitudes, whereby most people actually tend to believe individually that it is fine to give and receive gifts in exchange for expedited service or other perks at the health facility.

The corresponding interventions would look quite different depending on which of these two scenarios is true. In the first case, the intervention could seek to somehow disclose the actual preferences of people to dispel the expectation that bribery and gift-giving in health services are collectively endorsed and rewarded. In the second case, the intervention would likely involve some manner of awareness-raising and educational activity to inform people about the negative impacts of bribery and gift-giving in the context of the provision of public services, and especially in health where such actions may literally entail life-or-death consequences.

Currently, we are conducting preliminary research activities that will help us finalise the intervention design and the measurement approach. Piloting of the research tools also will give us indicative evidence on the prevalence of the relevant attitudes, social norms and behaviours. The research team and I look forward to reporting on our progress as we move forward with the project’s implementation.

Leave a Reply

Your email address will not be published. Required fields are marked *

Post comment