What to do with social norms? Exploring the relation between a social norm and actual behavior

(1)

What to do with social norms?

Exploring the relation between the social norm prohibiting saving in secret and actual behavior in Eastern Province, Zambia

Sandra G.C. Bleeker

Master Thesis African Studies

(2)

What to do with social norms?

Exploring the relation between the social norm prohibiting saving in secret and

actual behavior in Eastern Province, Zambia

Sandra G.C. Bleeker

Master Thesis African Studies

Universiteit Leiden

The Netherlands

14 July 2020

Supervisor:

Prof.dr. M. Dekker

Second reader:

Dr. D.J. Money

Photo title page: men and women from Mbang’ombe village who participated in the group and individual interviews, the female interpreter and me on the right. ©Sandra G.C. Bleeker

(3)

Abstract

This study investigates the normative framework and the experienced reality of a social norm to enable effective policymaking, which improves the financial situation of the underserved in Eastern Province, Zambia. Social norms are becoming increasingly part of social science research, but working with social norms is not straightforward. This thesis contributes to this field of knowledge by giving in-depth qualitative knowledge of how a social norm influences people in rural Zambia. A previously conducted quantitative research revealed tensions between the desire for individual control over money and social and personal norms prohibiting secrecy, advocating the communal over personal interests. During a fieldwork period of ten weeks, seven group and twenty-two individual interviews were conducted. The group interviews show the content of the social norm and how salient it is in the communities where the interviews were conducted. The individual interviews shed light on how a social norm and its behavioral prescriptions are personally experienced. The main conclusion is that it is context-dependent on how social norms influence people's behavior choices, and this differs with time and personal circumstances. The behavioral prescriptions of the social norm prohibiting saving in secret do not guide all community members' actual behavior all the time. Some people sporadically elude or violate the norm secretly, while others do it openly, and again others never transgress. This norm is dynamically and flexibly interpreted depending on people's living situations. Effective policies should include this variety of households and living situations to improve the financial situation of the underserved.

(6)

Acknowledgements

Thank you for reading my master thesis. This work is the product of many different people, while my name is on top. Without a thank you note to everyone who helped me, who worked with me, who challenged me, this thesis cannot begin.

Firstly, I would like to thank my supervisor, Marleen Dekker, for providing me the opportunity to engage in this subject and for her useful feedback and comments. Even though we got stuck in the Covid-19 pandemic, we managed to communicate well. Thank you, Marleen, also for the Westerdijk premie contributed, which enabled the research team to pay a small sum of money to the participants in the villages.

Then I would like to thank the organization and employees from Financial Sector Deepening Zambia. Thank you for being flexible and allowing me to start with the research instead of the internship. This has secured my data collection, especially with the unforeseen global health crisis. Thank you, Kingsley, Msanide, Floyd; thank you, Thandiwe, for all the work and help. A very big thank you goes out to the Director and all amayi and asibambo (crew members) from Tikondane Community Centre. Thank you for my stay, for the interviews, for the fun, and the insights you gave me. Thank you especially Jason, Tigris, Ennie, Alice, Christine, Martha and Doris. It is a shame that we could not say our proper good-byes, but I will return again – and please take care of my cat?

I also want to thank all the participants from the different interviews. Thank you for the openness, thank you for sharing your knowledge on the topic. I count myself so lucky for your willingness to engage with me and the other research team members.

At last, I thank my personal support system made up of family and friends. It has been a crazy ride so thank you for sticking with me.

Thanks everyone.

(7)

Abbreviations

Abbreviation Description

FGD Focus Group Discussion

FSD Financial Sector Deepening (Zambia)

MCD Moral Case Deliberation

(8)

List of figures

Figure 1: Interview Participants Mbang'ombe Village. ______________________________ 1 Figure 2: Conditions necessary for adherence to social and descriptive norms. __________ 14 Figure 3: Map of interview locations. ___________________________________________ 21 Figure 4: Second round data analysis – village ____________________________________ 61 Figure 5: Second round data analysis – town _____________________________________ 61 Figure 6: Third round data analysis – Chimwa Village ______________________________ 62

(9)

1. Introduction

Human behavior is complex and the subject of research, in- and outside academia. Finding out why people behave in specific ways, which motives they have, and how they expect others to behave is of interest to researchers in developmental studies (Harper et al., 2018), economics (Blay et al., 2018), psychology (Cialdini et al., 1990), and many other disciplines. The studies mentioned recognize that social norms, descriptive norms, and social and personal expectations are part of decision-making regarding behavior choices. However, it is not clear-cut which norms and expectations guide which behavior and how they are related. This thesis aims to investigate one social norm and its relation to people's actual behavior in a qualitative way to contribute to the body of literature on social norms in development studies.

Most social norm studies are primarily quantitative (e.g., Stankov, 2011; Perugini & Gallucci, 2001; Stavrova & Fetchenhauer, 2015). These example studies measure social norms in different ways, without questioning if, why, and how social norms are the driving force behind behavior. Studies that balance quantitative and qualitative methods can map an overview of social norms and their influence on behavior that is the most representative of reality (Bicchieri, 2017). One such study is a research project that investigated how social norms influence financial decision-making in rural Zambia (Barr et al., 2020). This thesis offers further qualitative depth to this mixed-methods study.

1.1 Focus and Scope

In the financial inclusion debate, it is seen as significant to know how people come to financial decisions because this sheds light on how well they understand financial systems and how well financial services are designed for the financially underserved (Quiñonez, 2017; FSD Zambia, 2020). The recently conducted quantitative study of Barr et al. (2020) in 22 communities in Eastern Province, Zamia, revealed that individuals desire control over money, even if this means compromising household efficiency. One way to have individual control over money is to save in secret; however, there is a social norm prohibiting this behavior. The individual desires to have and control money on one side, and the widespread disapproval of saving in secret on the other are the focus of this thesis. It seems that saving in secret cannot coexist with widespread disapproval of individual behavior. Thus, either saving in secret does not happen, or the behavior prescriptions of the social norm prohibiting saving in secret are not as respected as portrayed by the research participants in Barr et al. (2020).

This research explores the normative framework of a social norm and its lived reality experienced by rural and peri-urban inhabitants of Eastern Province, Zambia. Cristina Bicchieri’s model of social norm identification provides the basis for this framework. Her ‘rational reconstruction’ of what a social norm is, offers an outline for quantitative and qualitative research and is the central theory of this thesis. Social norms can be followed automatically, the ‘heuristic route’, or through rational consideration, the ‘deliberational route’ (Bicchieri, 2006: p 4-5). Social norms do not need to be followed to influence daily life. Norm avoidance is also behavior influenced by a social norm (ibid). For example, a person will

(10)

not ask for help, to not be held accountable for repaying the favor he asked for: the social norm of reciprocity is actively evaded – and thus influencing behavior choices.

How social norms influence people's behavior choices is context-dependent, and it differs with time and personal circumstances. This thesis shows how the behavioral prescriptions of the social norm prohibiting saving in secret affect community members' behavior. Some people sporadically elude or violate the norm secretly, while others do it openly, and again others never transgress. This norm is dynamically and flexibly interpreted depending on people's living situations. This knowledge of how and when precisely the norm is adhered to or violated can be used to make financial inclusion policies more constructive in the long term (Kinzig et al., 2013). Understanding how this social norm operates can also fuel norm change to make the norm more empowering in the long run (Harper et al., 2018).

1.2 Research question and rationale

Social norms become increasingly part of social research and policymaking on local and international levels (Harper et al., 2018; Towns, 2012). Social norms are a force to be reckoned with, as they can be destructive (Scharbatke-Church & Hathaway, 2017) and empowering (Kinzig et al., 2013). How this 'reckoning' is done most useful is not apparent, partially because social norms are not universal and partially because academia is not unanimous about the definition and usability of social norms. To 'reckon' with social norms, one should first know how they operate. The main research question of this thesis is:

WHAT ARE THE BEHAVIORAL PRESCRIPTIONS AND LIVED REALITY OF THE SOCIAL NORM PROHIBITING SAVING IN SECRET IN EASTERN PROVINCE,ZAMBIA

?

The following six sub-questions assist the answering of the main question:

(1) What is a social norm?

(2) Is there a social norm that prohibits saving money in secret?

(3) What are the behavioral prescriptions with regards to saving in secret?

(4) To what extent are these behavioral prescriptions followed in the private sphere? (5) How does the aspect of time influence norm adherence?

(6) What is the lived reality of the social norm prohibiting saving in secret?

This study's main contribution is to map the normative framework and the experienced reality of a social norm to enable effective policymaking, which improves the financial situation of the underserved in Eastern Province, Zambia. By conducting different types of group interviews and formal and informal individual interviews, this research offers qualitative information and methods to the study of social norms. This thesis also contributes to an

(11)

understanding of rural and peri-urban households in Zambia and how the social norms and codes of conduct between spouses are interrelated and influence each other.

1.3 Thesis outline

Chapter two structures the theoretical framework within which this thesis is written. The most fundamental theory for this thesis is derived from the model of Cristina Bicchieri (2006) and the previously conducted quantitative research into financial decision-making and social norms in Zambia (Barr et al., 2020). This chapter answers sub-questions (1) and (2).

Chapter three describes the methods of data collection, analysis, and usage in the empirical chapters. Even though the methods in this research are not originally part of social norm research in the Social Sciences, they show that qualitative group and individual interviews can uncover meaningful information about social norms.

As it becomes clear from Chapter two, there are several instances of tensions between what is ‘ought to be done’ and what people prefer to do. The most significant tensions between adhering to the norm and fulfilling individual desires to control money are the paragraphs of Chapter 4. These tensions represent the normative framework of the social norm. There are noteworthy differences in the meaning and significance of the social norm prohibiting saving in secret between the village and town. This chapter is divided along the lines of those geographical areas and answers sub-questions (3), (4), and (5).

Chapter five explores the social norm's lived reality in terms of norm violation with sanctions as consequences. When norm deviation occurs, and it cannot stay hidden, nor can it be justified in the community, different types of sanctions can be used as punishments. Sub-question (6) is answered in this chapter according to the different experiences of the communities in the rural and peri-urban areas.

(12)

2. Theoretical framework

This chapter gives insight into the theory on social norms, their presence in everyday life, and how they are perceived by those who live with them. The first section explores the definition of social norms and answers sub-question (1). The following section will look into norm eluding behavior and the social acceptance of this. The previously conducted quantitative study into social norms and financial decision-making in Zambia is discussed next, which provides an answer to sub-question (2). The last section offers short reflections on the theory.

2.1 Social norms

Social norm theory is found across several disciplines, such as social psychology (Stavrova & Fetchenhauer, 2015) and sociology (Sullivan, 2009), but also neurology (Bas-Hoogendam et al., 2017; O'Callaghan et al., 2016) and behavioral economics (Blay et al., 2018). Most definitions describe how social norms act as rules or 'social facts' and mention others' expectations as an incentive to follow a particular prescribed behavior pattern (e.g., Sullivan, 2009; Koursaros, 2017; Roex & Rözer, 2018). Other definitions are more elaborate than the former when describing how social norms come into play in situations where self-interest and the interests of others collide (O’Callaghan et al., 2016). There is no single definition of the term ‘social norm’ nor is there agreement on how they influence behavior (Cialdini et al., 1990). Scholars arguing against norm influence conclude that all behavior can be ascribed to the following of certain norms and that human behavior is only sometimes in line with the dominant social norms (ibid). The arguments for the usefulness of social norm research emphasize that social norms are context-dependent and situational cues offer individuals in new social settings assistance with their decisions on how to behave (ibid). Certain social norms will be activated in certain situations at certain times, but this activation depends on time, context, and person (Bicchieri, 2006).

Another complication with the definition of social norms is that the noun ‘norm’ has more than one meaning, both in academia and everyday life. One meaning is 'what is commonly done', and the other is ‘what is commonly approved’ (Cialdini et al., 1990). Cialdini, Kallgren, and Reno recognize this difference in the definition of norms, and they emphasize that both segments ‘represent separate sources of human motivation’ (1990: p. 203). These different sources of human motivation are the bases for descriptive and injunctive norms. From the perspective the researchers give in their article, descriptive norms are those norms that influence our behavior by offering the evidence what others do. Descriptive norms portray what is commonly done and that this is beneficial. Injunctive norms, on the other hand, urge behavioral choices based on moral rewards and sanctions. These norms prescribe what ought to be done in a particular situation to socially 'fit in'. Besides these two 'external' categories of norms, personal norms affect behavior through the personal approval or disapproval of one's conduct (ibid).

Cristina Bicchieri takes the definition offered by Cialdini et al. and develops it further to provide a testable, workable definition (Bicchieri, 2006). With the 'rational reconstruction of what a social norm is', Bicchieri develops a model of four necessary conditions for a social

(13)

norm to be followed and creates a definition shareable across the many disciplines (2006: p. 10). The four conditions are the contingency condition, the conditional preference condition, the empirical expectations condition, and the normative expectations condition.

Condition 1 is the contingency condition: it is a prerequisite that people know that a specific behavioral rule exists and applies to the situation. Condition 2 is the conditional preference

condition: preference for conforming to a social norm is dependent on two types of

expectations. When those expectations are met, that individual has a predisposition to conform. In certain situations, these expectations can be present. The individual 'will exhibit a preference for the norm and thereby experience positive utility for conformance and negative utility for nonconformance’ (Blay et al., 2018: p. 195). In similar situations, those expectations can be absent, and conformance does not need to be experienced as positive or necessary.1

Condition 2a captures the first type of social expectations: empirical expectations. These are the expectations that enough other people conform to the norm. What number of persons is perceived as ‘enough other people’ – what Bicchieri calls 'a sufficiently large subset of the community' – varies with the different thresholds people have (2010: p. 298). Condition 2b covers normative expectations. This condition comes in two versions, 2b and 2b'. The first comprises normative expectations without sanctions. These are expectations of an individual that others expect her to act norm accordingly. The second type of normative expectations prevails in situations in which the individual expects that those others prefer that she conforms to the norm, Condition 2b’. Those others are even willing to sanction (reward) the individual if she discards the norm (adheres to it) to secure norm adherence. These expectations are called normative expectations with sanctions (Bicchieri, 2006). The injunctive norms of Cialdini, Kallgren, and Reno, Bicchieri calls social norms; descriptive norms are the same in both models. Bicchieri’s conceptualization also acknowledges the diversity of norms, and it considers different mechanisms for conformity for each norm. The definition this thesis is working with, Bicchieri’s rational reconstruction of social norms, consists of the four elements explained above. Only when all four are ‘present’, a social norm is guiding the behavioral choices in a particular situation for an individual. This model recognizes that social norms depend on expectations and that those expectations differ between individuals. What one experiences as normative expectations, another internalized, and only needs empirical expectations to activate norm adherence. For the latter, the norm is descriptive and not injunctive. Figure 2 gives an overview of the conditions necessary for social and descriptive norms.

A social norm is a shared belief of a specific context community that contains behavioral prescriptions that people prefer to follow when enough relevant people follow the prescriptions and when they know that other people expect or prefer them to conform.

1_{Section 2.2 further investigates norm eluding behavior, which occurs when conformity is not necessary, and} when the norm's behavioral prescriptions are conflicting with personal interests (Bicchieri, 2010).

(14)

2.2 Norm eluding and violating behavior

Social norms often exist in situations where personal interests and public benefits conflict with each other (Bicchieri, 2010). An example of this is when a husband wants to drink beer with his friends, while there is little money available. His interest is to drink beer, but the communal (household) interest is to save up that money to provide for the future. For certain people, empirical expectations and the knowledge that other people think they should follow a particular norm is enough to follow that norm. However, others only follow the norm when there are positive or negative consequences to following or discarding that norm (ibid). For the former, there is probably no apparent conflict between their personal and the collective interests: they experience congruence between the social norms' behavior prescriptions and personal values (Hassan et al., 2019). For the latter, their personal interests are likely conflicting with the normative expectations of society. The way people deal with the behavioral prescriptions of the norm in question could be through negotiation between norm and personal interest, a rejection of the norm and the experience of sanctions, or the repressing of personal values and the experience of frustration or stress (ibid).

Because of the different conditions necessary for norm adherence, there are different forms of norm violation. It is not uncommon for certain norms to be openly defied when sanctions against these violations are insignificant or lacking (Bicchieri, 2017). When the violator knows that she can get away with it, she will have less trouble with transgressing. Other norms are openly defied when the violator wants to defy that which the norm symbolizes on a larger scale, like social taboos (ibid). For example, women who publicly join self-defense classes in a patriarchal society where gender-based violence is highly prevalent, such as Nepal, is one such norm violating group (Standing et al., 2017). Most women who lived in temporary shelter

Descriptive Norm Condition 1: Contingency Condition 2: Conditional Preference Condition 2a: Empirical Expectations Social / injunctive Norm Condition 1: Contingency Condition 2: Conditional Preference Condition 2a: Empirical Expectations Condition 2b: Normative Expectations without sanctions Condition 2b’: Normative expectations with sanctions or

(15)

camps after earthquakes in Nepal wanted to participate in self-defense classes but were shy and uncomfortable if men came to watch (ibid). The young generation of women feminists, outside of the shelter camps, take these classes openly: they are proudly 'no longer accepting they are second class citizens' (ibid, p. 56). They meet with resistance, especially from their in-laws, but they persist in challenging the patriarchal values that make women second class citizens. Another form of open defiance of a social norm can be when people experience that norm as suffocating or harmful. However, only a small portion of the context community is transgressing and doing so proudly (Bicchieri, 2017).2

Norms can also be, often on a larger scale, secretly violated or simply evaded. People defy social norms more quickly when the outcome of this defiance can stay hidden (Bicchieri, 2010). At the same time, when it is widespread knowledge that a large part of the context community is actively violating a norm, normative pressure will start weighing less (Bicchieri, 2017). It can also be that people experience a discrepancy between their empirical and normative expectations. When they hear a particular behavior occurs, while it is prohibited, it becomes clear that this violating behavior is accepted at least by the norm violating part of the community (ibid). When the relevant expectations are absent or conflicting, the preference for norm adherence can disappear, especially when the opposing personal interests are still in place. An example is illegal downloading of music or series: while people know – perhaps vaguely – that it is illegal to do so, the knowledge that so many others violate the norm overrules any normative authority (ibid). This awareness makes it less attractive to act norm accordingly, and more enjoyable to violate the norm and get the music or entertainment for free.

Hassan et al. researched how social norms influence identity formation among coming-of-age youths in Pakistan (2019). They found seven strategies young adults use to negotiate normative expectations and personal interests (ibid). For this thesis, the negotiation and conflict tactics are especially interesting, because they can only appear in contexts of tensions between the normative expectations from society and personal values. Norms can form the boundaries in which the youths make autonomous decisions (Strategy 3). Sometimes norms are slightly altered so that the autonomous decision can be approved by significant others (Strategy 5). When norms are too suffocating, they can be avoided or rejected (Strategy 6), or personal interests can get suppressed (Strategy 7).

Living in a high normative cultural context means that the subjects have a strong knowledge of what they should be doing or what they ought to do in everyday life and specific situations (Cialdini et al., 1990). Nevertheless, this is not necessarily congruent with what the individual wants or needs. The Pakistani young adults would, at times, choose which normative expectations to fulfill and so adhere to selected social norms, but not to others (Strategy 4). In their specific context, this would mean that they would act according to some family members' normative expectations. In other situations, they would question traditional values and not follow suffocating norms while embracing others.

(16)

2.3 Social norm research

Social norms are most notably researched in labs or experimental game settings: two or more people are paired together to 'play' a game. There are many types of games, and in each type, the participants have different roles, and there is a different aim. Often, one person is the decision-maker, and the other is the receiver, and the participants need to divide some money amongst themselves. In the labs or rooms where these games are played, the researchers control the conditions carefully. The most condition is the information the participants receive, to ensure that they do not use other motivations in their decision-making than social norms. The decisions people take are thought to depend on social norms, such as fairness, reciprocity, and benevolence. Through the games, it is possible to see what most people choose in certain situations (Bicchieri & Xiao, 2009). The participants' expectations can be manipulated to see if people start to make different choices when different expectations are prevalent.

Outside the lab, the main form of measuring social norms is through vignettes and surveys, wherein a particular focus is placed on normative expectations. In the field, it is often impossible to manipulate social expectations and other variables, so the games are not the best methods to understand how behavior works (Bicchieri, 2017). With Bicchieri's model, however, it is now possible to design vignettes and interviews to understand people's empirical and normative expectations. Bicchieri proposes a combination of lab tests and field experiments because both methodologies combined provide a realistic overview of the social practices under investigation.

2.3.1 Social norm research in Zambia3

One such study, where lab tests and field vignettes were merged, took place in Eastern Province, Zambia. The project aimed to investigate the social norms that guide and constrain the financial decision-making of Zambian men and women. The study consisted of experimental games to quantify and vignette surveys to identify social norms. In the allocation game, married couples were given a set of choices to allocate money between them: one of them was randomly chosen to be the decision-maker and the other the receiver. In the vignette conducted afterward, a set of choices was proposed to the participants, and they were asked to evaluate their appropriateness.

The research team used different workable definition strategies to identify the strength of social norms. One was to pinpoint the manner and pace with which social approval declines when an individual chooses to behave deviating or violating the behavioral prescriptions. This is in line with Condition 2b of Bicchieri's proposal of normative expectations without sanctions. Another way was to analyze how likely or severe punishment will be in case of norm violation. This comes down to Bicchieri's normative expectations with sanctions, Condition 2b’. However, Barr et al. use a third way:

(17)

it can be defined with reference to the strength of the correlation between the behavioral prescription and actual behavior.

From the research, it became clear that spouses are willing to compromise household efficiency to keep individual control over money. Of the two spouses, wives are even more willing to compromise when their actions can stay hidden; husbands are not more inclined to compromise when their actions can stay secret (p. 11). Additionally, social norms do not inform those decisions on compromising household efficiency to keep individual control over money. The answers given by the participants were not sufficiently in session to conclude this (p. 13). However, the quantitative data points out that there is a social norm 'forbidding saving in secret from one's spouse' (p. 20). The participants evaluated the appropriateness of possibilities to spend spare cash, such as ‘spending it on treats for the family’ or ‘saving it in an e-wallet without the knowledge of the spouse’. There is a clear but not absolute rejection of the choices to save money without the spouse's knowledge, see Appendix B.

My thesis research started from this point to identify the content of this social norm. It investigated how strong the correlation is between the social norm against saving in secret and actual behavior, and what the content and behavioral prescriptions are of the social norm prohibiting saving in secret.

2.4 Reflections and limitations

That there is no straightforward explanation in academia what a social norm is, is a difficulty and an opportunity. It is a strain because there is no foundation on which one can build further and find deeper meanings. On the other hand, it offers researchers like me a chance to investigate different perspectives and opt for new methods of working with and defining social norms in everyday life.

This research aims to show how qualitative data can help understand the social norms identified through quantitative research. This form of research has not been conducted often because of the divided opinions of researchers about social norm research. The gross of the theory presented here is based on experimental game theory performed in labs with their carefully controlled variables. The workable definition of Bicchieri is a framework to look at social norms differently; however, it is mostly based on lab-studies. I think Bicchieri's definition lends itself to qualitative research, precisely because individuals perceive social expectations differently.

(18)

3. Methods

This chapter presents the methods that were the foundation of the data collection and analysis in this research. The primary methods are two specific forms of group interviews. There were also individual interviews conducted to subtract more clarity and qualitative depth from the answers given in the group interviews. The group interviews form the basis of the normative expectations’ framework of the social norm prohibiting saving in secret in the rural and peri-urban areas. The individual interviews add depth to this framework, but they also provide information about empirical expectations and individual values. All English interview guides can be found in Appendix A.

First, the methods of data collection, analysis, and usage are summarized. Then the interviewees and study areas are discussed in more detail. Lastly, ethics and limitations are considered.

3.1 Data collection, analysis, and usage

3.1.1 Interview methods

The first group interview format was the Moral Case Deliberation (MCD) method (Tan et al., 2018). MCDs are used in the medical field to decide an ethical dilemma by following the ten steps described below. Usually, the MCD is completed in one sitting. Between eight to twelve participants with relevant backgrounds, a case presenter, and a facilitator participate. The cooperators help the case presenter make the 'best' choice in an ethical or moral dilemma, for example, ending the life support of a patient in a coma.

The MCDs in my research were set up to investigate the social norms regarding saving money (in secret) and 'decide' how or if one should save in secret. First, the procedure and the case are introduced. In my research, these steps took place on the day the research team visited the villages to introduce the research to the participants. The following step is the formulation of ‘the moral question and moral dilemma’. This step took place the days of the interviews, and thus the ‘real’ first step in the interviews. The questions started on general terms, whether a woman [man] could save in secret. Then the participants were challenged to think about what saving in secret means for the household and the spouse. Furthermore, the participants were asked to reflect on what moral damage is done by saving in secret. The interviewees were also asked about the meaning and moral damage of not saving in secret. Step four and five serve as clarification and analysis for and by each participant to understand the different motivations behind different choices fully. In step four, the questions were about general saving and spending behavior and the code of conduct between spouses regarding financial decisions. In step five, the research team looked for specific types of people for whom and relationships in which saving in secret occurs. This was an attempt to make the topic more approachable for the participants by enabling them to give examples of people who definitely can save in secret. Step six is where the participants can come up with alternatives. In step seven, participants are asked to make an individual choice. The choices made in step seven are discussed in step eight. In this research, the participants unanimously

(19)

agreed that saving together with the spouse was the proper way to behave, so there were no differing opinions. In step nine, the case presenter can make a well-informed and thought-through choice with the participants' help. The last step is an evaluation of the method, and participants are encouraged to give feedback.

The second form of group interviews was a semi-structured in-depth focus group discussion (FGD). Two FGDs were held at Tikondane Community Centre. This center is located in the peri-urban region at the outer side of Katete Town. The participants were staff members of the community center and living on or around the premises or in Katete Town. The participants were asked general questions concerning the research topic and questions reflecting on inconsistencies and ambiguities within the MCD data.

The first type of individual interview was a structured questionnaire. These interviews took place right after the MCDs in the villages. They were held to provide three participants of each MCD the opportunity to speak individually, without the rest. The participants were chosen by the interpreters or volunteered.

The second type was a semi-structured interview, and these were conducted after the FGDs were translated. After reviewing the FGD and MCD interview translations, new questions and ambiguities had arisen. There was still time and opportunity to organize ten semi-structured interviews, and I conducted them in English with members of the community center. Soon after I had conducted the tenth interview, the Covid-19 pandemic broke out, and I returned home.

3.1.2 Data analysis

The data analysis took place after I returned home from the field. Firstly, I went through all interviews without taking any notes. The second time, I went through the MCDs and the FGDs to write down key and recurring elements. The data was divided into three main themes: statements about Relationships, statements about Women, statements about Men. These statements were assorted per the two geographical areas of the interview locations: village (MCDs) and peri-urban (FGDs) see Figure 4 and 5 in Appendix C. The third time I went through the interviews, I arranged the data in the same three themes, but now divided them per location per gender, see Figure 6 in Appendix C. All these data groupings were compared with each other to provide the foundation for the empirical chapters.

The MCDs were employed to start a conversation and discussion in the groups in the villages, but it stayed out. I suspect it stayed out because most participants are convinced that saving in secret is not appropriate when in marriage. When most people agree, there is no need for a vivid discussion. The villages we visited were rather small, so if an individual has a different opinion and shares this with a group that rejects that kind of thinking, public shame or other difficulties could be the result. The FGDs took place in a peri-urban environment where people are less dependent on each other, and there is more need for individualism than in the village. Having a different opinion in an environment where there is space for that is less damaging, and the participants were freer to speak their minds in these discussions.

(20)

The individual interviews were analyzed after the group interviews. They were especially useful in understanding specific aspects in the MCDs and FGDs as many questions aimed to clarify inconsistencies in the group interviews' data. Not all questions from the interview guide were repeatedly asked when I deemed that sufficient clarification was already given on that specific topic. The answers were compared between the different genders and ages of the participants.

3.1.3 Data usage

The answers given in the MCDs are used to formulate the normative framework in which the norm's behavioral prescriptions are set (Chapter 4). The MCD interviews were separated by village and gender to enable comparisons between the village answers. The group interviews are referred to by type, gender, and village for clarification (e.g., ‘MCD Women Mbang’ombe’). To improve readability, Tikondane Community Center is shortened to 'Tiko' (e.g., ‘FGD Men Tiko’). The individual interviews are also referred to by type, gender, and place. The interviews are numbered to secure the anonymity of the participants (e.g., 'Individual Female Tiko 3' and 'Individual Male Chimwa 1’).

3.2 Study area and interviewees

Figure 3 gives an overview of the interview locations of this research. They lay within the Katete District of Eastern Province, where Barr et al. also conducted their study. It was essential to stay within the same area because then both studies took place in environments where – presumedly – the same social norms and values are salient. To avoid possible bias, the interviews for this thesis were held in communities not included in the study by Barr et al. For the MCD participants, the internship organization provided a small payment: ZKW20 for the villagers and ZKW50 for village heads. The interviews at Tikondane were conducted without payment. Instead, I provided soda drinks and small snacks for the contributors. The village heads and the center director allowed us to use the buildings where we held the interviews without extra cost. The total number of contributors to this research is sixty-eight, excluding the research team of four members.

3.3 Ethics

All participants were eighteen years and older and were asked consent for the interviews beforehand. Afterward, the participants signed or fingerprinted the consent form. Those forms are kept by me in physical form and kept for at least five years. At Tiko, we asked the participants to give their consent orally. The consent is recorded at the end of the interviews: the discussion leader asked the permission for translation and usage of the answers given. All participants consented to this in Chinyanja. Those parts of the recordings are isolated and saved as separate files on my external hard drive.

Most group- and individual discussions were recorded with an audio recorder. Concerning data storage, the raw audio files are saved on an external hard drive. It is in my possession, and I do not share it with other parties. Some interview fragments have been incorporated in

(21)

short podcasts, which can be freely accessed online. No personal information is shared without people's consent. One group photo is used as a cover photo, but the consent that I use it in any way was agreed to when it was taken. The plain text files and the 'polished' English transcriptions are saved on my laptop, secured with a password, and on my external hard drive. There are readme (Word) and metadata (Excel) files, explaining the contents of the separate folders containing the data of this research on this external hard drive. The content of the recordings is not damaging or endangering individuals, but they are documented entirely anonymous to secure this.

3.4 Limitations and reflections

The most substantial part of the interviews was conducted in Chinyanja, so I only knew what was being said after transcription and translation. I was not able to step in during the interviews to explore a specific topic or answer further. This was challenging during the translation, as I found inconsistencies and gaps in the data. However, these inconsistencies forced me to expand the data collection by conducting more individual interviews. I collected more personal opinions and explanations, which has aided my understanding of the topic considerably. The data from the individual interviews are irreplaceable and helped me to shed light on the differences between the rural and peri-urban areas.

One limitation is that the four interviews of the men in Chimwa have failed. The only way to use the data from those interviews is through the notes the interpreters had made. Those notes do not come close to the detailed transcriptions we would have made from the

(22)

recordings if they had succeeded. The notes capture the core of the discussion and, in that way, contribute value. The recordings from the men’s interviews in Mbang’ombe did succeed and were transcribed in detail. While the original planning spread the village interviews over four weeks, the interviews were conducted in the week of my arrival in Eastern Province. Even though this had a negative consequence that some recordings failed, in hindsight, this fast pace secured my data collection. Especially with the shortened available time due to the Covid-19 epidemic, I feel lucky to have collected all this data.

The method of MCD was not the most suitable interview form in retrospect. While translating, it became evident that the method was unclear to the participants and, to a certain degree, the interviewers. I concluded this after finding that the team emphasized different topics at times than I would have wanted or that seemed logical to me. The main aim of an MCD is to find individual alternatives and opinions within the social and ethical dilemma at stake: this did not happen. The group setting did bring a clear contextual overview of the normative expectations regarding saving in secret, which became the first empirical chapter.

The interviews in the villages were organized to prevent participants from joining more than one MCD. However, one female participant joined both the pilot discussion and the women’s MCD in Mbang’ombe. The woman is closely related to Chief Mbang’ombe, owner of almost all the land in and around Katete district. We did not send her away because she helped the female interpreter explain certain concepts so that the women could better understand the questions. Her presence was authoritative, which may have hindered the freedom of speech of the women in lower social ranks. She sometimes steered the conversation in a certain way of her favor, by cutting women off or answering for them.

While reviewing her interventions during my data analysis, I realized that her contributions often contradicted what the other women said. The other women would sometimes agree with her and say that she was right. At other times, however, the women would challenge her views too. It became clear that her presence was a physical attendance of the social norm: she would say what the women should do while also aware that that is not always compatible with actual behavior. Her interventions were a form of correcting behavior, that could only happen in certain instances. I take her interventions as a form of data: what was happening was a physical enactment of the social tensions this thesis is trying to point out. While I could not find much space for this realization in my thesis, I made a podcast about it.4

The focus group discussions at Tiko were held after I held a general presentation of my research topic. These unplanned interviews were prepared quite hastily, and they turned out too structured for my liking and repetitive. However, the fast paces of both the MCD set-ups and the FGDs challenged me to become a better and more adaptable researcher. I will take with me that it can be useful to act and organize fast, but that there should be time available to deepen the understanding of eventual inconsistencies.

(23)

4. Tensions between norm and reality

Section 2.3.1 showed that most participants from the study of Barr et al. find saving in secret ‘very inappropriate’. Their rejective stance reveals a clear awareness that prescriptions on saving’s behavior are in place. This level of awareness fulfills the Contingency Condition of Bicchieri’s social norm definition model. The sections in this chapter investigate the specific content of the empirical and normative expectations, Condition 2a and 2b of Bicchieri’s definition model, in two different geographical areas.

However, the primary focus is not what people say those expectations are, but the inconsistencies between the external expectations and personal values. While the quantitative data shows a trend to save together with the spouse, participants from village and town mention motivations justifying saving in secret. It seems that in this case, the fulfillment of both empirical and normative expectations cannot induce norm obedience for all members of the community. In the public sphere, the social norm is valued, as can be taken from the numerical data. However, ‘[people] say something and practice something else’ (Individual Female Tiko 9).

This chapter is divided into three parts. The first section is primarily based on empirical data from the MCDs. It looks at the correlation of the norm and reality in the village by answering sub-questions (3) and (4) for the rural area. The second section is based on the data from the FGDs and individual interviews at Tiko and answers the sub-questions (3) and (4) for the peri-urban area. The last section investigates how time and timing influence people’s preference for the adherence of the social norm prohibiting saving in secret and answers sub-question (5).

Social norms are more frequently openly violated when public sanctions are mild or absent (Bicchieri, 2017). The social norm against saving in secret is openly deviated from when a person has a justification that is recognized by the rest of the community. This recognition will limit or eliminate the social punishment against the norm violator. Open defiance is accepted when one spouse drinks or gambles heavily. It is noticeable for other people that the deviation of the social norm prohibiting saving in secret by the other spouse begins with the extensive misbehavior of the first. Heavily drinking or gambling usually occurs in open bars resulting in that other community members can observe this improper behavior. The deviation of the social norm under investigation is for the greater good of the household and therefore justified. The household’s well-being is more important than upkeeping the behavioral prescriptions because the greater good of the household is also the village's greater good.

The participants indicated that the situation and improper behavior have to be extremely compromising to the rest of the family that saving in secret is publicly allowed. A (repetitive) quarrel or a difference in understanding is not enough for the public. However, it can be enough reason for an individual to consider saving in secret. It is not uncommon that people practice against the norm in their private environment while pretending to behave norm accordingly in public. What this private environment looks like is context-dependent: it could

(24)

be entirely secret if there is no one to trust, but others ‘tell someone in the house that you are keeping money somewhere’ (FGD Women Tiko).

4.1 Norm and reality in the village

4.1.1 Structure of the household

In a rural area in Eastern Province, men and women often work together on the field and earn most of their income by selling their harvest once a year. The appropriate way of managing finances in a husband-wife household is to decide how to spend the earnings in beneficial ways (MCD Men Mbang’ombe). This can entail, after paying certain expenses and next farming year’s necessities, ‘buying something large’, like cows, or saving the money in a savings group (MCD Women Chimwa). There should be a certain level of cooperation, but the husband is the head of the house. He is entitled to lead in the decision-making process because the patrilocal, Christian set-up of the society vindicates this. In all interviews, village and town, the husband's role as the head of the house and the main decision-maker was recognized.

The husband will most likely decide how any earned money is spent. In particular families, the wife has a say about it, but that is not as widely acknowledged or elaborated as the man’s position. In the women’s MCD in Chimwa, the topic of deciding about earnings was discussed, and three different viewpoints came up:

(1) It is just the power of the man to tell me how I should use that money.

(2) The man is the one who decides about money, unless if a woman has a program with the same money, then maybe she can decide.

(3) When the money comes at home, the man needs to make a budget for a certain thing and the remaining money he gives to the woman so that she can budget for the things that they need at home because the woman is the one who stays at home and knows what is supposed to be bought.

(MCD Women Chimwa) The first point is seen as the ‘traditional’ way of living between husband and wife, where the husband has more authority in decision-making than the wife. The second viewpoint is similar to the first; however, now there is space for discussion of options and opinions. However, the emphasis on ‘maybe’ makes it evident that the husband is most likely to decide about that money. The last point is that most women in the discussion resonated with, but it is not clear if that is how they live with their husbands or if it is the way they want to live. In this situation, the husband entrusts the wife with a certain amount of household money to independently solve pressing household issues (MCD Women Chimwa).

In the men’s group discussion in Chimwa, two of these three viewpoints came up as well. When the men were asked whether it was correct for them to decide about money in the household independently, they answered:

(25)

(1) It is good, yes.

(2) No – in a situation where a man decides on his own, it ignores that you are a couple. (MCD Men Chimwa) The first answer overlaps with the first point of the women: it is appropriate that men decide about money because it is their power. Men are expected to provide for the house, so they need the power and space to do so. The second overlaps with the second and third points of the women: it is suitable that men consider their wives’ opinions. Both men and women are part of taking care of the household, so both opinions should be heard.

None of the participants in the villages stated that it is appropriate for a woman to decide on her own about money in a ‘normal’ situation. This normal situation is a marriage in which husband and wife live together in harmony and unity, without long-lasting frictions. Most interviewees agreed that making financial decisions individually is inappropriate and not beneficial for the family, whether done by the husband or the wife. It is proper that when one of the two spouses earned money, it is shown to the other and that a decision on what to do with that money is made (MCD Women Chimbanda).

The main argument is that men have more authority than women in (financial) decision-making because the woman lives in the husband's house, and he married her. The man marries the woman in the rural areas by paying her family a particular bridewealth or lobola (Individual Male Tiko 5). Marriage is a consensual affair among the largest ethnic group(s) in Eastern Province, the Chewa (Johnson, 2018). Through the lobola and the woman's consent, the husband earns the right to be the head of the house and decide about money the woman has brought home. One male participant explains what the most appropriate behavior is with regards to money spending:

Here in our village, the norm says that the man is the head of the house because he is the one who plays a bigger role. (…) We look at the man firstly as the head of the house of that family, the man gets his idea and gives it to his wife, but it does not mean that the man can do things on his own or get the money on his own from town, no.

(MCD Men Mbang’ombe) The male participants from this village advocate that they think men and women should be equal. In contrast, the women describe that, in their reality, there is a hierarchical difference between the genders:

Yes, [a man manages to save in secret] because a woman cannot go to a man and start searching [in] his trousers. (…) he wants to make sure that if there is any money in his trousers, it should go where he is going. However, for a woman, if she maybe finds money, she will tell the husband because she does not want to bring conflict in the home.

(26)

(MCD Women Mbang’ombe) Men are known to keep money on them, and they go to great lengths not to show their wives. However, women are better off showing the money they have because there will be conflict when they are found with money, which their husbands did not give (MCD Women Chimwa). Women and men have, in theory, the same rights to decide about the money of which both of them know. They have this right because they work together to bring the money into the home. The husband is leading the decision because he plays a more prominent role, which he earned by marrying the woman and paying the lobola. However, some men perceive paying lobola as making the woman their property (Individual Female Tiko 1). One male participant reflected on this:

This [not sharing of ideas before acting] is one of the reasons why many marriages fail because us men we normally have more authority over our women and not thinking about them. (…) when we have money, we just say it is me, it is me. We do not consider them. We forget all the problems that we have gone through together.

(MCD Men Mbang’ombe) Women perceive this behavior as overshadowing (MCD Women Mbang’ombe). It is friction between differing viewpoints on the code of conduct of husband and wife and how life should be lived. The participants frequently mention that the expectations of 'ways of living' in the household can be the origin of marital problems. If there is a 'mismatch' between these expectations of ways of living, then this friction can, over time, lead to conflicts. When a husband does not consider his wife's needs or suggestions, the wife can start feeling subjugated and mistreated. On the other hand, some wives want their husbands to take on most of the responsibilities, while the husband appreciates a wife with whom he could discuss different ideas (MCD Men Mbang’ombe). The husband can perceive his wife’s behavior as lazy or heartless, and he will wonder if he made a mistake marrying her.

4.1.2 ‘The way you live at home’

The ‘way’ husband and wife live together determines their preference for norm adhering behavior, also of the social norm prohibiting saving in secret from the spouse. Social norms are adhered to, partially to satisfy significant others' normative expectations and partially because adherence can be beneficial (Bicchieri, 2006). If a spouse does not find her husband's expectations significant, she has less reason to adhere to norms that are important to him. This section explores these differing 'ways of living' in greater detail by reviewing the level of 'unity, cooperation, and togetherness' in the marriage (MCD Men Mbang’ombe).

In rural families, there has to be unity for the husband and wife to accomplish their goals (MCD Men Chimwa). Unity within the marriage is the glue that holds people together and the motivation for ‘cooperation and togetherness’. The interviewees from the peri-urban region of Katete Town recognized this necessity for unity in the village family. Some of their statements on unity are:

(27)

It is important to do things together because in this way you can plan together and see how to make the future better, budget together. In rural areas, they always sit together. (Individual Female Tiko 1) Unity is very important in a family because, in our family or traditional set-up, we depend on each other. (…) That is why everyone should work together. In this big set-up, if there is no cooperation, the set-up cannot exist.

(Individual Male Tiko 3) Because we are able to help each other, your neighbors and your relatives will be on your side to help you. (…) Here, we use our own things, food, house. We do not spend too much here in the villages.

(Individual Female Tiko 8) The importance of unity is made clear through references to the traditional way of life and the fact that an individual cannot accomplish as much as a family. The last statement adds that rural life is organized not so much around money, but around sharing. All these motives to strive for unity are based on the conception that life in the village is most enjoyable and, in general, better in a united home, family, and community. The alternative, division or individualism, is seen as undesirable: the 'picture' of the household looks better when united (Individual Female Tiko 9). Togetherness within the household means that all members have the same goals and ideals. Cooperation is that all members act and work together in ways that can bring the family to those goals. Marriage makes man and woman one body and makes them act as one, according to the Christian marriage (i.a. MCD Women Chimwa). Husband and wife should be trusting each other, they should want the same things and work together, and they should stand by each other at all times. They should fulfill the gendered division of tasks, wherein the man is supposed to take care of the income, and the wife is supposed to take care of the husband and house.

Caring for keeping up the gendered division of roles and the correct code of conduct is genuine for most couples at the beginning of their union. In this time, they are still new to each other and are still trying to impress the other with correct behavior (MCD Women Mbang’ombe). Ordinarily, this newness starts wearing off after some time, and instances of dishonesty and mistrust occur. Spouses get to know each other's behavior, and they adapt their own behavior correspondingly (MCD Women Mbang’ombe). It often happens that after a short time in marriage, people find out they do not know and understand each other (Individual Male Tiko 2). The man and the woman have different goals and ideals, and some are unwilling to adapt and work together. Especially if there was a no or a too short period of courtship and the couple rushed into marriage, the two will start doubting one another after a while, and the internal unity starts disappearing (Individual Male Tiko 5). Without unity, the marriage is missing that vital glue.

(28)

Some participants stated that when a lack of unity and abundance of mistrust cannot be resolved, the couple should divorce (FGD Men Tiko). If there are children involved, divorce becomes a less attractive option, especially for women in the village. Women are supposed to take care of the children, and they will end up in a challenging position when they do not have income (Individual Female Tiko 8). That is why ‘uneducated women’ and women without any independent means have to accept the husband’s improper behavior because they will be worse off divorced (FGD Women Tiko). According to most participants from the peri-urban interviews, the idea of husband and wife becoming one body is a perception of the past. When asked to estimate the level of trust between village spouses, all ten individual interviewees said that there is never complete trust between the partners. Two reasons stand out:

[There is] not a hundred percent trust. (…) A family looks honest (…), but when it comes to savings, they do it separately. It shows they do not trust each other. (…) After selling the harvest, the husband returns to the village with nothing; all the money wasted. (…) Many of these will divorce (…) or saving on her own.

(Individual Female Tiko 9) In certain cultures, it is understood so well that unless certain things are kept aside secretly, there is a time that death comes (…) and the relations will come and take everything and leave the woman [or man] empty’.

(Individual Male Tiko 5) The woman’s opinion shows that husband and wife in the village set-up need to have discipline to make do with the yearly income. However, it often happens that one wastes that money – on drinking, prostitution, or affairs – and then the couple either breaks up or saves separately to make the marriage work. However, they keep up a façade of honesty and good morals for the outside world.

The second point touches another reason that spouses should not be absolutely trusting of each other: cultural traits. This habit of property grabbing happens mostly among the Ngoni, Chewa, and Nsenga people, so those spouses will make sure to hide something of value at their parents’ house ‘to have something in the event of future breakup’ (Individual Male Tiko 5). This ‘arrangement’ produces dishonesty and disunity.

There are serious trust issues between husband and wife, judging from the ease of money wasting and the potential property grabbing of extended family members. Some of those trust issues find their roots at the foundation of the marriage:

This happens a lot in the villages. Especially among those who (…) have not been to school, all they can do in the village is to get married; there is nothing else. If we (…) court for three or four months, I could pretend to be a lamb, while I am a goat. The time is too short. The foundation is very poor; that is how [people] can be together without trust. Some might love each other. Others do not.

(29)

(Individual Male Tiko 2) The foundation of a marriage is defined as how well husband and wife are fit for each other. When the personal expectations of 'ways of living' that both spouses initially differed, the foundation is weak or bad. In the village, especially in the public sphere, marriage is portrayed as the best way to live. A unified home is better and more enjoyable than an individual or divided house. According to the participants at Tiko, marriage is the 'only thing to do' for villagers. Marriage is getting by in the best way. Besides, an unmarried person has less social status than a married person: the unmarried person is often made fun of (Individual Male Tiko 5).

Many marriages are based on normative pressure and prioritization. Putting up a picture of unity for the outside world is, in some cases, all that unity is. Lack of ‘real’ unity lowers the weight of violating the norms that are in place to bind the household together because the spouses do not care for fulfilling the other’s expectations. There is no preference for cooperation and togetherness, so the normative pressure from within the household is not binding to work together. There is still that pressure in the public sphere because it is still not accepted that the spouses save individually in the open if it is not necessary. Saving money in secret becomes a reasonable option to meet personal needs if separation is not desired. The money that one spouse saves alone could be a fund for individual needs – such as soap, sanitary pads, or beer – to avoid conflicts today, or insurance for the future when divorce is considered.

4.1.3 Conclusion

The behavioral prescriptions within the household are that both spouses keep to the gendered division of tasks and roles, in which the husband leads the family. As the household head, the husband is the primary decision-maker and income generator because of the lobola paid and amenities provided by him. However, he should not act on his ideas only – he is the one who took the woman in, so why would he not want to share decisions with her? (MCD Men Mbang’ombe). Women are expected to take care of the household and the husband and work on the field if necessary. She is not allowed to make independent decisions about money. Husband and wife are supposed to budget together to achieve the highest goals they can while saving money to prepare for difficult times, and they should not be hiding money from each other.

Depending on the level of 'real' unity in a marriage, the possibility of violating certain norms becomes imaginable and even preferable over norm conformity. The foundation of the marriage and 'the way they live at home' are both determining factors for norm conformity because they define the significance of the spouse’s expectations. For husbands and wives, there are – in theory – the same behavioral prescriptions, whereby cooperation and togetherness play a fundamental role. In practice, husbands can effortlessly violate the behavior prescriptions by keeping money on their own because they have the right to decide about that money. When a wife finds it essential to have (more) salt or soap in the house, but

(30)

her husband denies her money for this, saving in secret can become a reality to provide for those personal needs. Especially when she knows that her husband is keeping money for himself, she can judge his behavior as unfair and improper, which can justify the improper saving in secret she conducts.

4.2 Norm and reality in the peri-urban

4.2.1 Structure of the household

Katete town is a peri-urban area, where the rural and urban ways of life meet each other. There are tar and gravel roads, most houses are connected to the electricity grid, but cooking is done on charcoal because the grid is not reliable. In this area, farming is not the only way of generating income, but most families have fields on which they or hired laborers work. Where the rural and urban meet each other, hybrid ways of life come into existence. It can be challenging to estimate what others think and prefer because there is not one narrative on proper and improper behavior. This hybridity influences acceptable and unacceptable; it loosens up morals because people from different backgrounds live together in the same area, judging behavior in different ways.

One difference between the participants from the community center and the villages is that the former expect a monthly salary, and the latter are used to getting their primary income once a year (Individual Female Tiko 1). Katete town has a blend of income-generating activities that people participate in, which results in a monthly or otherwise regular pay-out. In the village, most households rely on farming for their primary income, which comes once a year after the harvest is sold. In town, it is easier for women to participate in other income-generating activities such as cleaning, selling garden produce, or poultry in the market. There are also more job opportunities for men, such as (bicycle-)taxi drivers, workers in a shop, bar, or restaurant. In the village, some have home shops where garden produces, cooking oil, salt, or talk time are sold. Villagers often do small 'piece works’, such as molding bricks or building a shed. These activities, however, are never abundant enough to sustain a family. In town, people are usually involved in a few different jobs to earn their income.

In a family living in town, both husband and wife usually bring home a salary, with which they budget. However, just like in the village, the husband is usually the one with the first call, because he is the head of the household (FGD Women Tiko). Nevertheless, there are three different viewpoints:

Lady A: Yes, the husband is the head of the house.

Lady B: Not nowadays, because if it is my money and I work for it, he does not have any authority or power over my money.

Lady A: No, you should not say that, because he, the man, has allowed you to work. Even if you work, you need to trust one another. If you love and trust one another, you need to use that money together.

What to do with social norms? Exploring the relation between a social norm and actual behavior

What to do with social norms?

Sandra G.C. Bleeker

Master Thesis African Studies

What to do with social norms?

Exploring the relation between the social norm prohibiting saving in secret and

actual behavior in Eastern Province, Zambia

Sandra G.C. Bleeker

Master Thesis African Studies

Universiteit Leiden

The Netherlands

14 July 2020

Supervisor:

Prof.dr. M. Dekker

Second reader:

Dr. D.J. Money

Table of Contents

Abstract

Acknowledgements

Abbreviations

List of figures

1. Introduction

?

2. Theoretical framework

3. Methods

4. Tensions between norm and reality