The effect of microfinance on happiness of small business owners in rural Tanzania : the happy man is not he who seems thus to others, but who seems thus to himself

(1)

THE EFFECT OF MICROFINANCE ON

HAPPINESS OF SMALL BUSINESS

OWNERS IN RURAL TANZANIA

the happy man is not he who seems thus to others, but who seems thus to himself

Author: Guido Andrea

Student numbers: 12016063 (UvA) 2614592 (VU) Delivered on: July 1st_{, 2018}

MSc program: Entrepreneurship Thesis supervisor: Dr. Joeri Sol

(2)

Preface

This is the master thesis: “The effect of microfinance on happiness of small business owners in rural Tanzania”. It contains research on the concept of microfinance in rural Tanzania.

This document is written by Guido Andrea who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

I wish to thank Kathrin, whom without I would not have managed, our supervisors, Joeri Sol and Enno Masurel, for their invaluable insights, feedback and guidance, our hosts, Nsubili and Musabila, whom without we would not have seen any wildlife, our assistants, Laurent, Irene and Mery and, finally, all of the experts and respondents that allowed us to collect the data we have.

(3)

Abstract

This study researches the effects of microfinance on happiness of small business owners in one of the poorest countries on earth: rural Tanzania, Africa. It is theorized that funds attracted through microloans are not only used for enterprise purposes, but also as a way to improve the general quality of life of the borrower. The research question centralized in this study is: what is the effect of microloans on happiness of small business owners in Tanzania? Hypotheses that have been drafted imply a comparison of two groups to distinguish (statistically) relevant differences in happiness, in personal wealth and in financial distress. The latter variable is expected to have a negative effect on the former two variables. Additionally, the moderating effects of borrowing longer (period) and more (amount) is also examined. Data has been collected on a treatment and control group to the extent that a valid comparison between the two, through applying a variety of statistical analyses, is possible.

The results of this study show that microfinance does not yield a significant difference in happiness between borrowers and non-borrowers in rural Tanzania. Explanations for the absence of a statistical difference can be found in the economic slow-down at the time of this study, too narrow a research period and too high a degree of poverty in the context in which the study took place. The short borrowing periods associated with microloans cannot make a substantial difference in the lives of these subsistence entrepreneurs or the negative side effects of borrowing, in this study concretized through financial distress, are too abundantly present. This study contributes to the discussion on the effects of microfinance, on which, as of yet, no general consensus has been reached.

(4)

3.2.2.ECONOMY 23 3.2.3.POLITICS 24 3.2.4.ENTREPRENEURSHIP 24 3.2.5.MOROGORO 24 3.2.6.MZUMBE UNIVERSITY 25 3.2.7.SUBO 25 3.3QUALITY OF RESEARCH 27 3.3.1.QUESTIONNAIRE 27 3.3.2.SENSITIVITY 29 3.3.3.DATA COLLECTION 29 3.3.4.DATA ANALYSIS 31 4. RESULTS 32 4.1GENERAL INFORMATION 32 4.2QUALITATIVE 32 4.3QUANTITATIVE 35 4.3.1.DEMOGRAPHICS 35 4.3.2.BUSINESS CHARACTERISTICS 36 4.3.3.LOANS 37 4.3.4.SCALES 39 4.4TESTING HYPOTHESES 40 4.4.1.GROUP COMPARISON 40 4.4.2.T-TESTS 41

4.4.3.REGRESSION (AND CORRELATION) 42

4.4.4.CONCLUDING HYPOTHESES 47

(5)

5.1HAPPINESS AND PERSONAL WEALTH 48

5.2FINANCIALLY DISTRESSED 49

5.3THEORETICAL AND PRACTICAL IMPLICATIONS 49

5.4BIAS AND (THUS) LIMITATIONS 50

5.5DISCREPANCIES IN THE DATA 51

5.6AVENUES FOR FUTURE RESEARCH 52

6. CONCLUSION 53

REFERENCES 54

APPENDIX I. ENGLISH QUESTIONNAIRE 59

APPENDIX II. SWAHILI QUESTIONNAIRE 62

APPENDIX III. SOURCES QUESTIONNAIRE 65

APPENDIX IV. QUALITATIVE DATA COLLECTION 69

1. LENNARD SLUITER (MSC) 71

2. DR.EMIEL EIJDENBERG 73

3. RAEL ADHIAMBO (MSC) 74

4. PROF.DR.ENNO MASUREL 75

5. GERA VOORRIPS &JUDITH ROSENBRAND –TRIPLE JUMP 76

6. DR.RAFAEL PEREZ RIBAS 77

7. DR.JOERI SOL 79

8. LAURENT DEOGRATIUS (MSC)&ANUSIATHA FAUSTINI GODFREY (LEGAL)–SUBO 80 9. H.B.MGOTOH (VILLAGE CHAIRMAN)&A.AHMED (VILLAGE EXECUTIVE OFFICER) 82

10. DR.ALBOGAST MUSABILA 84

11. KASANDA JOHN – LOCAL ENTREPRENEUR 85

11. EUGENE GRIES &IRIS DE NOBEL – DUTCH EMBASSY 86

APPENDIX V. INDEPENDENT T-TESTS AND CHI-SQUARE TESTS 87

INDEPENDENT T-TESTS 87

CHI-SQUARE TESTS 89

APPENDIX VI. FACTOR ANALYSES AND DIMENSION REDUCTION 94

FACTOR ANALYSIS -PERSONAL WEALTH 94

RELIABILITY ANALYSIS 1– PERSONAL WEALTH 98

RELIABILITY ANALYSIS 2– PERSONAL WEALTH 99

FACTOR ANALYSIS - HAPPINESS 100

RELIABILITY ANALYSIS – HAPPINESS 103

APPENDIX VII. NORMALITY OF DATA 104

(6)

APPENDIX IX. INDEPENDENT T-TEST SUM HAPPINESS AND PERSONAL

DEVELOPMENT 107

APPENDIX X. DESCRIPTIVE STATISTICS GROUP COMPARISON 109

APPENDIX XI. HIERARCHICAL ANALYSIS 111

APPENDIX XII. CORRELATION MATRIX ALL VARIABLES 115

APPENDIX XIII. CORRELATION MATRIX MICROLOAN SPECIFICS 117

APPENDIX XIV. MODERATOR MODELS 119

MODERATOR PERIOD 119

(7)

1. Introduction

Microfinance is considered an effective tool in poverty alleviation (Tedeschi, 2008) and women empowerment (Cheston & Kuhn, 2002) or both (Khandker, 2005; Afrane, 2002). Since its inception, a myriad of articles has been published on the subject. From this, we gather that microfinance is a domain that includes microloans or -credits, microsavings and microinsurances (Banerjee, 2013; Leone & Porretta, 2014). The impact of microfinance, and more specifically of microcredits, enjoys an increased interest of scholars, because its effects have split the field. Knowledge on the impact of microfinance is partial and contested. Accurately described by Hulme (2000), on the one end of the spectrum we find studies that argue that microfinance has “beneficial economic and social impacts” (p. 79) and on the other end of the spectrum there are scholars that “point to the negative impacts that microfinance can have” (p. 79). Additionally, as a result of various forms of bias, “quantifying [these] positive impacts has proven difficult” (Tedeschi, 2008, p. 504).

In other words, some scholars argue that microcredits are effective, while others do not. Using panel data from Bangladesh, Khandker (2005) found that microfinance indeed contributes to alleviating people from poverty and empowering women. These findings are impaired by work from Banerjee et al. (2015). They report on increased business investments and profits, but they also consider these changes to be marginal in a rapidly growing local economy. Copestake, Bhalotra and Johnson (2001) found that higher profits and household income only reached Zambian entrepreneurs that take out a second loan, while loan obligations made other borrowers worse off. The most striking example of partial and contested data are two analyses of the same dataset by Pitt and Khandker (1998) on the one hand and Murdoch (1999) on the other. The former find that household consumption as a result of microloans increased, especially when women borrowed funds. However, as a result of applying simpler estimations, the latter found no other impact than that it reduces consumption volatility (Roodman & Morduch, 2014). These conflicting results were never subjected to further analysis.

Although in 2006 awarded the Nobel Peace Prize, Muhamed Yunus, the patriarch of microfinance, in 2011 suffered severe critiques, voiced through The Micro Debt documentary. Additionally, in 2010 various media outlets reported on deaths associated to overindebtedness of microborrowers (e.g. Melik, 2010). Although some may argue that the reported evidence is circumstantial, this negative publicity seemed to have cast a shadow over the social goals of

(8)

microfinance. This was around the time that enthusiasm on microfinance ebbed slightly and researchers began expressing an increased interest in the actual effects of microloans. Karlan and Zinman (2011) theorize that disrupting informal lending activities (borrowing from family or the community) in itself is a negative side-effect of microlending, as these forms of borrowing are relatively safe and without constraints. Undoubtedly the high interest charges associated with microfinancing might indebt those in need even further and, briefly touched upon earlier, psychological effects of overborrowing can “do more harm than good” (p. 1278). Wanting to alleviate the poor and hungry from poverty without resorting to charity, Yunus came up with a sustainable way to contribute. Rather than giving money to the poor, he borrowed them money for them to advance their business. As the concept of microfinance developed, and subsequently research on the subject as well, it became apparent that money that was borrowed also served non-enterprise purposes. Health, housing, education and general (household) consumption would also advance as a result of attracting loans (Stewart et al., 2010). Basu and Yulek (2004) argued that such privately spending of microloans is why for policy makers and researchers it can be difficult to measure the actual effects of microlending. Although scarce, some studies have been dedicated to measuring this impact, included under the banner of subjective well-being. Despite the borrowers having available more money to consume or invest, Karlan and Zinman (2011) fail to document a significant relation of microfinance with subjective well-being.

Such qualitative aspects in turn are considered to be important determinants of happiness. Shin and Johnson (1978) distinguish “feeling happy” and “being happy” (p. 477). Generally speaking, feeling happy has a short-term character and is often referred to as a state of euphoria: the presence of pleasure and the absence of pain. Being happy, on the contrary, is a general consideration of one’s life: is it considered a life in which all of the objectives form a harmonious and satisfying whole?

In this study, we will focus on the qualitative impact of microloans. It is theorized that non-enterprise expenditure directly contributes to an improved quality of life. We will take this debate one step further and measure the effect of microloans on the happiness of the borrowers, that is expected to improve as a result of an improved quality of life. Combining this subjective consideration – the happiness of the entrepreneur – with something objective as a microloan, will generate additional interesting insights that contribute to the debate on the effects of microloans. Regardless of context, measuring happiness and subjective well-being will deem

(9)

valid. As Lyubomirsky and Lepper (1999) argued, people can be happy despite personal obstacles, tragedy or lack of any great love or wealth, while others, surrounded by the comfort of a wealthy life, experience unhappiness. Centralized in this study is the following research question:

What is the effect of microloans on happiness of small business owners in Tanzania? According to the United Nations, sub-Saharan Africa is amongst the poorest regions of the world. In fact, globally, 767 million people live below the poverty line of coming around of $1.90 per day, of which the majority lives in Southern Asia and sub-Saharan Africa (United Nations, 2018). One of the countries in the latter part of the world is Tanzania. Even though the Tanzanian economy has shown strong improvement since the mid 1980’s, the most recently reported figures indicate that about half of its population lives below the international poverty line (The World Bank, 2011). This country on the eastern coast of Africa is rich in resources but is faced with a significant challenge in fighting poverty. Tanzania is therefore a viable environment to research the prerequisites and effects of microfinancing. This study contributes to the scarcity of quantitative data in the country of Tanzania, as was voiced by Basu and Yulek (2004) and the need to understand evidence from sub-Saharan Africa, as was voiced by Van Rooyen, Stewart and De Wet (2012).

This thesis is structured as follows. From an elaborate literature review, a theoretical framework will be constructed. This theoretical framework is the basis upon which hypotheses are drafted. Data collection, both qualitative and quantitative, in the third chapter will be elaborated upon. Then, in chapter four, these data will be analyzed, of which the results in chapter five will be discussed. The last chapter contains concluding remarks and an answer to the previously drafted research question.

(10)

2. Literature review

2.1 Defining and demarking microfinance

In the mid-1970’s, it was Muhamed Yunus that thought of a way to help the poor (Hulme, 2009). Instead of focusing on charity, he thought of a way to offer a sustainable solution to alleviate from poverty the hungry and deprived people he saw in rural Bangladesh. Having an economic background, he came up with the idea of borrowing small amounts of money to the poor so that they could advance their business and repay the money that was borrowed in a timely manner. At first, his experiment failed, but after drafting a rough model – which became widely known as the “Grameen model” – he started achieving small successes. The concept of microfinance was born. With many foreign aid agencies wanting to help the country of Bangladesh, amongst others through monetary support, the conditions for further developing his organization were excellent.

Microfinance relates to “financial services offered specifically to poor, low-income households and micro-enterprises” (Stewart et al., 2010, p. 11). These microenterprises generally operate in the informal economy because barriers to enter the formal economy are generally (too) high (Kolstad, Nygaard, & Fjeldstad, 2006). These financial services are not limited to making available monetary support, but also comprise investing (savings) and insurance (risk management). In other words, microfinance is the overarching definition that consists of three separate domains: microloans, -credits or -debt, microsavings and microinsurances. Although not universal, microfinance services may go hand in hand with training, education and technical assistance.

Microfinance jumped in the gap that existed between the institutionalized financial system and those microentrepreneurs having no access to monetary support at all (Basu & Yulek, 2004). This gap existed, because for commercial banks, underdeveloped economies formed no fertile soil to set up their activities. The poor generally lack knowledge to invest their borrowings to the most productive use, lack of documentation makes it difficult to assess creditworthiness and collateral is required to mitigate risks (Sengupta & Aubuchon, 2008). Physical collateral in underdeveloped economies is considered a luxury and lack of administration on property rights makes it difficult for landowners to meet strict collateral requirements. Therefore, social collateral replaced the function of physical collateral, which is one of the ‘rules’ of the Grameen model as well. Lenders, generally women, are organized in groups. A group would then

(11)

collectively become responsible for a debt. This kind of collateral benefits from reputational effects, concretized through so-called peer screening processes (Hulme, 2009). Women would be able to select who they would want to take out a loan with, based on their knowledge of group members and their relative stance within a community. Not being able to meet repayment obligations might result in reputational damage, social isolation, restricted access to other required business assets or physical force (Armendáriz de Aghion & Morduch, 2000).

2.1.1. Collective versus individual lending

Formalization of borrowing money took on informal sources of finance. Buckley (1997) from research in Kenya, Malawi and Ghana concluded that institutional finance replaced loans from friends and family, merchants and traders, as well as rotating savings and credit associations (ROSCAs). These informal associations mean that group community members would join together and pool their savings. One member of the group then used these savings, repaid them, only for the funds to move to another member of the group. The difference between group-lending from a microfinance institute and ROSCAs, is that money does not need to be saved first, but that a group can take out a loan immediately. Group-lending then still benefits from the peer screening process, monitoring other members of a group and so-called enforcement capacity (Wenner, 1995). Usage of the formal financial sector generally was limited to savings deposits. However, this also changed, as microfinance increased in popularity, for example on the African continent, as a way to store liquidity, rather than having to buy assets such as gold (Basu & Yulek, 2004).

In slightly developed environments or transition economies, physical collateral is increasingly being offered to take out microloans. This development goes hand in hand with a departure from the traditional group-lending to an increased focus on individual lending. Banerjee (2013) argues that it might have to do with the fact that joint liability can lead to “punishing potentially innocent borrowers” (p. 498). Armendáriz de Aghion and Morduch (2000) argue that group-lending for the wealthier entrepreneurs imposes certain limitations, as they are restricted to the maximum amount a group can or thinks it can borrow.

2.1.2 Borrowing to women

The Grameen model became a panacea for poverty reduction (Khandker, 2005). Although disputed by many (Buckley, 1997; Morduch, 2000; Stewart et al., 2010), poverty reduction was

(12)

in fact one of the pillars on which the Grameen bank was built. Another pillar related to the fight for gender equality, which in Bangladesh – “a predominantly Muslim country traditionally viewed as culturally conservative and male-dominated” (Morduch, 1999, p. 1571) – was not the standard.

Targeting women yielded favorable results, as loans taken out by women are more efficiently spent than loans taken out by men (Pitt & Khandker, 1998). So, borrowing to women empowers them, but also increases chances of the entire family benefiting from it (Kabeer, 2001). Goetz and Gupta (1996) at the end of the last millennium already found that reaching women was a success, but that now a “second-generation” problem announced itself: women actually being able to decide what to spend money acquired through microloans on, as well as increasing productivity, mobility, social status and access to markets.

The tool of microfinance has been developed as a result of populations in developed countries partially being excluded and in Lesser Developed Countries (LDCs) entirely being excluded from traditional financial service providers (Leone & Porretta, 2014). Through the years, the term ‘financial inclusion’ was coined to differentiate between those able to rely on financial services and those unable to (for example Dev, 2006). Since microfinance can play its part in any economy, it has achieved a global presence, with slightly different nuances per context, but generally following the same rules, namely: those of the Grameen model. Auwal (1996) found that the philosophy of the Grameen Bank enjoyed an increased interest from academia from all over the world. He researched the adaptability of the model in two different cultures: in the United States and in Malaysia. He found that the differences in organizing the model relate to “government policies, organizational skills, exogenous economic and political factors, historical experiences and acceptance of certain modes of operation” (p. 46).

2.1.3 Issues of microfinance

Making available monetary support in the broadest sense means providing very small loans to those in need. In this thesis, we focus only on formal lending activities, often initiated by (branches of) commercial or state-owned banks, post offices and now increasingly microfinance institutes (Basu & Yulek, 2004). At the same time, we exclude informal lending activities like village moneylenders and loansharks (Sengupta & Aubuchon, 2008). “Moneylenders are defined as those individuals who spend a significant part of their time lending money, usually for short period and sometimes unsecured by collateral.” (Buckley,

(13)

1997, p. 1084). Formal lending institutes generally tend to be more lenient and flexible to not stain their reputation and with it ensure financial sustainability (Banerjee, 2013).

This desire for sustainability can lead to microfinance institutes valuing economic goals over social goals. Lewis (2008) explains that this tendency can cause microfinance institutes to turn from replacing loansharks to microloan-sharking or, in other words, charging the generally already impoverished target audience exorbitant interest rates. These interest rates ought to be high as a result of high costs (e.g. locating borrowers, recording repayments etc.) and high uncertainties (Banerjee, 2013), but Lewis mentions annual interest rates of “more than 100 percent” with a “worldwide average of 31 percent” (p. 56). Instead of reducing poverty, it then amplifies it. This conflict between alleviating poverty, empowering women and microfinance institute sustainability is what has sparked the debate on the effects of microfinance.

Edward and Olsen (2006) untangle the triangle of poverty alleviation, women empowerment and microfinance sustainability. They conclude that sustainability is increasing in popularity, which has a negative effect on the social goals of alleviation and empowerment. In fact, although Brau and Woller (2004) argued that “women invest the loans in productive activities or in improving family welfare more often than men” (p. 11), women empowerment is losing track in comparison to poverty alleviation.

Placing this discussion in a larger context, similarities to the debate between institutionalists and so-called welfarists are notable (Brau & Woller, 2004). The former are convinced that microfinance institutes should be able to maintain on their own, diminishingly relying on the support of governments and beneficiaries. Nongovernmental organizations (NGOs) in this play an important part. Welfarists, however, see an important role to be played for beneficiaries, as without their support, the poorest of the poor are unreachable. Murdoch (1999) questioned the relevance of microfinance for seriously destitute households, while leaving open the possibility of relevance for non-destitute households that operate below poverty lines. This debate touches on the so-called outreach of microfinance institutes: to what extent do microloans reach those most in need? Or is there a trade-off between profitability of a microfinance institute and depth of outreach? Although also not undisputed, Murdoch (2000) in a later study found that “if financial self-sufficiency is desired, then the very poor will not be reached by MFI services” (Brau & Woller, 2004, p. 5).

(14)

2.1.4. Goals of microfinance (institutes)

The dynamics of a microfinance institute are multifold. On the one hand, we see that such institutes were established to help the poor, but on the other hand, they are the ones being helped. If a microfinance organization can rely on the financial support of beneficiaries, then the financial bottom-line becomes less important and the institute can focus more on the social goals related to microfinance.

Monetary support to MFIs deviates from investing in the traditional sense of the word: the interest rates are less important than are the effects of the money that was invested – “doing well by doing good” is an important mantra. With the bottom line becoming increasingly important, MFIs tend to focus more on the lenders that are commercially more interesting, rather than essentially helping the ones who, from a theoretical point of view, would need the funds the most. This is understandable from a sustainable point of view but is unfavorable from a social point of view. Therefore, in measuring the impact of a microfinance institute it is crucial to first establish the reason of existence of that institute. Is it to become profitable? Or is it to help those in need?

In their work, Battilana and Dorado (2010) elaborate on the coming into existence of hybrid commercial microfinance organizations. These organizations were the result of an explosive demand in nongovernmental organizations that offered loans to the poor. Therefore, these organizations started combining two logics: the logics of development and banking. This way, these microfinance organizations would be able to manage on their own, while still pursuing the social goals related to microfinance. The study entails a comparative study of two organizations in Bolivia, active in the early nineties. The research found that these hybrid organizations needed to create an organizational identity that found a balance between the logics they combine.

“The success of the microfinance sector has been impressive. Across a wide variety of models, reported loan repayment rates, even among the poorest clients, often exceed 95%.” (Pronyk, Hargreaves, & Morduch, 2007, p. 1925). This is a prime example of academia celebrating the effectiveness of microfinance. Brau and Woller (2004) even argued that with such high repayment rates (e.g. upper 90 percent in many cases), the subject of microfinance would allow for adoption in mainstream financial services. However, some scholars found that these high repayment rates are not a valid indicator of effectiveness. Most poignantly was that Bateman

(15)

and Chang (2009) in their research on repayment, found that repayment of existing loans occurred amongst others by attracting new loans. As such, credit programs denoting high repayment rates “indicates only that borrowers are able and willing to repay, it says virtually nothing about impact on enterprise operations” (Buckley, 1997, p. 1091), let alone about impact on other aspects.

We cite Buckley, because in 1997 he found that credit was barely used for non-enterprise purposes. However, more recent studies found that funds attracted through microfinance are not only invested in the business of a microentrepreneur (Stewart et al., 2010; Hulme, 2009), but are also employed as a means of making available education (for children) and improving general health, nourishment and housing. These qualitative aspects will in this study be collected under the banner of ‘personal wealth’. This dichotomy between the business and personal life is the first step in roughly drafting the theoretical model of this thesis. Illustration 1 allows us to build further and illustrate how the connection between microfinance and happiness is further concretized.

2.2 Microfinance in Africa

Financial inclusion in Africa remains an obstacle. Its value in development has been acknowledged widely, but access for small and medium enterprises (SMEs) remains low (Demirgüç-Kunt & Klapper, 2012). Mobile money is increasing in popularity, with almost the entire Tanzanian population being aware of its existence and about half of its population using such services (InterMedia, 2012). Financial inclusion is increasingly being achieved through a growing linkage between the banking system and the microfinance movement (Basu & Yulek, 2004). As a result, the gap between the formal and informal economy is narrowing and microlenders increasingly are able to rely on formal banking services.

In line with the developments in Bolivia, in the 1980s the African continent too enjoyed a mushrooming of government and NGO-sponsored microfinance programs. However, the focus of these programs was on women empowerment, as was found by Mayoux (1999). In the extension of swelling critiques on microfinance as a tool in poverty alleviation and business development, she researched the effects of microfinance on women empowerment on the African continent. Generally, microfinance in this context became the successor to funds attracted through lending vehicles like ROSCAs and SACCOs.

(16)

Differences between the African and Asian context, she found, were multiple fold. First, African women are generally more involved in production and market activities, which is why they are generally more inclined to use microfinance and related services for their own good. This is in line with the relatively small gender gap in Africa as compared to other regions, as argued by Demirgüç-Kunt and Klapper (2012). Second, a certain degree of separation between man and woman within the household leads to the woman retaining income for herself and the family. The downside to this, is that the man tends to withdraw his own income as well, further decreasing his responsibility within the household. She ends with emphasizing the focus on financial sustainability of the microfinance movement in Africa, which, as has been theorized before, can have a detrimental effect on the social goals of microfinance. Stewart et al. (2010) in extensive review work of microfinance research on the African continent are explicit: “We conclude that some people are made poorer, and not richer, by microfinance, particularly microcredit clients.” (p. 6).

(17)

2.3 Happiness

Microfinance can have a positive effect on the general quality of life. An improved quality of life, in turn, has a direct effect on happiness. But: what is happiness? And how can happiness in the relative short timeframe and relative little writing capacity be brought back to a comprehensive and understandable whole and as well be linked to a study on the subject of microfinance? To these questions, in this chapter we offer an answer.

The first horde in researching happiness is finding agreement on terminology. Is happiness the same as well-being? Where do we place life satisfaction relative to these two concepts? In research that aimed to examine definitions and experiences of happiness, Delle Fave et al. (2011) came to define happiness as “a condition of psychological balance and harmony” (p. 185). Prominently associated to this condition are family and social relations. From this, we can collect that happiness is an idea, a subjective consideration of the life of an individual. This subjectivity (implicitly relative to objectivity) is a first dichotomy in research on the subject of happiness.

Another dichotomy is the distinction by Shin and Johnson (1978) between “feeling happy” and “being happy” (p. 477). This distinction theoretically is best captured by Ryan and Deci (2001), who describe two philosophies within the field of research on happiness: hedonism and eudemonism. Albeit many controversies surrounded what exactly entails hedonism, it is generally accepted that it reflects well-being to consist of (short-term) feelings of pleasure or happiness. From this, one could conclude then that well-being is something of a domain that comprises both pleasure and happiness. Eudemonism, in turn, entails that well-being lies in “the actualization of human potentials” (p. 143). The pursuit of happiness goes further than just finding as many moments of pleasure as possible, but true happiness lies in the expression of virtue – that is, in doing what is worth doing. It is theorized that this process, the pursuit of doing what is worth doing, can have a detrimental effect on the well-being of an individual. This then means that eudaimonia cannot be equated to well-being.

Other prominent works focus on the effect of personality and characteristics on happiness: “What type of people are likely to be well or happy?” (Ryan & Deci, 2001, p. 149). Considering the big-five personality traits, it has consistently been found that extraversion and agreeableness were consistently positively associated with subjective well-being. Neuroticism, on the other hand, was consistently negatively associated with subjective well-being.

(18)

Another relevant consideration in assessing happiness are emotions: are people that experience positive emotions more than negative emotions considered to be happier? Yes, according to Diener et al. (1999). They argue that subjective well-being consists of the presence of positive mood, the absence of negative mood and life satisfaction. Together, these three components are often summarized as happiness.

Other related constructs are life satisfaction, psychological or subjective well-being and happiness. In a study frantically titled “What can economists learn from happiness research?” by Frey and Stutzer (2002), it is argued that this precise terminology will be used when empirical research is cited. However, otherwise these labels are used interchangeably. Diener (2000) for example says: “It is this approach to defining the good life that has come to be called ‘subjective well-being’ (SWB) and in colloquial terms is sometimes labeled ‘happiness’” (p. 34). Diener further argues that SWB is compartmentalizable and that similarities between life satisfaction and happiness are irrefutable.

Happiness comes from a variety of sources. At the risk of diving into a philosophical debate, in this study we will distinguish four well-known accounts of a happy life (Von Wright, 1963). The first relates to a form of materialism: having (as opposed to doing) certain things that give pleasure. The second relates to a state of mind, an equilibrium, in which needs and wants on the one hand and satisfaction on the other come together harmoniously. The third category is one that in the modern-day life is considered “doing what you love”. A fourth category can be found in relative status. Some academia argue that comparison with others and past experiences significantly affect happiness as well. This comparison, but the assumption that attracting microloans generally leads to an improved quality of life, is the basis for the first hypothesis to be tested in this study:

H1: Small business owners that use microloans are happier than those who don’t. At the basis of this first hypothesis to be tested stands the assumption that microloans are spent on the indicators previously attributed to personal wealth of the borrower. These indicators are non-economical and, thus, difficult to measure. However, borrowers have more funds at their disposal, meaning that an increased expenditure on personal wealth is a good possibility after having taken out a loan. In fact, Afrane (2002) found that, despite the general short-term character of microloans, business investments right after the fact of borrowing usually drops. From this, we gather that small business owners that (are able to) spend more on personal wealth

(19)

display an increased sense of well-being. Taking this one step further: small business owners that have been able to rely on microloans longer and to a larger extent display an even higher degree of happiness. This also touches on financial distress, which will be discussed later, as a first acquaintance with microloans might raise feelings of insecurity, which can have a negative effect on happiness. Microloan amount and microloan period then become variables of influence. From this, the following hypothesis and two sub-hypotheses have been drafted.

H2: Small business owners that use microloans spend more on items related to

personal wealth than those who don’t.

H2A: Small business owners that have used microloans longer (period) spend more on personal wealth than those who don’t.

H2B: Small business owners that have used more microloans (amount) spend more on personal wealth than those who don’t.

2.3.1. Financial distress

Theorizing further on the determinants of happiness, we learn that an individual’s economic situation can be of particular relevance (Prawitz, et al., 2006). Financial distress is the degree to which someone experiences money problems. “Norvilitis et al. (2003) demonstrated that perceived financial well-being is related to one’s overall psychological well-being.” (as cited in Archuleta, Dale, & Spann, 2013, p. 51). It is not necessarily availability or unavailability of economical means, but more the perception of one’s financial situation.

The concept of financial distress is generally attributed to firms and their performance (Opler & Titman, 1994). It is one of the domains within the field of corporate finance. Terms like debt, bankruptcy and restructuring are popular. Although these slightly touch onwhat we here set out to include, financial distress in this study relates to financial problems of the individual. The concept of financial or economic distress can be divided in two areas: objective and subjective aspects of employment and income (Benson et al., 2003). Objective conditions, like being unemployed or not being able to make ends meet, are valid indicators of economic distress. Subjective conditions, like fear of losing a job or worries about money, may also indicate economic distress. What we gather from this are two things: 1) economic and financial

(20)

distress will in this study be used interchangeably and 2) the subjective consideration of this distress is in line with earlier theorized subjective considerations of happiness.

In constructing a bridge between the concept of financial distress and microfinance, it can go one of two ways. Either having a microloan has a reductive effect on feelings of financial distress or it increases feelings of financial distress. As we discussed before, theoretically the making available of monetary support to destitute households in itself is a positive development. This way, households can increase their business investment or their general consumption, with it invest in their future and eventually lift out of poverty.

However, borrowing money to an entrepreneur at a 36% annual interest rate means that this entrepreneur in the next year has to increase its income by at least 36% to meet the interest demands of the loan. Moreover, since we found that microloans are not necessarily used to invest in a business, at least for the short term the borrower has to generate extra funds without investing in its capacity to generate extra funds. Therefore, the borrower is expected to experience an increased sense of financial distress. Additionally, as theorized by Schicks (2013), debt can have negative psychological and sociological consequences for the borrower. From this, we draft the last hypothesis to be tested in this study:

H3: Small business owners that use microloans have higher financial distress than those who don’t.

(21)

3. Methodology

In this chapter, we will elaborate on the methodology applied in this research. Respectively the research design and context of the research will be discussed. Then, steps taken to collect scientifically sound data will be explicated and, ultimately, the process of data analysis will be discussed.

3.1 Research design

Microfinance is the overarching definition that roughly consists of three separate subjects: microcredits, microsavings and microinsurances. Research on the latter two subjects lags behind (Banerjee, 2013), but the subject of microcredits no longer is in its infancy. In 2004, Brau and Woller already argued that research on microcredit had matured from infancy but had not reached the phase of maturity yet. Since no general consensus on (the effects of) microfinance has been reached, we conclude that research on the subject is in an intermediate stage. According to Edmondson and McManus (2007), intermediate theory research “frequently integrates qualitative and quantitative data to help establish the external and construct validity of new measures through triangulation” (p. 1165).

Happiness is a research domain as old as time, which has been researched in a wide array of different contexts. A great number of articles has been published on the subject and, even though the field is still evolving, a general consensus on the subject has been reached. Therefore, this research field can be considered to be in a mature phase. Bringing together these two relatively distant subjects, however, has not been done (frequently) before.

Therefore, in this study a hybrid research method has been applied. The combination of qualitative and quantitative data helps to elaborate on a phenomenon, to provide preliminary tests of relationships – between microfinance and happiness – and to promote insight and rigor (Edmondson & McManus, 2007).

3.1.1 Qualitative

Since this study explores the relation between microcredits and happiness in the unfamiliar context of Tanzania, Africa, an increased understanding of the context is desirable. This increased understanding of the context is centralized in 3.2 Research context and will be adjusted and expanded through interviews with local experts. These interviews will make up

(22)

the qualitative part of this research. To ensure relevance of these interviews, the experts will have to have at least three years of working experience in the field of microfinance or entrepreneurship or they should have an increased knowledge on the local context of the Mzumbe region in the Mvombero district.

3.1.2 Quantitative

To test the earlier drafted hypotheses, quantitative data will be collected through a pre-drafted questionnaire. Important in drafting this questionnaire, is that items used to measure the separate characteristics of the microloan, the development of personal wealth and happiness have proven their functionality as much as possible in previous studies to increase chances of success. The validity of the instruments, expressed in a Cronbach’s alpha, will be discussed in the results section.

Popular in research on the effects of microfinance is comparing two groups (for example: Pitt & Khandker, 1998). Karlan (2001) says that “such studies have risen in popularity because they are cheap, easy to implement, and often encouraged by donors” (p. 75). In comparing two groups, it is important that they are as similar to each other as possible. In this study we want to minimize external effects and only measure the effect of having a microloan on happiness. Therefore, in collecting data, we aim to find respondents that have a microloan and respondents that do not have a microloan but applied for one in the last year. That way, we know that the starting situation for both groups is similar and comparable. Should this not be the case, for someone not to have a microloan might be a reason because they have sufficient funds of themselves. The effect of not having a microloan on happiness might then be diluted and, thus, the groups cannot validly be compared to one another.

For the two sub-hypotheses that are based on the assumptions that microloan amount and microloan period have an effect on degrees of happiness, we can build a moderator model. It is theorized that these two variables have a strengthening effect on this relationship.

(23)

3.2 Research context

The country of Tanzania has shown strong improvements in its battle against poverty. In 2007, 59.9% of its population lived below the poverty line. In 2011, this percentage had decreased to 49.1%. Although improved, the majority of the Tanzanian population still has to come around of less than $2,- per day, about 12 million people still live in poverty and about 4 million people still live in extreme poverty (The World Bank, 2015).

3.2.1. Geographic and demographic

The United Republic of Tanzania is a country in east-Africa. The country as we now know it came to be as a congregation of Tanganyika and Zanzibar, which happened on the 26th_{of April,} 1964. Tanzania is bordered by Kenya and Uganda to the north, Rwanda, Burundi and Congo to the west and Zambia, Malawi and Mozambique to the south.

The country in 2016 had 55 million inhabitants (The World Bank, 2018). The country’s capital city is Dodoma. Its region counts 2.3 million inhabitants. In the region of Dar Es Salam, the major commercial city of the country, 5.5 million people live. The Morogoro region, the focal context of this study, counts 2.4 million inhabitants (2012: 2.2 million). The Tanzanian Shilling (TZS) is the national currency. The national language is Swahili, but English is also widely used.

3.2.2. Economy

The Tanzanian Growth Domestic Product has denoted impressive growth rates of around 7% the past years. Its GDP in 2016 was 47.34 billion USD and the Growth National Income was 900 USD per capita. For comparative purposes: the Dutch GDP in 2016 was 777.228 USD billion and the GNI was 46.610 USD per capita (The World Bank, 2016).

It has been widely acknowledged that the Tanzanian economy shows great potential and prospects for attaining higher growth and development levels. This is due to its many natural resources – recent years have been characterized by an increased export in gold – and its relative stable political situation. The agriculture in the country plays an important role, as it employs about 75% of its population (Ministry of Industries and Trade, 2015). The most important crops in terms of export are coffee (2016: 224.2 billion TZS), cashew nuts (2016: 756.9 billion TZS) and tobacco (2016: 783.8 billion TZS), of which the latter is strongly represented in the

(24)

Morogoro region. Other relevant economic sectors are tourism, energy, investment, mining and trade.

3.2.3. Politics

As indicated, the current political situation in Tanzania is relatively stable. In 2015, the current president, Dr. John Pombe Joseph Magufuli, has been selected as the fifth president, after elections that denoted a high turnout as a result of the Ukawa coalition that challenged the reigning party of CCM (BBC News, 2015). Previous opposition was weak and divided, which is why Tanzanians considered elections to be of less importance. The government is committed to good governance, rule of law and respect for human rights. Its three biggest challenges include ensuring macroeconomic stability, implementing its development-oriented budget and enabling the private sector to play a more significant role in the development of Tanzania (The World Bank, 2018).

3.2.4. Entrepreneurship

This focus on the private sector is also acknowledged by The World Bank. With its sister organization, the International Finance Corporation (IFC), it has developed a strategy to proactively develop infrastructure projects, focusing on power, oil and transportation. Since the majority of the Tanzanian population is employed in the agricultural sector, commercialization of this sector is a priority. IFC emphasizes strengthening financial markets, making available financial services to micro, small and medium enterprises. In this, it supports the initiative of microfinance as well.

Microfinance on the African continent dates back to the 1950s. Back then, it was not microfinance as we know it today, but it showed similarities to the informal lending vehicles of SACCOs, ROSCAs and ViCoBas (Village Community Banks). However, the microfinance movement remains relatively small and concentrated in a number of countries. It is undisputedly increasing in popularity, with an emphasis on regulations and infrastructure to accommodate the movement further (Stewart et al., 2010).

3.2.5. Morogoro

The Morogoro region in 2012 was populated by 501,794 households. Out of these households, 70% were in rural areas and 30% in urban areas. The average rural household size consisted of

(25)

4,4 persons (urban: 4,1 persons). main occupation in the Morogoro region, as comparable to the entire country, is farming (National Bureau of Statistics, 2016).

One district within the Morogoro region is the Mvomero district, which houses 14,1% of the total inhabitants of the region. The district in 2012 counted 312.109 inhabitants. Hereof, 276.447 and 35.662 people lived in rural and urban areas respectively. The annual growth rate of the population is 1,9%. In terms of literacy rate, we see a significant difference between rural and urban areas as well. In 2016, the literacy rate of Tanzanians in the urban Morogoro region aged five years and older was 86.4%. In rural Morogoro, this percentage was 65.1%. Narrowing the gap that exist between rural and urban areas in Tanzania make up one of the focal points of the aid strategy of The World Bank (2018).

3.2.6. Mzumbe university

The villages around the Mzumbe university count around 1600 households and 6000 inhabitants (Ahmed & Mgotoh, personal communication, April 19, 2018). There are around 800 farmers and 200 food sellers. Not all of the inhabitants were born in the area but migrated there because the presence of the Mzumbe university offers economically interesting opportunities.

The university, other than obvious educational and developmental advantages, also has economical relevance. As a result of the student population, which enjoys an increased degree of welfare, it is for entrepreneurs interesting to be situated adjacent to the university campus (Musabila, personal communication, April 19, 2018). That is also why markets have been established on the northern and on the southern side of the campus. These markets, complimentary to various shops and a cafeteria on campus, offer basic needs like food and beverages, offered by (improvised) restaurants, bars and shops, as well as copy shops, hairdressers, clothing and tailor services and even leisure activities, such as public football viewings and billiard. As will be elaborated on in the next chapter, most of these entrepreneurs operate in the informal economy, which has been widely acknowledged to be of relevant importance to the Tanzanian economy (Adams, da Silva, & Razmara, 2013).

3.2.7. SUBO

During this study, we closely collaborated with a local microfinance institute: SUBO Financial Consultant and General Supplies LTD. This institute has been established in September, 2017.

(26)

Although widely represented in Morogoro Town, microfinance in the vicinity of Mzumbe was not yet abundantly available (Musabila, personal communication, April 19). The organizations that offered microloans are generally branches of larger organizations like FINCA and BRAC. SUBO is, as of yet, the only institute that has a local office.

About 50% of the villagers, about 70% of the agricultural sector and almost all food sellers depend on microloans (Ahmed & Mgotoh, personal communication, April 19, 2018). SUBO jumped in the gap that existed. This microfinance institute at the time of research had 272 customers and can thus be considered a small organization. Its purpose is to become financially sustainable, which is why it is currently in the process of building a trusted customer base that has proven its ability to repay debts and interest charges in time: “We don’t give to people that can’t repay” (Deogratius, personal communication, April 17, 2018). As the organization is still in development, business processes are characterized by informality.

Interest charges amount to 20% per month, however, the amount to be repaid over a three-month 200.000 TZS loan is 270.000 TZS. The first three-month, the borrower is expected to repay half of the loan, or 100.000 TZS and 20% over the initial borrowing amount, or 40.000 TZS. Then, outstanding debt amounts to 100.000 TZS. The second month the borrower is expected to repay half again, or 50.000 TZS and 20% over the outstanding debt, or 20.000 TZS. After the third month, the total loan amount of 50.000 TZS and 20% over this amount, or 10.000 TZS has to be repaid. The total amount to be repaid on a 200.000 TZS loan then amounts to 270.000 TZS. This amount is evenly divided over the three months, meaning that the borrower per month has to repay 90.000 TZS. This equals about 34,17 EUR.

Most of the customers of the institute operate in the food industry, which is also why this research has opted to focus on this industry. Subsequently, the majority of the customer base is female, as selling food is generally considered a woman’s job. These women operate in the informal economy, registration of businesses is low. Repayment has been organized to occur on the first day of each month. If a repayment deadline is missed, the borrower discredits themselves. The organization only gives out new loans if sufficient potential borrowers have applied for one.

(27)

3.3 Quality of research 3.3.1. Questionnaire

In close collaboration with Professor Masurel, a final English version of the questionnaire was delivered. This version was translated by a local bilingual to Swahili. To ensure validity, after translating from Swahili to English, we then asked another bilingual to translate the Swahili version back to English. This process is in line with the approach applied by Brislin (1970), who argued that double-translation ensures comparability and equivalence in meaning. Although this process minimizes the risk of getting lost in translation, Buckley (1997) recognizes “that not all African languages contain discreet and accurate translations for terms such as credit, debt, interest rates, repayment cycles, collateral and so forth, so meanings may change or at least differ in translation.” (Buckley, 1997, p. 1088). The questionnaire itself and an overview of the sources from where items were collected can be found in Appendix i. English questionnaire and Appendix iii. Sources questionnaire respectively.

3.3.1.1. Personal information

In personal information, we inquire on gender, which has ever played an important role in (research on) microfinance, age, education, marital status and household size. This personal information relates to the “resources” that were distinguished by Shin and Johnson (1978) as being one of the factors that have an influence on happiness. “The resource cluster included achieved resources (sex, race and age) and inter-personal relationships (marital status, the number of children and the number of people living together).” (p. 480).

3.3.1.2. Personal wealth

Although in most studies labeled differently, for example Khandker (2005) makes mention of “investment in human capital (such as schooling)” and “reproductive health issues” (p. 266), in this study we will also include items related to “needs and wants”, as labeled by Shin and Johnson (1978). While at first this may seem vague, it actually shows overlap to an elaborate consideration of studies conducted on measuring the quality of life in South-Africa (Higgs, 2007). It is argued that the assessment of basic needs and wants, like health, education, housing, standard of living, leisure time and consideration of the community and local government, says something about the general state of happiness of a human being. This might especially be relevant in the African context, where such basic needs and wants are not the standard.

(28)

Additionally, as a household starts to enjoy increased welfare, consumption increases as well (Mahjabeen, 2008). Therefore, after acquiring funds through a microloan, we expect to see an increased expenditure on personal wealth. To capture a wide array of these areas, we will include five items that measure expenditure on housing, education, healthcare, number of meals consumed in a day and general standard of living.

3.3.1.3. Happiness

By measuring happiness of two different groups, we expect to distinguish intangible effects of microcredits. Regardless of how the money is spent, there might already be a difference in happiness from a comparative point of view (“I received microcredits and he did not”). What we set out to measure is how the attracted loan is spent and whether this has a significant effect on happiness.

In this study, we have adopted two items from a measure of subjective happiness as developed by Lyubomirsky and Lepper (1999). These two items have been adopted, because they have shown high correlations with other happiness measures. They have been adjusted to fit to the questionnaire drafted for this study, which means that the answering possibilities have been adjusted from a 7-point Likert scale to a 5-point Likert scale. The second item, “compared to most of my peers, I consider myself” also relates to the relative life comparison as argued by Shin and Johnsons (1978).

Other items of the in total seven items that measure happiness are adopted from a popular study by Diener et al. (1985). Diener (2000) argued that combining measurements of happiness avoids respondents from answering the questions in a socially desirable way. The other adopted items have shown high internal consistency and high temporal reliability. Here too, the answering possibilities have been adjusted from a 7-point Likert scale to a 5-point Likert scale to be compatible with this study. Although inquiring on the life conditions of someone living in poverty might seem misplaced, we were by Dr. Sol and local experts reassured that this would not be an issue.

3.3.1.4. Financial distress

In the questionnaire we included two items to measure financial distress of a respondent. This is in line with suggestions by Diener (2000), who argued that “researchers need to use measures of both pleasant and unpleasant affect, because both are major components of SWB” (p. 35).

(29)

This variable can be considered explanatory. Should no significant relation be found between expenditure on personal wealth and degrees of happiness of the respondent, then in financial distress we might find an explanation. Prawitz et al. (2006) argued that economic distress was a good predictor of lower levels of well-being. Financial distress can directly be linked to overindebtedness (see also 2.2.1. Financial distress).

The items have been taken from a study conducted by Mills et al. (1992), because of their applicability and because the authors reported a high alpha of .81. They included four items, of which we adopted two: “I often experience money problems” and “I spend a lot of time worrying about financial matters”. The items were made compatible to the questionnaire used in this study, which means that the 4-point Likert scale was adjusted to a 5-point Likert scale, ranging from “strongly disagree” to “strongly agree”.

3.3.2. Sensitivity

To achieve sensitivity, a narrowly defined industry was selected (Davidsson, 2004). Data collection focused on entrepreneurs active in the food sector and, more specifically, fruit vendors and so-called mama lishas. The latter group are best defined by entrepreneurs, generally women, hence mama, that prepare and sell food. These entrepreneurs operate in the informal economy. We focused on this industry as a form of convenience sampling.

Another issue of sensitivity is the period about which information is collected. Many scholars before us have stumbled upon the issue that comparing two groups over a certain period of time deems difficult, because “Over the last year” does not mean the same as “Over the last 12 months”. Additionally, inquiring on the last year of business operation requires an explication of which periods we are comparing. Is it compared to the year before? Two years before? Since we want to know more about the actual effects of microloans, we have phrased the questionnaire so that a respondent compares the daily life of twelve months ago to the current daily life. 3.3.3. Data collection

To ensure that data collected in this study is valid, several measures were taken. Prior to collecting data, we conducted a pilot study on a market in Morogoro town (N = 9). This market was purposely selected to not dilute any data collected in the Mzumbe area. The findings were used to refine the questionnaire. One of the findings of this pilot was that revenues in the past three, six and twelve months were hard to distinguish. Instead, we rephrased our question to the

(30)

average revenue per day. If that was difficult to remember, then the number of dishes sold, in combination with the average price, was requested. For fruit vendors, we learned, average revenues were easier to recollect. Another important take-away was related to the microloan about which we inquired. We first asked the respondent to fill out questions on microloan specifics (questions #41 through #52) about their most recent loan. However, to best measure the effects of the microloan, we adjusted this to the oldest loan within the past twelve months. The researchers in the process of data collection enjoyed help of three local research assistants. These assistants were clearly instructed on how to assist respondents and guide them through the survey. The survey itself contained disclaimers to ensure honesty and integrity. The survey was in Swahili entirely. Any irregularities in the process of data collection or any questions could immediately be directed to the researchers, as they remained present at all times. Additionally, entrepreneurs were as much as possible visited on site, so at the place where they conducted business, to maximize their comfort.

In the process of collecting at least one hundred respondents to our questionnaire, at least fifty of which acquired a microloan, the target group, and at least fifty that did not acquire a microloan, the control group, we applied a form of convenience sampling. In close collaboration with the microfinance institute, we were able to target enough respondents that used a microloan. Crucial in obtaining data on two comparable groups, was adhering to stringent advice of Dr. Sol. He indicated that only collecting data on a microloan and a non-microloan group might result in comparing two too different groups of respondents. Instead, a group that got a loan and a group that did not get a loan, but applied for one, would allow for the best comparison.

In surveying the respondents, we emphasized that cooperating did not affect their position relative to the microfinance institute. Additionally, no compensation was made available. However, in the process of collecting data, we found that asking the customers to come to the office of the microfinance institute would not generate sufficient responses. Instead, and to accommodate comfort to respondents, we visited them on site. In terms of sampling, selecting all customers and non-customers from their customer base or selecting all food vendors from one village the internal validity of results was best secured. A third strategy comprised randomly selecting food vendors from nearby towns with a mixture of microloan specifications. In that case, external validity would be highest. However, as we will later see, generalizability of results is difficult.

(31)

3.3.4. Data analysis

We collected both qualitative and quantitative data. These data will in the next chapter be processed and analyzed. The interviews will briefly be summarized and the key take-aways will be included in a separate chapter. Summaries and notes can be found in the appendices. The quantitative data will be analyzed using SPSS. First, to ensure comparability of the two groups, demographics and business characteristics will be discussed. Then, the specific microloan characteristics will be analyzed. This will function as a stepping stone to t-tests to distinguish any statistically relevant differences between the treatment and the control group. Additionally, we will conduct a hierarchical regression analysis on the total population, as well as a simple regression analysis on only the treatment group. Within this simple regression analysis, a moderator model using the loan period and loan amount as moderator variables will be constructed.

(32)

4. Results

In this chapter, we will report on the results of this study. First, some general information about these data will be given. Second, since we collected both qualitative and quantitative data respectively, these data as such will be processed and analyzed. Finally, interpretation of these data will function as the foundation to the discussion in the next chapter.

4.1 General information

In total we conducted 13 interviews (N = 13), of which six in the Netherlands and seven in Tanzania, and we collected 118 responses to our questionnaire (N = 118). Four survey candidates indicated not being willing to cooperate. Of this total number of respondents, one respondent indicated having had a loan in the last year, but he or she did not fill out the remainder of the survey. Also, one respondent indicated having had a microloan of 5.000.000 TZS for a period of twelve months. With an average loan amount of around 350.000 TZS and generally lending period of only three months, this response is considered an outlier and will subsequently be excluded from any correlation and regression analyses.

4.2 Qualitative

The data collected from interviews have been used throughout this report. The most relevant takeaways will here be reported on. The interviews have been recorded and transcribed. A summary of each of the interviews, as well as an overview of whom we interviewed, can be found in Appendix iv. Qualitative data collection. References will only include last names of the interviewees.

We conducted a total of 14 interviews, of which three revolved around refining our questionnaire. Of these three interviews, only one has been summarized. The adjustments taken from the other two refinement-interviews have directly been implemented in the questionnaire. These adjustments have been included alongside the sources from which the different items have been adopted. See also Appendix iii. Sources questionnaire.

The first few interviews revolved around getting acquainted with the local Tanzanian context. Both Sluiter and Eijdenberg have experience with the country of Tanzania and Adhiambo is from Kenya. They informed us on practicalities prior to our travel, but also on how to approach

(33)

the process of data collection (e.g. use research assistants and ask local experts to translate our questionnaire).

This was also recommended by Masurel, who also indicated that inquiring on profits might be a sensitive issue in the African context. However, in interviews that followed with local experts and Ribas, we learned that inquiring on profits would not be a problem in terms of sensitivity. The latter even expected that distinguishing between information on the past three, six and twelve months would not be a problem. Ribas, but Voorrips and Rosenbrand, both from Triple Jump, as well, indicated that microloans occasionally replace funds borrowed from friends and family. Triple Jump is an organization that supports, advises and guides the development of promising financial intermediary institutions in developing countries. Voorrips and Rosenbrand indicated that the effects of microfinance are very context specific. Adhiambo too indicated that repayment of loans sometimes happens by depending on friends and family.

Adhiambo also indicated that in the contract a borrower usually signs prior to retrieving a microloan, it is stated that the funds are for enterprise purposes. However, it is not unusual that the microloan is spent on other aspects like housing, education or healthcare as a result of the entrepreneurs operating in the informal economy as well. This is where we stumble upon a discrepancy that we gathered from out interview with Deogratius as well, as he also indicated that the microloan has to be spent on the business, but that it does not always happen as such. Moreover, Deogratius helped us in the process of data collection which, as we will elaborate on later, could have led to socially desirable answers (ergo: microloans are for business investments, but we included questions on other expenditure areas as well).

The connotation of the interview with Ribas was generally negative in terms of effects of microfinance. Measuring its effects is difficult and quickly results in a situation in which two entirely different (groups of) entrepreneurs are compared to one another in terms of size or productivity. “I wouldn’t be too optimistic about results”, he indicated and explained by how microfinance really slows down innovation. If money is readily available, then failure is less likely. The first step to innovation (and a key component of entrepreneurship) is failure. Moreover, if an entrepreneur has the ability to invest in his business and requires employees to take care of the growth the business experiences as a result thereof, then it is tempting to rely on family members. As a result, children might stop going to school. If then the business improves, from a microeconomic perspective it has a positive effect, but from a macroeconomic perspective, it has not.

The effect of microfinance on happiness of small business owners in rural Tanzania : the happy man is not he who seems thus to others, but who seems thus to himself