Social security systems in Tanzania 
MAGH1MBI3 
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
A.S.T.MCHOMVU1, FELICIAN S.K.TUNGARAZA2 AND SAM 
ABSTRACT 
The paper starts by examining the concept of social security in 
Tanzania, showing that there are three key issues in social 
security which have not been adequately addressed by existing 
social security schemes and need immediate attention. The paper 
then examines the nature and forms of social security in Tanzania 
in a historical perspective. It shows that non-conventional social 
security instruments have failed to promote equitable economic 
growth and have been heavily biased in favour of the well-off at 
the expense of the poor. The effectiveness of traditional and 
informal social security systems has been impaired by changes 
that have taken place since the colonial period. Formal social 
security schemes are riddled with problems. The development of 
formal social security has been gradual and the introduction of 
structural adjustment programmes has led to the decline of 
formal security schemes. In terms of coverage, formal social 
security schemes cover only 6% of the population and focus on 
only a few risks. 
Phase I: Overview 
Introduction 
IN THE PAST TWO DECADES, research on social security in Tanzania has 
bifurcated between the study of modern and traditional and informal 
1 Tutor, National Social Welfare Training Institute, Dar es Salaam 
2 Senior Lecturer and Head of Department of Sociology, University of 
Dar es Salaam 
3 Associate Professor, Department of Sociology, University of Dar es 
Salaam 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 11
social security schemes. The focus has been mainly on formal social 
security and particularly on the protection of workers against contingencies 
associated with temporary or permanent loss of income. Formal 
social security schemes cover only 5% of the labour force and 6% of 
the population. Most of those covered are working in the urban formal 
sector and these schemes provide few benefits (Bossert 198-7, 
Tungaraza 1988, Wangwe and Tibandebage 1999). 
Since the main aim of social security is to help individuals maintain 
a reasonable standard of living when faced with social and economic 
contingencies, there has been an increasing recognition of the need 
to understand the adequacy of benefits. Studies by Mangangila (1976), 
Mataba (1983) and Mlyansi (1991) show that the amount of benefit 
provided by various formal security schemes in Tanzania is insufficient 
to avert poverty and they have thus failed to provide adequate social 
protection to members in distress. Moreover, these schemes are faced 
with a host of other problems, such as the small rates of interest 
awarded annually to members, the evasion of payment of premiums 
by employers and a lack of transparency, the delaying of benefits and 
disparities in risks covered. 
The literature on traditional and informal social security systems 
has focused on the organizational and administrative structures and 
social relations among members of the family, the kinship, neighbourhood 
and community groups and types of transfers going on within these 
systems (Bossert 1987, Bakari 1988, Tungaraza 1993, 1994 and 
2000a). These systems operate in both urban and rural areas. A number 
of studies on informal social security systems have shown that they 
have failed to promote equitable economic growth and that these 
programmes are biased in favour of the rich. Furthermore, they show 
that poverty alleviation programmes have failed to identify the real poor 
(Sender and Smith 1990). The lack of accountability and inadequate 
resources allocation of these programmes has also been shown. 
The purpose of this paper is threefold. Firstly, to examine the concept 
of social security in the Tanzanian context. Secondly, to identify social 
protection issues and thirdly, to assess the strengths, weaknesses, 
12 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
opportunities and threats of social security systems so as to arrive at 
viable suggestions for enhancing their effectiveness and efficiency. 
The concept of social security 
Much of the literature on social security in Tanzania has defined social 
security as the protection provided by society to its members through 
public measures against the economic and social distress which 
otherwise would be caused by the stoppage or reduction of earnings 
arising from contingencies. In most cases the contingencies are 
associated with either the temporary or the permanent loss of income 
(Haule 1994, Omari 1994, Mallya 1994). Examples of contingencies 
which could cause economic and social distresses include sickness, 
employment injury and occupational disease, maternity, old age, 
invalidity, death and unemployment. Such a definition does not 
adequately fit the socioeconomic context of developing countries such 
as Tanzania. It overemphasizes the role of the public sector and the 
related net costs and expenditures while overlooking social security 
as an investment in human capital. It also provides limited guidance to 
effective poverty alleviation. Most people in Tanzania are faced with 
insecurity caused by chronic or structural poverty. This insecurity 
arises mainly from insufficient economic development. But the 
removal of chronic social insecurity faced by the poor does not remove 
the social insecurity associated with risks emanating from conventional 
contingencies such as loss of employment, disability, old age and death. 
The fundamental social security needs of the poor in Tanzania are 
the result of chronic or structural poverty and only secondarily of 
conventional social insecurity. An attempt to define social security 
should therefore take into account both the conventional definition as 
well as the basic needs definition of social security which portrays the 
situation in Tanzania. Holzmann and Jergensen (2000) offer such a 
definition. According to them, social security consists of "public 
interventions to assist individuals, households and communities better 
manage risk and to provide support to the critically poor". Underlying 
this definition is the idea that social security is both a springboard and 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 13
a safety net for the poor. It regards social security as an investment in 
human capital and focuses more on the causes than on the symptoms 
of poverty. Finally, this definition considers individuals, households 
and communities as vulnerable to multiple risks from different sources, 
both natural and man-made. 
Based on the above definition, some social security issues can be 
identified in Tanzania. These include poverty, equity and the right to 
social security. Each of these key issues will be discussed in turn. 
Social security issues in Tanzania 
Poverty 
Poverty in Tanzania remains pervasive. Studies conducted in the 1990s 
based on surveys using random sampling techniques indicate that 
mainland Tanzania ranked as the second poorest country in the world 
with a GNP per capita of US$100. Other studies based on household 
surveys estimated the per capita expenditure at US$281, thus indicating 
an underestimation of the national income and expenditure 
(Bagachwa 1994, and Tinios et al., 1993). Whichever is the correct 
figure, most Tanzanians live in poverty, though this is more widespread 
in rural than in urban areas (Jazairy et al., 1992; Tinios et al., 1993). 
Common to all these studies is the attempt to establish the nature 
and extent of poverty in all areas of Tanzania. These use income as an 
indicator of poverty, a methodology that has been criticized elsewhere 
(e.g. Republic of Tanzania 1999). As an indicator, income reveals little 
of matters of empowerment and participation, literacy levels, the health 
status and food security of individuals and tradition and norms. It is 
because of the unsatisfactory nature of this indicator that the United 
Republic of Tanzania in 1999 published a book on poverty and welfare 
indicators to assist researchers. 
According to this report, poverty is a state of deprivation that 
prohibits a decent human life. Poverty results from many mutually 
reinforcing factors such as a lack of productive resources to generate 
material wealth, illiteracy, the prevalence of diseases, discriminatory 
socioeconomic and political systems, natural calamities such as 
drought, floods and HIV and man-made calamities such as wars. The 
14 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
above notwithstanding, we do not know who are the real poor in Tanzania. 
Moreover, despite many studies on poverty alleviation, there is 
no agreed poverty line against which to gauge various efforts to 
eradicate poverty. Sender and Smith (1990) show that it is a myth to 
perceive rural people as a homogeneous group characterized by 
egalitarian forces. We need to understand the heterogeneity of the poor 
people in Tanzania. 
Despite the literature on poverty in Tanzania, most studies do not 
link poverty and social security policies and programmes. Perhaps the 
only exception is the study by the ILO(1982). According to this, the 
basic needs of Tanzanians were in danger of being eroded by the marketoriented 
policies advocated by the World Bank and the International 
Monetary Fund. It advised Tanzania to pursue economic growth with 
redistribution as well as to provide public welfare services so as to 
ensure the satisfaction of basic human needs. Social security was not, 
however, perceived as one of the strategies needed for poverty 
alleviation. 
Equity 
Equity denotes fairness or social justice. It implies that people's needs, 
rather than social privileges, should guide the distribution of opportunities 
for well-being. Equity requires reducing unfair disparities as 
well as meeting acceptable standards for everyone. Pursuing equity in 
social security means trying to reduce unfair and unnecessary gaps in 
social security, while working efficiently to achieve the greatest 
improvements for all. In policy terms, therefore, we have to aim for 
social security policies that are redistributive. This implies, firstly, the 
protection of the vulnerable by systematically reducing inequities in 
the distribution of incomes and resources through a guarantee of basic 
income and the conditions for living in dignity. Secondly, it implies 
providing the conditions for autonomy, so everyone can participate fully 
in their society. Thirdly, it implies enforcing, protecting and legitimating 
social and political citizenship. 
Studies on equity issues in Tanzania include the works by Awiti 
(1973), Putterman (1986), Coulson (1982), Ndulu (1982), Green 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 15
(1977), Bank of Tanzania (1981) and Tungaraza (1988). When examining 
the distribution of land in Ismani, Iringa Region Awiti (1973) concluded 
that rich peasants, representing about 9% of households, held approximately 
53% of the total land under cultivation. A similar study by 
Putterman (1986) in some selected villages in Arusha, Morogoro, 
Mbeya, Iringa and Ruvuma regions reaffirmed the existence of inequality 
in the distribution of landholding. 
Inequality in the distribution of resources is also found at the level 
of social infrastructure, such as educational institutions and health 
facilities. Coulson (1982), Ndulu (1982) and Tungaraza (1995) 
indicate that regions such as Dar es Salaam, Kilimanjaro, West Lake, 
Ruvuma and Tanga have more schools and health facilities per inhabitant 
than regions such as Shinyanga, Dodoma, Kigoma, Tabora and Mwanza. 
Social inequalities also exist within the regions. In his study of the 
social sector in Shinyanga, for example, Tungaraza (1998) found that 
health facilities, educational institutions and water supplies were 
unevenly distributed between the districts. The way these resources 
are distributed corresponds, to a great extent, to the differential 
regional distribution of incomes. Regions with comparatively fertile 
soils, a favourable climate and a relatively reliable rainfall, such as 
Kilimanjaro, Arusha, Tanga and West Lake, have also higher average 
incomes than other regions. Tungaraza (1988) observes that income 
inequalities found at the regional level are also found at individual and 
household levels in all areas of Tanzania. Differences in the quality of 
arable land, the availability of social infrastructure and educational 
attainment have led to the concentration of economic activities in 
certain regions. The observed income inequalities at individual and 
household levels are a result of many factors. Among them include the 
different economic development of individual regions, the low 
productivity of agriculture, the different levels of educational achievement 
and the existence of differences in salaries paid by employers. 
Most of the above studies were conducted in the 1970s and 1980s 
and therefore cannot give us a true picture of the inequities in the 
distribution of social security resources and incomes in contemporary 
Tanzania. Furthermore, the above studies are silent on the extent of 
16 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
redistribution of social security resources across generations and 
within a generation. Such an understanding is important so as to redress 
the unfair and unnecessary social gaps in social security in Tanzania. 
The right to social security 
There is only limited information on the right to social security in 
Tanzania. The only study on this aspect is the work by Tungaraza and 
Mapunda (2000). They point out that social protection is a human rights 
issue. According to Article 22 of the Universal Declaration of Human 
Rights of the 10th December 1948 everyone, as a member of society, 
has the right to social security. The State therefore is obliged to ensure 
a minimum standard of'material welfare to all its citizens on a basis 
wide enough to cover all the main contingencies of life. This includes 
general basic social support for all citizens, regardless of contribution 
or employment. In Tanzania, the right to social security for all citizens 
is stipulated in the Constitution of the United Republic of Tanzania of 
1977, as amended in 1984, 1995 and 1998. Some of these rights 
include the right to social welfare at times of old age, sickness or 
disability and in other cases of incapacity. Tungaraza and Mapunda 
(2000) argue that the reason this right is not conferred upon citizens 
is due to a shortage of financial resources. In addition, however, a lack 
of accountability, a lack of prioritization and deficiencies in the 
enforcement of these rights have denied people their rights to social 
protection. This right has not been realized because of a lack of 
awareness on the part of the members of society. 
The second right to social security is the right to live and to 
protection of life by society. Despite the existence of this right, many 
people have died from sickness, hunger and floods. Moreover, many 
people still live in sub-standard housing, while many households do 
not meet their members' basic needs because of economic hardships. 
Thirdly, Tungaraza and Mapunda (2000) show that, although contributors 
and beneficiaries of various social security schemes in Tanzania 
have the right to participate in matters pertaining to social protection 
either directly or through their representatives who are freely elected 
by them, in practice these members do not directly participate in board 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 17
meetings nor are they represented by members who are freely elected 
by contributors of the funds. Contributors and beneficiaries of various 
social security schemes do not have an opportunity to influence the 
decision-making process and to monitor the administration of social 
security schemes. 
The right to social protection conferred under the Constitution of 
the United Republic of Tanzania of 1977, as amended in 1984, 1995 
and 1998, should also confer the right to benefits. Tungaraza and 
Mapunda (2000) argue that the scale of benefits payable under the 
different schemes should be known to members. Benefit formulae and 
conditions for payment should therefore be clearly spelt out in the laws 
establishing the respective schemes. Protecting the rights of the 
beneficiaries is the main responsibility of any social security scheme 
and benefits should be paid promptly. Mechanisms for appeal should 
be established in law so that aggrieved contributors and beneficiaries 
can pursue their rights. 
The nature and forms of non-formal and formal social security 
Social security systems in Tanzania fall into two main groups, namely, 
non-formal and formal social security. Formal social security systems 
refers to conventional social security programmes such as pension 
schemes, health insurance and workmen's compensation schemes. 
Non-formal social security systems include macro-economic policies 
and programmes to promote equitable economic growth and employment, 
investment in and the development of social infrastructure and 
specialized programmes for alleviating poverty and traditional social 
security systems and practices. 
Non-formal social security arrangements 
Non-formal social security can be divided into two: non-conventional 
social security instruments (such as public interventions not geared 
directly at social protection) and traditional forms of social security. 
Non-conventional social security instruments 
Non-conventional public interventions include macro-economic policies 
and programmes to promote equitable economic growth and 
18 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
employment, investments in and the development of social infrastructure 
and programmes such as health, education and training, housing 
and water supply. It includes specialized programmes for alleviating 
poverty such as credit schemes, integrated rural development, rural 
works, co-operatives, community-based organizations and non-governmental 
organizations. 
Various studies (see Mtatifikolo 1994) have shown that nonconventional 
social security instruments have failed to promote the 
equitable economic growth which can harness the productive capacity 
of the large number of unemployed and underemployed working 
people in the all areas. Many of the past fiscal and monetary policies 
and policies dealing with investments, trade, prices, finance and others 
have been heavily biased in favour of the well-off at the expense of 
the poor, the unemployed and underemployed. Indeed, this is the area 
where the greatest potential for poverty alleviation through structural 
changes in the economy lies. Such a change would, at the same time, 
hasten the process by which the poor can be integrated in the mainstream 
formal economy and join the formal social security system as 
they would have graduated from being primarily concerned with 
survival to covering their contingency needs. 
These studies also show that many of the State-sponsored programmes 
for the alleviation of poverty have failed to reach the poor and 
improve their standard of living for a host of reasons. Firstly, these 
programmes are often flawed in their design and delivery system, 
resulting in the leakage of the benefits to the rich people. One example 
of this is the ambiguity in the selection criteria that purport to identify 
the intended beneficiaries. Secondly, they display a lack of focus or 
exclusivity in the targeting of beneficiaries. Thirdly, they show a lack 
of accountability and incentives and, lastly, the resources allocated for 
these programmes are inadequate. 
The State's lack of financial resources as well as its lack of political 
will are other factors leading to its failure to provide adequate social 
protection to the poor. Tungaraza (1990:62) shows that Tanzania's 
social security expenditure quota (defined as the proportion of social 
security expenditures on gross domestic product) in the 1960s, 1970s 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 19
and 1980s has been remarkably low, thus indicating the low priority it is 
given. 
Non-governmental organizations and self-help organizations have 
been able to reach only a limited though significant, number of the poor. 
The reasons for this include a lack of financial resources which hamper 
their efforts to scale up their otherwise effective programmes and 
sustain them over time and their weak administrative base, since they 
are limited to working in a small area. 
Traditional and informal social security arrangements 
A number of studies have attempted to understand the nature and forms 
of traditional and informal social security arrangements in Tanzania. 
Among them include works by Bossert (1987), Tungaraza (1988, 1993, 
1994 and 2000b), Victor (1997) and Urassa (2000). Bossert (1987) 
identifies four institutions of traditional social security in pre-colonial 
Tanzania, namely, domestic groups, which live, produce and consume 
together, kinship organizations (lineages, clans), the neighbourhood 
and the chief. According to Bossert, in the past there was mutual 
assistance and economic co-operation among these groups and each 
was obliged to help the other in case of special difficulties or during 
ceremonies. 
The neighbourhood community is generally composed of relatives. 
Social relations between them are therefore determined by the rules, 
norms and obligations that obtain between relatives. Thus, relations of 
obligations and commitment hold between kinsmen, as well as between 
neighbours who are not kin. The chief is the ritual head, the supreme 
judge and the military leader of his people. Apart from controlling all 
the local resources, in times past the chief guaranteed the survival of 
his subjects in case of hardships. A successful chief was therefore 
expected to keep a great stock of grain, to be generous and hospitable 
towards his subjects and to provide them with food in times of 
emergency. 
There was a difference between the provisions for the protection of 
people from each contingency. For example, old people with many 
children were adequately protected while childless old people had 
almost none. Although there were provisions for medical, nursing and 
20 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchonivu, Tungazara and Maghimbi/TANZANIA REPORT 
ritual care in case of illness, the effectiveness of this was limited by the 
low level of medical knowledge. 
The changes that took place during colonial period and after 
Independence have led to significant changes in the traditional social 
security practices. Studies conducted in the 1980s and 1990s show 
that the domestic group (family institution) still provides support to 
close relatives only (Bossert 1987, Tungaraza 1988 and 1993). Even 
the family as a social security institution is increasingly failing to 
provide sufficient protection to its members against various contingencies 
(Tungaraza 1994, Urassa 2000). This has led to the proliferation 
of other forms of informal social security systems. In urban areas there 
are money exchange networks (popularly known as Upatu), the pooling 
of labour networks and pooling of resources (Tungaraza 1993 and 1994, 
Victor 1997). Victor (1997) shows that in urban Morogoro District 
there are Chama cha kufa na kuzikana, associations concerning with 
death and burial affairs; Jumuia, (religious groups formed by people 
who belong to the same denomination); tribal associations; community-
based organizations and UMASIDA (a mutual health insurance in 
Dar es Salaam). 
Informal social security systems which currently operate among 
colleagues, neighbourhood groups and friends have been criticized (see 
Tungaraza 1993 and 1994) for failing to provide long-term protection 
against various risks and for providing inadequate protection to their 
members. This is because coverage depends on one's ability to 
contribute to the group. Those who are unable to contribute are 
automatically disqualified from membership. Their lack of financial 
resources means that these groups cannot alleyiate the poverty of their 
members and they are able only to provide short-term protection. 
Moreover, these forms of social security are confined to a small 
geographical area. Finally, they are dominated and guided by customs 
and traditions which were normally biased towards men. 
Formal social security schemes 
Studies that have investigated the nature and forms of formal social 
security systems in Tanzania include the study by Bossert (1987), 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 21
Tungaraza (1988), Mlyansi (1991), Wangwe and Tibandebage, (1999). 
According to Bossert (1987), formal social security systems in 
Tanzania are organized according to three major principles; social 
insurance, public support and social assistance. Other principles 
include the principle of obliging an employer to grant security and the 
principle of compulsory saving. The principle of insurance is the basis 
of the Parastatal Pensions Fund (PPF), The National Social Security 
Fund (NSSF), the Public Service Pensions Fund (PSPF), the National 
Health Insurance Fund (NH1F) and the Workmen's Compensation 
Scheme. All these schemes are contributory and cater for employees 
in the formal sector of the economy. For example, the Parastatal 
Pensions Fund (PPF) caters for employees holding pensionable offices 
in parastatals, the National Social Security Fund (NSSF) caters for all 
parastatal employees who do not hold pensionable offices and 
employees in the private sector and the Public Service Pensions Fund 
(PSPF) caters for all civil servants who hold a pensionable office in 
the civil service. 
The public support principle forms the basis of the public health 
services in Tanzania and non-contributory pension schemes such as the 
Teachers Service Commission of 1965, the Defence Forces Gratuities 
and Pension Scheme of 1966 and the Specified State Leaders scheme 
of 1999. The principle of social assistance provides benefits to people 
affected by social risks who can prove they have no means of their own 
and do not get support from their own relatives. Examples of assistance 
given include support to beggars and peasants, in case of agricultural 
risks, floods, victims of the MV Bukoba, a ship that sank in Lake 
Victoria and victims of earthquakes in Rungwe District, Mbeya Region. 
The principle of compulsory saving is the basis of provident funds. 
Each member has an individual account into which members and their 
employers pay the prescribed contribution. When members are due 
for benefit (due to old age,.invalidity, death and sickness), they receive 
the legally prescribed payments from the accumulated account. 
The principle of employer liability applies in case of the continuation of 
wage payments due to illness as per the Regulation of Wages and 
Terms of Employment Order of 1972. This principle also applies in 
22 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
case of compensation payments when employees are dismissed in 
terms of the Severance Allowance Act of 1962. It applies in case of 
certain payments at death in terms of the Employment Ordinance of 
1957. Formal social security schemes.are riddled with problems. 
Among these include delay in processing benefits, unnecessary 
bureaucratic red-tape, lack of transparency, disparities in risks covered 
and poor governance (Mangangila 1976, and Bossert 1987). 
The origins and development of formal social security 
There are a few studies on the origins and development of formal social 
security in Tanzania. These include studies by Bossert (1987) and 
Tungaraza (1988). Bossert examines the origin and development of 
modern social security from pre-colonial period up to-1983 while 
Tungaraza explores the period from 1961 to 1988. According to 
Bossert (1987), formal social security in Tanzania can be traced back 
to the period of colonial penetration, first, by the Germans (1885- 
1918) and then by the British (1918-1961), when a new political 
system was established alongside the introduction of plantations and 
the imposition of a hut tax which forced the African people either to 
migrate into various economic enterprises or to participate in cash 
crop production. It was during this period that some social security 
measures were introduced, such as pensions schemes, compensation 
payments to workers injured while performing their duties, the 
distribution of food to the rural population during years of poor harvest 
and the provision of health services and education for serving colonial 
government officials, thus forming a nucleus of the present formal 
social security system. 
Bossert (1987) argues that the formal social security in Tanzania 
was initiated by the colonial administration and not the result of 
pressure from African political leaders and workers. These measures 
were initiated in the medical field, followed by old age pension 
schemes and, later, by workmen's compensation schemes. 
From Independence in 1961 up to the late 1980s, the post-colonial State 
pursued policies that led to political, economic and social changes in 
the country. This involved, among other things, the development of 
agriculture and industry, the expansion of education and health and 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 23
the settlement of rural people into Ujamaa Villages. These changes had 
an impact on the development of formal social security schemes. According 
to Bossert (1987) and Tungaraza (1988), in addition to the rapid 
development of health facilities in the rural areas and the expansion of 
education, family allowances and the subsidization of corn meal were 
established together with many other new social security schemes. The 
expansion of social security schemes during this period was a result of 
both the agitations of trade unionists and government's own initiatives. 
After Independence urban-based public health services were put in 
place. After the Arusha Declaration in 1967 the development of social 
security was characterized by the expansion of rural health, the provision 
of family allowances and tax relief from 1980 for employed couples, 
the continuation of payment of salaries at times of illness (1972) and 
paid maternity leave (1975). 
A number of studies exist on the development of formal social security 
systems from 1985 to the present, that is, the period of structural 
adjustment policies and globalization in Tanzania. Among these include 
one by Haule (1994) who says that structural adjustment policies have 
led to the loss of members from the social security schemes such as the 
Parastatal Pension Fund and the National Social Security Fund because 
of retrenchment. This has led to the drain on the resources of the 
social security schemes due to the payment of benefits to thousands of 
workers who were withdrawn prematurely. 
The devaluation of the Tanzanian Shilling, as a result of structural 
adjustment, caused a serious erosion of the financial assets and solvency 
of social security schemes. Social security schemes in this period 
tended to develop in an ad hoc manner in response to specific needs 
or pressures rather than in accordance with any long-term national 
strategy. The result has been disparities in provision, the duplication 
of administration and a failure to relate scarce resources to national 
priorities. 
In the health sector structural adjustment policies have led to 
reforms such as the introduction of user fees in the health sector 
(Warioba 1998, Msoka 1998 and Tibandebage and Mackintosh 2000). 
This has led to the resurgence of illnesses such as malaria, upper 
24 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchamvu, Tungazara and Maghimbi/TANZANIA REPORT 
respiratory infections and TB as well as the non-compliance of patients 
with treatment regimes, because neither urban nor rural people can 
afford medical care. On access and exclusion, Tibandebage and 
Mackintosh (2000) conclude that there is considerable exclusion 
because health care is not affordable. Consequences of exclusion for 
patients include death, borrowing to pay for treatment and putting the 
lives of other members of the community in danger. 
Coverage of social security schemes and kinds of 
benefits provided 
Various studies (see Bossert 1987, Tungaraza 1988 and Wangwe and 
Tibandebage 1999) on conventional social security have shown that the 
coverage of formal social security schemes is not comprehensive in 
terms of the population and risks covered. Currently, such schemes 
cover only 6% of the population and about 5% of the active labour force 
in the country. In terms of gender, the majority of the people covered 
are men, since men constitute most those employed in the formal 
sector. In terms of risks, these schemes focus on a few benefits 
(Bossert 1987, Wangwe and Tibandebage 1999). They cover old age, 
disability, survivorship, illness, maternity, occupational accidents and 
diseases. Benefits are of two kinds: benefits in kind, where members 
are entitled to medical services from their employers and financial 
benefits in the case of illness and maternity. Provident funds give lump 
sump benefits to their members while social security schemes based 
on the social insurance principle provide initial lump sum payments 
followed by monthly benefits to their members. The same applies to 
non-contributory pension schemes. 
The literature on the benefits available shows that conventional social 
security schemes have failed to protect members in distress so they 
may be self-respecting, self-sustaining and valuable members of society 
(Mangangila 1976, Mataba 1983, and Mlyansi 1991). For example, the 
rates of interest awarded annually to members of provident funds have 
invariably been negative in real terms and the lump sums paid out 
generally represent no more than a few month earnings. 
Adequate protection against hardship should be the fundamental 
guide in the establishment of the benefit structure of a national system 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 25
of social security. Protection should be based on average earnings and 
should be related to family needs. There should be, for example, a 50% 
supplementary allowance for an aged wife or the younger widow and 
the surviving children of an insured individual. In Tanzania the criteria 
used to fix the level of social protection are secrets only known to the 
social security schemes. 
Strengths, weaknesses, opportunities and threats 
Strengths 
• There exist different forms of social security in Tanzania. These have 
a system and structure which can be studied and improved. 
• Members of non-formal and informal social security systems seem 
to have developed a sense of ownership of these systems which 
thus have the potential to be expanded and improved. 
• There is some commitment from the government to act as a guarantor 
of all existing formal social security systems in the country. 
Weaknesses 
• Formal social security schemes cover very few people and the 
benefits paid are inadequate. 
• They are characterized by delays in making payments to members, 
bureaucracy, inefficiency and poor governance. 
• They are usually monopolistic and workers are forced by law to 
join, with no choice between schemes. 
• They are inflexible and do not provide a forum for members to have 
• There is no possibility of portability of members' benefits from one 
• There is no systematic and comprehensive national social security 
a say in the way their contributions are invested. 
scheme to another. 
policy which governs all social security programmes and activities in 
the country. 
• Social protection is still regarded not as a right of all citizens but as 
a privilege. 
Opportunities 
The commitment of the government as a guarantor to formal social 
security can be used as an avenue to strengthen these systems. The 
26 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
current emphasis on investigating the potential of non-formal social 
security systems in the provisions of social protection can be exploited 
in extending social security coverage to those who are now excluded. 
Threat 
Government policy, legal and financial support have focused on largescale 
formal social security schemes such as the NSSF and PPF but 
there is no articulate programme on social protection of people outside 
the formal sector. Such thinking has continued to dominate policy 
maker and thus continue to be a threat to non-formal social security 
development. 
Conclusion 
Conventional social security programmes cover a relatively limited 
proportion of the population and have not reached the urban and rural 
poor. They do not provide adequate social protection to members, thus 
leaving members in poverty when faced with various contingencies. 
The poor rely on non-conventional social security programmes which 
serve them only inadequately. This is exacerbated by the erosion of 
traditional social security practices as a result of increasing urbanization 
and the process of globalization as the extended family system 
breaks down and as population pressure impinges on traditional use 
of common property. 
There is a need to understand the social security needs of different 
categories of poor people in Tanzania, the sources and types of risks 
confronting the poor and the different arrangements and strategies 
(informal, market-based and public) for risk reduction, mitigation and 
coping used by individuals and the government. In order to have a real 
picture of the inequities in the distribution of resources and incomes in 
contemporary Tanzania, we must understand the extent of redistribution 
across generations and within a generation in both conventional 
and non-conventional social security schemes and programmes. 
Furthermore, we need to know the transactions and hidden opportunity 
costs involved in informal risk sharing arrangements. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 27
28 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mdhomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Phase II: Co-operatives and Social Protection 
ABSTRACT 
In the 1950s and 1960s Tanzania had the third largest co-operative 
movement in the world. These co-operatives provided economic and 
social protection to members so that poor peasants could sell their 
crops even in years of bad world market prices. The services provided 
by co-operatives, like education and trusteeship for peasants 
who took out loans, collapsed when the government abolished cooperatives 
in 1976. They were re-introduced.in 1982 but, due to 
their abolition, they had lost capital, personnel and members. The 
current co-operatives are much weaker than the pre-1976 ones and 
cannot provide the same kind of protection they once did. Cooperatives 
have still a great potential for social and economic 
protection but much change in the government policy on cooperatives 
is needed. 
Introduction 
THERE IS NO SINGLE INSTITUTION besides the co-operative that has played 
a greater role in the lives of Tanzanians. The strong co-operative 
movement in Tanzania served members and non-members all over the 
country and in the 1960s Tanzania had the largest co-operative 
movement in Africa. Co-operatives in Tanzania operated on the 
principle that the economically and socially strong should shoulder 
the weak. The decline of the co-operative movement since the 1970s 
has resulted in the erosion of the protection they provided to their 
members. Current co-operatives are weak and have suffered from 
obstructive and ineffective government policies(Kennedy, 1994:191). 
The term "co-operative" is ordinarily associated with the consumer 
co-operative store established in 1844 at Rochdale, England, of 28 
people. The term is also used to describe collective self-help. What 
made Rochdale special is that it worked and so became the basis for a 
worldwide movement. Rochdale provided the rules by which co- 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 29
operative societies could function. These rules (including voluntary 
and open membership, democratic control, a limited return to capital, 
the ownership by members of any surplus, member education and cooperation 
among co-operatives) are adaptable and can be utilized by all 
classes and all nations (Melnyk, 1985:3-4). 
The concept of the co-operative is fluid and the principles of cooperatives 
offer organizational flexibility. Co-operation has taken a 
great variety of forms. In some case only loose associations are 
formed, such as the Italian fishermen co-operatives for administering 
member insurance. In other cases co-operation is rigidly organized and 
cohesive, such as the Israeli kibbutzim with community ownership of 
work. Between these two extremes there is a large range of co-operative 
types (Melnyk, 1985:3). 
Co-operatives and social protection 
Co-operatives have helped to provide social protection in modern 
Tanzania. They have contributed to social stability and passed on to their 
members the values and norms that promote social order and stability. 
Modern co-operative institutions based on Rochdalean principles were 
introduced in mainland Tanzania (Tanganyika) in the 1920s. Their legal 
status has been defined in various laws, starting with the Co-operative 
Societies Ordinance of 1932. 
Modern co-operatives in Tanzania are basically a rural phenomenon. 
They are owned and run by peasants and their main activity is buying 
members' and non-members' crops for export. The early co-operatives 
enabled peasants to reap the trading profit, which had before gone to 
Asian merchants. Co-operatives provided security because they bought 
peasants' crops even in years of very low prices on the world market. 
Co-operatives also provided social services support to their members. 
In Moshi older peasants told us that the Kilimanjaro Native 
Co-operative Union (KNCU) provided scholarships for its members' 
children and supported primary and secondary education. The KNCU 
built Lyamungo Secondary School in 1962, which was given to the 
government in 1965. It even opened its own business school (KNCU 
College of Commerce) in 1956. The latter was closed in 1962, possibly 
30 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomyu, Tungazara and Maghimbi/TANZANIA REPORT 
due to conflict between the established co-operative leadership and 
the emerging political leadership (Tanganyika got Independence in 
1961 but the nationalist struggle entered party politics only in 1954). 
The decline of the co-operative movement 
Modern co-operatives were abolished by the government in 1976 and 
re-established in 1982 (Bantukari 2000:40-43). The abolition contributed 
to the current.bad state of co-operatives because they lost both capital, 
personnel and members after abolition. By 1976 the co-operatives had 
built up an an encouraging level of capital through retained earnings, 
cash reserves, members' deposits and cash investments in various 
holdings. These resources were all frozen when co-operatives were 
abolished (Bantukari 2000:45). 
When co-operatives were re-introduced they had lost their monopoly 
in buying peasants' crops in their respective areas and they now 
have to compete with crop merchants. This is part of the structural 
adjustment policies which have helped to weaken co-operatives. 
Modern co-operatives now are very weak: for example, they buy only 
about 10% of the peasant cotton crop although in the past they handled 
the whole crop. 
In 1968 the co-operative movement in Tanzania was the largest in 
Africa, handling £27,500,000 worth of the country's annual exports. 
This was half the value of the country's exports. Only in Israel and 
Denmark did co-operatives market a greater proportion of a nation's 
overseas business (UPA 1968:85). The peasants interviewed in focus 
group discussions in Same, Dodoma, Moshi and Mwanza dismissed 
the suggestion that the current (post-1982) co-operatives provide 
social protection, saying that the old co-operatives were better placed 
to do so. The old co-operatives had well-defined functions which 
helped peasants to consolidate their social and economic position. 
These functions were: 
• Distributing farm implements and other agricultural inputs; 
• Acting as trustees for peasants who took loans from credit agencies; 
• Running special funds for members' education; 
• Constructing primary and secondary schools; 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 31
• Supporting higher education through scholarships; 
• Building factories for processing agricultural products bought from 
members (e.g. cotton gunneries, coffee curing, oil processing mills 
and maize mills). 
Findings from field observations in Moshi, Dodoma and Mwanza 
indicate that modern co-operatives have been in decline since 1976. 
When they were re-introduced in 1982, they were allowed to form 
without share capital subscriptions in the strict and practical sense. 
Members of co-operatives were required to pay only non-refundable 
dues and co-operatives were thus unable to build up capital. Although 
the law was changed again in 1991, current co-operatives lack cropbuying 
funds. Another researcher observed that in 1960 co-operatives 
were able to meet 33.5% of the funds required to purchase crops but 
by 1996.they could raise only 0.7% of the funds (Banturaki 2000:41). 
Membership 
Focus group discussions conducted in Chome and Ugweno divisions 
(Same and Mwanga districts) revealed that co-operatives have very few 
members at present and most peasants in the country are forced by 
circumstances to sell their crops to crop merchants. In Kilimanjaro, 
Mwanza and Dodoma regions before 1976 most peasants were 
members of co-operatives. In most villages in Moshi district (now Hai, 
Rombo and Moshi districts) all peasants were members of the KNCU 
but only a tiny fraction are members today. The total co-operative 
membership in the whole countryside has dropped from serving about 
20% of the peasants in 1976 to about 5% in 1982 and 3% in 1991 
(Banturaki 2000:49-50). This figure obscures the fact that, before 
1976, co-operatives served almost all peasants because they were not 
restricted to their own membership. 
The peasant informants were clear that they got a better deal when 
selling crops in the pre-1976 period than now. Crop merchants who 
buy most of the crops in the country now do not pay the second 
payment, which is the difference between the price co-operatives give 
32 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
members up front and what they actually get in the world market after 
deducting costs like interest on money advanced by banks. 
The commercialization of co-operatives 
The 1982 co-operative legislation placed the co-operatives' apex 
organization under the patronage of the ruling political party and 
membership was almost compulsory, but not clearly defined, because 
every village in the country was supposed to be a political wing of the 
ruling party and a multi-purpose co-operative society. After the 1991 
co-operative law, membership has become voluntary and peasants are 
not obliged to sell their crops to co-operative societies. At the same 
time co-operatives have introduced attempts to make their members 
pay share capital. This has contributed to the low number of members. 
Co-operatives have in the 1990s become more elitist and are controlled 
by the richer peasants in the villages. The poorer peasants have thus 
been disenfranchized and marginalized. 
In Kilimanjaro, Mwanza and Dodoma regions, where the cooperatives 
were surveyed, no poor peasants have been "chased" from 
co-operatives but most poor could not meet the conditions for joining 
them after 1991 and were therefore sidelined. In Bukoba, for example, 
the peasant must pay the share capital in full and at one time and be 
able to sell to the co-operative over 200kgs of coffee per year 
(Bantukari 2000:42-43). The situation is the same in Dodoma, Mwanza 
and Kilimanjaro. Most of the poorer peasants cannot afford the time 
waiting to sell their crops to bureaucratic co-operatives. They are thus 
forced to sell to crop merchants, who offer lower prices. 
Sometimes co-operatives themselves cannot offer price incentives 
and they have thus lost their traditional role in protecting peasants. It 
was reported in one newspaper (Family Mirror 31 July-6 August 
2001), for example, that co-operatives in Kagera region borrowed 
peasants' coffee in the previous year's coffee season (that is, 2000). 
In 2001, under the cloud of last season's debts, these co-operatives 
have anriounced a price of Tshs. 80 per kilogram of dried unpulped 
robusta coffee. At the same time private traders from Uganda were 
prepared to pay Tshs. 170 per kilogramme for the same coffee. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 33
Operating problems in co-operatives 
It has also been reported by the Minister for Co-operatives and Marketing 
in Daily News 25 June 2001 that co-operatives in Tanzania are 
now in an appalling state. By May 2001 they owed banks Tshs.17.8 
billion. This is very discouraging as co-operatives had accumulated 
this mammoth debt despite the cancellation of their Tshs. 44 billion debt 
by the government in the last ten years. This was revealed during the 
general meeting of the Arusha Co-operative Union (ACU) on 23 June 
2001. The ACU itself owes banks more than Tshs. 5.8 billion and it 
owes other creditors Tshs. 899 million. 
The Minister said that the ACU was found by an investigating 
committee to have incurred a ioss of more than Tshs. 145 million. The 
was the result of poor supervision of ACU projects and "ghost" 
payments (payments not backed up by the delivery of goods, or services 
or goods or services not measuring up to the value of payments). The 
ACU had been approving loans to members in a shady manner. It did 
not conduct proper financial supervision, auditing and employment and 
this, according to the Minister, contributed to frequent financial losses. 
The ACU had a small investment base and conducted business in oldfashioned 
ways. Its members were not adequately sensitized to the 
benefits of co-operatives. If things continued in this way, said the 
Minister, the ACU would not be able to survive in a liberalized market 
economy with fierce competition. Poor leadership in co-operatives 
contributes to the endless woes in co-operatives today. 
All these problems seem to be the order of the day in contemporary 
Tanzanian co-operatives. In focus group discussions peasants in 
Kilimanjaro, Dodoma and Mwanza regions mentioned the same 
problems over and over again. They wereparticularly discouraged by 
the practice of ghost payments made by co-operative officials and the 
regular occurrence of outright theft. More data on this theme will be 
presented later. The Minister promised that his ministry is committed 
to revamp and popularize co-operatives in Tanzania to enable them to 
run profitably and that his ministry will ensure that the current Cooperative 
Act is reviewed. This is a positive attitude, but we note that 
major policy reform of co-operatives is needed in order to enable them 
34 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mohomvu, Tungazara and Maghimbi/TANZANIA REPORT 
make a meaningful contribution to the economic advancement and 
social security protection of the Tanzanian peasant. The role of crop 
merchants, for example, must be clearly spelled out and, when crop 
merchants threaten the peasants' security, the co-operative must be 
given priority in policies on marketing and agricultural reform. 
It seems clear that co-operatives specializing in crop buying are in a 
serious crisis and cannot provide the social and economic protection 
they used to before 1976. Their members cannot be solely blamed for 
the poor state of their co-operatives. Contradictory policies over the 
years, including policies which have abolished strong and viable cooperatives 
which poor peasants toiled for years to build, have also 
contributed. The government should view the co-operative as an 
economic venture with an important social function and formulate 
policies conducive to their evolution. This will occur only when there 
is a change in government thinking and when it is able to view the 
peasant co-operative as a tool for social security protection. 
From in-depth interviews we found that co-operatives lack capital 
to buy crops and to provide social services to members. They are 
unable to attract credit for their members and, even if such credit were 
available, few creditors would accept the guarantee of shaky cooperatives 
for their members. Many co-operative unions are under 
receivership because of their huge debts. These include co-operatives 
in areas where they were strong before 1976. Some of the bigest of 
these under liquidation are the Dodoma Region Co-operative Union 
(CRCU), Tanga Region Co-operative Union (TARECU), (Shinyanga 
Region Co-operative Union (SHIRECU), Lindi Region Co-operative 
Union (LIRECU), Ruvuma Region Co-operative Union (RURECU) 
and Buha Co-operative Union. 
Case study I: The CRCU co-operative 
The CRCU covers Dodoma region. It was abolished by the government 
in 1975 and a new CRCU was established in 1984 as CRCU 1984 Ltd. 
The problems it faces are similar to those faced by other co-operatives 
all over the country and its situation will serve as a case study. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 35
After the 1991 Co-operative Act every co-,operative must have formal 
members and those served by the co-operative are not automatic 
members. Every co-operative member has to contribute shares: if a 
co-operative has no contributions it ij flot registered. Thus membership 
has dwindled in the 1990s, as we saw earlier. 
CRCU 1984 Ltd. had clear objectives including: 
• improving the standard of living of its members; 
• conducting business on behalf of peasant farmers in Dodoma 
region; 
• buying mixed crops like maize, millet, sunflower, finger millet, 
peanuts, chicken peas, oil seeds, honey and beeswax. 
Three hundred or more primary co-operatives in Dodoma region joined 
CRCU but each village contributed a share capital of only Tshs. 
1,500.00 (about US$2). The initial membership capital contributed by 
all co-operatives amounted to Tshs. 399,000 only. Thus CRCU did not 
have the capital to procure crops. To salvage this situation CRCU took 
'out a loan of Tshs. 1,225,000,000 from the National Bank of Commerce 
(NBC). Having failed to service its loan, CRCU ultimately became a 
giant debtor. The total value of CRCU was only Tshs. 94,000,000. 
Co-operatives continued to decline financially and in 1992 the 
Ministry of Agriculture and Co-operatives put up a budget of Tshs. 
35,000,000,000 to rescue them. This sum was distributed to different cooperative 
unions in the country. In Dodoma it was claimed that CRCU 
was buying crops which did not exist (these are also known as "vacuum 
crops" in Tanzania). There were also claims that crops had been bought 
but they did not tally with value of money paid. Funds were misused to 
purchase items not budgeted for. It was claimed that luxury cars were 
bought and officers of the union were paid unnecessary allowances. 
Between 1984 and 1992 the management of CRCU changed five times 
and no manager completed two years of service. Thus no manager 
stayed long enough to ensure that a proper system was set up and to 
make sound plans for crop buying and exporting or selling locally. 
Under these conditions, adequate measures to pay back bank loans 
were not formulated. From 1984 to 1990 the leadership of CRCU was 
• very weak 
36 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
The CRCU continued to make heavy loses and there were many cases 
of cheating. For example a CRCU lorry leaving for Dar es Salaam could 
be loaded with 10 tons of crops but, when it reached a government 
checkpoint such as a weighing bridge, the load would register only 7.5 
tons. Or the record in the lorry might show 10 tons but the actual 
weight would read 15 tons; but no manager or board member to.ok any 
measure to prevent this. According to the co-operative officer for 
Dodoma rural district, the point was reached when CRCU started to 
default in paying its staff salaries. In March 1997 a liquidator (LART) 
was appointed for CRCU and the six other unions mentioned above. 
These assets went into receivership but the last time money was 
received in March 1998 when LART handed over Tshs. 1,200,000. The 
liquidator was supposed to be paid Tshs. 200.000 every month and a 
watchman Tshs. 17,000 per month. Two lorries, one Fiat (10.9 tons) 
and one Scania (9 tons) were handled over to the liquidator. By mid 
2001 these lorries had not been sold. Before liquidation the Scania 
lorry was hired out for a six month period to a businessman from Moshi 
at a rate of Tshs. 650,000 per month. After one year the businessman 
came to renew the contract and since then he has not been seen in 
Dodoma. Before liquidation the Fiat lorry was taken to Dar es Salaam 
for maintenance: the bill came to Tshs. 1,200,000. This lorry was 
dumped at Buguruni Garage and parts of the lorry were sold in bits. 
All that is left is scrap. The watchmen has left and abandoned CRCU 
property worth Tshs. 13,000,000 at Gulwe. Three godowns in Kondoa 
have also been abandoned. LART has so far not facilitated the liquidation. 
By April 1997 there were 22 ex-staff who owed CRCU Tshs. 
9,600,000. Trade creditors owe CRCU Tshs. 136,000,000. Peasants 
and other Tanzanians understand the potentials of co-operatives in 
improving their position but they feel they have been let down by those 
in authority: In some areas peasants are trying to reform their cooperatives 
or build new ones. 
Case study II: Mvumi Makulu Zabibu Co-operative 
In the Dodoma region vines grow well but there is no local market for 
grapes and wine. A government wine-pressing mill was closed down 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 37
in the early 1990s and since then the problems of the growers have 
increased. In 2001 they were in the process of forming a vine growers 
co-operative society (Mvumi Makulu Zabibu Co-operative Society) 
in an attempt to protect their status. This has deteriorated sharply after 
structural adjustment measures which have removed government 
subsidies 'for crop inputs and reduced the traditional protection given 
to co-operatives. The anarchist application of trade liberalization has 
allowed the importation of cheap wines (mostly from South Africa) 
into Tanzania to the point where the nascent local wine industry and the 
security of vine growing peasants are threatened. 
Co-operative now have very few members. From in-depth interviews 
in the villages these appear to be the richer peasants who are able to 
wait for longer periods before selling their coffee and other crops. Poor 
peasants cannot do this because of pressing requirements for cash to 
buy social services like health and education and for necessities like 
food, cloth, kerosene and building materials. The 1991 Co-operatives 
Act treats share capital as a necessary feature of business that can be 
transferred seamlessly to a co-operative enterprise. Members of a cooperative 
are therefore not able to exercise their rights until they have 
paid their shares, their entry fees and any other dues prescribed in the 
by-laws of the particular co-operative society. Although the 1991 Cooperative 
Act does not explicitly state that payment of share capital is 
a condition for registration, co-operative officers and the local elite 
have enforced this. The local dlite, which consists of rich peasants, may 
have its own interest in dominating other peasants socially. Thus, even 
if co-operatives become economically viable, they may in the long run 
provide protection only to rich peasants and not the poor majority. 
The 1991 Co-operative Act recognizes co-operatives as private 
institutions estabfished and managed by members. It also decrees that 
the co-operative movement in Tanzania is based on the international 
principles adopted by the International Co-operative Alliance. This act 
doe not lay out a prescribed co-operative structure. Strictly speaking, 
therefore, when rich peasants organize themselves in co-operatives and 
leave their poorer neighbours at the mercy of the naked market, they break 
no law. There is thus a need for the Ministry of Co-operatives and Mar- 
38 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghlmbi/TANZANIA REPORT 
keting to take a broader view of co-operatives as an instrument that can 
help poor peasants in production and marketing. Co-operatives need to 
be addressed in a broad policy framework to improve their capacity to 
provide social security and economic protection. This can be done 
through special schemes to help poor peasants to join by lending them 
resources to buy share capital in existing co-operatives or to form their 
own. 
Co-operatives are poorly managed but this problem is not beyond 
correction. The peasants also felt that co-operatives need more 
government support in auditing and supervision. We can add that the 
1991 Co-operative Act misses the point. Co-operatives are not only 
commercial institutions but are also, traditionally, mutual support 
organizations. The 1991 Co-operative Act clearly states that the cooperative 
society must appear commercially viable before it is registered 
and a feasibility study has to be submitted with the application for registration. 
Under the 1982 Co-operative Act everyone living in a village 
where a co-operative operated was a de facto member of the co-operative 
society. The 1991 Co-operative Act bases membership on purely economic 
criteria and members' choice and the area of operation of a co-operative is 
not defined in terms of a neighbourhood or administrative boundary, as 
was the case in the 1982 Co-operative Act. The 1991 Co-operative Act 
endorsed the voluntary nature of co-operation, which is one of the cooperative 
principles, but not the social aspect of co-operation. The idea 
of the strong supporting the weak in co-operation is missing. 
Savings and credit societies 
Besides co-operatives dealing with crops the other category of modern 
co-operatives are savings and credit societies. These are very important 
in light of the fact that most Tanzanians are absolutely poor. So far 
these societies have not reached as many people as either the old or the 
current crop co-operatives. Membership of a savings and credit cooperative 
depends on a regular income or the ability to pay and 
traditionally, therefore, these societies have drawn their members from 
salaried employees. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 39
Primary school teachers seem to be leading in the country in forming 
savings and credit societies. In Dodoma and Kilimanjaro regions credit 
and savings societies were noticed. In Moshi town there was a teachers 
saving and credit society and one for the staff of the Northern Diocese 
of the Evangelical Lutheran Church of Tanzania. There was one for the 
staff of Co-operative College Moshi and in Dodoma one for teachers. 
All these savings and credit societies function in the same way. 
Members, who are employees, authorize their employers to make a 
monthly contribution from their salary and send it to the savings and 
credit co-operative on their behalf. When members take out a loan they 
authorize the employer to make a monthly deduction from their salary 
on their behalf and send it to the society. The contingencies thus 
covered are many and range from the death of a family member to 
raising a small sum of capital for building part of a house or injecting 
capital in a small business. 
Saving and credit societies faGe problems similar to other cooperatives. 
They have very limited capital because their salaries and 
thus their deductions, are small. No capital from outside has been 
injected into the societies and members can borrow only after monthly 
deductions and loan repayments have been made. They thus cannot 
borrow on short notice and have to queue for loans which can only be 
made at the end of the month. Because only small sums can be 
borrowed members have not been able to break out of the vicious cycle 
of continuously contracting small loans which cannot significantly 
change their economic situation. Unless there is major reorganization 
of these co-operatives they will continue to play a small role in poverty 
reduction and social security protection. 
The teachers savings and credit association in Dodoma rural district 
started by contributions of Tshs. 1,000 per share which has risen to 
Tshs. 5,000 per share. A single member can own up to 20% of all 
shares. The society has a membership of 14,000 teachers: it is the 
largest saving and credit association in the central zone of Tanzania. 
Total collections in one month amounts to Tshs. 9,000,000 and the same 
amount is given out each month as loans; thus very small loans are 
given out. However the members felt that these loans were important. 
40 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Co-operatives and women 
The argument in this paper is that co-operatives have a great potential 
in providing social protection for the household. The focus on the 
peasant is not accidental but is a consequence of historical reality. Cooperatives 
formed a formidable social movement in the 1950s and 1960s. 
They had built up funds and could offer economic and social incentives 
like scholarships to members and other peasants who lived where cooperatives 
operated. Social conditions in Tanzania have not changed, 
except insofar as poverty has actually increased in the 1980s and 1990s. 
About 70% of the population are still small-scale peasants. The middle 
class is very small, even in urban areas, where most people are petty 
traders or are involved in the informal sector. The fact, however, that cooperatives 
in Tanzania have always been dominated and controlled by 
men is a key to one of the weaknesses of co-operatives. Since their 
inception in the 1970s modern co-operatives concentrated on the 
marketing of cash crops. Colonial and post-colonial extension effort 
always targeted the male head of the peasant household as the 
"farmer". Few women have thus been members of co-operatives, even 
after Independence. 
The situation is the same in other countries in Africa. Government 
support, devised to increase cash crop production, goes to men who 
attended courses at farm institutes, are granted loans and are taught 
improved methods of crop production. Women have largely been 
excluded from those benefits. Stereotypes about gender roles seriously 
inhibited the capacity of women to support themselves on the land (Pahl 
1988:352). 
In Dodoma, Mwanza and Kilimanjaro regions it was not difficult to 
observe the well-known phenomenon of the asymmetrical division of 
labour between men and women. Women shoulder more than their 
rightful share in producing crops, in domestic labour and in milking 
animals. In Dodoma and Mwanza men take animals to pastures. However 
in Kilimanjaro this occurs only in the plains. In the mountain zone 
where most of the people live, the feeding of livestock is only done by 
women. This is very laborious work, which sometimes involves much 
walking. Animal fodder is not easily available because of high human 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 41
population density. Land and animals are owned and controlled by men. 
Whatever security is provided by co-operatives, they tend to favour 
men and marginalize women, who provide so much of the labour in 
peasant households. 
In theory after the 1991 Co-operative Act women have been able to 
join co-operatives on their own after paying the necessary dues. 
However in real life the registered co-operative member is still the 
male head of the household. Yet it is well known that, when the social 
and economic condition of women improves, social protection for a 
greater section of the population (including children) occurs. Much 
reform is therefore needed in co-operatives to improve the status and 
role of women. In Rochdalean co-operatives, encouraging women to 
establish their own co-operatives may help to improve their position. 
However this assumes that other reforms will be put in place, like 
changes in the social attitudes and reforms to ensure that more women 
own land, animals and crops. 
Conclusions: building a new co-operative community 
These problems are widespread all over the county. Given the reality 
of the socioeconomic conditions in Tanzania we argue for strengthening 
the rural co-operative so that even a minimal degree of social 
protection for the majority can be guaranteed. 
The antithesis of the co-operative is the crop merchant. Quite often 
these are large companies from Western Europe but there are also local 
crop merchants, some of whom are involved in processing peasant 
crops like cotton and cashew nuts. The way that structural adjustment 
and liberalization was introduced in Tanzania left little or no protection 
to the peasant co-operative. Subsidies (such as those on fertilizers) 
for extremely poor people have been attacked by the World Bank and 
IMF. The strength of co-operatives in Tanzania has been reduced by 
policies like structural adjustment. It is not clear why the government 
cannot provide some systematic and long term protection to peasants' 
co-operatives when governments in Western Europe, North America 
and Asia protect their farmers and offer them subsidies. When one of 
the authors of this paper visited co-operatives in Japan in 1998, he 
42 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
witnessed a very strong co-operative movement where small-scale 
farmers and fishermen thrive. Ordinary co-operative members and 
leaders there said that government financial support was crucial to their 
success and pointed out that loans given to co-operatives by government 
through local government were charged at a much lower interest 
rates than those charged by financial institutions. 
Crops merchants, unlike co-operatives, are not bound to treat 
peasants well when it comes to crop prices and other business matters, 
so the strong arm of capital kills the poor peasant and forces betteroff 
peasants to organize against their poor neighbours. The security 
and even survival of poor peasants is at risk, as they cannot singlehandedly 
confront crop merchants who are backed by industrial capital 
and financial institutions within and outside the country. 
The peasants feel that they have been let down when it comes to cooperative 
policy. Peasants worked really hard to build co-operatives 
and they had high hopes of them. The decline of co-operatives has 
resulted in the decline of farming and thus a lower output. As a result 
peasants are unable to build strong institutions without help. 
Co-operatives have a great potential in breaking up the old division 
of labour which oppresses women and young people. Thus they can help 
to liberate the whole nation if they are given the necessary organizational 
and financial support. The government needs to formulate clear 
and long-term policies on co-operatives and peasants and others must 
be confident thaf, if they build strong co-operatives, they will be given 
legal protection. The abolition of co-operatives must end and the 
authorities may even be persuaded to accept the idea that the abolition 
of co-operatives in the 1970s was wrong. 
Co-operatives have a managerial and organizational advantage 
because they can be flexible. They can be formed at community level 
with people who know each other well. The Ministry of Co-operatives 
and Marketing can help to remould all rural (peasant) co-operatives, to 
incorporate a savings and credit scheme. One of the greatest problems 
for Tanzanians is the lack of credit to purchase machinery, seeds and 
other inputs and to develop their farms. Co-operatives can be directed 
to undertake social innovations and better-planned agricultural inno- 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 43
vations like the wider use of high-yielding varieties of seeds (HYVS), 
irrigation and fertilizers. Co-operatives can thus help bring a social and 
economic revolution in Tanzania through an agrarian revolution. This 
will make them strong instruments of social protection for peasant 
households. 
The government must promote the co-operative as a movement and 
as a community, if it is to be successful (Melnyk, 1985:135-136). The cooperative 
community can serve as the intermediate stage between 
individuals and families on the one hand and the government and wider 
society on the other. As they have the organizational and managerial 
advantage of being small and consisting of members who know each 
other well, this will help in building the trust to undertake both small 
and, later, (when there is experience and capital) large, social and 
economic ventures. 
One scholar has identified barriers to the evolution of local classes 
of capable entrepreneurs in Africa (Kennedy 1994:191). Among these 
barriers are obstructive and ineffective government policies and the 
competitive threat from foreign-capital, endowed with considerable 
advantages not available to local entrepreneurs. After the 1991 Cooperative 
Act social reformers and co-operative members and potential 
members expect legal reform which clearly favours the small peasant. 
44 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Phase II: Mutual Aid 
ABSTRACT 
The paper investigates the role of mutual aid in the provision of 
social protection. Methodological triangulation involving survey 
interviews, focus group discussions, in-depth interviews, documentary 
reviews and participant observations were used to 
collect data. Findings show that mutual aid groups of different 
types have existed in different epochs during periods of peace, 
prosperity and even during calamities. Some mutual aid groups are 
organized on gender lines. These differences are also observed at 
the level of age groups, educational attainment, marital status and 
religious affiliation. Mutual aid groups in rural areas cover almost 
all kinds of contingencies while those in urban areas cover specified 
risks. Although mutual aid groups provide short-term protection 
against soc4al and economic risks, their capacity for protecting 
members against long-term risks is impaired. The paper concludes 
by making suggestions for improving mutual aid groups so that they 
may provide adequate protection against all risks as well as for 
removing the inequities inherent in the system. 
Introduction 
For well over a decade there has been a growing interest among 
scholars on the topic of mutual aid (Wasserman and Faust, 1994). This 
interest has been driven by the importance of mutual aid in providing 
multiple sources of support to individuals. Some studies in a number 
of socio-cultural settings throughout the world show that mutual aid 
ensures the chances of survival for individuals as well as increasing 
the amount of welfare and enjoyment of life with the least waste of 
energy (Kropotkin 1902, House 1981). Others show that mutual aid 
based on neighbourhoods can best handle immediate emergencies 
(Murdock 1949, Sussman and Slater 1963, Sussman 1965, Litwark and 
Szelenyi 1969). 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 45
In addition, studies by Lomnitz(1968), Aldous (1969) and Wasserman 
and Faust (1994) show that members of mutual aid groups view their 
actions as interdependent. Each individual in the mutual aid group has 
ties to other individuals, each of whom in turn is tied to other people. 
Moreover, the relational ties or linkages between actors are viewed as 
channels for the transfer or flow of resources (material or nonmaterial). 
The linkages between individuals are due to their perceived 
vulnerability to multiple risks from different sources such as illness, old 
age, death and bad weather. These shocks hit individuals in an unpredictable 
manner or cannot be prevented. As a result, individuals take 
precautionary strategies such as creating informal exchange-based risksharing 
mechanisms through extended families, mutual gift-giving and 
crop-sharing arrangements. 
In societies characterized by a great incidence of risks and severe 
deprivation, a low degree of development of formal systems of social 
protection, the limited coverage of public support, resource constraints 
and the relative powerlessness of the poor and vulnerable, such as 
Tanzania, mutual aid is of great importance for the survival of members. 
Indeed, Burgess and Stern (1991) view mutual aid in developing 
countries as an indispensable mechanism for delivering support to the 
poor and vulnerable. These arrangements are backed by value systems 
and moral codes that emphasize the need for co-operation among 
people living or coming from the same community (Platteau 1991). 
Literature on this topic with particular reference to the relationships 
between mutual aid and the provision of social protection in Tanzania 
is generally lacking. The few studies that deal with mutual aid have 
generally been limited to the definition, prevalence or characterization 
of mutual aid groups (Lugalla and Mbwambo 1999). Moreover, the 
nature of mutual aid groups, membership, mutual aid structures, their 
governance, the contingencies covered and the benefits offered to 
members, together with the extent of support given to members, remain 
to be specified. The goal of this paper is to illuminate these features to 
show that mutual aid arrangements have been providing much-needed 
social protection against social and economic distress for members of 
46 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
the community. Before going into detail, methodological issues will be 
discussed in the next section. 
Methodology and study areas 
This is a descriptive study carried out in the month of February 2001. 
The purpose of using this study design was to get information-rich 
cases in study communities but, at the same time, through focus group 
discussions and in-depth interviews, explore and confirm information 
on mutual aid activities. Secondly, it was intended to enable the research 
team to understand the whole socioeconomic and cultural picture of 
the individual members in various mutual aid groups under investigation. 
Thirdly, the research team wanted to understand how various 
mutual aid activities provide social protection to members in those 
study areas. 
The study was carried out in Mwanza, Dodoma, Dar es Salaam and 
Kilimanjaro. The reasons for selecting these study sites included 
portraying the existence of people engaged in different economic 
activities namely, cattle-herding, farming and fishing and with different 
kinds of mutual aid activities. For example, Dodoma was selected 
because of its central position in the country and the existence of 
pastoralists and farmers and the area experiences drought in some 
years. Mwanza has farmers, pastoralists and fishermen, besides being 
the second largest city in the country and the centre of the Lake Zone 
region. Kilimanjaro was selected because of the existence of farmers 
and strong mutual aid activities. Dar es Salaam was chosen because it 
is the largest city in the country, with various types of mutual aid 
activities and inhabitants of various ethnical backgrounds. 
The methods of data collection used included structured and semistructured 
interviews, in-depth interviews, focus group discussions, 
participant observation and documentary reviews. Survey interviews 
were administered to a total of 200 respondents selected by purposive 
sampling. The research team initially visited village chairpersons and 
introduced themselves and the purpose of the study. The team asked 
the village chairperson to introduce the team to members of the com- 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 47
munity. After establishing a rapport with the community, adults aged 
18 years or above who were found to have mutual aid experience were 
asked to participate in the study. 
The study focuses on the experience of people in mutual aid activities. 
A total of four focus group discussions (FGDs) were held in 
each study site. FGDs were divided in two groups, namely, men and 
women. Each of these groups was subdivided into groups of young 
people (18-35 years), adults (36-59 years) and old people (60 years 
or above). The size of FGDs ranged from 7 to 12 participants. The 
criteria for selecting participants in the FGDs included their membership 
in a mutual aid group or activity, their residence in the study site and 
their ability and willingness to participate in the discussion. Discussions 
in the FGDs lasted between one and half and two hours. 
Characteristics of respondents 
A total of 200 respondents were interviewed. Out of these, slightly over 
half (54%) were men. Two in five (40%) of the respondents were 25- 
34 years old while nearly a half (49%) were 45-54 years. Respondents 
who were aged 15-24 years constituted 6.5% of the total study 
population and only 4.5% consisted of people aged 55-64 years. The 
education level of respondents varied considerably. Over half (54%) 
of the respondents had primary school education and slightly over a 
third (38%) had secondary education. Very few (5.5%) of the respondents 
had adult education and the remaining 1% had post-secondary 
education. Nearly three quarters (73%) of the respondents were 
married while 13% were single, 13% were widowed and 1% were 
divorced. Slightly over half (50.2%) of the respondents were Christians, 
48.3% were Moslems and the remaining 1.5%% were traditional. 
The nature of mutual aid groups (kiwili) 
According to Kropotkin (1902) and Smith (1999), mutual aid of 
various kinds have been practised in different epochs during periods 
of peace, prosperity and even during periods of calamities. In modern 
times it is common to find people jointly constructing roads and canals, 
building health facilities and schools. Findings from all four study areas 
48 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Table i: Membership size of mutual aid groups 
Nos. of people in your mutual 
61 + 
30-60 
11-29 
5-10 
For % of respondents 
aid group 
20 
40 
20 
20 
indicate that some of the mutual aid groups have been in existence 
long before the participants in the study were born. For instance, 
Bacheye, Basumba batale and Bagalu groups in Mwanza have been 
operating in the study areas for some time, as reported by Mzee 
Mabula who said, "My grandfather belonged to a Bacheye group when 
I was born in 1939. He remained a member until he died five years ago." 
In Kilimanjaro, particularly in Same and Mwanga, similar mutual aid 
groups called makamba exist. Other mutual aid groups are of recent 
origin. The Wacheza Mpira group, for example, was established in 1995 
while Umoja wa Kandaya Ziwa, Dodoma was established in December 
1998. Similarly, Umoja wa Wanakaskazini, Dodoma (UWAKA) was 
established in 1998. All the above testify that mutual aid was found aO 
over the study areas and this is still the case in these areas today. 
The size of mutual aid groups varies greatly. The smallest consists 
of five or less people and the largest has 100 members. In some parts 
of Same District such as Mpinji, Kirangare, Bwambo and Mhero, for 
example, about 80% of all households belong to a mutual aid society. 
The typical mutual aid society, known as kiwili, consists of three male 
heads of household and their wives (see Table I). 
Discussions in various focus group discussions indicated that 
medium-sized mutual aid groups are preferred for two reasons. First, 
given the technology used and problems in organizing labour, a large 
unit will hinder the efficient perfprmance of tasks. Secondly, it is 
easier to manage and control small or medium-sized groups. According 
to Tischler (1994) large numbers threaten the nature of the group. In 
addition, there are always fears tljat the larger the size, the greater the 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 49
chance of resistance to socialization in group norms by some of the 
members, thereby undermining the group's solidarity. This is disastrous 
to social protection that is hinged upon mutuality and trust. 
Findings from survey interviews in Mwanza, Dodoma, Dar es Salaam 
and Kilimanjaro indicate that eight different types of mutual aid groups 
existed, namely, those dealing with farming, livestock-keeping, 
business, house-building, carpentry, cooking, wedding and funeral 
activities. Nearly a third (31%) were involved in activities related to 
funerals; 18% to weddings, 24.4% to business while 16% to farming 
activities. Most of these groups were confined to rural areas. 
Women predominated in mutual aid groups dealing with business 
and cooking. Men predominated involved in activities considered to 
be physically demanding, such as farming. In study sites where 
pastoralism is the main economic activity, all the respondents involved 
in herding livestock were men. In groups involving house-building, 3% 
of the respondents were men and only 1.6% were women. Nineteen per 
cent of respondents dealing with wedding groups were men, while 
16.5% were women. Men constituted 32% of respondents in groups 
dealing with funerals while women constituted 29.7%. All this can be 
explained by gendered division of labour, although this cannot be 
generalized to all mutual aid activities. In the case of kiwili, for example, 
both men and women participate in building stone terraces in Mamba 
and Chome areas in Same District. Women collect stones and help in 
moving soil as well as cook for members of the mutual aid group, men 
break the stones and build the terraces. 
When mutual aid groupings are analysed in relation to age groups, 
a different picture emerges. Respondents aged 15-24 predominate in 
groups dealing with business (65%), while 27.8% of the respondents 
aged 55-64 years predominate in groups dealing with farming. The level 
of education attainment may explain the differences observed. Most 
respondents aged 15-24 years have had primary school education and 
are therefore better able to deal with business activities. Respondents 
aged 55-64 years are seldom literate and therefore able to deal mainly 
with farming. Respondents aged 45-54 years predominate in group 
dealing with wedding ceremonies. This can be explained by the fact 
50 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
that this age-group has sons and daughters of marriageable age and 
therefore have to participate more than other age groups in wedding 
ceremonies. Respondents aged 25-34 years and those aged 55-64 
years predominate in funerals groups. Energetic young people have to 
dig the graves and perform the funeral ceremonies while elderly people 
have to attend. In areas such as Mwanza old people are associated with 
witchcraft. If they fail to attend funerals they are suspected of bewitching 
the deceased person. 
The education levels of respondents seems to influence their involvement 
in mutual aid group activities. Respondents with no education 
tend to be involved in group activities related to agriculture (37.5%), 
business (38%) and funerals (37.5%), while those with post-secondary 
education level tend to be involved in activities dealing with weddings 
more than in other mutual aid group activities. 
The religious affiliations of the respondents also influences the 
types of mutual aid activities they participate in. Respondents who are 
traditional are involved more in agriculture (38%), weddings (24%) 
and funerals (38%) than in other activities. Christians are involved in 
business (27.6%) and Moslems in carpentry (3.6%) and cooking 
(4.6%). The marital status of the respondents also influences the types 
of mutual aid activities they participate in. Single respondents tend to 
engage in agriculture (19%) and business (31%) and married respondents 
in weddings (18.5%) and funerals (33%). 
Reasons for participating 
Members explain that they participate in group activities in the 
expectation that, in the future, they may also find themselves in 
difficulty and in need of rescue. As noted by a participant in a FGD 
with male adults in Mwanza, 
It is important to help each other, particularly when your 
colleagues are in difficulties. This is because I, myself, might be 
faced with a similar problem in future. If I am they will also assist 
me or bail me out. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 51
Other motives mentioned in the study sites include emotional comfort, 
showing cohesion or integration and avoiding social exclusion. 
Membership in a mutual aid group 
The process of becoming a member in a mutual aid group is not hard. 
Members make contact with someone who is already a member. 
Findings from Mwanza, Dodoma, Dar es Salaam and Kilimanjaro 
revealed that 75% of the respondents knew somebody who was already 
a member in a mutual aid group. These included relatives such as 
sisters, uncles, brothers and aunts; friends and neighbours. 
Over half the respondents said their mutual aid societies had no 
newly-recruited members and only slightly over two in five said that 
their mutual aid group actively recruited new members. Findings from 
Mwanza and Dodoma indicated that there are regulations governing the 
admission of new members in the mutual aid groups. Potential members 
normally approach existing members and get acquainted with other 
members, the group's activities and the terms of reference for membership. 
Thereafter the potential members submit an oral or written 
(especially for burial societies) application for membership. The applicant 
must have a referee who also acts as a guarantor from a member 
within the mutual aid group. The mutual aid group leaders consider the 
application. If found suitable, the potential members will be admitted 
into the group. Potential members must also possess their own capital. 
The guarantor acts as a character reference. 
Apart from considering the behaviour of the potential recruit, other 
qualities examined include: the person's attitude to work, nonquarrelsome 
behaviour, a sound mind and the ability to keep secrets. 
It goes without saying that in this kind of co-operation members will 
always avoid the less motivated and problematic person. In addition, 
in urban-based mutual aid groups new members must pay membership 
fees and regular members must pay monthly premiums. For example, 
Umoja wa Kanda ya Ziwa, Dodoma has imposed a membership fee 
of Tshs. 3000 payable before admission to the group. In addition, a 
monthly contribution of Tshs. 1000 is required. Umoja wa Wanakaskazini, 
Dodoma admits people from Kilimanjaro and Arusha regions 
52 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
only. Widowed women from other regions whose husbands come from 
Kilimanjaro and Arusha regions can be admitted and women from these 
regions who are married to husbands from other regions are also 
allowed to join. In addition, a member pays admission fees and annual 
contributions and must contribute money whenever burials occur. A 
penalty is imposed or membership is frozen if this rule is broken. 
The structure and governance of mutual aid groups 
Evidence from the study areas suggests that mutual aid groups consist 
mainly of close friends, relatives and acquaintances coming from the 
same community. They are people who know each other well and who 
have common interests. In rural areas, these people have known each 
other since childhood. The groups display a feeling of solidarity and 
group identity, of intimate relationship, together with the wish of 
members to keep in close contact with one another and stay with one 
another. 
There are two types of leadership in mutual aid groups. The first are 
leaders who participate in group activities. Ninety-two per cent of the 
respondents said their leaders occupied the top place in the hierarchy 
of their group. Eighty-four per cent of farmers, 57% of livestockkeeping 
groups, 47% of housing construction groups and all business, 
road construction, wedding and funeral groups were constituted in this 
way. Most (88%) of the respondents said their leadership is permanent. 
Of these, 50% said their mutual aid group has one leader: the remainder 
have two leaders. When asked how the leaders of their mutual aid 
groups came into power, 54% said they were already in office when 
they joined these groups. The remaining 21% said their leaders were 
elected. Participants in a FGD with men in Nyanghomangho Village, 
Mwanza Rural District, also said that farming mutual aid groups 
elected their leaders. At the apex there is Malamala, below whom is 
the King. Next is the Mtemi then Mtwale and Lunashi or messager. 
When there is farming activities, Mtwale informs the Mtemi and sends 
the Lunashi to inform members of the mutual aid group. 
The second category of mutual aid groups consists of those in 
which all members of the group were equal, with temporary leaders. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 53
Leadership is bestowed on a member for a particular activity for the 
course of that task. About 8% of the respondents in the study areas 
mentioned this kind of mutual aid group. 
Mutual aid groups have in-built democratic procedures based on 
shared community values. As noted by a participant in a FGD with 
women aged 18-34 years in Buhongwa Village, 
When there is an issue, all members of the group are informed of 
the meeting and the time and the venue of the meeting. They all 
discuss and deliberate the issue until they come to a consensus. 
Once the decision is made, it binds all members. 
Survey interview data shows that these members participate in the 
planning and implementation of group activities. All respondents in 
farming groups participate in the planning and implementation of their 
group activities, as do all members in livestock-keeping groups, funeral, 
wedding, housing and road construction groups and 71% of those in 
business groups. When asked how members participate in governance 
and administration, 42% said they participated in planning how to 
execute the work, 58% said in decision-making and 100% said in attending 
meetings. 
Constraints: relationship with local government 
Despite the seemingly good governance at grassroots level, there are 
a number of constraints in the relationship between mutualaid groups 
and local government. According to the Human Development Report, 
Shinyanga Region, Tanzania (1998), the relationships between government 
institutions and other institutions at the village level vary. 
Whereas in some villages traditional institutions have been incorporated 
into the formal structure of village government, in others they have 
not. In some cases, the government is reported to have undermined the 
operations of traditional institutions. In some cases communities are 
satisfied with some formal and informal institutions, but in others they 
are not. 
The Shinyanga Human Development Report reveals that existing 
institutional arrangements constrain the potential for good governance 
54 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
at village level. Government structures are complex and overlap in 
functions. The government is characterized by excessive red tape and 
cumbersome procedures at all level of social service delivery. Indeed, 
existing structures do not work to promote self-governance at grassroots 
level. The representatives of the people are unable to fulfil their 
objectives promptly or effectively simply because the issues discussed 
during full council meetings do not reflect the needs of people. 
Contingencies covered 
Kropotkin (1902) argues that the purpose of mutual aid is to ensure the 
maintenance and further improvement of the welfare of their members. 
Individuals who adhere to the principle of mutual aid have the best 
chances of survival. Findings from the study sites show that a great 
number of mutual aid groups or associations are formed, covering 
many contingencies. Indeed, mutual aid in Tanzania, like in other developing 
countries, is part of daily life. For example, the main purpose of 
Umoja wa Kanda ya Ziwa, Dodoma mutual aid society is to help 
members whenever they are faced with a death and during funerals. 
The purposes of Umoja wa Wanakaskazini, Dodoma are broader and 
include strengthening co-operation among members, particularly in 
matters concerning death, funerals, burials, illness and social services. 
This society has even broader objectives, such as protecting and 
preserving the rights of members and ensuring that members have 
gainful employment. Indeed, these broad objectives of Umoja wa 
Wanakaskazini, Dodoma appear in their constitution. 
The objectives of most mutual aid groups(83.9%) and societies in 
the four study areas centre around helping members in times of 
hardship as well as in times of celebration. In case of hardship, 
members come to the rescue of any other member who has fallen into 
distress owing to an event considered to be a misfortune such as illness 
or accident, death, crop damage by wild animals, destruction of a house 
or fields by floods or fire. The labour or any other kind of help 
provided under the scheme does not give rise to formal and immediate 
payment. Instead, assistance is offered against the promise of 
reciprocal help at any future indeterminate time. 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 55
Getting loans through mutual aid groups is cited as another objective 
(10.7%) and so is the include provision of credit and savings (2.6%) 
and ensuring the success of the activity (2.8%). 
The objectives of groups are related to the types of contingencies 
covered. Discussions in FGDs with men in Buhongwa Village, Mwanza 
Rural District, indicated that they covered contingencies such as court 
fines, funerals, death and burials. One participant said: 
If cattle belonging-to a member of our group are stolen the 
responsibility of our mutual aid group will be to sit down and 
discuss how we can help by contributing either money or cattle. 
In cases where a member needs to have his or her farm cultivated, 
the group will jointly do so. Also we help members who 
have lost their dear relatives, sons or daughters through contributing 
money, food and all activities related to burials. 
Other mutual aid groups and societies cover similar contingencies, 
including transporting the deceased person to his or her place of birth 
and helping with funeral arrangements. 
Mutual aid groups are dynamic in the types of contingencies 
covered. Whereas they covered funerals, illnesses and weddings in the 
past, currently mutual aid groups focus on providing credit and savings 
services to members and help people threatened by hunger and funerals. 
Benefits offered to members 
Findings from survey interviews in the study areas indicate that four 
main types of benefit are offered to members. These include food, 
capital, help with funerals and getting help at any time. Analysis of the 
types of benefits by gender indicates that provision of food was 
mentioned mostly by male respondents (36.7%). More female 
respondents (30. 1%) than men (14.3%) cited the provision of capital 
as a benefit and it was mainly men who mentioned help with funerals. 
In the case of the death of a member or the close relative of a 
member, financial benefits and assistance with all the necessary 
preparations for burial ceremonies in the form of labour services such 
as cooking, buying coffins and psychological assistance are provided, 
56 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
as noted by a participant in one of the FGDs conducted in Dodoma: 
"We stay with those who have lost a relative so as to console them for 
three days while at the same time helping them with their domestic 
activities". 
Data from U WAKA on the financial benefits offered to members faced 
with a death in their family show that between 1998 and 1999 a total of 
43 deaths occurred. UWAKA contributes money as burial expenses 
ranging from Tshs 45,900 to Tshs 670,000 depending on whether the 
deceased is a baby or an adult. 
Members of various mutual aid groups and societies get materia 
and non-material assistance, such as helping sick members or thei 
relatives with transport to health facilities, visiting a sick person whc 
is hospitalized and giving food. The amount of benefits provided varies 
according to the magnitude of the crisis. As pointed out by one participant 
in a FGD with men in Nyanghomangho Village, Mwanza Rural 
District: 
We provide assistance to a member according to the type of risk 
or amount requested by the victim. For example, if a member is 
fined Tshs, 20,000 by a court, the 40 members of the mutual aid 
group will be asked to contribute Tshs. 500 each. If he is 
charged Tshs. 100,000 each member will be required to contribute 
Tshs. 2,500. 
It is evident from the above that some members, particularly the richer 
peasants, can get more benefits than others. Money accrued from 
various group activities is also spent on purchasing farming equipment 
to be distributed to members. 
The capacity of mutual aid to provide social protection 
In this section an attempt is made to assess the capacity of mutual aid 
to provide adequate.social protection to members. Available evidence 
from in-depth interviews and FGDs show that kiwili type of mutual aid 
group has helped in the past ten years to bring about agrarian change 
in the mountains of Same District. Single households without the 
capital to hire labour cannot build stone terraces, or do so very 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 57
slowly. Farmers in Mamba and Chome areas, Same District, however, 
know that these terraces, which retajn higher moisture and manure, are 
very productive in this hilly terrain. The crop output and incomes have 
improved in the past ten years for households that have been able to 
build stone terraces through kiwili mutual aid societies. Peasants have 
increased their incomes by growing high-value crops like onions, Irish 
potatoes, ginger, wheat and beans as well as other crops, like maize and 
cabbage, on their stone terraces. They have thus helped to increase 
food security in these areas. This directly addresses the contingency 
of famine, which is a constant menace to poor peasants in Tanzania. 
FGDs with men in Nyanghomangho Village, Mwanza Rural District 
and Mvumi Dodoma show that mutual aid support is adequate insofar 
as a member does not go hungry because of lack of food. Sixty-six per 
cent of respondents, consisting of 71.2% of the men and 61.4% of the 
women, said the benefits accrued from this aid was adequate. The 
reasons given for finding the benefit inadequate included the view that 
many social problems still needed to be addressed (53%) and that it 
depended on the weather (47%). During bad weather, members are 
likely to be affected by famine. 
Various studies have shown that mutual aid helps to build strong and 
cohesive communities, fosters trust between citizens and helps develop 
norms of solidarity and reciprocity that are essential to Stable communities. 
Our findings lend support to these propositions by showing that 
participants who were involved in mutual aid groups have developed 
solidarity and a group identity. They were ready to help each other in 
times of hardships as well as in times of happiness. 
Conclusion 
The study has shown that mutual aid still plays a big part in Tanzania. 
Many people are involved in mutual aid activities of various types. They 
staunchly maintain their mutual aid activities and endeavour to 
reconstitute them where they have ceased to exist. Small and mediumsized 
mutual aid groups predominate due to the resulting efficiency 
and the ease of managing and controlling the groups. 
58 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Members of mutual aid groups make joint decisions after long deliberations 
in matters of mutual interest, including issues related to the 
social protection of members. Despite the seemingly good governance 
at grassroots level, the study discerned constraints in the relationship 
between mutual aid groups and local government institutions. As far 
as contingencies are concerned, the study shows that, with the exception 
of urban-based mutual aid groups, all risks are covered. The 
benefits offered to members vary according to the type of risk and 
amount of benefit. Mutual aid activities have enabled members to 
survive various hardships, especially lack of food, clothes and shelter. 
They have a better chances of survival than non-members. 
The above notwithstanding, the efforts embodied in mutual aid are 
of little importance simply because they cannot substantially and still 
less permanently, alleviate the misery of most the people in Tanzania 
who are in poverty. If these groups, however, were given the necessary 
organizational and financial support they could provide much-needed 
long-term social protection of members against all social and economic 
contingencies. 
As shown in the paper, mutual aid groups have a managerial and 
organizational advantage, simply because members know each other 
well and trust each other. They need only to be encouraged and guided 
as to how best to organize community-based social security schemes, 
using their own organizational structures. This requires, among other 
things, the formulation of a national social security policy. In this way, 
all community members can be attracted and protected against various 
social and economic contingencies both short-term and long-term. 
References 
Awiti, A. 1973. Economic differentiation in Ismani, Iringa Region. The 
African Review 3 2 209-239 
Bagachwa, M.S.D. 1994. Poverty Alleviation in Tanzania: Recent 
Research Issues, Dar es Salaam University Press 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 59
Bakari, A.H. 1988. Traditional social security as practised in contemporary 
Tanzania's urban centres. In: Between Kinship and the State: 
Social Security and Law in Developing Countries, Benda-Beckmann, 
Von F. et al. (eds.), Dordrecht, Foris Publications 
Bank of Tanzania 1984. Twenty Years of Independence 1961-1981: A 
Review of Political and Economic Performance, Dar es Salaam 
Bantukari, J.A. 2000. Co-operatives and Poverty Alleviation, Dar es 
Salaam, Team Publishers 
Coulson, A. 1982. Tanzania: A Political Economy, Oxford, Clarendon 
Press 
Green, R.H. 1977. Tanzanian political economy goals, strategies and 
results, 1967-74: notes towards an interim assessment. Towards Socialism 
in Tanzania, Mwansasu B.U. and Pratt C. (eds.), University of 
Toronto Press, Toronto 
Holzmann, R. and Jorgensen, S. 2000. Social Risk Management: A 
New Conceptual Frameworkfor Social Protection and Beyond, Social 
Protection Discussion Paper No.006 
ILO 1982. Basic Needs in Danger: A Basic Need Oriented Development 
Strategy for Tanzania. ILO, Geneva 
Jazairy et al., 1992. The State of the World Rural Poverty: An Inquiry 
into its causes and consequences. London, International Fund for 
Agricultural Development 
Kennedy P. 1994. Political barriers to African capitalism. The Journal of 
Modern African Studies, 32 2 
Kropotkin, P. 1902. Mutual Aid: A Factor of Evolution, London, 
Pelican Books 
Litwark, E. and Szelenyi, I. 1969. Primary group structures and their 
functions: kin, neighbours and friends. American Sociological Review, 
34:465 
Mallya, W.J. 1994. The role of NGOs in providing social welfare and 
social security for women in Tanzania. In Gender and Social Policy: 
Experiences of Tanzania, Klemp L. (ed.), Dar es Salaam, Friedrich 
Ebert Stiftung 
60 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
Mataba, M.S.L. 1983. The Social and Economic Consequences of Old 
Age and Compulsory Retirement: Retired People 1982-83 and the 
Retiring People, B.A. Dissertation, University of Dar es Salaam 
Melnyk, G. (1985), The Search For Community: From Utopia to a Cooperative 
Society, Montreal & Buffalo, Black Rose Books Mlyansi, P. 
1991. Social Security in Tanzania: An Analysis of the Conditions 
Influencing the Plight of Teachers in Arusha and Dar es Salaam, M. A. 
Dissertation, University of Dar es Salaam 
Msoka, A.C. 1998. Cost sharing and patients' response to treatment in 
an urban district hospital. Proceedings of the Seventeenth Annual 
Scientific Conference of the Tanzania Public Health Association, 
November 23-26,1998 held at Mkonge Hotel, Tanga 
Mtatifikolo, F. 1994. Implications of public policies on poverty and 
poverty alleviation: the case of Tanzania. In Bagachwa, M.S.D (ed.), 
Poverty Alleviation in Tanzania: Recent Research Issues, Dar es 
Salaam University Press, 91-122 
Murdock, G. 1949. Social Structure, New York, Macmillan 
Ndulu, B.J. 1982. Unequal regional distribution of economic opportunities 
in Tanzania and affirmative policy efforts towards 
•equalization. ERB Paper 82 8, Dar es Salaam 
Omari, C.K. 1994. The gender dimension of public action for social 
security and welfare. In Klemp L. (ed.), Gender and Social Policy: 
Experiences of Tanzania, Dar es Salaam, Friedrich Ebert Stiftung 
Sender, J. and Smith, S. 1990. Poverty, Class and Gender in Rural 
Africa: A Tanzanian Case Study, London, Routledge 
Pahl R.E. ed., 1988. On Work: Historical, Comparative and Theoretical 
Approaches, University Press of Africa (UPA) (1968) (for Ministry of 
Information and Tourism United Republic of Tanzania), Tanzania 
Today, Nairobi. 
Sussman, M.B. 1965. Relationships of adult children with their parents 
in the United States. In Shanas and Streib (eds.), Social Structure and 
the Family: Generational Relations, New Jersey, Prentice-Hall 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 61
Sussman, M.B. and Slater, S.B. 1963. Re-appraisal of Urban Kin Networks: 
Empirical Evidence, Presented at meeting of the American 
Sociological Association, Los Angeles 
Tibandebage, P. and Mackintosh, M. 2000. Access, Exclusion and 
Quality of Care in Dar es Salaam, Mbeya and Coast Region. Research 
report supported by DFID 
Tinios et al., 1993. Household Consumption and Poverty in Tanzania: 
results from the 1991 National Cornell-ERB Survey. Paper prepared 
under the auspices of the CFNPP, Africa Economic Policies Project 
supported by USAID. 
Tischler, H.L. 1996. Introduction to Sociology, Forth Worth, Harcourt 
Press 
Tungaraza, F.S.K. 1988. Notwendigkeit und Moglichkeit von Sozialpolitik 
in Tanzania. Unpublished Ph.D Thesis, University of Augsburg 
Tungaraza, F.S.K. 1993. Social networks and social care in Tanzania. 
Social Policy & Administration, 27 2:141-150 
Tungaraza, F.S.K. 1994. Changes in traditional social security systems 
and their impact on women in Tanzania. In Klemp L. (ed.), Gender and 
Social Policy: Experiences of Tanzania, Dar es Salaam, Friedrich Ebert 
Stiftung 
Tungaraza, F.S.K. 1995. Access to health care among women in 
Tanzania. Proceedings of the 14th Annual Scientific Conference of 
the Tanzania Public Health Association TPHA held at the KCMC, 
Moshi, Tanzania, October 9-13, 52-71 
Tungaraza, F.S.K. 1998. Shinyanga Region Human Development 
Report: Social Sector Study. Report for the UNDP under the technical 
support of the IDS, Sussex 
Tungaraza, F.S.K. 2000a. Older People, Social Security and Alternative 
Government-Community Partnership Support Mechanisms in 
Tanzania. Paper presented at the National Ageing and Older People 
Policy Workshop, Dar es Salaam 
Tungaraza, F.S.K. 2000b. Networking among street children as a strategy 
for surviving and coping up with urban life: a case study of Kisutu- 
Gerezani Streets, Dar es Salaam. In Lugalla J.L.P. and Kibassa C.G. 
(eds.), Poverty and Street Children in East Africa: Coping Strategies 
62 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 17 NO 2 JULY 2002
Mchomvu, Tungazara and Maghimbi/TANZANIA REPORT 
and Street Children's Health, Hamburg/Muenster, Lit Verlag, forthcoming 
Tungaraza, F.S.K. and Mapunda, G. 2000. The National Social Security 
Policy, Policy Paper. Dar es Salaam. 
United Republic of Tanzania 1999. Poverty and Welfare Monitoring 
Indicators. Vice President's Office, Dar es Salaam 
Urassa, J.A.E. 2000. The Intergenerational Social Protection Bonds 
among the Chagga of Kilimanjaro. Unpublished B.A. Dissertation. 
University of Dar es Salaam 
Victor, P. 1997. Informal Social Security Systems and Poverty Alleviation 
among the Elderly in Contemporary Tanzania: A Case Study 
ofMorogoro Urban District. Unpublished B.A Dissertation, University 
of Dares Salaam 
Wangwe, S.M. and Tibandebage, P. 1999. Towards Social Security 
Policy and Reform in Tanzania. Paper presented at the National Social 
Security Fund Symposium, Dar es Salaam. 
Warioba, R.1998. Accessibility to health services and user fees: A focus 
on adolescents in Dodoma, Tanzania. Proceedings of the 17th. Annual 
Scientific Conference of the Tanzania Public Health Association, 
November 23-26, 1998 held at Mkonge Hotel, Tanga 
VOL 17 NO 2 JULY 2002 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 63