Determinants of Rural Household Saving I

International Journal of Research in Social Sciences
Vol. 6 Issue 11, November 2016,

ISSN: 2249-2496 Impact Factor: 6.278
Journal Homepage: http://www.ijmra.us, Email: [email protected]
Double-Blind Peer Reviewed Refereed Open Access International Journal - Included in the International Serial
Directories Indexed & Listed at: Ulrich's Periodicals Directory ©, U.S.A., Open J-Gage as well as in Cabell’s
Directories of Publishing Opportunities, U.S.A
Determinants of Rural Household Saving in

Tanzania
Dr FelicianMutasa*
Isaack Michael Mchumi **
Abstract
This paper examines saving motives of rural household in Tanzania. Descriptive and
econometric method of analysis is used to analyse data from 810 households gathered through
structured questionnaire. Further, it employs logistic regression to determine association between
saving motives and demographic characteristics. Triangulation of data is done through focus
group discussion to verify data collected via the survey questionnaire. Data was collected in
thirteen districtsin quasi-randomizedzones. Accordingly, of the two categories of saving
motives, majority of rural household prefer motives under the livelihood category. As for twelve
motives analysed education motive is top priority saving motive. Further education level is the
only independent variable that increases the logit of the dependent variable that means it
increases the odds of an individual to choose livelihood saving motives. The study found
majority of rural households arerigid towards changing saving motives.
Keywords:Saving motives, livelihood motives, non-livelihood motives
*
Dr FelicianMutasa is Senior Lecturer and Dean Faculty of Arts and Social Sciences, The
Open University of Tanzania, Dar es Salaam, Tanzania.
**
Ph.D. Candidate, Department of Economics,Faculty of Arts and Social Sciences, The
Open University of Tanzania, Dar Es Salaam, Tanzania.
299 International Journal of Research in Social Sciences

http://www.ijmra.us, Email: [email protected]
ISSN: 2249-2496 Impact Factor: 6.278
1. Introduction
Saving behaviour is an old phenomenon. For example, saving matter is written in the book of
Genesis chapter 41 and the Gospel of Saint Mathew chapter 25 in the Holy Bible being scriptures
of over two thousand years ago. As such people including the poor save for a variety of reasons
most of which are aimed at risk management and risk coping. Dercon (1996) and Ravi (2006)
underscored reasons for saving by poor households in two categories: 1) ex-ante protection
against risk and 2) ex-post management of risk and that in the relative absence of complete credit
and insurance markets rural households save to mitigate and coping with risks associated with
income uncertainty.
It is commonly agreed by many studies, most poor rural households in less developed countries
do not have ready access to saving facilities in banks or other formal financial institutions.
Instead, they use alternative, informal vehicles for their saving, such as livestock, gold and other
precious metals, jewellery, and housing materials or other stock of physical goods. They also
maintain cash at home, or may deposit savings with a friend, family members, or moneylender.
Or they may participate in rotating savings and credit associations with trusted family members
or neighbours (Rutherford, 2000; FSDT, 2006 and Nga, 2007). On whether poor save or not, the
study by Rutherford (2000) affirms that there are considerable amount of researches disapprove
the commonly believed view that many rural households in less developed countries are too poor
to save. Saving by rural households has been confirmed by many empirical studies beyond doubt
including those done in Africa like works of Chowa (2012); Mirach and Hailu (2014); Teshome
et al. (2013); Precious and Asrat (2014); and Nigus (2015).
Studies confirm own saving is the most important factor for a country’s investment. According
to Attanazio and Szekely (2000) saving is an alternative means to accumulate assets in the
absence of credit and insurance markets, the capacity to save becomes one of the main vehicles
of social mobility and of enhancing future income-earning possibilities. Horioka (1990),
Wakabayashi and Mackellar (1999); Kitamura et al. (2001); and Upender and Reddy (2007) their
empirical studies expound household saving contributes a lion’s share of gross saving in the
biggest economies such as China, Japan, USA, and India; these countries are among countries
with higher household saving in the world.

According to the World Bank indicators report of 2010 released in 2011, Tanzania national
accounts had gross domestic savings of about 16.85 percent gross domestic product (GDP) of
the year (Gross domestic savings calculated as GDP less final consumption expenditure (total
consumption)). Further, the report indicates that over the past 20 years, the value of gross
domestic saving indicator has fluctuated whereby in 2008 gross domestic saving was 10.31
percent GDP. There is evidence that per capita household saving is on increase alongside rural
financial services growth in Tanzania. For example IFAD (2013) reports about 3000 Rural
Saving Credit and Cooperative Societies (RUSACCOS) in Tanzania with saving per capita
increased to TanzanianShillings 294,397 in 2012. As Dercon (1996) and Ravi (2006)
underscored two main categories of reasons for saving by poor households namely: 1) ex-ante
protection against risk and 2) ex-post management of risk that in the relative absence of complete
credit and insurance markets rural households save to mitigate and coping with risks associated
with income uncertainty.
However, economic theory provides a number of motives which pull or push households to save
including life cycle motives, permanent income motives, bequest motives, precautionary
motives. Study by Dauner (2004) outlines a number of reasons and motives of household saving
including (i) to decrease vulnerability to shocks i.e. income, health, death, etc; (ii) to accumulate
lump sums; (iii) to cater for life-cycle needs i.e birth, wedding, death; (iv) to furnish for
investment in human, physical and social capital; (v) to bequeath relatives and friends; and to
obtain credit. The priority saving motives by rural household differ among countries and even
within a country. Furthermore, saving motives play different role on social and economic aspects
of the household as some motives advance social attributes while other motives promote
economic prospects; the type of motives pursued by household has connotation on poverty
reduction. This paper examines saving motives and relationship between saving motives and
demographic characteristics of rural household in Tanzania.The paper therefore underscores
“what pushes or/and pulls” rural household in saving.

2. Literature
Like many other subjects in economics, modern theories and models of saving and consumption
originated in studies of saving behaviour conducted in western countries, developed countries,
and market economies (De Nardi et al., 2009). Among well-known models of saving in this
context are the Absolute-Income Hypothesis (AIH) of Keynes; Permanent-Income Hypothesis
(PIH) of Friedman; and the Life-Cycle Hypothesis (LCH) of Ando and Modigliani. Similar to
Chowa et al. (2012) this paper presents the underlying theories of household saving in three
perspectives: 1) an individual-oriented perspective; 2) a social perspective; and 3) an institutional
perspective.
(a) Individual-oriented perspectives

Saving theories under an individual perspective include neoclassical economics, economic
psychology and behavioural economics. Neoclassical economic theory assumes that individuals
are rational beings who respond in predictable ways to change incentives. From this perspective,
there are two broad determinants of individual behaviour: opportunities (or constraints) and
individual preferences (Chowa et al., 2012). The starting points for much neoclassical economic
research on saving and asset accumulation have been the Life Cycle Hypothesis (Albert Ando,
Franco Modigliani and Richard Brumberg) and the Permanent Income Hypothesis (Friedman
Milton). Other models of saving developed by economists as an extension of LCH and PIH
economists include buffer-stock models and augmented life cycle models which try to
incorporate bequest motives and precautionary motives.
Economic psychology theory unlike neoclassical economic theory this perspective do not
assume that people behave in a rational manner and have perfect knowledge. The perspective
assumes that personality characteristics and attitudinal variables affect saving and asset
accumulation. Jevons (1965) and Marshal (1961) although they approve neoclassical economic
they also believed that there are various psychological characteristic that influence the temptation
to spend and forego saving. As such there are some established psychological models on savings
behaviour by psychologists including those of Katona (1975); and Olander and Seipel (1970).
For example Katona’s theory of saving (1975) is partly determined by income and partly by
some independent intervening factors. Two important factors are the ability to save and

willingness to save. Ability to save refers to those who can save, whereas willingness to save is
related to the degree of optimism and pessimism of economic conditions (Katona, 1975).
Psychological and sociological theories of saving consider additional determinants of saving and
asset accumulation, including personality characteristics, motives, aspirations, expectations, and
peer and family influences. Some of the propositions emphasize the effects of relatively stable
personality characteristics on asset building. Other psychological and sociological propositions
assume that saving-related preferences and aspirations are not fixed and in fact seek to explain
how motives, aspirations, and expectations are shaped. The propositions that emphasize relative
stable personality characteristics typically come from psychology. For example, psychologists
have examined the effects of thrift, conscientiousness, emotional stability, autonomy,
extraversion, agreeableness, inflexibility, and tough-mindedness on saving. The propositions that
seek to explain how motives, aspirations, expectations, and even preferences are shaped come
from both sociology and psychology.
Behavioural economics perspective does not assume that people are rational and all-knowing. It
integrates insights from psychology and economics. Behavioural economics qualifies some of
the unrealistic assumptions of standard economic models of human behaviour, such as
unbounded rationality, unbounded will power, an unbounded selfishness (Shefrin and Thaler,
1988); Aisle, 1975; Angeletos, et al., 2001; Laibson, 1997; Mullainathan and Thaler, 2000;
Shefrin and Thaler, 1998; Thaler 1981) explain that behavioural economics decisions are
influenced by common human characteristics such as self-control and ability to delay
gratification, mental accounting, use of rule-of-thumb, default options, and hyperbolic
discounting. However, given scant studies in developing countries little is known about the
explanatory powers of these factors on saving behaviour of poor rural income households.
Behavioural theorists have identified a number of common human characteristics that shape
financial behaviour, including lack of self-control (people tend to place too much weight on
current consumption relative to future consumption); limited cognitive abilities (people do not
always learn from their mistakes, and people tend to be overwhelmed by too many choices);

inertia (people tend to continue doing what they are currently doing); the tendency to interpret
default options as advice; and the tendency to use mental accounting techniques.
(b) Sociological perspective

According to Chowa et al (2012) this perspective entails social stratification theory referring
essentially to a distribution of power in society. The divisions in society, based on economically
conditioned power, are called classes, which refer to any group of people that is found in the
same economic situation (D’Souza, 1981; Weber, 1967). Class and social stratification have
strengths in explaining the factors affecting savings behaviour among low-income households
because class relates to the possession (or lack) of resources (economic or otherwise) necessary
for individuals and households to save and build up their assets. Individuals and families in lower
economic classes have limited access to information, resources, and services that can help them
save and accumulate assets over time. This study also found some respondents were saving to
invest in assets accumulation like bicycle and motorcycles in order to acquire higher status-qou
in their families and community. Occupation and education are among important class related
factors which are explained by sociological perspectives.
(c) Institutional perspective

Institutional theory asserts that individuals and households are faced with institutional level
factors that make it impossible or difficult to save. The main hypothesis of institutional theory
assumes that low-income individuals and families are unable to save and accumulate assets
primarily because they do not have the same institutional opportunities that higher income
individuals and households receive (Beverly and Sherraden, 1999; Sherraden, 1991). Institutions
in the institutional theory refer to purposefully-created policies, programs, products, and services
that shape opportunities, constraints and consequences. Seven institutional-level dimensions
have been hypothesized to influence saving and assets accumulation. These dimensions are
access, information, incentives, facilitation, expectations, restrictions, and security (Beverly and
Sherradenet al., 1999; Beverly et.al., 2008; Sherraden and Barr, 2005; Sherradenet al., 2003).

2.2 Empirical literature

There are plenty of empirical studies on household saving most of them done in developed
countries in Europe, North America and Asia. On the contrary very few studies on saving
especially rural household saving have been done in poor countries especially in Africa. Kraay
(2000) identified important saving motivations in China to be precautionary reasons and target
saving motives. Also Wei in 2009 pointed out reasons underlying high household saving in
China as social safety net, low level of financial development, life-cycle motives, culture, habit,
corporate saving, and unintended consequences of social policies. According to Abdelkhaleket
al. (2009) in Morocco saving is used as buffer stock to help households copy with uncertainty of
both income and needs implying that insurance or precautionary reason is prime motive for
saving.
In the United States, Stilglitz (1993) claims rich individuals save a considerable amount, often
more than they need for their own retirement. Similar findings were noted by Lawrence
Summers of Harvard University and Laurence Kotlikoff of Boston University who claim that
wealthy people in United States save relatively larger amount for bequest implying that bequest
motive was more important, but target motive to finance education for students missing
scholarships found to be existing. However, the studies found less need for precautionary saving
in the United States as there is an effective insurance system covering a range of risks facing a
household. Kotlikoff and Summers (1981) obtained surprising finding that intergenerational
transfers account for about 80 percent of total household wealth in United States. By contrast,
Modigliani (1988) obtained diametrical opposed finding that intergenerational transfers account
for only about 20 percent of total household wealth in the United States.
Kitamura et al. (1994) found that, the presence of well-established social security system and
generous public pension programs increases consumption expenditures of workers households in
Japan. The accumulation of wealth by Japanese households starts very early and lasts until very
late in life, with unconsumed wealth transferred to the next generation in the form of bequest
(Hayashi, 1997). Barthold and Ito (1991) found that about one-third to one-half of household
assets is obtained by bequest in Japan. Retirement and housing motivations found also to be
important in Japan (Horioka and Watanabe, 1997).

Fehr and Habermann in 2008 found that tax incentives motivated individuals saving in Germany.
They explain that as many other OECD countries before, Germany also introduced a programme
to promote the development of private saving in 2001. The program was similar to individual
retirement accounts (IRAs) in the United States and United Kingdom. Saving was mainly for
life-cycle motives. Also, study by Borsch-Suppan and Essig (2003) found that more than 40
percent of Germany households save regularly a fixed amount and about 25 percent households
plan their saving and have clearly defined saving target in mind. Most of Germany household
saving is in form of contractual saving, such as saving plans, whole life insurance and building
society contracts. Thus makes the flow of saving rather unresponsive to economic fluctuations,
such as income shocks. Also the study found most households prefer to cut consumption if ends
not met.
Banks and Tanner (1999) reviewed the economics of household saving in United Kingdom
(UK). The key findings were: total wealth in the UK was held in the form of liquid financial
assets, housing, pensions and life insurance; some inequality in the distribution of wealth would
be expected, given economic theories of the way households accumulate wealth over life-cycle;
the 1980s were a period of dramatic change in ownership of key assets such as housing, pensions
stocks and share; in spite of the proliferation of new saving vehicles, majority of the people still
hold the majority of their wealth in conventional forms such as interest bearing accounts at the
bank or building society; most individuals do not typically hold large amounts of financial
wealth; tax-privileged saving vehicles have been taken up relatively widely, but are held
predominantly by wealthier households; and almost one-ten of the population have no assets at
all and this proportion has been increasing over time.
Using a GMM-system estimator and a balanced panel of 258 Norwegian farm households, Sand
in 2002 found that traditionally in Norway farm households have relatively high saving and low
marginal propensity to consume.
Upender and Reddy in 2007 done a study in India and found that the estimate of constant
income elasticity of household saving to be more than unit implying that the marginal propensity

to save is higher than the average propensity to save, all else equal. Another study by Unny
(2001) found positive factors influencing saving in India including level of income, income
inequalities, value of assets and level of education, however, dependency ratio and number of
male children had negative influence. According to Salam and Kulsum (2001) Indian
government has policy in place promoting saving and capital formation as primary instrument of
economic growth and that saving is used to finance increasing requirement for investment. In
India, household sector saving provide bulk of national saving.
According to Waweru (2011) in Kenya SACCOs are seen as vehicles for resource mobilization
and gateways to economic prosperity for families especially those in the lower and middle
income category. Kibetet al in 2009 outlined determinants of saving in Kenya including type of
occupation, household income, age, and gender of household head, level of education,
dependency ratio, service charge, transport costs and credit access. Study by Ellis et al in 2010
found that, in Kenya, savings are used to undertake productivity-enhancing investments and
education provision. As expected, rural inhabitants found to save more for agricultural
investments while urban inhabitants tend to save for other purposes, such as starting a business.
Individuals with a better education are more likely to save and invest than those with less
education. Men and women exhibit similar patterns of behaviour in terms of saving for
investment purposes.
Boring in 2010 found several solid determinant of household saving behaviour in Uganda
namely: the age of the respondent (not just the age of the head of household), literacy, higher
education, formal sector employment, entrepreneurial activity, and attitudes about life’s current
state. Marital status and whether or not the respondent is sole responsible for the household
financial decisions is statistically insignificant regarding whether or not to save but quite
important regarding institutional choice. Also wealth found to play a significant role in the
decision to employ informal and non formal institutional saving methods. Ssemakula in 2007
conducted a review of Rural Speed a USAID funded project on saving promotion radio programs
in Uganda, he found that saving campaigns on radio generally demystified the thinking that
Ugandans do not have the saving culture except there was a general lack of information on
saving. The study found that project beneficiaries assert that they save in order to secure loans to

establish or expanding their enterprises. The USAID assisted project aimed at a broad-based
public awareness campaign with the aim of promoting the value of saving money. The programs
included saving related key topics such as reasons for saving, a potential saver, where to save,
why to save, and limitations of saving by rural households.
Hardly few studies on household saving have been done in Tanzania including findings by
Lwoga et al (1999). The findings concur with studies like those done by Johnson and Rogaly
(1997) and Rutherford (1999) that the poor use their saving for a variety of reasons which
include daily expenditure, consumption smoothing, and accumulation to meet life-cycle needs
and events and financing of emergencies.
3. Methodology
This paper presents cross-sectional data collected through structured questionnaire and focus
group discussion in thirteen districts in Tanzania. The data is all primary, as such secondary data
on rural households in Tanzania is scant. However primary data deemed to present more recent
state of affairs than secondary data. Mixed methods approach using questionnaire method and
focus group discussion facilitated triangulation of data. Further, quasi-randomized design
approach was adopted to ensure sample representatives are selected from sixgeographical
zonesin the country i.e. north, south, central, west, east for mainland, and Zanzibar.
Rural area in Tanzania constitutes largest part both geographically and population wise which
according to NBS (2014) rural population was 70.4% out of 44,928,923 people in 2012. Rural
area is characterized by poor transport infrastructure thus making accessibility difficult in some
areas; there are many ethnic groups (over 120) with diverse culture, norms, traditions, taboos,
customs and behaviours; there is different livelihood systems including crop farming,
pastoralism, mixed farming, and off farm activities (rural micro small enterprises); rural is
populous, relatively poor and illiterate than urban Tanzania. Therefore, sampling strategy is
designed to take into consideration of this diversity.
Probability and non-probability sampling methods namely simple random sampling, convenience
(or accidental) and purposive (or judgemental) sampling were used to establish sample

respondents. The representative sample administrative regions in each zone are: Kusini Pemba
and Mjini Magharibi regions for Zanzibar zone; Kilimanjaro and Manyara regions for north
zone; Iringa and Lindi regions for south zone; Dodoma region for central zone; Mwanza region
for west zone; Tanga, Morogoro and Pwani for east zone. Therefore data collection is done in
thirteen rural districts in eleven regions in Tanzania (both mainland and Zanzibar) with each
zone represented by at least one rural district. Table 3.1 shows information on geographical
location and respondents’ occupation.
Table 3.1 Respondents occupation and geographical location

No District Region Zone Data collection method and Major
samples livelihood
Questionnaire FGD activity of
respondents participants respondents
West Mjini Fisheries &
district Magharibi farming
1 (Unguja) Zanzibar 75 10
Mkoani South Fisheries &
2 (Pemba) Zanzibar 70 10 farming
Kondoa Dodoma Pastoralism
3 Central 60 10 (Maasai)
4 Sengerema Mwanza West 45 10 Farming
5 Mheza Tanga East 35 10 Farming
6 Ulanga Morogoro East 60 10 Farming
Kilombero Morogoro Pastoralism
7 East 45 10 & farming
8 Rufiji Pwani East 35 10 Farming
9 Mufindi Iringa South 90 Farming
Iringa Iringa Farming
10 Rural South 75 10
11 Kilwa Lindi South 95 Farming

Katesh Manyara Pastoralism

12 North 65 10 & Farming
Moshi Kilimanjaro Farming
13 Rural North 50 10
TOTAL 810 110
Source: Field data
Data collection was done via structured questionnaire and focus group discussion. In order to
control biasness and reinforce independence,structured questionnaire and focus group discussion
were administered by one person each. The questionnaire was administered in its original
language (English) by enumerator who was conversant in both English and Kiswahili (national
and local language). Enumerator performed questionnaire pretesting exercise prior to the actual
field data collection. Structured questionnaire was administered in all thirteen districts while
focus group discussions were done in eleven as presented in table 3.1. There were 810
respondents of the questionnaire whereas there were eleven focus group each with 10
participants making 110 people. The total sample size of respondents for both questionnaire and
focus groups is 920 households. Household is the unit of study therefore study data was collected
from household heads or their represenatives.
The focus group discussions were done among beneficiaries of one Belgian Technical
Cooperation (BTC) financed project known as Kilombero and Lower Rufiji Wetlands
Ecosystem Management Project (KILORWEMP) and four IFAD financed projects namely
Rural Micro Small and Medium Enterprise Support Programme (MUVI); Marketing
Infrastructure Value Addition and Rural Finance Programme (MIVARF); Agricultural Services
Support Programme/ Agricultural Sector Development Programme-Livestock Zanzibar
(ASSP/ASDPL-Zanzibar); and Belgian Funds for Food Security (BFFS). Table 3.2 presents
details of geographical locations of focus groups.
Table 3.2. Names and dates of focus group discussions

no Names of Focus Groups District Date
1 Fuoni Dairy Association West District June 2012

2 Farm Field School in Kendwa Mkoani May 2012

3 Pangalua Village Water User Association Kondoa October 2013
4 Farmer Group Sengerema April 2013
5 Sayuni SACCOS Galangala Village Katesh April 2013
6 Maduma Farmers Mheza April 2013
7 Farmer group Iringa Rural April 2013
8 Juhiwangumwa Wildlife Management Area Rufiji February 2015
Community Based Organisation
9 Mbuti Beach Management Unit Kilombero March 2015
10 Chokoachoko Community Based Forest Ulanga March 2015
Management
11 SACCOS in Umbwe Moshi Rural July 2012
Source: Field data
The framework of analysis of data is constituted by descriptive analysis which is used to estimate
descriptive statistics i.e. frequencies, percentages and cross tabulations and logistic regression
analysis is used to estimate the study model.
Econometric Model
Logistic Regression is used to establish relationship between dependent variable and independent
variables. Independent variable are age, sex, education, marital status, family size, occupation
and income whereas saving motiveis the dependent variable. Similar independent variables have
been used in many studies on determinants of household saving. The model assumes dependent
variable is in two categories “dichotomy dependent variable”: (i) Livelihood motives refer to
saving motives that contribute to poverty reduction they include business motive, retirement
motive, precaution motive, education motive, house motive, land motive, assets motive, extra
living cost motive, taxes and loan repayment motive and (ii) non-livelihood motives refer to
motives that do not contribute to poverty reduction such as leisure and travel motive, luxury
motive, entertainment motive, wedding motive, and funeral motive.
Therefore, the logistic regression model is presented by logistic function in equation (1):

f(y) =1/1+e-z (1)
The variable Z is defined as

z=0+ 1X1+2X2+3X3+4X4 +5X5+6X6+7X7
The logistic regression in equation (1) re-written with variable Z defined

f(y) =1/1+e-0+ 1X1+2X2+3X3+4X4 +5X5+6X6+7X7 (2)
Where:
f(y) = Dependent variable (livelihood and non-livelihood saving motives)
0 = intercept
1-7 = regression coefficients
X1- X7 = predictors (age, education, marital, sex, income, occupation, family size)
The variable x in equation (2) represents individual factors or independent variables affecting the
dependent variable, whereas f(y) represents the probability of a particular outcome (dependent),
given the set of determinants (factors). Therefore variable z is a measure of the total contribution
of all the factors used in the model and is known as the logit.
Each of the regression coefficients describes the size of the contribution by the predictor. A
positive regression coefficient means that the predictor increases the odds (likelihood) of
outcome, while a negative regression coefficient means that the predictor decreases the odds of
outcome; a large regression coefficient means that the predictor strongly influences the odds of
the outcome; while a near zero regression coefficient means that the predictor has little influence
on the odds of outcome.
4. Analysis and Results

The total number of respondents participated in the questionnaire survey are 810 with 96.4
percent response rate. The high responses rate implies that the subject was well understood and
interesting to the respondents. There was a good balance of gender whereby of the total
respondents 52.3 percent are men and 47.7 percent are women. Further the results implies that

the subject was interesting to both female and male gender. Respondents who claim not to have
attended school at all is 12 percent, 62.7 percent have primary education, and 22.2 percent have
secondary education while only 3.1 percent have managed to study up to tertiary level (College
or university education). Results of education level of respondents reflects situation of literacy in
many developing countries including Tanzania whereby reports show high illiteracy in rural
areas. Marital status of the respondents include: 53.3 percent are married, 36.2 percent are single,
3.7 percent are widow, 3.1 percent are divorcees and 3.3 percent are separated. Results show
occupation of the respondents as farm and off farm activities 95.1 percent and 4.9 percent are
employed. As for the family size on average every respondent has four dependants.
Table 4.1 Aggregated results

Description Results (%)
Preference of livelihood motives 74.1
Preference of life-cycle model 85.4
Education as first priority 66.8
Rigid to change saving motives 82 .9
Savings used to buy household assets 56.2
Finance illiteracy 79.4
Source: Field data
Table 4.1 presenting aggregate results of data collected from respondents, rural households are
saving for motives in the livelihood category which are capable to reduce poverty. Also the
results confirm that rural household follow life-cycle model meaning that they do save for
retirement. They save money for use when they are old and retired from production activities or
employment. Further the results show that education is priority motive of saving by rural
household as 66.8 percent of the respondents rank education as top priority motive to save. This
is positive results since education has great impact in poverty reduction. Also the results show
that rural household is not willing to change motives for saving. On one hand this aspect is good
in case of livelihood motives but on the other hand it is negative in case of non-livelihood
motives. For example if household doesn’t want to change the motives to buy agricultural inputs

such as seeds, fertilizers etc. it means that agricultural production would increase. But if
household was not willing to change non-livelihood motive like entertainment for livelihood
motive like saving for education it means that illiteracy remain with the household. Further the
results show rural households use savings to buy durable items or household assets by 56.2
percent. Finance illiteracy is prominent with 79.4 percent are without basic finance knowledge.
The results were supported by majority during focus group discussion thus confirming data
collected via structured questionnaire.
Table 4.2 Disaggregated results

Findings/Results (%)
Description Gender Age
Men Women Young old
Preference of livelihood motives 71.7 76.7 73.4 82.8
Preference of life-cycle model 85 85.8 89.9 74.1
Education as first priority 66.7 66.8 67.1 62.1
Rigid to change saving motives 80.9 87.1 85.3 64.9
Savings used to buy household assets 59.4 52.6 54.2 82.8
Finance illiteracy 74.5 84.7 80.7 62.1
Source: Field data
Disaggregating results by gender and age the findings show slightly differences based on gender
and age. Women has slightly higher rate for livelihood than men this is explained by the fact that
women especially in the rural are the ones who ensures food for the young children in the family
unlike men who may spent savings on entertainment e.g. localdrinks. Old people have slightly
higher preference of livelihood than young it is perhaps because old people have farms and may
need more savings than young to support agricultural production. As regard to education motive,
men and women are indifferent, however, young people place higher preference on education
motive than old people thus confirming the fact that they are the one getting education thus it is
positive that young people would prefer saving for education than old people. The results show
that old people are slightly flexible in changing motives for saving whereas men, women and
young people seem rigid in changing motives for saving. This can be explained by the fact that

old people have short period remained on life span thus they may be willing to revise their
motives especially those of long term nature like infrastructure, valuable assets and investment.
Old people seem too illiterate than old meaning that young generation has more education on
finance matters than old people.
Table 4.3 presents results on ranks of saving motives in terms of preferences. Respondents
assigned ranks in order of preference from 1st to 12th with top priority motive ranked 1st and
least priority motive ranked 12th. Results show that preferences were relatively different on
seven saving motives thus there is no dominant rank with relatively high score. However, results
show that five saving motives education ranks 1st (50.9%), bequest ranks 9th (28.9%), marriage
ranks 10th (38.6%), leisure and entertainment ranks 12th (74.9%), and non-specific ranks 11th
(45.6%) respondents were indifferent thus there is a dominant rank with high score. The results
imply that these saving motives are common to many. Results show high scores of non-specific
motives means that there are other motives for saving by rural households besides eleven
motives presented in this study.
Table 4.3 Results on saving motive preferences

Score (%)
N Saving 10t 11t 12t
o Motive 1st 2nd 3rd 4th 5th 6th 7th 8th 9th h h h
17. 14. 14. 10. 10. 11.
1 Retirement 3.2 6.7 8 2 8 7 2 4 6.0 2.5 1.1 0.4
Extra 15. 18. 20. 11.
2 living cost 1.7 4.8 5.4 9.6 8.0 8 5 1 9 2.2 0.9 0.2
Illness ,
disaster
(precaution 28. 20. 16. 12.
3 ary) 9.3 9 1 3 3 6.7 2.8 1.6 0.5 0.6 0.0 0.0
50. 13.
4 Education 9 1 8.8 6.5 4.4 4.0 4.2 3.6 1.9 1.0 0.2 0.0
5 Marriage 0.7 0.6 0.5 1.1 1.5 1.2 2.3 6.2 14. 38. 29. 2.5

0 6 6
House & 15. 19. 12. 12. 12.
6 land 6 1 0 5 7 8.6 9.9 5.6 1.7 0.6 0.7 0.4
Assets
(durable 10. 12. 11. 11. 15. 11.
7 items) 7.0 9 1 5 7 1 2 8.3 6.9 4.1 0.2 0.1
Leisure &
entertainme 15. 74.
8 nt 0.5 0.2 0.4 0.9 0.4 0.5 0.5 0.9 1.9 3.7 3 9
Taxes &
loan 10. 14. 17. 18. 14.
9 repayment 0.5 1.7 4.9 7.2 0 6 0 6 4 6.9 1.7 0.7
11. 15. 14. 15. 13. 10.
10 Business 8.9 5 2 9 9 8 7 4.0 2.3 1.6 0.6 0.0
Non 22. 45. 17.
11 specific 0.2 0.1 0.1 0.4 1.0 0.5 0.4 1.7 8.9 2 6 9
11. 17. 28. 14.
12 Bequests 0.9 1.7 2.0 4.2 6.8 7.7 1 2 6 9 2.8 1.5
Source: Field data (2015)
Notably, views of participants in the group dicsussions held in eleven districts are perfectly
inline with study results collected via the structured questionnaire. Therefore information
collected through questionnaire interview is valid to support conclusion of this paper.
Econometric results
Table 4.4 presents results of logistic regression. The -values presents coefficient of
determination for each independent variable in the model. The results show positive relationship
between level of education and livelihood outcome. This results is consistent with the fact that an
educated individual is likely to make rational decision in choosing saving motives that can
reduce poverty. Marital status seem to be negatively related to livelihood saving motives. This
results was perhaps influenced by young persons as they may be concerned with cost of marriage

therefore they choose to save in order to cover for the costs related with wedding ceremonies.
Further, occupation is negatively related to the dependent variable. Perhaps this results imply
that once a person has steady income source from employment he or she may not care about use
of savings for livelihood since the person has stable income source from employment to take
care the role of savings. Therefore saving here seems important aspect to persons with
entrepreneurial spirit or persons with unpredictable income. The other factors in the model are
more or less neutral meaning that they actually do not influence dependent variable.
Table 4.4 Logistic regression results

95.0% C.I. for
EXP()
Expected
sign  S.E. Wald df Sig. Exp() Lower Upper
Sex + 0.94 0.33
-0.225 0.231 1 0.799 0.508 1.256
7 0
Age cohort + 0.13 0.70
-0.174 0.467 1 0.840 0.337 2.099
9 9
Education + 2.06 0.15
1.264 0.879 1 3.539 0.632 19.837
level 6 1
Marital status + 1.44 0.22
-1.019 0.847 1 0.361 0.069 1.900
6 9
Dependants + 0.87 0.34
-0.249 0.266 1 0.780 0.463 1.313
level 6 9
Occupation + 5.05 0.02
-1.746 0.777 1 0.174 0.038 0.799
5 5
Income level + 0.04 0.83
-0.069 0.323 1 0.933 0.495 1.757
6 0
Constant 4.35 0.03
2.562 1.228 1 12.961
2 7
Source: Field data

5. Conclusion
Generally, based on the findings, it is possible to conclude that rural households in Tanzana are
saving for sound causes. With regard to the effect of explanatory variables on household saving
motives in the model, therefore (i) Education level is the only factor that positively influence
chances to select livelihood saving motives which is capable to reduce poverty; (ii) Marital status
of an individual negatively affects chances to save for livelihood motives thus persons who are
not married would save to finance cost of future marriage wedding events; (iii)Occupation of a
person affects chances to choose livelihood saving motive; (iv)Rural households follow life-
cycle model meaning that they also save for retirement; and (v)The study has found rigidity in
switching off saving motives among rural households. Policies emphasing provision of financial
education to boost financial literacy e.g. saving, bookkeeping, financial statements, costing,
interest, dividends, pricing; fostering financial and insurance markets for increased financial
inclusion and insurance markets in rural areas would enhance household saving for investments
and agricultural production for poverty eradication. The above are underlying policy
recommnedations suggested by this paper.
References
[1] Abdelkhalek, T., Arestoff, F., Freitas, N., M. and Mage, S. (2009) A Microeconometric
Analysis of Households Saving Determinants in Morocco, JEL, Universite Paris Dauphine
[2] Ainsle, G. (1975) Specious Reward: A Behavioural Theory of Impulsiveness and Impulse
Control.Psychological Bulletin, 82, 463-496
[3] Angeletos, G.M., Laibson, D., Repetto, A., Tobacman, J., & Weinberg, S. (2001). The
Hyperbolic Buffer Stock Model: Calibration, Simulation, and Empirical Evaluation. Journal of
Economic Perspectives, 15, 47-68
[4] Attanasio, O. and Szekely, M. (2000) Household saving in East Asia and Latin America:
inequality, Demographics and all that. Annual conference on development economics, World
Bank
[5] Banks, J and Tanner, S (1999) Household Saving In the UK, The Institute of Fiscal
Studies, London

[6] Barthold, T. and Ito, T. (1991) Bequest Taxes and Accumulation of Household Wealth:
U.S.-Japan Comparison. In Ito, T. & Kruger, A., Eds. The Political Economy of Tax Reform.
Chicago: The University of Chicago Press, Pp.235-290.
[7] Beverly, S. and Sherraden, M. (1999) institutional determinants of saving: implications
for low-income households and public policy, Journal of socio-economics, 28 457-473Beverly,
S., Sherraden, M., Cramer, R., Williams, T., Nam, Y., and Zhan, M. (2008) Determinants
ofAsset Holdings in McKernan and Sherraden, M. (eds), Asset Building and Low-Income
Families (pp. 89-152) Washington DC: Urban Institute Press
[8] Boring, E.I. (2010) An Empirical Analysis of Household Saving Behaviour in Uganda,
Georgia State University
[9] Borsch-Supan, A. and Essig, L (2003) Household Saving in Germany: Results of The
First Save Study, NBER, Massachusetts
[10] Bank of Tanzania (2012) Banking Supervision Annual Report, Dar es Salaam, Bank of
Tanzania
[11] Chowa, A.N.G., Masa, R.D., and Ansong, D. (2012) Determinants of Saving Among
Low-Income Individuals in Rural Uganda: Evidence from Assets Africa. Advances in Applied
Sociology Vol.2, No.4, pp280-291
[12] Dauner, I. (2004) Mobilizing Saving: Key Issues and Good Practices in Saving
Promotion, Swiss Agency for Development and Cooperation, OSDC, February 2004
[13] De Nardi, M., French, E., Jones, J.B. (2009) Life Expectancy and Old Age
Savings,American Economic Review, American Economic Association, vol. 99(2), pages 110-
15,
[14] Dercon, S. (1996) Risk, Crop Choice, and Saving: Evidence from Tanzania, Economic
Development and Cultural Change, Vol. 44, No.3 (Apr., 1996), Pp.485-513, The University of
Chicago
[15] D’Souza, V.S. (1981) Inequality and its Perpetuation: A Theory of Social Stratification,
New Delhi: Manohar
[16] Ellis, K., Lemma, A. and Rud, J.P. (2010) Financial Inclusion, Household Investment and
Growth in Kenya and Tanzania, Overseas Development Institute, London

[17] Fehr, H and Habermann, C. (2008) Private Retirement Saving in Germany: The Structure
of Tax Incentives and Annuitization, Department of Economics, University of Wuezburg,
Wurzburg
[18] FSDT (2006) FinScope National Survey on Access to and Demand for Financial Services
in Tanzania,
[19] Hayashi, F., Altonji, J. and Kotlikoff, L. (1996) Risk Sharing Between and Within
Families,Econometrica, 64, 261-94.
[20] Horioka, C.Y. (1990) Why Is Japan’s Saving So High? A Literature Survey. Journal of
The Japanese and International Economies, 4, 49-92
[21] Horioka, C. Y. and Watanabe, W. (1997) Why Do People Save? A Micro-Analysis of
Motives for Household Saving in Japan. Economic Journal, 107, 537-552
[22] IFAD, (2013) Supervision report of Marketing Infrastructure, Value Addition and Rural
Finance Support Programme (MIVARF) Tanzania
[23] Jevons, W.S. (1965) The Theory of Political Economy (5th ed.) London: Macmilla
[24] Katona, G. (1975) Psychological Economics. New York: Elsevier
[25] Kibet, L.K., Mutai, B.K., Ouma, D.E., Ouma, S.A., and Owuor, G. (2009) Determinants
of Household Saving: Case Study of Smallholder Farmers, Entrepreneurs and Teachers in Rural
Areas of Kenya, Journal of Development and Agricultural Economics Vol. 1(7), Pp.137-143
[26] Kitamura, Y., Takayama, N. and Arita, F. (2001) Household Saving in Japan Revisited,
University of Venice, Research in Economics (2001) 55, 135-153
[27] Kotlikoff, L. (1988) Intergenerational Transfers and Saving, Journal of Econometric
Perspectives, 2, 41-58.
[28] Kotlikoff, L.J. and Summers, L.H. (1981) The Role of Intergenerational Transfers In
Aggregate Capital Accumulation, Journal of Political Economy, Vol. 89, No.4 (August), Pp.
706-732.
[29] Kraay, A. (2000) Household Saving in China, World Bank Economic Review, Vol.14,
No.3
[30] Laibson, D., (1997) Golden eggs and hyperbolic discounting. The Quarterly Journal of
Economics, 112, 443-477

[31] Lwoga, C., Maximabali, F., Rutherford,S., Mutesasira, L., Safe, S., Gogadi, B., Mukami,
N., Seiya, G., Maeda, S,. And Tairo, J.N. (1999) Use and Impact of Saving Services among the
Poor in Tanzania, Microsave
[32] Marshall, A. (1961) Principles of Economics: and Introductory Volume (9th ed) London:
Macmillan
[33] Mirach, T.H. and Hailu, Y.M. (2014) Determinants of Household Saving in Ethiopia: A
Case of North Gondar Zone, Amhara Regional State, International Journal of Development
andEconomic Sustainability, Vol.2, No.4 pp37-49, UK.
[34] Modigliani, F. (1988) The Role of Intergenerational Transfers and Life Cycle Saving In
The Accumulation of Wealth, Journal of Economic Perspectives, 2, 15-40.
[35] Mullainathan, S., and Thaler, R. (2000) Behavioural Economics, Cambridge, MA:
National Bureau of Economic Research
[36] NBS, (2014) 2012 Population and Housing Census, -Basic Demographic and Socio-
Economic Profile
[37] Nga, M. T. (2007) An Investigative Analysis Into The Saving Behaviour of Poor
Households in Developing Countries: With Specific Reference to South Africa, Unpublished
Masters Thesis, Department of Economics, University of The Western Cape
[38] Nigus, H.Y., (2015) Determinants of Household Saving in Gedeo Zone, Southern
Ethiopia, Journal of Economics and Sustainable Development, Vol.6, No.7
[39] Ölander, F., and Seipel, C. M. (1970). Psychological approaches to the study of saving
(Studies in Consumer Savings No. 7). Urbana- Champaign, IL: Bureau of Economic and
Business Research, University of Illinois.
[40] Ravi, S. (2006) Uncertainty and Saving Decision, Indian School Of Business
[41] Rutherford, S. (2000) the Poor and their Money, New Delhi: Oxford University Press
[42] Salam, A. and Kulsum, U. (2001) Saving Behaviour in India: An Empirical Study,
Department of Economics, AMU, Aligarh
[43] Sand, R. (2002) The Propensity to Consume Income From Different Sources and
Implications For Saving: An Application To Norwegian Farm Households, Nord Trondelag
Research Institute
[44] Shefrin, H.M. and Thaler, R.H. (1988) The Behavioural Life-Cycle Hypotheses.
Economic Inquiry, 26, 609-643

[45] Sherraden, M. (1991) Assets and the poor: a new American welfare policy. Armonk,
N.Y: M.E. Sharpe, Inc.
[46] Sherraden, M., and Barr, M. (2005) Institutions and inclusion in saving policy. In
Retsimas and Belsky, E. (Eds), Building assets, building credit: creating wealth in low-income
communities (pp. 286-315) Washngton DC: brookings Institution Press
[47] Sherraden, M., Schreiner, M., and Beverly, S. (2003) Income, Institutions, and Saving
Performance In Individual Development Accounts, Economic Development Quarterly, 17, 95-
112
[48] Ssemakula, M. (2007) Saving Promotion Radio Programs, Chemonics International Inc.
[49] Stiglitz, J.E. (1993) Principles of Microeconomics, W.W. Norton & Company, Inc.
Stanford University, 10-641.
[50] Teshome, G., Kassa, B., Emana, B., and Haji, J. (2013) Determinants of Rural Household
Savings in Ethiopia: The Case of East Hararghe Zone, Oromia Regional State, Journal of
Economics and Sustainable Development, Vo.4, No.3
[51] Thaler, R. H. (1981) Some Empirical Evidence On Dynamic Inconsistency, Economic
Letters, 8, 201-20
[52] Upender, M. and Reddy, N.L. (2007) Saving Behaviour in the Indian Economy
International Journal of Applied Econometrics and Quantitative Studies Vol.4-1
[53] Unny, C.J. (2001) Determinants of Saving of Rural Households in Kerala, Christ College,
Irinjalakuda, Thrissur, Kerala
[54] Wakabayashi, M. and MacKellar, L (1999) Demographic Trends and Household Saving
in China, IIASA, Laxenburg, Austria.
[55] Waweru, K.M. (2011) An Investigation Into The Cash Balance Management Challenges
In Saving and Credit Cooperative Societies (SACCOS) in Nakuru County, Kenya, School of
Business and Public Management, Mount Kenya University
[56] Weber, M. (1967) Class, status and party. In Bendix, R. and Lipset, S.M. (Eds), Class
Status and Power: Social Stratification Comparative Perspective (pp. 21-28). London Routledge
and Kegan Paul

Abreviations and Accronyms

AIH Absolute Income Hypothesis
BTC Belgian Technical Cooperation
GDP Gross Domestic Product
IFAD International Fund for Agricultural Developmnet
LCH Life-Cycle Hypothesis
NBS National Bureau of Statistics of the United Republic of Tanzania
OECD Organisation for Economic Cooperation Developmnet
PIH Permanent Income Hypothesis
SACCOS Savings and Credit Coorperative Society
USAID United States Agency for International Developmnet
Funding
This research is sole funded by authors themselves. Therefore, authors hereby declare that there
is no substantive conflict of interest construed to influence the results or interpretation of the
manuscript.


Determinants of Rural Household Saving I

Uploaded by

Copyright:

Available Formats

Determinants of Rural Household Saving I

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Determinants of Rural Household Saving I

Uploaded by

Copyright:

Available Formats

International Journal of Research in Social Sciences

Vol. 6 Issue 11, November 2016,

Determinants of Rural Household Saving in

Keywords:Saving motives, livelihood motives, non-livelihood motives

299 International Journal of Research in Social Sciences

300 International Journal of Research in Social Sciences

301 International Journal of Research in Social Sciences

(a) Individual-oriented perspectives

302 International Journal of Research in Social Sciences

303 International Journal of Research in Social Sciences

(b) Sociological perspective

(c) Institutional perspective

304 International Journal of Research in Social Sciences

2.2 Empirical literature

305 International Journal of Research in Social Sciences

306 International Journal of Research in Social Sciences

307 International Journal of Research in Social Sciences

308 International Journal of Research in Social Sciences

Table 3.1 Respondents occupation and geographical location

309 International Journal of Research in Social Sciences

Katesh Manyara Pastoralism

Table 3.2. Names and dates of focus group discussions

310 International Journal of Research in Social Sciences

2 Farm Field School in Kendwa Mkoani May 2012

311 International Journal of Research in Social Sciences

f(y) =1/1+e-z (1)

The variable Z is defined as

The logistic regression in equation (1) re-written with variable Z defined

4. Analysis and Results

312 International Journal of Research in Social Sciences

Table 4.1 Aggregated results

313 International Journal of Research in Social Sciences

Table 4.2 Disaggregated results

314 International Journal of Research in Social Sciences

Table 4.3 Results on saving motive preferences

315 International Journal of Research in Social Sciences

316 International Journal of Research in Social Sciences

Table 4.4 Logistic regression results

317 International Journal of Research in Social Sciences

318 International Journal of Research in Social Sciences

319 International Journal of Research in Social Sciences

320 International Journal of Research in Social Sciences

321 International Journal of Research in Social Sciences

322 International Journal of Research in Social Sciences

Abreviations and Accronyms

323 International Journal of Research in Social Sciences

You might also like