Tuesday, September 17, 2002
|
|
|
FACT SHEET
The Tanzanian Household Budget Survey (HBS), conducted in
2000/01 by the National Bureau of Statistics (NBS), is the largest-ever
household budget survey in Tanzania, and one of the largest in Africa, covering
more than 22,000 households. It represents an important achievement for the
NBS, an independent executive agency of government.
Oxford Policy Management (UK) and its partners assisted NBS
in sampling, data processing, analysis and report writing. The survey was
supported by the Tanzanian development partners: and specially by the United
Kingdom Department for International Development. Funding was also provided by
Governments of Sweden (SIDA), Canada (CIDA), Netherlands (The Royal Netherlands
Embassy), Denmark (DANIDA), Japan (JICA), United States (USAID), Norway and
UNDP.
Today’s report presents the results of the survey.
More than
22,000 households interviewed
A nationally-representative sample of 22,178 households was
interviewed – between 12 and 24 households in each sampled area and around
1,000 in each of Mainland Tanzania’s 20 regions[1].
The information was collected for both urban and rural areas.
The questionnaire was designed to record a whole range of
individual and household characteristics, including:
n
household
members’ education, economic activities and health status
n
household
expenditure, consumption and income
n
ownership of
consumer goods and assets
n
housing
structure and building materials, and
n
household
access to services and facilities.
Fieldwork was between May 2000 and June 2001.
Assessing trends during the 1990s
The 2000/01 Tanzanian Household Budget Survey gives trends
over the 1990s by comparing its data with the 1991/92 HBS (see table on page
6).
The 2000/01 HBS used a questionnaire very similar to the one
in 1991/92, so the two data sets are reasonably comparable. The National Bureau
of Statistics conducted both surveys, with a number of staff involved in both.
In order to maximise the comparability of the estimates over time, the 1991/92
data were completely re-analysed. While this analysis was informed by earlier
work on the 1991/92 data, a number of substantial changes were made. These
included re-weighting the 1991/92 data to bring the population share of each
area more into line with population estimates.
Taken together, the two Household Budget Surveys enable
in-depth analysis and represent a powerful resource to further inform poverty
policy. It is expected that other analyses will be undertaken in the future,
both inside and outside the National Bureau of Statistics.
Poverty
and other social indicators: main results
Unsurprisingly, the 2000/01 HBS confirms that income poverty is high and many social
indicators are poor in Tanzania. It also shows large gaps between urban and
rural populations. At one extreme, Dar es Salaam is substantially better
off than the rest of the population by almost all measures; at the other, rural
households are much poorer than their urban equivalents in almost all respects.
On the whole, differences in poverty and social indicators
between men and women are smaller than geographical differences. Women have lower incomes than men, though it is not possible to
assess differences in consumption between individuals. The proportion of
households headed by women increased during the 1990s but they are no poorer
than those headed by men. Adult women have lower levels of education than adult men, but current school enrolment rates are
slightly higher for girls.
Many measures of welfare
show modest improvements during
the 1990s. The economy diversified and household consumption increased. The
proportion of the population that is poor has fallen slightly, although
absolute numbers have risen due to population growth. However, improvements are
often associated with rising inequality.
Many improvements have been concentrated in urban areas, particularly Dar es
Salaam, while more limited in rural areas.
In detail
Household demographic composition. There has been a large rise
in the proportion of households headed by a woman – from 18 per cent in 1991/92
to 23 per cent in 2000/01. Urban areas other than Dar es Salaam have the
highest proportion of such households. There has been a fall in household size
– from an average of 5.7 to 4.9 people – and a small rise in the proportion
headed by the over-65s.
Household construction and facilities. There has been an increase
in the proportion of households living in dwellings built with modern materials
– concrete, stone, cement and metal. Nationally, 84 per cent of households own
the house they live in, although in urban areas over a third rent privately.
There was very little change in tenure over the
decade. Overall, 10 per cent of Tanzanian households are connected to the
electricity grid. Coverage is much greater in urban areas (59 per cent of
households in Dar es Salaam and 30 per cent in other urban areas). In rural
areas, only two per cent of households are connected. Around nine per cent of
households use electricity as their main energy source for lighting, while 84
per cent depend on paraffin. Use of electricity for cooking is less common
(about one per cent of households).
Some 93 per cent of households report having use
of a toilet; over 90 per cent in rural areas. There was no change over the
decade.
Average distance to a number of important
services fell over the 1990s, but in rural areas there are long average
distances to some important facilities: households are on average 37 kilometres
from a bank and 18 kilometres from a police post.
Education. A quarter of Tanzanian adults have no education
and 29 per cent can neither read nor write. Some 30 per cent in rural areas
have no education, compared with only eight per cent in Dar es Salaam and 13
per cent in other urban areas. Women are about twice as likely as men to have
no education. Rural women in particular have missed out, with 41 per cent
unable to read or write. Improvement in the level of adult education over the
decade was limited. There was a rise in the highest standard achieved for those
with primary schooling but no apparent decline in the proportion of adults
without education.
Almost two-thirds of Tanzanian households are
within two kilometres of a primary school; even in rural areas 58 per cent are
within this distance. This suggests distance is not a major impediment to
primary schooling for most households. Average distance to a primary school
appears to have lengthened slightly over the decade. Households are much
farther from secondary schools: a quarter of rural households are more than 20
kilometres distant.
Health. In rural areas, some 28 per cent of individuals
were ill in the four weeks preceding the survey, compared with 24 per cent in
urban areas. Children under five and older adults are the age groups most
likely to be ill or injured. Women report more illness than men.
For individuals who have been ill, the most
commonly reported complaint is fever/malaria – reported in 69 per cent of
children and 60 per cent of adults. Some 69 per cent of individuals who had
been ill had consulted a health-care provider. Even in rural areas, 67 per cent
reported consultation of some kind. Some 54 per cent of individuals who
consulted a health-care provider used a government service. The private sector
is an important service provider in both urban and rural areas. Users are more
likely to report dissatisfaction with government providers than with private
ones. Long waiting times, lack of drugs and, in the case of regional hospitals,
high cost are the most commonly reported problems with government facilities.
High cost is also the most common complaint about private providers. Most
households are reasonably close to primary health-care facilities. The average
distance to these facilities appears to have shortened slightly over the
decade.
Drinking water. Overall, 43 per cent of Tanzanian households use
an unprotected source of drinking water, including unprotected wells and
springs and surface water such as rivers and lakes. Some 39 per cent use piped
water and another 16 per cent protected wells or springs. People in urban areas
have better drinking water supplies than the rural population. Rural households
must also travel farther to their supply, with only 49 per cent within a
kilometre of it. This compares with 73 per cent of households in urban areas.
Productive activities and productive assets. Although most Tanzanians
still depend on agriculture, households have diversified their economic
activity. Some 70 per cent are now headed by an individual who works in
agriculture or fishing, compared with 75 per cent in 1991/92. There has also
been a drop in government and parastatal employment – from 5.2 to 2.5 per cent
of adults. There has been a rise in private sector employment and in
self-employment, which are now the main activities of 40 per cent of adults in
Dar es Salaam and 31 per cent in other urban areas.
Women have experienced the largest reduction in
agricultural activity: down from being the main activity of 77 per cent in
1991/92 to 63 per cent in 2000/01. Men saw a smaller fall, but have been more
affected by the shift from government and parastatal employers to the private
sector.
Some 62 per cent of children aged five to 14
undertake some form of work; slightly over half combine work with study. Girls
are more likely to work than boys – 64 per cent of girls and 59 per cent of
boys reported this. When they work, girls are most likely to do so in the
household business or undertaking household chores. Boys are more likely than
girls to work in agriculture or outside the home.
Household consumption. Average consumption per
person is highest in Dar es Salaam – 2.6 times higher than the rural average of
about 8,500 TShs per month. Household consumption rose by around 17 per cent in
real terms between 1991/92 and 2000/01. Dar es Salaam saw the biggest rise –
around 47 per cent – whereas rural areas witnessed only around 11 per cent. As
a result, the gap between urban and rural areas widened.
Some 65 per cent of household consumption
expenditure is on food. Rural households spend the highest proportion on food;
those in Dar es Salaam the lowest.
Education and medical expenses each represent
about two per cent of average household expenditure. Dar es Salaam households
spend the biggest proportion of their income on health and education; rural
households the lowest.
Income poverty and inequality. The rise in household
consumption was accompanied by a small increase in inequality. The Gini
coefficient, a measure of how unequally expenditure is distributed, rose from
0.34 to 0.35.The richest 20 per cent of the population now account for 44 per
cent of household spending, compared with 43 per cent in 1991/92. The biggest
rise in inequality was in urban areas, particularly Dar es Salaam.
There
was a small fall in income poverty of about three percentage points over the
decade. Some 36 per cent of Tanzanians now fall below the basic needs poverty
line and 19 per cent below the food poverty line, compared with 39 per cent and
22 per cent in 1991/92 (see How the poverty lines are
drawn on
next page).
The absolute
number of individuals living in poverty increased during the 1990s because of
population growth. Using national population projections, there are now 11.4
million Tanzanians below the basic needs poverty line compared with 9.5 million
in 1991/92.
Poverty profile. Households with many members and those with a
large proportion of dependants are particularly likely to be poor. Households
headed by someone who is economically inactive are also more likely to be poor.
Households that depend on agriculture have somewhat higher levels of poverty
than average, particularly those relying on sale of livestock. Poverty levels
are also strongly related to the education of the head of household. Some 51
per cent of individuals are poor if the head has no education, compared with
only 12 per cent when the head is educated above primary level.
Over the 1990s, poverty declined most among the
employed and self-employed, particularly private sector employees. In contrast,
poverty increased in households with a head who was economically inactive or
uneducated and in those with a large proportion of dependants. Only 50 per cent
of children aged seven to 13 from the poorest households were studying compared
with 66 per cent from other households. This risks creating a cycle of
deprivation, since the survey also shows that adults’ incomes are strongly
related to their education. In the case of adults, the poor do not report
higher levels of illness and injury than wealthier households; for their
children the reverse is the case.
Reported income. Despite the overall importance of agriculture in
rural areas, some 40 per cent of rural household income comes from sources
outside their own farm production. Employment and self-employment account for
around 71 per cent of income in Dar es Salaam. In other urban areas these
sources contribute 57 per cent of income and agricultural sources provide
another 20 per cent.
Rural households depend on a wider variety of
income sources than urban households; 65 per cent report more than three
sources. Such diversification seems to be an important way for rural households
to raise their incomes, since households with a larger number of sources have
higher mean incomes.
Individuals with a tertiary education earn
almost four times the income of the least educated. There are also large
differences between average incomes of men and women, men earning 1.9 times
more.
How the
poverty lines are drawn
The poverty lines define the minimum expenditure
necessary to meet basic human needs. The survey used two lines: the food
poverty line and the basic needs poverty line.
The food
poverty line represents
the expenditure necessary to eat sufficient calories. It is based on the food
basket consumed by the poorest 50 per cent of Tanzanians. Average quantities
consumed per adult equivalent are estimated for every food item. Average unit
prices are also calculated. The approximate calorific values of these foods are
calculated. The food basket gives the share of consumption accounted for by
each item. The level is set so that the sum of calories is 2,200 per day, the
minimum necessary for survival. The food basket defined by these two parameters
is then priced to give the food poverty line.
The basic
needs poverty line
is higher than the food poverty line. It includes the
cost of other essential items of expenditure, such as clothes and other
non-food consumption. This is done by calculating the share of expenditure that
goes on food in the poorest 25 per cent of households. Multiplying the food poverty
line by the inverse of this share inflates it to allow for non-food
consumption. [In 2000/01, the food poverty line was 5,295 TShs compared
with 2,083 TShs in 1991/92. The basic needs poverty lines were 7,253 TShs in
2000/01 and 2,777 TShs 1991/92 (all in nominal prices) (Note that at this moment, 1 US dollar
is equivalent to 950 Tanzanian Shillings.
Key Indicators from the Household Budget Surveys
Indicator 1991/92 2000/01
THE HOUSEHOLD AND HOUSING
Average household size 5.7
4.9
Mean percentage of dependants 40
42
Percentage of female-headed households 8
23
Percentage of households with a modern roof 36 43
Percentage of households with modern walls 16 25
Average number of persons per sleeping room 2.6 2.4
Percentage of households with electricity 9
12
Percentage of households using a toilet 93
93
Percentage of households owning a radio 37 52
EDUCATION, HEALTH AND WATER
Percentage of adult men with any education 83 83
Percentage of adult women with any education 68 67
Percentage of adults literate –
71
Primary net enrolment ratio –
59
Percentage of children age 7-13 years studying 57 61
Secondary net enrolment ratio (forms I-IV) – 5
Percentage of households within 2 km of a primary school 66 63
Percentage of ill individuals who consulted any health provider - 69
Percentage of households within 6 km of a primary health facility 75 75
Percentage of households with a protected water
source 46 55
Percentage of households within 1 km of drinking water 50 55
ECONOMIC ACTIVITIES
Percentage of adults whose primary activity is agriculture 73 63
Percentage of children age 5-14 years who are working – 62
Mean area of land owned by rural households (acres) – 6.0
Percentage of households with a member with a bank account 18 6
CONSUMPTION AND POVERTY
Average consumption exp. per capita (2000/01 TShs, 28 days) * 8,686
10,120
Percentage of consumption expenditure on food 71 65
Percentage of population below the food poverty line 22 19
Percentage of population below the basic needs
poverty line 39 36
Percentage of total consumption by the poorest 20 percent of population 7 7