|
|
 |
|
Street Vendors in Africa: Data and Methods
Jacques Charmes
Director of Research,
French Scientific Research Institute for Development and Co-operation
(ORSTOM. Paris)
March 1998 (revised October 1998)
Paper prepared for the United Nations Statistics
Division, the Gender in Development Programme of the United Nations
Development Programme (UNDP) and the project "Women in Informal
Employment: Globalizing and Organizing" (WIEGO).
Defined by the XVth International Conference
of Labour Statisticians as a new concept of labour force in 1993,
the informal sector comprises units in the household sector, as defined
by the System of National Accounts (SNA) and which, therefore, are
unincorporated enterprises or do not hold a complete set of accounts,
including:
i) units - registered or not - without permanent employees,
ii) units with permanent employees and which are, alternatively or
simultaneously:
- unregistered units, or
- units which do not register
their permanent employees, or
- units which employ, on
a continuous basis, less than a given number of persons, according
to the legislative codes (fiscal or social) or to the practices
of survey statisticians when they design the scope and coverage
of enterprises surveys.
As broadly defined, the international concept
distinguishes between two sub-categories of informal sector units:
i) "the family enterprises" comprised of independent or own-account
owners, family workers, apprentices and casual workers, and with no
permanent employees; and ii) "the micro-enterprises" comprised of
units with less than 5 to 10 employees (or jobs), or which do not
register them, or which are not registered as enterprises.
Activities in the informal sector which are particularly difficult
to capture in statistical surveys are:
- home-based workers, be they outworkers (or piece-rate workers) or
own-account workers, and more generally, all activities undertaken
in domestic premises,
- itinerant or seasonal or temporary jobs on building sites or road
works, - second jobs or pluri-activity,
- street vendors.
The outworkers and the street vendors are probably the most important
and the most challenging for a better understanding of the informal
sector. This paper deals with street vendors. It presents data on
this category of workers from several African countries and discusses
the concept or definition used to identify street vendors and methods
of enumeration.
I. Selected data on street vendors: share in employment and levels
of incomes
The importance of street vendors in informal employment, and especially
in female employment, is attested by specific and comprehensive data
collection operations in various countries, particularly in Africa
where this type of activity is widespread.
Standard labour force surveys and population censuses are usually
not good instruments to identify street vendors. The main sources
for measuring street vendors have until now been the establishments
censuses or special enumeration. The number of street vendors can
be compared with total employment in the informal sector, as measured
through establishments censuses. However, it should be noted that
establishment censuses do not include outworkers or workers in domestic
premises.
Street vendors represented around 30 per cent of informal employment
in Niamey, Niger, in 1982 (Charmes 1982) and 38 per cent (or 31.2
per cent of total employment) in the 5 main towns of Guinea in 1987
(DGSI/PAGEN, 1987). In Guinea, commercial activities accounted for
more than 80.6 per cent of total employment in street vending and
street vendors for more than 63.5 per cent of total commercial employment.
Unfortunately, it is not possible to disaggregate these data by sex.
Sex disaggregated data were collected and tabulated in the 1992 surveys
in Benin (Maldonado, 1994 and Charmes, 1996). Using the same methods
of estimation as in Guinea and Niger, street vendors in Benin represented
80.7 per cent of all economic units enumerated in urban areas (the
10 major cities), and women accounted for 75.2 per cent of the street
vendors. These figures corresponded to 68.5 per cent of urban informal
employment and 63.5 per cent of total urban employment (outworkers
were not taken into account). Women's share rose to 61.5 per cent
of total labour force engaged in the informal sector, and to 52.8
per cent of the urban informal labour force. Street vendors accounted
for 33.7 per cent of total urban informal labour force, of which 76.7
per cent were women. As a whole, female street vendors represented
about a quarter (25.9 per cent) of the total urban labour force engaged
in the informal sector, and 23.7 per cent of the total urban labour
force (modern sector and outworkers included).
If we consider that Benin is not a special case in this respect -
and there is no reason to believe it is - then the preceding figures
should be regarded as illustrative indicators of the numerical importance
of street vendors in sub-Saharan economies, at least in West Africa.
Street vendors are more than one third of the urban informal labour
force and more than 30 per cent of the total urban labour force; female
street vendors account for more than a quarter of the urban informal
labour force and for more than 20 per cent of the total urban labour
force.
Figures on levels of incomes earned by street vendors are more scarce,
but when they have been collected, they proved to be higher than the
usual assumptions and theories. Usually considered as subsistence
or survival activities, street trade provides incomes several times
higher than the legal minimum wage (4,2 times in Niamey, 1982). A
high proportion of street vendors (around 40 per cent) declared that
they were not seeking a job in the modern sector. They preferred going
to work in a fixed and well established shop. Ten years later with
the effect of structural adjustment programmes on the purchasing power
of urban populations, street vendors' incomes in the cities of Benin
were still favourable when compared with the legal minimum wage (ranging
from 1.1 to 4.5 times the minimum wage according to the kind and the
branch of activity).
In Latin America (PREALC, 1988) and in Asia, studies undertaken on
street vendors referred to the working poor and activities for survival,
and remained qualitative. In Africa these activities have been tentatively
approached on a statistical basis. It is interesting to note that
at the 1987 International Conference of Labour Statisticians when
the concept of informal sector was discussed for the first time, the
delegate from Kenya objected to the use of the term "moonlighting"
to describe it. In his country, the term "Jua Kali" ("under the burning
sun" in Swahili) was the phrase used to refer to street vendors, the
main component of the informal sector. Far from moonlighting, they
were operating in the burning sun and not deliberately concealing
their activities, although they were unregistered. |
| |
II. Methods for
measuring the street vendors
1. Content and characterization of the concept
When dealing with the question of the enumeration of street vendors,
the place of work is the first and main criterion used to characterise
this category. In a few countries, a location of work question was
added in a short and very rudimentary way to the questionnaires of
population censuses (for instance in Morocco or in Tunisia). One aim
of the question was to separate the outworkers from the mass of the
independent or own-account workers. The question was as follows: "what
is the location of work or the type of workplace?" The alternative
answers were:
- an enterprise or a workshop
- the home
- mobile.
Labour force surveys may offer a greater choice of answers: for instance,
the Mali 1985 Demographic Survey distinguished the following 10 types
of location (this classification has been used in all subsequent labour
force or informal sector surveys):
- enterprise
- shop or workshop
- building sites or road works
- fixed market
- mobile market
- home with specific outfits
- home without specific outfits
- street
- mobile
- other.
It would be useful to add collective courtyard to this list.
Other countries (for example Guinea in 1987, Niger in 1982) have further
specified street vendors in fixed locations by distinguishing:
- street vendors with only bowls, baskets or mats
- street vendors with stools
- street vendors with tables (called table-owners or dressers in Niger)
- street vendors with porch-roofs or sheds, or window dress.
The categories for mobile street vendors (hawkers and peddlers) were:
- walking street vendors
- street vendors with cart, bicycle, etc...
The recent and rapid expansion of this segment of the informal sector
labour force which operates outside an enterprise's premises has enlarged
the concept of street vendors in many countries to the category of
street workers including, among others, the following: tailors specialised
in mending, carrying their sewing machines on their heads, hairdressers
carrying their stools, cheap and fast meal sellers, cycles and motor
vehicles' repairers and so many other services workers. Such workers
for a long time have taken to the pavements and the streets of the
towns. More recently manufacturing activities such as furnitures'
makers or metal workers are leaving the courtyards to work on the
street. The share of street vendors in the crowd of street workers
has tended to drop. To take account of these changes, two items should
be added to the classification of location:
- street vendors without fixed premises
- street vendors with rudimentary fixed premises.
In addition to the binary classification: sedentary/non sedentary;
an intermediate category, semi-sedentary, was added to the 1992 census
of establishment and informal sector in Benin. The semi-sedentary
category aimed at accounting for those activities undertaken in the
streets but with a kind or appearance of rudimentary fixed premises,
which provide not only self-employment, but also, eventually, more
or less permanent jobs to family members or other members of the labour
force. The non sedentary street vendor may have a fixed place in the
street, but has to remove the goods at the end of the day. The semi-sedentary
street worker may leave his intermediary and final products on the
spot.
Such structural changes in the characteristics of street workers imply
that surveys on these types of workers need to come closer and closer
to classical informal sector surveys. They need to distinguish production,
trade and service activities, and no longer consider street activities
as a phenomenon that will disappear.
It should be emphasised that greater attention needs to be given to
similarity between street workers and outworkers whose tasks are often
sub-contracted from large firms. Street vendors or workers might not
be as independent as they appear: they may purchase or hire from the
same supplier the goods they sell, or they may be given the goods
by the supplier who pays more or less the equivalent of a salary.
The occupational status of street vendors is not easy to identify.
As for outworkers, it is a challenge for data collection. Similar
issues of questionnaire design and data collection method are raised
in enumerating street vendors and homeworkers.
Two additional issues should be considered related to the differences
between street vendors in rural and in urban areas. First, street
vending is not only an urban phenomenon. Perhaps even more than in
urban areas, in rural areas the non-agricultural labour force is located
outside enterprises' premises. Vendors are particularly numerous along
roads that cross villages or at the cross-roads, and many farmers
or family workers who are classified in primary activities for their
main jobs, are road vendors or market vendors for their second (annual
or seasonal) jobs. Second, trade and sales activities may concern
goods produced by the same persons on their farms or in their homes
and this represents a conceptual and methodological difficulty in
rural as well as in urban areas. To sell self-produced goods should
not be considered as a different activity from producing them, except
if there is a kind of transformation (such as crushing the grains
or cereals, but this will not be the case for fruit or vegetables)
or if they have been carried long distances to be sold in market places.
This is not a marginal point concerning the measurement of women's
activities. It is probably an important source of underestimation
of their contribution as far as this contribution is limited to commercial
margins and does not take the value added in the production process
into account. This is important for a correct enumeration of street
vendors, because many will have declared themselves as producers in
households surveys while they will be registered as vendors in establishments
censuses. This must be clarified through a set of questions in these
latter operations.
2. Data collection methods
As already mentioned, the location or place of work is the main feature
for identifying street workers. Population censuses have been able
to provide information on the number of people involved in these activities
as far as they have put these items in the questionnaires, through
a specific question, or at least as categories of response covering
occupation and status in employment (specifically own-account workers
in status of employment). Further progress would require a more detailed
classification of occupational statuses. Moreover, the international
standard classification of occupations also identifies a category
of hawkers and peddlers in commercial occupations, but it is relegated
to the 4 digit sub-category and as such is very rarely released in
the official publication of the results.
A technique for estimating home-based workers and street vendors from
the available sources is as follows: the residual obtained by comparing
the results of a population census (or labour force survey if no census
is available) and establishment census can be interpreted as consisting
of outworkers or undeclared or non registered workers in given branches
of activity and as street vendors in other commercial branches. Until
now, establishments censuses have been the most efficient source to
capture the street vendors segment, when these operations decided
to include them in their scope as in the three countries already mentioned
(Niger, Guinea and especially Benin). As far as they have fixed places
in the streets (tables or stools or even baskets), the enumeration
is not difficult. In addition in Benin, 1992, the mobile street vendors
who often gather in a few places known by everybody, were enumerated
after a campaign through the mass media. A vendor, once enumerated,
was given a badge to avoid double counts. Mobile street vendors in
urban Benin accounted for more than 40 per cent of the total number
of street vendors (informal transportation being included in this
figure).
In Benin, Guinea and Niger, enumeration was accompanied by the administration
of a short questionnaire collecting a few items on the demographic
characteristics of the vendor and the characteristics of the activity,
leaving to a second sampling phase collection of all the data required
by researchers and policy makers. Given the difficulties of sampling
this population, alternative methods can be used, for instance the
administration of a detailed questionnaire in the first and unique
phase, to 1 out of 2 or 3 or 4 street vendors enumerated.
Street vendors remain a major contradiction and possibly a black hole
in the mixed surveys, i.e. the two-stage survey which is recommended
to capture the various segments of the informal sector. At the first
stage, a representative sample of households is selected: all own-account
workers and employers in the informal sector are enumerated in the
selected households. At the second stage, all economic units of these
informal operators are surveyed with an establishment questionnaire
and preferably on the worksite. While own-account workers are to be
interviewed at their workplace, it is difficult or even impossible
to find the workplace of street vendors. Based on their declaration
within the household, they are usually interviewed at home and the
objective of checking their declarations with the observations is
not achieved, leaving room for underestimation and misunderstandings.
This does not mean that mixed household surveys are useless to measure
the numbers and characteristics of street vendors. But they certainly
need to be followed by in-depth surveys in the field and in establishments
to prevent as much as possible the underestimation of these activities.
3. Questionnaire design for a special street vendor survey
It is important to ensure that the design of the questionnaire for
a street vendor survey takes into account the main issues concerning
the informal sector, and those relating to gender. In addition to
the questions on location described in the preceding paragraphs, the
questionnaire should include the following topics:
a) demographic and socio-economic characteristics of the person: age,
sex, educational level, matrimonial status, family status (living
with parents or head of household -- it may be interesting to develop
a complete household questionnaire which would detail size of family,
occupational status of the head, number of persons employed, unemployed,
at school, sources of income, etc... ), geographical origin or place
of birth, date of arrival.
b) biographic characteristics and previous job experience: rural to
urban migration itinerary and various activities undertaken before
entering the present job; comparative degree of satisfaction and level
of income with the present situation; questions on the modalities
of installation, help received from relatives, ease of entry or barriers
and obstacles encountered due to the regulatory framework or to other
reasons should be put.
c) present activity: detailed description (type of goods or services
sold), type of installation (see above). This section should be developed
along the lines presented above. Examples of questions follow:
i) Are you?
- an independent own-account worker working alone : with other family
workers, (several hours a day, or only when you are absent) / with
casual employees
- associated with other persons on an own-account basis, (in such
a case, develop questions on the type of association)
- a wage employee, (in such a case, develop questions on the employer
and the type of salary)
ii) Do you (or your family) produce the goods you sell? (in such a
case, develop the questions on production)
iii) Do you buy the goods you sell?
- from several shopkeepers or persons: always the same / not always
the same
- from one single shopkeeper or person: always the same / not always
the same (in such a case, does this person sell the goods to other
street vendors like you)
- from a big store
iv) What are the conditions of purchase?
- at the same price as for any customer
- goods are paid for when they have been sold
- goods are consigned against payment of a fixed amount which will
be completed after sale of the whole: in this case, is there a charge
for credit
v) At the end of the day (yesterday), how much have you sold?
vi) Have you deducted from this amount what you have taken?
- for eating
- for the needs of the family
- for other needs
vii) Of this amount, how much are you going to use to buy new goods
to sell the next day?
- What do you do when you cannot save enough money to reconstitute
the amount of goods you are used to sell?
- What do you do when you save more money than you need to spend to
purchase the goods you sell?
The major difficulty in trying to reconstitute the income earned from
street vending is choice of the period of reference: if the day (or
the week) will be the most adequate for the receipts and for the personal
and family consumption expenditures, the purchase of goods will probably
require a different reference period.
Another important but easier question is the duration of work: number
of hours per day, number of days per week, number of months per year.
It is also important to ask what are the nonworking months. This is
to verify whether street vending is still associated with seasonal
agricultural activity and if it is related to rural-urban migrations.
In Niger for example, due to the intensity of rural-urban migration
during the agricultural season, the authorities put street vendors
on trucks which brought them back to their villages where labour shortages
were observed.
d) free entry and competition: these aspects of street vending are
as important as the questions related to income, because they are
at the core of theories of development and the determination of policy
interventions by the State, donors agencies and grass-roots organisations.
While ease of entry continues to be a characteristic of street vending,
street vendors have constant fights with authorities and competing
vendors. As Victor Tokman put it (PREALC, 1988), street vendors must
defend their sites against the municipalities, the established shopkeepers
and the other vendors who try to limit the number of entries. It is
an important challenge in surveys on street vendors to measure the
degree of freedom and competition, and to identify the hidden barriers
that keep the poor from developing their private initiatives.
Among various questions on these topics, three are emphasised here:
i) What conditions made it easy for you to begin vending and what
conditions made it difficult ?
- completely free to set up
- received the help of other people (or vendors), and what kind of
help
- from the family
- from the extended family
- from the village (or district)
- from the ethnic group
- from others
- had to make payments (in kind, in cash, in per cent of receipts,...)
to
- the agents of the municipality
- the agents of other administrations
- the established shopkeepers
- the other street vendors
- other individuals representing the shopkeepers, the other vendors,
keepers or protectors of the street, the block, the district
- other payments
ii) And if you have a place, is it easy or difficult to keep your
site, and why ?
- various payments that you have to make
- various obligations that you have to respect
iii) Among all these payments and obligations, which ones do you consider
as usual or natural, and which ones as excessive or exorbitant?
The questionnaire should also cover the following topics:
e) urban policies and taxation: preparation of this section will require
information on the various formal and informal taxes that street vendors
have to pay.
f) Social insurance, traditional solidarities and social capital:
how do street vendors manage when they fall sick or when they have
an accident which prevents them from working? Can they rely on their
traditional cooperative networks and are the investments they have
made in this form of social capital more profitable and efficient
than the payments they may make to informal protectors in the market?
There are different means of measuring the social capital (Charmes
1998). A survey on street vendors - unless it is part of a household
survey - will provide qualitative responses (rather than measures
in terms of expenditures and time use) concerning the ways and means
by which street vendors cope with the costs of social insurance and
can rely on the social capital they have accumulated.
g) needs, organisation and support: finally, a last section of the
questionnaire should deal with the wishes of the street vendors, the
ways and means they imagine or require to improve their working and
living conditions. |
| |
III. Steps to improve
measurement of street vendors
As a first step, some data on street vendors may be obtained by household
surveys (labour force as well as informal sector surveys). These surveys
will be more useful if more detailed and appropriate classifications
were introduced.
In connection with the classification of status in employment (ICSE),
a specific question on the place of work need to be introduced in
household surveys. This should not be difficult if an agreement is
reached on a classification of worksites. The solution adopted by
a few countries, to develop the classification of statuses only for
the independent own-account workers, is insufficient, given the rapid
changes that are also occurring in the categories relating to street
vendors: employers, employees, and family workers. This issue is also
relevant to the identification of home-based workers. The current
ICSE does not allow clear distinctions within the labour force. Place
of work needs to be introduced in all the categories of the ICSE.
Another improvement in classifications is needed to clearly identify
what activities are to be put in the category of street vendors. Censuses
and surveys will enumerate street workers, including those involved
in manufacturing (tailors, furniture makers, etc.) and services activities
(transportation, restaurants, hairdressers, etc.) and a necessary
distinction will have to be made later, on the basis of a clear classification
and in terms of geographical areas, differentiating urban streets,
rural roads and cross-roads, and in general mobile and permanent markets.
The implementation of an establishment census and follow-on survey
approach, is the only means to capture the diversity and complexity
of the category of street vendors. A household approach is too far
from the reality in the field of work, and a mixed survey (household
+ establishment) would have two major shortcomings:
- in the sample, the number of street vendors would be insufficient
to seize the diversity of the activities because of the over-representation
of the most common activities
- it is necessary to interview the respondent at his workplace, at
least at one stage of the survey, but it would be extremely difficult
to come back to the street vendors' places, on the basis of household
declarations.
As to the questions to be specifically and appropriately addressed
in the surveys of street vendors, such as: number of vendors, levels
of incomes, degree of freedom and barriers to competition and private
initiative, role of solidarities and social capital, they require
a comprehensive and qualitative preliminary approach, adapted to the
specifics of each country, which will be used as an input in the design
of the methodology and questionnaires.
Efforts need to be made in order to capture:
- the real incomes (through the adaptation of the period of reference
and the formulation of the questions) as well as the real costs and
expenditures made necessary for the exercise of these activities
- the reality and consequences of free entry, free competition, informal
and formal taxes and barriers
- the linkages between street vendors and the modern sector (sub-contracting),
and between street vendors and the countryside (rural-urban migration)
- the living conditions of street vendors, the social protection which
they can afford, through official urban or national policies, traditional
solidarities, and social capital.
References
Charmes J. (1982): Le secteur non structure ? Niamey. Rapport d'une
enquete. Bureau International du Travail. Programme des Emplois et
des Comp?tences Techniques pour l'Afrique (PECTA). Addis Abeba - 104
p.
Charmes J. (1996): Situation et perspectives de la population active
et de l'emploi au B?nin: 1979-2002. Minist?re du Plan, de la Restructuration
Economique et de la Promotion de l'Emploi, PNUD, Cotonou, 58p.
Charmes J. (1998): Ostentation, solidarit?s, protection sociale. Des
d?penses ostentatoires ? l'accumulation d'un capital social. Vers
une convergence des interpr?tations des ph?nom?nes ?conomiques dans
les soci?t?s africaines? ORSTOM- Universit? de Versailles-St Quentin
en Yvelines-C3ED.Multigr. 4p.
DGSI / PAGEN (1987): Recensement des ?tablissements dans les villes
de Conakry, Kindia, Mamou, Lab?, Kankan (Guin?e). Minist?re du Plan
et de la Cooperation Internationale, Direction G?n?rale de la Statistique
et de l'Informatique, Conakry, 1987.
ILO (1972) : Employment, Incomes and Equality. A Strategy for Increasing
Productive Employment in Kenya. ILO, Geneva.
King K. (1996): Jua Kali Kenya. James Currey, London, 1996.
Maldonado C. (1994): Recensement des ?tablissements en milieu urbain
au B?nin. PEESI / PNUD / BIT / INSAE, Gen?ve, Cotonou.
PREALC (1988): Sobrevivir en la calle. El commercio ambulante en Santiago.
PREALC, Santiago, 173p. |
| |
|
