12.4 Frames for informal sector surveys#
12.4.1 Definition of the informal sector#
The informal sector represents a very important part of the economy in many countries. SNA 2008 recommends the use of Resolution II of the Fifteenth International Conference of Labor Statisticians (ICLS) 1993 (🔗) in defining the informal sector. The Resolution requires the informal sector to be defined as a subset of household unincorporated enterprises, comprising those enterprises that:
produce at least some output for the market; and
are less than a specified size in terms of the number of persons engaged, or of employees or of employees employed on a continuous basis; and/or
are not registered under specific forms of national legislation, such as factories or commercial acts, tax or social security laws, professional groups’ regulatory acts, or similar acts, laws or regulations established by national legislative bodies.
Enterprises that are involved in agricultural production may, optionally, be included or excluded; conceptually, they should be included.
The registration (third) criterion is becoming the most commonly used in practice.
It is important to recognize that these criteria were a practical compromise that provide a framework for a definition of the informal sector, rather than a single, unambiguous definition. They reflect variations across countries in how the informal sector is actually defined. For example, the Philippine Statistics Authority defines the informal sector as comprising business that are a single proprietorship in terms of legal organization, a single establishment in terms of economic organization and have a total employment of less than ten persons.
On the other hand, the Guidelines for Building Statistical Business Registers in Africa (🔗) suggest defining the informal sector as the set of non-agricultural household enterprises that have market production but that are not registered with the administrative source(s) on which the coverage of the SBR is based. This definition is in accordance with the Resolution and has the virtue of making the formal/informal boundary very visible and easy to implement. It means that:
the suite of business surveys based on the SBR and administrative data measure the formal sector;
non-agricultural household market enterprises not included in the SBR constitute the informal sector; and
agricultural household market enterprises not included in the SBR constitute the informal agricultural sector.
At the 2018 International Conference of Labour Statistician (ICLS), the ILO received a strong mandate to initiate a revision of the 15th ICLS resolution concerning statistics of employment in the informal sector and the 17th ICLS guidelines concerning a statistical definition of informal employment. A working group has been established, and a new set of standards based on current definitions and country practices will be prepared for the next ICLS, presently scheduled for October 2023.
12.4.2 Relationship between the informal sector and SBR#
As noted in the Chapter 12.3 — Statistical Business Register, a modern SBR is based on administrative data and provides coverage only of the businesses registered with the administrative sources, thereby excluding most or all (depending upon the definition) of the informal sector. The temptation to try and expand the coverage of the SBR to include the entire informal sector should be resisted as there is no sustainable way of doing so.
Thus, frames for surveys of the informal sector, or for an economic census that includes the informal sector, must be otherwise derived. This is addressed in the subsections below.
Three situations are discussed:
frame for a survey of informal businesses with visible premises, based on area enumeration;
frame for a survey of informal businesses conducted by households, based on a household frame, and
frame for an economic census with an informal component.
The following subsections also indirectly address some of the problems in providing survey frames that are faced by countries that do not have a modern SBR either because there are no suitable administrative sources, or because the NSO cannot access them.
12.4.3 Measurement of the informal sector#
The ILO publication Measuring informality: A statistical manual on the informal sector and informal employment (🔗) is a technical guide for NSOs. It focuses on technical issues that are specific to the production of statistics on informal employment and the informal sector. It presents the currenzt international standards on employment in the informal sector that were adopted by the ICLS in 1993 and the guidelines on informal employment adopted by the 17th ICLS in 2003. It provides practical guidance on implementing the standards and presents three basic approaches regarding the type of survey that can be used to collect the informal sector data, together with the strengths and limitations of each. They are:
household-based surveys with a labour force component, notably labour force surveys;
establishment-based surveys and censuses of production, and
mixed surveys, where the first phase is a household survey, and the second phase is an establishment survey of the businesses obtained in the first phase; the first phase can be a standalone survey or the module of an existing household survey.
Although it describes these approaches in detail, the Manual should not be regarded as a comprehensive guide to labour force or other household surveys, or to enterprise surveys for which there are dedicated manuals. Also, it should be noted that a population census or household survey may collect data relevant to the informal sector (such as occupation, industry, class of worker, and place of work of household members) without specifically having an informal sector module.
The following paragraphs describe the provision of frames for three types of surveys involving the informal sector and the mixed survey approach.
Frame for survey of informal businesses with visible premises
As noted in the Manual, a comprehensive survey frame for informal businesses can be created only by area enumeration. It may benefit from information obtained by an earlier enumeration, for example from an economic census that has been conducted not too long previously. In any case, the country is divided into enumeration areas and a list of all businesses with visible premises within a sample of these areas is obtained and used as the basis for the survey frame. The set of businesses thus identified will not define the informal sector in accordance with the ICLS93 framework as businesses without visible premises are excluded.
As such a list is likely to fall rapidly out of date, unless the survey is to be repeated annually there is little or no point retaining the list for the next cycle. There is certainly no point in attempting to use it to create or update an SBR.
Frame for survey of informal businesses conducted by households
Such a frame can be created by area enumeration of households, or, more probably, by use of a household frame that has already been created for some other purpose, for example the conduct of a labour force survey or a population census. Each household in the household frame is asked if they conduct a business and those that do form the frame for the informal sector survey of the businesses. During the course of the survey, those businesses that are found to be in the SBR are eliminated from the survey if the SBR is considered to define the formal sector.
Such a frame gives broader coverage of the informal sector than a survey based on area enumeration of visible premises as it includes businesses conducted by households from their own homes without visible premises.
Frame for an economic census with informal sector component
The appropriate method for constructing a frame for an economic census depends upon the target coverage of the census. If the target coverage of the census is all businesses in the formal sector, then the SBR can provide the frame, and the results can be used to update the SBR. There is no coverage of the informal sector.
If the target coverage of the economic census is all business with visible premises, then area enumeration as described above is required to cover the informal businesses with visible premises.
In a dual-frame approach, the resulting frame is used in combination with a frame from the SBR that covers businesses in the formal sector. To avoid duplication, businesses that are in the SBR frame and in the area-based frame, are identified, marked as duplicates and included only in the SBR frame.
Feedback from the census concerning businesses in the SBR is used to update the SBR. As the area-based frame cannot be maintained and is likely to fall rapidly out of date, there is no point in attempting to maintain it or use it to expand the coverage of the SBR.
If the target coverage of the economic census is all incorporated businesses and all household businesses, then a household-based frame as described above is required to cover the household businesses. It is used in combination with the SBR in a dual frame approach (as also described above). Again, feedback from the census concerning businesses in the SBR is used to update the SBR, but data about the informal sector business should not be used to expand the coverage of the SBR.
Mixed surveys including the 1-2-3 Survey Approach
The 1-2-3 survey is a type of mixed modular survey. It is an arrangement of three interlocking surveys, as shown in Figure 15. Sometimes only the first two phases are applied, referred to as a 1-2 survey.
The first phase is the creation of the frame by a household survey such as a labour force survey. Persons that are identified as owners and operators of informal household market enterprises form the frame for Phase 2. In Phase 2, more detailed information is collected regarding the business activities. Phase 2 can be carried out at the same time as Phase 1 or later (but not too much later). If later, a sub-sample can be used. Phase 3, if conducted, involves collecting data on consumption and living conditions.
Fig. 15 1-2-3 Survey#
The 1-2-3 survey approach was first used in Mexico at the end of the 1980s. Initially designed to study the informal sector, the approach was gradually extended to measure and monitor poverty and governance. Over the last few years, the approach has spread to many countries in Africa, Asia and Latin America. The first two phases of a 1-2-3 survey are a labour force survey and an informal sector survey. The third phase is an income and expenditure survey, administered to a subsample of households identified in the first phase, and designed to estimate the weights of the formal and informal sectors in household consumption, by product and type of household. Ad hoc surveys can be added to any of the three phases to obtain additional information on, for example, access to micro-credit, social insurance and taxes.
12.4.4 Examples of national practices#
Afghanistan – Integrated business enterprise survey, 2009
The integrated business enterprise survey carried out in Afghanistan was the first survey of its type in the country. It covered all non-agricultural economic units in all urban and most rural areas and accounted for about 81 per cent of the total rural population. As it lacked a complete up-to-date sample frame for all units in the country, it used a dual sample frame. A list-based frame was used for large establishments (generally with 20 or more workers) and an area frame for villages in rural areas and ‘controller’ (or enumeration) areas in urban areas. The area frame was constructed using the 2004 population census list of villages and ‘controller’ areas. The population censuses did not provide information on the concentration of establishments or their industries, so this information could not be used to create strata. Areas in urban and rural areas were therefore stratified only by province; in urban areas ‘controller’ areas were also stratified according to the concentration of establishments, with areas with markets or a high concentration of enterprises forming one stratum separate from other areas.
All large enterprises in the list-based frame were included in the sample.
The sample-based on the area frame was selected in two stages. In the first stage, 150 villages in rural areas were selected using systematic sampling with probabilities proportional to population. In urban areas, 150 controller areas were also selected with systematic sampling but with equal probability. Each controller area was divided into around five smaller segments, from which two were selected, one with the highest concentration of units and another at random.
In the second phase, all economic units within the selected villages and segments were listed. The listing included all units operating in fixed premises, with the exclusion of those already in the list-based frame, as well as all household-based and itinerant activities. This was achieved by inquiring in each dwelling, whether some economic activity was being carried out there or was based there. Activities carried out within the dwelling were included in the list, as well as activities of household members that were carried out in no fixed premises. Activities carried out by household members in fixed premises outside the dwelling were not included in the area frame, as they were listed at the village or controller area in which the activity was carried out.
Ethiopia – Census of economic establishments, 2004
The first census of economic establishments in Ethiopia was carried out by the Central Statistical Agency (CSA) in February-March of 2004. The census covered all urban establishments that were set up with the aim of making profits and covered both public and private establishments. The census included units that sold or produced in open markets or in a fixed location and mobile businesses in all sectors, with no limitation as to the number of persons engaged and the amount of income. It excluded civil service institutions and defence and other non-profit-making government and non-governmental organizations.
The fieldwork for the census was carried out without many problems. However, the reporting quality of some data items, particularly those on capital and revenue, was considered to be unreliable and the NSO was not capable of setting up a reliable SBR based on the data collected. Thus, the CSA’s subsequent thrust is to establish a reliable SBR by conducting well-designed censuses of economic establishments and enterprises, and to use administrative data for supplementing and maintaining it.
National survey of micro-enterprises (ENAMIN), Mexico
Over the period 1992 to 2012 INEGI, Mexico conducted its National survey of micro-enterprises (ENAMIN) on a regular basis. The survey encompassed all non-farm activities without prejudging if they were formal or informal. For trade, transport, construction and services, it included units with six or fewer workers (including the owner or manager); for manufacturing, it included units with 16 or fewer (including the owner or manager). Mexico adopted only the two first stages of the 1-2-3 survey because an income and expenditure survey was already well established.
The first phase of the survey was the labour force survey, which has a questionnaire that already included the necessary elements for identifying entrepreneurs in the informal sector, as well as employees and contributing family members. In the second phase, the survey was directed to heads of non-farm micro-economic units (including professionals) operating either with or without premises, whether the business in question is the main or a secondary activity.