Tool 8 - Analysing Employment Distribution

 

Introduction Objectives Key Questions Steps Step 1     Define the categories of actors Step 2     Determining employment at each level Step 3     Calculate the employment distribution at different levels of the value chain Step 4     Analysis of the employment distribution contribution Step 5     Determine the impact of Governance on employment Step 6     Determine the impact of technology structures on employment Step 7     Determine the employment variability over time What Should be Known after Analysis is Complete

Introduction

 

As with income distribution, analysing the distribution of employment within the value chain is central to understanding how to increase the participation of the poor.  Understanding how employment is distributed along the chain provides the necessary start to determine opportunities for employment generation. The distribution of employment and access to employment by different wealth classes can be analysed to identify employment opportunities. See also the mapping suggestions in Value Chain Toolbook - Part Two (Tool 2) for visual representations of employment distributions.

 

Analysing distribution of employment is not only an analysis within a particular value chain but also recognises that individual actors participate in a number of different value chains at the same time. For example, a farmer may be involved in several agricultural crops and several handicraft activities as a means of income diversification. In the same way a trader might be involved in trading multiple agricultural products at the same time or at different times depending on the season. Therefore, livelihood strategies made by various actors are influenced by labour constraints and any analysis must take this into account.

 

The second part of this tool looks at whether there is room for improvement in the distribution of labour and how this can be done, taking into account seasonality in demand and availability of labour and also the competitiveness between labour intensive and labour saving upgrading strategies.

 

Objectives

 

Key Questions

 

Pro-poor dimensions in the distribution of employment are:

 

• What are the opportunities / barriers for the poor to find employment in the value chain?
• Is it more interesting for the farmer to work on his own farm or switch to wage labour within or outside the chain?
• Which age groups do have the chance to access employment?

 

Steps

 

Step 1     Define the categories of actors

 

To analyse employment distribution within a value chain it is important to first categorise actors.  The mapping of the value chain as discussed in Value Chain Toolbook - Part Two (Tool 2) provides a map of actors within categories and this can be used as a basis to add employment specific information.

 

 

There can be different types of farmers, collectors, wholesalers and retailers. As was the case with defining the categories for income levels along a value chain in Value Chain Toolbook - Part Two (Tool 7), the most important categorisation for pro-poor value chain analysis is based on income levels (a distinction between poor and non-poor actors).

 

For example, for flower retailers in Hanoi (Vietnam) there are at least three different broad categories; hawker, retailers in open air markets, and retailers in their own flower shops. These retailer categories are very much related to the different wealth levels, with hawkers being the poorest. Other examples of categories that could be used are presented in Box 1 below.

 

Tool 8 - Box 1: Examples of other categories of actors

Categories	Dimensions
Skills	Unskilled, low-skilled, high-skilled
Gender	Male or female
Ethnicity	Different ethnic types
Business Type	Micro, small, medium, large
Period	Day labour, temporary labour, permanent labour
Status	Family, hired
Origin	Temporary migrant, permanent migrant, locally hired

  

Take Note

Within specific groups it may be important to look at age distribution. For example, in rural Vietnam it is becoming obvious that the average age of farmers is increasing because younger people find it easier and more attractive to find employment elsewhere. Even if employment opportunities exist this does not mean it is open to each age group, gender or social group.

 

Step 2     Determining employment at each level

 

By comparing the distribution of employment across each level of the value chain a comparison of opportunities for the poor at various levels of the chain can be made. This is complementary to the analysis of the incomes accruing at each level of the chain.

 

Employment at each level of the value chain can be determined in different ways:

 

1. Wholesalers:  Conducting a survey of wholesalers is generally not too time-consuming. Be aware of seasonal variations; in the off season the number of wholesalers is much smaller than in the main season.
2. Retailers: Based on the total traded volume of a product in a value chain and the daily turnover of a retailer one can calculate how many retailers are involved. But if additional time is available count all retailers in a sample area (e.g. open air market retailers) and then apply the figures to calculate the retailers in a total area. For example, count the total number of open air markets in a city (e.g. 130) and then take a random sample of various open air markets (e.g. 15). Visit these open air markets, count the number of retailers in these markets or ask the market administrator (if present) how many booths he rents out. Calculate the average number of retailers per open air market and multiply by 130 to get a rough estimate.
3. Transporters: Estimate the total volume of sales, and the typical volume per transport unit (e.g. trucks, motorbike, carts, boats). Then estimate the number of people required per transport unit, the time required to transport, and the number of full time equivalent employees (FTEs) this generates. 
4. Processors:  Identify the number of processors in an area from official sources (e.g. registration certificates); identify the number of informal processors from key informant interviews.
5. Collectors: Conduct interviews with village leaders or commune heads. Estimate the number of collectors under each trader/wholesaler. Estimate the total volume of sales, and the typical volume per transport unit. Then estimate the number of people required per transport unit, the time required to transport, and the number of FTEs this generates.  
6. Farmers: Estimate the number of farmers based on hectarage under each crop and yields (related to traded volumes). Cross check with district authorities for official figures. Obtain information on sales of key inputs sold by input providers at bottleneck points (e.g. seed). Be sure to distinguish between smallholders and commercial farmers.
7. Hired labourers: Estimate from partial budgets and scale up.
8. Input suppliers: Seed, fertiliser, nurseries, breeding station owner. Estimate volumes demanded in the market and volumes provided by the average input supplier. Estimate average employment per input supplier and estimate the total number of FTEs this generates.
9. Service suppliers: Extension, design, marketing etc. Estimate how much of the services provided by the suppliers feed into the specific chain (and not to other chains). 

  

A fast way to get an idea of the number of actors in a value chain is to carry out interviews with wholesalers. Wholesalers are often located in just a few locations and there is usually a small number of wholesalers compared with the number of farmers, collectors or retailers. Through a combination of census counts (counting the total number of wholesalers in a certain location)  and interviews with a number of wholesalers it is possible to get a good estimate of the total traded volume of a product in the value chain (e.g. tons of avocados, or number of roses). Conducting interviews with the other actors in the chain to estimate their typical turnover allows an estimation of how many actors are involved.

 

As many actors in agricultural value chain are only involved seasonally, it could be useful to convert the collected employment data into a standardised indicator. This allows comparisons among various value chains, for example using the number of FTEs as the main indicator for the employment created by a certain value chain. One just simply defines or agrees on how much labour days per year are considered 1 FTE, for example 240 days. If someone only works for 120 days, this is accounted as a half FTE. It is also important to consider both direct and indirect employment – in administration and ancillary services. In another example, farmers can hire labour to work on lower valued crops while they concentrate their own labour on higher valued crops.

 

Take Note

For a quick insight in the employment generation by a value chain focus resources on the use of participatory analysis tools with wholesalers and transporters. They are often concentrated in just a few locations (saves time in visiting) and have a very good overview of traded volumes and the various upstream and downstream channels.

 

Due to employment diversification strategies, the employment in one value chain may be only a small fraction of the total employment of a household; especially for service activities all along the chain. The share of employment represented by the value chain should be calculated to accurately model livelihoods and livelihood responses. In the example in Table 1 below, the share of employment in different livelihood activities was calculated for farming households in Laos across different income levels.

 

Tool 8 - Table 1: Average utilisation of labour by livelihood activities in Lao PDR 

Farm and Non-farm Activities	Poor	Average	Better-Off
Rice	41.0%	36.3%	35.3%
Root and Tuber Crops (e.g. cassava, potato)	1.7%	0.8%	0.0%
Upland Crops (e.g. maize, other cereals, legumes)	11.2%	9.9%	6.7%
Vegetables	10.7%	9.9%	2.8%
Perennial Crops (e.g. rubber, coffee, pepper)	1.2%	2.6%	1.3%
Annual Industrial Crops (e.g. sugarcane, cotton)	1.0%	1.2%	0.2%
Fruit Trees	2.1%	2.1%	4.9%
Fishing and Aquaculture	0.4%	1.3%	2.2%
Small livestock (e.g. poultry, pigs, goats)	5.4%	10.2%	9.5%
Large Livestock (e.g. cattle, buffalo)	1.8%	3.9%	7.3%
Non-Timber Forest Products	5.0%	1.8%	0.9%
Forest Products	2.6%	2.1%	0.9%
Other Farm Activities	2.7%	0.5%	0.0%
Handicrafts and Weaving	3.0%	1.7%	0.9%
Off-Farm Work and Remittances	10.2%	15.8%	27.2%
Total	100.0%	100.0%	100.0%

 Source:^[1]

 

Step 3     Calculate the employment distribution at different levels of the value chain

 

Conduct field surveys to obtain an indication of the different dimensions of employment at each level of the chain according to the category. These surveys can be short and simple, just to get some idea of turnover volumes per actor (e.g. mean harvested number of roses per farmer per year; or average annual traded volume per collector per day/month/season/year), income levels, or the number of hired labourers.

 

Comparison of employment over different stages in the chain should be undertaken according to the various categories developed in Step 1. This gives a picture of the distribution of benefits to individuals within the framework of enterprises at each level of the value chain.

 

An example of this is given in Figure 1 below, which shows the different numbers of actors at each level of the shrimp value chain in Bangladesh. This can be extended to describe the different categories of actors (poor, non-poor, self-employed, wage earners).

 

 

Note: Faria, Aratdar and agents are specific types of middlemen engaged in the shrimp value chain in Bangladesh

Source:^[2]

Tool 8 - Figure 1: Example of employment over different stages in the value chain

 

Take Note

Estimating the levels of employment at each level of the chain is difficult. The information often does not exist and large assumptions need to be made. For example, if total volumes of production are known, and the average production per farmer can be estimated, then employment at the farm level can be calculated. Similarly, average volumes of trade by individual wholesalers can give an estimate of the number of wholesalers in the value chain.

 

Step 4     Analysis of the employment distribution contribution

 

Comparing the distribution of employment across each level of the value chain enables a comparison of benefits actors are getting at various levels of the chain. This is complementary to the analysis of the margins and the profits at each level of the value chain. However, an analysis of the employment gives a more accurate picture of the true distribution of benefits at each level of the value chain, as it reflects the often vastly different number of players at each level of the chain. A matrix can be developed that shows the numbers of actors by category at each level of the chain; see the example in Table 2 below.

 

Tool 8 - Table 2: Example of analysing the number of actors at each level of the chain 

		Farmer	Collector	Trader	Wholesaler	Input supplier	Service suppliers	......
Number of People	Poor Average Better-off
Volume of sales	Poor Average Better-off
Number of people	Unskilled Low-skilled High-skilled
…

 

Tool 8 - Box 2: Example of employment impact evaluation 

Within the framework of the GTZ Value Chain development program in Vietnam an avocado value chain analysis was carried out in Dak Lak Province. As avocado trees are mostly grown as shade trees or windbreakers around coffee fields, the avocado sector in Dak Lak has not been very visible for policy makers. On average a farmer has about five avocado trees, which might suggest that avocado is not an important product in Dak Lak. Based on data collected during a rapid diagnostic appraisal and a short survey among the 98 major avocado wholesalers in Dak Lak province it was possible to calculate the number of persons involved in the avocado sector. This example only makes estimates of the avocado sector in Dak Lak and does not include all the employment involved of wholesalers and retailers in HCMC, Hanoi and all other cities to which the avocados are transported. Based on the census it was estimated that during the main avocado season, 337 ton of avocados per day are exported from Dak Lak to other provinces in Vietnam. This figure was obtained through very short interviews (max 20 min per wholesaler) with almost all avocado wholesalers in Dak Lak province. These 337 ton per day are only exported during the main season, which lasts four months. Avocado is also traded during the other eight months of the year but in very small volumes. Employment analysis was focused on the main season only, so the data presented below are an underestimation of the employment generated by the sector. Sector size in Dak Lak: Avocados exported by Dak Lak wholesalers 337 ton/day 40,410 ton/season Harvested number of trees 3,368 trees/day 404,100 trees/season Number of farmers involved 674 farmer/day 80,820 farms/season Number of collectors involved 1648 persons/day   Harvested area 22 ha 2,649 ha Truckloads 42 truckloads/day 5,051 truckloads/season   In addition to the 100 avocado wholesalers there are also about 1648 active collectors. These actors play the most critical role in the avocado value chain as they harvest and collect the avocados. They visit the farmers and harvest one or two trees per visit. In total about more than 80,000 farmers are involved, with an estimated harvested area of more than 2,600 ha.   Assumptions for these calculations: Average harvest per tree 100 kg/tree Mean no. of trees per farmer 5 trees/farmer Turnover per collector 200 kg/day Number of trees per ha 150 trees per ha Average truck load 8 ton/truck   These data do not include the employment the sector generates for a business service provider like the bamboo basket makers. All avocados are transported in large bamboo baskets, with each basket containing about 100 kg of avocadoes. This means that every day about 3,368 bamboo baskets are required. As the baskets are recycled and data was not collected about this no estimate was made of the employment generation for bamboo basket makers, but it must be significant.   It was further calculated that the total value added of the avocado sector in Dak Lak province was almost US$ 7 million in every main season. With these data and the employment estimates it was possible to create an increased awareness among provincial policymakers about the economic importance of the avocado sector in Dak Lak.

 Source:^[3]

 

Comparing employment across value chains is a good indicator of alternative activities which households could undertake. In the example in Table 3 below, the value chains for five different sectors in Zambia are compared for employment and income. The results indicate that the domestic horticulture, cotton and tobacco sectors are the ones with the most employment, and that there are significant opportunities for wage employment in the tobacco and export horticulture chains.

 

Tool 8 - Table 3: Income distribution and employment across value chains in Zambia 

Value Chain	Sector Earnings (US$ million)	Wage Employment	Small Farmers	Earnings per Person (US$/day)
Cotton	81	2,300	280,000	$1.30
Tobacco	63	92,000	23,000	$2.49
Sugarcane	65	4,000	1,692	$51.91
Export Horticulture	55	14,500	2,500	$14.71
Domestic Horticulture	116	10,000	525,406	$0.98

 Source:^[4]

 

Step 5     Determine the impact of Governance on employment

 

With this step, a researcher can compare employment across sub-chains of the value chain that have different governance structures (e.g. informal linkages versus contract linkages). The analysis in the steps above can be ungrouped by governance structures. In the example in Box 3 below, the value chain for cotton in Zambia is separated into three governance chains, which show the levels of employment at each value chain stage.

 

Tool 8 - Box 3: Example of employment across different governance structures in cotton in Zamia

 

Source:^[5]

 

Step 6     Determine the impact of technology structures on employment

 

Compare employment across different sub-chains of the value chain that have different technology structures (e.g. supermarket chains versus traditional retailing chains, village rice mills versus commercial rice mills, smallholders versus commercial farms).

 

For example, the development of a supermarket is expected to decrease employment of the poor, due to the use of capital-intensive versus labour-intensive technology in supermarket distribution. Thus, to achieve poverty alleviation objectives, the diversity of retail distribution, including distribution by small-scale markets, should be maintained as much as possible.

 

Also, the opportunities for the poor to participate in the supermarket-driven chain as supplier or trader of produce tend to be fewer because of stricter quality and consistency of supply requirements by supermarket chains as compared to less advanced types of retail distribution.

 

Finally, because of prices tending to be higher in supermarkets as compared to e.g. small scale markets, the poor (as consumers) may suffer if cheaper alternatives are not available.

 

Tool 8 - Box 4: Example of employment generation for poor in a supermarket-driven chain in Vietnam 

Summary of investigated issues on the poor’s access to supermarkets and other Domestic Value Chains (DVCs) in Vietnam.   The case studies show that poor farmers as producers have no direct access to supermarkets because of the requirements of the latter in terms of safety (for vegetables) and quantities (for all products).

 Source:^[6]

 

Step 7     Determine the employment variability over time

 

Look at the changes in employment over time, both within the year (seasonality), as well as between years. Timelines of changes in employment across different sub-chains over a long period (e.g. 5 years) can be very informative and useful.

 

Tool 8 - Table 4: Seasonal labour patterns in houysan village, Savannakhet Province

Activity	Labour		Month
	Males	Females	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Wet Season Rice	50%	50%
Irrigated Rice	50%	50%
Vegetables	10%	90%
Maize	40%	60%
Sweet Potato	40%	60%
Tobacco	80%	20%
Resin	50%	50%
Bamboo Shoots	20%	80%
Daily Labour	60%	40%
Alcohol Making	0%	100%
Blacksmith	100%	0%

 Source:^[7] 

 

Tool 8 - Box 5: Survey questionnaire for calculating distribution of employment 

Labour Use Schedule  Get the farmer to list all farm and non-farm activities and sources of income and livelihood. Put them into the categories below. Using 200 seeds, ask the farmer to partition and weight each activity according to total household labour use over the year. For After the farmer has finished weighting review the results with the farmer. Do pair-wise comparisons between the cells, asking the farmer to verify that the relative weightings are correct. Farm and non-farm activities Jan-Feb Mar-Apr May-Jun July-Aug Sept-Oct Nov-Dec Rice ___% ___% ___% ___% ___% ___% Root and Tuber Crops (cassava, potato etc) ___% ___% ___% ___% ___% ___% Upland Crops (maize, other cereals, legumes etc) ___% ___% ___% ___% ___% ___% Vegetables ___% ___% ___% ___% ___% ___% Perennial Crops (rubber, coffee, pepper etc) ___% ___% ___% ___% ___% ___% Annual Industrial Crops (sugarcane, cotton, etc) ___% ___% ___% ___% ___% ___% Fruit Trees ___% ___% ___% ___% ___% ___% Fishing and Aquaculture ___% ___% ___% ___% ___% ___% Small livestock (poultry, pigs, goats, etc) ___% ___% ___% ___% ___% ___% Large Livestock (cattle, buffalo, etc) ___% ___% ___% ___% ___% ___% Non-Timber Forest Products ___% ___% ___% ___% ___% ___% Forest Products ___% ___% ___% ___% ___% ___% Other Farm Activities ___% ___% ___% ___% ___% ___% Handicrafts and Weaving ___% ___% ___% ___% ___% ___% Off-Farm Work (Not Including Remittances) ___% ___% ___% ___% ___% ___%   Check Sum Total=100%

 Source:^[8] 

 

The questionnaire above can be implemented in the field using a large sheet of card paper, which can be laminated to allow repeated use. The respondent can place seeds on each of the boxes to represent their labour use. The example in  Figure 2 shown below is an analysis of a farming system in Mindano, Philippines. The picture indicates that the household spends an equal amount of time over the year "saging" their banana trees (weeding and cutting on a regular basis) and taking care of their single cow "Baka". They have a second field where they plant maize in July-Oct and rotate with sweet potato ("camote") and squash. Finally, under the banana trees they plant a small bit of taro ("gabi") which they harvest one year later (hence the activities all occur in the Jan-Feb period). 

 

 

Tool 8 - Figure 2: Example of analysing labour utilization using participatory approaches in the Philippines

 

The results of individual respondents can be aggregated within specific categories (location, income level) and presented in a tabular format as shown below.

 

Tool 8 - Table 5: Average use of labour (%) by livelihood activities - poor families in Houysan Village, Lao PDR

Farm and Non-farm Activities	Jan-Feb	Mar-Apr	May-Jun	Jul-Aug	Sep-Oct	Nov-Dec	Total
Rice		5.6	7.4	14.6	6.2	12.0	45.8
Root and Tuber Crops (e.g. cassava, potato)
Upland Crops (e.g. maize, other cereals, legumes)		1.8	2.6	3.4	2.8	2.0	12.6
Vegetables	4.6	3.2	1.2		3.2	3.4	15.6
Perennial Crops (e.g. rubber, coffee, pepper)
Annual Industrial Crops (e.g. sugarcane, cotton)
Fruit Trees
Fishing and Aquaculture
Small livestock (e.g. poultry, pigs, goats)	0.6	0.6	0.6	0.6	0.6	0.6	3.6
Large Livestock (e.g. cattle, buffalo)
Non-Timber Forest Products			1.0	6.4	7.0	1.4	15.8
Forest Products	2.0	1.2	1.0	0.8	0.8	0.8	6.6
Other Farm Activities
Handicrafts and Weaving
Off-Farm Work and Remittances
Total	7.2	12.4	13.8	25.8	20.6	20.2	100.0

Source:^[9] 

 

The table above can be used to carry out additional analysis which can be presented in graphical format, such as the distribution of labour over the year (data presented in the final row of the table). 

 

Source:^[10]

Tool 8 - Figure 3: Graphic presentation of grouped labour use

 

Similarly, an analysis can be carried out to show the labour constraints over time, which may indicate when hired labour is used, and what changes to the production system may need to be put in place to alleviate labour shortages. Using the example questionnaire in Box 6 below, a graphical representation of seasonal labour constraints can be constructed.

 

Tool 8 - Box 6: Example of survey questionnaire for calculating labour constraints 

What are the seasonal labour constraints for the farmers? Get the farmer to place a þ or a x in the appropriate row for each month.     Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Surplus Labour r r r r r r r r r r r r Enough Labour r r r r r r r r r r r r Lack of Labour r r r r r r r r r r r r

Source:^[11]

 

 

 

Source:^[12]

Tool 8 - Figure 4: Graphic presentation of labour constraints by different household types over the year 

 

What Should be Known after Analysis is Complete

 

After having followed all the steps, the key questions outlined below should be able to be answered:

 

1. What are the differences in employment within and between different levels of the value chain?
2. What are the impacts of the distributional outcomes of the value chain on the poor and other disadvantaged groups, both currently and in the future?
3. What are the changes in employment that result from the development of various types (e.g. vegetable trade through traditional open air markets versus modern supermarkets) of value chains?
4. What  is the variability of employment and risks to livelihoods within and between various levels of the value chain?
5. What is the impact of various governance systems on employment distribution between and within various levels of the value chain?
6. What  is the impact of various value chain technologies on employment distribution between and within various levels of the value chain?

 

Useful Examples

Example 1: {Title}

 

{words}

 

Example 2: {Title} 

 

 

{words}

 

Links to Other Examples

 

Author	Title	Description of Tool	URL Link