Basic features of rural small-scale forest-based processing enterprises in developing countries.

by
Dr. Yacob Fisseha
Centre for International Programs Michigan State University, USA

Introduction

The objective of this paper is to review briefly existing information on the contributions, characteristics, changes and constraints of the forest-based small-scale processing or manufacturing enterprises (FB-SSIs)^1/ in developing countries. The review is based mainly on a number of comprehensive, uniform surveys carried out by Michigan State University in collaboration with host institutions under a programme which had primary financial support from the Agency for International Development of the United States.

^1/The acronyms SSE and SSI stand for respectively the small-scale enterprises sector as a whole (i.e. including service activities) and small-scale industrial (i.e. manufacturing and repair) enterprises subsector. Corresponding acronyms for the forestry based subgroup are FB-SSE and FB-SSI respectively.

The countries chosen for the review here are Jamaica, Honduras, Zambia, Egypt, Sierra Leone and Bangladesh. At FAO's request, the author extracted information on forest-based processing enterprises from the general data bases on all off-farm employment. His detailed review of the findings was submitted to FAO in late 1984 which FAO subsequently published in early 1985 (FAO, 1985). Many of the points briefly mentioned in the present review are discussed there in detail.

For Zambia, a one-shot survey specifically for the forest-based small-scale enterprises was carried out for FAO in 1985. The present review is compiled from a report on that survey (Fisseha and Milimo, 1986).

Socio-Economic Statistics of the Surveyed Countries

Under the 'Economic Indicators' section of Table 2, seven key variables are shown which the Handbook of International Trade and Development Statistics of the United Nations calls “indicators of economic development”. The objective of choosing these indicators is to help examine how the FB-SSIs in particular and all small-scale industrial enterprises (SSIs) in general may perform at different stages or settings of economic development and how their roles and characteristics may vary accordingly. To facilitate such examination, the countries in Table 2 are listed in order of decreasing per capita GDP which ranges from US$ 1,212 for Jamaica to a low $143 for Bangladesh. The remaining countries are well distributed in-between. The remaining indicators generally exhibit a consistent pattern with the corresponding level of per capita GDP in each country.

Given the distribution of the economic development indicators in Table 2, the question one could ask is whether the characteristics and attributes of the FB-SSIs in the low income countries would be consistently and predictably different from those in the high income countries. Much of the information to follow does in fact suggest such consistency. What may not be so apparent is however, the implication of such consistency or trend for FB-SSI sectoral growth, extent of enterprise diversity and the level and nature of economic linkages with other sectors. The series of studies discussed here were not specifically designed to answer all such questions. In fact a more focussed time series study across countries would be more appropriate to analyze some of these issues.

Table 2: Socio-Economic Statistics of countries in which surveys were carried out

Economic indicators	Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Population (Millions)	2.2	3.7	5.6	42.3	3.5	88.5
Population Density (Per Km2)	200	33	8	42	49	615
Economic indicators for 1980:
Per Capita GDP (US$)	1212	674	671	464	374	143
Literacy rate (%)	96	57	50	38	7	26
Population in rural areas (%)	50	70	57	49	83	92
GDP from agriculture (%)	8	26	16	21	30	49
GDP from manufacturing (%)	15	14	18	12	7	8
Manufacturing employment (% of total employm't)	17	27	14	28	11	38
Commercial energy per cap. (Kg. coal equiv)	1260	190	490	473	92	46

Sources: For Zambia, see Fisseha and Milimo, 1986: for the others, see FAO, 1985.

General Description of the Surveys

The different country studies of the FB-SSIs were carried out between 1974 and 1985 (see Table 3). Four of them were however done between 1978 and 1982. Experiences gained in earlier studies were very useful to improve subsequent ones. Whenever flow data of inputs and outputs were collected, the survey was divided into two stages: Phase I dealing with simple enumeration or census of enterprises and Phase II designed to collect flow data from individual entrepreneurs through once or twice a week repeated visits over a period of one year. The relatively much fewer cases for Phase II represented a random sample from the enumeration results of the Phase I survey.

The studies carried out in Sierra Leone and Jamaica covered the whole country while for the remaining countries, the coverage was partial. The proportion of the national population actually surveyed during the enumeration stage ranges from almost 41 percent in Jamaica to 3 percent in Bangladesh. For those countries partially covered, the proportion of their population represented by the survey was much higher; in Zambia, for example, it was 60 percent.

In many cases, areas represented by the survey were stratified according to population sizes of localities from which a random, and sometimes purposive, sample of localities was drawn for complete enumeration. The sampling proportion usually ranged from 4 percent for the rural localities to 100 percent for the major urban areas.

A working definition used to delineate a small-scale enterprise used the following criteria: (a) any non-farm activity or business pursued on a commercial basis as a source of primary or secondary income; (b) total number of people 'employed' or engaged equal to or less than 50 (in Jamaica, 25); (c) privately owned (and organizationally unaffiliated with a national or international corporation); and (d) business operation from a fixed location.

As already noted, the surveys were carried out both in urban and rural areas. According to the U.N. definition of 20,000 or more people for urban however, localities of such size were not found in the regions enumerated in Honduras.

The discussions in the following sections deal almost exclusively with the manufacturing enterprises (SSIs and FB-SSIs) and very little on the service or trade group. A detailed study of the service group would be very interesting too. In some ways, the service or trade (commerce) group show much more linkages (through trade) with large scale enterprises than the manufacturing group; they also often account for significant employment.

Table 3: General Characteristics of the Small-scale Enterprises (SSEs) Surveys

Variable	Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Year of Study	1978-80	1979-80	1985	81-82	74-75	78-80
Country coverage	C	P	P	P1	C1	P
% of population enumerated	40.7	7.2	8.0	5.7	18.5	3.0
Urban(U) or rural (R)	R & U2	R	R & U	R & U	R & U	R & U
Upper SSI size limit (No. of workers)	25	50	50	50	49	50
No. of SSEs3 counted	9,500	2,120	16,776	15,881	9,195	57,184
Percent of SSEs in manufacturing (%)	37	100	80	44	56	100

Sources: See Table 2.

Notes:

1. C and P stand respectively for complete and partial.

2. R and U stand respectively for Rural and Urban localities. Although the enumeration in Honduras is designated 'rural', it simply means that urban localities satisfying the U.N. definition were not found in the surveyed regions - all other towns were included otherwise.

3. In the case of Egypt, 19938 dairy processing enterprises were excluded as being too intimately part of agriculture to be grouped under manufacturing.

Except in Honduras and Bangladesh, the other surveys included both manufacturing and commerce (trade or service) enterprises. The proportion of manufacturing enterprises (SSIs) relative to all enterprises (SSEs) ranges from 37 percent in Jamaica to 80 percent in Zambia. The atypically high proportion for Zambia is due primarily to a large number of local beer brewers in the country side.

The rest of this paper briefly reviews the highlights from the findings of the different country studies on the forest-based manufacturing activities which are identified as FB-SSIs for short. For detailed coverage of the overall findings, the following major works should be consulted: Liedholm and Chuta (1976) for Sierra Leone; Bangladesh Institute of Development Studies (BIDS) (1981) for Bangladesh; Fisseha (1982) for Jamaica; Stallmann and Pease (1983) for Honduras; Davies et al (1984) for Egypt; and Milimo and Fisseha (1986) and Fisseha and Milimo (1986) for Zambia and FAO, (1985) for a summary of the above country studies.

Composition and Basic Characteristics of the Forest-Based Small-Scale Industries

In this Section some basic characteristics and attributes of the forest-based small scale manufacturing enterprises (FB-SSIs) will be highlighted. The topics discussed include the enterprise type composition of the FB-SSIs, their relative magnitudes in terms of enterprise numbers and employment and some basic characteristics that typically distinguish such small-scale enterprises.

FB-SSIs: Their Composition

The diversity of activities or enterprise types in the FB-SSI group differs from country to country. For the most part the difference is the result of differences in forest-based raw material endowments or availability. There are a few dominant ones that tend to be found in almost all the countries surveyed. The two dominant (composite) activities are carpentry/furniture production and basket/mat/hat making. Except in Bangladesh, one of these two activities accounts almost for two-thirds of the total FB-SSIs in each country. Each of these two activities of course typically produces basic consumer goods: traditional baskets primarily for the low income group and modern wooden furniture for the middle income. Raw material for baskets is usually readily available while that is not the case for furniture and indeed part of it might have to be imported.

A furniture making workshop in Bangladesh

Table 4: Composition of the small-scale forest-based manufacturing enterprise group

FB-SSI types	% of total number of enterprises
	Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Saw-milling/pitsawing	0.8	3.2	5.6	-	0.1	0.9
Carpentry/ Furniture	23.1	71.4	14.3	23.8	66.8	27.2
Wood Carving/ bamboo/cane	12.5	0.2	11.9	-	5.9	11.6
Basket/mat/ hat making	63.5	10.6	60.3	70.4	23.8	32.4
Others	0.1	14.6	7.9	5.8	3.4	27.9

Sources: See Table 2.

Notes:

1. The 'Other' category includes activities such as broom making in Honduras, fuelwood in Zambia, agricultural tools in Egypt, and container making and agricultural tools in Bangladesh.

2. Many FB-SSIs do not specialize in the production of one item; so classification sometimes depends on the most dominant or important activity.

Without aggregation into the seven entries shown in Table 4, the varieties of activities could easily reach at least 20. Except as broad indicators of general orders of magnitude, the figures on the table are not strictly comparable across countries since Zambia for example includes enterprises producing fuelwood while that is not usually the case with the other countries.

Tapping trees for resin - an essential raw material

The Relative Magnitude of The FB-SSIs

Table 5 presents the relative magnitudes of enterprise groups within the bigger context of the small-scale industrial enterprises (SSIs). Except in Jamaica, the FB-SSIs are usually the second largest single category after garments in the number of enterprises. In Zambia, the most dominant group is the traditional beer brewers followed again by the forest-based activities.

The overall distribution of the SSE activities across the countries does not seem to follow any pattern by per capita income levels. This may however be due to the fact that the enterprise groups are simple aggregates of enterprises at different levels of technological sophistication; no distinction is made for example as to whether they are traditional (household) or modern (factory type) activities. It should be pointed out also that the SSI category of enterprises account for about 37, 80, 44 and 46 percent of the total SSE enterprises (which include service or trade activities) respectively in Jamaica, Zambia, Egypt and Sierra Leone. Service or trade activities were not enumerated both in Honduras and Bangladesh.

Within the FB-SSI group, woodwork tends to account for one-half to two-thirds of the total number. Again there are departures from the pattern in Table 5 as shown for Honduras and Sierra Leone where woodwork clearly dominates the other forest-based products. Incidentally, the figures for the two countries point out the dominant role the non-wood activities play in raising the share of the FB-SSIs relative to all SSIs.

Table 5: Industrial composition and relative magnitude of the small scale manufacturing enterprises (SSIs) in areas surveyed (by enterprise numbers)

Enterprise Groups		Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
All SSIs:
1. Numbers		13343	3703	169870	34465	50135	57184
2. % of SSEs		37	-	80	44	46	-
Relative magnitudes, within SSIs total (%):
1. Garments		36. 1	39.9	5.0	46.3	60.1	51.8
2. FB-SSI:
	(a) Wood works	13.6	10.8	12.4	7.7	9.8	8.1
	(b) Others	23.8	1.8	18.5	17.1	4.9	12.5
Subtotal FB-SSI		37.4	12.6	30.9	24.8	14.7	20.6
3. Foods		2.2	25.0	4.0	8.8	3.3	20.3
4. Metals		3.2	2.8	4.6	3.8	15.4	2.1
5. Leather & shoes		8.2	2.8	0.8	2.3	2.3	0.6
6. Non-metallic minerals		1.0	11.4	2.6	5.2	2.9	1.4
7. Repairs		8.8	3.4	1.4	7.8	0.7	2.0
8. Other manufacturing		3.1	2.1	51.4	0.9	2.1	1.2
Total (%)		100	100	100	100	100	100

Sources: See Table 2.

Notes:

1. 'Others' under the FB-SSIs refers to non-wood forest-based activities.
2. 'Repairs' includes bike, auto, appliance, etc.
3. The high figure for 'Other' in Zambia is due to the large number of traditional beer brewers.

The Relative Employment Size of the FB-SSIs

The ultimate interest on the small-scale industrial enterprises lies not in their raw numbers but in the economic contributions they make to national welfare. One such contribution is employment. Table 6 displays the relative sizes of employment corresponding to the relative enterprise magnitudes shown in Table 5. A prominent characteristic displayed in Table 6 is the small size of the enterprises. This number is the result of dividing the total number of employment by the corresponding total number of enterprises shown in Table 5. Generally the relative contribution to total SSI employment closely follows the picture shown in Table 5. Also, food processing, woodworks and repairs tend to have larger number of workers compared with the rest; on the other hand, garment and the non-wood FB-SSIs show the opposite characteristics.

The largest source of labour input for an enterprise is the family, that is both the proprietor or owner/operator and family members. It accounts for percent of the total labor force.

Constructing wheels from wood - as their fathers did before them

Some Basic Characteristics of the FB-SSIs

There are some basic features that uniformly characterize small-scale forest-based enterprises (FB-SSIs) that are typically found in many developing countries. Some of these features include the simple nature of their operations, their rural locational bias in aggregate, their relative accessibility to the poor, the landless and women and finally the dependence of the enterprise on the role of the family. Some of these characteristics are directly due to an enterprise's small size with relatively small investment and simple operational organization.

Table 6: Industrial composition and relative magnitude of the small scale manufacturing enterprises in areas surveyed (by employment)

Enterprise Groups		Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Mean No. of workers per enterprise		2.2	2.2	1.7	1.9	1.8	3.8
Relative magnitude of SSI types, by employment (%):
1. Garments		24.8	24.6	4.3	37.6	51.4	58.1
2. FB-SSI:
	(a) Woodworks	15.0	14.1	13.9	9.6	14.8	5.1
	(b) Others	19.8	2.2	19.3	14.3	5.5	7.9
Subtotal FB-SSIs		34.8	16.3	33.2	23.9	20.3	13.0
3. Foods		6.8	37.5	5.3	11.9	4.6	22.4
4. Metals		5.7	2.9	3.8	4.9	18.3	1.4
5. Leather & shoes		5.3	2.8	0.5	1.9	1.4	0.3
6. Non-metallic minerals		1.7	10.4	2.4	8.6	2.7	1.7
7. Repairs		16.2	5.1	1.2	9.0	1.9	1.3
8. Other		4.7	0.4	49.2	2.2	1.5	1.8
Total (%)		100	100	100	100	100	100

Sources: See Table 2.

Notes:

1. 'Others' under FB-SSI refers to non-wood forest-based activities.
2. 'Repairs' includes bike, auto, appliance, etc.
3. The high figure for 'Other' in Zambia is due to the large number of traditional beer brewers.

Table 7 shows that a majority of the enterprises operate from the home. In Zambia and Egypt more than three-fourths of the enterprises use the home or homestead as a workshop-cum sales area. In fact, many of them (for example 88 percent in Zambia) operate from the open space near the home. In Zambia only 0.6 percent of all the SSEs owned any powered machines (Milimo and Fisseha, 1986) and in Jamaica the total number of machines per enterprise was only 0.7 (Fisseha, 1982).

A large number of the enterprises are found in demographic and economic settings that are clearly rural or semi-urban in character. In all the countries, over 80 percent of the FB-SSIs are found in such localities. Accordingly, a large share of the FB-SSIs employment is found in the rural areas in close proximity to agricultural activities. In many cases a FB-SSI activity is pursued as a complement to agricultural employment or to provide supplementary income.

The table also shows that the proportion of women in ownership of FB-SSI enterprises ranges from about 3 percent for Bangladesh to 65 in Egypt. Similarly, the proportion of females in the labor force ranges from 6 percent in Honduras to 31 percent in Egypt. In the case of the FB-SSIs, the employment of females is to a large extent identified with the non-wood activities. In fact in some cases (e.g., in Jamaica) where employment is not traditionally open to them, particularly such as in wood-works, women tend to own (and manage) similar businesses.

Except in Bangladesh, the proportion of enterprises operated by the owner only ranges from 60 to 65 percent. The FB-SSI activity is indeed a family affair. Except in Sierra Leone, the family contributes over 50 percent of the labour force and about the same extent of the labour input hours.

Almost all the above characteristics are manifestations of the small size of the enterprise. Table 7 also gives some information on the size of the enterprises as a group. The average size of the enterprises in each of the countries falls roughly between 2 and 4 people. In fact, the number of enterprises with a labor force of above 10 account for less than 1 percent in all the countries except Honduras.

Except in Sierra Leone, the average number of annual hours per worker could for all practical purposes be taken as 1000. On the basis of a 40-hour week, this amounts to only 50 percent of the 'available' work time. This however is a reflection of the rural location of the vast majority of the enterprises where available time may be shared between farm and non-farm activities. In Zambia about 60 percent of the proprietors continued to have income from agriculture in addition to their FB-SSI income; in fact, for almost two-thirds of these, farming was the primary job before they started the FB-SSI enterprise.

With respect to the mean values of total investment and value of production or sales, there is wide differences among the countries. The figure for Bangladesh is conspicuously out of line relative to both the number of annual hours and the amount of total investment. Given the level of per capita income as shown in Table 2, it is not clear why the value of production is also that high.

Table 7: Summary of basic characteristics of the FB-SSIs

Attribute		Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Proportion of total FB-SSIs (%):
One-person operations		58	59	69	69	-	36
Production at home, not workshop		52	72	81	76	-	-
Rural location:
	- Enterprises	88	100	96	80	99	97
	- Employment	79	100	95	65	96	-
Women's share:
	- Ownership	32	10	12	65	-	(3)
	- Labour force	30	6	12	31	-	21
% family members in
	- labour force (No.)	82	51	86	89	(41)	73
	- hours worked	68	57	-	89	34	-
Mean Values:
No. of workers per enterprise		2.2	2.2	1.7	1.9	1.8	3.8
Total investment(US$)		3030	1055	-	-	431	255
Hrs. worked annually per worker		990	1247	1205	1712	2004	836
Annual production value per firm(US$)		4979	2536	-	1501	1384	2362

Sources: See Table 2.

Notes: Tile number of hours per worker for Zambia is estimated from the one-visit survey.

There are also wide differences within countries. Differences due to size and location are shown for example for two countries in Table 8. The variations in output due to these two characteristics is particularly consistent and almost predictable. Thus, the turnover of Jamaican enterprises with labor force greater than 5 is about two and half time greater than those with labor force less or equal to five; a corresponding similar relationship holds for the urban-rural attribute. Further breakdown would show even bigger variations. The mean output for the one-person operations in the small population stratum, for example, is $212 in Honduras compared to $16,053 for enterprises employing 6 or more people in the largest stratum. The corresponding means for Jamaica are $1,037 and $25,172. Clearly Table 8 shows the need for greater appreciation of the potential effects and feasibility of various policy interventions on enterprises characterized by different sizes, locations and other distinguishing features.

The aim of a project may, for example, be to expand output among producers through increased availability of raw materials (assuming potential market is available). Availability of just raw materials may not be enough however to attain the objective; improved production skill, added capital investment or aggressive marketing strategy may also be required. The capability of different enterprises to meet these requirements may vary by factors such as enterprise size or location and by proprietor training (experience) background. All these are of course empirical issues.

Economic Contributions and Sectoral Viability of FB-SSI Enterprises

The economic contribution of the FB-SSIs is evaluated from four criteria in this section: Employment, worker training, contribution to gross domestic product (GDP) and returns to family labor. The information on sectoral viability is very limited in all the countries. Using the available data, the variables or proxies examined for sectoral viability include (a) relative rates of return to family labour hour and rates of return to invested capital on the financial side and (b) changes in over time in the number of enterprises and the size of employment (both of which may give an indication of the overall business environment.)

Delivering the household furniture

Table 8: Differences in average capital stock and annual output among Honduran and Jamaican FB-SSIs due to size and location characteristics

Enterprise types or category		Honduras			Jamaica
		Capital Stock	Output	Value Added	Capital Stock	Output	Value Added
Labour force size:
	- 5 or less	884	2058	1515	2675	10844	6239
	- 6 or more	6232	15692	7714	11033	25171	22517
All sizes		1055	2536	1776	3513	12352	7952
Location:
	- Rural	393	1156	867	3152	8190	4564
	- Urban	2278	5164	3234	3719	22567	16564

Sources: Survey data.

Notes:

1. 'Labour force size' stands for the total number of people working in the enterprise (i.e., including the proprietor, family members, apprentices and hired workers).

2. All values are given in US$ at the exchange rates of US$1 equal to 1.78 Jamaican dollar (1979) or 2.00 Honduran Lempira (1980).

3. Capital figures for fixed assets and tools are based on current replacement values rather than on initial purchase prices or present 'book values'.

4. Data for Jamaica are not adjusted by strata weights as is the case in other reports.

Economic Contributions from the FB-SSIs

Information on economic contribution is shown in Table 9. In Jamaica and Zambia, employment in FB-SSIs ranges respectively from about 14 to 29 percent of the total manufacturing employment in the economy (i.e., both small and large scale industrial enterprises combined). It is difficult to tell how many of these people would have been unemployed in the absence of the FB-SSI activities. In Zambia about 16 percent of the owner/operators said that they were ‘unemployed' before they started the present FB-SSI activities. Given that proprietors could be expected to be more aggressive or enterprising in their economic pursuit compared with their workers, it is possible that the percentage of unemployment among the latter is much higher.

The second type of contribution shown in Table 9 is the on-the-job technical and business training of workers. This kind of training is cheap and effectively practical, benefitting even the large scale sector as some are eventually employed by it. To provide technical training under formal setting for such a large number of workers could have cost, for example, perhaps thousands of dollars annually. As for business training, many of them while serving as trainees or skilled workers are directly exposed to the steps and strategies of operating an enterprise so that over time some of these trained people go on to open their own shops - a few growing to be large scale producers.

In developing countries the contribution of the entire manufacturing sector to GDP usually lies between 10 and 20 percent (see Table 2). The contribution of the small-scale industrial enterprises (SSIs) to GDP typically lies on the other hand between 2 and 4 percent depending on the level of economic development and the enterprise definitions used. This could imply that anywhere from 10 to 40 percent of the manufacturing contribution could come from the small-scale (SSI) subsector. The forest-based SSIs in turn contribute about 20% to 25% of the SSI share.

The final contribution from the FB-SSIs is the income or return to family labour on FB-SSI activities. In terms of raw numbers, this ranges from US$308 in Bangladesh to $2 494 in Jamaica. When one remembers that these averages are generated from smaller economic units only, the actual income levels for many families may indeed be much higher than the per capita income shown in Table 3. Of course, an average picture obscures finer details and there must be many individual enterprises in serious financial trouble or even going under. Nevertheless, on aggregate the income levels are reasonable.

When FB-SSI earnings are compared with the prevailing wage rate in an FB-SSI or in agriculture (the most likely alternative employer) the prevailing wage rate is only about two-thirds of the hourly return to FB-SSI family labour in most of the countries (Egypt excepted). Looking at it from a different angle, the return to an hour of family labour on FB-SSI activity is about 50 percent higher compared with the prevailing agricultural hourly wage rate. Working in one's own FB-SSI activity was therefore much more remunerative than paid employment either in the FB-SSI subsector or in agriculture.

How does hourly return for family labour in FB-SSI compare with the corresponding return for agricultural activities? Only Thailand and Sierra Leone had a simultaneous study of both farm and non-farm activities which allows one to look at families that were involved actually in both activities. Even then, the return specifically for FB-SSI activities cannot be separated from other SSI activities. In Thailand, SSI activities used up about a third of the total labor hours and generated about 20 percent higher return per family hours than farm activities. Comparing between farm and non-farm activities in Nigeria, Matlon et al (1979) say “... returns tended to be higher in non-farm than in farm enterprises.” They attribute this higher return per SSE hour to higher capital cost associated. Thus, not only are SSE enterprises more remunerative relative to farm enterprises but rural families generate more return per hour of SSI activities than in farm work. As there are other important economic and non-economic factors to consider, it does not necessarily follow that rural families should concentrate their attention on non-farm labour at the expense of farm activities.

Table 9: Major economic contributions of forest-based small scale enterprises

Contributions		Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Total employment - ('000)		10.2	-	137.4	-	18.0	-
Workers trained (No.)		1663	-	4647	-	-	-
Share of FB-SSI in total SSI:
	- value added (%)	47	16	-	18	23	18
	- value of production (%)	49	14		19	27	-
Net return to family labour (US$)		2494	1221	-	495	371	308
Prevailing wage as a % of return to FB-SSI family labour hour		61	64	-	85	63	62
Rate of return on investment in FB-SSIs, (%)		33.0	41.7	-	27.5	32.2	54.2

Sources: See Table 2.

Notes:

1. Financial figures (all in US$) and number of trainees are given on annual basis.

2. 'Return to FB-SSI family labour hour' is before opportunity cost of family labor is deducted.

Sectoral Viability or Growth of the FB-SSI Activities

The interest here is of course on sectoral viability in each country and not the relative profitability (and hence viability) of individual enterprises characterized by specific size, market, technology and management parameters. One can look at absolute growth in the FB-SSI subsector in at least two fundamental ways: average growth in individual enterprises as can be identified by measuring changes over time in sales, employment or investment; or growth in the number and composition of enterprises and their aggregate employment whether their income or investment is expanding or not. Study of time-series data would be necessary to measure over time particularly for the first approach. As such data are rarely available however, many of the country surveys have focussed on description of the static picture at certain points in time over the life of the enterprise.

Does the fact that the FB-SSI return per family hour of work is higher than the prevailing wage indicate that there should be a ready supply of entrepreneurs to open their own 'little' shops? Or does the fact that the rate of return on FB-SSI investment ranging from 28 percent in Egypt to 54 percent in Bangladesh (see in Table 9) mean that there should be a rush of external funds into the subsector? Both of these expectations are unrealistic both from experience and from analysis of the characteristics of small-scale enterprises. Investment capital, initial or operating, is usually a major problem for many small-scale producers and could effectively bar many potential proprietors from entering the subsector. Speaking of this same problem both in Sierra Leone and Nigeria, Matlon et al (1979) say, “Entry costs associated with even the most labour intensive technique and at the smallest production scale were still substantial when compared with the incomes and investment capacity of those households.” They add, “Finally, the Sierra Leone data also found that capital requirements were the primary factor limiting the participation of low income households in high return off-farm occupations.”

^2/For a fuller discussion of issues related to finance see the paper by Brunton (In this publication).

As to the availability of external investment fund, there are two things to point out. First, institutional loans are practically unavailable for small-scale producers due to a number of factors; second, the high rates of return may in fact be specific to the particularly small capital base on which rate of return is calculated or the way a small-scale enterprise is organized and operated, i.e., the organically inseparable relationship among funds belonging to the proprietor, the household and or the family. And because the absolute size of the income generated from such enterprises is relatively small, not many people with funds would be willing to invest and personally manage at the level of scale typical of such SSEs. Thus, the hypothesized potential attractiveness for investment may not be as great as first impressions would suggest.

On the same line of thinking, the question of whether such high rates of return encourage existing proprietors to invest more and expand their activities will be examined relative to the general availability of funds discussed in the next section on problems and constraints facing FB-SSIs.

Finally, several of the present studies suggest that the number of enterprises in the FB-SSI group has been growing or at least not declining over the recent past. Most of the reported growth is based on proprietors' own subjective assessment of the business environment. Entrepreneur perceptions of trends in number of enterprises in the surrounding locality, changes in one's own output as well as the general perception of total market demand are taken into consideration in such assessment. According to this subjective approach, the number of enterprises consistently exhibits growth or expansion trend. The survey technique is, however, unable to take account of enterprise mortality as only entrepreneurs of surviving units are interviewed.

There have been attempts to use structured, objective approaches to measure changes in number of enterprises and their employment. In Sierra Leone (Chuta et al, 1981) the Phase I enumeration survey mentioned in Section 1 was repeated 6 years later in those localities with population over 2,000 to see such changes. The repeat survey showed that employment among the total SSEs has grown at an annual rate of 3.2 percent. In Jamaica, the number of enterprises was growing at an annual rate of 2 percent in 1979/80 although the country was going through severe economic recession. In Bangladesh it is estimated that the SSI enterprises are growing at an annual rate of 4.7 percent. For Jamaica and Zambia, estimated annual equivalent growth races for employment and Investment over the life of the enterprise are given below:

Annual Growth Rates		Jamaica	Zambia
Labour Force:
	Urban	9	7
	Rural	2	3
Investment:
	Urban	4	30
	Rural	20	12

It seems certain though that in many developing countries the number of SSE enterprises grows both in time of national economic expansion and in time of contraction - the latter accompanied with a shrinking general enterprise income. During economic expansion, the growing purchasing power of consumers makes it attractive for new entrants; and during economic contraction, the SSE sector acts as a last resort for those losing their jobs particularly in the large scale sector - this was seen to be the case for example, in Jamaica (Fisseha, 1982). By serving also as the refuge of the unemployed or displaced, it probably helps to diffuse social disturbance or upheaval that could otherwise accompany any economic recessions. It should be pointed out though, that the increase in the number of enterprises during times of economic contraction applies primarily to the very small ones that do not require a sizeable capital investment.

Finally, the situation with the FB-SSIs is very peculiar in that if their local raw material supply suddenly diminished, they would face particularly severe problems as very few can afford imported substitutes. Such a phenomenon would have a drastic effect on their numbers and employment. The next section will discuss to what extent raw materials shortages, among other things, have been a problem.

Major Problems and Constraints facing Forest-Based Enterprises

In this section, problems related to raw materials, finance and product demand/market for FB-SSIs will be briefly reviewed. These three problems are chosen for emphasis because they are the ones most frequently reported by FB-SSI producers. Table 10 shows the ranking of these three problems subjectively considered by FB-SSI producers as being the most serious problems faced.

Raw Material Problems

Raw materials are increasingly the most worrying problems faced by FB-SSI producers simply because in many areas their availability is progressively diminishing. In many developing countries the availability of raw materials, particularly of wood, is the limiting factor for growth or even sustenance of the handicraft industry.

When raw material problems are primarily due to the declining availability of domestic natural resources (e.g., reeds, cane, bamboo, rattan, wood, etc.) of a country, there is not much that could be done in the short-run to increase the supply, particularly for bamboo, rattan and wood, but a more efficient use of what is available could be encouraged. However, besides questions of natural endowment of these resources, there are others, perhaps equally significant, that could cause raw material problems.

They include issues such as the distribution or pricing systems and the efficiency of regulatory bodies mandated to oversee the utilization (exploitation) and conservation of such resources. In Thailand and Philippines, the respective teak and rattan problems were related to the restrictive tendency of regulatory bodies although the destructive pattern of harvesting these raw materials was also contributing to the problem. On the other hand, the major raw material problem in Jamaica was lack of foreign exchange to buy lumber or other imported raw materials such as finishing items. Also with the lumber in Jamaica and rattan in Philippines, inadequate skill or knowledge of processing such raw materials was a contributing factor to the wastage and thus shortage of the raw material.

The second major source of raw material problems is the marketing system. Due to poor transportation or monopoly distribution, resources are unavailable or inaccessible. They may also have high or unstable prices. To some extent such were the problems in Bangladesh and Egypt.

Table 10: Ranking of entrepreneur perceptions of main problems facing FB-SSIs

Problem area	Ranking of problem
	Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Finance	1	(1)	3	--	1	(1)
Raw material	2	(3)	2	1	2	(2)
Demand/market	3	--	4	2	3	(7)
Others	--	(2)	1	3	--	(3)

Sources: See Table 2.

Notes:

1. The number '1' shows that the problem was ranked “most important”.

2. Numbers in parentheses show that the ranking is for the whole SSI group not for the FB-SSI only.

3. Only when they are among the top three ranks are the 'other' problems indicated.

Although prices of products were also rising just as fast or even more, the annual price increase rates of key FB-SSI raw materials were 27 percent in Jamaica (compared to 31 for key products), 18 in Zambia (15 for products) and 20 in Bangladesh (same as for products). Since raw material account for only 10-25 percent of the value of the finished product, the above price increases were not necessarily unbearable but it would cause serious working capital shortage. Where prices fluctuate significantly, they make forward planning virtually impossible.

Small-scale producers would probably be able to cope better (say, via substitution of cheaper, scrap or recycled materials) against shortage than their larger counterparts. However, the small ones are always at a disadvantage when it comes to volume price discount, duty free import and inability to get loans. In fact lack of working capital would prevent them from avoiding the effects of rapid raw material price escalation by purchasing stocks when raw material prices are moderate (e.g. during seasons of plenty).

The proportion varies from country to country and by enterprise type but raw materials generally account for 15 to 25 percent of the gross value of output. In Table 11 the raw material-output ratio is given for different enterprise sizes and locations for Jamaica and Honduras. The Cable seems to indicate that larger and more urban enterprises tend to have higher raw material-output ratio. It is not clear from the table whether larger enterprises, which are mostly found in the urban areas, have higher ratio because they use more expensive (e.g., high value or imported) raw materials or whether they are inefficient in their use of it. On the other hand some small, rural forest-based enterprises produce their own raw materials - materials whose value may not be costed other than through labor hours that went into their production. As a result they would tend to show lower material-output ratio compared with their larger urban counterparts. The actual reason may be a combination of these two possibilities.

When one looks just at the one-person operations, the more urban areas tend to have a higher ratio than the more rural ones (e.g., 0.20 vs 0.14 for Honduras and 0.33 vs .13 for Jamaica). When the location effect is minimized by looking into areas which are not so rural nor so urban, the high ratio for the larger enterprises still persists for Honduras but is less consistent for Jamaica.

Table 11: Raw material-output ratios for different enterprise size and location - examples from Honduras and Jamaica

Enterprise size (i.e No. of workers per unit)		Ratio of value of raw materials to value of output
		Rural	Semi-rural	Semi-urban	Urban	ALL
Honduras:
	1	.14	.17	.20	-	.18
	2-5	.14	.16	.20	-	.19
	Over 5	-	.26	.42	-	.38
	All sizes	.14	.20	.27	-	.26
Jamaica:
	1	.13	.16	-	.33	.16
	2-5	.13	.12	.19	.26	.19
	Over 5	-	.13	.29	-	.24
	All sizes	.13	.14	.23	.27	.20

Sources: Survey data, Jamaica (1979/80) and Honduras (1980)

Notes:

1. 'Rural' refers to villages or hamlets with a population of 2,000 or less.

2. 'Semi-rural' refers to localities with a population of below 20,000 which are characteristically found in rural settings.

3. 'Semi-urban' refers to localities with a population size of 20,000 or more. According to this definition, there were in fact no semi-urban localities in Honduras at the time of survey but only rural and semi-rural.

4. 'Urban' here refers to the capital city, Kingston.

Problems Related to Finance

Looking at Table 10, finance seems to be the major concern of producers. Finance problems could be conveniently divided into two groups: investment (initial or expansion) and working capital. Investment capital issues refer to the availability of adequate funds to invest up to the optimum enterprise size or available technique of production. With working capital problems the concern is with adequacy of funds to meet recurrent expenses (including modest stocks of inputs and products) on a timely basis. Historically, the focus of discussion on SSI financial issues was limited to investment capital. There is growing evidence though that working capital needs among SSI proprietors are just as important, if not more, than the former. For example, about 90 percent of the finance problems in Jamaica were due to working capital shortages. In Honduras almost two-thirds of the proprietors who got credit used it for working capital needs (Kelley and Coronado, 1979).

The average initial investment ranges from US$50 in Zambia to $450 in Jamaica. Close to three-fourths of the source for this kind of finance is personal savings and gifts from relatives and friends (Table 12). Loans, most of which again come from friends and relatives, accounts for less than 10 percent.

Over 80 percent of the expansion capital also comes from personal savings and re-invested profits. An examination of a random sample of almost 1000 small-scale (farm and non-farm) enterprises showed that less than one percent had ever received a loan from a formal financial institution (Matlon, et al, 1979). Small may be beautiful but when it comes to credit, it appears to be unattractive!

Outside finance for working capital is much harder to come by although the need for it could be very vital due to rising prices of inputs, depressed demand, seasonality of production and credit sales of output without compensating credit purchase of inputs. Commercial banks are not successful in making loans to the very small enterprises. Their interest rates and required collateral are very high to guard against what they perceive to be costly and 'risky' loans to SSEs; if ever a loan application is successful, the time it takes to process it and sometimes the manner of disbursement of funds are extremely slow. For example, the size of the collateral in Bangladesh and Jamaica was respectively 300% and 400% of the outstanding loan from formal institutions. Informal loans on the other hand, though expensive in terms of interest rates, are less tediously available and accessible without much need for exorbitant collateral values.

The question of financial needs of small-scale producers and how to address them is unfortunately highly controversial and sometimes lost in political argument which may be divorced from reality. Sometimes there is the tendency to downplay the importance of financial constraints among small-scale producers and to disapprove concessionary loans or provide service to help them tap the commercial financial market. On the other hand, it would not be correct to ascribe every problem to financial constraints and therefore advocate for 'easy' loans to all and sundry SSI producers. Depending on the general economic environment, the type of enterprise and the time in the production cycle, there may be a situation of genuine financial need. And again depending on the problems preventing SSI producers from accessing the private financial market (formal or informal), there may be justified reason for public institutions to step in and facilitate things.

Table 12: Sources of initial and expansion capital for SSIs

	Haiti	Bangladesh	Sierra Leone
Sources of initial capital (%):
Personal Savings & Gifts	72	73	76
Loans	12	4	6
Other sources	16	23	18
All initial capital	100	100	100
Sources of expansion capital (%):
Personal savings and re-invested profits	81	89	88
Loans (all sources)	19	11	8
Other sources	0	0	4
Total	100	100	100

Source: FAO, 1985

Urban SSI enterprises are relatively more capital intensive than their rural counterparts; as a result they have a much higher labour productivity. In Honduras, for example, the capital labour (hours) ratio for rural enterprises is 0.05 compared to 0.16 for those in the urban areas; the corresponding labour productivity (output-labour ratio) is respectively 0.97 and 1.67. In Jamaica too, capital productivity for rural enterprises was 4 times higher than their urban counterparts.

Demand or Market Problems

Forest-based small-scale producers primarily cater to the needs of rural and relatively low income urban consumers. A question which is often asked is whether as consumer income rises the demand for FB-SSI falls, that is whether they are 'inferior' goods in economic jargon. The few studies carried out suggest that this is not the case: thus in Sierra Leone the income elasticity of demand during the survey was 0.88 showing that for every 10 percent increase in income, consumption of the FB-SSI goods would increase by 8.8 percent.

It would appear that development strategies that promote income re-distribution as well would result in more consumption of (hence more employment from) the small-scale enterprise products. There are obviously a number of relevant issues to look into in order to establish whether in fact this will be the result.

The question of market and marketing strategies by FB-SSI producers is an area like management that may not have been thoroughly studied. FB-SSI producers carry out very little market promotion activities in the formal or modern sense of advertisement. However there are a number of, albeit informal, ways that producers try to promote their products with both existing and potential clients. In Jamaica, increased product promotion was positively associated with increased profitability.

Charcoal packed ready for the market

Forward linkage of FB-SSI products with the large scale producers either through open market sale or sub-contracting is not that substantial. The little sub-contracting that exists seems to be within the SSI subsector itself between the upper size category of the small-scale producers and the smaller ones. There seem exceptions to this rule particularly in some Asian countries.

Finally, export is another important outlet for FB-SSI producers either through direct export or indirectly through the sale of products to tourists or to merchants that cater to tourists. In Jamaica and Bangladesh it was estimated that about 10 percent of the total FB-SSI production value is eventually exported. In Thailand 6.5 percent of the larger SSI furniture and fixture production value was estimated to be exported (see Boomgard, 1983).

A village co-operative collecting wood

Constraints Related to Managerial/Entrepreneurial Skills

The caliber of SSI managerial quality is central both to the viability of an enterprise and to the success of external intervention efforts. One of the distinct characteristics of SSI proprietors is that they are both investors and managers. There is no capacity for hiring capable management from outside. Such a situation may not be that bad after all. The high rate of return on SSI investment observed earlier seems to be due to the fact that the entrepreneur or owner is the planner and manager of the enterprise. It is this kind of relationship that makes it possible to conclude that small is indeed possible.

As the enterprise grows many suspect that management could be a limiting factor. The question of management weakness even in the absence of expansion is also something that should be looked into. There are in fact evidences that the latter problem could be a more serious problem. In Jamaica about 20 percent of the proprietors failed to identify one or more of key cost items in their enterprise. About 17 percent of them in fact failed to identify 3 or more cost items. Such problems could inevitably lead to diminished business income and subsequently even to closure. Problems of the latter type would probably improve with experience and perhaps some education on improved practices. Table 13 presents some attributes of FB-SSI producers that may influence their efficiency. Practices that could be even more crucial (such as information gathering, production techniques and marketing strategies) are not shown there. Such practices could on the other hand be the object of intervention whose success may depend on the kind of attributes shown in the Table 13.

The age of the proprietor combined with the age of the enterprise could indicate duration or extent of experience in a particular line of industry. Having been trained earlier as an apprentice could also help one to be exposed to the various facets of business activities.

The level of record keeping is very low in all the countries except Sierra Leone. The role of record keeping in improving business profitability is not however always supported by empirical evidence. There probably is a level in business activities when record keeping becomes absolutely necessary for efficient operation. And for many others, keeping of records also has its cost: unfortunately it may not be cost effective for all FB-SSI producers to keep or use written records.

The information in Table 13 could potentially provide some clues as to the level of entrepreneurial quality. It does not directly shed light on the potentiality for success of individual enterprises. Such information can successfully be inferred from sound business analyses and a close scrutiny of historical experience of each proprietor within the prevailing economic (market) and policy environment. Since availability of flow information on inputs and outputs over a production cycle is the exception rather than the norm for small-scale enterprises however, the feasibility of a comparative full financial or business analyses is almost impossible - unless, that is, time is taken to collect the necessary data to carry out such financial analyses.

Furthermore, if one is to pay adequate attention to other related activities that compete for the proprietor's funds and time, the reference period required for such data collection is bound to be rather lengthy; as a result, quantitative business analyses for intervention purposes end up of necessity limiting the exploration to the larger and urban enterprises that keep records, thus leaving out from the picture the smaller enterprises which comprise a lion's share of the SSI subsector.

Depending on the objective at hand, it may not be necessary to have full data nor to depend exclusively on quantitative evaluations. Equipped with the needed background knowledge of certain activities and industrial groups, useful data might be collected quickly to generate clues or rough comparative magnitudes of certain variables. For example, raw material procurement costs may indicate differences in management qualities.

Whatever quantitative estimates are available should be supplemented with semi-quantitative and subjective evaluations. In fact, business viability investigations dealing with non-quantifiable variables, both internal and external to the enterprise, will be just as important.

Table 13: Management characteristics of FB-SSI producers or proprietors

Proprietor or producer attributes	Jamaica	Honduras	Zambia	Egypt	Sierra Leone	Bangladesh
Proportion of enterprise work hours contributed by proprietor (%)	64	53	-	(84)	(34)	-
Average age	40	41	38	49	-	-
Started own business (%)	98	99	(78)	49	-	42
Literacy rate (%)	97	74	58	-	33	59
Years of education	6	-	-	1	-	-
% who hire apprentices	78	52	(63)	28	(90)	25
% who keep business records	7	9	9	-	21	4

Sources: FAO, 1985; and Fisseha, 1982.

1. Figures in parentheses refer to all SSI enterprises.

2. 'Started own Business' refers to the percentage of proprietors who started their business from scratch.

3. 'Record keeping' means those who keep some sort of records - not necessarily complete ones.

Bibliography (Fisseha)

Badr, Mahmoad, Nadia El Sheikh, James Seale, Stephen Davies, Abdel Azim Mostafa, and Abdel Rahman Saidi., 1982 “Small-Scale Enterprises In Egypt”: Fayoum and Kalyubiya Governorates Phase I Survey Results. MSU Rural Development Series, Working Paper No. 23. East Lansing: Michigan State University, Department of Agricultural Economics.

Bangladesh Institute of Development Studies (BIDS)., 1981 “Rural Industries Study Project:” Final Report. Dacca, Bangladesh.

Boomgard, James J., 1983 “The Economics of Small-Scale Furniture Production and Distribution in Thailand.” Unpublished Ph.D. dissertation, Department of Agricultural Economics, Michigan State University. East Lansing, Michigan.

Byerlee, Derek, Carl Eicher, Carl Liedholm and Dunstan Spencer., 1983 “Employment-Output conflicts, Factor-Price Distortions, and Choice of Technique: Empirical Results from Sierra Leone.” Economic Development and Cultural Change. 31(2): 315-336.

Chipeta, Mafa E., 1984 “Rural Non-Farm Employment and Income Generating Activities in Developing Countries: The Contribution of Wood-Processing Activities.” Draft. FAO; Rome.

Chuta, Enyinna and Carl Liedholm., 1979 “Rural Non-Farm Employment: A Review of the State of the Art.” Michigan State University Rural Development Papers, Paper No. 4. East Lansing: Michigan State University Department of Agricultural Economics.

Chuta, Enyinna and Carl Liedholm., 1975 “The Role of Small-Scale Industry in Employment Generation and Rural Development: Initial Research results from Sierra Leone.” African Rural Employment Paper No. 11. East Lansing; Michigan State University, Department Agricultural Economics.

Chuta, Enyinna and Carl Liedholm., 1982 “Employment Growth and Change in Sierra Leone Small-Scale Industry, 1974-80.” International Labour Review. 121(1): 101-113.

Davies, Omar., Yacob Fisseha, and Claremont Kirton., 1979 “Small-Scale, Non-Farm Enterprises in Jamaica: Initial Survey Results.” MSU Rural Development Series, Working Paper No. 8. East Lansing: Michigan State University, Department of Agricultural Economics.

Davies, Stephen., et al., 1984 “Small Enterprises in Egypt: A Study of Two Governorates.” MSU International Development Papers, Working Paper No. 16. East Lansing: Michigan State University, Department of Agricultural Economics.

FAO., 1985 “The Contribution of Small-Scale Forest-Based Processing Enterprises To Rural Non-Farm Employment and Income in Selected Developing Countries.” (Based on the work of Yacob Fisseha).

Fisseha, Yacob., 1982 “Management Characteristics, Practices, and Performance in the Small-Scale Manufacturing Enterprises: Jamaican Milieu.” Unpublished Ph.D. dissertation. East Lansing: Michigan State University, Department of Agricultural Economics.

Fisseha, Y and Omar Davies., 1981 “The Small-scale Manufacturing Enterprises in Jamaica: Socio-economic Characteristics and Constraints.” MSU Rural Development Series, Working Paper No. 16. East Lansing: Michigan State University, Department of Agricultural Economics.

Fisseha, Y and John Milimo., 1986 “Rural Small-Scale Forest-Based Processing 1986 Enterprises in Zambia: Report of a 1985 Pilot Survey.” FAO, Rome.

Haggblade, S., Jacques Defay and Bob Pitman., 1979 “Small Manufacturing and Repair Enterprises in Haiti: Survey Results.” Michigan State University Rural Development Series Working Paper No. 4. East Lansing: Michigan State University, Department of Agricultural Economics

Industrial Management Co., Ltd., - Industrial Finance Corporation of Thailand (IMCL-IFCT) ed. 1983 “Rural Off-Farm Employment in Thailand.” Bangkok, Thailand.

Kelly, John C. and Luis Coronado., 1979 “A Profile of Rural Small Industries in Selected Areas of Honduras: A Report for the Agency for International Development.” Tegucigalpa: Honduras.

Liedholm, Carl and Enyinna Chuta., 1976 “The Economics of Rural and Urban Small-Scale Industries in Sierra Leone.” African Rural Economy Paper No. 14. East Lansing: Michigan State University, Department of Agricultural Economics.

Matlon, Peter J., 1979 “Income Distribution Among Farmers in Northern Nigeria: Empirical Results and Policy Implications.” African Rural Economy Paper No. 18. Department of Agricultural Economics, Michigan State University. East Lansing, Michigan.

Matlon, Peter J. et al., 1979 “Poor Rural Households, Technical Change and Income Distribution in Developing Countries: Two Case Studies from West Africa.” African Rural Economy Working Paper No. 29. Department of Agricultural Economics, Michigan State University. East Lansing, Michigan.

Mead, Donald C., 1983 “Subcontracting in Central Java: Report on Consultancy.” Department of Agricultural Economics, Michigan State University. East Lansing, Michigan.

Milimo, John T. and Yacob Fisseha., 1986 “Rural Small-Scale Enterprises in Zambia: Results of a 1985 Country-Wide Survey.” MSU International Development Papers. Working Paper No. 28. East Lansing: Michigan State University, Department of Agricultural Economics.

Spencer, Dunstan S.C., Derek Byerlee and Steven Franzel., 1979 “Annual Costs, Returns, and Seasonal Labor Requirements for Selected Farm and Non-Farm Enterprises in Rural Sierra Leone.” African Rural Economy Working Paper No. 27. Department of Agricultural Economics, Michigan State University. East Lansing, Michigan.

Stallmann, Judith I. and James W. Pease., 1983 “Rural Manufacturing in Three Regions of Honduras.” Draft report. East Lansing: Michigan State University, Department of Agricultural Economics.