Small-Scale Manufacture of Footwear (ILO - WEP, 1982, 228 p.) CHAPTER V. FRAMEWORK FOR NATIONAL DECISIONS (introduction...) I. Comparison of alternative footwear products I.1 Capital and labour requirements for different technologies and types of footwear I.2 Product choice, resource allocation and the satisfaction of basic needs II. Economic and social appraisal of alternative technologies and products II.1 Social versus private benefit-cost analysis II.2 Appraisal of alternative technologies for type 1 footwear II.3 Data requirements and sensitivity analysis III. Towards a national footwear manufacturing strategy III.1 Factors influencing a national footwear strategy III.2 Comparative advantage, social considerations and product choice III.3 Backward linkages and availability of raw materials III.4 Employment, technological choice and skill requirements III.5 Other factors

Small-Scale Manufacture of Footwear (ILO - WEP, 1982, 228 p.)

CHAPTER V. FRAMEWORK FOR NATIONAL DECISIONS

In Chapter IV, the nature of technological choice and the selection of least-cost (combinations of) techniques in footwear manufacture were explored for various types of footwear and/or scales of production. Where necessary, the overall feasibility of alternative footwear projects in terms of financial requirements and private profitability was indicated.

In this chapter, the analysis of alternative production technologies, scales of production and product choice is placed in the broader framework of national decision making. The financial appraisal of alternatives in terms of domestic market prices, relevant to the private entrepreneur, is therefore extended to a so-called economic and social appraisal emphasising the use of accounting prices. The latter should better reflect the true scarcity of the factors of production (e.g. labour, capital) and development objectives than market prices, which, in many developing countries, may be distorted as a result of taxation, protection, monopolies and other market imperfections.

Prior to evaluating the economic and social impacts of technological alternatives, least-cost combinations of techniques used to produce the different types of footwear covered by this study are analysed in depth in section i. Special attention is paid to the impact of the choice of various types of footwear_; resources allocation and the satisfaction of basic needs.

Section II deals with the appraisal of footwear projects on the basis of economic and social accounting prices rather than domestic market prices, and illustrates the effect of using different prices on the choice of appropriate, least-cost combinations of techniques. It also includes a brief discussion of the data required to conduct such evaluations, and shows how the social feasibility of various footwear projects may be affected by changes in accounting prices.

The last section of the chapter briefly outlines various aspects of a national footwear manufacturing strategy on the basis of the findings of chapters IV and V. Various factors which may influence such a strategy are discussed, and their effect on product choice, scale of production and technology choice is indicated. As this chapter is mainly concerned with national as distinct from private sector decision making, it is primarily addressed to planners, and policy makers in national or local government departments, development organisations and financial institutions.

I. Comparison of alternative footwear products

I.1 Capital and labour requirements for different technologies and types of footwear

The major conclusions reached in Chapter IV with regard to the nature of technological choice in footwear manufacture can be summarised as follows: (1) Generally, the choice of technology affects the number and required skill of production employees (direct labour) and the volume and composition of plant and machinery (fixed capital). (2) Capital-labour substitution in manufacturing a particular type of footwear is characterised, to a large extent, by the substitution of skilled labour, rather than unskilled labour for capital. Thus, labour-intensive technologies are characterised by both a large number and a high proportion of skilled workers. (3) When comparing least-cost technologies for different types of footwear, variations in labour requirements across products are substantial, but variations in fixed capital per employee are comparable to those found for technologies used to produce a given type of footwear. The nature of capital-labour substitution across products is such that capital-intensive products tend to be associated with a high proportion of skilled labour. (4) At low scales of production, the least-cost technologies are relatively labour-and skill-intensive.

Table V.I provides estimates of the capital requirements per employee for the most labour-intensive, most capital-intensive and least-cost technology used in the production of different types of footwear. These estimates are summarised from the Ethiopian case study reported in Chapter IV. The capital-labour ratios for the least-cost technology are the same as those included in the last line of Table IV.3 in Chapter IV.

Table V.I Fixed capital per employee for manufacturing 1,200 pairs per day of different types of footwear with different technologies in Ethiopia (in 1972 Ethiopian dollars)

Type of footwear	Most labour-intensive	Least-cost	Most capital-intensive
Stitched leather, cemented-on shoes - type 1	1.856	3,952	5,100
Stitched leather, moulded-on shoes - type 2	2,650	4,953	6,016
Welded PVC, moulded-on shoes - type 4	8,158	10,324	10,828
Stitched PVC, cemented-on shoes - type 5	2,305	3,393	4,343
One-shot moulded PVC sandals - type 6	10,536	10,536	17,140

Source: McBain (1977)

Table V.1 shows that for leather upper shoes, opportunities for capital-labour substitution are greater between alternative technologies for a particular type of footwear than between different types of shoes. The reverse is true, however, in the case of footwear made of synthetic material. When all types of footwear are considered together, variations in capital-intensity are definitely greater across products than across technologies used for the manufacture of a particular type of footwear.

I.2 Product choice, resource allocation and the satisfaction of basic needs

When discussing the results of Table IV.3, it was concluded that, if effective demand for footwear would be such as to justify only one type of shoe or sandal to be marketed at a time, the higher prices varieties would be preferred. Compared with the synthetic and inexpensive type 4 shoes and the very cheap one-shot plastic sandals (type 6) the higher-priced type 2 shoes and type 5 sandals appear to be considerably more appropriate in terms of resource allocation and scale of production. Several reasons may be advanced to explain why product choice based on the latter two factors leads to this result.

First, whereas the higher-priced footwear types require more capital, returns are sufficiently high to justify their production when measured against competing investment possibilities. Second, the higher-priced types generate, proportionally, much more employment than the less expensive varieties. Although the latter require an absolute number of skilled labour that is far below that needed for the higher-priced types, it is likely that the type of skills required for footwear types 4 and 6 is far more scarce in developing countries than the type of skills necessary for the other, more craft-based products. Consequently, the factor requirements corresponding to the labour-intensive, high-priced products seem a priori much more in line with relative scarcity of factors of production in developing countries than do factor requirements associated with less expensive footwear. Third, the damage caused to existing small-scale footwear production could be substantial if footwear types 4 or 6 were to be marketed, because these products must, by necessity, be produced in large-scale enterprises. Finally, insofar as the high-priced types of footwear use leather materials, important backward linkages to domestic leather tanneries may take place. Such linkages are likely to be absent in the case of synthetic materials which, in most developing countries, will have to be imported. Thus, in terms of resource allocation, employment creation, surplus generated and backward linkages, the higher-priced footwear types 2 (or 1) and 5 appear most attractive.

Viewed from the demand side, however, the expensive type 2 or 1 shoes and type 5 sandals are typically suited to serve the higher income brackets and export markets. In terms of price and product characteristics (appearance, durability, comfort, protection, repairability, maintenance and cleaning) these types of footwear are not likely to suit the means and needs of the majority of low-income consumers in developing countries. In contrast, the inexpensive type 6 sandals have very low status and wear out faster than the price differential with other types would justify, but are the only low-priced type of footwear that may be afforded by a large majority of people. Type 5 sandals are of better quality than type 6 sandals, and their production makes use of a technology well suited to the resource endowments of the majority of developing countries. However, their price clearly puts them beyond the reach of the poor consumer.

A low footwear price suited for the purchasing power of low-income groups may therefore create an effective barrier to the production of relatively inexpensive footwear that can be considered appropriate in terms of product characteristics and production technology. Currently, no simple alternative that may fit both requirements (i.e. use of an appropriate technology to produce an appropriate type of footwear) can easily be suggested. To this effect, a market survey would have to be conducted to ascertain effective demand for inexpensive footwear, the effect of threshold prices, and the consumer's appreciation of product performance characteristics.

Next, it should be investigated whether labour-intensive technologies which make use of local raw materials can be applied. In the affirmative, the unit cost of production should match the threshold price,, and the footwear quality be such as to make the latter competitive with type 6 sandals. If these conditions do not apply, a policy decision will have to be taken with a view to reaching a balance between the satisfaction of basic needs on the one hand and the efficient use of local resources on the other.

II. Economic and social appraisal of alternative technologies and products

II.1 Social versus private benefit-cost analysis

Thus far, the analysis of footwear manufacture technologies and that of. alternative types of footwear and scales of production have been conducted in terms of resource costs and sales revenues valued at actual or expected market prices. As a rule, this is the common practice followed by private entrepreneurs, and the obtained results are therefore primarily of interest to them. Applying conventional discounted cash flow (DCF) analysis, the technology and/or type of product which add most to the firm's net present worth (i.e. maximises the net present value (NPV)) are generally adopted and may be considered the best technology or type of footwear from the viewpoint of the firm.

For a number of reasons, however, the best technology from the firm's point of view is not necessarily the appropriate technology from a society's point of view. First, the development objectives of a national, regional or local government usually differ from the objectives of a firm, in particular with regard to employment and income distribution. Second, a number of effects which matter to the government are not or are considered differently by the firm (certain indirect effects, external effects such as skill formation or pollution). Third, market prices actually paid or received do not often properly reflect the scarcity of products and factors of production. Fourth, direct taxation is a cost to the firm, but not to society for which it is a transfer to the government.

When the private and government views on appropriate technology may be expected to differ, the application of social benefit-cost analysis becomes mandatory. Starting from a financial DCF analysis, a social benefit-cost analysis aims at measuring the social benefits and costs of alternative technologies and/or products in terms of their contribution to development objectives and of their use of scarce resources at the actual proposed project location. The evaluation methodologies used in social benefit-cost analysis emphasise the use of accounting prices, not only to correct for distorted market prices but also to reflect development objectives, in particular growth, employment creation and income distribution.

In a number of cases, accounting prices can simply be expressed as conversion factors, with a view to adjusting the market values derived from the financial analysis into social values used in benefit-cost analysis. Similarly, the social net present value can be considered the selection criterion: the technology or product showing the highest social NPV is defined as the "appropriate" technology or type of product. However, as the actual selection mechanism may include more elements than can be taken care of in a NPV criterion, some additional criteria may have to be considered as well.

The substantial progress made in the derivation of consistent sets of accounting prices for investment appraisal is reflected, among others, in the work of Little and Mirrlees (the OECD Manual (1968) and its successor volume (1974)), the UNIDO Guidelines (1972) , the World Bank study of Squire and Van der Tak (1975), and, most recently, in the UNIDO Guide (1978). The latter suggests that project appraisal be broken down into five stages, each of which can be considered a measure of the net benefit of the investment proposal. According to the Guide (1978, p. 3), the five stages consist of:

1. Calculation of the financial or commercial profitability at market prices, using conventional cash flow analysis,

2. Conversion of the standard cash flow table into accounting prices to obtain the net benefit at economic or efficiency prices.

The next steps attempt to adjust the economic value of the net benefits for distributional impacts, viz.:

3. Adjustment for the impact on saving and investment,

4. Adjustment for the impact on income distribution,

5. Adjustment for the production or use of goods whose social values are considered less than or greater than their economic values (luxury consumer goods and basic needs goods, respectively).

Starting from the standard financial analysis (stage 1) the use of accounting prices and adjustment factors permits a complete economic (stage 2) and social appraisal (stages 3-5) by appropriately correcting the original market value of benefits and costs.

Prior to applying social benefit-cost analysis to alternative footwear manufacture technologies or to alternative types of footwear, two remarks should be made about the nature of accounting prices. As long as project alternatives are located in the same area, accounting prices will be the same for each alternative. However, if project alternatives are located in different areas, as may be the case when comparing small-scale alternatives to a large-scale firm, accounting prices are likely to differ from one location to another location. This may be particularly true for the social cost of labour (i.e. the shadow wage rate) since the opportunity cost of the latter is not the same for all locations.

The use of accounting prices with a view to influencing the choice of project and/or technology, reflects policy objectives such as the reduction of unemployment or the focus of development efforts on special groups or regions. Project selection is, of course, not the only instrument to achieve certain development objectives, and its efficiency will have to be weighted against other policy measures. The extent to which a government may wish to use project selection as an instrument of development policy therefore co-determines the actual value of accounting prices. For example, if income distribution is of no interest to a government, social accounting prices will coincide with economic accounting prices.

Accounting values of benefits and costs may be adjusted with a view to improving income distribution (see stages 3 and 4 above). This may be illustrated with respect to the shadow wage rage (SWR). Considering stage 3, if a marginal addition to savings and investment to sustain future growth is considered more valuable than a marginal addition to consumption, if wage income entails a greater commitment to consumption than non-wage income, and if the government considers project selection as an instrument to influence the relation between aggregate consumption and investment, then part of the commitment to consumption which arises from employing more labour can be considered a social cost to society, necessitating an upward revision of the SWR.

Considering stage 4, if wage payments resulting from a project accrue in particular to low-income groups or if the project is located in a depressed area, and if the government is committed to improve the interpersonal and interregional distribution of income through the selection of projects, part of the wage payments is considered a social benefit to society, and will result in a downward adjustment of the SWR. Depending on a project's location, its impact on special socio-economic groups, and commitment to additional consumption, the SWR in the economic analysis will thus require adjustments in the social analysis to reflect various distributional concerns.

II.2 Appraisal of alternative technologies for type 1 footwear

(a) The data base

Following the approach indicated above, the social appraisal of alternative technologies and/or types of footwear can in practice be applied in two steps. First, an economic appraisal is undertaken by converting benefits and costs in the financial analysis into accounting values expressed in economic or efficiency prices. The latter are meant to reflect the real scarcity of commodities and factors of production and serve a purely allocative purpose. Differences between the economic and financial analysis therefore include the effects of correcting for market distortions and imperfections, external effects and taxation.

Next a social appraisal is conducted by adjusting the accounting values of benefits and costs in the economic analysis with a view to reflecting the impact on income distribution and the production or use of goods to which a special social value is attached. The actual conversion of economic into social values can be achieved either by the use of social accounting prices (which convert financial values directly into social values) or by the application of social adjustment factors.

An example of the conversion of benefits into accounting prices is presented in Table V.2 with respect to the machine-intensive type 1 shoe factory reported in chapter IV. Following the Little-Mirrlees approach, the accounting price for imported goods is simply obtained by removing all import duties and surcharges from their market value. The adjustment for the market value of unskilled labour reflects the prevailing unemployment rate, and yields a shadow wage rate which is lower than the market wage. Locally produced inputs are adjusted to allow for the accounting value of labour and imports. As the domestic market for shoes is protected against imports, the extent of protection is removed from the sales value. Due to the corrections needed to bring domestic prices in line with the equivalent of world market prices, no exchange rate adjustment needs to be made. For the sake of convenience, the discount rates are assumed to be the same in both analyses.

Because import duties on most intermediate inputs and equipment are relatively high in Ghana (approximately 40 and 45 per cent, respectively), and import duties on competing types of shoes are reported to be relatively low (around 10 per cent), value added and net operating profit are negatively protected. Net profits and the profit ratio are therefore substantially higher at accounting than at market prices as shown in the last column of Table V.2. The corresponding input structure also appears more in line with the earlier sectoral data mentioned in Chapter IV.

Table V.2 Conversion of cost and benefit items from market prices into accounting prices, Ghana, machine-intensive factory (¢ thousand)

Cost or benefit item		Market value	Import duties	Conversion factor	Accounting value
1.	Direct materials	1,186			828
	Imported	1,140	351		789
	Local	46		0.85	39
2.	Electricity	1		0.69	1
3.	Spares, tools and equipment	11	1		10
4.	Office overhead costs	29			24
	Imported Local	4 25	2	0.85	2 22
	Intermediate inputs (1+2+3+4)	1,227			863
5.	Staff and skilled labour	97		1.0	97
6.	Other labour	27		0.75	20
7.	Depreciation	22			16
	Net operating profit (8-1 through 7)	427			624
	Value added (8-1 through 4)	573			757
8.	Ex-factory sales	1,800		0.90	1,620
9.	Fixed capital	552			394
	Imported	469	146		323
	Buildings	63		0.85	53
	Other local	20		0.90	18
10.	Working capital	901			641
	Imported	820	252		568
	Local	81		0.90*	73
	Total capital (9+10)	1,453			1,035
	Net profit at 10% (8-1 through 7-10% of capital)	282			520
	Net operating profit/ capital %	29.4			60.3

* Weighted average

Source: Calculated from McBain (1977) using additional assumptions.

(b) Economic and social appraisal of alternative footwear manufacture technologies

The full economic appraisal of alternative technologies to produce 1,200 pairs of men's cemented shoe per day is summarised in Table V.3. Whereas the most machine- and most labour-intensive technologies are the same as in Table IV.2 (although their valuation is, in this case, different), the least-cost combination of techniques at accounting prices differs from that at market prices as shown by the number and composition of production workers. As in the case of the financial appraisal, differences in the attractiveness of the three technologies, as indicated by the net annual profit, are relatively minor, and the conclusions reached in Chapter IV apply for the economic analysis as well.

A comparison of the characteristics of the least-cost technology identified on the basis of market prices (Table IV.2) with those identified on the basis of economic accounting prices (Table V.3) shows, somewhat unexpectedly, that the social least-cost technology uses 12 fewer production workers than the least-cost technology identified on the basis of market prices. This interesting result, in a labour-surplus economy, illustrates, in a seemingly unusual fashion, the sensitivity of the least-cost mix of techniques for changes in relative factor prices. First, the high rates of import duty on capital equipment and the relatively small difference between market and accounting wage costs makes the machine-intensive processes relatively less attractive in financial than in economic terms. As a result, the least-cost technology is nearer to the labour-intensive end of the range of technologies when appraised in market instead of economic accounting prices.

Second, as observed in Chapter IV, capital substitues to a large extent for skilled rather than for semi- and unskilled labour when technological alternatives available for a given type of footwear are considered. Hence, at accounting prices, the relatively more capital-intensive combination of techniques implies a shift in the skill mix in favour of processes employing less skilled and more unskilled labour. This shift in the skill mix is reinforced by the lower shadow wage rate for unskilled relative to skilled labour. As a result, the reduction of the total number of production workers by 12 is accompanied by a substantial change in the skill mix of workers: in the economic appraisal, 30 fewer skilled production workers but 18 more semi-and unskilled production workers are employed. The reduction in the number of workers is therefore concentrated in the category of skilled labour. This result is consistent with the assumptions underlying the accounting price determination of labour, viz., that in a labour-surplus economy, unemployment is mainly concentrated among the semi- and unskilled labour and skilled labour can generally be considered a scarce factor of production.

Table V.3 Economic characteristics of producing 1,200 pairs per day of men's cemented shoes with different technologies, in Ghana, valued at accounting prices (¢ thousand in 1972 prices and relative to sales, unless indicated otherwise)

Technology	Most machine-intensive		Most labour-intensive		Least-cost at accounting prices
Cost or benefit item	Value	Ratio	Value	Ratio	Value	Ratio
1. Intermediate inputs	863	0.533	857	0.529	857	0.529
2. Total wages	117	0.072	146	0.090	129	0.080
Staff and skilled labour	97	0.060	120	0.074	104	0.064
Other labour	20	0.012	26	0.016	25	0.016
3. Depreciation	16	0.010	7	0.004	10	0.006
New operating profit (4-1-2-3)	624	0.385	610	0.377	624	0.385
Value added (4-1)	757	0.467	763	0.471	763	0.471
4. Ex-factory sales	1,620	1.000	1,620	1.000	1,620	1.000
5. Fixed capital	394	0.243	184	0.114	249	0.154
6. Working capital	641	0.396	649	0.400	649	0.400
Total capital (5+6)	1,035	0.639	833	0.514	898	0.554
Net profit at 10% (4-1-2-3-10% of capital)	520		527		534
Net operating profit/ capital %		60.3		73.2		69.5
7. Staff (No.)	36		36		36
8. Skilled production workers (No.)	49		90		61
9 . Other production workers (No.)	70		92		86
Total employed (7+8+9)	155		218		183
Fixed capital/production worker ()	3,311		1,011		1,694

Source: Calculated from McBain (1977), using additional assumptions.

II.3 Data requirements and sensitivity analysis

The data required for the financial, economic and social appraisal of projects can be distinguished in two groups. The first group consists of data which refer to the physical characteristics in terms of inputs and outputs of the proposed project and to their valuation at actual or expected market prices. These data are obtained from engineering studies and actual quotations for the financial analysis reflecting the local conditions under which the project is expected to be constructed and operated. In the actual project preparation activities, such data are normally prepared by project engineers and financial analysts.

Most of the additional information for the economic and social analysis refers to "national parameters", such as the social rate of discount, and are, in principle, applied to all projects. These data, mainly accounting prices, are usually prepared by government planners at a central or regional planning office and require the expertise of project economists. In addition, project-specific circumstances may require adjustment or even separate estimation of some of the accounting prices. For example, a nationally determined SWR may have to be adjusted to account for local circumstances, or the accounting price of a special type of footwear be estimated separately if only the accounting price for an "average" type of footwear is available. Such estimates can usually be made independently by professional economists familiar with social benefit-cost analysis.

Even when a project has been properly prepared and appraised, it should not come as a surprise that, in reality, plant performance never turns out exactly as originally foreseen due to such factors as changes in the cost of major inputs, in the rate of capacity utilisation, in the pattern of learning, in the scale of operation, etc. A number of these factors were considered separately for the case of the 1,200 pairs per day men's cemented shoe factories in Ghana and Ethiopia by McBain (1977). The results of this analysis confirm several findings reported for products other than footwear. Low wages and high capital costs, underutilisation of capacity, and small-scale production tend to favour labour-intensive least-cost mixes of techniques. Economies of scale and higher product quality standards introduce a bias towards more machine-intensive techniques. Finally, shift working, changes in labour productivity, in the cost of direct materials and in working capital requirements have little effect on the composition of the least-cost mix of techniques.

III. Towards a national footwear manufacturing strategy

III.1 Factors influencing a national footwear strategy

A national footwear strategy concerns primarily the footwear manufacturing sector as a whole rather than individual projects. It therefore involves a number of factors which are not necessarily relevant to each individual project (linkages, organisation of production, training, price and market policies, tariff policies). However, as sectoral policies directly affect individual investment decisions, the analysis of individual projects is an indispensable input in the formulation of a sectoral strategy. For example, if changes in the least-cost combination of techniques arising from the use of accounting instead of market prices are of a systematic nature in most footwear projects, this finding should have important implications for sectoral, and possibly overall economic and fiscal policies. Another example refers to the existence of various fiscal policies (e.g. high protective tariffs against imports of footwear, exemptions of duties on import of intermediate inputs) biased in favour of large-scale footwear manufacturers, and which discriminate against small-scale producers. If such biases are confirmed at the project level, it is through sector-wide, and possibly economy-wide policy measures, that such a situation can be reversed at the project level.

Findings from this and the previous chapters suggest the following, non-exhaustive list of factors which may influence a national footwear strategy, in particular with regard to product choice, level of scale, and production technology. The interaction between sectoral and project considerations is briefly mentioned whenever this is relevant.

(1) Size and composition of the market. Analysis of domestic and export markets is an integral part of project analysis. Sectoral studies could indicate which sections of the domestic market can best be supplied by home production and which should be supplied through imports (if at all). Export possibilities, by type and quality of product, should be investigated. As far as the domestic market is concerned, its regional dispersion will affect transportation costs and the scale of production.

(2) Purchasing power of various economic groups, with a view to determining the type and quality of footwear for both the home and export market. The identification of the various sub-markets is part of the financial analysis. When certain types of footwear are considered a basic necessity, a (differential) social premium can be attached to the output of special types of footwear so that social benefits will exceed economic benefits when appraising an investment proposal. If preference for the domestic production of basic needs is very strong, cost-effectiveness analysis may suffice.

(3) Size of the initial capital investment. If the profitability of the footwear factory which exhausts the overall investment budget of the footwear manufacturing sector is relatively high, there is a case for requesting more investment funds to be allocated to the footwear sector at the expense of other, less profitable sectors of the economy. The present value of capital costs should be estimated on the basis of the social discount rate.

(4) Availability of material inputs, mainly finished leather. Whether domestically produced or imported, the accounting price of the major raw materials should be used in order to ensure that the true scarcity of these materials is effectively taken into consideration.

(5) Backward linkages (tanneries, local production of machinery and equipment). Expansion of the footwear sector may cause additional demand for finished leather, other intermediate inputs, and capital goods. Such linkages should be considered at the project level once it has been shown that no alternative marketing opportunities for supplying sectors outside the proposed footwear project exist. At the sectoral level, the estimated additional demand generated by backward linkages could serve as useful information for the respective supplying sectors.

(6) Employment and income generation for unskilled labour. These factors depend on the rate of expansion of the footwear manufacturing sector and the nature of the technology adopted. The consequences of policies to foster employment and income generation for unskilled labour can best be analysed at the project level through a low shadow wage rate.

(7) Availability of key skills. These are reflected in the corresponding accounting price, and partly determine the most appropriate technology. Changes in the availability of local skills can be brought about by special training programmes and should be reflected in a relative decrease in the future accounting price of skilled labour.

(8) Foreign exchange savings. Information on the foreign currency component follows from the financial analysis. The valuation of foreign exchange earnings or uses is taken care of in the economic analysis through the choice of a proper accounting price of foreign exchange.

(9) Scale of production. The choice of a scale of production follows from considerations under (1), (2), (3), (6) and (7), as well as sectoral policy preferences regarding centralisation or decentralisation of production.

(10) Organisation of production. It determines the extent of product and process specialisation in separate production units.

(11) Sectoral price, tariff and fiscal policies. These follow, ideally, as implementation measures to ensure successful operation of existing and new footwear projects and should be consistent with national policies.

Some of these factors will be considered in more detail below. As countries differ considerably in size, resource endowments and stage of development, the best combination of products, scale of production and technologies is likely to be different for different countries. The presentation of a strategy for a particular country seems therefore of limited usefulness and will not be attempted in this study.

III.2 Comparative advantage, social considerations and product choice

In terms of resources, especially materials and labour, many developing countries are well placed to expand their production of footwear. Where basic skills are available and the necessary materials are of sufficient quality (either from domestic tanneries or imported, see sub-section III.3 below), the present level of wages in most developing countries gives them a distinct cost advantage over high-income countries. The high level of labour productivity resulting from machine-intensive methods of production in the developed countries is more than offset by the high level of wages, so that labour costs per unit of output are substantially higher than in most of the less-developed countries, even though labour productivity in the latter countries could be relatively low.

Because of the variety of footwear products which can be produced at competitive prices, developing countries are faced with two major options with respect to potential markets to be served: (1) The development of a domestic industry, the market for which can be initially based on import substitution, and subsequently on increasing per capita incomes.¹ (2) The development of an export industry based on the considerable comparative advantage which developing countries can be shown to possess. Whereas the cost advantage will generally be sufficient to face international price competition, quality aspects are equally crucial for a successful penetration of the export market. Hence, the level of skills, management and quality control necessary to enter the international markets will generally be high, and is likely to be very different from the lower technical levels at which domestic products are manufactured. In the absence of

¹ Being a basic needs good, the demand for footwear increases relatively fast with increasing incomes among low-income groups. experience in modern footwear assembling methods, the development of production units for the domestic market seems therefore essential to acquire the technical and managerial experience necessary to compete successfully in the export markets.

As shown in Chapter IV and Section I , the manufacture of high-quality leather-uppered footwear appears very attractive. The use of valuable leather materials for the production of the more expensive types of footwear is justified by the high price these products command, the favourable economic surplus and private profitability, the high level of employment creation resulting from the generally labour-intensive way in which these products are manufactured, and the type of craft-based skills required. By contrast, the most inexpensive types of footwear, which are meant to serve the large majority of the population, are based on synthetic materials and require large-scale, capital-intensive production units, which create very few jobs, require specialised technical skills, have less backward linkages, if any, and are a potential threat to small-scale producers of low quality leather footwear. It is therefore suggested that future research and investment be directed towards the development of those low-priced, medium-quality footwear products that can be produced in a more labour-intensive way, make use of local, inexpensive materials and are complementary to rather than competitive with existing small-scale producers.

In cases where certain types of footwear are favoured because they constitute basic needs goods (e.g. plastic sandals), such a preference may be expressed by adding a (differential) social premium to their economic value when appraising alternative footwear projects. This is particularly relevant when low-income consumers cannot afford the initial expense to buy high-quality footwear, but the product price of the cheaper substitute matches the threshold price for low-income groups. Obviously, such a social premium will be difficult to establish in actual situations. In such a case, a practical device is to estimate the switching value of the premium, i.e. the amount by which the economic benefits of producing a particular type of footwear will have to be raised in order to make the project socially preferable to other types of footwear. In this way, the social cost of satisfying certain basic needs (employment forgone, less use of local materials, fewer foreign exchange savings) can be ascertained and be weighted against other alternatives.

III.3 Backward linkages and availability of raw materials

(a) Backward linkages

The major backward linkages resulting from the operation of footwear factories refer either to finished leather (used for the production of leather footwear) or to synthetic materials, mainly plastics, used in the manufacture of synthetic footwear. Minor backward linkages include the production of various tools and pieces of equipment used in footwear manufacturing and that of intermediate inputs such as nails, glue, fabrics, etc.

Backward linkages are of interest in the only case where these tools, equipment and materials can be produced locally since imports do not contribute to an increase of the national value added. Furthermore, the footwear manufacturing sector may need to compete against other sectors and/or exports in order to acquire the needed inputs (e.g. leather) whenever the supplying sectors cannot expand their production (supply constraint). In this latter case, benefits derived from backward linkages would be limited. The use of leather in the local manufacture of footwear instead of exporting it does not yield backward linkages with the leather producing sector.

(b) Raw materials

The decision as to which raw materials should be favoured is not an easy one to make. This may be illustrated by the following examples. If a country has no local capacity to supply plastics suitable for footwear production, but does have a number of leather tanneries producing semi-finished (wet blue or crust) and finished leather of different quality, the decision to manufacture leather types of footwear may seem obvious at first sight, but could nevertheless be premature, or even incorrect, when based on the availability of local leather only. First, if local production of plastics used for footwear is not competitive with imports, the decision not to produce plastics locally is a rational one: the foreign exchange impact of an inefficient production unit would certainly exceed the foreign exchange savings through import substitution. However, the production of sandals with imported plastics may still be economically or socially feasible (i.e. no project should be penalised for its using imported synthetic materials). Second, when leather has a potential export market, the use of locally produced finished leather in domestic footwear production implies that potential foreign exchange earnings are foregone. Under these circumstances, it should first be ascertained which alternative use of leather yields the highest return. This factor may further justify the import of plastics for the production of sandals.

The quality of locally produced leather, reflecting both the quality of hides and skins and local tanning capabilities, is an equally important factor. If high-quality hides and skins are locally available, but tanning capabilities are yet insufficient to produce high-quality finished leather, a decision must be taken regarding the stage at which leather can best be processed (wet blue, crust or finished). In such a situation, it could well be rational to export semi-finished leather and, at the same time, import finished leather for domestic purposes. Meanwhile, the local capacity to finish leather could be gradually built up. If, on the other hand, hides and skins are of a medium to poor quality and export prospects for leather are therefore less favourable, local processing up to the stage of finished leather could be justified when the leather can be used for the manufacture of medium-quality leather footwear meant to serve local markets.

III.4 Employment, technological choice and skill requirements

For typical conditions in developing countries, the least-cost combination of techniques to manufacture footwear products turns out to be labour-intensive in terms of fixed capital per worker. Thus, the adoption of the latest developed-country technology cannot be generally recommended for a less-developed country.

Savings in capital costs and the creation of additional employment can be substantial when employing a labour-intensive or intermediate technology instead of a machine-intensive technology. Both the Ghanaian and Ethiopian case studies show that application of the most machine-intensive instead of the most labour-intensive technology implies almost a doubling of fixed capital requirements, a reduction of total employment of about 30 per cent, and hence an increase in the capital-labour ratio by a factor of 3. However, the effect on overall profitability of choosing least-cost technologies, though positive, is less pronounced. This finding carries two important implications: (1) it explains why firms adopting latest developed country technology easily survive in an environment of low wages and high capital costs, and (2) a strong concern with employment creation need not imply important sacrifices in other areas, although the implementation of labour-oriented development strategies may face serious problems in view of the first implication.

It may also be recalled that the nature of capital-labour substitution in manufacturing a particular type of footwear product appears to be such that capital substitutes to a larger extent for skilled than for unskilled labour (i.e. labour-intensive technologies are characterised by both a large number and a high proportion of skilled workers). When comparing least-cost technologies for different types of footwear, variations in labour requirements across products are substantial, but variations in fixed capital per employee are comparable to those amongst technologies for the same type of footwear. The nature of capital-labour substitution across products is such that capital-intensive products tend to be associated with a high proportion of skilled labour.

As far as leather-upper shoes are concerned, opportunities for substitution are greater between alternative technologies for a particular type of footwear than between different types of shoes. The reverse is true, however, when footwear made of synthetic material is considered. When the various types of footwear made of different materials are considered simultaneously, variations in capital-intensity are definitely greater across products than between technologies to produce a particular type of footwear.

The higher priced leather-upper shoes generally demand an absolute number of skills far higher than that for the less expensive varieties made of synthetic materials. It is likely, however, that the type of skills required for the manufacture of synthetic footwear (both in production and maintenance activities) is far more scarce in developing countries than the type of skill necessary for the other, more craft-based products.

III.5 Other factors

A number of additional factors should be taken into consideration when formulating a national footwear production strategy. These factors, already mentioned under points (7) to (11) in section III.1 are briefly discussed below.

(a) Foreign exchange savings

In general, the majority of developing countries -especially small to medium size countries - do not yet have a strong capital goods sector. Thus, equipment used in many industries - including the footwear industry -must generally be imported. The foreign exchange component of fixed capital equipment used in footwear manufacturing will generally be high whenever one or more of the following conditions apply:

- use of machine-intensive processes instead of labour-intensive ones to produce the same type of footwear,

- choice of capital-intensive footwear products (e.g. plastic sandals) instead of labour-intensive products (low quality leather shoes),

- adoption of large scales of production.

The choice of raw material may also increase the fraction of foreign exchange in total production costs.

On the other hand, the production of footwear may generate substantial foreign exchange earnings through exports to neighbouring developing countries and/or developed countries. Whenever exports are intended for developed countries, the required footwear quality and the volume of exports should generally be such as to 'require the use of capital-intensive technology and the adoption-of large scales of production, thus offsetting a relatively large proportion of foreign exchange earnings (especially if quality leather must also be imported). In general, foreign exchange earnings should be larger than foreign exchange expenditures. However, the difference may not always be so large as to justify the export of footwear to developed countries.

On the other hand, export of footwear to neighbouring developing countries could be particularly attractive since the required footwear type and quality should be similar to those in demand at home, and since the volume of exports need not be too large. Consequently, labour-intensive or intermediate technologies and small scales of production may be adopted for the production of this type of exports. In general, these exports should have a large positive impact on the balance of payments than would exports of footwear to industrialised countries.

(b) Skill requirements

Whereas the machine-intensive production processes and products require fewer skills than do the more labour-intensive alternatives, the nature of the skills differ considerably from the more craft-based skills traditionally used in the labour-intensive methods of production. At higher scales of production, this effect will be reinforced, both at the technical and managerial levels. Large-scale, machine-intensive footwear factories will therefore normally require special training programmes and/or availability of expatriate personnel.

(c) Scale of production

Least-cost combinations of techniques tend to become more labour- and skill-intensive at low scales of production. In the modern, organised sector, economies of scale through savings in fixed capital, staff and skilled labour favour the set-up of larger production units. Compared with small- and medium-scale producers in the modern sector (employing up to 50 workers), however, artisanal or handicraft production of footwear in the traditional, unorganised ("informal") sector appears more attractive. Compared with producers in the large-scale modern sector, artisanal production of footwear, usually carried out in units employing less than 10 or even less than 5 workers, is less profitable, but creates considerably more employment (in the Ethiopian case 3 to 4 times as much). Policies which favour the profitability of small footwear production units (e.g. through improvement of techniques, training) in markets where they can be expected to remain competitive, should therefore be encouraged. Policy measures may include the creation of better marketing facilities, better access to and conditions for obtaining credit, ensured supply of essential inputs, improvement of production methods to increase productivity, etc.

In addition to these measures, policies which encourage specialisation by small-scale and artisanal enterprises can substantially improve their profitability. Details of such policies were discussed in Chapter IV.

(d) Sectoral price, tariff and fiscal policies

When contrasted with an analysis based on local market prices, the analysis of alternative technologies based on economic and social accounting prices has important policy implications for sectoral price and trade policies and overall fiscal policies. In general, these policies should be designed in such a way as to induce the selection of those least-cost combinations of techniques that suit a particular country best in terms of its development objectives and real scarcity of factors of production. To this effect, (selective) price and tariff measures and various incentive schemes will have to be utilised. For instance, if the prevailing system of market prices is such that, in a labour-surplus environment, no appreciable labour-intensive techniques are selected, but rather machine-intensive varieties are preferred, the following measures to redress this situation may be applied: (1) high import tariffs on imported capital goods, (2) no application of low interest rates "to stimulate investment" indiscriminately, (3) wage subsidies, implemented through fiscal exemptions, to encourage the use of labour-intensive methods and equipment, (4) no excessive tariff protection which would remove incentives to become competitive and enable machine-intensive technologies to survive easily. Although the specific measures to be taken will differ from country to country, the examples mentioned serve as an illustration of the type of measures conducive to the adoption of more labour-intensive technologies.