Section 4: Market saturation and adding-up

4.1 INTRODUCTION

The objective of this section is to explore the degree to which the adding-up problem, or the possibility that increased production and exports may lead to a proportionately lower increase or decline in sale revenues, is relevant in selected NTAE markets, countries or group of countries. The analysis of the adding-up problem adheres to that by Imran and Duncan (1988) and Akiyama and Larson (1994) and is based on the estimation of the elasticity of export revenue with respect to volume. Sub-section 4.2 contains a brief discussion of the methodology, whilst sub-section 4.3 presents the results of the econometric estimation of demand and supply functions of selected NTAE markets, the calculation of the elasticity of export revenue with respect to volume for each country and commodity under examination and discusses the extent to which the adding up problem is relevant for the selected NTAEs.

4.2 METHODOLOGY

The adding-up problem was first discussed by Bhagwati (1958) in the context of immizerizing growth. In the body of research that followed, several authors examined the adding-up problem from the partial equilibrium perspective focusing on particular commodities that are important, as far as revenue and economic development are concerned (Schiff, 1994; Akiyama and Larson, 1994; Imran and Duncan, 1988). The objective of such analyses remains the assessment of optimal commodity strategies for individual countries and for country groups or regions and the provision of recommendations for trade, investment and lending policies. A production strategy for a small country that, in isolation, faces an infinite price elasticity of demand may not be optimal for a group of small countries that together face an inelastic demand both at home and at the export market, as an increase in production may deteriorate their terms of trade.

The assessment of the relevance of the adding-up problem for a particular commodity is based on estimation of the corresponding elasticities of demand that a country or a group of countries face. Imran and Duncan (1988) and subsequently Akiyama and Larson (1994) provided a framework for the estimation of elasticities of demand and the analysis of the adding-up problem. For a particular country or region and commodity, the authors state the adding-up problem in terms of the elasticity of export revenue with respect to volume (ERV) as follows:

where h is the price elasticity of demand facing the country, p is the price of the commodity and C' is marginal cost of production. The price elasticity of demand facing the country (that is the elasticity of the demand at home, as well as at the export market) is defined as:

where and denote the price elasticity of demand at the world market and the elasticity of supply for the rest-of-the-world respectively, whilst m^row denotes the market share of the rest of the world and m^h the market share of the country under examination.^[30]

The above relations demonstrate the following:

(i) The elasticity of export revenue with respect to volume exported usually lies between 0 and 1. The ERV will be equal to 1 when the market share of the country under examination m^h is very small and will be equal to zero in the unlikely case of zero marginal costs, or when the country under examination has a very large market share.

(ii) The higher the world price elasticity of demand is, the higher the price elasticity of demand h facing the country, and the higher the elasticity of export revenue with respect to volume. This suggests that traditional agricultural commodities are likely to encounter adding-up problems, as the corresponding price elasticities of demand tend to be low in absolute value. For example, the own-price elasticities for wheat in the World Food Model of FAO lie between -0.20 and -0.60, whilst those for maize lie between -0.20 and -0.55. Traditional commodities that are produced by developing countries are also subject to low elasticities of demand. For example, estimated own-price demand elasticities in the World Coffee Model of FAO lie between -0.10 and -0.37. Moreover, the fact that market shares in traditional agricultural exports are usually concentrated exacerbates the adding-up problem.

(iii) The higher the elasticity of supply of the rest-of-the-world is, the higher the price elasticity of demand h facing the country, and the higher elasticity of export revenue with respect to volume. Elastic supply in the rest-of-the-world results in an increased share of the rest-of-the-world in world production and a decreasing share of the country under consideration, suggesting higher elasticity of export revenue with respect to volume.

(iv) The higher the market share of a country m^h, the lower the price elasticity of demand h facing the country, and the lower the elasticity of export revenue with respect to volume, as large producers may exert power over the price, resulting in decreased revenue as sales increase.

4.3 ESTIMATION AND RESULTS

In order to calculate the elasticities of export revenue with respect to volume (ERV), the own price elasticities for demand h and the elasticities of supply of the rest-of-the-world for each country under consideration were estimated for selected NTAEs, namely: asparagus, avocados, green peas, green corn, cabbages, green beans, mangoes and pineapples.

The estimation of world demand and rest-of-the-world supply functions was carried out utilizing FAO data on production, consumption and trade of the commodities under consideration for the period 1970-2000. Due to the lack of data on prices, the export unit value at the world level was used as a proxy.

World demand per capita was taken to depend on the real price of the commodity and the real Gross Domestic Product per capita which is used as a proxy for income.^[31] World demand functions were specified as static and thus the estimated elasticities can be thought of as reflecting the response of economic agents to price changes in the long run. Several specifications were tried in order to take into account the simultaneity bias, as both quantity consumed and produced at the world level and the price of each commodity are endogenous variables. Initially world demand functions were estimated by 2SLS utilizing lagged prices and GDP per capita as instruments. However, it was decided to introduce a dynamic element in the rest-of-the-word supply function. Supply was taken to depend on the lagged ratio of the world price to the unit import value of fertilizer which was used as a proxy for input costs. Consequently, as the specification implies that quantity produced is pre-determined, the world demand functions were specified in their inverse form, with the real price being depended on quantity consumed per capita and the real GDP per capita, thus taking into account the endogeneity of the world price. Both world demand and rest-of-the-world supply functions were estimated by OLS.

The estimated world demand own-price elasticities are presented in Table 4.1.^[32] The estimates suggest that the selected NTAEs have higher own-price elasticities than traditional agricultural commodities and therefore, may be less likely to be subject to adding-up problems, depending on the market shares of the countries under examination.

Table 4.1: Estimated own-price demand elasticities for selected NTAEs

For each country under examination, the estimated own-price demand elasticities were used in conjunction with estimates of the elasticity of supply for the rest-of-the-world and the corresponding market shares in order to calculate the ERV (see equations (1) and (2)). Sub-sections 4.3.1 to 4.3.9 present the estimated own-price elasticities for demand and the corresponding elasticities of export revenue with respect to volume (ERVs) for each commodity and country under examination.

4.3.1 Asparagus

Table 4.2 presents the estimates on the asparagus market. These indicate that South American asparagus producing countries, as a group, do not face serious adding-up problems with ERVs lying between 0.95 and 1.00.

Table 4.2: Asparagus: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

However, it appears that China faces a relatively inelastic demand reflecting its large share in the world production (0.84), suggesting that an increase in production will have a negative impact on the price of asparagus partly offsetting the increase in sale revenues.

4.3.2 Avocados

The estimates of own-price elasticity of demand and ERVs are shown in Table 4.3.

Table 4.3: Avocados: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

The estimates indicate that most of the South American avocado producing countries, with the exception of Mexico and Colombia, face highly elastic demands with ERVs ranging from 0.90 to 0.99. For Mexico, an important producer of avocados with a production share of 0.36l, the ERV was estimated to be equal to 0.66, suggesting an increase in production and exports will result in depressed prices and a proportionately lower increase in revenue.

The adding-up problem is not directly relevant to small groups of African and Asian avocado producers, with the possible exception of Indonesia. However, simultaneous expansion of avocado production in Africa and Asia will result in a proportionately smaller increase in export revenues.

4.3.3 Cabbages

Table 4.4 shows the estimated own-price elasticities for demand and the corresponding ERVs for selected cabbages producing countries. The estimated elasticities indicate that China which has a production share of 0.4, faces a relatively inelastic demand (-1.32) and a low ERV, suggesting that given an increase in volume will depress prices and will result in a proportionately smaller increase in revenue.

Table 4.4: Cabbages: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

The ERV for Asia as a single entity, excluding China, also suggests that adding-up may constitute of a problem, as a simultaneous expansion of production and exports will deteriorate the terms of trade in Asian countries. Both South American and African cabbage-producing countries face high elasticities of demand and appear not to be subject to adding-up problems.

4.3.4 Green beans

The estimated own-price elasticities for demand and the corresponding ERVs for selected green beans producing countries are presented in Table 4.5. An inelastic own-price elasticity of world demand (-0.70) in conjunction with a relatively high share of China in the green beans world market (0.34) suggests that China has an ERV close to zero, indicating that increases in sales of cabbages by China will have a detrimental impact on the world price and subsequently on the export revenues of all exporting countries.

Table 4.5: Green beans: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

Asia, as an entity has an ERV of 0.76, indicating that a simultaneous expansion in production and exports may also depress market prices and deteriorate the terms of trade. However, Far East countries only (Bangladesh, India, Indonesia and Thailand) do not appear to be subject to an adding-up problem.

African and South American countries both in isolation and as a group face high elasticities of demand and high ERVs, indicating that increases in production and exports will bring about approximately equal increases in revenue.

4.3.5 Green corn

Table 4.6 presents the estimated own-price elasticities for demand and the corresponding ERVs for selected green corn producing countries. The estimates suggest that individual producers, with the possible exception of Nigeria for which the ERV is estimated to be equal to 0.82, are not likely to experience a slowdown in their revenue following an increase in sales.

Nevertheless, a simultaneous expansion of production and export sales by Asian, South American and especially African countries may result in a decrease in world prices and a proportionately smaller increase in export revenues, suggesting that the adding-up problem is relevant as far as green corn is concerned.

Table 4.6: Green corn: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

4.3.6 Green peas

The estimated own-price elasticities for demand and the corresponding ERVs for selected green peas producing countries are presented in Table 4.7. The estimates indicate that an increase in sales of green peas by India and China is likely to result in a proportionately lower increase in revenues as their ERVs amount to 0.39 and 0.52 respectively.

Table 4.7: Green peas: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

Algeria and Morocco, as well as all South American countries, both in isolation and as a group, face high elasticities of demand and do not appear to be subject to an adding-up problem. The estimated ERV for Egypt is equal to 0.89, suggesting that an expansion in production and exports may result in a decrease in world prices that will partly offset the corresponding increase in export revenues.

4.3.7 Mangoes

Table 4.8 presents the estimated own-price elasticities for demand and the corresponding ERVs for selected mango producing countries. Among the Asian countries, India and China face an elasticity of demand of -1.64 and -3.87 and an ERV of 0.39 and 0.74 respectively, indicating that an expansion in production and exports by these countries will result in a proportionately lower increase in revenues.

Table 4.8: Mangoes: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

Individual African and South American countries, with the exception of Mexico, face highly elastic demand and, in the event of an increase in production, they are not likely to experience an increase in revenue that would be proportionately lower than the corresponding increase in sales. However, both Africa and South America as single entities are likely to experience adding-up problem following a simultaneous increase in production and sales both at home and the world market.

4.3.8 Pineapples

The estimated own price elasticities for demand and the corresponding ERVs for selected pineapple producing countries are presented in Table 4.9. Amongst the Asian mango producers, Thailand, Philippines and China face relatively low price elasticities of demand and are likely to experience a slowdown in their revenue, following an increase in production. Asia, as an entity is subject to serious adding-up problem. The estimated ERV of 0.17 indicates that a simultaneous increase in export sales by Asian countries will result in export revenues increasing by only a small proportion.

Table 4.9: Pineapples: own-price elasticities for demand (h) and elasticities of export revenue with respect to volume (ERVs)

African and Asian countries, with the exception of Nigeria, face highly elastic price elasticities of demand. Nevertheless, the adding-up problem is relevant as both Africa and South America, as single entities, have relatively low ERVs.