Search for a command to run...
Supermarkets are traditionally viewed by development economists, policymakers, and practitioners as the rich world's place to shop. The three regions discussed here have a great majority of the poor on the planet. But supermarkets are no longer just niche players for rich consumers in the capital cities of the countries in these regions. The rapid rise of supermarkets in these regions in the past five to ten years has transformed agrifood markets at different rates and depths across regions and countries. Many of those transformations present great challenges—even exclusion—for small farms, and small processing and distribution firms, but also potentially great opportunities. Development models, policies, and programs need to adapt to this radical change. This paper describes the transformation of agrifood systems in Africa, Asia (excluding Japan), and Latin America. First, we describe the traditional retail and wholesale system in the midst of which emerged modern food retailing and its procurement system. Second, we discuss the determinants of and patterns in the diffusion of supermarkets in the three regions. Third, we discuss the evolution of procurement systems of those supermarkets, and consequences for agrifood systems. At the end, we hint at emerging implications for farms and firms in the region. As development proceeded in the currently developed world, and is proceeding in the three developing regions under study here, markets shift from fragmented, local markets (such as village markets with wholesale and retail functions) to larger, centralized wholesale markets. This “de-fragmentation” tends to occur first in dry goods such as grains and later in “fresh products”—fruits and vegetables, meat, fish, eggs, and milk. There is progressive fresh-food market integration through the rise of medium/long distance trade and the establishment of specialized production areas, as one would expect from the theory of specialization and comparative advantage. This integration is accelerated by urbanization and improvements in roads, and thus takes place at different rates over regions, countries, and zones. Governments have also intervened to spur growth in the fresh foods and grains wholesale sector, such as in Brazil in the 1970s/1980s and in China now. Governments have also intervened directly in grain wholesale and even retail marketing, such as the Fair-Price Shops in India and the (now defunct and eclipsed) Foodstuff Stores in China. Governments seldom, however, intervened in the fresh food retail sector that continued, until the recent rise of supermarkets, to be dominated by mom and pop stores, street fairs, and central markets. That is, traditionally, a major change occurred in the wholesale sector with only gradual effects on the food retail sector. In the latter stages of these changes in wholesale markets in Europe and the United States were concomitant changes in the retail sector, with the advent of self-service stores and then consolidation of the retail sector via the rise of supermarket chains in the past fifty to eighty years. A reversal of the traditional causal direction then occurred: retail transformation deeply changed the wholesale sector and thus the conditions faced by farmers. Below we show that a similar retail transformation has already made great headway in most countries of the three developing regions in only one decade. The determinants of the diffusion of supermarkets in developing regions can be conceptualized as a system of demand by consumers for supermarket services, and supply of supermarket services—hence investments by supermarket entrepreneurs. Both functions have as arguments incentives and capacity variables. On the demand side, several forces drive the observed increase in demand for supermarket services (and are similar to those observed in Europe and the United States in the twentieth century). Demand-side incentives were as follows. First, urbanization, with the consequent entry of women into the workforce outside the home, increased the opportunity cost of women's time and their incentive to seek shopping convenience and processed foods to save cooking time. Second, supermarkets and large-scale food manufacturers spurred the secular reduction in processed food prices. Demand-side capacity variables were as follows. First, real mean per capita income growth in many countries of the regions during the 1990s, along with the rapid rise of the middle class, increased demand for processed foods (the entry point for supermarkets as they could offer greater variety and lower cost of these products than traditional retailers due to economies of scale in procurement). Second, rapid growth in the 1990s in ownership of refrigerators meant ability to shift from daily shopping in traditional retail shops to weekly or monthly shopping. Growing access to cars and public transport reinforced this trend. The supply of supermarket services was driven by several forces, only a subset of which overlap with the drivers of initial supermarket diffusion in Europe and the United States. The supply-side drivers were three. First, foreign direct investment (FDI) was a crucial factor. The development of supermarkets was very slow before (roughly) 1990, as only domestic/local capital was involved. In the 1990s and after, FDI was crucial to the take-off of supermarkets. The incentive to undertake FDI by European, U.S., and Japanese chains, and chains in richer countries in the regions under study (such as chains in Hong Kong, South Africa, and Costa Rica) was due to saturation and intense competition in home markets and much higher margins to be made by investing in developing markets. For example, Carrefour earned three times higher margins on average in its Argentine compared to its French operations in the 1990s. Moreover, initial competition in the receiving regions was weak, generally with little fight put up by traditional retailers and domestic-capital supermarkets, and there are distinct advantages to early entry, hence occupation of key retail locations. Attracting FDI were policies of full or partial liberalization of retail sector FDI undertaken in many countries in the three regions in the 1990s and after (e.g., China in 1992, Brazil, Mexico, Argentina in 1994, various African countries via South African investment after apartheid ended in the mid 1990s, Indonesia in 1998, India in 2000). Overall FDI grew five to ten fold over the 1990s in these regions (UNCTAD); growth of FDI in food retailing mirrored that overall growth. A second crucial supply-side factor was the revolution the past decade in retail procurement logistics technology and inventory management. New practices included efficient consumer response, ECR, an inventory management practice that minimizes inventorieson-hand, and use of internet and computers for inventory control and supplier-retailer coordination. These appeared first in developed countries and then in the late 1990s and early 2000s swept developing countries among leading chains, through home-office guidance for local branches of global chains, and knowledge transfer and imitation and innovation by domestic supermarket chains. These changes were in turn key to centralizing procurement and consolidating distribution in order to “drive costs out of the system,” a phrase used widely in the retail industry. Substantial savings were thus possible through efficiency gains, economies of scale, and coordination cost reductions. China Resources Enterprise, for example, notes that it is saving 40% in distribution costs by combining modern logistics with centralized distribution in its two large new distribution centers in southern China. These efficiency gains fuel profits for investment in new stores, and, through intense competition, reduce prices to consumers of essential food products. The incentive and capacity determinants of demand for and supply of supermarket services vary markedly over the three regions, within individual countries, and within zones and between rural and urban areas at the country level. Several broad patterns are observed. First, from the earliest to the latest adopter of supermarkets, the regions range from Latin America to Asia to Africa, roughly reflecting the ordering of income, urbanization, and infrastructure and policies that favor supermarket growth. The overall image is of waves of diffusion rolling along. The first wave hit major cities in the larger or richer countries of Latin America. The second wave hit in East/Southeast Asia; the third in small or poorer countries of Latin America and Asia including, for example, Central America and Southern then Eastern Africa. By this time, secondary cities and towns in the areas of the “first wave” were being hit. The fourth wave, just starting now, is hitting South Asia. Latin America has led the way among developing regions in the growth of the supermarket sector. While a small number of supermarkets existed in most countries during and before the 1980s, they were primarily financed by domestic capital and tended to exist in major cities and wealthier neighborhoods. That is, they were essentially a niche retail market serving at most 10–20% of the national food retail sales. However, by 2000, supermarkets had risen to occupy 50–60% of national food retail among the Latin American countries, almost approaching the 70–80% share of the United States and France. In a single decade Latin America had the same development of supermarkets that the United States experienced in five decades. The supermarket share of food retail sales for the leading six Latin American countries averages 45–75%: Brazil has the highest share, followed by Argentina, Chile, Costa Rica, Mexico, and Colombia. Those six countries account for 85% of the income and 75% of the population in Latin America. Supermarket sectors of other countries in the region have also grown rapidly, but these started later and from a lower base. For example, supermarkets accounted for 15% of national food retail in Guatemala in 1994 and today account for 35% (Reardon and Berdegué). The development of the supermarket sector in East/Southeast Asia is generally similar to that of Latin America. The “take-off” stage of supermarkets in East/Southeast Asia started, on average, some five to seven years behind that of Latin America, but is registering even faster growth. The average processed/packaged food retail share over several Southeast Asian countries—Indonesia, Malaysia, and Thailand—is 33%, but is 63% for East Asian countries—Republic of Korea, Taiwan, and Philippines—(ACNielsen). A rough rule of thumb, applicable from Latin America, is that the share of supermarkets in fresh foods is roughly one-half of the share in packaged foods, hence roughly 15–20% in Southeast Asia and 30% in East Asia outside China (and Japan). The 2001 supermarket share of Chinese urban food markets was 48%, up from 30% in 1999. Assuming the urban share of the total Chinese population to be approximately one-third, the total national packaged/processed food retail share of supermarkets is around 20%, similar to the share for supermarkets in overall food retail for Brazil or Argentina in the early 1990s. However, the rate of store growth is three times faster in China in 2003 than it was in Brazil and Argentina in the 1990s. The most recent venue for supermarket take-off is in Africa, especially in Eastern and Southern Africa. South Africa is the front-runner, with roughly a 55% share of supermarkets in overall food retail and 1,700 supermarkets for 35 million persons. The great majority of that spectacular rise has come since the end of Apartheid in 1994. To put these figures in perspective, note that 1,700 supermarkets is roughly equivalent to 350,000 mom and pop stores or “spazas” in sales. Moreover, South African chains have recently invested in thirteen other African countries as well as in India, Australia, and the Philippines. Kenya is the other front-runner, with 300 supermarkets. Zimbabwe and Zambia have fifty to hundred supermarkets each (Weatherspoon and Reardon). Second, within each of the three very broad regions there are large differences over subregions and countries. Usually, these can be supermarket-growth-ranked according to the variables in the supply and demand model presented above. In Latin America, for example, Brazil with a 75% share of supermarkets in food retail store sales can be contrasted with Bolivia with at the most 10%; in developing Asia, Korea with 60% can be contrasted with India with 5%; and in Africa, South Africa with 55% can be contrasted with Nigeria with 5%. Third, the take-over of food retailing in these regions has occurred much more rapidly in processed, dry, and packaged foods such as noodles, milk products, and grains, for which supermarkets have an advantage over mom and pop stores due to economies of scale. The supermarkets' progress in gaining control of fresh food markets has been slower, and there is greater variation across countries because of local habits and responses by wetmarkets and local shops. Usually the first fresh food categories for the supermarkets to gain a majority share include “commodities” such as potatoes, and sectors experiencing consolidation in first-stage processing and production: often chicken, beef and pork, and fish. In Brazil, where the overall food retail share of supermarkets is 75%, the share in Sao Paulo of fresh fruits and vegetables is only 25%. This kind of rough “three to one” ratio is typical in the regions. This difference is also not uncommon in developed countries: in France, supermarkets have 70% of overall food retail, but only 50% of fresh fruits and vegetables. The convenience and low prices of small shops and fairs, with fresh and varied produce for daily shopping, continues to be a competitive challenge to the supermarket sector, with usually steady but much slower progress for supermarkets requiring investments in procurement efficiency. Despite the slower growth in supermarkets' share of domestic produce, it is staggering to calculate the absolute market that supermarkets now represent, even in produce, and thus how much more in other products where supermarkets have penetrated faster and deeper. For example, Reardon and Berdegué calculate that supermarkets in Latin America buy 2.5 times more fruits and vegetables from local producers than all the exports of produce from Latin America to the rest of the world! This should be contrasted with the nearly exclusive focus on produce exports in government and donor programs to spur growth in agricultural diversification and to help producers gain access to dynamic markets. Fourth, the supermarket sector in these regions is increasingly and overwhelmingly multinationalized (foreign-owned) and consolidated. The multinationalization of the sector is illustrated in Latin America where global multinationals constitute roughly 70–80% of the top five chains in most countries. That supermarket sector growth is substantially driven by FDI from outside these regions differentiates supermarket diffusion in these regions from that in the United States and Europe. The tidal wave of retail FDI was mainly due to the global retail multinationals, Ahold, Carrefour, and Wal-Mart, smaller global chains such as Casino, Metro, Makro, and multinationals such as and In some larger countries, domestic chains, in with global multinationals, have the For example, the top in Brazil is with Casino, of France, since and the top in China is the national with some stores, in and in 2003 as a of and two The rapid consolidation of the sector in those regions is in the United States and Europe. For example, in Latin America the top five chains per country have of the supermarket sector 40% in the United States and in The are for example, of each on food by are now in The consolidation takes place mainly via foreign of local chains (and by larger domestic chains smaller chains and These multinationalization and consolidation the supply of supermarket diffusion and retail multinationals have access to investment from and to that is much than is the by their domestic The multinationals also have access to practices in retail and logistics some of which they developed as domestic firms have they have had to similar these firms had to with global multinationals or had to from their (e.g., the national or national as from the diffusion model the and patterns of diffusion have over large and small cities and and over and poor consumer In there has been a from supermarkets' only a small niche in capital cities serving only the rich and middle well the middle in order to deeply into the food markets of the have also from cities to and in some countries, already to small towns in rural 40% of smaller towns now have supermarkets, as many towns even in countries supermarkets are now rapidly the top cities of China in the and are to smaller cities and to the poorer and more and and The to products for retail rest with the procurement in supermarket chains. in the United Chile, or they are under several from supermarket under intense competition and are between the traditional retailers fresh local products on one side, and efficient global on the other The procurement to this by and costs and the varied demand of procurement seek to and products with and of supermarkets usually that they have to procurement systems to and outside of the traditional wholesale systems because the latter their because they to out the cost by the with the American or the produce in the study regions is by poor and public infrastructure such as chains, and among and is usually and in and in to The be due to market such as and market Several broad patterns of changes are observed in the procurement that First, there is a of procurement As the number of stores in a supermarket there is a to shift from a procurement to a distribution serving several stores in a or a region several This is by procurement and increased use of centralized increased of also occur in the procurement and in the produce distribution efficiency of procurement by coordination and other it increase transport costs by of the products. Usually retailers have a or where they from to centralized procurement as economies of scale and and on the and of the For example, we observed a small in an in China that invested recently in a distribution for processed/packaged foods but continues to buy fresh foods from the market By a national invested in a large for packaged/processed foods and has recently a large for fresh foods as produce has a and these products have a in profits and The top three global retailers have made or are more centralized procurement system in all the regions in which they a centralized procurement system in most of its centralized its procurement in France, Carrefour has been to its procurement system in other countries. For example, in 2001 Carrefour a distribution in Paulo to three million with fifty to in the Southeast centralized its procurement systems in chains, such as China Resources of Hong stores in southern are also centralizing their procurement systems. is in retail in China and has large stores in the of and In of growth its million investment in China over the five a shift from procurement to a centralized system of procurement each is large distribution centers were in The distribution in is and be to stores and Second, there is a logistics improvements to procurement To some of the transport costs that with supermarket chains have (and that This that supermarket practices and which almost with the The of by supermarket chains and in Argentina the use of logistics by retail are in Asia. For example, a supply for in and production practices to supply and the efficiency of their chains in the three regions increasingly to a in the same as the supermarket logistics and wholesale distribution with other is the Carrefour distribution in Brazil, which is the of a of Carrefour with major and global of China in that it would a large distribution to be with and global distribution for fruits and vegetables in is in with of the Third, there is use of specialized The changes in logistics have supermarket chains new or the traditional wholesale system. The supermarkets are increasingly with specialized to and of their These specialized and and and on of the supermarkets. The and of the specialized has in of players and between the and the domestic food markets. Moreover, there is emerging that supermarket chains produce they to mainly via specialized For example, functions as the of most stores of the supermarket in Central America, as for in Africa. Fourth, the rise of and is new in one of the most markets in the food sector, the produce sector. as incentives to the to with the and over time investments in (such as and to the the products. The retailers are of and the of products with of processed and and and products under with the supermarkets. Supermarket chains have with processing firms, in turn with processed fruits and vegetables are under the for the supermarket in Costa Rica, and various firms produce under the products for the As retail sales of products to such are to increase in Latin America and Asia. food retailing in these regions in the with little use of and the emerging a rapid rise in the of in the supermarket sector (and other modern food sectors such as scale food and food The rise of for and of food products, and the of the of public is a crucial of the of in the procurement systems. In these as of coordination of supply chains by over many regions or countries. and the and efficiency and of a also be to a that the public are in all the markets in which the retail be as for or public (Reardon and In this can as competitive the sector (and other by The evolution of in the supermarket sector in these regions is also driven by between the by the in developed countries and in developing countries. many small and are it to the of supermarkets, and are being from their procurement The procurement practices of supermarkets and large are the of the for and first-stage To to advantage of and and and a of development for the small and sector. Development that mean programs and in be three or chains can up to 50% or more of the supermarket sector in development programs and policies to with just a of This is an and an and of and
Published in: American Journal of Agricultural Economics
Volume 85, Issue 5, pp. 1140-1146