Asian Journal of Agriculture and Development, Vol. 6, No. 2 51

The Transformation of Food Retail in the Philippines

Glory Dee Romo University of the Philippines Mindanao, Philippines Email: [email protected]

Larry Digal University of the Philippines Mindanao, Philippines Email: [email protected]

Thomas Reardon Michigan State University, USA Email: [email protected]

ABSTRACT

From a small base by the end of the 1990s, modern retail (the chains selling at least some food) in the Philippines grew very rapidly in the 2000s, at thrice the rate of the country’s GDP growth. Reaching 13 billion USD of overall sales by 2010, 5.25 billion USD of this amount came from food sales. While much of the policy debate about market development focuses on export markets, we will show that supermarkets in the Philippines already sell twice the value (volume priced at export prices) of food that is exported – and modern retail is growing faster. Modern retail has reached 45 percent of urban food retail – already beyond the share of the middle class in the population, and about 35 percent of the national food market – from a tiny fraction of that fi gure two decades ago.

Modern food retail is itself rapidly transforming – with a rise of fresh produce sales, industry concentration, format diversifi cation off-mall into formats that permit greater market penetration, and emerging procurement system modernization. While traditional food retail (and even most aspects of modern food retail) in the Philippines track international experience, the lack of major foreign presence even after retail foreign direct investment (FDI) liberalization in 2000 is a puzzle about which we present hypotheses. Given the already large and increasing importance of food retail modernization in the Philippines, this theme should enter the agrifood research mainstream and be the subject of systematic fi eld survey analysis in order to start discerning its impact on consumers, farmers, wholesalers, and processors. 52 Glory Dee Romo, Larry Digal and Thomas Reardon

THE TRANSFORMATION OF FOOD RETAIL wholesale, and a third in the retail segment. IN THE PHILIPPINES This implies that the latter is important to food costs and food security for the large share of Starting from a common traditional food the population that is urban (nearly half in the retailing system of small shops, wet markets, Philippines) and the many rural households and hawkers (as was similarly common that are net buyers (Balisacan 2000). Moreover, in the US, Western Europe, and Japan), a retail transformation appears to cause or at “supermarket revolution” took off in the early least encourage changes in upstream segments to mid-1990s in developing countries (Reardon (processing, wholesaling, and farming). et al. 2003; Reardon and Timmer 2007). The Such changes may be in sectoral structure, spread of supermarkets has taken place – and technologies, in institutions, or organization of continues to do so – in three waves. the food system. The fi rst wave countries experienced Despite the growing importance of the supermarket sector takeoff in the early to mid- modern food retail sector in the Philippines for 1990s. These include much of South America the overall agrifood economy [noted in various and East Asia outside China and Japan, self-published consulting reports, working paper Northern-Central Europe and the Baltics, bulletins, newspaper and magazine pieces, and and South Africa. These fi rst wave countries conference papers in the grey literature, such saw supermarket diffusion in a single decade as PDFI (1999), Digal and Concepcion (2004), that took some fi ve decades to happen in the Cabochan (2005), IBM (2007), Nielsen (2008), U.S. and the U.K. The second wave countries and Macabasco (2009) and others], there has include Mexico and much of Southeast Asia, been only one refereed academic journal article Central America, and Southern-Central Europe. on the subject – that of Digal (2001). This The third wave countries are those where the body of grey literature as well as the limited supermarket revolution did not take off until academic work on the subject has posited that the late 1990s or early 2000s; these areas modern retail has been growing quickly, with include parts of Eastern and Southern Africa, its take-off mainly in the late 1980s and early some countries in Central and South America, 1990s. It has also been observed that retail “transition East Asia” (China and Vietnam), formats have tended to change over time, Russia, and India. The modern retail in a subset and that procurement systems appear to be of the third-wave countries, especially China, modernizing with greater direct procurement Vietnam, Russia, and India, is growing very from processors and some emerging relations quickly, often at three to four times the rate of with specialized wholesalers, concessionaires, their rapidly growing GDPs per capita. and agribusiness companies as suppliers. The transformation of food retail Besides the dearth of treatment in the in developing countries is an important academic research literature, there are several development research topic in part because it important gaps in information in the existing treats change in roughly a third of the agrifood literature; namely: economy – as value chain research estimates 1) The journal article by Digal treats only that about a third of the value-added and the 1990s. We show in this paper that costs occur in the farm segment of the food while modern retail grew quickly from supply chain, a third in the processing and a small base in the 1980s and 1990s, Asian Journal of Agriculture and Development, Vol. 6, No. 2 53

it was very small by the end of the on upstream actors in the supply chain). We 1990s compared to what it has become shall address the following sets of issues: by 2010. Thus, the local academic First, and the main issue addressed by the literature does not capture the recent paper, is what does empirical evidence show decade of deep and rapid change. concerning the following: (a) growth of modern 2) Where it offers quantitative food retail sales including overall “banner sales”, assessments, the literature tends to be that is, food and non-food sales; (b) change in based on industry experts’ guesses of the product composition of sales over food the size of sales (such as that of fresh product categories with diversifi cation above produce), or the use of offi cial data that and beyond the traditional (for modern retail) does not allow breakdown by types category of ambient processed, into frozen and of formats or product categories of chilled (dairy, meat, fi sh), beverages, and fresh sales and are not easily available for fruits and vegetables; (c) change away from the each year. The literature also does not traditional base of department stores toward present statistics with cross-year data a format diversifi cation into hypermarkets, that show modern retail growth rates, supermarkets, and convenience stores; (d) sales by format, distinction of food spatial diffusion of modern retail away from sales of various categories compared to its initial base mainly in Metro Manila (MM) overall sales, and the gross productivity into other regions (i.e., into other provinces in of retail (in terms of sales per sq meter). Luzon, Visayas, Mindanao); and (e) change 3) Until now, literature in the Philippines in gross sales per square meter as a rough presents some company websites proxy for productivity. Note that based on the and key informant/case studies of international literature that we refer to when procurement system change, but presenting the results and recent-historical there is as yet no published study information from the Philippines, each of these about the procurement systems for fi ve axes represents an axis of modernization of either processed or fresh foods by a food retail, and an impetus for and step toward substantial cross section of modern further modern retail diffusion. retailers. Second, while we fi nd that the Philippine 4) There is as yet no published survey- situation and the modernization of retail in the based study of farmers, processors, country shares many characteristics with other or wholesalers involved in supplying developing countries, there are particularities modern retail compared with those that it has in common with relatively few other are not. developing countries that are also experiencing This paper seeks to make a contribution to retail transformation. Of interest is the the literature on modern retail in the Philippines continuing central role of domestic capital mainly by presenting new data addressing the in the transformation, while elsewhere retail fi rst two sets of gaps noted above (growth and FDI has played a far larger fomenting role. We composition of modern retail), and briefl y explore the issue of the determinants of retail reviewing existing case studies and hypothesis- transformation that the country shares with presenting literature on the third and fourth most other developing countries, which ones gaps (procurement system change and impacts are unique to the Philippines, and why. 54 Glory Dee Romo, Larry Digal and Thomas Reardon

Third, and treated most briefl y in the paper the third way because the small individual due to scant empirical evidence, we review the shops in the traditional sector may be formally (mainly gray) literature on the modernization registered small grocers, or be informal shops of the Philippines’ procurement systems of or stalls, or mobile carts or hawkers. Moreover, processed and fresh foods which may have modern retail may also be small stores such as impacts on processors, wholesalers, and a chain of convenience stores. Below then we farmers – the impacts of which have not yet defi ne and justify the set of segments we range been published in the Philippines. These points under traditional and modern. are stated more as hypotheses to guide much- needed empirical research. The Traditional Retail Segment Our purpose is to present a substantial new set of empirical fi ndings, in particular on the The traditional retail system in the fi rst set of issues above; and for the other two Philippines is very similar to what is found sets of issues, to assess the state of knowledge in other developing countries or in developed and draw implications from these in terms of countries “historically” (before the advent of agrifood economic research gaps and agenda. modern retail, hence in the early 20th century). To this end, we draw on data from three types The student of retailing would note nothing of secondary sources. The fi rst and main source unique or specifi c to the Philippines in terms we use is a proprietary data set from the UK- of its small scale, types, formats, and spatial based PlanetRetail, one of the leading statistical density; in fact, the description below of the services on retail in the world, providing detailed types and characteristics of each one could sales data on the leading modern retailers over easily be written for Turkey, Peru, Indonesia, the past decade. These data have not been India, or the US, historically. The traditional published for the Philippines. The second actors are as follows: source is a systematic review of secondary 1) The wet market is a set of stalls data and literature from government or offi cial selling fruits and vegetables, either sources and academia, and third, statistics and grouped outside or under one roof, case studies from major consulting services. either periodic (called a talipapa) or The paper proceeds as follows: Section 2 permanent (called a palengke). This briefl y discusses the categories and defi nitions may be stand-alone or appended to of traditional and modern retail in the a wholesale market. It may operate Philippines; Section 3 presents new data on daily at fi xed hours in urban areas, or trends in growth and nature of modern retail weekly on particular days. It can also over the past decade; Section 4 surmises the have fi sh, meat, or poultry depending determinants of the patterns found in Section 3; on the region, city, or neighborhood. and Section 5 concludes. As everywhere in the world, prices in the wet market are not fi xed and CATEGORIES OF FOOD RETAIL bargaining is the norm. Shopping is IN THE PHILIPPINES vendor-administered, not self-service. 2) Small shops, called sari-sari, are like There are at least three ways of categorizing the “mom and pop” stores in traditional retailers: informal versus formal, small versus retail systems in other countries. large, and traditional versus modern. We choose They are typically from tiny to small/ Asian Journal of Agriculture and Development, Vol. 6, No. 2 55

medium scale, single-owner, often run The Modern Retail Segments by a husband and wife and perhaps a family member or employee, not self- The defi nition of modern retail service, and typically carry either dry and transformation of the retail sector goods/foods such as grains (rice and wheat fl our), packaged foods, and non- The retail literature does not have hard- food FMCG (fast-moving consumer and-fast rules as to the defi nition of modern goods such as detergents) or wet goods retail. From common usage in the literature, we (produce, poultry, or meat). assemble the following key characteristics and 5) Sari-sari stores may deliver to homes, some of their ambiguities. and may offer credit to some of their First, an early (historically in the literature) regular clientele (suki), although recent dividing line between traditional and modern studies in other countries show that retail is the use by the latter of “self-service.” small shops in urban areas provide Second, a certain scale of operation credit only to a few select clients, characterizes modern retail. In popular despite the persistent image that they discussion of modern retail in developing sell a lot on credit and that this is a countries, it is often assumed that a modern persisting advantage for them (Minten retailer is a large store – as in a supermarket et al. 2010). Minten wrote a case on or hypermarket; however, from a retail Delhi; there is no recent published test research perspective such as we use, this is of this hypothesis in the Philippines. not so. Modern retail is composed of larger- The only study we found on this is by than-traditional independent stores e.g., small Dannhaueser (1977) for Dagupan City, supermarkets, and chains. While the scale can showing even then that only a minority be in the store (such as hypermarket), it can also of small shops extended credit to a very be in the chain (and hence the aggregate volume few of their customers. of the enterprise) – thus a chain of small (even 3) Various mobile small retailers, such very small) stores, where all the stores (of a as push carts (kariton) and hawkers given format) follow roughly a similar pattern (maglalako) who peddle their wares of retailing and procurement, is invariably on foot. The push cart would typically classifi ed as modern retail. have a perishable product, such as fresh Third, it is often assumed that a modern produce or dairy products in limited retailer has a large assortment. But a modern assortment, and deliver to homes or retailer can either have a broad or narrow station at a particular point and sell selection. For example, a convenience store or at certain hours. There is sometimes a single category store – like 7-11 or the bakery a small cluster of push carts selling chain of Jollibee’s, has a more narrow selection different specialties. than a supermarket – while a large supermarket/ 4) Medium-sized stand-alone shops hypermarket like SM has a wide variety of or traditional groceries are another thousands of types of products. element of what we include here in Fourth, there is often an assumption “traditional retail.” that modern retailers have “modernized” 56 Glory Dee Romo, Larry Digal and Thomas Reardon

procurement systems – characterized by buying In sum, at odds with popular discussions, a in bulk, buying direct from producers, and using formal retail research perspective on the criteria sophisticated inventory management practices. for modern retail reveals that there are fairly However, again, the popular assumption is simple and fl exible basic criteria – that of a not the necessary criterion for modern retail. minimum scale either of an independent store Independent supermarkets or chain stores of or a chain of stores of any scale per outlet, plus any outlet scale have procurement systems self-service. That then leaves for a given modern which fi gure on a spectrum that goes from one retail outlet or chain a wide variety of possible extreme, “traditional procurement systems” practices (from “similar-to-traditional shops” (buying from the spot market or a traditional to “cutting-edge modern practices”) in retail wholesaler off-market, and having delivery environment and sourcing, and from small to made to each store), to the other end, “modern large in store size, and product assortment. That procurement system” (buying directly from breadth of defi nition is useful because of the processors or farmers, or from specialized/ great variety of forms, shapes, and approaches dedicated wholesalers, under contract, using that modern retail takes in the developed and private standards, and having delivery made developing regions in general – and, as we show to centralized distribution centers or DCs), or below, in the Philippines. some system which falls anywhere in between Our criteria for modern retail are close those extremes (Reardon et al. 2003). to, but are different in some ways from recent Fifth, is the popular assumption that modern mainstream discussion on this theme. For retail has a certain “environment” that differs example, IBM (2007) defi nes the term as sharply from traditional retail – that modern stores with self-service, in chains, and owned stores have air-conditioning, electronic check- by corporate entities. They leave out the large out counters, credit card systems, packaging independent supermarkets with one or a few produce, and other such amenities. But again, stores or several malls in a mall company that there is nothing inherent in modern retail about has supermarkets within the malls or shopping the ambience. For example, formats targeting centers, that can be the “nodes” of development the lower income segments may adopt a “no of chains as we have in the Philippines and frills” ambience – for the appearance or for elsewhere, and in fact constitute an important reality or both – of having lower costs and thus part of the supermarket chains in the small and lower prices; they may eschew air conditioning, medium sized chain association, Philippine packaging of produce, and instead stack discount Amalgamated Supermarkets Association (PAG- packs and bulk produce on crude shelves. ASA). Sixth, we include somewhat incongruously At least in theory, the shift from the traditional “cash & carry” stores into our analysis of modern to the modern retail form is a transformation retail. In developing countries in general, and as that increases effi ciency in the retail segment practiced in the Philippines in particular, these via economies of scale and scope; control over stores are a mix of wholesale (to small retail supply chain processes in order to reduce waste shops and hotel, restaurants and catering or and do away with the many middlemen found HORECA clients) and retail to families who in traditional supply chains; and to have the want to buy in bulk. scale and resources to invest in capital-intensive inventory management technologies. Asian Journal of Agriculture and Development, Vol. 6, No. 2 57

The evolution of modern retail formats 1) Small chains of small scale supermarket in the Philippines outlets added stores in established cities and moved to new cities – generally on the First, “traditional” modern retail arose same island. in the Philippines in the 1960s-1970s, and 2) Large and medium domestic companies such mirrors similar types of retail emerging in as SM, Robinson’s, Rustan’s, Gaisano, and other developing countries at about the same Benison Waltermart built many new malls period. This early foundation had two pillars. and developed medium-scale supermarkets On the one hand, mainly commercial families and hypermarkets to anchor them, as started small chains of small supermarkets. An well as maintaining department stores as example is Rustan’s Supermarket, founded in anchors. Some of these companies, like 1970. On the other hand, department stores, and SM Investments, became conglomerates then department stores as anchor stores in malls, with large operations in real estate and began to proliferate in the 1980s. The largest of banking as well. The mall operations these companies, SM, started as a single shoe themselves were part of real estate activity store that opened in 1958. Then it transitioned to as they are essentially an infrastructure/ a chain in the 1960s, added clothing and house services-augmented land-rental operation; and offi ce wares to be a department store in the for example, the 40 SM malls have 12,000 1970s, then into a mall with an SM Department retail tenants, all renting. Store as anchor in 1985. In the 1990s, a number 3) Some of the above medium and large of other mall companies with department stores companies have begun rapidly rolling out as anchors, and also transitioning into having stand-alone (i.e., not part of malls) small supermarkets as anchors entered the scene – to medium supermarkets and hypermarkets such as Robinson’s, Gaisano, and others. to increase spatial penetration, such as Second, the current confi guration of Savemore of SM. modern retail in the country arose after the 4) Foreign cash & carry chain SHV Makro of fi nancial crisis in the late 1990s and then the Netherlands, with 7.2 billion USD of developed very quickly in the 2000s, as we banner sales globally, and the warehouse- show below. There was a sudden and extensive club chain PriceSmart of the US, with 1.3 proliferation of formats designed to greatly billion USD of banner sales, entered the increase and accelerate penetration of the Philippines in the early 2000s after the various market segments, from lower income partial liberalization of FDI. Both were to middle and upper income, from suburbs to later bought by domestic leaders SM and dense inner cities, from “destination” shopping Puregold respectively. Note that these to near-residence shopping, and from diverse chains are relatively small; compare them to inventories to focused offerings. Here we list leading global chains’ warehouse and cash the formats as part of our defi nition of modern & carry format sales: Metro C&C (with 47 retail, and then in the subsequent section trace billion USD of global sales) and Wal-mart’s their growth with data. The formats and their formats of this type (with 8.8 billion USD roll-out are as follows: of sales). 58 Glory Dee Romo, Larry Digal and Thomas Reardon

5) Domestic cash & carry chains include develop; these are larger than convenience the leading domestic player, Suy Sing stores but smaller than supermarkets (IBM, Commercial, with four large distribution 2007), designed to penetrate dense cities and centers acting as cash & carry outlets. also rural towns as they do not need large initial 6) Domestic warehouse/club chains such as customer bases. A similar development of “hard Puregold – which arose from a duty-free discount stores” will probably occur, like small chain – emerged in the early to mid-2000s. supermarkets but focused on a narrow range of Older and smaller chains like Uniwide items with steep discounts, to penetrate poor compete with them. areas. There may also be small format fruit 7) Foreign convenience stores (chains and vegetable chains such as those developing belonging to large companies) entered – in China and Indonesia, to compete more most notably President Chain Store/7-11 forcefully with wet markets. (Taiwan-based, with operations in Taiwan, The trend we identify below in the increase China, Philippines, and Vietnam) and in hypermarkets will likely continue and even Ministop (the convenience store chain of increase, as medium-sized hypermarkets have the giant mainly Asian-region multinational been used as destination or entertainment hubs AEON based in Japan). in various parts of Asia, serving in a sense as 8) Large foreign drug/personal care store mini-malls that are easier to establish where chain AS Watson (Hong Kong) entered in real estate markets are tight. a joint venture with SM. Other non-grocery chains include the Body Shop and foreign GROWTH AND CHANGE IN MODERN FOOD nutrition chains like GNC. Large domestic RETAIL OVER THE 2000s chains like Mercury Drug compete with them. Both sell grocery, including some Modern food retail has grown and changed food. in nature quickly over the past decade in the 9) Foreign chains of forecourt stores (usually Philippines. We present data on these changes attached to gas stations) such as Shell and in the following fi ve subsections. The fi rst Chevron entered. discusses our database; the second, the growth 10) Bakery chain stores also emerged as a in “banner sales” (food plus nonfood sales); the format of Jollibee, the fast food chain. third, the growth and change in food sales and their categories, as part of grocery sales; the Third, it is likely that there will be fourth, sales by format over time; and fi nally, continued development of several formats that the spatial and socioeconomic diffusion path of will maintain and possibly increase the pace modern retail. of store penetration into cities and towns, and increasingly into rural areas. The Database There will probably be more off-mall expansion of small format stores, given similar Tables 1 to 4 are based on proprietary data trends in other Asian countries. At present, obtained from the PlanetRetail website at current the main format is the convenience store, but date (www.planetretail.net) which tracks sales, following developments in other countries in formats, store numbers, and various business the region, “neighborhood” format stores may operations of retail companies in a number of Asian Journal of Agriculture and Development, Vol. 6, No. 2 59

countries. PlanetRetail is one of the globally 53 member companies which collectively own most popular and respected retail data fi rms. We 148 outlets, 20 warehouses, 1 hypermarket, acquired, selected and processed the raw data. 11 supermarkets in malls, 35 supermarkets The data in Tables 1a to 4 show banner sales of in commercial complexes, 63 stand-alone all products, food and nonfood; grocery sales supermarkets, 19 convenience stores, and one which include food and fast moving non-food wholesaler. items such as detergents; food, which includes The second missing set are the chains and ambient processed products like polished rice, independents in the Philippines Association of noodles, and so on; frozen products like ice Supermarkets, Inc. (PASI), formed in 1969. cream; chilled products like milk, meat, and PASI has 40 member companies with 555 small fi sh; alcoholic and non-alcoholic beverages; and medium supermarket outlets, of which 68 and fresh fruits and vegetables. We show data are within Metro Manila (MM), 463 outside for 1999 to 2010 – the latter being estimated by MM but in Luzon, 20 are in the Visayas, and 4 PlanetRetail from stores and sales information are in Mindanao (Macabasco 2009, citing PASI and tends to track well the realized sales. data). Of the outlets, 184 are not included either The set of companies in Tables 1-4 have in the PlanetRetail list (which shares President all been followed by PlanetRetail for the Chain/Philippine Seven and Waltermart with the Philippines, with the following ranking of PASI list) or the mall list below (where Metro banner sales in 2010; PlanetRetail estimates Gaisano is included). this for the full year from the fi rst quarter The third missing set of chains or stores information compared to the year before: (1) not followed by PlanetRetail include a number SM Investments, (2) Puregold, (3) Robinson’s, of the supermarket chains or independents that (4) Rustan’s, (5) Mercury Drug, (6) AS Watson, operate as “department store-cum-supermarket (7) Benison (Waltermart), (8) President Chain anchor or supermarket anchor” in malls. The (7-11), (9) SHV Makro, (10) Jollibee Bakeries, great majority of malls have as an anchor (11) Uniwide warehouse chain, (12) Pricesmart a department store with a food fl oor and/or club/warehouse chain, (13) a set of “forecourt” supermarket or hypermarket that sells food. We chains (PTT, Chevron, Total, Shell), as well as assembled a list of 103; we assume there is one a nutrition chain (GNC). supermarket per mall, hence 103 supermarkets. The above 13 represent the largest chains, in Our second list includes only mall companies, particular with Luzon as their central location. of which there are 41; we make the conservative The data shown for them are, however, an assumption that there are two supermarkets per under-estimate of modern food retail in the mall company, thus 82 supermarkets. Hence Philippines. Several sets of retailers are missing for the mall company list, we have a total of from the above. 185 small/medium supermarkets. Note that The fi rst missing set are the chains and the fi gures above do not include those that independents in the Philippine Amalgamated are already listed in our tables (hence covered Supermarket Association, Inc. (PAGASA), in the PlanetRetail list which includes SM, formed in 1986. PAGASA notes that its member Robinson’s, and Waltermart). chains tend to be small ones, and focus on the From the above three sets, we derive what B, C, and D consumer segments or the lower we think is a conservative estimate of 517 middle and poorer groups. The association has modern retail company outlets selling food that 60 Glory Dee Romo, Larry Digal and Thomas Reardon

are not covered by PlanetRetail. Most of the retail vis-a-vis total retail because of the large outlets in malls and on the PASI and PAGASA amount of informal sector retail. Interestingly, lists are of the size of supermarkets on average. the NSO data and our estimates are close. From PlanetRetail we derive the sales in 2010 Using the adjustment factor of 33 percent to the of a typical supermarket of Robinson’s, a fairly PlanetRetail data, and noting that its modern small supermarket with 1,500 sq meters of sales retail fi gure for banner sales is 3,786 million space and 6.3 million USD per year of sales. USD in 2005, we get 5,035 million USD – below Based on this, 517 (retail outlets) times 6.3 the NSO’s 5,251 fi gure by only 4 percent. It is million (annual sales) is 3,257 million USD. reassuring that the aggregate of the data set that From Table 1 we note that the total banner sales we used for the detailed composition discussion of the 13 companies followed by PlanetRetail checks closely against the broad census-based in 2010 is 9,663 million USD. Summing 3,257 offi cial fi gures. million and 9,663 million, we get a total of Table 1b shows the levels of sales and growth 12,920 million USD. So PlanetRetail data of modern retail in the Philippines. The rows need to be adjusted upward by 33 percent (a show the company (but not the brands, which conservative estimate that misses stand-alone are the names of the sub-chains with particular independents and smaller chains not part of the formats, such as 7-11 or SM Supermarkets), three sets above) to get the “correct” number and the nationality of the capital (and whether – PlanetRetail’s total plus our estimate of the or not it is in a joint venture with a foreign or missing. We return to these aggregate fi gures Philippine company). below when discussing Table 1. The columns show, for 1999 to 2010 (with Moreover, note that Tables 1-4 also show the the latter an estimate by PlanetRetail), the data from chains that started and stopped (such banner sales (food plus nonfood). The sales as Makro and Pricesmart) that are followed by fi gures are in millions of USD in nominal terms; PlanetRetail, and then their sales after being under those levels are percentages, which depict acquired by other companies. Moreover, for a year-on-year growth relative to the year before; given company, we present data only from the the latter rates are presented in simple average sub-chains that sell some or mostly groceries. in the last column. The salient results are as We likewise excluded non-retail operations follows: such as real estate or fast food chains. First, modern retail sales grew very quickly in the Philippines over the 2000s. The Growth of Modern Retail in the Philippines average year-on-year increase was 26 percent – in the 2000s compared with the average year-on-year growth of the GDP of the Philippines of roughly 9 Table 1a shows the government’s National percent from 1999-2008, both in current prices. Statistics Offi ce (NSO 2005) data for formal Seen another way, GDP increased 2.12-fold sector fi rms; it does not include informal sector for the period indicated, while modern retail sari-sari stores or stalls in palengke. The table banner sales for these fi rms increased 7.1-fold. compares retail fi rms with 20 employees or This suggests that modern retail gained share, more (large scale), versus those with less than displacing traditional retail over the decade. 20 (small scale). Note that these data cannot These modern retail growth rates are similar to be used to estimate the share of large-scale other “second wave” countries in the Southeast Asian Journal of Agriculture and Development, Vol. 6, No. 2 61 less ATE of ATE than 20 rms fi No. of 20 or more ATE of ATE less ATE of ATE than 20 (million USD) Total Revenue Total 20 or more ATE of ATE less ATE of ATE than 20 rms fi No. of 20 or more ATE of ATE less ATE of ATE than 20 rms (2001, 2003, 2005) (million USD) fi Total Revenue Total 20 or more ATE of ATE 20 less than ATE of ATE rms fi No. of 20 or more ATE of ATE 2001 2003 2005 less 20 or ATE of ATE 152 818 112 50,617 122 753 95 50,751 184 1,041 158 52,388 (million USD) Total Revenue Total 20 or more 3,373 1,620 668 76,378 4,094 1,960 637 76,262 5,067 1,745 934 68,147 ATE of ATE : NSO (2005) Non- specialized stores retail trade in stores (a) Retail sale of food, beverages & tobacco in specialized stores (b) (a)+(b) 3,525 2,437 780 26,995 4,217 2,713 732 127,013 5,251 2,786 1,092 120,535 Industry Description Table 1a. Total revenue and number of formal sector retail 1a. Total Table of less than 20 - small players ATE of 20 or more - big players; ATE Employment; Total Average - ATE June 2009) 47.53 (exchange rate as of 11 1 USD = PhP Source 62 Glory Dee Romo, Larry Digal and Thomas Reardon 9% Grth Ave. Ave. 26% 19% 21% Ann. 21 10% 486 20% 156 0 269 17% 76.2 0.932 5 57% 20 71 11% 608 233 13% 104 255 203 0 75.9 (0%) (-5%) (-5%) 1.499 (25%) (61%) (-33%) 9 11 113% 19 61 724 231 105 297 281 13% 192 71.2 (0%) (1%) (-16) (-6%) (-5%) (-5%) 1.930 (19%) (29%) (10%) (100%) 5 39 18 62 901 253 137 414 102 308 193 76.8 (na ) (7%) (2%) (0%) (-5%) 2.432 (24%) (10%) (31%) (39%) (26%) (10%) (100%) (254%) 9 72 19 63 499 272 126 147 315 188 79.6 (4%) (8%) (7%) (2%) (6%) (2%) 1068 (-3%) 2.778 (19%) (21%) (24%) (14%) (85%) (100%) to year-on-year growth. 72 73 61% 73 20 78 10 511 276 154 149 334 214 86.9 (na ) (9%) (2%) (2%) (1%) (0%) (5%) (6%) 1239 3.203 (16%) (22%) (15%) (24%) (14%) (10%) 96 23 86 39 236 144 669 104 399 295 272 98.8 (7%) 1423 (-3%) 3.786 (14%) (15%) (53%) (18%) (32%) (31%) (42%) (15%) (20%) (10%) (27%) (290%) 27 96 48 118 263 121 169 807 145 489 331 253 1739 (-7%) 4.488 (11%) (19%) (22%) (19%) (76%) (21%) (39%) (17%) (23%) (12%) (12%) (23%) (-16%) 35 60 117 117 299 144 323 950 183 651 381 120 151 2412 (-3%) 5.682 (14%) (22%) (91%) (18%) (39%) (26%) (27%) (30%) (33%) (15%) (25%) (25%) (-40%) 41 66 330 122 532 202 522 731 140 (4%) 1126 2784 6.596 (10%) (16%) (65%) (19%) (15%) (10%) (16%) (17%) (37%) (12%) (17%) (10%) 45 68 114 114 353 724 193 584 748 148 (7%) (2%) (6%) (3%) 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1238 3377 (-7%) (-4%) 7.592 (36%) (10%) (12%) (10%) (21%) (15%) 55 388 183 (4%) 2010 1537 4307 (10%) (64%) (16%) (22%) (16%) (13%) (24%) (28%) (24%) (40%) Estim. (107%) : Authors’ calculations from raw data from PlanetRetail. GDP from World Bank (2010). from World calculations from raw data PlanetRetail. GDP Authors’ : : Sales are only by formats that sell some grocery; banner sales all food and non-food sales. Figures in parentheses refer SM (Philippine, US, French, and Dutch/UK)) Watson) since 2002 and Netherlands (Makro) chain Watson) since 2007 subsidiary), majority in Oct 2000 2008-2010 owned/operated by SM 6. AS Watson (Hong Kong) 60/40 JV with AS Watson and AS Watson (Hong Kong) 60/40 JV with AS Watson 6. GDP in current prices billions of USDGDP 167 11. Uniwide Warehouse stores (Phil.) Uniwide Warehouse 11. 118 2. Puregold (Phil.), warehouse/retail/wholesale 1187 Company, main brand, and nationality of capital Company, 7. Benison (Waltermart) 7. Benison (Waltermart) 224 12. Pricesmart (US) 13. PTT, Chevron, GNC, Total, Shell grocery forecourts Total, Chevron, GNC, 13. PTT, 5. Mercury Drug (Phil.) 678 4. Rustan’s (Phil.) 4. Rustan’s 845 3. Robinson’s (Phil.), JV with Japanese chain (Ministop) 3. Robinson’s Aeon chain since 2000 a division of 1.SM Investments (Phil.), JVs with Hong Kong (AS 8. President Chain (7-11) (Taiwan via Malaysian (Taiwan 8. President Chain (7-11) Total banner sales in current prices billions of USDTotal 9.663 9. SHV Makro (C&C) (Netherlands via Asian subsidiary), 9. SHV Makro (C&C) (Netherlands via 10. Jollibee Bakery (Phil.), retail 141 Sources Table 1b. Growth of modern retail in the Philippines (1999-2009), millions USD Table NOTE Asian Journal of Agriculture and Development, Vol. 6, No. 2 63

Asian region such as Thailand, Indonesia, in 2000 to a projected 1.54 billion by 2010 (6- and Malaysia, but slower than the third wave fold); those of Puregold from a mere 73 million countries such as China. in 2004 to a stunning 1.19 billion by 2010 (16- Second, there is evidence of some pro- fold); and SM’s from 608 million in 2000 to an cyclicality of modern retail growth, with estimated 4.3 billion USD in 2010 (7-fold). In spikes in growth after recovery from the Asian contrast, the lower-ranked fi rms like Uniwide economic crisis in the late 1990s and again barely grew in sales over the decade. after the fi nancial crisis of 2008-09. This is Fourth, compared to some third wave corroborated in the trade press concerning countries, the share of pure cash & carry sales and growth expectations of the retailers. wholesale chains is relatively low. This may be However, while that holds for the whole group due to several factors. A number of the chains of chains, the leading companies such as SM such as Puregold are in warehouse format and and Puregold had banner sales growing much serve both retail and consumer clients. Several quicker than both the GDP and the lesser rivals, fi rms that are named cash & carry also sell even during the recessions. This suggests they retail. Finally, there are large wholesale players were gaining market share both from modern with multiple distribution centers that act as a rivals and the traditional retailers. chain of modern cash & carry (exemplifi ed by Third, there was a strong U-shaped pattern Suy Sing Commercial, www.suysing.com, the in trends in concentration over the decade. In leading grocery distribution company that arose 2000, the C3 or the share of the top 3 companies from the main wholesale market, Divisoria). was 73 percent (very concentrated); this dipped Fifth, there are no major global chains to 58 percent by 2005 as growth in the market (Carrefour, Wal-mart, Tesco, Metro, etc.) and liberalization of investment rules brought among the top four. In the next section we in other players. However, by 2010 the sector present hypotheses to explain this. This had reconcentrated to a C3 of 75 percent – phenomenon is strikingly unlike similar retail essentially at European and Latin American contexts elsewhere in Asia, such as three of levels of retail concentration. This concentration the top four in Malaysia, two of the top four in is most likely to be somewhat less at a national Taiwan, Thailand, and Indonesia, and one of the level as some of the fi rms among those not top four in South Korea. However, during the followed by PlanetRetail were growing quickly, 2000s a number of foreign retailers of groceries mainly off-Luzon. did enter the Philippines, solo or in joint venture This trend of reconcentration appears to (JV) with national fi rms1. Aside from 7-11, one be due to the two foreign fi rms Makro and could say that the foreign chains that entered Pricesmart rising in the fi rst half of the decade and stayed as JVs with large local players (AS and then being acquired by SM and Puregold Watson, Ministop, Makro), or came and then respectively. Add to this the very fast growth were acquired fully (Pricesmart) are either by the three leaders – with Robinson’s food regional (not global) multinationals or second- formats’ sales growing from just 255 million tier global chains. Other countries in which

1 Notable are Hong Kong’s AS Watson in JV with SM; Japan’s Ministop as a JV format of Robinson’s; Taiwan’s President Chain (7-11); Netherlands’ SHV Makro (later bought by SM) and U.S.’s Pricesmart (later bought by Puregold), as well as foreign forecourt chains Chevron (US), Total (France), Shell (Dutch-English), PTT (Thai)) and nutrition chains (GNC of the US). 64 Glory Dee Romo, Larry Digal and Thomas Reardon ions of USD ient temperature), alcoholic and non-alcoholic drinks, tobacco, pet 64% 52% 65% 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 2010 Estimate 486 [58%] 426 [57%] 412 365 295 236 [59%] 195 184 177 158 [56%] 761 [50%] 595 [48%] 514 413 346 292 [44%] 195 187 144 97 84 [33%] 88 cation of items sold fi [share of banner sales] 2399 [56%] 1800 [53%] 1501 1168 791 619 [44%] [share of banner sales] 524 437 672 [57%] 349 410 [57%] 301 296 183 248 [41%] 96 198 55 [57%] 41 * * 4. Rustan’s 4. Rustan’s a) Grocery [share of banner sales] Ambient b) c) Chilled and frozen 99 (23%) 138 (32%) 87 (23%) 120 (31%) 49 (23%) 65 (30%) 35 (24%) 42 (28%) b) Ambient (non-chilled processed food) b) c) Chilled and frozen 665 (31%) d) Beverages (alcoholic and non-alcoholic) 492 (31%) 248 (12%) e) Fruits and vegetables 181 (11%) f) Share of food in grocery (USD)2. Puregold a) Grocery 410 (19%)Ambient (non-chilled processed food) b) 72% [1727] 302 (19%) c) Chilled and frozen 207 (10%) 239 (40%) 150 (27%) 71% d) Beverages (alcoholic and non-alcoholic) 154 (10%) 146 (40%) 75 (13%) e) Fruits and vegetables 56 (10%) f) Share of food in grocery [USD] 46 (13%) 3. Robinson’s a) Grocery [share of banner sales] 97 (16%) 78% [525] 94 (17%) 56 (9%) 59 (16%) 47 (8.5%) 15 (41%) 78% 62.5% 34 (9%) 5 (14%) 54 (21%) 20 (9%) 6 (16%) 80% 3.4 (9%) 32 (14%) 16 (7%) 51% [172] b) Ambient (non-chilled processed food) b) c) Chilled and frozen 182 (26%) d) Beverages (alcoholic and non-alcoholic) 141 (25%) 90 (13%) e) Fruits and vegetables f) Share of food in grocery [USD] 68 (12%) 141 (20%) 65% [571] 108 (20%) 66 (9%) 67 (25%) 51 (9%) 29 (11%) 50 (19%) 25 (9%) 17 (22%) 4 (5%) 12 (16%) 7 (9%) 1. SM a) Grocery Company, classi Company, Table 2. Grocery share of banner sales and food grocery modern retail in the Philippines (1999-2009), mill Table * Grocery includes fruits and vegetables, chilled frozen foods, ambient products (processed foods in packages stored at amb care, household and health beauty care products. Asian Journal of Agriculture and Development, Vol. 6, No. 2 65 0 90 [31%] 85 83 77 70 71 [31%] -- 30 (14%) 22 (15%) (continued) 51 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 (13%) 00 00 2010 6 (3%) 6 (3%) 0 0 23 (9%) 21 (9%) 14 (9%) 6 (9%) 17 (8%) 14 (8%)13 (5%) 12 (5%) 7 (9%) 8 (5%) 6 (8%) 4 (5%) 14% [38] 14% 14% 14% Estimate 51 (33%)34 (22%) 44 (31%)20 (13%) 31 (22%)18 (12%) 17 (13%) 15 (12%) 24 (34%) 16 (23%) 9 (13%) 8 (13%) 47 (23%) 40 (23%) 8 (9%) 7 (8%) 44 (10%)58 (13%) 38 (10%) 20 (9%) 12 (8%) 34 (16%) 29 (16%) 15 (17%) 12 (17%) 80% [134] 79% 83% 78% [379] 77%50% [104] 50% 76% 35% 75% 33% cation of items sold fi 7. Benison/Waltermart a) Grocery [share of banner sales] 167 [75%] 144 [75%] 151 136 108 77 [74%] 54 5. Mercury Drug a) Grocery [share of banner sales] 207 [31%] 178 [31%] 1596. AS Watson 116a) Grocery [share of banner sales] 101 271 [70%] 246 [70%] 230 209 183 164 [70%] 107 88 71 [70%] c) Chilled and frozen d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] b) Ambient (non-chilled processed food) b) f) Share of food in grocery [USD] c) Chilled and frozen b) Ambient (non-chilled processed food) b) f) Share of food in grocery [USD] Ambient (non-chilled processed food) b) c) Chilled and frozen d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables e) Fruits and vegetables Company, classi Company, d) Beverages (alcoholic and non-alcoholic) Table 2. Grocery share of banner sales and food grocery sales... Table 66 Glory Dee Romo, Larry Digal and Thomas Reardon (continued) 79% 79% 79% 11 (10%)11 19 (10%) 14 (10%) 38 (36%)20 (19%) 68 (36%)15 (14%) 36 (19%) 26 (14%) 51 (36%) 27 (19%) 20 (14%) 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 2010 1 (1%) 0.7 (1%) 0.4 (1%) 0.4 (1%) 8 (13%)6 (10%) 7 (13%) 6 (10%) 9 (13%) 7 (10%) 7 (13%) 5 (10%) 80% [54] 80% 80% 80% 62% [90] 64% 65% 64% 11 (10%)11 9 (10%) 5 (10%) 4 (10%) 24 (40%) 23 (40%) 29 (40%) 21 (40%) 12 (10%)44 (41%) 10 (12%) 36 (41%) 6 (12%) 21 (42%) 5 (12%) 17 (41%) Estimate 10/60 (17%) 10 (17%) 12 (17%) 9 (17%) cation of items sold fi 9. Makro (cash & carry) a) Grocery [share of banner sales]10. Jollibee Bakery-retail a) Grocery [share of banner sales] (food) Uniwide warehouse 11. 141 [100%]a) Grocery [share of banner sales] 68 108 [72%] 67 [57%] 66 181 64 [56%] 195 [72%] 69 60 153 134 48 66 138 137 69 39 145 [71%] 82 [57%] 10 84 9 83 5 78 60 9 59 [57%] 5 91 8. President Chain (7-11) a) Grocery [share of banner sales] 145 [79%] [80%] 118 111 96 77 69 [80%] 62 50 49 48 56 [79%] c) Chilled and frozen d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] c) Chilled and frozen d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] b) Ambient (non-chilled processed food) b) Ambient (non-chilled processed food) b) b) Ambient (non-chilled processed food) b) c) Chilled and frozen d) Beverages (alcoholic and non-alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] Company, classi Company, Table 2. Grocery share of banner sales and food grocery sales... Table Asian Journal of Agriculture and Development, Vol. 6, No. 2 67 81% 3 (8%) 6 (17%) 5 (14%) 41 [57%] 41 22 6 [55%] 15 (42%) (continued) 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 418 148.4 70.4 65%26%10% 64% 28% 10% 59% 25% 10% 2010 Estimate 71% [3805] 71% [1274] 66% [500] cation of items sold fi : Sales are only by formats that sell some grocery; food does not include tobacco : Authors’ calculations from raw data PlanetRetail Authors’ : 13. 5 Forecourt chains Grocery [share of banner sales] (food) Simple ave share of grocery in banner sales, selected years 43 [78%]Simple ave share of food/grocery in 37 [82%] millions of USD Simple ave share of ambient in grocery, 34 selected years Simple ave share of fruits and vegetable 28 in grocery (for chains carrying them) Fruits and vegetables sold 23 in millions of USD 18 [85%] 17 15 15 14 12 [62%) 13 12. Pricesmart a) Grocery [share of banner sales] b) Ambient (non-chilled processed food) b) c) Chilled and frozen d) Beverages (alcoholic and non- alcoholic) e) Fruits and vegetables f) Share of food in grocery [USD] Company, classi Company, Table 2. Grocery share of banner sales and food grocery sales... Table NOTES Figures enclosed in [brackets] indicate share of banner sales Figures enclosed in (parentheses) indicate share of grocery sales Source 68 Glory Dee Romo, Larry Digal and Thomas Reardon

domestic retail has maintained the great majority further below. This can be seen as part of the of modern retail include India (where retail FDI general process of modernization of food retail has not yet been liberalized), South Korea (after in the Philippines – where food retail moves the rise to near hegemony of large domestic out of being a mere “fl oor” in a department fi rms that became multinationals), and Chile (in store (around the world a typical “early stage” an earlier stage when major local players had of modern food retailing), into supermarkets, dominated real estate locations, pushed out, hypermarkets, and convenience stores. At the prevented, or bought out multinationals before same time, non-food single branded stores like these large domestic fi rms were later acquired Marks & Spencer or the Gap arise and eat into by multinationals). non-food share. Moreover, the shift in relative importance of grocery that follows the usual Product Category Composition (international) product market penetration of Food Sales by Modern Retail stages of modern retail – from nonfood, to ambient processed products, to chilled/frozen, Table 2 shows the shares of grocery sales in to fresh produce – has been occurring in the total banner sales, the share of food in groceries, Philippines. and the shares in food of its component Second, the food share in grocery rose on categories: ambient, which are non-chilled simple average from 66 percent to 71 percent. processed food such as noodles, rice, potato But again, this slow rise in the simple (over chips, and so on; chilled and frozen foods such chains) average masks rapid change among as dairy, meats, fi sh, and so on; alcoholic and the top chains: SM’s leapt from 51 percent to non-alcoholic beverages; and fresh fruits and 72 percent, Robinson’s from 52 percent to 65 vegetables. Tobacco, a minor share of grocery, percent, and a major new player (Puregold) is excluded. Grocery less food is composed of emerged suddenly with 78 percent of its grocery nonfood FMCG (fast-moving consumer goods) sales from food. This spelled a spectacular such as detergents, soap, and beauty items. jump in the importance of modern retail in the Several salient points hold some surprises. Philippine food sector – from 500 million USD First, the share of grocery in banner sales in 1999 to 3.8 billion USD in 2010. Recalling rose from 59 percent to 65 percent over the our rough and conservative estimate that the decade and thus the share of non-grocery fell; PlanetRetail coverage may neglect some 33 these are mainly nonfood durables and semi- percent of food-selling modern retail, we can durables like clothing, kitchen appliances, etc. reasonably add 33 percent to the above estimate, This slow general shift masks the fast change to come to 5.05 billion USD of food sales by in the two front-runners, SM and Robinson’s; modern retail in 2010. grocery went from 41 percent to 56 percent Note that modern retail food sales are over the decade in SM, and from 33 percent to double the Philippine agrifood exports – and 50 percent in Robinson’s. This increase both thus can be said to have double the impact on refl ects the rising role of food formats in these the agrifood sector2. Yet the issue of agrifood leading companies and the decline in importance exports far dominates over modern food of their department store formats, discussed retail in public policy debate on food sector

2 Philippine exporters sold 2.64 billion USD for the top 20 agrifood products per FAOSTAT for 2007. Asian Journal of Agriculture and Development, Vol. 6, No. 2 69

development. We hope that stakeholders and disaggregated by North Asia (at 48 percent in concerned policymakers will take note of the 1999 and 73 percent in 2007) and Southeast above fi gures so that this gross imbalance may Asia (at 34 percent in 1999 and 46 percent be redressed. by 2007). For the Philippines, the fi gure is 45 It is challenging to go from our estimate percent in 2007 versus 43 percent in 2005, of 5.25 billion USD of modern food sales in close to the 42 percent we calculated above. the urban market (as that is where the modern Note that the urban food market is, in volume, retailers operate) directly to a calculation of the roughly 70 percent of the total food market in share of modern urban food retail. The problem 2000 – the rest being the rural market. Hence, is that we do not have statistics on the sales by the share of modern food retail in the overall informal sector retailers, and thus do not know national market would be around 35 percent, the total size of the pie. Two rough methods can and is growing much faster than GDP. be used to resolve the issue. Third, another indicator of food retail On the one hand, we can use household modernization is the relatively low share of expenditure survey data to construct the ambient processed food in total grocery at 26 denominator, and then use the estimate of food percent, (similar over the decade), and its share in sales by modern retail as the numerator. NSO total food at 26 percent/71 percent, or 37 percent. (2005) shows total urban household expenditure Historically, as well as recently in developing on food and nonfood items to be 810 billion PhP, countries, the earliest category penetration of which we convert approximately to 16.2 billion modern retail has been in ambient products – USD. We then take the share for food consumed those that can be stored, whose costs can be at home (38.6 percent) from the 2000 Family driven down by economies of scale of storage Income and Expenditure Survey by NSO, and and distribution, and on the basis of which derive an urban food retail market of 6.26 billion modern retail usually gains its fi rst competitive USD for the same year. We then double that to advantage over traditional retail. An example 12.5 billion USD as an approximation of the of this is in Hong Kong in the 1970s-1980s, food market at the end of the decade; the GDP when supermarkets vanquished rice shops (Ho slightly more than doubled in nominal terms, 2005). As this share was already fairly low at but typically the food share would decline as the start of the decade, it appears that the gains per Engel’s Law. Dividing into that the modern in “non-ambient” foods had occurred already in retail food sales of 5.25 billion USD, we get a the 1990s when modern retail was in rapid early rough share of 42 percent. Interestingly, we see establishment. next that this is close to the fi gure from urban One of the “diversifi cation” categories consumer surveys by Nielsen Company. (away from the traditional ambient category) On the other hand, we can consult consumer is the chilled and frozen product category such surveys of buying habits from modern retail. as meat, dairy, frozen prepared meals, and the We know of only one large-sample survey like. Over the decade, SM experienced a rise in of this type done in urban areas, which is by the share of chilled and frozen products from 14 Nielsen (2008). Nielsen reports that by 2007, percent to 19 percent; and Robinson’s, from 16 the share of modern retail in grocery retail in percent to 20 percent. This appears to be driven Asia (excluding Japan) was 52 percent, up by: (1) a rise in households with refrigerators, from a mere 35 percent in 1999. The data is (2) rising incomes, (3) the rise of large food 70 Glory Dee Romo, Larry Digal and Thomas Reardon

manufacturers such as San Miguel, and (4) products, many of which are grown by small investment in cold shelves in the outlets and and medium farmers. On the other hand, the cold/cool chain in the distribution centers and exports, mainly bananas and pineapples, tend to truck fl eets. come from medium and large growers. Hence Fourth, at once surprising in the Philippines the supermarket-market may have more impact but expected from international trends, on small and medium farmers over time than Philippine modern retailers’ marketing of fresh might the export market, assuming present produce has increased quickly in the 2000s. patterns continue. In simple averages over companies, the table The signifi cance of this development shows that fresh produce is 10 percent; this should be seen in context. The most “non- share is similar to the typical 10-15 percent of traditional” product in modern retail is fresh supermarket sales that one sees internationally produce. Historically (in the 20th century in the – and it is also near the 11 percent share of fruits US and Western Europe), fresh produce retail and vegetables in food from the NSO household was consistently the last product category to be survey of 2000. penetrated by modern retail. In the US, it took Again, the aggregate average which 40 years after the advent of supermarkets in the remained steady at 10 percent over the decade 1920s for fresh produce to be a signifi cant item masks the leap from 7 percent to 10 percent of or for them to be even sold in supermarkets. SM’s grocery sales coming from fresh produce. This was simply because people (in the US) Overall, modern retail’s fresh produce sales traditionally bought fresh produce from tiny leapt from 70 million USD in 1999 to 418 produce shops, wet markets, and street hawkers million USD in 2010. and pushcarts. This same “lagged” penetration Compare this latter fi gure fi rst with the of fresh produce has been experienced in overall urban produce retail market: using developing countries, but with a rapid rise in the estimate of the urban food market as 12.5 the 2000s. billion USD in 2010, and the 11 percent share in produce (NSO 2005), the produce market is Format diversifi cation and sales gross yield roughly 1.375 billion USD. The share of modern trends in modern food retail retail in it is 30 percent – some 12 percent below the share in overall food of 42 percent, but that Table 3 shows the shares of various formats gap is typical internationally. in modern retail sales in the Philippines Moreover, compare the 412 million covering the period 1999-2010, in millions of USD with the exports of fresh fruits from the USD. Given the widely differing core formats Philippines in 2007 (for the top items: bananas, of the various companies, we discuss them in 856 million USD; pineapples, 148 million groups. The department store-cum-supermarket USD; and mangoes, mangosteens, and guavas, food-sales fl oor is, in the Philippines, the 35 million USD altogether). Exports of fresh traditional start-up format of modern retail – produce are more than twice as important at as it was in many countries. Similarly, the rise present as the produce sales of the modern of other formats separate from the department retail market in the Philippines, but the latter store base tended to go fi rst into supermarkets is growing fast. Moreover, the fresh produce and then into the larger hypermarkets and sold by supermarkets includes a broad range of warehouses, and then into the smaller formats Asian Journal of Agriculture and Development, Vol. 6, No. 2 71 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 2010 Estimate cation of items sold fi 1.SM a) Hypermarket-warehouse b) Supermarket c) Convenience store 2.Puregold a) Hypermarket-warehouse b) Supermarket 22% c) Convenience store 3.Robinson’s a) Hypermarket-warehouse b) Supermarket 100% c) Convenience store 42% 4.Rustan’s 100% a) Hypermarket-warehouse b) Supermarket (9%) c) Convenience store/NS5.Mercury Drug (9%) a) Hypermarket-warehouse b) Supermarket (39%) (13%) c) Convenience store (22%)6. AS Watson (37%) (12%) a) Hypermarket-warehouse (21%) (9%) b) Supermarket c) Convenience store/drugstore 0 (19%) (9%)7. Benison/Waltermart 8% a) Hypermarket-warehouse (19%) b) Supermarket 100% c) Convenience store 0 0 100% 8. President (7-11) 0 a) Hypermarket-warehouse 37% b) Supermarket 100% 0 (11%) c) Convenience store 0 (27%) (8%) (18%) 0 0 0 0 (11%) 100% (20%) 0 0% 0 37 (12%) 33% (9%) (12%) (24%) 0% (14%) Company, classi Company, Table 3. Shares of various formats in modern retail banner sales the Philippines (1999-2009), percent over total Table 72 Glory Dee Romo, Larry Digal and Thomas Reardon

b 100% (continued) a

2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 b 2010 100% 100% Estimate cation of items sold fi Warehouse b C&C; a : : Authors’ calculations from raw data PlanetRetail. Authors’ : Sales are only by formats that sell some grocery 9) Makro a) Hypermarket-warehouse b) Supermarket c) Convenience store 10. Jollibee Bakery-retail a) Bakeries Uniwide warehouse 11. a) Hypermarket-warehouse b) Supermarket c) Convenience store 12. Pricesmart a) Hypermarket-warehouse 100% 100% Note: Legend b) Supermarket c) Convenience store 13. 5 forecourt chains Source Company, classi Company, Table 3. Shares of various formats in modern retail banner sales the Philippines.... Table Asian Journal of Agriculture and Development, Vol. 6, No. 2 73

such as convenience and neighborhood stores. retail seeks to tap the convenience market, to That shift is an indicator of modernization, and penetrate dense urban spaces and supplant a second phase, of modern retail itself. corner mom & pop stores, and to grow with To understand the context of format spreading highways. differentiation, note some of the different Second, for the three leaders with department formats of leading chains, with an example stores that also sell some food, the department from SM. The hypermarket of SM (Super SM) store share of the total has declined rapidly as in 2010 has on average 10,833 sq meters of the other formats rose: for SM over the decade, sales area and 39 million USD in annual sales; 67 percent to 34 percent; Robinson’s, 77 percent the supermarket format (SM Supermarket), to 44 percent; and Rustan’s, 72 percent to 67 covering 5,616 sq meters, generated 30 million percent. USD in sales (compared to 28 million USD in Third, shown only partly in Table 2 is 2008). Note however that the sizes of the formats the trend of modern retail growing outside of are self-defi ned by the chains; a Waltermart shopping centers or malls. This was shown “hypermarket” has 14 million USD sales a year, in the case of convenience stores above. The only about half the sales from what SM calls a same case was also true for supermarkets. supermarket. There are several salient points in An example is the Savemore format of SM. the table. These are somewhat smaller than their regular First, there has been a “bifurcation” in supermarkets, and are located outside malls or format development that one has seen in other shopping centers. The format started in 1999, countries. and had only grown to 10 stores by 2008, On the one hand, there has been a marked but then jumped to 26 by the end of 2009; 18 growth in hypermarkets and warehouse stores. more are slated for 2010 alone. The company The share of SM sales in hypermarkets went announced that this is explicitly designed to from none to 22 percent by decade’s end; penetrate dense urban areas, capture the market Puregold and Waltermart uses only this format, near residences, locate near dense streams of as Makro did and still do, along with Uniwide street traffi c, and near wet markets. Fresh meats and Pricesmart; Rustan’s moved it from 12 and produce are emphasized (PlanetRetail.net, percent to 39 percent of its sales over the accessed 15 May 2010; The Philippine Star decade. This format is attractive partly because, 2008). We expect this trend to continue and in the words of SM’s food retail director, it is accelerate – and with it, a renewed impetus for a “destination shopping” point, where shoppers modern retail diffusion. (consumers or retailers) incur the transaction While the trend for modern retail to spread costs to get to it and want to fi nd a wide variety beyond malls and shopping centers continues, it of products in a one-stop shop. is worth refl ecting on why malls or commercial On the other hand, there has been rapid centers have played a prominent role so far in development of convenience stores, both modern retail diffusion in the Philippines – as as overall companies such as President (7- they did in the US in the 1950s-60s. We surmise 11), and as proliferating forecourt stores in several reasons. Malls provide the functions gasoline stations. This is a typical trend in both of clustering of stores and services and hence developing and developed countries as modern economies of agglomeration, diversity for 74 Glory Dee Romo, Larry Digal and Thomas Reardon

customers, and shared expenses for services Table 4 shows sales in millions of USD and infrastructure. They have parking areas, per thousand sq meters of sales space for 2000, entertainment areas and other amenities, and so 2005, 2009, and 2010. This cannot be interpreted have become leisure centers that are protected as total factor productivity because it is not net from heat and cold, congestion, and crime and of costs, but rather provides a rough indication parking problems. of performance. Excluding drugstores from the It has been common both in Asia and in simple average as they have much higher gross other regions for shopping center development sales per sq meter as expected, we fi nd that or management companies to either have sales per sq meter at fi rst declined by 7 percent their own “brand” department store, or be in from 2000 to 2005, and then shot up by 35 partnership with one, and use that as an anchor in percent from 2005 to 2010. This may have been the mall. By this path, many malls came to have because in the fi rst half of the decade, sales department stores-cum-supermarkets, such as strategies that were tried were less adapted to SM, Robinson’s, Rustan’s, and Gaisano malls in the context; weaker retailers (which were later the Philippines, but that phenomenon is repeated bought out) were still operating; or competition in many developing countries especially in the patterns were spatially such that there was some 1970s-90s. Powerful investment companies redundancy; or some combination of these such as the ones mentioned and similar others reasons. A reversal of several of them, such as combined their real estate acquisition and change in strategies and possibly a decrease in management with mall unit rentals and the competition in given areas (due to acquisitions) development of their own anchor chains, starting may be responsible for the rise of sales per sq with department stores and later extending to meter. Comparing across companies, it is striking other formats. how divergent the fi gures were in 2000 and even The typical trend that we now see in the in 2005 and how convergent they have become Philippines, and which occurred internationally, by 2010. This suggests that competition may is the development of those formats “off-mall” as be inducing imitation and diffusion of practices stand-alones, where they then began to compete that smooth out variation in performance over with the rising modern retail chains that did most surviving chains, with a few exceptions. not start as mall anchors. Typically, as is now occurring in fi rst wave developing countries Spatial and Socioeconomic Paths and historically in developed countries, the of Diffusion of Modern Retail mall-based supermarkets eventually became minor players and the main modern food retail Spatial diffusion took place in stand-alone stores – which could penetrate far more densely the neighborhoods The diffusion path of modern retail over and city centers. It appears that the current provinces and islands in the Philippines trend that one sees, for example in SM with essentially mirrors international experience. its continued development of stand-alone Reardon and Timmer (2007) note that modern supermarkets and hypermarkets (at rates higher retailers tend to spread in waves over areas – than development of mall-based supermarkets) countries in a region, zones in a country, and will be the dominant trend in the Philippines in over socioeconomic group market segments the next decade. – fi rst in the largest and/or richest market, and Asian Journal of Agriculture and Development, Vol. 6, No. 2 75 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 3.51 2.61 2.8 2010 Estimate cation of items sold fi : Authors’ calculations from raw data PlanetRetail. Authors’ : 1. SM2. Puregold 3. Robinson’s4. Rustan’s5. Mercury Drug6. AS Watson 7. Benison (Waltermart) 8. President Chain (7-11)9. SHV Makro (Cash & Carry)10. Jollibee Bakery-retail Uniwide Warehouse11. 3.47Simple average of all 3.5 3.63Simple average excluding Mercury 4.7 3.35AS Watson Drug & 7.7 3.66 3.41 3.0 4.6 3.29 7.5 7.03 3.03 na 2.0 6.9 4.4 1.9 1.24 1.81 2.22 3.19 3.85 2.09 5.3 3.6 2.31 6.2 1.8 3.24 6.8 2.02 2.03 4.44 5.7 3.6 2.9 1.04 3.29 Company, classi Company, Table 4. Modern retail in the Philippines: banner sales/sales area (1999-2010) millions of USD/thousands sq meters Table Sales are only by formats that sell some grocery. Source 76 Glory Dee Romo, Larry Digal and Thomas Reardon

Table 5: Pattern of spread of SM and Robinson Malls in the Philippines, 1980s to 2009

Other Luzon Visayas’ three Metro Manila Mindanao Total provinces main islands 1980s SM = 1 1 Robinson’s = 2 Robinson’s = 1 Robinson’s = 1 1990s 9 SM = 4 SM = 1 Robinson’s = 4 Robinson’s = 11 Robinson’s = 5 Robinson’s = 3 2000s 53 SM = 10 SM = 15 SM = 3 SM = 2 Total 63 Source: SM Prime Holdings (2009); http://www.philippinecountry.com/philippine_malls/major.html#plaza%20fair

then into smaller and/or poorer ones. Retail Manila but has a small presence in the Visayas, chains often use the procurement system or and none in Mindanao. Other fi rms tend to have supply base of the initial base area to supply the limited local presence in a particular island. As stores in the next area or market. That is why the large chains move to new provinces and supermarkets are found well beyond the middle islands, some of them will likely acquire or buy class and the large cities, selling to the food out local chains. markets of the poor. The leading companies have followed Roughly 65 percent of the population and the “typical” pattern one fi nds internationally. at least 75 percent of the retail market in the Table 5 illustrates this by showing the 63 mall Philippines (despite its being an archipelago openings of SM and Robinson’s from the 1980s with some 7,000 islands) are in three focal to the present. There was but 1 opening in the points – the island of Luzon, the three main 1980s, 9 in the 1990s (from 1990 to 1997, islands of the Visayas region, and the island stopping abruptly with the fi nancial crisis, and of Mindanao (PDFI 1999). Within these areas, then resuming in 2000), and 53 in the 2000s. sharp concentrations of population and markets The initial openings were in Metro Manila, then exist, such as those in Metro Manila and the simultaneously spreading to other provinces in provinces around it, in and around Cebu, in and Luzon, and gradually to the islands of the Visayas around Davao, and so on. region. The last step has been the openings of However, given that these three areas both chains in Mindanao, ingressing into the are separated by sea and due to the attendant moderate to densely occupied areas dominated relatively high transportation costs, chains by the regional chains. fi nd it easier to spread over areas on a given large island. A large chain may often do this, Socioeconomic diffusion after opening initial positions in each of the main islands. The Luzon-based companies SM For the market penetration of the various and Robinson’s have spread pan-Philippines socioeconomic strata, there are two parallel (over the three areas); Gaisano has spread over trends. The fi rst, starting roughly in the 1970s, Mindanao and the Visayas main islands; and is the continuity of the set of smaller and local Rustan’s has spread extensively over Metro chains or independent modern retailers who Asian Journal of Agriculture and Development, Vol. 6, No. 2 77

catered to the gamut of income classes in their modern retailers that led to falling costs and areas, but with an emphasis on the B, C, and eventually prices which allowed them to D consumers. These chains are typifi ed by gain market share from traditional retailers the members of PAGASA. The second is the – fi rst in processed foods and staples, and evolution of the larger chains that tended to eventually in fresh produce. start with As and Bs, and moving into the Cs Below we use the Philippines as an example in the second half of the 2000s. Part of this of the applicability of these determinants. socioeconomic shift is simply a correlate of the spatial shift noted above. The emphasis on the As and Bs continues to a large extent for Demand-side Factors Robinson’s and Rustan’s. On the other hand, SM is taking a tack that explicitly focuses First, there has been rapid urbanization. The on moving into the market of the C’s; this is urban share of the population was 39 percent in supported by cost-cutting measures on the 1980, 48 percent in 1990, 52 percent in 1995, procurement side discussed below, and by the 57 percent in 2000, and projected to be 73 rolling out of the Savemore format emphasizing percent (the northern Asia rate) by 2020 (US price discounts. Department of Commerce 2001). Because urban incomes were and are higher and grew faster DETERMINANTS OF RETAIL than rural incomes, the share of the urban food TRANSFORMATION IN THE PHILIPPINES economy in the overall food market is already, IN INTERNATIONAL CONTEXT we estimate, about 75 percent by 2010. Second, incomes have grown rapidly in the Reardon et al. (2003) lay out several 2000s, at a real GDP compound growth rate of determinants of the take-off of modern retail 5.1 percent over 2000 to 2008, similar to the 5.2- in developing countries roughly in the early 5.5 in Indonesia, Malaysia, and Thailand, other 1990s: second wave countries in the retail diffusion 1) Demand-side factors such as urbanization, grid, as per PlanetRetail data. rise in incomes, and improvements in Third, female participation in the labor transportation and proliferation of vehicles force (working outside the home) increased; for and refrigerators; these reasons were women aged 24-54, the share went from 51.5 necessary but not suffi cient as they already percent in 1980 to 59 percent by the end of the existed before the grand take-off; 1990s (Lim 2002). 2) An important supply side reason in many Fourth, migration from the Philippines, countries was the deregulation of commerce and the effect of remittances and return of and retail foreign direct investment (FDI) in workers, may also play a role. The importance particular; this led to a large infl ux of FDIs of migration from the Philippines, with some and competitive domestic investments; 8.2 million Filipinos working abroad and their however, FDI has not been the driver in all remittances of about 17 billion USD or 13 countries – e.g., in India, Russia, China, and percent of GDP in 2007 (Ruiz 2008) are well Chile, domestic investments have been the known. However, there have been no empirical main drivers; studies of whether migration has some effect on 3) Procurement system modernization by 78 Glory Dee Romo, Larry Digal and Thomas Reardon

the rise of supermarkets in the Philippines (or In Metro Manila alone, public transport in other countries with heavy migration such as vehicles tripled between 1980 and 2001 or Mexico, China, or India) – such as via the cultural double the rate of city’s population increase. effects of staying in countries with food systems Public motorcycles for the same period rose dominated by supermarkets (as Cabochan from 4,801 to 51,768; cars from 1,461 to 27,322; 2005 posits). However, modern retailers have utility vehicles (jeepneys, company buses) from “voted with their pesos” concerning the idea 27,202 to 28,250; public utility buses from of the link between migration and shopping at 3,578 to 8,232; or a total of 37,042 to 115,572 modern retail. For example, SM has set up 11 for the said period (www.klima.com.ph 2010). “Global Pinoy Centers” in their malls – giving Many bus and jeepney lines have stops at the special discounts to Overseas Filipino Workers malls and shopping centers, which facilitates (OFWs), with a remittance center, and facilities this shopping. These patterns occurred in other for voice and video contact with migrants/ cities as well. Moreover, the recent completion OFWs for families. of LRT and MRT (railway transport) lines Finally, in other countries, the increase contributes substantially to bringing people to in vehicle ownership is associated with the malls; some of their major stations are close more mobility and thus with better access to to the shopping centers. supermarkets. However, it appears that the This means that in the Philippines, modern rapid increase in public transport may be a more retail growth can be relatively independent of important determinant in the Philippines. vehicle ownership growth. This might (as a Passenger cars and motorcycles grew hypothesis) be a reason for the disproportionate from about 10 to 23 per 1,000 population from persistence of malls in the country relative to 1980 to 2000; however, this number was still other countries where they tended to have a well below Thailand and Indonesia (Nagai et greater share in earlier stages. It appears that al. 2003). IBM (2007) shows results from a routes of public transportation move along Nielsen survey of how shoppers commute to major urban arteries, along which malls are shop. Fifty-seven (57) percent of Filipinos strategically positioned. use public transport – compared to only 3 percent in India, 19 percent in Indonesia, and 2 percent in Taiwan. Car use in the Philippines Supply-side Factors - Investment is 10 percent. As refl ected in the fi gures cited, and Procurement the relative importance of public transportation in urban Philippines is greater compared to Why are there no global food retailers similar countries in the same general category operating in the Philippines? of income – and it increased extremely quickly over the past two decades. About 78 percent of No leading global chain has entered the total daily person trips rely on buses, jeepneys, Philippines , although many have shown interest. taxis, and tricycles run as public transport. From 2002 to 2007, Wal-mart, Carrefour, Asian Journal of Agriculture and Development, Vol. 6, No. 2 79

Metro, Tesco, Casino, and others expressed heavily – Luzon’s population of 46 million in interest, and some of them sent missions to 2007 is close to that of South Korea; Mindanao’s explore concrete options, but none entered3. Yet 22 million is close to that of Malaysia. all of them made public statements about how We have already seen that SM has started attractive the Philippines is for modern retail its investments inter-island (as a recent step, investment, and how much potential there is. but now steadily), as have Gaisano and others. Moreover, some of the internal discussion Foreign chains would face no differential higher in the Philippines about why the foreign chains cost than local players. Moreover, foreign chains have not come en masse features various have faced inter-island costs in Indonesia, hypotheses about the contexts that are supposed across-mountain costs in many countries, and to have “scared off” foreign chains. We critique vast distance shipping in Russia and China. those hypotheses as follows. These costs do slow down expansion but have The fi rst hypothesis presented in the debate not deterred investment elsewhere. The same is that “traditional retail is important in the goes for the costs of corruption. Philippines Philippines and too ingrained to overcome, ranks among the highest in the world as per and that scares off foreign chains.” We already www.worldaudit.org/corruption/, but Russia, noted that the traditional retail found in the Vietnam, and Indonesia rank close, and each of Philippines is the same as those in all the other these three are major retail FDI destinations. countries where there have been a supermarket Third, it is posited that “incomes are low and revolution – starting with the US and Western thus not attractive to foreign chains.” We have Europe and then into today’s developing regions. already noted the following: that incomes are The foreign chains have already met the same growing as quickly in the Philippines as they are challenge of initial conditions in other countries in other FDI destination countries in the region; in the region and elsewhere. Moreover, the that there is a substantial middle class; and that massive rise of modern retail in the Philippines the urban food economy constitutes some 75 undermines this hypothesis. percent of the overall food economy, and thus Second is the claim that “transaction is a substantial target. Moreover, even without costs are high in the Philippines because it is much FDI, modern retail in the Philippines has an archipelago with 7,000 islands; corruption already adapted (using standard international costs are also high – and all of that scares off approaches, nothing specifi c or indigenous to foreign chains.” Transaction costs are indeed the Philippines) to consumer needs beyond the relatively high because of the need for ocean middle class. Modern retailers in the country shipment between the three areas (Luzon, have already started to penetrate the food Visayas, and Mindanao), and internal roads in markets of the urban poor, similar to what its many areas are poor (Lidasan and Castro 2009). counterparts have done in many fi rst and second However, the largest two islands have fairly wave countries. To see this, compare the share of dense infrastructure and are comparable to other food retail of modern retailers (45 percent) with countries where foreign retailers have invested the share of the middle class in the Philippine

3 For example, Metro considered the possibility in 2001 (PlanetRetail 2001, September 28) and then actively explored the possibility (PlanetRetail 2007, September 12) – but it did not enter. Wal-mart (as well as the Casino chain of France) was reported to have had discussions with the Department of Trade and Industry in 2001 (PlanetRetail 2002, January 2) when the retail trade law was being liberalized, and then actively explored entry in March 2002; it was supposedly waiting for the expiration of the foreign ownership clause under the retail trade law at the end of that month to fi rm up its plans – but it did not enter (PlanetRetail 2002, 5 March). 80 Glory Dee Romo, Larry Digal and Thomas Reardon

population estimated at 25 percent (Senauer companies with equity of 200 million USD or and Goetz 2003). more. In contrast to the points above, we present Aside from the act, the Constitution of 1987 what we think are reasons driving the lack also stated that foreign fi rms could not own land, of investment in the Philippines by the large but could only lease up to 75 years; fi rms that foreign chains: (1) FDI liberalization that is only are of 60 percent Filipino ownership or more partial, (2) perception that conditions on JVs are can own land – so that joint ventures can do so disadvantageous, and (3) diffi culty of access to (Catindig 2001; Aldaba 2005). These provisions real estate. We lack the data to rigorously assess were intended to: (1) Cushion the impact of the these points, but present key information and act on large and medium players, (2) keep out structural conditions to tentatively support our small foreign fi rms that would penetrate into arguments. traditional strongholds of small retailers such First, retail FDI liberalization has been only as small towns and dense poor urban areas, (3) partial, with signifi cant constraints remaining. support local manufactures. The act stopped For a half century there was a socio-political short of imposing various provisions one sees movement in the Philippines in favor of in retail regulations in other countries, such as requiring that a retail company has Philippine limiting opening hours, or zoning so that large citizenship. This was aimed against foreigners, retailers cannot go into dense areas where many in particular Chinese retailers, and then against small retailers operate. colonial control of business. It culminated in A debate ensued after the passage of the the Retail Trade Nationalization Law of 1954 2000 act wherein positions were taken that the (PDFI 1999). Starting in the mid-1980s, there provisions for minimum equity, local content, were a series of liberalizations of the Philippine and no land ownership (except in JVs where economy. By the mid-1990s, political pressure they were limited to 40 percent) would be “deal had built up to include retail in that liberalization. killers” for interested foreign chains (Manuzon An intense 5-year debate ensued. 2002). We concur only with the point on land The upshot was that the 1954 law was ownership restriction, but not with the one on repealed in 2000 with the passage of the equity – that it is a serious limitation to foreign Retail Trade Liberalization Act. Some of the chains. These provisions of the act were actually important clauses of the act include: (1) Foreign meant to keep out what are generally considered fi rms with equity destined for the Philippines in international retail as “tiny fi rms”; in fact it less than 2.5 million USD could not enter; (2) is diffi cult to identify any retail chain investing foreign fi rms with between 2.5 and 7.5 million in Asia that is not larger many times over the USD could only own up to 60 percent for the minimum requirements noted above. fi rst two years of the implementation of the new Second, foreign chains perceive the act, and then could own wholly afterwards; (3) conditions of JVs to be disadvantageous. fi rms with more than 7.5 million USD equity Moreover, there appear to be linkages between could own wholly, if any given store costs more the real estate and retail markets that make it than 830,000 USD to establish; (4) for at least diffi cult for foreign retailers to access retail 10 years, fi rms had to source at least 30 percent locations. We obtained unique and confi dential of their inventory from Philippine makers; (5) information to support these hypotheses, at fi rms in the top categories have to have parent least with one key informant (who preferred Asian Journal of Agriculture and Development, Vol. 6, No. 2 81

to remain anonymous) – albeit an important and makes several points, which can serve as one from a large foreign retail chain that had hypotheses for future survey-based analyses. made a careful study of the Philippine market First, it appears from webpages of the for potential entry. By the nature of the issue, leading chains and the reports that cite them, any sample size to assess our hypothesis would that the leading food retailers in the Philippines be tiny – a few fi rms that lead global retail. But are using modern technologies for inventory their points resound with many comments made and sales management such as the well-known by other key informants in our fi eld research. effi cient consumer response (ECR) techniques The global retailer informant noted the (ECRP 2005), centralized distribution centers, following: (1) 6-7 main business families and direct purchase from large suppliers such as dominate both retail and real estate, and they San Miguel or Universal Robina. have cross-shares across the service industry and Second, the distribution interface of production/supply sector; (2) the retail margins processors/food manufacturers and retailers on the surface appear low but are covered by – modern or traditional, but apparently more rebates from suppliers and sales of real estate at among modern retailers – appears to have high prices; (3) it is very diffi cult to get access modernized. Traditional in the Philippines, to land for rent or purchase except via tie-ups and common to many countries, is the system with the main retail/business families who are whereby processed or semi-processed foods at the same time the dominant real estate fi rms. move from small to medium manufacturers, The above explanation of the real estate slaughter-houses, mills, and so on, to wholesale access constraint, combined with our points markets or terminal markets, where they are de- critiquing the other possible explanations of bulked and distributed to retailers by stockists context constraints and equity requirements, on regular delivery routes (Dannhaeuser appear suffi cient to explain why there has not 1977). Over time in the Philippines, as in other been a signifi cant entry of large foreign fi rms, developing countries, the latter system has and probably there will not be in the medium waned, driven by several factors, the main one term. That has not, however, constrained the being the concentration of the food processing Philippines from experiencing a very rapid sector. modern retail transformation. While the modern retail sector took off in the Philippines mainly in the 1990s, the food Procurement system modernization processing sector had by that time already been not only growing, but concentrating rapidly. There is very little published research Concentration of processing that precedes that of found on the key subject of procurement system retail is common in developing countries. Hence, modernization among modern retailers in the in the 1990s, national and later multinational Philippines – or the impacts of system change regional giants grew rapidly, with the rise of on suppliers either among the processing fi rms San Miguel or Universal Robina Corporation, or farmers. The few studies that exist are grey for example, adding to a pre-existing base of literature of self-published reports but there canned fruit enterprises such as Dole. are no academic articles on the issue; this These large fi rms set up their own agent represents a major gap in published research. networks and eventually large distribution The available information is of several types centers from which they distributed fi rst to 82 Glory Dee Romo, Larry Digal and Thomas Reardon

traditional retailers such as sari-sari stores food sales. While much of the policy debate and then to modern retailers as the latter rose. about market development focuses on export It appears, from our key informant interviews markets, we have shown that supermarkets in and the webpages of the processors and modern the Philippines already sell twice the value of retailers, that the interface is more direct and the food that is exported, and modern retail is developed between the large processors and growing faster. retailers – such as the distribution center of From a tiny fraction of this fi gure two San Miguel delivering large volumes of diverse decades ago, modern retail has reached 45 products direct to the distribution centers of percent of urban food retail, already beyond SM, Robinson’s, Rustan’s, and so on. This trend the share of the middle class in the population, – and the bulk buying advantages it confers on and about 35 percent of the national food modern retailers – aligns with the international market. Modern food retail is itself rapidly trend (Reardon and Timmer 2007). transforming with a rise of fresh produce sales, Third, the few existing publications that industry concentration, format diversifi cation treat modern retail’s procurement of fresh off-mall into formats that permit greater market foods, fruits and vegetables in particular, tend penetration, and emerging procurement system to show the emergence of some procurement modernization. While traditional food retail, and modernization in the mid-2000s. For example, even most aspects of modern food retail in the Digal and Concepcion (2004) note that some Philippines track international experience, the of the leading chains began using specialized lack of major foreign presence even after retail or dedicated wholesalers and concessionaires, FDI liberalization in 2000 is a puzzle about rather than the retailers themselves buying from which we presented hypotheses – specifi cally the wholesale markets. There is some evidence related to access of retail space by foreign from Mindanao that this shortened supply chain retailers being an important constraint. increases vegetable quality (Concepcion et al. Given the already large and increasing 2006). importance of modernization of food retail in Again, to our knowledge there have not been the Philippines, this theme should enter the any academic journal publications of systematic mainstream of agrifood research and be the survey-based studies of supermarkets’ relations subject of systematic fi eld survey analysis in with produce farmers, their fresh produce order to start discerning its impact on consumers, procurement practices, or their effects on farmers, wholesalers, and processors. It is too traditional retailers, consumers, or prices. soon, based alone on the existing publications on These are important gaps in research that merit the subject, to identify specifi c policy measures urgently addressing. needed to equip suppliers to take advantage of a trend that appears to be on track to be the CONCLUSIONS dominant demand-side force in the Philippine food economy within a decade. Field-based From a small base by the end of the 1990s, empirical research using representative farm modern retail (chains selling at least some food) and processor surveys is thus urgently needed grew very rapidly in the 2000s at thrice the rate to inform the emerging debate. of GDP growth, reaching 13 billion USD of overall sales by 2010, of which 5.25 billion is Asian Journal of Agriculture and Development, Vol. 6, No. 2 83

ACKNOWLEDGEMENT

We are very grateful for funding from the Asian Development Bank and helpful guidance and comments from Kelly Bird.

REFERENCES

Aldaba, R.M. 2005. The Impact of Market Reforms on Digal, L.N. and S.B. Concepcion. 2004. The Case Competition, Structure and Performance of the of the Philippines: Regoverning Markets, Philippine Economy. Discussion Paper Series Securing Small Producer Participation in No. 2005-24. Makati: Philippine Institute for Restructured National and Regional Agri- Development Studies. food Systems. Accessed May 15, 2010. www. regoverningmarkets.org/en/resources/china/ Balisacan, A.M. 2000. Growth, Redistribution, and china_country_report_2004/ Poverty: Is the Philippines an Exception to the ERCP (Effi cient Consumer Response Philippines). 2005. Standard Asian Story? Journal of the Asia Pacifi c Philippines “out-of-stock” study results. Trends Economy, 5 (1&2): 125-140. (1). Manila: Marketing and Communications Committee of ECR Philippines. Banal III, C.R. 2006. Breaktime: The Store of the Show. Philippine Daily Inquirer, November 14. http:// Ho, S-C. 2005. Evolution versus Tradition in Marketing services.inquirer.net/print/print.php?article_ Systems: The Hong Kong Food Retailing id=32399 Experience. Journal of Public Policy and Marketing, 24 (1): 90–99. Cabochan, C.V. 2005. “Philippines: Modern Retail Trade and Policy Implications.” Paper presented IBM. 2007. Food Retail Formats in Asia: Understanding at the Annual Meeting of the Pacifi c Economic Format Success. Atlanta, Georgia: IBM Global Cooperation Council, Pacifi c Food System Business Services for the Coca-Cola Retailing Outlook 2005–2006, Kunming, China, May 10- Research Council Asia. 13. Klima. 2010. Land Transport Conditions: Existing Catindig, T.A. 2001. The ASEAN Competition Law Modes. http://www.klima.ph, Accessed May Project: The Philippines Report. Japan Fair 15. www.klima.ph/cth/overview/lt_condition/ Trade Commission, Accessed May 15, 2010. existing_modes/public.php/ (accessed 15 May http://www.jftc.go.jp/eacpf/02/philippines_r.pdf 2010).

Concepcion, S.B., M.O. Montifl or, L.N. Digal, Lidasan, H, and J. Castro. 2009. “Philippine Intermodal E.T. Rasco, and P.J. Batt. 2006. Consumers, Logistics Systems and Policies”. Presented at Market Intermediaries, and Farmers: Different the Logistics Conference, Makati, Philippines, Perceptions of Vegetable Quality. Stewart February 5. Postharvest Review, 2 (3): 1-5. Lim, Lin Lean. 2002. Female Labour Force Dannhaeuser, N. 1977. Distribution and the Structure of Participation. Background paper for the Retail Trade in a Philippine Commercial Town United Nations Population Division, Expert Setting. Economic Development and Cultural Group Meeting on Completing the Fertility Change, 25 (3): 471-503. Transition. Accessed March 1. www.un.org/ esa/population/publications/completingfertility/ Digal, L. N. 2001. An Analysis of the Structure of completingfertility.htm the Philippine Retail Food Industry. Philippine Journal of Development, 28 (1): 13-54. 84 Glory Dee Romo, Larry Digal and Thomas Reardon

Macabasco, D.R. 2009. The Supermarket Business in Reardon, T. and C.P. Timmer. 2007. “Transformation the Philippines. Food and AgriBusiness Monitor, of markets for agricultural output in developing August. Pasig City, Philippines: Center for Food countries since 1950: how has thinking changed?” and AgriBusiness, University of Asia and the In Handbook of Agricultural Economics edited Pacifi c. by R.E. Evenson, and P. Pingali, 2808-2855. Amsterdam: Elsevier Press Manuzon, M. 2002. Retail Blast: The Retail Trade Industry Might just be the Country’s Next Reardon, T., C.P. Timmer, C.B. Barrett, J. Berdegue. Economic Growth Driver. Philippine Business, 2003. The Rise of Supermarkets in Africa, September 2. www.philippinebusiness.com.ph/ Asia, and Latin America. American Journal of archives/magazine/vol9-2002/9-2/industry.htm Agricultural Economics, 85 (5): 1140-1146. (accessed 15 May 2010). Ruiz, N.G. 2008. Managing migration: lessons from the Minten, B., T. Reardon, and R. Sutradhar. 2010. Food Philippines. Migration and Development Brief 6, Prices and Modern Retail: The Case of Delhi. Migration and Remittances Team Development World Development 38 (12): 1775-1787 Prospects Group. The World Bank, Washington, D.C., August 11. Nielsen. 2008. Retail and Shopper Trends, Asia Pacifi c 2008. Hong Kong: Nielsen Retail Services. Sanchez, E. 2006. Uniwide, Suy Sing OK lease agreement. Philippine Daily Inquirer, February NSCB (National Statistical Coordination Board). 2008. 13. http://www.accessmylibrary.com/article- Philippine Statistical Yearbook 2008. Manila: 1G1-141988450/uniwide-suy-sing-ok.html. NSCB. Senauer, B. and L. Goetz. 2003. The Growing Middle Nagai, Y., A. Fukuda, Y. Okada, and Y. Hashino. 2003. Class in Developing Countries and the Market Two-wheeled vehicle ownership trends and for High-Value Food Products. WP 2003-02, issues in the Asian region. Journal of the Eastern Food Industry Center, University of Minnesota. Asia Society for Transportation Studies, 5: 135- 146. SM Investments. 2009. SM Prime Holdings 2009. Manila: SM Investments. NSO (National Statistics Offi ce). 2005. Annual survey of Philippine business and industry wholesale The Philippine Star (no author stated). 2008. Savemore: and retail trade, repair of motor vehicles, The Neighborhood SM. www.philstar.com, motorcycles, and personal and household goods. April 28. http://www.philstar.com/Article. Manila: NSO. aspx?articleId=58729.

PlanetRetail. 2001. Metro Cash & Carry Takes on the US Department of Commerce. 2001. Population trends World, 28 September. Philippines. U.S. Department of Commerce, Economics and Statistics Administration, Bureau PlanetRetail. 2002a. Next Port of Call – The Philippines. of the Census. PPT/92-11RV2 Issued February PlanetRetail.net, 2 January. 1996, with corrections added November 2001. Washington, D.C. PlanetRetail. 2002b. WAL-MART For the Philippines. www.PlanetRetail.net, 5 March. World Audit. 2010. 2009 Country Corruption Rankings. worldaudit.org, Accessed May 15 2010. http:// PlanetRetail. 2007. METRO to enter Philippines?. www.worldaudit.org/corruption.htm. www.PlanetRetail.net, 12 September.

PDFI (Policy and Development Foundation, Inc.). 1999. Retail Trade Policy and the Philippine economy. Report submitted to the Trade and Investment Policy Analysis and Advocacy Support of the Philippine Exporters Association. Manila: PDFI.