Click the image for more details A clear view of where the value lies in China's supermarket industry over the next decade: 'small food supermarkets' are a major opportunity. The supermarket industry in China is undergoing rapid change, especially in the last two years with the emergence of 'big store + delivery' models represented by Hema and 7Fresh. Discussions about the future evolution of offline supermarket formats have become a hot topic in the market. Offline supermarkets require large investments, have low net profit margins, and long payback periods; for example, supermarket properties often have leases of 15 years or more. Therefore, it is crucial—even a matter of life and death—to anticipate trends early and precisely position supermarket formats that meet consumer needs. We believe that the core driver of supermarket format changes lies in societal shifts. Insight into long-term internal trends in Chinese society will enable early identification of the supermarket formats that best meet consumer needs, allowing companies to make precise layout decisions, avoid resource waste, and help investors choose rational investment targets. We believe that the customer unit for supermarkets is the 'household,' and the decline in average household size in China, leading to changes in supermarket shopping behavior, is the most critical factor affecting the evolution of supermarket formats. The best reference is our neighbor Japan, where the average household size has been continuously declining for decades, now below 2.5 people, while more women participate in the workforce. The result of these changes is that for Japanese households, the efficiency of large-scale, infrequent shopping trips has greatly diminished, and the total consumption of fresh produce has continued to decline. These changes have been reflected in the evolution of Japan's supermarket formats. In recent years, the general merchandise supermarket (GMS) format has been shrinking, privately operated specialty food stores have been closing, and 'small food supermarkets' have begun to rise. Note: General merchandise supermarkets (GMS) refer to large supermarkets with sales areas ranging from several thousand to over ten thousand square meters, selling food, daily chemicals, household goods, small appliances, and other general merchandise. Specialty food stores are primarily privately operated, such as fish shops and tofu shops. 'Small food supermarkets' refer to small food-focused supermarkets of 200-500 square meters, like AEON's MY BASKET shown on the right. We believe that any change in the retail industry is essentially an improvement in the efficiency of retail activities, and this process inevitably leads to a redistribution of retail channels. In China, with the rise of internet e-commerce giants like Alibaba and JD.com, e-commerce penetration has reached world-leading levels, far exceeding Japan's. For low-frequency, highly standardized products like daily chemicals, home textiles, and small appliances, e-commerce channels have clear cost and efficiency advantages over GMS. For food, large GMS stores have lower procurement efficiency and profitability compared to small food supermarkets with a coverage radius of within 500 meters. As Chinese society changes, this gap will become more pronounced. Meanwhile, specialty food stores, due to less concentrated procurement and limited private management capabilities, will likely become chain-operated and consolidate their product offerings, gradually evolving into small food supermarket formats. Given that China's social trends are similar to Japan's, we believe the future direction of supermarket format changes will be the same. People will increasingly prefer small food supermarkets over GMS, and small food supermarkets will take over the channel role of wet markets and traditional specialty food stores. Small food supermarkets will flourish across China, while GMS will continue to shrink. In the future, there will be no 'China's Walmart' or 'China's Costco.' Over the past 17 years, AEON, Japan's leading supermarket company, has seen its market share in the GMS business rise from 19.8% to 53.69%, but due to the overall contraction of GMS, its revenue and profit growth have been very slow. In contrast, AEON's chain of small food supermarkets has grown rapidly, with market share, revenue, and profit all achieving relatively fast growth. This case is worth our consideration. We also propose a small food supermarket format of 400-500 square meters with a small coverage radius, which we believe suits China's characteristics, for the reference of investors and companies currently engaged in such businesses. Currently, there is no leading brand or company in China's small food supermarket segment; the scale is generally small, which corresponds to the huge potential of this market. The fastest-growing player is Fresh Legend (Shengxian Chuanqi), which has opened a hundred stores. Also worth watching are JD.com, which has outstanding logistics advantages and plans to open 300-500 square meter 'Four Seasons Preferred' (Siji Youxuan) small food supermarkets to complement 7Fresh, and Sun Art Retail (Gaoxin Retail), which has high operational efficiency and may enter the small food supermarket segment with 'Hema Xiaoma' targeting both B2B and B2C ends. For GMS, attention should be paid to the risks associated with companies that have expanded aggressively in recent years. I. Japan's GMS business continues to shrink, while small food supermarkets grow steadily According to data from Japan's Ministry of Economy, Trade and Industry, sales in the GMS format fell from 7.91 trillion yen in 2001 to 5.79 trillion yen in 2017, a decline of 25.4%, with an average annual compound contraction rate of -1.93%. In contrast, small food supermarkets saw steady growth in turnover, rising from 41.9 trillion yen in 2001 to 44.7 trillion yen in 2017. In terms of store numbers, small food supermarkets have been increasing, with a 22.8% growth over the past five years, while large GMS stores have been decreasing, with a 3.5% decline over the same period. This trend is more pronounced in densely populated areas. For example, in the Tokyo metropolitan area, the number of GMS stores decreased by 36 (a cumulative decline of 22.64%) by 2017, while small food supermarkets increased by 309 (a cumulative growth of 34.33%). Specialty food stores, mainly privately operated, such as butcher shops, fish shops, and tofu shops, have seen even sharper declines, with a nationwide drop of over 40%. II. Deep attribution of format changes: declining household size drives the shift, and 'small food supermarkets' thrive accordingly The core customer for supermarket shopping is not simply an individual consumer but a household. Consumers purchase daily necessities for the entire family, so changes in household composition directly affect supermarket shopping activities. Japanese society has experienced aging and a declining birth rate in recent decades, directly leading to a continuous decline in average household size. In 1990, the average household size was 3.05 people, similar to China's current situation. By 2017, it had fallen to 2.47 people, meaning that over more than two decades, the average Japanese household lost more than half a person. There are no signs of reversal. The direct impact of smaller households is a reduction in the variety of ingredients needed per shopping trip. More importantly, the efficiency of a large-scale, time-consuming shopping trip to a GMS decreases. A straightforward example: a family with more members is far more motivated to prepare meals at home than a smaller family or a single person. Similarly, for a two-hour GMS shopping trip, the efficiency and time cost-effectiveness of buying ingredients for five people versus two people are completely different. Examples of two scenarios At the same time, more Japanese women are entering the workforce, reducing the number of full-time housewives and increasing dual-income households. More work responsibilities mean less time for daily shopping, and Japanese women's shopping time has been declining in recent years. Unlike shopping for clothing, cosmetics, and other highly experiential, self-rewarding purchases, daily food shopping is less experiential. The easiest time cost to cut is daily food shopping. Shopping at a closer small food supermarket rather than a GMS saves time. According to a study published in a local economic journal, 'Spatial and Temporal Characteristics of Urban Residents' Shopping Behavior and Its Influencing Factors,' the time cost factor for choosing a location for vegetable and food shopping is 0.416, while for daily necessities it is 0.266, for outerwear 0.096, and for home appliances 0.077. This shows that for high-frequency food purchases, consumers value time cost far more than for other categories. These changes have directly impacted supermarket formats. Consumers' desire to make large-scale trips to GMS has declined, and they increasingly prefer small food supermarkets or convenience stores near their communities, which take less time and offer essentially the same benefits as GMS shopping. The decline of specialty food stores in Japan is essentially due to similar reasons as the decline of GMS. Specialty stores are often privately operated, with limited ability to procure and manage complex food categories. They often only handle one or a few major categories, so consumers must visit multiple stores to buy all the ingredients for a meal, taking far more time than a one-stop trip to a small food supermarket. Two different shopping models Moreover, with declining household size, the efficiency of visiting specialty stores also decreases. For example, fish at specialty stores is often sold whole, while small food supermarkets sell pre-cut portions. For smaller families or singles, shopping at a small food supermarket is clearly a better choice. According to statistics from the Japan Supermarket Association for 2010, 2012, and 2015, the proportion of Japanese people choosing to shop at small food supermarkets rose from 79.3% to 86.5%, far higher than the 52.5% for large GMS. III. Small food supermarkets fill the channel gap, improving Japanese consumer satisfaction and generating clear positive social effects During these years, GMS and specialty food stores have shrunk, seemingly leaving no place to buy groceries. However, thanks to the development of small food supermarkets, food shopping has become more convenient than before. Japan classifies people who live more than 500 meters on foot from any food retail outlet and those over 65 who cannot drive as 'food desert' residents. From 2005 to 2015, the number of elderly people over 65 with shopping difficulties grew by 30.4%, and their share of the total population rose from 20.1% to 26.3%. However, due to the growth of small food supermarkets, the proportion of the population within the defined 'shopping difficulty' area actually fell from 26.4% to 24.6%. Surveys also show that the proportion of Japanese residents who believe measures are needed to address food shopping inconvenience has declined, from 48.3% in 2011 to 38.3% in 2017. In metropolitan areas where small food supermarkets have grown most rapidly, this proportion dropped even more sharply, from 50% to 27.9%, indicating improved satisfaction with food shopping convenience. This improvement in consumer satisfaction with food shopping convenience occurred against a backdrop of a significantly larger base and proportion of people with shopping difficulties, and a contraction of over 40% in both GMS and specialty food stores. It can be seen that the development of small food supermarkets not only filled the gap left by the shrinking GMS and specialty stores but also enhanced consumer comfort and satisfaction. To some extent, this shows that small food supermarkets, which concentrate high-frequency daily food purchases, have improved the efficiency of retail and even Japanese society as a whole, producing positive social effects. We believe that it is precisely the changes in Japanese society, especially the decline in average household size, that have driven the continuous contraction of GMS and the gradual rise of small food supermarkets. IV. China's social changes are similar to Japan's, and the GMS format will enter a long night 1. China's average household size is also on a one-way decline According to data from the Ministry of Public Security, since the reform and opening up, China's average household size has been continuously declining, from 4.66 people per household in 1978 to 3.10 in 2016. According to the 'China Family Development Report' released by the National Health and Family Planning Commission in 2015, the average household size had already fallen to 3.02 people in 2015, and the 'three-person household' is moving toward a 'two-person household.' The same report predicts that the number of households in China will exceed 500 million for the first time by 2040 and remain around 500 million by 2050. According to UN projections, China's population will be slightly lower than today, at 1.35 billion by 2050. If these two figures are used, the average household size in 2050 will be about 2.7 people, meaning that the average household size will continue to decline in the future. This trend is even more pronounced in cities. For example, in Beijing and Shanghai, the average household size for registered residents is 2.53 and 2.68, respectively. These figures only account for registered residents; if the large migrant populations in these cities were included, the numbers would be even lower. It is clear that China's social trends are similar to Japan's. We believe that as the average household size continues to decline, the GMS format will also shrink in China, while small food supermarkets will gradually rise. Signs of this are already emerging. 2. China's population is concentrating in large cities and urban agglomerations, approaching Japan's distribution China's population density reached 145 people per square kilometer in 2017, more than three times the world average and far higher than the US's 35 people per square kilometer. China's dense and concentrated population is similar to Japan's; for example, the eastern region of China has a higher population density than Japan. The success of Walmart's GMS model in the US is closely related to the relatively dispersed population, where low-density living suits large-scale, infrequent shopping trips. For high-density populations, the advantages of small food supermarkets become apparent. China's current urbanization rate is 58.5%, lower than Japan's 91.6%. It is a certainty that China's urbanization rate will continue to rise, further concentrating the population in several major urban agglomerations and increasing residential density, which is more suitable for the development of small food supermarkets. 3. China's e-commerce penetration is higher than Japan's, making GMS more substitutable China's e-commerce penetration has reached 19.9%, higher than the US's ~10% and far higher than Japan's 5.79% (Japan's smartphone penetration is only 39%, with many still using flip phones). Growth remains strong, with a 35.1% increase in 2017, compared to Japan's 6.0%. GMS stores allocate significant space to daily chemicals, home textiles, small appliances, and other categories with higher gross margins. However, these items have low purchase frequency and weak experiential value, making them highly substitutable by pure e-commerce channels. As e-commerce logistics systems, represented by JD.com, become more mature and scalable, e-commerce platforms can cover more low-priced general merchandise and offer faster, more convenient home delivery than self-pickup. Since e-commerce commission rates for such items are often lower than the entry fees charged by GMS, suppliers tend to prefer e-commerce for cost reasons. We believe that in terms of efficiency, cost, and experience, low-priced general merchandise is better suited to pure e-commerce channels, which will gradually replace the role of GMS for these products, further impacting GMS operations. 4. Current situation of China's supermarkets: growth slowing, large supermarkets beginning to decline The growth rate of supermarket industry sales has slowed significantly compared to a decade ago. In 2017, despite a favorable macroeconomic environment, large supermarket sales experienced their largest decline in history. The slowdown in large supermarket sales stems from two factors: first, a decline in the operating area of large supermarkets, which fell by 5.91% at the end of 2017; second, a deterioration in their operating performance. According to the '2017 China Chain Top 100' released by the China Chain Store & Franchise Association (CCFA), the average sales per store for large GMS saw negative growth of -2.3%. Overall, the decline of large GMS in China is beginning to show. Based on the above, we judge that the overall market size of GMS in China will show a long-term shrinking trend. V. Japan's leading supermarket company: GMS market share gains do not help performance; 'small food supermarkets' drive growth We have already described the decline of GMS and the rise of small food supermarkets in Japan over recent decades, and we judge that China will follow a similar trend over the next long cycle. China's supermarket brands are relatively fragmented with low market concentration. There is a general expectation that leading GMS companies will increase their market share, and perhaps a GMS company with ultra-high market share like Walmart or Costco will emerge. However, we judge that China's GMS market will continue to shrink. Even if market share in the GMS business increases, will it truly bring profits to companies? Under such market trends, can a strategy of expanding through GMS store openings still create giants like Walmart or Costco? To answer these questions, the operating conditions of Japan's leading listed supermarket companies are worth in-depth study, as they offer strong reference for judging the future development of China's listed supermarket companies. Based on the essence of efficiency and cost, and the Japanese example, our judgment is that China will never see a Chinese version of Costco or Walmart. 1. Under the industry backdrop, a significant increase in GMS market share does not help AEON's performance Since 2001, AEON, as Japan's leading supermarket, has continuously increased its market share in the GMS business, from 19.85% in 2001 to 53.69% in 2018, occupying half of Japan's GMS market. However, with the overall GMS business shrinking, the increase in market share did not lead to a proportional increase in revenue. Over 17 years, AEON's GMS revenue had a compound growth rate of 4.04%, with revenue expanding by 96%, less than doubling. In the last decade, AEON's GMS market share increased by 13%, but sales only increased by 221.2 billion yen, a growth of 7%. During the period of significant market share gains, sales barely achieved slow growth, but profits from its GMS business saw a substantial negative growth. According to AEON's GMS profit disclosures starting in 2011, AEON's GMS achieved a profit of 46.4 billion yen in 2011, but only 10.5 billion yen in 2018, with net profit margin falling from 1.71% to 0.34%. It even recorded a loss in 2015. Ito Yokado (part of the 7&I Group), the second-largest player, has had an even harder time in the GMS business, with profit margins consistently below 0.5% in recent years. Combining the situations of AEON and Ito Yokado, we can see that even for leading companies, despite significant market share gains, their growth and profitability are very limited against the backdrop of an overall shrinking GMS market. 2. Small food supermarkets lead AEON's growth Although the GMS business has been struggling, AEON's small food supermarket (MY BASKET) is on the rise. While expanding its scale, its profitability far exceeds that of the GMS business. Since 2001, AEON's small food supermarket turnover has risen from 416.2 billion yen to 3.23 trillion yen, an expansion of 7.76 times, with a compound growth rate of 12.83%. Its market share has also increased from 0.99% in 2001 to 7.24% currently. AEON has separately disclosed profits for its small food supermarkets since 2011. Due to large-scale mergers and acquisitions and expansion expenditures, net profit growth has been slower than revenue growth, with a 7-year compound growth rate of 3.34%. Compared to the GMS business, small food supermarket profits surpassed GMS profits in 2015, and operating profit margins also exceeded GMS in 2015, despite larger expansion expenditures. VI. Where is the next opportunity for China's supermarkets? We have analyzed that the driving factors in Chinese society are very similar to Japan's, and the future development of GMS and small food supermarkets is likely to replicate Japan's trends. At the same time, China is a relatively dynamic society, while Japan is relatively static. China's retail industry is still in its early stages of development. The changes in China's retail industry, like the development of e-commerce in China, will certainly come more rapidly than in Japan, and opportunities are born in this transformation. So where is the best opportunity for China's supermarket industry in the future? Based on the above analysis, we believe it is not in increasing GMS market concentration, but in the broad development prospects of small and medium-sized supermarkets with Chinese characteristics. Among these, it is likely that a scaled chain of small food supermarket brands will emerge. Combining Japan's situation and China's characteristics, we propose a small supermarket of 400-500 square meters, with a trade area radius of 300-500 meters, and an average ticket of less than 30 yuan, focusing on food. We believe this is a relatively suitable format for the Chinese market, and the store opening strategy should adopt a 'dense network' model. We expand on this below: 1. Chinese people have a rich dietary structure, requiring relatively more fresh produce categories China is famous for its diverse diet, and correspondingly, the variety of ingredients is greater than in other countries. AEON's MY BASKET small food supermarkets sell almost exclusively food, with SKU counts of 2,000-3,000, but fresh produce categories are not numerous. Japan has strong agricultural protection policies, so the variety of vegetables and fruits on the market is limited, and meat basically does not include lamb or duck. Overall fresh produce consumption in Japan is also smaller. In China, the variety of fresh produce should be greater, and more variety means greater management difficulty, requiring more space to display goods. Therefore, from the perspective of household ingredient diversity, China's small food supermarkets should be larger than Japan's. This chart shows changes in the proportion of food expenditure items in Japan over the past 50 years. 2. China's average household size is larger than Japan's, so purchase volumes may increase proportionally The consumer unit for supermarket shopping is the household, and the size of the household directly determines the quantity of goods purchased per trip. Japan's chain small food supermarkets typically range from 200-300 square meters, with AEON's MY BASKET around 150-250 square meters and a trade area radius of about 300 meters. With Japan's average household size of 2.3-2.4 people and an average ticket of 700-800 yen (about 40-50 RMB), considering China's average household size of 3.0-3.1 people, the quantity of goods purchased per trip in China should be about 1.3 times that of Japan. If MY BASKET is a successful replicable chain format, then from a purchase volume perspective, China's small food supermarkets should be 1.3 times the size of Japan's. 3. China is relatively younger than Japan, so per capita food consumption is higher Compared to Japan's super-aged society, China's age structure is younger, with a median age about 10 years younger, at 36.7 years. According to research on the relationship between dietary energy needs, age structure changes, and food consumption, people's food intake begins to decline around age 50, especially meat and poultry intake, which drops rapidly. Age structure changes also affect the proportion of the working-age population. Research shows that a 10% increase in the population aged 65 and over leads to a 5% decline in meat demand. Currently, China's population aged 65 and over is 11.4%, far lower than Japan's ~25%. Japan's elderly population share is much higher than China's, so from the perspective of per-household purchase volume, for a family of three, China's purchase volume should be higher than Japan's, and accordingly, small food supermarkets need larger operating areas. Based on the above analysis, we propose that the relatively reasonable area for a small food supermarket with Chinese characteristics should be: 200 (based on AEON's small food supermarket) * 1.3 (household size ratio) * 1.1 (age adjustment) * 1.5 (ingredient abundance adjustment) = 429 square meters. 4. The coverage radius of a Chinese-style small food supermarket should be within 300-500 meters According to Japan's Ministry of Agriculture, Forestry and Fisheries, a household more than 500 meters from a food retail store is defined as having difficulty purchasing food, and measures are provided for such people. With GMS covering a 2-3 km radius, 95% of residents in the area would be defined as having food shopping difficulties, so Japan's approach is to open more small food supermarkets rather than GMS. China's latest policies are similar. For example, according to the 'Beijing Street Commercial Ecological Configuration Standard Guidance' issued in 2018, by 2020, the 'basic guarantee' configuration standard will be achieved: each community should have no fewer than 2 vegetable retail stores and no fewer than 1 convenience store, with a service radius of less than 500 meters as residential supporting facilities. This shows that the policy direction in China and Japan is basically the same. Taking AEON's MY BASKET as an example, its service radius is about 300 meters. Therefore, combining the policy directions and mature cases of China and Japan, the service radius of China's small food supermarkets should be within 300-500 meters. 5. The average ticket for a Chinese-style small food supermarket should be around 20-25 yuan According to the Japan Ryutsu Shinbun, AEON's MY BASKET has an average ticket of about 700-800 yen, equivalent to 40-50 RMB. Since small food supermarkets are currently mainly developed in urban areas, considering the per capita income of core cities in China and Japan, the average ticket for China's small food supermarkets should be around 10-12 yuan. However, China's average household size is about 1.3 times that of Japan, and because China's convenience store and vending machine formats are not as mature as Japan's, small food supermarkets may sell some high-margin convenience store core products like cigarettes and drinking water. Therefore, we believe the average ticket for China's small food supermarkets should be in the 20-25 yuan range. Although the average ticket is low, the convenience of small food supermarkets encourages customers to increase purchase frequency, while also ensuring freshness. VII. China's small food supermarkets are in their infancy Currently, there is no leading brand or company in China's small food supermarket segment; the scale is generally small, which corresponds to the huge potential of this market. The players that best match our positioning for future Chinese small food supermarkets are Fresh Legend and JD.com's planned 'Four Seasons Preferred' (Siji Youxuan). Fresh Legend stores are mainly 200-300 square meters with around 1,800 SKUs. It is currently growing rapidly, with a hundred stores in Hefei, and recently completed a 300 million yuan Series B financing, valuing the company at 3 billion yuan. JD.com, which has already opened the medium-sized food supermarket 7Fresh, is also planning to open 'Four Seasons Preferred' to complement 7Fresh. According to reports, Four Seasons Preferred will be positioned at 300-500 square meters, with around 2,000 SKUs, serving a 500-meter radius. Sun Art Retail's 'Hema Xiaoma' is positioned as a smaller version of Hema and may also enter the small food supermarket segment. Daily food consumption is a trillion-yuan market, and the space for small food supermarkets is vast, with prominent scale effects. It is highly likely that a leading brand in small food supermarkets will emerge in the future. The biggest problem with opening small stores is logistics: small stores have limited scale, smaller delivery volumes per trip, and higher restocking costs compared to larger stores. We believe that JD.com, which has invested heavily in logistics over the years and has a strong nationwide logistics network, has a first-mover advantage in opening small food supermarkets at scale. Sun Art Retail, which has been cautious in opening large GMS stores in recent years, can leverage its existing RT-Mart GMS stores and supply chain to extend into small food supermarket operations. Small food supermarkets are in their infancy, and we will wait and see how they develop in China. VIII. Summary Any change in the retail industry is essentially an improvement in the efficiency of retail activities, and this process inevitably leads to a redistribution of retail channels. Consumers will always gravitate toward more efficient channels, leading to channel redistribution. Social changes cause dynamic shifts in the procurement efficiency of various channels; for example, the development of smartphones has greatly improved the efficiency of online shopping. For the supermarket industry, long-term social trends will also drive changes in the efficiency and benefits that various supermarket formats provide to consumers. The Japanese supermarket transformation analyzed at length in this article is a good example. It is crucial to anticipate trends early and achieve rational retail format layout. Due to the low profit margins in the supermarket industry, investing in GMS expansion may achieve short-term growth, but the long-term risks are very high. This article, using Japan as an example, provides a forward-looking analysis of the relationship between long-term social changes in China and the development of the supermarket industry. We judge that the biggest opportunity in China's supermarket industry lies in small food supermarkets, while Hema and 7FRESH will form the main players in large and medium-sized food supermarkets. The risks for large GMS are high, and even an increase in market share is unlikely to significantly boost performance. We hope this article can serve as a catalyst for discussion and provide a reference for retail practitioners and investors.