Retail

There Is No Such Thing as “Chinese Consumption”

This summer, I crossed China on a diagonal. The big cities all look alike; their consumers do not. Decoding the August figures, city by city.

Introduction – The diagonal

All summer, I travelled across China on a diagonal: Beijing, Shanghai, Wuhan, Guangzhou, Shenzhen. What strikes you first is the similarity. The same elevated expressways spanning the neighbourhoods, the same new air-conditioned metros, the same malls with the same brands on the same floors. A Western executive who runs through these cities on a three-day business trip draws a natural conclusion: it is one and the same homogeneous market.

Think again. Beneath the twin infrastructure, ways of consuming diverge profoundly, even between neighbouring cities. Shanghai and Suzhou are 25 minutes apart by high-speed train. Guangzhou and Shenzhen, half an hour. In each of these pairs, the consumers are nothing alike.

The August retail figures have just come out. They are the perfect demonstration, on one condition: do not read them in aggregate.

I. Shanghai: read the composition, not the total

Let’s start with the figure making the headlines. Shanghai retail fell 6.7% in August, after -6.3% in July. Over eight months, -1.0%. The country’s historic number-one consumer city is in negative territory, while the national figure shows +0.4%.

Most readings stop there. That is precisely the mistake.

Break it down. Automobiles account for nearly 20% of retail sales at Shanghai’s major retailers, the highest share of any major Chinese city. In August, they collapsed by 38.8%. Over eight months: -19.1%. This single category drags the city’s total down by 4 to 5 points. In other words: most of the “Shanghai decline” is the decline of a single aisle.

The rest of the picture describes a city consuming normally. Food: +2.9% in August, +4.2% over eight months, which matches what anyone can see on the ground, with crowds at Sam’s Club. Cosmetics: +2.6%. Home appliances: +16.5% in August.

And part of the story is missing from the statistic itself. 社零 (shèlíng, total retail sales) counts only physical goods and food service. Tourism, exhibitions, shows, the night-time economy, everything that makes Shanghai buzz in summer, is left out. The Chinese have a phrase for this paradox: 体感热、数据冷, “the feeling is hot, the data is cold.”

II. Suzhou, 25 minutes away

Now take the high-speed train from Hongqiao. Twenty-five minutes later, you are in Suzhou: 13 million inhabitants, one of the wealthiest populations in China, and a unique fabric of champion counties: Kunshan, China’s top county by economic strength, Changshu, Zhangjiagang, all among the national elite.

Suzhou retail in August: around +3%, the best performance of China’s nine major cities. Above all, a structure with no weak spot: food up by double digits, food service around +9%, home appliances +10%, even jewellery in positive territory. Suzhou is the only major city where no category is plunging.

Why such a contrast with its neighbour? Four structural reasons. Suzhou’s retail depends far less on premium cars, the stricken category. Property costs two to three times less than in Shanghai: at comparable income, a Suzhou household has more discretionary income, the exact mechanism we documented in the City H note. Its statistical perimeter includes wealthy, high-spending counties, whereas Shanghai, 100% urbanised, has no such cushion. And its propertied middle class, built on industry rather than finance, is less exposed to income compression and the wealth effect (property and the stock market).

The high-speed train puts Suzhou 25 minutes from Shanghai. Statistically, they are ten points apart.

III. Guangzhou and Shenzhen: old money vs. new money

The same exercise works in the Pearl River Delta, with one additional variable: demographics.

Guangzhou is a city of long-established prosperity. A trading capital for centuries, with a dense family fabric, settled generations, retirees with comfortable pensions and accumulated wealth. People consume out of habit and culture: 食在广州, “Guangzhou is where you eat.” The figures confirm it: beverages +19.4%, solid growth in food service, and even cars in positive territory. The Cantonese consumer is well established, and spends accordingly.

Shenzhen is the exact demographic opposite. A city of migrants, with an average age of 32.5, eight to ten years younger than Beijing or Shanghai: the youngest of China’s megacities, populated by highly paid tech workers. On paper, the purchasing power is there. In behaviour, it is not: these employees live under the constant threat of layoffs in an industry that practises the 35岁危机 (the “age-35 crisis,” the age beyond which tech rarely hires); most are first-generation arrivals with no local family wealth or safety net, and a mortgage sized to today’s salary. The result: massive precautionary savings and hesitant consumption. Telecoms are booming (+35%, this is Shenzhen, after all), but total retail is barely positive and gold is falling.

Same province, half an hour by train. On one side, old wealth that spends with serenity; on the other, high earners who do not dare. A uniform “Greater Bay Area” retail plan misses both cities at once.

IV. The car case: when municipal policy manufactures the numbers

A legitimate objection: don’t these gaps simply reflect different consumer profiles? Not only. The purest demonstration is the car market, where a single national shock, the price war (-18.5% nationally in August), produces three opposite outcomes depending on a variable nobody looks at: municipal licence-plate policy.

In Beijing, plates are allocated by lottery. Demand is rigid: whoever finally wins a quota buys immediately, without waiting for the next price cut. The result: electric vehicles keep growing (+4.7% over eight months); only combustion cars are declining.

In Shanghai, combustion-car plates are sold at auction, at around 90,000 yuan at the peak, and the car mix is the most upmarket in the country. This system creates a double-trigger wait-and-see effect. Holders of combustion plates, anticipating their depreciation as electrification advances, rush to sell while the asset is still worth something. Buyers, meanwhile, no longer have any reason to hurry: the green (electric) plate is free, and the days of bidding quickly for fear of paying more for a plate the following year are over. The only price that moves is the vehicle’s, and it falls every quarter. Sellers in a hurry facing buyers in none: the exact configuration that crushes prices. The result: electric and combustion sales collapse together in value.

In Guangzhou, plates are easy to obtain and the local carmaker GAC irrigates the market: electric vehicles jump 28.7% and cars contribute positively to the city’s retail.

Three municipal regimes, three patterns of car consumption, three retail figures. And what plates do to cars, commercial licences, local subsidies and municipal industrial priorities do to other categories. The hidden variable in Chinese retail is not income. It is the local institution.

V. Nine cities, almost half a continent

Now step back. China’s nine leading cities (Beijing, Shanghai, Guangzhou, Shenzhen, Suzhou, Chongqing, Wuhan, Hangzhou, Chengdu) have a combined population of 187 million, equivalent to Germany, France and the Netherlands combined. Their annual retail, around 14,000 billion yuan, is comparable in order of magnitude to the combined retail of Germany, France and Italy.

In August, this “continent” spread across a ten-point range:

Five readings of this table.

✦ First, the range: ten points between Suzhou and Shanghai, more than the growth gaps between European countries.

✦ Food and beverages are holding up everywhere. Double-digit growth in Suzhou (+17%), Chongqing (+16%) and Hangzhou (+13%), and beverages at +19% in Guangzhou; the most modest gain is Shanghai’s, at +2.9%, but it is still positive. From one end of the continent to the other, the fundamentals remain solid: China eats, drinks and entertains. This universal base is the only constant in the table.

✦ Cosmetics are still holding, between +2.6% in Shanghai and +4.9% nationally in August. In an environment described as “in the doldrums,” beauty keeps growing wherever it is measured.

✦ The statistic does not count services, which structurally disadvantages cities like Shanghai. Tourism, exhibitions, shows, the night-time economy, everything that makes a Shanghai summer, is excluded from 社零. Beijing thus posts +4.5% in services while its retail declines. The cities furthest along in the shift to experience-based consumption are mechanically the most penalised by a thermometer designed to count objects.

✦ Finally, cars are the category where local policy manufactures the numbers, as shown above: positive in Guangzhou, resilient in electric in Beijing, collapsing in Shanghai. And it is the only case in the table where a city sees its heaviest category plunge: in the other eight cities, the weak spots are low-weight categories (office supplies, jewellery). Most of the “Chinese decline” comes down to one category, in one city.

The same category can therefore be your best and your worst market depending on the city: gold falls 9% in Shanghai and rises 12% in Beijing; telecoms are up 35% in Shenzhen and down 14% in Shanghai. No one reads Germany, France and Italy through a single figure, a single strategy, a single distribution plan. There is no reason to treat these nine cities any differently.

Takeaways for executives

1. Never budget China on the aggregate figure. August’s +0.4% is the average of a -6.7% and a +3%. An average of booms and collapses is not operational information: it is a statistical artefact.

2. Audit your organisation’s geographic bias. A China P&L run from Shanghai imports Shanghai’s pathologies (wait-and-see behaviour, overweighting of stricken categories) into its reading of the whole country. Ask yourself where the anecdotes that feed your executive committees come from.

3. Map the municipal institutions that govern your category. The licence-plate case applies everywhere: licences, local subsidies, commercial zoning rules. Two cities that are twins in infrastructure can be opposites in regulation, and it is regulation that makes the numbers.

4. Read the figures where your revenue is made, and in your category, never anywhere else. If 60% of your China revenue comes from five cities, it is those five cities’ statistical releases you need to track, not the national aggregate. And within those cities, your category: in the same month of August, telecoms were up 27.3% and brand stores down 10%. The sentence “Chinese consumption is in the doldrums” contains no useful information for your business; your category’s figure in your cities contains all of it.