《經(jīng)濟學(xué)人》題圖。

During the Spring Festival holiday, which this year lasted from February 15th to 23rd, China regroups and regathers. People cross the country on fast trains to join their families, watch dancing robots on TV and hand out red packets of crisp banknotes to younger relatives. But above all, they gather to eat. In a café in Fuzhou, a southern city, locals and tourists ate cheesecake and drank kombucha. One customer ordered wontons wrapped in “swallow skin” sheets, which mash together sweet-potato starch and pounded pork. “I really like eating,” said Yu Huan, another customer, who works in fashion in Shanghai. “It’s one of the ways I obtain happiness.”
今年從2月15日持續(xù)至23日的春節(jié)假期期間，中國舉國歡慶，團聚一堂。人們乘坐高鐵跨越山河與家人相聚，在電視前觀看機器人舞蹈，向年輕晚輩遞上裝滿嶄新鈔票的紅包。但最重要的是，他們圍坐共餐。在南方城市福州的一家咖啡館里，本地居民與游客品嘗著芝士蛋糕，啜飲康普茶。一位顧客點了用“燕皮”包裹的餛飩——這種薄皮由紅薯淀粉與捶打過的豬肉糜混合制成。“我真的很喜歡吃，”在上海從事時尚行業(yè)的顧客余歡表示，“這是我獲得快樂的方式之一。”

This year the National Bureau of Statistics (NBS) got into the spirit of things by revealing, for the first time, exactly how much Chinese consumers spend on food. The number emerged from a revision of the consumer-price index. The new weights imply that food (excluding dining out, booze and tobacco, with which it is often mashed together) accounted for 17.2% of household consumption last year. The equivalent figure for America was less than 8%.
今年國家統(tǒng)計局也應(yīng)景地首次公布了精確的居民食品消費數(shù)據(jù)。這一數(shù)據(jù)源于居民消費價格指數(shù)的權(quán)重調(diào)整。新權(quán)重顯示，食品（不包括外出就餐、酒類和煙草——這些常被合并統(tǒng)計的類別）占去年家庭消費支出的17.2%。而美國的對應(yīng)數(shù)字則低于8%。

These percentages confirm China’s passion for food. But they also have a less comforting implication. China may be far ahead of America in dancing robots and high-speed trains, but it still lags far behind on one of the oldest measures of economic development: Engel’s law. It states that as their income increases, people devote a smaller share of it to sustenance. This regularity, discovered almost 170 years ago by Ernst Engel, a German economist, is one of the “most enduring relationships in economics”, according to Richard Anker of the University of Massachusetts, Amherst. It can be used to predict food spending. But it can also be used in reverse, to infer incomes. Other things being equal, Engel declared, the share of outlays devoted to food is “the best measure of the material standard of living of a population”.
這些百分比印證了中國對美食的熱情。但它們也揭示了一個不那么令人寬慰的含義。中國或許在機器人舞蹈和高鐵領(lǐng)域遙遙領(lǐng)先于美國，但在衡量經(jīng)濟發(fā)展的最古老指標(biāo)之一——恩格爾定律上，卻仍遠(yuǎn)遠(yuǎn)落后。該定律指出，隨著收入增長，人們用于維持基本生存的支出占比會下降。這一由德國經(jīng)濟學(xué)家恩斯特-恩格爾在約170年前發(fā)現(xiàn)的規(guī)律，被馬薩諸塞大學(xué)阿默斯特分校的理查德-安克爾稱為“經(jīng)濟學(xué)中最經(jīng)久不衰的關(guān)系之一”。它既可用于預(yù)測食品支出，亦可逆向推演收入水平。恩格爾曾斷言，在其他條件相同的情況下，食品支出占比是“衡量人口物質(zhì)生活水平的最佳尺度”。

Engel discovered his measure in data painstakingly collected by others. Edouard Ducpétiaux, a Belgian jurist, tabulated the budgets of 199 households across all nine provinces of his country in the 1850s. Frédéric Le play, a pioneering sociologist, gleaned similar figures from 36 families across Europe, gaining their confidence through praise, small gifts and “interesting conversation”.
恩格爾的這一發(fā)現(xiàn)源于他人艱辛收集的數(shù)據(jù)。19世紀(jì)50年代，比利時法學(xué)家愛德華-迪克佩蒂奧統(tǒng)計了本國九個省份199戶家庭的收支賬目。先驅(qū)社會學(xué)家弗雷德里克-勒普萊則通過恭維、小禮物和“有趣的交談”贏得信任，從歐洲36個家庭中采集了類似數(shù)據(jù)。

Ducpétiaux and Le play had “delivered the pearls”, admitted Engel, “but not the string”. What tied the data together was the consistent relationship between dosh and nosh that he spotted. Reviewing the law 150 years later, Mr Anker found the lix was still easy to discern across over 200 countries. Even China’s NBS takes it seriously. “The Engel coefficient”, it said last year, is an “important indicator for measuring the standard of living of residents”.
“他們獻上了珍珠，”恩格爾坦言，“卻未提供串珍珠的線。”而將這些數(shù)據(jù)串聯(lián)起來的，正是他發(fā)現(xiàn)的收入與飲食支出之間穩(wěn)定存在的關(guān)聯(lián)關(guān)系。150年后重新審視這一定律時，安克爾發(fā)現(xiàn)這種關(guān)聯(lián)在200多個國家中依然清晰可辨。甚至連中國國家統(tǒng)計局也高度重視該指標(biāo)，曾于去年明確指出：“恩格爾系數(shù)是衡量人民生活水平的重要指標(biāo)。”

Several economists trust this measure more than they trust China’s official income figures. In 2014 Emi Nakamura and Jón Steinsson of the University of California, Berkeley, and Miao Liu of Boston College used Engel’s finding to cast doubt on the country’s growth and inflation statistics. They compared households in 2006 with those that reached a similar income two years later. They discovered that the later households were still devoting substantially more of their budgets to food. Perhaps they were not quite as prosperous as the official figures claimed.
多位經(jīng)濟學(xué)家對此指標(biāo)的信任度甚至超過對中國官方收入數(shù)據(jù)的信任。2014年，加州大學(xué)伯克利分校的中村恵美與約恩-斯坦松及波士頓學(xué)院的劉淼運用恩格爾的發(fā)現(xiàn)，對中國經(jīng)濟增長與通脹統(tǒng)計數(shù)據(jù)的準(zhǔn)確性提出質(zhì)疑。他們將2006年的家庭數(shù)據(jù)與兩年后達到相似收入水平的家庭進行對比，發(fā)現(xiàn)后者仍將預(yù)算的顯著更高比例用于食品支出。這表明這些家庭的實際富裕程度可能不及官方數(shù)據(jù)所宣稱的水平。

Engel’s law is also a source of concern for Adam Wolfe of Absolute Strategy Research, a consultancy. He points out that the official Engel coefficient (which includes spending on cigarettes, alcohol and dining out, as well as food) has mysteriously stopped falling, despite China’s reported growth. These items accounted for 29.3% of consumption in 2025, the same as eight years before. This “violation” of Engel’s law, Mr Wolfe argues, suggests that China has suffered a “severe development setback”.
恩格爾定律也引起了絕對戰(zhàn)略研究公司亞當(dāng)-沃爾夫的擔(dān)憂。他指出，盡管中國報告了經(jīng)濟增長，但官方恩格爾系數(shù)（包含香煙、酒類、外出就餐及食品支出）卻已神秘地停止下降。這些項目在2025年占消費支出的29.3%，與八年前基本持平。沃爾夫認(rèn)為，這種對恩格爾定律的“違背”表明中國遭受了“嚴(yán)重的發(fā)展挫折”。

But Engel’s law has a wrinkle: dining out. When people eat at a restaurant, café or stall, they are not just buying food. They are also paying for the cooking, washing-up and ambience. Mr Anker once did his own fieldwork to quantify this point. He bought noodles and steamed buns in street markets in xi’an, a city in western China. He also patronised McDonald’s and Outback Steakhouse in Massachusetts. Rather than eat the dishes, he weighed their ingredients, then estimated their cost. He calculated that the Chinese street food cost up to 30% more than a similar meal at home. McDonald’s cost 150% more. The steak: 233% extra.
但恩格爾定律存在一個復(fù)雜因素：外出就餐。當(dāng)人們在餐館、咖啡館或攤位用餐時，他們購買的不僅是食物，還支付了烹飪、洗碗和環(huán)境氛圍的費用。安克爾曾通過自己的實地調(diào)查來量化這一點。他在中國西部城市西安的街市購買了面條和包子，同時也在馬薩諸塞州光顧了麥當(dāng)勞和澳拜客牛排館。他沒有食用這些餐點，而是將食物拆解后稱量其中的食材并估算其成本。他計算出，中國街頭食品的價格比在家制作類似餐食高出30%；麥當(dāng)勞高出150%；而牛排則高出233%。

Yu’s law
余的定律

If restaurant meals are included in calculations of Engel’s law, the weight of food spending may be overstated. But excluding them poses the opposite danger. Awkwardly, the NBS did not disclose this month how much the Chinese spend on dining out. Nor did it provide a narrower measure of food consumption, excluding dining out, for the years before 2025. That makes it hard to know whether eating out has been propping up the Engel coefficient.
若將外出就餐納入恩格爾定律的計算，食品支出的權(quán)重可能被高估。但將其排除則面臨相反的風(fēng)險。令人尷尬的是，國家統(tǒng)計局本月并未披露中國居民在外就餐的支出金額，也未提供2025年之前剔除外出就餐的、范圍更窄的食品消費衡量數(shù)據(jù)。這使得人們難以判斷外出就餐是否在支撐恩格爾系數(shù)。

Figures from Wind, a financial-data platform, provide a clue. They show that restaurants and other “catering services” rose from 5% of consumption in 2017 to 7.4% in 2024 (the latest figure available). Such numbers can also be deducted from the official Engel coefficient to arrive at a narrower measure of past food spending. This calculation suggests that food’s weight was as high as 20.7% in 2017, well above the 17.2% for 2025 that the NBS has just revealed. In other words, if dining out is subtracted, food’s weight in Chinese consumption has continued to fall. The country has not broken Engel’s law after all.
金融數(shù)據(jù)平臺Wind的統(tǒng)計數(shù)字提供了線索。數(shù)據(jù)顯示，餐館及其他“餐飲服務(wù)”在消費中的占比從2017年的5%上升至2024年（可獲得的最新數(shù)據(jù)）的7.4%。這些數(shù)字也可以從官方恩格爾系數(shù)中扣除，從而得出過去更狹義的食品支出衡量值。這一計算表明，2017年食品權(quán)重曾高達20.7%，遠(yuǎn)高于國家統(tǒng)計局剛剛公布的2025年17.2%的水平。換言之，如果剔除外出就餐，食品在中國消費中的權(quán)重持續(xù)下降。中國終究并未打破恩格爾定律。

In Fuzhou, Ms Yu provides corroboration. She came to visit restaurants not family. She has tried seafood hotpot, peanut soup and local fish balls. “Food makes up the biggest part of my budget,” she confesses. But that’s no economic setback. She is limited less by her wallet than by her stomach. “As one person, I can’t eat that much,” she says. “So that’s why I stayed for five days.”
在福州，余女士的實例為此提供了佐證。她此行是為了去餐館吃飯，而不是探親。她品嘗了海鮮火鍋、花生湯和當(dāng)?shù)佤~丸。“食物占了我預(yù)算的最大部分，”她坦言。但這并非經(jīng)濟挫折。限制她的，與其說是錢包，不如說是胃口。“一個人吃不了那么多，”她說，“所以我才待了五天。”