Global Top, Local Bottom
Why, on distribution, someone is always telling you to look down
@Mythogen Engine · Oct 21, 2026
“Snatching Rice from the Beggar's Bowl”, Part 2 | Previous: “Suffer If You Don't Work?” | Next: “The Fraud You Can See, the Numbers Nobody Measures”
Not resenting the rich, not hating the poor: struggle is not competition, and in the end it reaches every floor.
The ranking and the end of the month
Ah Yan is thirty-two and works as an office administrator at a small company in Hong Kong, earning HK$22,000 a month. The title on her business card looks better than the number on her payslip.
One evening she tries an online “global income ranking calculator”. She enters her monthly pay, and the screen says she ranks in the top quarter of the world's adults. She takes a screenshot to send to a friend, then doesn't. That afternoon her landlord had told her the rent goes up next year; after rent, her pension contribution and her credit card, what is left this month is not enough for a new phone.
Under a video about the wealth gap, she comments that she has nothing left by the end of the month. The first reply: “You're already rich by global standards. Kids in Africa don't even have food, and you're complaining?” The second: “Keep this up and people from the Third World will flood in, and you won't even have this job.”
One says she is too rich to complain; the other says she is about to be dragged down by the poor. The two replies point in opposite directions, but they share one thing: both tell her to look at people poorer than her. Neither asks where the money went.
1. Where you stand
Workers in Hong Kong, Taiwan and developed economies generally occupy an awkward position: top tier by global income, lower-middle by local living standards.
The 2026 World Inequality Report lets us measure that position. The poorer half of the world's adults earn on average about €5,100 a year, or about €425 a month; the threshold for the global top 1% is about €250,000 a year. The report uses purchasing-power-parity euros, that is, euros adjusted so that prices are comparable across countries; the same applies below. The top 10% globally, about 560 million adults, take 53% of global income; the bottom 50%, about 2.8 billion people, get 8% (World Inequality Report 2026).
Ah Yan earns HK$22,000 a month, slightly below Hong Kong's median monthly employment earnings of HK$23,600 (C&SD data, reported by UDN), or about HK$260,000 a year. To compare this with those numbers, Hong Kong dollars must first be put on the same scale. At the World Bank's private-consumption PPP rate, about HK$6.06 equals one international dollar (World Bank data), so her annual income is worth about 44,000 international dollars; converted roughly at euro-area PPP, that is about €30,000 in PPP terms, about six times the average income of the poorer half of the world. That already makes her richer than most people on earth, yet still short of the global top 10% — the report puts that threshold at about €65,500 a year, and she is not even halfway; as for the top 1% line, her income would need to be about eight times what it is now. She is neither at the top nor in the Third World, but stuck in the middle — high on the global ranking, low on her own city's lease.
The calculator at the start of this essay was built by this site (global income ranking calculator). Checking it while writing, I found the first version was wrong: it stitched together data on different bases into one curve and treated the report's euros as US dollars, which briefly “promoted” Ah Yan into the global top 10%. After the fix, she lands in the top quarter. Same person, same pay; only the definitions changed, and the ranking moved by more than ten percentage points. A ranking looks precise to the decimal point, but behind it is a chain of choices: which currency, which basket of prices, adults or all people, before or after tax. Any income ranking, including this site's, should be asked which definitions it uses before you look at the rank it gives you.
Hong Kong has an extra layer. A family that bought early may rank high globally by wealth, and that wealth is not entirely illusory: once the mortgage is paid off there is no rent, which amounts to an extra monthly income; sell up to emigrate, downsize or move somewhere cheaper, and paper gains become real money — what Hong Kong people call “making it ashore”.
But making it ashore has two conditions: getting in early, and finding a way out. If you sell but still have to live in Hong Kong, you sell high and buy back just as high; most of the gain just goes round in a circle on paper. Late buyers are another story: prices have fallen nearly 30% from their 2021 peak, and negative-equity cases rose to more than 40,000 in Q3 2024, a 21-year high (RFA); only after prices stabilised did they fall to 4,356 in Q2 2026 (HKMA).
The same flat is wealth for those who got in early and found the exit; for those who bought at the peak on a full mortgage, it may be a negative number. This kind of wealth depends on when you got in and which door you leave by, not on what is left at the end of the month — as with another, much larger paper fortune discussed below.
Taiwan's version comes from another report. Allianz's Global Wealth Report 2026 shows Taiwan's net financial assets per capita in 2025 at €164,470, about NT$5.94 million, fifth in the world after the US, Switzerland, Denmark and Singapore (Focus Taiwan). When the news came out, people online split into two camps: one said Taiwan turns out to be so rich, the other said they felt nothing of the kind. Finance writer Wang Po-ta took the report apart on Facebook and argued both camps were right, because the “feeling nothing” has three layers.
The first layer is distribution. The richest tenth of Taiwanese hold about 60% of financial assets (cnyes). Using the report appendix's figure of 61.8%, Wang calculated that the richest tenth average about NT$36.7 million each, while the other nine-tenths average about NT$2.52 million, less than half of NT$5.94 million. And NT$2.52 million is still a mean; the median of those nine-tenths can only be lower.
The second layer is debt. Taiwanese household debt rose 6.3% in 2025 to €714.2 billion (Business Today). The report counts mortgages as liabilities but does not count homes as assets; a young family that has just taken on a NT$10 million mortgage with a few hundred thousand in savings may have negative net financial assets in the report. What they feel every month is the mortgage side.
The third layer is stock versus flow. The NT$5.94 million measures a stock built up over decades of saving plus rising asset prices, not income earned each year. Allianz itself notes that of every €5 of new household wealth worldwide in 2025, about €4 came from rising asset prices, while new global savings actually fell 5.4% (Global Views). Taiwanese households' securities rose 17.7% in a year, while deposits rose only 4.9% (Business Today). Those with shares saw their wealth take off with the market; for those saving from wages, fifth place has little to do with them.
The Hong Kong and Taiwan versions say the same thing: rankings measure stocks and averages, while life is lived in flows and medians. Ah Yan's place on the global ranking and the balance in her account at the end of the month are two different numbers.
2. The same money does not buy the same life
Purchasing power parity solves one problem and creates another. It evens out prices across countries so that a dollar in Hong Kong can be compared with one in Bangladesh; but it uses a national average basket of goods, weighted by the whole society's spending. Ah Yan's basket is not the average basket. At the HK$9,000 to HK$12,000 a month that people online often quote for a 200 sq ft flat, rent already takes 40% to half of her pay; the average basket does not give rent more weight because she hands half her income to a landlord.
Only over a longer span can you see how that basket has grown heavier. In 2000, Hong Kong's median monthly employment earnings were HK$10,000 (Hong Kong data reprinted in China's National Bureau of Statistics yearbook); for February to April this year it is HK$23,600 (UDN), a rise of about 140% over two decades. Over the same period, the Rating and Valuation Department's private domestic price index, with 1999 as 100 (Syfe), stood at 323.2 in June this year and peaked at 398.1 in September 2021 (Sing Tao): prices more than tripled, and at the peak came close to quadrupling.
The rent story is more complicated. In the same index series, rents reached 205.8 in June this year, a record (Sing Tao), about double 1999 — a rise similar to wages. But that doubling started from an already very high base: wages and rents rose together, so rent's share of income did not fall, while the threshold for buying moved further away.
A plate of baked pork chop rice tells a different story. In 1984, Fairwood sold baked pork chop rice for HK$8.50 (HK01); on a 2010 Fairwood menu, the pork chop rice with a drink was HK$35; in 2026 a Fairwood rice plate with tea or a soft drink, such as the chilli chicken rice, costs HK$58, or HK$64 with the slow-cooked soup. Over the later part of that period, median employment earnings rose from HK$11,000 in 2010 (Hong Kong data reprinted in China's National Bureau of Statistics yearbook) to HK$23,600 this year. In other words, a month's median pay bought about 310 plates in 2010 and about 400 now. In Hong Kong, food has actually become relatively cheaper; what has outrun wages is property.
But there is a gap in this number that statistics cannot see. Many Hong Kong people feel that fast food has become more expensive, worse-tasting and smaller; the HK$8.50 plate of 1984 was cooked from real ingredients in a kitchen, while today's rice plate tastes more like a pre-made product heated in the shop. A price index measures the price of a plate, not what is missing from it. The Taiwan Railway pork chop bento mentioned below, still NT$60 but with a shrunken chop and fewer sides, is the same thing: the price hasn't changed, what's inside the box has. So “a month's pay buys 400 plates” is only half right — it buys plates, not necessarily the plate of those days; the big fall in quality and quantity is not counted.
Taiwan's numbers make the gap even clearer. In 2025, average regular monthly earnings of Taiwanese employees were NT$47,884, up 3.09%, the largest rise in 26 years; after inflation, real regular earnings rose 1.4% (TechNews citing DGBAS). But the same statistics show nearly 70% of employees earning below the average; median regular earnings are only about NT$38,319 (PTS).
Over a longer span, Taiwan's wages look more like standing still. In December 2018, real regular monthly earnings after inflation were NT$38,235, lower than NT$38,398 in 2001 — back 17 years (TVBS). Over the same period, Taipei's house-price-to-income ratio rose from about 6 in 2002 (Global Views) to 15.57 in Q1 2025 (Economic Daily News): two decades ago, a Taipei household could buy a median home with six years of income, spending nothing; now it takes fifteen and a half.
Memories of living costs often live in a lunch box. People online recall that a Fulong bento once cost NT$50–60 and now costs NT$90 (United Daily News); Taiwan Railway's classic NT$60 pork chop bento is still around, but the chop, sides and rice have all shrunk, and it is sold in limited numbers (Taiwan News); office workers in the north have long complained that they cannot find a bento under NT$100 (China Times).
The bento went from NT$50–60 to nearly NT$100, a rise of over 50%; real wages are about where they were in 2001; and a Taipei home went from six years of income to more than fifteen. Wages keep up with the official average price level; they do not keep up with bentos, leases and house prices.
Prices have two sets of books too. Taiwan's consumer price index rose only 1.66% in 2025, but taking June that year as an example, CPI rose 1.37% year on year, a 51-month low, while food rose 2.82% and eating out 3.44% (TVBS); the housing index and rent index both hit record highs. House prices are another order of magnitude: from 2018 to 2024, the Ministry of the Interior's housing price index and the Sinyi index rose an average of 6.58% and 8.12% a year, far above overall CPI and rents; Academia Sinica calls the gap between official figures and what people experience “felt prices” (NOWnews citing Academia Sinica).
Official inflation measures the average person's inflation. The lower someone's income, the larger the share of eating out and rent in their spending, and the closer the inflation they feel is to bentos and leases rather than the average 1.66%. Pay up 3%, bentos and rent up 3% to 4%, house prices up 7% a year — on the global ranking, their place does not move at all; in their own city, a home drifts further away.
Price and wage levels in Hong Kong and Taiwan differ a great deal: both are high in Hong Kong and both low in Taiwan. What the two share is not the level but the direction: wages follow average inflation, while living costs follow rent and eating out. Global rankings use purchasing power to even out the world, but the lives of Ah Yan and that Taiwanese worker are counted in leases and menus.
3. How much of this position did you earn?
Ah Yan is richer than most people in Africa mainly not because she works harder than they do, but because of where she was born.
The economist Branko Milanović ran a calculation: suppose everyone has only two characteristics they can barely control — which country they live in, and which income class they belong to within it. These two alone explain more than half of global income inequality; his conclusion is that effort or luck can play only a limited role in the global income distribution (Milanović, 2015). He calls the income gap that comes with birthplace a “citizenship premium” or “citizenship rent” (Demos interview).
The idea has an uncomfortable symmetry. We generally dislike the children of the rich — within one country, inheriting wealth by birth is seen as unfair. But across countries, “inheriting” an income several times higher by birthplace is something few people feel they need to justify.
This is not meant to make Ah Yan feel guilty. The citizenship premium describes two sides of the same thing: if most of her position is luck, then most of the position of someone her age in Africa is luck too. That being so, explaining either side's situation by “hard work” or “laziness” does not hold up.
4. The first trap: guilt
“You're already rich by global standards, why complain?” sounds like a moral reminder; its real function is to make people shut up.
Its logic: as long as someone in the world is worse off than you, you have no right to complain. Follow this logic and no one can ever complain, because someone is always worse off; in the end the only person entitled to complain is the poorest person in the world — who is not online.
The effect of this trap is to swap a local question of distribution for a global competition in misery. Ah Yan's question was: why is there no money left after rent? Those who answered changed it to: how much better off are you than Africans? The first question points to landlords, land prices, wages and the tax system; the second points only to her own conscience.
Guilt will not give that child in Africa an extra meal either. If Ah Yan holds back her complaints, her rent will not go down and Africa will not get more food; the only change is that no one is left asking about distribution at home.
5. Factfulness: the most respectable gag order
The guilt trap has a most respectable version, written up as an international bestseller.
In 2018, Swedish physician Hans Rosling, with his son Ola Rosling and daughter-in-law Anna Rosling Rönnlund, published Factfulness. Bill Gates called it one of the best books he had ever read and announced he would give a copy to every US college graduate that year (Wikipedia). The book's argument is simple: the world is much better than you think; you think it is getting worse because ten instincts distort your judgment.
The book gets some things right. It rejects dividing the world into “us” and “them”, and reminds readers that people in poor countries are making progress too — on this, this essay agrees. The problem lies in what it cannot see, and that happens to be Ah Yan's end of the month.
First, its ruler stops at the top. Factfulness divides the world into four income levels: under $2 a day is Level 1, $2–8 Level 2, $8–32 Level 3, and over $32 Level 4 (Wikipedia). Level 4 has no upper limit. Ah Yan, on HK$22,000 a month, earns about $90 a day and is Level 4; Musk, whose fortune grew by more than $500 billion in a year, is also Level 4. A ruler whose markings stop at $32 can measure how Level 1 rises to Level 2, but cannot measure that inside Level 4 one person has nothing left after rent while another becomes richer overnight than 3.8 billion people combined. Where the distribution problem is most extreme is exactly this ruler's blind spot.
Second, it counts only what can be counted. Income, life expectancy, vaccination rates and school enrolment are all improving; but a plate of baked pork chop rice getting smaller and worse, a Taiwan Railway bento shrinking, a month's pay buying more plates with less on each — none of this appears in its charts. When progress is defined as what can be measured, decline that cannot be measured is as good as not having happened.
Third, it turns structural problems into psychological bias. Factfulness says people think the world is getting worse because ten instincts, such as the “negativity instinct” and the “fear instinct”, are at work. As a framework for checking headlines, this is useful; but applied to Ah Yan it becomes something else: her frustration with her lease and wages is renamed a cognitive bias to be corrected. The problem is not landlords, land prices and distribution, but her head.
Fourth, its poverty line is itself contested. The book's progress story relies on the World Bank's extreme-poverty line of about $2 a day. In 2019 the anthropologist Jason Hickel wrote a rebuttal to Bill Gates arguing that the line is absurdly low; using a more reasonable line of $7.40 a day, the number of poor people has risen rather than fallen since 1981, and the share has stayed around 60% (The Guardian). Hickel's calculations were in turn disputed by Steven Pinker, Max Roser and others (Reason); but the dispute itself shows that how much better the world is getting depends on where you draw the line.
Finally, look at who is recommending it. The person who gave this book to every US graduate is one of the richest people in the world. That does not make the book wrong, but it shows one thing: the story that “the world is getting better” suits those at the top particularly well — it directs attention to progress at the very bottom rather than to the share at the top. Hickel describes Gates and Pinker as going so far as to argue that, since the same forces bringing the rich enormous wealth are also eliminating poverty, people should not complain about rising inequality (The Guardian).
“The world is getting better” and “my end of the month is getting worse” can both be true. When the first is used to deny the second, data becomes a tool for silencing people — and the most respectable kind.
6. The second trap: fear
The second reply has a fuller version that circulates widely online: rich countries send medicine and food to poor countries; death rates fall and populations explode, but no matching industries develop; poor countries keep producing poor people, who then flood into rich countries and drag down their living standards.
The story is vivid, but several key links in it run against the data.
First, the poor are not growing in number. On the $3-a-day poverty line, the number of people in extreme poverty worldwide fell from about 2.3 billion in 1990 to about 831 million in 2025 (World Bank); on the old $2.15 line, the share fell from nearly 38% in 1990 to an estimated 8.5% in 2024. Of course, the World Bank's lines are low; critics note that at $5 a day four in ten people are still poor, and that excluding China, the number of poor in sub-Saharan Africa has risen (CADTM). “The whole world is getting better” is not the whole truth; but neither is “poor countries are breeding themselves ever poorer, ever faster”.
Second, fertility in poor countries is falling too. Sub-Saharan Africa's fertility rate peaked near 6.8 in the late 1970s and has fallen since (World Bank data); Africa as a whole was about 4.0 by 2024 (UN World Population Prospects 2024). Slower than Asia, but in the same direction.
Third, aid is too small to explain rich countries' difficulties. In 2024, official development assistance from OECD Development Assistance Committee members totalled $212.1 billion, just 0.33% of their gross national income (OECD). A country giving away a third of one per cent of its income is not enough to leave its own workers with nothing after rent.
The most ironic thing about this story is its own mechanism. It says the poor have many children because they lack social security and rely on their children in old age. Then the answer should be social security — once people no longer need many children to secure their old age, fertility falls. The fear story points at the poor, but its own mechanism points to protection.
The same logic has a local Hong Kong version: tightening welfare on one side, importing foreign labour on the other (see Part 1 of this series, “Suffer If You Don't Work?”). Both start from the very bottom, while land prices and assets stay untouched.
7. Where the money is
Both traps tell people to look down. Look up, and the numbers are these: the richest 10% globally own 75% of the world's wealth, the poorest 50% only 2%; the richest 1% own 37%, more than eighteen times the poorer half (World Inequality Report 2026). New wealth is shared the same way: from 1995 to 2021, 38% of new global wealth went to the richest 1%, while the poorest 50% got 2% (World Inequality Report 2022).
Most of this wealth exists as assets. In the US, the richest 10% of households hold 87.4% of household stocks and funds; the top 1%'s share rose from 42.8% in 1989 to 50.2% in 2026, with the rise concentrated in the top 0.1% (Federal Reserve Distributional Financial Accounts, compiled). Rising productivity, rising profits and rising share prices all end up in the hands of those who own shares.
Beyond wealth, income is flowing the same way. US Bureau of Labor Statistics data show that in Q2 this year, the share of nonfarm business output paid to workers as compensation was just 52.8%, the lowest since records began in 1947; for most of the postwar period it stayed between 60% and 66% (BLS). Over the same period productivity has been rising: up 2.9% year on year in Q1 this year, in step with the surge in AI investment; the job platform Indeed's research arm suggests these productivity gains may be shifting from workers to investors and business owners (Indeed Hiring Lab). Efficiency is up, while workers' share has fallen to its lowest in more than seventy years.
Taiwan has a similar curve. Employee compensation as a share of GDP was above 50% in the early 1990s and fell to 43.1% in 2024, a record low; in the same year, operating surplus, which reflects corporate profits, rose to 35.06% (Economic Daily News). Explanations differ. DGBAS attributes it mainly to industrial structure: in capital-intensive industries such as semiconductors, much of the value added goes to equipment depreciation, so even with high engineers' pay, labour's share is low (TechNews); Academia Sinica academician Hsieh Chang-tai argues that a long-weak New Taiwan dollar has shifted the gains of growth to the owners of exporting firms (Storm Media). Other research finds that the gap between real wages and productivity in 2002–2014 came mainly from export prices falling relative to consumer prices, rather than from changes in labour's share (Lin Yi-ling and Yang Tzu-ting). The explanations vary, but the direction of the numbers is not in dispute: labour's share of the economy's gains has never been smaller in recent years.
The most extreme example came in June this year. With SpaceX's listing, Musk's fortune briefly passed $1 trillion; Oxfam calculated that at that point he alone would be richer than the poorest 46% of the world combined, about 3.8 billion people, whose total net worth is about $890 billion (Oxfam). His fortune had grown by more than $550 billion over the previous year (Oxfam Ireland); soon after the listing it fell back with the market, and the title of trillionaire lasted only a short time (IBTimes).
Over $500 billion more in a year came from valuation, not labour; the fall did not require him to do more or less of anything either. This is financialised wealth: it can exceed the combined wealth of 3.8 billion people overnight and lose a chunk within weeks, and neither has anything to do with how much work he did that year.
The same analysis adds one more line: a tenth of Musk's fortune would be enough to end extreme poverty worldwide (Oxfam). Improving the lot of the poorest does not require Ah Yan to be downgraded. The room is at the very top, not in the middle.
8. The other side
In fairness, two objections deserve to be taken seriously.
The first comes from moral philosophy. The philosopher Peter Singer and others argue that by global standards, the middle class in rich places is genuinely rich, and so has a duty to help those poorer; “the money is at the top” cannot be a reason for the middle to do nothing. The argument has force: Ah Yan is not Musk, but what she spends on takeaway each month might be a month's income for a family in extreme poverty.
The second comes from economics. Globalisation has indeed exposed workers in some rich places to competition. The “elephant curve” of economists Lakner and Milanović showed that from 1988 to 2008, incomes of the global middle rose sharply while the lower-middle classes of rich countries sat in a trough, which was for a time read as stagnation for those classes.
But that reading was later questioned. The UK's Resolution Foundation reanalysed the same data and argued that the claim of general stagnation among the lower-middle classes of rich countries does not hold; the problem lies mainly in especially unequal growth in the US (Resolution Foundation); the US think tank PIIE also noted that the trough comes largely from the collapse of the Soviet Union and Japan's stagnation, not from globalisation itself (PIIE).
The two objections can stand together, and both point to the same place. The pressure of globalisation is real, but on whom it falls and how hard depends on how each country distributes; the middle does have a duty to those poorer, but that duty and the top's duty to everyone differ in scale by a factor of far more than a thousand. Admitting that you are relatively well off does not mean shouldering the share that belongs to those at the top.
The problem is that how each place distributes is increasingly out of its own hands. Capital finds its own way out: factories can move, profits can be shifted to low-tax places; even after weighing political risk, it still flows globally. People cannot move that fast: they have rent, children's schooling, parents to care for. So any place that tightens on its own, whether through antitrust, labour protection or taxes on capital, hears the same line: tighten, and capital leaves (on how the tax base is hollowed out as a result, see section 13 of Part 1). In 2021, more than 130 jurisdictions agreed to a 15% global minimum corporate tax (the OECD's “Pillar Two”) precisely to close this route; in January 2026 the OECD agreed a “side-by-side” arrangement exempting US multinationals from most of the rules (KPMG, EY). Even a 15% floor had to open a door for the largest source of capital.
But one thing cannot move: land. Land prices, public housing and public markets are exactly what a local government can act on without waiting for any global agreement.
9. Protections dismantled
At this point it is easy to conclude: if the problem is distribution, the answer is to hand out more money. But Hong Kong once had another answer: not handing out money, but holding down the cost of living structurally. In 2024, 29.4% of Hong Kong's population lived in public rental housing and 15.3% in subsidised sale flats, together nearly half (Housing Bureau). This was a universal protection on living costs: a ceiling on private housing, a ladder onto the market, and no need for the government to hand out cash to every household.
Taiwan took another road. Government-built social housing totals about 127,000 units (CNA citing the National Land Management Agency, end-July 2026), against a total housing stock of about 9.576 million homes (Liberty Times): under 1.5% of the stock. Taiwan relies on home ownership and family: the household home-ownership rate in 2025 was 83.95%, a 16-year low (Economic Daily News citing DGBAS). One relies on public housing, the other on ownership; when a Taipei home takes fifteen and a half years of income, both roads are starting to clog.
How Hong Kong's protection was dismantled, how income limits screen out ordinary workers, and why the means test can measure only payslips is the subject of Part 3 of this series. Here, one point is enough: holding down living costs requires structure, and the biggest item in that structure is land rent. Dismantle the structure and patch it with cash handouts, and the money ends up flowing to land rent. The question to ask is who collects that rent.
Closing: one pie, cut thinner every year
Back to Ah Yan's two replies. One tells her to feel guilty about her position, the other to be afraid for it; both tell her to look at people poorer than her, so her gaze stops at children in Africa and migrants flooding in, and she never asks where the money went: to that €250,000-a-year line, let alone to the man who gained more than $500 billion in a year.
The question is not whether to compare upward or downward. The whole structure sends money to the top, and the share left below shrinks every year; yet the media, society and everyone in it naturally compare only with the people beside them: the neighbour upstairs, the neighbour downstairs, colleagues, newcomers, families on welfare. However they compare, they are fighting over the same pie, cut thinner every year. This is not competition but zero-sum struggle: every extra bite you take is a bite your neighbour loses; and those who took most of the pie are never part of the comparison.
Even the people the public sees as “the rich” are mostly not at the top. Small and medium business owners, factory owners and shopkeepers in traditional trades are “above” in workers' eyes; but they too pay rent, and depend on the goodwill of landlords, banks and platforms, their profits whittled away layer by layer by rent, commissions and price pressure from big clients. Workers think the boss is stingy, the boss thinks workers don't try hard enough, and each treats the other as their problem; in fact both are stuck on the same pie, just sitting in different places.
Once the structure is clear, the questions change: productivity has risen so much, why haven't wages kept up? Assets have appreciated so much, why has it all landed with those who own assets? Improving the lot of the poorest needs only a small part of the top, so why should it be borne by the middle and the bottom fighting each other?
Population size was never the reason Ah Yan has nothing left after rent. Distribution is.
“Snatching Rice from the Beggar's Bowl” series: Prologue | Part 1 “Suffer If You Don't Work?” | Part 2 “Global Top, Local Bottom” | Part 3 “The Fraud You Can See, the Numbers Nobody Measures” | Part 4 “Those Who Leave and Those Who Arrive” | Final part “After the Counting”
The next essay asks: why would a protection that worked for decades fail for no stated reason, while official figures cannot say why?
References
The full list of sources is linked inline throughout; Chinese-language sources keep their original links. Key references: World Inequality Report 2026 and its Technical Notes (thresholds in PPP euros, 2025 prices, adults 20+, pre-tax national income); World Inequality Report 2022; Milanović, 2015; US BLS labor share, Q2 2026; Oxfam, 2026; Rosling, H., Rosling, O. & Rosling Rönnlund, A.: Factfulness, 2018; Jason Hickel in The Guardian, 2019; Housing Bureau: Housing in Figures 2025; KPMG and EY on the Pillar Two side-by-side package (January 2026).