Bernie Sanders, who is running for president in 2020 gave a speech a few days ago. He makes a claim which goes like the following:

Today the world's richest 26 billionaires, 26, now own as much wealth as the poorest 3.8 billion people on the planet, half of the world's population.
Link to video

This is a surprising figure, at least to me (though maybe it shouldn't be.) I'm just wondering if it's true.

Edit: My question has been marked as a potential duplicate. I can't really click "Yes, that solved my problem" because they are two different claims. One claim is about the wealth comparison of 5 billionaires to the poorest half of the world's population, and the other claim is with regard to 26 billionaires. However if anyone thinks it's a duplicate, I'm happy to have the question closed.

  • 5
    I have made an Excel spreadsheet summing up the worth of billionaires at Forbes (forbes.com/billionaires), here: 414soft.com/billionaires.xlsx. Free to use. Here is probably a good place to investigate further: data.worldbank.org, resources from Oxfam can be found here: oxfam.app.box.com/s/f9meuz1jrd9e1xrkrq59e37tpoppqup0/folder/…
    – Jacob Oscarson
    Jun 24, 2019 at 12:49
  • 84
    Likely, everyone here owns more than the poorest 1 billion people combined. Statistics like this can be deceiving.
    – Sjoerd
    Jun 24, 2019 at 23:15
  • 11
    I'd even argue that wealth calculations are altogether meaningless. You could be a multi billionaire who owns the equivalent of a train company just before air travel really took off and bankrupted them all. Ultimately wealth is highly volatile and at the top end, based on fuzzy valuations that can change very quickly.
    – Stephen
    Jun 25, 2019 at 4:41
  • 48
    @Stephen though you might be hard pressed to find an ex-Billionaire who lost his money and then went on to not have safe drinking water and had to take his kids out of school at 12 years etc. -If you look at 'i lost all my money' stories, they always are still rich afterwards, just because 1/1000th of a Billion is still a million. Any Billionaire can loose 99.9% of his money and still be a millionaire... most could lose an even higher percentage.
    – bukwyrm
    Jun 25, 2019 at 5:34
  • 14
    @bukwyrm Which is exactly what Sjoerd is talking about, really; it's surprisingly hard for people in developed countries to understand what "poor" means. If you lose all of your wealth overnight, you're still not poor (especially if you still have a job; it's much worse for e.g. retirees). An ex-billionaire who suddenly needs to pay taxes on the billions he no longer has will go bankrupt; but he still lives in a developed country, with all the massive infrastructure, work productivity etc. At the other end, people tend to look at "$1 per day!" and forget "also has cows and fields".
    – Luaan
    Jun 25, 2019 at 7:58

7 Answers 7


Short Answer: True.

Every year, in time for Davos (i.e. the Annual Meeting for the World Economic Forum), Oxfam releases a report about the state of inequality. Here's a link to their latest. Additional link to calculations down at the bottom of the page. https://www.oxfam.org/en/pressroom/pressreleases/2019-01-18/billionaire-fortunes-grew-25-billion-day-last-year-poorest-saw

I'm pretty sure the US senator was referring to that.

Now, a couple of things to note:

  1. If Oxfam's research were bad (it's pretty darn good), it would be discredited by many many business folk and of course top-rated economists - especially since they habitually present this at the most august fora they can get into. The WEF is up there, obviously.

  2. While Oxfam reports are "grey literature", serious academics like Sen, Stiglitz and Piketty have made the same point over the past 10 years, albeit differently nuanced - academic writing generally does not aim at the same shock factor that INGOs aspire to.

  3. There is a distinction between income inequality, wealth inequality... and other important inequalities (e.g. gender). Oxfam's 26 vs 3.8 Billion refers to wealth inequality.

Update: See footnote 1 for the link to the 26 richest (and their worth) and footnote 2 for the link to the poor (World Bank data-tables). Both notes on page 76 of the Oxfam report. https://oxfam.app.box.com/s/f9meuz1jrd9e1xrkrq59e37tpoppqup0/file/385579400762

  • 87
    Criticis of oxfams statistics usually focus on their choosen measurement, which is net wealth. Net wealth, in this case, means that the "poorest" are those heavily in debt. Their debt is also subtracted from whatever wealth anyone in the "poor half" has. In short, if you are completely broke but debt free, you are (by oxfam measurements) richer than the entire poorest billion.
    – Guran
    Jun 24, 2019 at 11:32
  • 70
    @Guran True, but of course the counterpoint (and I presume Oxfam's argument) is that ignoring debt doesn't make much sense. Taking out a million dollar loan does not make you a millionaire. In any case, ignoring those with net debt still gives a similarly small number. Jun 24, 2019 at 15:19
  • 41
    I think you are missing the point. If a newborn baby is richer than a billion people, does that tell us something about the newborn baby? Or the billion people? Also consider the possibility that someone like Donald Trump is a net debtor of, say, a billion dollars (probably true during one of his bankruptcies). Are we really holding up a person like that as a sign of inequality because he's so poor?
    – Brythan
    Jun 24, 2019 at 23:19
  • 22
    @Brythan Trump never went bankrupt. Several businesses he was involved with went bankrupt. Jun 25, 2019 at 2:54
  • 21
    @Acccumulation It makes sense when you consider the whole economy. For every debtor, there is a creditor - as long as the creditor can reasonably expect the debt to be repaid, the math works out with the creditor's net worth including the debt as a positive asset, and the debtor's as negative, adding up to zero. It still doesn't make much sense in the context of wealth comparison, mind - being able to afford a debt is again something that makes you relatively rich; you can use the debt to fund a venture and earn higher returns than otherwise. Many poor regions are poor because they lack that.
    – Luaan
    Jun 25, 2019 at 8:03

That statistic is at least plausible, and likely true. Children have essentially no wealth, so to start out the richest are already ahead of about 25% of the world (about 1.9 billion people) simply by having a positive net worth. The next 1.9 billion is made up of the impoverished adults, young adults that are just starting careers, working class populations, and middle class adults.

Wealth is misleading when used as a statistic because it's largely a function of saving over time, the rule of thumb for savings is that they double about every 7 years when invested wisely. The top 26 billionaires are people that have been extremely wealthy for decades. The thing to keep in mind with these numbers is that the average retiree will also have as much wealth as about 50-60% of that number, simply because most haven't had time to accumulate any wealth.

  • 18
    This answer was posted before it was migrated to Skeptics.SE. Here, it doesn't meet the standards. Please edit.
    – Oddthinking
    Jun 25, 2019 at 7:02
  • 7
    The statistic is considering worldwide population. The average retiree of the world will have not even close to 50-60% of the top 26 people's wealth, unless he happens to be in the US.
    – Gnudiff
    Jun 25, 2019 at 8:36
  • 5
    Even in the US the average retiree will be extremely far away from that. Jun 26, 2019 at 4:36
  • 5
    Wait, you are claiming that "wise investment" leads to long-term 10% return? I mean, "during a period of unprecedented world economic growth in history expansion invested in a country that won 3 major world conflicts without taking significant damage"? Ok. But, 2000 to 2019 was 4.9% annual in that same country. When talking 10-year returns, cherry picking a century and after-the-fact country is going to get you bad data.
    – Yakk
    Jun 26, 2019 at 16:49
  • 2
    "rule of thumb for savings is that they double about every 7 years when invested wisely" I doubt this is true for most people, obviously, you can claim that people that do not meet this doubling are not true Scotsmen. en.wikipedia.org/wiki/No_true_Scotsman Jun 26, 2019 at 22:00

Short Answer: True.

Oxfam's research is available on their website https://www.oxfam.org/en/research/economy-99 and is based on "Credit Suisse Global Wealth Data book 2016"

https://www.oxfam.org/en/pressroom/pressreleases/2017-01-16/just-8-men-own-same-wealth-half-world http://publications.credit-suisse.com/tasks/render/file/index.cfm?fileid=AD6F2B43-B17B-345E-E20A1A254A3E24A5

I have no reason to dispute this evidence at this time.

Long Answer: Lies, Damned Lies, and Statistics.

The methodology that Oxfam uses is net wealth NOT wealth. What this means is that when they talk about the bottom 50% - they're talking about people in net debt (think medical student) PLUS people with small positive sums (e.g. a rice farmer in China). They combine this wealth together, and they then compare this to the billionaires in question.

As an illustration, imagine we had 5 people in the world, with net wealth's as follows:

-2 -1 0 1 2

From this you can make the following claim:

The richest person has more wealth than the bottom 80% of the population!

However, you can also make the following claim:

The poorest person has more wealth than the bottom 60% of the population, including themselves!

How? Simple, because -2 + -1 + 0 = -3. -3 is smaller than -2. Hence the poorest person it richer than the bottom 60% even though he's part of the 60%. It's completely true... but also very misleading if you don't understand the methodology being used.

Looking at the numbers

If you look at the Global Wealth Databook, on page 148 you'll find the following summary:


0 1 2 3 4 5 6 7 8 9

(Wealth share, in percent)

-0.4 0.0 0.1 0.2 0.3 0.6 1.1 2.3 6.8 89.1

Note that the bottom 40% = -0.4 + 0 + 0.1 + 0.2 = -0.1.

So as long as your net wealth is above -0.1% of world wealth, that is -271.27Billion*, then you are richer than the poorest 40% of the world... even if you are deeply in debt and part of that 40%.

Also note that the bottom 50% = -0.4 + 0 + 0.1 + 0.2 + 0.3 = 0.2

0.2% of world wealth is ~ $542.54Bn. So if you get a group of people whose net wealth is higher than that (say the top billionaires) then you can claim they're richer than the bottom 50%!

This ignores the fact that the bottom 50% actually own far more - but they've got debts that 'net out'.

*(not sure that figure in $, but let's assume it's world wealth (pg 146) 271.270TRN*-0.1%)

  • 7
    Oxfam has directly addressed this criticism: it isn't as large as you suggest: It changes the top 1%'s share from 48.1% to 47.9% Jun 25, 2019 at 18:44
  • 2
    @TemporalWolf good link, but that is not the claim being addressed in this question. The question is about the top 26 people, not the top ~75 million.
    – NPSF3000
    Jun 25, 2019 at 19:12
  • 6
    @temporalwolf, fair enough but Oxfam insists on making these claims even after writing that article (2015). So if anything, it just proves that Oxfam is deliberately using misleading statistics.
    – NPSF3000
    Jun 25, 2019 at 19:40
  • 5
    I'm trying to understand why you're convinced this is misleading: again, even removing the negative net wealth group doesn't largely affect the sum. And the position that some significant fraction of that debt is held by rich people who 'net out' has yet to be backed by a source. Jun 25, 2019 at 19:56
  • 4
    @temporalwolf, Oxfam make a claim, and that is what is being addressed. If they made a different claim, with different methodology then that is what I'd be addressing. Take it up with Oxfam.
    – NPSF3000
    Jun 25, 2019 at 20:29

Oxfam published this claim based on calculations in a January 2019 briefing paper with a rather laborious title: Public Good or Private Wealth? Universal health, education and other public services reduce the gap between rich and poor, and between women and men. Fairer taxation of the wealthiest can help pay for them.

In particular, in Figure 3: Wealth of the bottom 50% of global population and the accumulated wealth of top 50 billionaires, 2017 and 2018:

Figure 3

The data for the calculations is sourced from Forbes (for billionaires) and Credit Suisse (for bottom 50%).

A note explains a minor caveat:

Wealth information from both sources are for two different months: March for Forbes and June and December for Credit Suisse in 2018 and 2017, respectively. Strictly speaking, this means that 26 (43) billionaires had as much wealth in March 2018 (2017) as half the population did in June 2018 (2017).


Others have pointed out the catch to subtracting the debt of the people with negative net worth from the wealth of people with moderate incomes. No need to repeat that.

Let me also add that such statistics depend greatly on how one calculates wealth, and the definition can make the results very misleading.

For example, you often read statistics like this page, https://www.givewell.org/international/technical/additional/Standard-of-Living, that says that economists calculate "the number of people living on under $1.25 a day at about 1.4 billion worldwide".

A little thought will show that that cannot possibly mean that their lifestyle is comparable to that of someone in America or Western Europe trying to live on $1.25 per day. I sincerely doubt that it would be possible to buy sufficient food to live for $1.25 a day, never mind other necessities of life.

Not to say that these people are not extremely poor. I certainly wouldn't want to switch places with them. But they're not THAT poor. They get these statistics by ignoring many forms of real wealth. Many of these people are subsistence farmers. They own land for farming, or live in a culture that pays little attention to the idea of legal title to land. What would be the market value of this land if it was in Los Angeles? By that standard some of these peasant farms would be quite wealthy. They may hunt animals for meat and use their skins for clothing. What is the market value of this? These things are not counted in these statistics. Etc.

On a less dramatic scale, you often see statistics contrasting the wealth of the richest Americans to the average Americans. According to this page, https://www.thebalance.com/american-net-worth-by-state-metropolitan-4135839, "The concentration of wealth in the U.S. continues to deepen as the top 1 percent of wealthiest U.S. households now holds 24 percent of liquid wealth. Non-affluent households, representing 70 percent of U.S. households, control less than 10 percent of the nation’s liquid wealth."

Of course rich people are richer than poor people -- duh -- but the situation is not as extreme as these numbers make it sound. The subtle key word in that statistic is "liquid". "Liquid wealth" is cash, stocks and bonds, and the like. It does not include houses, cars, appliances, furniture, and other such assets.

Suppose person A owns a house worth $200,000 with no mortgage, a brand new $30,000 car bought with cash, a $50,000 boat, and tens of thousands in other assets, but has only $5,000 in the bank, while person B lives in an apartment and doesn't own a house, doesn't own a car but rides the bus, and in general has few material possessions, but he has $50,000 in the bank. Would you say that B is 10 times as rich as A, because only the cash counts as "wealth"?

Realistically, most people spend a considerable percentage of their income on things that are quickly consumed: food, gas for the car, electricity, etc. After that they use their money to buy things that they want to use in their daily life: houses and cars and furniture and the like. It's only after they have these things that they put substantial money into savings and investments. (I'd say most Americans put too little money into savings, but that's another story.)

So suppose you have two people. A has earned $500,000 so far in his life. Of this he has spent $200,000 on things he's consumed, $250,000 on durable assets like a house and car, and he has saved $50,000. B has earned $750,000. Of this he has spent $250,000 on things he's consumed, $300,000 on durable assets, and he's saved $200,000. So B earned 50% more money than A, but his liquid assets are 400% of A's. (I just made up those numbers, I'm not claiming they represent any actual case histories, but I think they're not implausible. There are people out there in these general ballparks.) Yes, B is richer than A, but counting just liquid assets exaggerates the difference wildly.


Do the 26 richest billionaires own as much wealth as the poorest 3.8 billion people?

Taking the question absolutely literally: No

This figure is based only on net worth. Specifically the net worth of multiple different people subtracting the debts owed by one person from the assets owed by completely different people.

So the people in question do own significantly more wealth than it claims.

Further: if Bob down the street owes a million dollars and has no assets his debt doesn't cancel out my own assets.

This figure is reached by not only looking at net value, it also adds up the net value of multiple people.

26 richest billionaires between them own 1.4 trillion.

top two-fifths of the 3.8 billion people referred to in the Oxfam stat. That’s 1.4 billion people; between them, they are worth $2.2 trillion.

So that group does have more wealth than the 26 richest billionaires

The oxfam report reaches it's number by subtracting the money owed by people like Jérôme Kerviel from that 2.2 trillion.


let’s look at just the top two-fifths of the 3.5 billion people referred to in the Oxfam stat. That’s 1.4 billion people; between them, they are worth $2.2 trillion. And they’re a subset of the 3.5 billion people who between them are worth $1.7 trillion.

The first lesson of this story is that it’s very easy, and rather misleading, to construct any statistic along the lines of “the top X people have the same amount of wealth as the bottom Y people”.

The second lesson of this story is broader: that when you’re talking about poor people, aggregating wealth is a silly and ultimately pointless exercise. Some poor people have modest savings; some poor people are deeply in debt; some poor people have nothing at all. (Also, some rich people are deeply in debt, which helps to throw off the statistics.)


How is it that the US can have 7.5% of the bottom decile, when it has only 0.21% of the second decile and 0.16% of the third? The answer: we’re talking about net worth, here: assets minus debts. And if you add up the net worth of the world’s bottom decile, it comes to minus a trillion dollars. The poorest people in the world, using the Credit Suisse methodology, aren’t in India or Pakistan or Bangladesh: they’re people like Jérôme Kerviel, who has a negative net worth of something in the region of $6 billion.


The result is that if you take the bottom 30% of the world’s population — the poorest 2 billion people in the world — their total aggregate net worth is not low, it’s not zero, it’s negative. To the tune of roughly half a trillion dollars. My niece, who just got her first 50 cents in pocket money, has more money than the poorest 2 billion people in the world combined.

Or at least she does if you really consider Jérôme Kerviel to be the poorest person in the world, and much poorer than anybody trying to get by on less than a dollar a day.

There is massive wealth inequality but the oxfam methodology is misleading and this claim as worded is technically false. A subset of people from the worlds poorest 3.8 billion do in fact control more wealth than the 26 richest billionaires


wealth is not comparable to debt at a 1:1 rate. Bob and I could come to an agreement that he pays me 1 dollar and next week I'll owe him 100 trillion dollars, it would be meaningless beyond the total value of my net assets. Very little wealth would have been created or changed hands but by oxfams analysis I would then personally cancel out about half the worlds total wealth.

  • 4
    The problem here is you're choosing a different definition for the word "wealth" than was used in the original report. As usual with these types of debates, you're 100% right with one set of definitions, and 100% wrong with another set of definitions. The fact that you disagree about what core terms actually means makes any meaningful debate impossible.
    – barbecue
    Jun 26, 2019 at 19:33
  • 1
    your counter-example is a reductio ad absurdum, which is a reasonable approach in deductive reasoning, but we're not dealing with deductive reasoning here. I read Oxfam's argument and I see merely an oversimplification, not insanity.
    – barbecue
    Jun 27, 2019 at 14:08
  • 4
    @barbecue the insanity is that deciles 2-5 are populated by dirt farmers in the poorest countries in the world, but the bottom decile is populated by people from advanced democracies who happen to have student loans, be underwater on their car loan or mortgage, have credit card debt, and billionaires hanging out at Davos with the 38, except they happen to be leveraged right now. Negative equity is not a dirt-farmer problem, these do not belong with the others, and they distort the average to uselessness. Jun 27, 2019 at 14:20
  • 1
    @harper accusing the opponent of insanity is a tactic, not a meaningful argument.
    – barbecue
    Jun 27, 2019 at 14:23
  • 1
    @harper did you or did you not use the word "insanity" to describe the opposing argument? Simply "mentioning a word" is not the same thing as using that word to describe something. I could simply mention the phrase "pedantic twit" but that doesn't mean I'm calling you a pedantic twit. On the other hand, if I did call you a pedantic twit, that would be an argumentum ad hominem, not a valid argument against your claim. Basically, you're saying that pointing out an ad hominem argument is itself a logical fallacy, which is pretty silly.
    – barbecue
    Jun 27, 2019 at 15:34

We don't know

This claim is not based primarily on data but on interpolations of the underlying data. By interpolations, I mean that they look at actual data from one year or one country and then look at smaller clumps of data from a different year or country. They assume some level of proportionality and fill in values for the missing data.

Some quick examples. The Credit Suisse report on which the Oxfam claims are based makes claims about 2018. But the United States did not produce data for 2018. Its most recent report is 2016 (and apparently it produced reports every three years from 2007 to 2016). Andorra, Brazil, and Russia did not produce wealth distribution data at all. See tables 1-3 (sources) on pages 12-13 and 1-5 (data) on pages 15-16 of the 2018 Databook PDF. Or see the introduction on page 4 for a description of their estimation methods.

Sweden and Denmark each have only one or two sources of partial data. In the published year, we can say that the single richest household in Denmark had more wealth than at least 60% of the households in the same country. For Sweden, we can make the same claim about the richest adult and the poorest 60% of adults. Of course, we could also say that about a newborn baby, as the data shows that the poorest 60% in Sweden and Denmark in the years for which that statistic was reported had negative net worth in aggregate. From table 1-5.

It's interesting to compare Sweden and Denmark to the US by this measure. For the US, that was only true at the 30% mark. At 40% and above, the households had positive net worth (table 1-5 again). By this measure, the US had lower inequality than Sweden and Denmark in the measured years. This runs contrary to the normal perceptions of the three countries. I leave it up to you if this is because this statistic is flawed or if the normal perceptions are.

  • 12
    Is this (the first couple of paragraphs, at least) falling for the Nirvana Fallacy? I have no doubt the data collected is not 100% and not perfect. I have no doubt that assumptions need to be made for economic modelling. But is there any reason to consider this model to be insufficient for the purpose?
    – Oddthinking
    Jun 25, 2019 at 7:06
  • @Oddthinking I couldn't answer that without original research. For example, if we split the underlying data into two pieces, how well does one piece predict the other using their estimation method. What would that show? We don't know. My point here is simply that they are using estimation, but the claim doesn't show that. And, as I point out in the last paragraph, comparing the base data has odd results, which may mean that it is not equivalent. It's not just that the data is not 100%. They are basing the claim on something less than 50% of the underlying data.
    – Brythan
    Jun 25, 2019 at 7:17
  • @Oddthinking I invite you to read page 4 of the Databook on their methodology. They make estimates from estimates, average over them and use correlations between their estimates to create estimates for wealth based on income. Some gems: "The best source of data for this purpose is household balance sheet data, which are now provided by 49 countries, although 25 of these countries cover only financial assets and debts. (...) The results are supplemented by econometric techniques, which generate estimates of the level of wealth in countries that lack direct information for one or more years."
    – sgf
    Jun 27, 2019 at 11:52
  • "We use direct data on the distribution of wealth for 35 countries. Inspection of data for these countries suggests a relationship between wealth distribution and income distribution, which can be exploited in order to provide a rough estimate of wealth distribution for 133 other countries, which have data on income distribution but not on wealth ownership"
    – sgf
    Jun 27, 2019 at 11:52
  • "The high skewness of wealth distributions makes sampling error important. Non-sampling error is also a problem due to differential responserates – above some level wealthier households are less likely to participate – and underreporting, especially of financial assets. (...) To compensate, wealthier households are over-sampled in an increasing number of surveys, such as the US Survey of Consumer Finances and similar surveys in Canada, Germany, Spain, and several other EU countries.
    – sgf
    Jun 27, 2019 at 11:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .