$2 a day

by John Q on September 21, 2012

I’m trying to find information on the effects of the 1990s welfare reform (surprisingly difficult, suggestions welcome) I came across this NYT article by Jason de Parle which included the following striking result (link added.[^1]

Luke Shaefer of the University of Michigan and Kathryn Edin of Harvard examined the share of households with children in a given month living on less than $2 per person per day. It has nearly doubled since 1996, to almost 4 percent. Even when counting food stamps as cash, they found one of every 50 children live in such a household

The result is striking because of the $2 figure, which is derived, not from a US poverty line, but from the World Bank Poverty line for developing countries. These children aren’t just poor by American standards – they would be considered poor in sub-Saharan Africa.

[^1]: Why does the NYT link internally for things like “Social Security” but not to the studies it is quoting?



Marxy Marx 09.21.12 at 10:16 am

“Why does the NYT link internally for things like “Social Security” but not to the studies it is quoting?”

Because linking to external studies is unlikely to increase web traffic.


InfoPump 09.21.12 at 10:17 am

Why does the NYT link internally for things like “Social Security” but not to the studies it is quoting?

Links are automatically generated from a dictionary of links.


ajay 09.21.12 at 10:20 am

Why does the NYT link internally for things like “Social Security” but not to the studies it is quoting?

Because it thinks it will encourage readers to click through to other NYT pages, boosting traffic and thus ad revenues. Linking to an external site, they think, is like inviting your readers to go away and not come back.


ajay 09.21.12 at 10:22 am

On the study, I am amazed that it’s actually possible to survive on $2 a day in the US – simply from the point of view of getting enough calories and protein to keep yourself alive.


Tim Worstall 09.21.12 at 10:43 am

John, sorry, but this isn’t an entirely accurate number. For the usual reason about US poverty statistics.

SIPP income measure we use includes labor
market earnings, pension and retirements,
cash income from public programs (but not
in-kind transfers), asset income (dividends,
interest and rents) and reported income
from friends and family members outside
the household (including child support),
and from informal sources.”

It is simply a fact that most of the (almost certainly too little in total) poverty alleviation that the US does comes in in-kind transfers and the EITC through the tax system. The SIPP income measure excludes all of these. Section 8, Medicaid, SNAP, EITC etc etc. In terms of the aid to the poor that it does include it is only the cash welfare payments, the very thing that paper is exploring the changes in.

Yes, they do another calculation which includes SNAP. But not any of the other in kind programs.

Just for clarity this is not a statement that the American poor are just fine. It is only to point out that the statistics being used are not comparable with the poverty statistics of any other country. Because everyone else measures poverty after all tax and benefits, whether in kind or cash. The US, for some perverse reason, measures poverty before the effects of much that is done to alleviate poverty.


rf 09.21.12 at 11:01 am

“Surprisingly difficult, suggestions welcome”

I remember reading some things a number of years ago by this person on that general area (research interests seem relevant)


Although it’s more on the policy side than the economics.
I can’t say I’m too surprised by the £2 figure, wasn’t Gina Rinehart lobbying for something similar in Australia? And I can’t imagine life on welfare in any part of the south of Britain being much more lavish. (What is it £10 a day?) I think it’s to grasp how broke some people are at the minute


afinetheorem 09.21.12 at 11:15 am

This strikes me as misleading. Read the original article (it’s only 5 or so pages long, and essentially does nothing except pull the low end of income out of SIPP with no discussion of the sample size for these types of incomes, how data was collected, etc.) I have lived, twice, in neighborhoods that are objectively among the poorest urban districts in the country. A colleague of mine taught in Pine Ridge, usually listed as the poorest county in America. Even in this places, I think you will struggle to find anybody with living standards compatible with $2/day income.

Two reasons for this. First, the measure is $2 a day or less in cash income for at least one month. This is going to pick up a lot of people in between jobs, or moving, etc; I have certainly had months in the past few years with zero income for those reasons. Second, as noted above, there has been a massive transition toward non-income transfers (and not just SNAP) since the Clinton-era reforms, so it’s not even clear the measures can be compared.

This isn’t to say that poverty doesn’t remain a problem – it clearly does. But, for one, W. J. Wilson’s measure of “extreme poverty”, where your neighborhood has 40% or more of its households under the poverty line, is lower than it was in the early 1990s even accounting for the financial crisis and its effects. So for the really worrisome type of poverty, it’s not obvious that things have become materially worse in the past 15 years.


faustusnotes 09.21.12 at 12:21 pm

My brother was offered a job in 1994, in the midlands of the UK, for 1 pound an hour. He was living in Devon, and was expected to move across the country for it. I think he even had to scrabble to find some excuse not to, in order to avoid getting into trouble with social security (which I recall was 59 pounds an hour, 20% more than his wage at the job would have been). I guess a pound an hour in the UK in 1994 would be about 2.50 now (can’t be bothered with inflation calculations). So about $3 an hour?


faustusnotes 09.21.12 at 12:21 pm

I meant to say .. but $3 a day I find hard to believe.


chris 09.21.12 at 12:39 pm

@6,7: I can’t get your numbers to make sense. Maybe you meant to say 59 pounds a week? (But that would be ~50% more than the wage, unless it was a really long hours job.)


bianca steele 09.21.12 at 12:59 pm

The US, for some perverse reason, measures poverty before the effects of much that is done to alleviate poverty.

It would be difficult to plan for alleviating poverty otherwise, I’d think.

Even the paragraph Tim Worstall quotes above is misleading, because it doesn’t list sources like food banks and clothes giveaways.


ajay 09.21.12 at 1:33 pm

9: It would be difficult to plan for alleviating poverty otherwise, I’d think.

Surely you need both? If you’ve got A, who’s making $10 and getting another $60 in benefits, his poverty doesn’t need alleviating as much as B, who is making $25 and getting nothing in benefits. But if you measure pre-benefits, A looks much worse off than B.


bianca steele 09.21.12 at 1:41 pm

I don’t see why the issue, about the US (according to Tim Worstall) not measuring poverty statistics the same way the UN (or whoever) does, has any relevance. I see two arguments here against the stats as they’re collected: one is that there is no poverty in the US so the stats are misleading, the other is something like “I’m pretty sure there are lots of programs that aren’t counted in that.”


ajay 09.21.12 at 2:20 pm

I think it was probably intended as a response to my disbelief that anyone could physically survive in the US on $2 a day, by pointing out that they are actually getting some other stuff that isn’t counted.


Chris Bertram 09.21.12 at 2:26 pm

But presumably, a subsistence farmer in sub-Saharan Africa with some crops and a goat is also getting “some other stuff that isn’t counted”, namely, some crops and a goat?


ragweed 09.21.12 at 2:53 pm

@CB #13 – that is one of the big issues with poverty stats in places like sub-Saharan Africa. There are subsistance farmers living on $2 a day who can grow their own food and live in houses they made themselves from local forest materials et. al. , none of which gets counted in the $2 per day cash economy. Then there are urban slum-dwellers living on $2 per day (or less) who really have no resources other than that $2 per day. The quality of life difference between the two can be substantial.


LizardBreath 09.21.12 at 3:03 pm

I used to live in Samoa, and that sort of effect is very visible there — it shows up in economic statistics as one of the very poorest countries in the world, but subsistence farming there is very practical and successful, which means most people are living much better than their cash income would suggest. It’s still a very poor country, but not horrifyingly so at all. (This applies only to 1993-94. I haven’t kept up how things are there currently at all.)


Ed 09.21.12 at 3:10 pm

As the other commentators have noticed, when you see the $2 a day income figure, or something similar, it is just capturing people who don’t live in the cash/ credit economy. For the U.S., this is news but the news is that there is an increasing amount of people who live by barter/ subsistence. This is not how the figure its usually spun.

Yeah, these people are poor, but its not like living on $2 a day in Manhattan.


EB 09.21.12 at 3:10 pm

Schafer and Edin seem to have used a methodology that did not distinguish between those so newly without income that they are not enrolled in SNAP, versus those who are. If you have no income and file for SNAP benefits, you will get a minimum of about $150 per month per person for food. Not to say that it’s a walk in the park to live on $5 per day, but far different from $2 per day.


Tzimiskes 09.21.12 at 4:05 pm

Well, I have a few citations that might be appropriate that I looked at when researching for an article on whether there is any real world evidence for the culture of dependency so beloved by the American right.

I’m not sure if you were only looking for publicly accessible pieces but what I have follows:

The most appropriate is probably:

The State of the Social Safety Net in the Post-Welfare Reform Era [with Comments and
Discussion]Author(s): MARIANNE P. BITLER, HILARY W. HOYNES, CHRISTOPHER JENCKS, BRUCE D.MEYERSource: Brookings Papers on Economic Activity, (FALL 2010), pp. 71-147
From the abstract:”We find that use of food stamps and noncash safety net program participation have become signif- icantly more responsive to the economic cycle after welfare reform, rising more when unemployment increases. By contrast, we find no evidence that cash wel- fare for families with children is more responsive, and some evidence that it might be less so. We find some evidence that poverty increases more with increases in the unemployment rate after reform, and none that it increases less. We find no significant effects of reform on the cyclical responsiveness of food consumption, food insecurity, health insurance, household crowding, or health.”

Evaluating Welfare Reform in the United StatesAuthor(s): Rebecca M. Blank Source: Journal of Economic Literature, Vol. 40, No. 4 (Dec., 2002), pp. 1105-1166

The Anthropology of Welfare “Reform”: New Perspectives on U.S. Urban Poverty in the Post-Welfare Era
Author(s): Sandra Morgen and Jeff Maskovsky
Source: Annual Review of Anthropology, Vol. 32 (2003), pp. 315-338
(I have to recommend something that suggests it is deconstructing the hegemonic discourse)

Welfare Transitions in the 1990s: The Economy, Welfare Policy, and the EITCAuthor(s): Jeffrey GroggerSource: Journal of Policy Analysis and Management, Vol. 23, No. 4 (Autumn, 2004), pp. 671-695

Hopefully this is on topic and will save you (or some poor grad student) some time slogging through J-stor.


Tim Worstall 09.21.12 at 6:07 pm

“I don’t see why the issue, about the US (according to Tim Worstall) not measuring poverty statistics the same way the UN (or whoever) does, has any relevance.”

One point of relevance would be that people do in fact compare US poverty figures with those of other countries. Without, umm, noting that rather important difference.

“These children aren’t just poor by American standards – they would be considered poor in sub-Saharan Africa.”

” I see two arguments here against the stats as they’re collected: one is that there is no poverty in the US so the stats are misleading, the other is something like “I’m pretty sure there are lots of programs that aren’t counted in that.”

I can even see two more arguments as well.

“We’re spending hundreds of billions a year on poverty and we’ve still got the same number of poor people. Thus spending on poverty doesn’t work and we should stop”.

This would be a very bad idea of course but not measuring the poverty alleviated by poverty alleviation efforts does leave one open to that argument. I even recall PJ O’Rourke making that very argument.

Or: “12% of Americans are living in poverty. When elected President I will double the EITC, issue 1 million more Section 8 housing vouchers and expand Medicaid”.

That was, roughly speaking, the John Edwards plan for poverty reduction. That hypothetical re-election campaign would have been pretty difficult when he did those things, spent hundreds of billions more on poverty alleviation (and I would support those policies myself quite possibly) and he managed to move the dial on that 12% of Americans in poverty by precisely not one single person.

What we would really like to do is measure poverty (or even inequality if you would prefer) both by market incomes and after all taxes and benefits. Everyone, including the US, produces measures by market numbers. Only the US does not produce the second set of figures. And we’d really like both so that we can see how much poverty is alleviated by what is spent on poverty alleviation.


bianca steele 09.21.12 at 6:29 pm

The argument for welfare reform was that there were people getting cash payments who could and would find work, and be better off, if they were cut off, and that there would be no significant hardship because all of them could find work if they wanted to. Whether or not hardship increased under the same definition as before seems like a perfectly good measure to me. But I don’t need to tell you that.


TheSophist 09.21.12 at 7:10 pm

From the Shaefer and Edin paper: When we consider SNAP benefits as equivalent to dollars, this reduces the number of extremely poor households
with children by about half. We estimate that SNAP currently saves roughly 1.4 million children from extreme poverty.

To be honest, I clicked through to the paper because I was incredulous that anybody (let alone 4% of households) could be living in that kind of extreme poverty in the US, and I assumed that there must have been a mistake in the way the data was being reported. But no, S and E do indeed say that. (Or at least talk about percentages of the overall set of households living in poverty, and math does the rest). If they’re close to right, this is horrifying.


Salient 09.21.12 at 8:19 pm

On the study, I am amazed that it’s actually possible to survive on $2 a day in the US – simply from the point of view of getting enough calories and protein to keep yourself alive.

In rural counties around here, food stamps (SNAP, WIC) have been steadily replacing cash resources. We’ve been doing a fair job in this state at shaming and intimidating state politicians out of reducing access to or benefits from food stamps, in part because there are whole counties where over half the population relies on them. Thankfully “food for poor families” is enduringly popular here, at least. But the program needs to be expanded dramatically in order to provide what it aims to.

It’s buried in the brief, but it looks like my local area’s microcosm is comparable to the national scenario: As figure 1 demonstrates, the number of households in extreme poverty shot up considerably between late 2008 and early 2011 when only cash income is considered. In contrast, when SNAP benefits are considered as equivalent to cash, the prevalence of extreme poverty remained virtually unchanged during the period after ARRA.

Worth noting the other big increase in SNAP benefits happened when welfare was butchered in the 90’s. The replacement of generalized benefits with food benefits is a decades-long trend.

As for rent, the other necessity, there’s a lot we tried to infer from interviewing the migrant student population. There are a lot of people living in their friend’s rented trailers, and a lot of people living in probably-condemnation-grade trailers that a farm owner planted in a corner of a lot to make some cash playing landlord. A few folks are apparently using food stamps to defray the cost of rent and other necessities, sometimes getting half the face value of the stamps in doing so. I don’t mean outright fraud, just fuzzy social living. You might live rent-free in a friend’s trailer and contribute to the groceries, without keeping track of whose food is whose. And there are a number of people who move around multiple times per year to find a new very low rent place willing to take a chance on them. (Generally you pay some upfront cost, try to get your Section 8 subsidy straightened out, fail to do so, make rent for a few months off whatever cash work you can get and move whenever you exhaust your landlord’s patience.)

Still, I remain about as mystified as you. How in the hell do people survive this level of awfulness? …


James Conran 09.21.12 at 8:31 pm

I found this working paper by Theresa Anderson, Katharine Kairys and Michael Wiseman an excellent aid to understanding what actually happened following the 1996 reform:



Bruce Wilder 09.21.12 at 8:38 pm

There is some pretty extreme poverty in the U.S. Maybe, not quite sub-Saharan Africa poor, but certainly poorer than anywhere in Western Europe. In many rural areas scattered across the country, subsistence farming (and drug abuse) keeps alive a hidden poverty that’s pretty intense. Beasts of the Southern Wild may be a more truthful portrayal than any abstract statistics.

The Yazoo delta of Mississippi has some of the poorest people anywhere in the so-called developed world, and government aid administered by a state government only Scrooge could love, is of little help.

But, I wonder just how many people, even in big cities, because of deliberately faulty bureaucratic design, end up living on $2/day. It is all fine and dandy to pontificate about how poverty should be measured after government aid programs, but from my personal experience as a volunteer trying to get help for some seriously jeopardized people, getting any aid at all can be a major, possibly insurmountable problem, for people, who are a bit adled by mental illness or drug abuse or just the stress of extreme poverty or abandonment by family, or who simply lack proof of residence or a mailing address.

People do fall out of the economy’s structure, and, given that government policy is to maintain a structure that has some room for, at most, only about 90% of the population, trying to help the 5% or so, for whom there’s no place, and never conceivably will be a place, is a hopeless task. Even if they manage to get help, thanks to the collaboration between neo-liberals and conservatives, that help comes with an expiration date, after which help becomes that much more difficult to obtain.

Anyway, that’s my long-winded way of saying that that $2/day figure doesn’t strike me as implausible, though, from a statistical analysis standpoint, it also represents the ragged edge of measurable.


Keshav 09.21.12 at 8:49 pm

One source on the effect of the 1990s welfare reforms is this paper by Moffitt and Scholz: http://www.nber.org/chapters/c11969. They document the change in the distribution of transfer income using SIPP data from 1984, 1993, and 2004.


LFC 09.21.12 at 9:37 pm

Re the OP — the one-in-50-children living in households whose income is less than 2$/day per person (even counting food stamps as cash). At least where I live, a smallish can of Coke in a convenience store costs more than $1. Because as ajay indicates upthread, you can’t physically survive in many or most parts of the US on $2/day, those families will be having to eat at soup kitchens, feeding centers, etc., and/or relying on people giving them money (which, as gifts, presumably are not counted as income in the relevant studies) and/or on non-cash expedients (e.g. barter, growing some food) as suggested above. (Otherwise the children in these families would not be simply hungry; they would be unable to survive at all.)


John Quiggin 09.21.12 at 11:12 pm

There’s USDA data on food insecurity which is broadly consistent with the results above. The relevant categories are “low food security”, which means getting by with help from soup kitchens and the like, and “very low food security” which means actually going hungry. The good news, if you like is that even in households with very low food security, children almost always get enough to eat, if not a nutritious diet



gordon 09.22.12 at 12:24 am

Maybe not as good as some of the suggestions already made, but there are a few EPI publications which might be useful:

Max Sawicky’s “The End of Welfare? Consequences of Federal Devolution for the Nation”

Bernstein and Chapman’s report “Falling Through the Safety Net”

H. Boushey on transition to work

Mishel and Shmitt



gordon 09.22.12 at 12:46 am


peggy 09.22.12 at 1:46 am

As another indication that some people in the US are very, very poor, since 1990 the life expectancy of white women without a high school diploma has decreased five years. This decline rivals the catastrophic seven-year drop for Russian men in the years after the collapse of the Soviet Union.
Curiously enough, the life expectancy of black women without a high school diploma has not decreased and now matches white women. Perhaps 73-74 years represents a new First World minimum standard.


John Quiggin 09.22.12 at 3:54 am

@Peggy I just saw that too – it’s horrifying


Tom Geraghty 09.22.12 at 8:00 am

CBPP has a bunch of stuff on welfare reform:



PeterC 09.22.12 at 12:42 pm

Important to remember that although as income poor as many in sub-Saharan Africa, the poor in the US have always been relatively better off in terms of rich begging, prostitution and other criminal opportunities to improve their lot. Indeed, if worse comes to worse many are even able to enlist. Those skills learned often of use in entrepreneurial criminal activity when they have finished their service.

What has happened to the johnquiggin blog site? Simply technical problems,or is it under attack?


Witt 09.22.12 at 1:40 pm

MDRC has done an enormous amount of research on welfare reform. See: http://www.mdrc.org/subarea_index_8.html

And: http://www.mdrc.org/project_8_11.html


Witt 09.22.12 at 1:44 pm

The Urban Institute also has a slate of papers marking the 16th anniversary of welfare reform: http://urban.org/16th-Anniversary-of-Welfare-Reform.cfm


Witt 09.22.12 at 1:50 pm

Unfortunately Public/Private Ventures went out of business this summer. I think they had also done some work on welfare reform (more on low-wage workers generally) but I can’t find their archives and am not at the office at the moment, so can’t check my own.

Mathmatica Policy Research has a large number of papers on welfare reform.


PeterC 09.22.12 at 1:52 pm

As far as poor getting sufficient calories … sadly garbage bins abound with plenty of discarded food, as do food courts abound with unfinished portions.

Something I expect never to forget is the sight of a man in his sixties or older moving from table to table “cleaning up”. That was in the early 1960s.


Witt 09.22.12 at 2:04 pm

Regarding the $2/day figure, it bears noting that there is in some cases a VERY large gap between policy and practice. In my state, reaching a welfare caseworker on the phone is near-impossible — you often have to call 15-20 times, their voicemail is often full or not set up, and they often don’t return calls.

Why this matters is that the system *also* requires an enormous (and ongoing) level of proof — pay stubs, doctor’s appointment slips, hourly wage records, etc. etc. If you are two working adults and you need childcare for your children, but one of you works in a hotel and the other in a restaurant, there may come a week when one of you is changing jobs and the other has his hours cut because it is a slow time.

As a result, you may have your child-care benefit cut because one spouse is “not working” and therefore must obviously be available to stay home with the kids. This may happen on a Thursday afternoon heading into a holiday weekend, meaning that you then have to scramble to figure out some kind of childcare for the following week, meantime asking your brand-new boss for time off so you can go in and argue with the case worker in person (due to aforementioned phone issues) to try to get the childcare benefit reinstated.

In the meantime you have a cash crunch and may not be able to afford food — or the time it takes to identify a food bank, confirm its hours of operation, travel there, and stand in line.

Alternatively, the system may simply fail to work — the transfer of your transportation benefit into your account may not happen, and it may take you 5-7 days to straighten it out. In the meantime, of course, you still have to get to work, so then you are spending your scarce cash money on tokens and not on food.

All of which is to say: It is completely plausible to me that there are people whose lives are sufficiently marginal that they regularly experience bouts or even long periods of exceptionally impoverished life.

Most of the discussion in this thread is on TANF, or Temporary Aid to Needy Families, but it bears noting that there are single adults with no children who are poor too. In my state, the governor this summer abolished General Assistance, aka cash for non-parents. This exceptionally modest $200/month payment was often used to tide people over during the several-years-long process of becoming eligible for federal Social Security Disability Income (SSDI).

When the payment was being abolished, welfare caseworkers were threatened with firing if they dared to answer distraught clients’ inquiries by suggesting they talk to their elected officials.

Most poor pople in the US aren’t in these dire straits, but we’re a country of 310 million people. Even a tiny fraction ends up being a lot of individual human beings.


purple 09.22.12 at 2:49 pm

Most people living on 2$ a day in poor countries are living off the land, often land they own or in some type of commons, i.e squatters area. It is in no way comparable to living in the US for $2 a day, where you have to buy everything.

This is not to say life is joyful, it’s just that most of the non-capitalist world does not mediate their life primarily through exchanges in money.


PeterC 09.22.12 at 3:13 pm

Purple I think has it back to front. The $2 a day of the world bank was not a “have $2 cash extra in the pocket on top of having everything for free off the land”.

In sub-Saharan Africa, the gains from garbage bins, food courts, begging, prostitution, bashing someone after they make a withdrawal from an ATM are simply not there.

However, the first world has worked hard to ensure that there are rewards from joining the military. That is the nature of “aid”.


Mao Cheng Ji 09.22.12 at 3:13 pm

The West has much better dumpster-diving opportunities, though.


PeterC 09.22.12 at 4:11 pm

Lets remember the poor are more appropriately called the “undeserving poor” as the term is used by the well-off right amongst whom are included many who have never done an honest days work in their lives. And here be especially careful. Undeserving means not that they don’t deserve to be poor but that they don’t deserve our sympathy or assistance. No they do deserve to be poor. They made that choice so let’s leave them to it.

But these undeserving poor are guilty of so much more than simple poverty. No. If you listen to the right, they’re also the culprits behind the GFC. Yes. They conspired to hoodwink the financial community and that is the cause, the sole cause of the global world’s problems. Maybe the poor are also to blame for AGW and cancer?


piglet 09.22.12 at 6:50 pm

I want to echo the point some have already made that accurately measuring income at the very low end of the distribution is methodologically just very difficult because that is where the informal economy is most relevant, often far outweighing any official economy tracked by even halfways reliable statistics. Further, I think it is worth pointing out that international comparisons of the income of the poorest are unreliable. PPP (purchasing power parity) calculations are not based on the prices relevant to the poor. For example, in the US a haircut is more expensive relative to food than say in Ethiopia. When PPP is calculated based on a basket containing both haircuts and food, the PP of a poor Ethiopian (who can hardly buy food) is likely to be overestimated. I remember years ago there was an academic debate about that issue but I haven’t been able to retrieve any reference to that. Any leads will be greatly appreciated.


Stephen P 09.23.12 at 12:57 pm

I’ve got an article in press that enumerates the distinct and deep methodological problems in trying to evaluate the effects of welfare reform. Can’t seem to locate an e-mail, but drop me a note and I’ll send along a pre-print version.


SlothropRedux 09.23.12 at 3:06 pm

It’s a longer read than you were probably looking for, but ” Both Hands Tied” is a great and sensitive book on the topic. Although its an ethnography of the people affected by welfare reform, it also cites a lot of the larger scale studies. Check it out: http://press.uchicago.edu/ucp/books/book/chicago/B/bo8615760.html


John Quiggin 09.23.12 at 9:58 pm

Thanks to everyone for these great suggestions.


terence 09.24.12 at 4:27 am

Chris Bertram wrote:
“But presumably, a subsistence farmer in sub-Saharan Africa with some crops and a goat is also getting “some other stuff that isn’t counted”, namely, some crops and a goat?”

Where it’s available the World Bank uses consumption poverty data. That is, data that takes into account income from subsistence activities and the like.

Ajay wrote:
“On the study, I am amazed that it’s actually possible to survive on $2 a day in the US – simply from the point of view of getting enough calories and protein to keep yourself alive.”
It must be possible: the World Bank’s poverty data is adjusted for purchasing power parity meaning that when you hear someone say there are X billion people living off less than $2/day globally, what they are actually saying is that there are X billion people living off less each day than you could have purchased in the United States with $2.50 in 2005 (I think 2005 is the year of reference).

So it’s possible, but appalling of course. The severity of global poverty is shocking. Equally so the fact that it’s even possible that the US might have pockets of poverty that are even remotely close to being as acute.

On the order of magnitude difference between poverty lines in most developed countries and those used by the World Bank see Lant Pritchett’s paper here:


Ragweed 09.24.12 at 9:03 pm

“Most people living on 2$ a day in poor countries are living off the land, often land they own or in some type of commons, i.e squatters area. It is in no way comparable to living in the US for $2 a day, where you have to buy everything.”

No – that is not true. SOME people living on $2 a day in poor countries do have significant subsistance options from farming et al. However, there are also plenty of very poor people, perhaps the majority, living on $2 or less a day in urban slums that do not have access to land or any other subsistance options (unless you count picking through garbage piles), and must buy everything. They live in massive slums and barrios, on hillsides that collapse in floods, etc etc.

The big issue with subsistance vs non-subsistance is that there are very different solutions – for a community that does have substiantial subsistance resources, their well-being is largely dependant on the maintenance of resources of the commons – forests, fisheries, cropland, clean water, etc. However, many development strategies focus exclusively on trying to increase cash income, so that the forest on which communities depend for subsistance harvests become “natural resources” to be exploited through timber harvest, subsistance fisheries are replaced by industrial fisheries, subsitance farming replaced by cash-crops, etc.

“Indeed, if worse comes to worse many are even able to enlist. Those skills learned often of use in entrepreneurial criminal activity when they have finished their service.”

I am not sure if that has changed recently, but since the 1990s, the US military largely rejected recruits with any criminal or juvenile history. The transformation of the military into a professional, volunteer army also brought with it an end to the notion that it was an alternative to doing time. Which is a shame in some ways, because, cannon-fodder issues aside, it was once a way for kids to get out of a bad situation.


Ragweed 09.24.12 at 9:12 pm

On the study, I am amazed that it’s actually possible to survive on $2 a day in the US – simply from the point of view of getting enough calories and protein to keep yourself alive.

In countries where there are substantial numbers of people living on less than $2 per day, there is also widespread malnutrition. The countries at the bottom of the UN Human Development Index have 25-50% of children under 5 experiencing stunting or wasting.


Davis X. Machina 09.25.12 at 12:59 am


ragweed 09.25.12 at 2:43 pm

This is late to this discussion, but there is another point that comes out of this – one that development economists have been discussing for some time. Global inequality and global poverty is becoming less clearly demarked as rich countries vs poor countries, and more as inequality within increasingly stratified countries. While there are still a host of countries at the bottom of the UN Human Development Index, increasingly, the worlds poor live in what are considered middle-income countries, with increasingly large gulfs between rich and poor in the same country.


Ed 09.26.12 at 4:13 pm

This is a good article about what it is like to live on less than $2 a day, or thereabouts, in the United States:



Mao Cheng Ji 09.26.12 at 5:20 pm

Comments on this entry are closed.