And the winner is ...

Actually a lot of respondents got this right, but the answer is counterintuitive as some also mentioned.   The  correct answer is that a poor person in a rich country is three times better off than a rich person in a poor country (given my definition of what "rich" and "poor" mean in this context). It is not even close.

The relevant numbers are below (in 2004 PPP-adjusted dollars). The reason we are tempted to answer "rich in a poor country" is the obvious wealth at the very top of the income distribution. But by the time you average the entire top 10%, the income level goes down quite a bit.

The next question would be to do the same for health indicators, say life expectancy.  My preliminary calculations indicate that the answer is the same--but I do not have a definite answer yet.

                       

	Per-capita GDP	Income share of …	Representative income of …
Poor country	$868	top decile in poor countries = 0.35	Rich individual in poor country = $3,039
Rich country	$34,767	bottom decile in rich countries   = 0.027	Poor individual in rich country = $9,387

Comments

[Pearl]

Oh, what a surprise. Seriously, were there really people out there who believed the contrary?

[Sean]

I believe someone mentioned this on the previous post, but I'll just say it again here. Though monetarily, a poor person in a rich country might be better off, this says nothing about the actual welfare. There are numerous studies showing that happiness is directly correlated to relative income. Further, though we would like to think that a poor person in a rich country would have social programs and safety nets, not to mention the possibility of upward mobility, this is often not the case. Case in point: the almost 50 million people in the US without health insurance.

[Bruce Webb]

Was this controlled by standard of living?

You have to be pretty well off in this country to have live in help, whereas inmuch of the world this is a standard feature of middle-class life. There are reasons why middle class Americans retire to Mexico even not speaking the language, the dollar goes farther there. It is not just a question of relative income, it is a question of actual exchange value. How much goods does that $3k buy compared to the $9k?

And I question trying to define 'rich' and 'poor' by decile and use that to define 'better off'. This builds in the assumption that income distribution has the same shape of curve. Whereas in fact we know that different countries have different degrees of wealth concentration. It seems a better way to capture this would be to turn the calculation around and break up the GDP by decile. For example you could define 'rich' as 'those controlling 90% of GDP' and then dividing that by the numbers in that category.

Arbitrarily defining 'rich' by population decile puts you at risk of the Bill Gates Syndrome. If nine guys are sitting in a bar and Bill walks end then everyone is worth $4.4 billion. In the Congo it might well be that fully 90% of national wealth is in the hands of the top 1%. Certainly it is ridiculous to suggest that you are somehow better off in the United States living on $9000 than Mobuto who was sitting on more than a $billion.

[spencer]

Like a lot of people I pretty much ignored point three of your conditions.

But I'm interested, exactly what countries are in the list
point 3 restricts the question to-- the bottom decile of countries?

[ds]

It is important to bear in mind that assumption 1 is unrealistic: We do not just care about our own consumption possibilities. Instead we constantly compare ourselves with others - If our real income remains constant but that of other members of the society rises, we feel actually worse off.

Even with greater consumption possibilities in a rich country, my welfare in a poor country is likely to be higher.
(Because my "rank" in the society is higher)

A realistic utility function should include this spillover effect.

[terence]

Bruce Webb,

By adjusting for purchasing power parity you have controlled for 'standard of living' (if I am understanding your question correctly). However, your point about the live-in help touches on an interesting argument: as I understand it, purchasing power parity figures tend to be calculated across a 'basket of goods' that includes all goods and services purchasable in a country. Yet in developing countries, services tend to be much cheaper (relative to basic goods) than they are in the developed world. Hence the live-in help being easier to come by. When purchasing power parity is used in poverty calculations, this becomes a potential methodological issue because services like live-in help are totally irrelevant to the poor, yet they are still included in PPP calculations (we're talking World Bank here, by the way) which leads to a lower poverty line and lower poverty figures than if your PPP basket was limited to simply items relevant to the poor. This is what is at the heart of Sanjay Reddy's critique of World Bank poverty measurements. You can read more at www.socialanalysis.org

cheers

Terence

[terence]

Sean and ds,

My copy of Richard Layard's book is at home, so I'm relying on my memory here - which is a risky thing to do at the best of times, let alone a Monday morning.

But I think you'll find that happiness becomes more of an relative function only after a point. Up until a certain per capita income there are, IIRC, pretty strong returns to absolute income. Only after that point do diminishing returns set in and other factors become more important. And I'd say that bottom decile countries are well below the key per capita income.

[terence]

Unfortunately it doesn't quite provide the answers to the questions about health, but Gap Minder's representation of the 2005 HDR makes for a very interesting watch (as opposed to read) with regards to the interaction between health and wealth and inequality:
http://tinyurl.com/yu7esc

[minderbender]

Spencer - I assume this list is accurate:

https://www.cia.gov/cia/publications/factbook/rankorder/2004rank.html

[Gallagher]

This answer is technically correct based on the way we've decided to look at it in this case. Sen has a famous article titled "Poor, Relatively Speaking" (Oxford Economic Papers, New Series, Vol. 35, No. 2 (Jul., 1983), pp. 153-169) that looks at the question from a different angle--because the poor man in the US makes RELATIVELY a lot less that the average person in his orbit he feels and thus to an extent is much poorer than the rich man in the poor country.

[sez.bad]

Dear Mr Rodrik,
May i pose a similar question for your users.
"Would you rather be an elite in a poor country or a nobody in a rich country?"

I would certainly like to know your answer, atleast.
Regards.

[terence]

sez.bad,

Once you mention the word elite the question becomes more interesting still, because elites also gain an additional benefit: power. And, through this, some opportunity to influence the circumstances that they make their lives in. People living in poverty are typically denied this.

[Rybinski]

I would agree with Gallagher, rich and poor is a relative concept. In my country, Poland, everybody is better off 15 years after collapse of socialism (life expectancy, number of houshold appl., number of cars, phones, infant mortality rates, you name it). But in opinion polls many people feel they lived better under socialism. It is probably becasue "I have a car, but my neighbor has a better one, so it makes me feel bad, under socialism we both had nothing and this felt better".
I think the title of Dani Rorik blog needs a correction. His line of thought is no longer unconventional, it will be the main stream in XXI century, see recent book by Tornell and Westermann.
Best regards,
Krzysztof

[John Maggs]

Since the result on rich/poor turns, in part, on the degress of inequality in rich countries, what is the result if we limit the "rich" field to the more unequal United States. Is it better to be poor in America than rich in the composite poor countries?

[Marc Lee]

Hands down, you are much better off being in the top 10% in a poor country than in the bottom 10% in a rich country. The analysis above is way too simplistic, and even the numbers are disputable (from sampling errors to use of income concept). To caricature this a wee bit:

Being in the top decile of a poor country likely means you have a good job, some level of status, opportunities for advancement and travel, education, better health and on and on. Being poor in a rich country (excepting students in transition, etc) is equivalent to a crappy job, no control over your circumstances, no opportunities for advancement or to live the type of life you find valuable, no education, poor health.

An important element missing is wealth. In most poor countries the richest decile in terms of assets controls upwards of 90% of the total. In rich countries the bottom decile has zero, if not negative, net assets.

Mr Rodrik, as posed above, would you really choose to be in the bottom decile in the US than in the top decile in Nicaragua?

[Dan K]

Marc, Nicaragua is poor. I think we'd all agree. But it is not near the bottom decile in PPP terms.

And it doesn't matter whether you calculate by listing countries or counting population, the bottom decile pretty much only includes sub-Saharan Africa countries plus maybe Afghanistan/NKorea and minus South Africa.

These are not cushy countries to live in. Only the true elite (Delong's upper .1%) is not actively looking for better opportunities abroad.

[Marc Lee]

That's a sobering clarification. I was pretty sure that Nicaragua was near the bottom. But if my choices are Sub-Saharan Africa, North Korea and Afghanistan, then maybe being anywhere in the distribution is not a pleasant choice. But then neither is being in the bottom 10% in the US. This one might just be a roll of the dice.

[William R. Rhoads]

Please ask Prof. Robert Frank of Cornell to answer your question? He says, if I understand him correctly, that relative rather than absolute income is more important in determining happiness and satisfaction with one's life. Personally, I would rather be in the lower decile in the U.S. than in the top decile in poor African countries, but I would rather be in the top decile in Latin America, in Colombia, Peru, or Chile, than in the lowest decile in the U.S.

[FA]

Dani, you know very well that if you depart from false statements, you cannot derive correct conclusions...

Even you have a broader idea of consumption that the one you use for this post...

[Ben]

The interesting thing, looking at the UNDP Gapminder report, it all depends on how you define "rich" and "poor." Dr. Rodrik uses the top and bottom decile, but if you change it to the top and bottom 1 percent, the standards of living for the extremely rich in poor countries is far superior in many facets to the standard of living of the super poor in wealth societies. The super rich in Latin America, China, India, and the Middle East are much better off than those on welfare in the US, UK, and Italy.

[Winton Bates]

Some of the above comments about the importance of relative income don't seem to be supported by what people say about how satisfied they are with life. See my blog: http://www.wintonbates.bigblog.com.au/post.do?id=183306