Ownership Discussion

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: OWNERSHIP: Re: Poverty and Economic Growth

The Dollar-Kraay article was discussed in the May 27 Economist and can be
downloaded with graphs from:

I am looking forward to comments by experienced econometricians (unlike myself)
who will try to reproduce the results.  I instinctively regard e-CON-ometric
work as giving relatively little insight if it is not accompanied by creditable
discussions of the mechanisms.  This paper was notably lacking in causal
mechanisms to explain the results.  Does growth hurt the poor in the absence of
social safety nets and do these results show that social safety nets are very
effective?  Or do these results show that growth in some sense automatically
helps the poor so countries don't have to worry so much about safety nets??
(many people might interpret it in the latter sense).  A number of other
questions arise.  When you look at before and after with over five year
intervals, is there a lot of hammering of the poor in the interrim?  How does
one try to reconcile these results with one's intuitive sense that the rich can
better take advantage of openness and globalization than the poor?  For
instance, if one might look at the impact of the internet on San Paulo, how
could one argue that the people in the slums would take advantage of the "new
economy" equally with the well-to-do people in the suburbs or downtown?  Jeff's
new book is all about the remarkable increases in inequality in the US.  How
could similar processes not be operating in countries that are becoming more
open and globalized, i.e., more like the US?

 Now let's look at what the concepts mean in terms of logic, regardless of the
econometrics and the underlying processes.

It is useful to keep in mind what the "one-for-one income growth" really means.
It is one-for-one in percentages.  Let say you have $1 a day income at the
beginning of the period counting as the growth episode and I have $10,000 a day
income.  At the end of the period, you have $2 a day income and I have $20,000 a
day income.  This is what the Dollar-Kraay article would call "one-for-one
growth" since both of our incomes increased by the same 100%.  However the
income gap went from 9,999 to 19,998, namely doubled.  Now you ask yourself.
When you or others read the Economist article, what was your mental model of the
"one for one growth"?  Did you realize they were only talking percentages so
that the income gap would actually be growing larger under "one for one growth"?
Did you realize that economists would describe such a doubling of the income gap
as "no change in income inequality" as indeed there would be no change in the
Gini coefficient (doubling all incomes leaves Gini the same) used to measure
income inequality?

There is another "trick" to bear in mind in interpreting this sort of argument.
They are comparing income to income without taking assets into account.  Let's
take the total dreamworld of equal absolute income (not percentage growth rates)
between the asset-rich and the asset-poor so there is zero income gap.  You have
$1,000 assets and I have $10,000 assets but our assets grow at the same amount
per time period, say at $200 a year.  Now a Korean-style crisis hits and
clobbers the assets of the asset-poor (say reducing your 1,000 to 100) while not
affecting the asset-rich who got bailed out by some friendly multi-lateral
institution.  But after the crisis, our incomes could remain the same in that
each of us could be getting $200 a year (you getting your job back but now
starting from $100 assets and me starting from $10,000 as before).  Then
economists could say "The crisis had no effect on incomes."   Indeed the income
gap not only remained constant, but remained at zero while the asset-gap
considerably widened as a result of the crisis.

When economists are dealing, it pays to watch the deck closely.


Keith Wilde <> on 06/08/2000 07:49:16 AM

Please respond to

Sent by:

To:   "H.F.(Bob)Fletcher" <Hfbobfl@Islandnet.Com>
cc:   Steve Kurtz <Kurtzs@Freenet.Carleton.Ca>, John Lambert
      <Jlambert@Worldbank.Org>, Jeff Gates <Jeffgates@Mindspring.Com>, Cog
      Ownership - Dan Bell <Ownership@Cog.Kent.Edu>

Subject:  OWNERSHIP: Re: Poverty and Economic Growth


Thank you. Very interesting indeed! Can you tell me what number of The
Economist contains the main article?  On my current job I don't get free
access to goodies of that kind.

And what do you copy cats think of this?

Keith Wilde
Ottawa, Canada
613 990-8125
613 747-6847
-----Original Message-----
From: H.F.(Bob)Fletcher <>
To: <>
Date: Wednesday, June 07, 2000 11:43 PM
Subject: Fwd: Poverty and Economic Growth

>Hi Keith
>Thought you might be interested.  Of course, the article doesn't address
>the potential impact of growth on the ecosystem or the longer term social
>impacts that may follow excessive resource exploitation.
>>From: "Elizabeth and Earlston Doe" <>
>>To: "Douglas Montgomery" <>,
>>         "H.F.(Bob) Fletcher" <>,
>>         "Ursula and Jim Mount" <>,
>>         "Ivo Imparato" <>
>>Subject: Poverty and Economic Growth
>>Date: Mon, 5 Jun 2000 17:53:15 -0400
>>X-Mailer: Microsoft Outlook Express 5.00.2615.200
>>The following has been circulated fairly widely among staff at CIDA.
>>It was forwarded to me by our son, Ed Doe, whose comment was that
>>it is very interesting, whether you agree with the conclusion or not. I
>>understand that this, which appeared in THE ECONOMIST, is comment on the
>>original article.  Ed also sent me a copy of the original article by
>>Dollar and Kraay (text only, no diagrams as yet).  If you are interested,
>>I would be glad to pass it on as well, (about 30 pages).
>>1. Poverty and economic growth
>>                IT IS striking that few, if any, of the backlash crowd
>>                that globalisation is bad for growth in an overall sense.
>>                Their complaint is rather that growth serves the interests
>>                only of the rich. As the prosperous become more so,
>>                inequalities widen and the poor are left out.
>>                This is a claim you could make about rich and poor
>>                countries, or about rich and poor people within any given
>>                country. Most backlashers seem to believe both versions.
>>                The evidence on the first has long been clear. Poor
>>                countries that cut themselves off from the global economy
>>                and fail in other ways to establish a platform for growth
>>                indeed stay poor; the rest do in fact "converge".
>>                As for the second version, difficulties in gathering and
>>                examining the data have clouded the issue. But a new paper
>>                by David Dollar and Aart Kraay of the World Bank puts
>>                matters straight. Its findings could hardly be clearer.
>>                Growth really does help the poor: in fact, it raises their
>>                incomes by about as much as it raises the incomes of
>>                everybody else.
>>                The authors look at data on growth, incomes and a variety
>>                of other variables for a sample of 80 countries extending
>>                over four decades. On average, incomes of the poor rise
>>                one-for-one with incomes overall. There is relatively
>>                variation around that average. If you plot incomes of the
>>                poor against overall incomes, the points all lie close to
>>                one-for-one straight line. For instance, the data yield
>>                episodes of at least five years in which overall incomes
>>                head grew by 2% or more a year. In all but six of these
>>                cases, incomes of the poor also rose. As the authors
>>                emphasise, this is not "trickle-down"?meaning that the
>>                get richer and then, after a while, the poor do better as
>>                well. The rich, the poor and the country as a whole are
>>                seeing their incomes rise simultaneously at about the same
>>                rate.
>>                The paper then looks at some other ideas about the course
>>                of poverty in development. The oft-cited "Kuznets
>>                hypothesis" holds that intra-country inequality increases
>>                the early stages of development and then falls later on.
>>                so, it now appears. Dividing the sample between rich
>>                countries and poor countries, Messrs Dollar and Kraay find
>>                that the link between incomes overall and incomes of the
>>                poor is, as before, roughly one-for-one in each case, and
>>                that in this respect the two sets are statistically
>>                indistinguishable.
>>                Another myth: in crises, the poor see the biggest falls in
>>                income. Again, dividing the sample into crisis and
>>                episodes, the authors find that the one-for-one link
>>                intact. (This is not to deny that a 10% fall in income
hurts a
>>                poor man more than a rich man. But if the claim is that
>>                incomes of the poor fall in crises by proportionately more
>>                than the incomes of the rich, it is wrong.) And yet
>>                myth: it is often argued that growth used to benefit the
>>                but in the new world economy no longer does so. Dividing
>>                the sample into two halves at 1980, the connection
>>                between incomes of the poor and incomes overall remains
>>                one-for-one in both periods.
>>                Finally, the authors ask whether particular policies and
>>                institutions have a systematically different effect on the
>>                poor. For instance, does globalisation increase
>>                intra-country inequality? The answer to that is no. The
>>                paper looks at the effect of openness to trade (measured
>>                by the sum of exports and imports relative to GDP) first
>>                incomes overall, and then on the distribution of income.
>>                finds that openness spurs growth to a statistically
>> significant
>>                extent, and has no discernible effect on distribution. In
>>                short, globalisation raises incomes, and the poor
>>                in full.
>>                Economists have long argued that the rule of law is
>>                in development. The authors confirm this: stronger
>>                rights promote growth. But do stronger property rights
>>                skew the benefits of growth away from the poor, as some
>>                might suppose? Again, no. The effect on distribution is
>>                statistically indistinguishable from zero. What about
>>                democracy? The income effects are small and statistically
>>                insignificant through both channels. Even primary
>>                surprisingly, has no perceptible pro-poor bias, although
>>                does, as expected, promote growth, and therefore helps
>>                the poor to that extent.
>>                Only two policies appear to have a systematically biased
>>                effect?that is, they affect the distribution of income as
>>                as growth in incomes overall. One is cutting inflation,
>>                the other is cutting public spending. Both of these raise
>>                growth, as you might expect. What you might not expect is
>>                that they also improve the distribution of income,
>>                the poor twice over.
>>                Surveys often show that the poor hate inflation more than
>>                the rich. Now you know why: the evidence in this study
>>                shows that inflation causes a proportionately bigger drain
>>                on the incomes of the poor than on the incomes of the
>>                The public-spending result seems more surprising. High
>>                public spending is often justified as a way to help the
>>                So far as their incomes are concerned, it seems to do the
>>                opposite: it retards growth, which directly reduces the
>>                income of the poor and everybody else, and then on top of
>>                that it tilts the distribution of income to the poor's
>>                disadvantage. "Social spending", the category of public
>>                expenditure most explicitly targeted on the poor, is
>>                neutral, having almost no effect one way or the other on
>>                either growth or distribution.
>>                It is hard to believe that this study is going to change
>>                backlashers' minds. After all, the authors are from the
>>                World Bank, so their work can be put in the bin unread.
>>                But perhaps it is not too much to hope that governments
>>                will be a bit less apologetic, a bit less pandering, now
>>                they have been shown so plainly that growth is as good for
>>                the poor as it is for everybody else.

      David Ellerman
      Economic Advisor to the Chief Economist
      World Bank, Room MC4-335
      1818 H St., NW
      Washington, DC 20433
      Ph: 202-473-6368
      Fx: 202-522-1158