TCS Daily

Bill Easterly Is About to Spoil the Poverty Party

By Tim Worstall - September 8, 2005 12:00 AM

The United Nations Development Program (UNDP) yesterday released its annual report, the Human Development Report. I don't think it's really going to surprise anyone to hear that while certain things are getting better, others are not and we should be trying harder. Put very broadly the past results and future challenges are:

        - 130 million people lifted out of extreme poverty
        - 2 million fewer child deaths a year 
        - 30 million more children in school 
        - 1.2 billion people gained access to clean water 

        - 2.5 billion still live on less than $2 a day; poverty reduction slowed in the 1990s 
        - 10 million preventable child deaths every year 
        - 115 million children still out of school 
        - More than one billion people still have no access to safe water; 2.6 billion lack access 
        to sanitation

As they point out, that is good but not good enough. The UN makes some sensible recommendations to speed things up.

        The Report decries what it calls 'perverse taxation,' under which the world's poorest 
        countries face the highest tariffs in rich countries, and examines the impact on the poor 
        of agricultural subsidies and protectionism in wealthy industrialized nations. Donor countries, 
        the Report shows, spend $1 billion a year aiding agriculture in developing countries and 
        $1 billion a day on domestic subsidies that undermine the world's poorest farmers. At 
        the same time, the report warns that the European Union and the US are restructuring 
        their subsidy programmes to limit the effectiveness of WTO disciplines.

        The overall effects of agricultural protectionist measures and subsidies in wealthy 
        countries, the Report estimates, cost developing countries close to $72 billion a year -- an 
        amount equivalent to all official aid flows in 2003.

The immediate destruction of the insane levels of agricultural subsidy in the rich world is the one thing that would increase human welfare the most. The idea is actually on the table as part of the Doha Round, with former US Trade Representative Robert Zoellick having proposed the so-called "zero/zero" option: The US will abandon its subsidies if the European Union will do the same (we will, unfortunately, be waiting for Satan to be wearing winter woolens before that happens).

There is, however, one slightly uncomfortable problem at the heart of the UNDP plans: The actual method by which they hope to improve the dismal realities of life for the poor. The basic analysis -- and this is true for the Millennium Development Goals, the plan from development economist and UN adviser Jeffrey Sachs and the UK's Commission for Africa -- appears to be wrong.

Now please, don't get me wrong, I'm all in favor of making the poor rich, for both moral and self-interested reasons, so the desires encapsulated in these plans are just fine and dandy by me. It is, however, an unfortunate truth about life, the universe and everything, that decent and just desires do not necessarily translate into useful or desirable action. It may even be that these plans will do some good, or at least not do any harm, but that isn't the point.

The Fallacy of the Poverty Trap

A new working paper by economist William Easterly shows us what is actually the problem. First, the current proposals are based on the following analysis:

        The UN Millennium Project and Jeffrey Sachs argue that it is the poverty trap rather than 
        bad government that explains poor growth of low income countries and the failure to 
        make progress towards the Millennium Development Goals (MDGs). Sachs says "the 
        claim that Africa's corruption is the basic source of the problem {the poverty trap} does 
        not withstand practical experience or serious scrutiny." Likewise the Millennium 
        Project says "Many reasonably well governed countries are too poor to make the investments 
        to climb the first steps of the ladder."

If this were true then the plans would have merit, so the question becomes, is this true? Easterly looks at a number of items to check:

        We can check further on some of the intermediate steps in the Big Push. Sachs said 
        that large aid increases would finance "...a 'big push' in public investments to produce a 
        rapid "step" increase in Africa's underlying productivity, both rural and urban." Over 1970-94, 
        there is good data on public investment for 22 African countries. These countries' 
        governments spent $342 billion on public investment. The donors gave these same countries' 
        governments $187 billion in aid over this period. Unfortunately, the corresponding 
        "step" increase in productivity, measured as per capita growth over this period, was zero.

Not looking all that good, eh?

        I think of this as a simple descriptive exercise to compare two alternative hypotheses: 
        (1) Divergence Big Time was due to a savings/technology poverty trap or (2) it was due to 
        bad government/institutions. The stylized facts emerging from this exercise support 
        (2) strongly over (1), confirming previous literature on institutions and development.

        I use three widely used measures of institutions: (1) the Polity IV measure again, now 
        averaged over 1960-2002, (2) the Freedom House measure of political liberties 
        (with the sign reversed, since an increase in this measure means less liberty), averaged over 
        all available years, which are 1972-2002, and (3) Economic Freedom in the World from 
        the Fraser Institute, averaged over all available years, which are 1970-2002. All measures 
        of institutions are strongly significant predictors of growth 1960-2002, and make initial 
        income negative in the regressions (significantly so in the IV regressions). The institutions 
        story makes Divergence go away in the more recent data as well.

        So which is it, bad government or the poverty trap? When we control for both initial poverty 
        and bad government, it is bad government that explains the slower growth. We cannot 
        statistically discern any effect of initial poverty on subsequent growth once we control for 
        bad government. This is still true if we limit the definition of bad government to corruption 
        alone. The recent stagnation of the poorest countries appears to have more to do with awful 
        government than with a poverty trap, contrary to the Sachs hypothesis.

No, it really isn't looking all that good. We seem to be locking ourselves into an argument over how much we should spend on a particular type of aid when the basic problem has been misdiagnosed. As the paper concludes:

        The classic narrative -- poor countries caught in poverty traps, out of which they need a 
        Big Push involving increased aid and investment, leading to a takeoff in per capita 
        income -- has been very influential in development economics. This was the original 
        justification for foreign aid. The narrative became less popular during the market-oriented 
        80s and 90s (even then the idea of the "takeoff" remained widely accepted, as it still is), 
        but has made a big comeback in the new millennium. Once again it is invoked as a 
        rationale for large foreign aid programs.

        However, the description of poverty traps, Big Pushes, and takeoffs as a justification for 
        foreign aid receives scarce support in the actual experiences of economic development. 
        The paper instead finds support for democratic institutions and economic freedom as 
        determinants of growth that explain the occasions under which poor countries grow more 
        slowly than rich countries.

There are various ways you can take this finding (and do remember, it is a working paper, others will no doubt wish to verify or contest the workings and conclusions), a call for more research perhaps, a vindication of pre-existing prejudices (I wrote something on the subject back here and Easterly uses similar markers for political and economic freedom as I did) or perhaps ignore it and suggest that the political capital invested in the process so far means that to rethink now is impossible.

Myself, I'm afraid I'm rather gloomy about the prospects, for once a bureaucratic bandwagon gets rolling it's almost impossible to stop it, even if it is on entirely the wrong track. I don't argue that there should be no aid, but I would take this paper as suggesting that we are about to do the wrong things with the money we have raised. If only we could get the UN, in its grand meeting this coming week, to understand two of the things about which Keynes was undoubtedly correct:

        Practical men, who believe themselves to be quite exempt from any intellectual 
        influence, are usually the slaves of some defunct economist. Madmen in authority, 
        who hear voices in the air, are distilling their frenzy from some academic scribbler 
        of a few years back.

Perhaps madmen is a little cruel for those who wish to make the world a better place and perhaps practical men is a little kind for those at the UN, but could I urge upon them this (possibly apocryphal) comment from the same late great economist?

        "When the facts change, I change my mind - what do you do, sir?"

If Easterly is correct in his paper then we're just about to waste $175 billion a year (roughly that 0.7% of GDP that the rich countries are pledged to) having misidentified the situation. I don't begrudge the spending (much) but I would like it spend on the correct problem.

Tim Worstall is a TCS contributing writer living in Europe. Read more of his writing here.


TCS Daily Archives