Should charity begin – and end – at home?

On many issues, such as higher education, the Centre for Independent Studies seems to care deeply about facts and data. But on certain topics, it seems only too happy to throw evidence out the window in favour of rhetoric. One of these is foreign aid. Last year, it was Helen Hughes. This year, it’s Deepak Lal. Neither Hughes nor Lal has done hard-core econometric research on the topic, and both are drawing conclusions that fly in the face of the best economic evidence, such as this study by Craig Burnside and David Dollar:

This paper uses a new database on foreign aid to examine the relationships among foreign aid, economic policies, and growth of per capita GDP. In panel growth regressions for 56 developing countries and six four-year periods (1970-93) the policies that have a large effect on growth are fiscal surplus, inflation, and trade openness. We construct an index of these three policies, interact it with foreign aid, and instrument for both aid and aid interacted with policies. We find that aid has a positive impact on growth in developing countries with good fiscal, monetary, and trade policies. In the presence of poor policies, on the other hand, aid has no positive effect on growth. This result is robust in a variety of specifications that include or exclude middle-income countries, include or exclude outliers, and treat policies as exogenous or endogenous. We examine the determinants of policy and find no evidence that aid has systematically affected policies – either for good or for ill. We estimate an aid allocation equation and show that any tendency for aid to reward good policies has been overwhelmed by donors’ pursuit of their own strategic interests. In a counterfactual we reallocate aid, reducing the role of donor interests and increasing the importance of policy: such a reallocation would have a large, positive effect on developing countries’ growth rates.

It’s quite reasonable to be sceptical about the relationship between aid and growth – indeed, the William Easterley vs Jeff Sachs debate over recent years has been fascinating, well-informed, and critically important. But please, let’s give a little more weight to careful analysis of the evidence, and less to anecdotes and prejudice. Any thinktank that wants to be taken seriously in the economic debate should probably make sure that its assertions are in line with the research.

This entry was posted in Thinktanks, Trade & Development. Bookmark the permalink.

29 Responses to Should charity begin – and end – at home?

  1. Sinclair Davidson says:

    Okay, I’ll bite. Which part of Lal’s op-ed don’t you like?

    He says

    …the humanitarian motive for giving aid may have justified transferring Western taxpayers’ money to poor people, but not to poor governments…

    Dollar and Burnside don’t disagree.

    The vast majority of foreign aid has failed to alleviate poverty. There are a few cases where it has improved the lot of poor people.

    Dollar and Burnside tell us when this happens.

    In badly run developing countries, governments channel aid to small elites.

    We know this too.

  2. Patrick says:

    I am interested, because whilst I think there is a good case for foreign aid, every formulation that makes any sense, like the study you cite, amounts to a good case for aid in principle but against aid in practice.

    Since we don’t live in principle but in practice, you know where that gets someone who wants to help foreign countries – campaigning against foreign aid, in favour of fairfree trade and investment and giving money to private charities.

    So Andrew, do you think that in practice we should be expanding foreign aid, or that we ought to focus on fairfree trade and investment?

  3. Andrew Leigh says:

    Sinc, much of the period that D&B analyse is the cold war. Much foreign aid wasn’t intended to help poor people, but to buy support from the leaders. I read D&B as saying that when western leaders actually set out to make a difference – funding good projects in countries with good policies – it improves lives. We should be working harder to identify good projects, not shutting off the spigot.

    Patrick, what the studies seem to show is that (both in principle and practice) good aid works, and (both in principle and practice) bad aid doesn’t.

  4. Sinclair Davidson says:

    I think we all agree that funding good projects in countries with good policies is a good idea. Indeed, these countries don’t need aid for very long. The question is about countries with bad policies – should we provide aid to them? This question turns on what it is we want to achieve. To the extent that dictators ‘steal’ the money but oppress their people less the aid may well improve lives, and be well spent. Yet, do the tax-payers of the OECD have a moral obligation to spend money in that manner? To the extent that aid is spent in the hope of improving lives and making individuals self-sufficient then it is a failure. i think we need to be very clear on what it is we hope to achieve.

  5. Russell says:

    Just to be clear … in the paper cited, ‘foreign aid’ seems to mean just one specific type of foreign aid: loans. To the ordinary person (who hears this Lal on Counterpoint) it sounds like all the varieties of aid are being condemned – although he sort of accepted the need for humanitarian emergency aid.

    Sinclair, most countries have good policies and bad policies, good politicians and bad politicians. It’s not as simple as you would have it. Do you have any evidence for your statement that: “To the extent that aid is spent in the hope of improving lives and making individuals self-sufficient then it is a failure.”

  6. Sinclair Davidson says:

    It’s not as simple as you would have it.

    Sorry, Russell. It is that simple. As you say, all economies have good and bad policies and good and bad politicians. On average, however, some economies have bad policies overall. These policies usually consist of fixed exchange rates, capital controls, high tariffs, and the like.

    The evidence at an anecdotal level is the entire continent of Africa – despite billions in aid it has not become economically prosperous. Systematic evidence can be found here

    and also here

    please note the article by Davidson (sorry Andrew, shameless self-promotion)
    and here

    and here

    Sorry, that all looks a bit untidy, but technophobe that I am, I don’t know how to enbed links into words.

  7. Russell Hamilton says:


    So if you go to the Ausaid site and look at Laos :

    you will find we give aid which helps, for example, to implement a land titling regime – which will give people the means to use their land as collateral etc . What’s wrong with this aid ??

  8. Sinclair Davidson says:

    Russell – I think you’re trying to change the subject. Have you read the links I posted above? Do you have any comment to make on them?

    As a general principle I’m all in favour of land titling – DeSoto is convincing on this point. I’m a bit confused as to why the Loas government is incapable of doing this themselves.

    But, nonetheless, most of the billions of dollars spent on foreign aid have the stolen by corrupt governments – as Lal indicates in his op-ed. This was not the policy design, consequently the policy can be described as having failed.

    Looking at AusAids site, I discovered that Australia budgeted to spend $34 million in Fiji this year promoting stablity and economic growth. So, in your opinion has that money been well spent?

  9. Russell Hamilton says:

    I didn’t say all foreign aid was good – I’m just disputing the general claim that ‘foreign aid’ is bad. The critics don’t seem to distinguish between the great variety of programs.

    Couldn’t we have just bought Fiji for $34 million?

  10. Patrick says:

    Surely, Andrew, your laudable concern with data and detail cuts both ways? Ie some critics might over-generalise their case, but that international aid, generally distributed with little concern for data and disregard for the relevant details, is in a bad way?

    Might we not have a moral obligation not to funnel money into such a poorly transparent area where so much of it will be wasted if not positively counterproductive??

    Given that there is a finite amount of it (money) after all, in the practical world, and you are supposed to study the appropriate allocation of scarce resources, IIRC 🙂

  11. Russell says:

    Sinclair – lunch time, so I can read your links … and they appear to be about aid in the form of ‘government to government transfers’ whatever they are. My point was that they aren’t the only type of aid.

    Recently AUSAID sent a librarian up to Cambodia to help them set up an archives / digitisation program for their National Library (I’m an assiduous reader of the glossy AUSAID magazine). Like land titling in Laos, I guess the Cambodians could have done something themselves, but it’s useful and nice to offer all that we’ve learnt from experience so that people in developing countries don’t have to start from the beginning and make the mistakes we did. It’s useful, neighbourly and we all learn something about each other.

  12. Sinclair Davidson says:

    I don’t think that foreign aid is ‘bad’ – I think it represents a stated policy failure. The policy was not to enrich despots. I agree there are a lot of neighbourly things that can be done, but these things aren’t expensive and don’t involve billions of dollars etc.

  13. Russell says:

    Sinclair. please stop making the same mistake. You say that you agree that good things can be done, ie that there is good foreign aid (as well as bad). So you can’t say that ‘foreign aid’ “represents a stated policy failure”, only that some foreign aid programs represent a stated policy failure.

    I suspect people like Lal and Hughes, and maybe you, want to take some failures in foreign aid and generalise those to discredit all foreign aid, which is just ridiculous because there have been many successsful programs which have fulfilled their stated policy aims. And they do that because they can’t stand ‘the government’ taking their money and spending it – and of course sometimes wasting it. Private charity, people deciding to use their own money for humanitarian purposes, fine; but governments doing it = bad. Well, you can’t generalise that ‘foreign aid’ doesn’t work, because the generalisation is just plain wrong.

  14. Pingback: Tim Worstall

  15. Sinclair Davidson says:

    Russell, I’m not judging aid by its intentions. I think we’d find that all aid (especially since the end of the Cold War) has good intentions. Those of us who criticise foreign aid look to outcomes. On a ‘bang for buck’ basis, the outcomes don’t look good. As Dollar and Burnside indicate the return to aid is higher in those economies with good policies and lower in those economies with poor policies. (I like that result, but our tame development economist doesn’t – he’s away at the moment, or I’d get more detail out of him). Unfortunately, there is a high correlation (IMHO in this case causation) between an economy being poor and having poor economic policy (see the Economic Freedom stuff for evidence).

  16. Patrick says:

    I agree with Sinclair, although I don’t presume to say for the same reasons. My fundamental point is that aid to governments is a really precarious proposition for moral reasons. On the supply side there is only a finite amount to be spent so a government as a collective spender has (I would argue) a moral obligation to allocate resources with some degree of regard for the collective interest, including in that ‘regard’ not simply wasting it.
    On the demand side, the problem in many ways is precisely that there is a demand side. By bolstering corrupt and venal bureacracies, we contribute not only to the undeserved (by any standard) enrichment of some but also to the continued, if not further, immiseration of many.

    I think aid should be given rarely and with great thought. I think the normal rules of public expenditure must apply – aid money must be spent in the pursuit of a defined and measurable goal, in such a manner that the actual allocation of funds is transparent, and that governments should not be any less strictly judged than another – by which I mean one should always take the announcement of a new aid package and replace ‘Donor Country A’, ‘the donor government’, etc with ‘Rupert Murdoch’ or ‘Macquarie Bank’ or ‘Shell’ and see what stinks. If you are instinctively repelled by the announced aid deal coming from one of the above, you ought to be repelled by it coming from your government as well.

    Above all, I think we should distinguish between ‘relief’ aid and foreign aid as it is perhaps most commonly perceived (long term flows to governments). Relief aid, which is temporary spending to meet immediate urgent needs is generally fine, and most people probably don’t oppose this. More generally, aid should be only with long-term goals in mind, such as not fostering corruption, nor wasting money on eg a water pipe network that will disintegrate in five years because no-one owns it nor has any incentive to maintain it.

    Sorry for the disjuncted nature of that.

  17. Russell says:

    I give up (just about) because we’re not speaking the same language. Sinclair I didn’t mention good intentions. My whole point, made in my first comment, is that the critics of aid are generalising the failure of some types of aid. But maybe that’s what “foreign aid” means in economics speak – those government to government transfers of money for general budgetary support etc.

    So what was wrong with us helping poor, backward little Laos, which probably does have a “corrupt and venal” government, by building a bridge to connect them to the outside world: the Lao people have definitely benefitted. Here’s what AUSAID’s magazine says ( they’ll send it to you gratis):

    “‘Now, after 10 years of
    operation, Laos, Thailand and
    Australia can all feel justifiably
    proud that the bridge is achieving
    this aim, helping those
    on both banks of the Mekong to
    forge enduring economic
    partnerships and extend their
    commercial reach,’ says Jonathan
    Thwaites, Australia’s Ambassador to
    ‘We’re seeing the Lao People’s
    Democratic Republic extend its
    international links by the
    construction of improved roads and
    new bridges, which will provide
    corridors of trade through mainland
    South East Asia. We are pleased to
    have had a hand in the beginning of
    this process.’

    The Friendship Bridge was so
    named to symbolise the long
    relationship Australia enjoys with
    both Laos and Thailand, and the
    cultural ties between the Asian
    neighbours. It is the first
    permanent crossing over the lower
    Mekong. Hundreds of international
    tourists use the bridge every day.
    Many find it an easy economic
    route to and from Bangkok and
    Vientiane, travelling by either bus
    or train. Traders take goods worth
    millions of dollars back and forth
    between their countries.

    For Sumphorn Manodham,
    President of the Lao Wood Products
    Exporter Group, the bridge is a
    major boost to business. ‘Having
    the bridge helps us to save costs and
    time in travelling and it helps
    access better facilities, such as
    hospitals, the airport and shopping
    centres. Export is faster,’ he says.

    Thongleuane Thammavong,
    who works at the Australian
    Embassy Medical Clinic notes that
    the Friendship Bridge has also
    helped reduce time for medical
    emergencies. Before the bridge, all
    emergency cases were sent to
    Nongkhai/ Udon Thani in Thailand
    via the extremely busy commercial
    port of Thanaleng.’There was only one ferry used
    for transport so if the ferry was still
    loading goods at the other side we
    would have to wait until they had
    finished before coming to pick us
    up,’ Thongleuane says. ‘This could
    take two hours. Now we can
    transport emergency patients over
    the bridge in the Australian
    Embassy Medical Clinic’s
    ambulance at any time without

  18. Sinclair Davidson says:

    they’ll send it to you gratis

    I send them, via the ATO, heaps of cash – nothing is gratis from the government. That is not the point here. Most Aid has the objective of alleviating poverty and unlifting peoples lives etc. To be successful, Aid should be temporary, yet it isn’t.

  19. Pingback: Club Troppo » Friday’s Missing Link

  20. Russell says:

    OK the ‘gratis’ was just teasing. Such a cheery publication though, all those beaming fuzzy wuzzies enjoying our generosity (Macquarie says that’s a derogatory term, but don’t you think it’s more of an affectionate term? bit like golliwog) …

    So the bridge isn’t temporary – what objections do you have to the bridge?

  21. Sinclair Davidson says:

    ‘fuzzy wuzzie’ is in the dictionary? Anyway, I’m sure it is a fine bridge. I have nothing against bridges (I hope they did an environment impact statement). I’m sure we could list every project and, in principle, each and every project would appear worthy. That’s not the point here. The road to hell is paved with good intention. What are the results of the billions that have been spent on aid since the end of WWII? Answer, not good.

  22. Patrick says:

    Sincs is right. If you start evaluating each aid project on no greater criteria than ‘goodness’ or ‘usefulness’ you will be here until forever (or bankruptcy).

    You are economists, try and remember that your raison d’etre is scarcity. If you act like scarcity isn’t a problem, or isn’t a real factor in policy making, we will stop believing we need you and then you will be dependent on internal aid.

  23. Helen Hughes says:

    In speaking to radio and general audiences Lal and Hughes did not deal with details of the econometric studies, with which they are of course, familiar.

    It is well established that Burnside and Dollar 1997 and Collier and Dollar 2002 World Bank papers used limited data. An improved data base was subsequently used by Collier (2001) and Collier and Dollar (2001) with less positive outcomes for aid effectiveness. But all these studies rely heavily in their limited conclusions, that is, where domestic policies are appropriate aid has positive effects, on unduly positive definitions of “appropriate domestic policy environments” as Vasques (2003) pointed out together with other methodological shortcomings.

    Boone 1994 and Easterly 2001 came to contrary conclusions.

    Clemens, Radelet and Bhavnani 2004 sought to show that if ‘relief’ aid was excluded from development’ aid a more favourable result for aid effectiveness could be obtained, again provided the economic environment was positive, but this finding has been critiqued by Vasquez (2004)

    The comprehensive survey of econometric studies of aid effectiveness by Gilbert, Powell and Vines, 1999 concluded: “there is now widespread skepticisms that concessional assistance does have positive effects on growth, poverty reduction or environmental quality”

  24. Andrew Leigh says:

    Helen, thanks for taking the time to comment. What troubles me is that it seems to me that your writings (and those of Deepak Lal) overstep the available econometric evidence. For example, Gilbert, Powell & Vines (1999) who you cite with approval, do not conclude that aid hurts countries, or even that it makes no difference. Instead, they argue that because of time inconsistency problems, conditionality is frequently ineffective (the phrase in your quote above should have been ‘conditional assistance’, not concessional assistance). But GPV conclude in praise of aid:

    Components of an answer to this question are now becoming clearer. Conditionality has not worked, or, at least, has not worked to anything like a sufficient extent to make this the main instrument by which the Bank advances development. The Bank should therefore step back from attempting to coerce developing country governments into adopting good policies. Instead, it should adopt a twofold strategy. First, it should make available policy advice, based on best practice and on experience, to all countries. This advice should be seen as a public good and should be subsidised or free to recipients (certainly to low income recipients), financed, as currently, from the Bank’s income from capital. Second, the Bank should finance good projects proposed by potentially good borrowers. Governments should propose projects to the Bank and then manage these projects, although Bank staff will often participate in the designing of these projects and even the writing of the proposals.

  25. Helen Hughes says:


    You have a strange view of the econometric evidence. If conditionality has not worked, how can aid possibly have worked? The Gilbert et al paper was predicated on the importance of conditionality of making aid effective. Deepak and I have taken the econometric studies very seriously, including the evaluation of the data used, which particularly in Burnside and Dollar was very weak, being based on income surveys which we know are not reliable for developed countries let alone developing countries. I have written about that in Policy magazine. The proxies and assumptions about ‘policies’ which are critical to these studies are subjective, in the Bank studies often unconvincing and sometimes ludicrous. Many of the coefficients used are imaginative rather than factually based. The more deeply one digs into the data and proxy issues the less convincing the findings, which most authors link to ‘good policies’ are. I understand that Fin Tarp who has written on this is tackling these issues in Stockholm in a major study.


  26. Andrew Leigh says:

    The Gilbert et al paper says at the outset “the imposition of conditionality is not crucial to the Bank’s intervention”.

  27. Aussie Equitist says:

    I can’t believe the rubbish that is being spewed out of the C(not-so)IS – both published on their site & permeating the mainstream media!

    I can recall one particularly despicable article was published in the mainstream media, which made patently false statements for the transparent purpose of demonising & vilifying single mothers!

    What is really scary, is that the Liebral-Coalition Dictatorship self-servingly plucked our so-called “Fair Pay” Commissioner from the CIS. Ditto re the RBA & ABC appointments – such board stacking trends are very disconcerting.

  28. Aussie Equitist says:

    Apologies for any unintended ambiguity in the final paragraph of my last post!

    I trust you all understood that: I was concerned about the board-stacking issue in general – not that I was suggesting that all such political appointments had been made from the dubious ranks of the CIS.

  29. Andrew Leigh says:

    AE: Ian Harper’s never worked at the CIS, to the best of my knowledge. It’s true he once gave a CIS lecture, but the same is true of Emerson, Tanner and Rudd.

Comments are closed.