Comments

Thanks Terence for another thought-provoking blog. Perhaps a suitable analogy that helps us understand why public sector capacity building in the Pacific performs sub-optimally is the Soviet tractor gearbox factory? This factory produces tractor gearboxes of a type and number because it had been planned in Moscow to happen this way, many years ago. As it has occurred, the factory ends up with more gearboxes than it can dispose of, yet the workers are so badly paid that they cannot survive. So they trade gearboxes on the black market for milk and bread from the state dairy producer and state bakery. The state wheat farm that gets the tractors that use the gearboxes is grossly inefficient because the gearboxes are not the right type for the farm, and they break down regularly anyway due to poor quality assurance in the factory. And the bread queue for ordinary citizens in Moscow gets longer and longer every day. In short, the incentives for individuals working within the system are poorly aligned with outcomes for ordinary people who depend on the system to work. The intellectual framework that supports a different, nuanced, flexible and demand-driven approach in the public sector reform business model of donors, now largely exists. However there is a great deal of institutional inertia at an operational level that resists the operational implications of the adoption of these principles. This inertia is based around norms and incentives - precisely what is discussed in this blog. Tobias rightly points out (in this blog and his paper) that capacity constraints in the Pacific are often closely linked with a small population base. But most efforts in the past to regionalise, outsource or externalise capacity have hit walls built of passive and sometimes outright resistance. It is difficult to escape the proposition put forth by Acemoglu & Robinson (http://whynationsfail.com/) last year that endemic capacity challenges are a political problem, and not a technical problem. I don't think this means that donors can't do anything about public sector capacity in developing countries. For example, in Solomon Islands, Australia has about A$240 million of influence to play with through its agencies. That's a fair degree of influence in Honiara. But how it is allocated is far more important than how much is allocated. If we think the Solomon Islands Ministry of Health has the right conditions and incentives to deliver public goods, then directing many more tens of millions into that organisation will nurture and support the dominant paradigm within that organisation, and deliver exponentially more of what is already being delivered. Or alternatively if it is riddled with malfeasance and maladministration, and delivers very little in the way of public goods, then we should expect that we will further nurture that culture by directing more funds in that direction. Consequently donors need a far greater and nuanced understanding of how institutions change, and more importantly they need to follow up with the will to alter how they allocate resources, if they are to play a positive role in making bureaucracies work better in the Pacific. Perhaps to make bureaucracies work better we need reform of the way we do public sector reform in the Pacific, more than we need another public sector reform program in the Pacific?

This is a great report on these lectures. Duflo's methods are so important to ensuring that aid assumptions are challenged. Her work on the important influence grandmothers can have on their granddaughters' health with just a small cash transfer is still in circulation, 13 years later: http://economics.mit.edu/files/732

The Garnaut imbroglio poses something of a dilemma for the Federal government. There’s a plausible (and ethical) argument to propose that Bob Carr’s foreign policy approach to PNG should maintain a balance between building the relationship and making it clear it does not condone bad policy (although the diplomatic nuances of such a posture are complex). But real politik within Australia would dictate that the government steer well clear of the tangled issue of Garnaut, the Sustainable Development Program and Peter O’Neill’s sensitivities. In an election year, Gillard and Carr certainly wouldn’t want to do anything that might stimulate greater controversy around the Manus refugee camps than already exists (and O’Neill is already encountering resistance to the initiative from Belden Namah). But most of all they wouldn’t want to jeopardise the newly strengthened and energised friendship with the PNG government, with Peter O’Neill being central to this. Papua New Guinea’s strategic importance in the China-Pacific era has dawned on our politicians and the need for a stable, friendly PNG has become a paramount goal. Bob Carr learned last year than intervening, even rhetorically, in PNG government actions is a dangerous place to be. One can perhaps criticise the PNG government in its approach to the Ok Tedi issue, but it seems a pity that Ross Garnaut didn’t handle the matter a little more sensitively himself.

Thanks Stephen for this. Without giving away anything of what will appear tomorrow(!) I would add 2 further items here. One is how PNG and Fiji will position themselves both within the MSG and the region more widely. On the one hand, there are indications that PNG is moving to leverage off its economic strength to gain greater diplomatic influence and on the other there are instances of these 2 collaborating yet more closely (possible investment in bemobile from Fiji is an example). And the Australia-PNG relationship whilst very warm at the moment will have a number of hotspots to negotiate including (but not limited to) pressure from elsewhere in relation to decolonisation issues, especially in relation to West Papua which then adds Indonesia into the mix.

Thanks MJ, I sort of agree with the direction of your comment, but wouldn't necessarily take it so far. I agree that it can be hard for donors to tackle malpractice or corruption head on, because this jeopardises relationships with partner governments. Yet at the same time there's reasonably good evidence to show that well intended aid reduces corruption, so it's not as if we get nothing for our buck here. I suspect though that most of this reduction, along with many of the improvements that we do sometimes see as a result of aid funded public service strengthening only occurs as a result of aid functioning as a countervailing force pulling against the destructive forces of local politics, which has a positive impact while it's there, but little by means of a positive legacy (with rare exceptions) after it is gone. I also agree that there are times when donors really do need to abandon efforts to 'strengthen partner systems' and try alternate mechanisms of service delivery, and that we don't do this enough. On the other hand I'd want to add to that statement that such alternatives are always second best: in the medium to long run there's no alternative to having the state provide health and education services and *if* we can use aid to achieve this, even if it's painfully slow an inefficient in the short term it's still a long term win. Whereas alternate delivery mechanisms are only really band aids. That being said, when you're bleeding a band aid is a very helpful thing. And in circumstances where we're achieving nothing or next to nothing through our strengthening partner system work then I'm all for abandoning this and then doing our best to make sure our aid delivers welfare benefits to the poor by finding alternate means of delivery. Thanks again for an interesting comment.

Nice typology. I would add two points to the general analysis. Firstly, in my experience developing country bureaucracies love to highlight the lack of capacity because they can then get lots of perks along side the capacity building that follows, and, moreover, because it implicitly gets them off the hook for corruption and apathy. Since in the development industry we spend a lot of time having to cooperate with the bureaucracies we often go along with this. It is one of the great euphemisms of international development, and one of the most unhelpful. Secondly all the big donors operate primarily through the local government as the implementing agency. So with huge dollops of cash to spend they throw a lot of money at the one thing that (almost) everyone can agree is a problem, without considering whether it is the primary constraint to improvements. At least some donor representatives around here recognise this problem, but any attempt to introduce competition in service delivery through the private sector or NGOs rarely gets anywhere for all the obvious reasons. And without local government agreement their hands are tied by head office's rules. Better not to spend the money, then, imho, but with domestic constituencies focused myopically on 0.7% GDP etc the tap stays open, pouring cash into the least effective pipe of them all.

Hi Tobias, Thanks for a very good comment. I agree that capacity is a significant problem in many cases, and I certainly wouldn't want to be seen to be downplaying it. And yet at the same time I think that if there are also serious problems with the incentive structure that trickles down from the political sphere improving capacity still won't change things that much, at least if we're not doing what we can to improve the incentive environment at the same time too. Surely if capacity building was a magic bullet would would have seen better sustained results from our work in this area? On the case of structures - excellent point. I wrote a post on Andrews and Pritchett's paper on this a couple of years ago: https://devpolicy.org/isomorphic-mimicry/ And I agree entirely with: "One approach that might help avoid both capacity problems and some political economy issues is ensuring that bureaucratic reform is informed by what can work to achieve the relevant outcome within constraints, rather then attempting to impose hugely ambitious imported models (accompanied by endless capacity-building programs) regardless of their appropriateness." Thanks for adding considerably to my original post.

The International Donors and Clients Community (especially in Africa) must deployed ICT and provide the appropriate Training, and used ICT like the business community has, to be able to have Return On Investments -ROIs. Gadema K. Quoquoi

You are absolutely right Terence - it is very hard for aid agencies, advisers, NGOs etc to do not least because it does not fit easily if at all into project design and delivery plans and it is also very hard to quantify. If and when it does happen it is almost always because the stars have somehow self-aligned rather than having been project-managed into it. I guess it is about remembering the importance and value of this and then having your (individual and institutional) antennae attuned in order to be able to contribute when the time is right.

Terrence – Thanks for another thought-provoking blog. I think most people would agree that problems with bureaucracies in developing countries can’t be reduced entirely to capacity constraints (the huge political economy literature, for example, supports your points about incentive compatibility). At the same time, I think there’s a logical problem with citing evidence that we aren’t good at building capacity as proof that we make too much of the problem. To me, capacity constraints are a problem especially because we aren’t particularly good at addressing them. In my experience, the severity of capacity constraints varies hugely by country and even by institution within countries (as you note) – so blanket conclusions about their relative importance are probably difficult to sustain. While of course never the only problem, I think there’s a lot of evidence (including our recent World Bank Policy Research Working Paper) that capacity constraints are a significant issue in many Pacific countries, given the inherent problem of finding enough people in small populations and small public sectors (in absolute terms) with the specialized skills required to support all of the various technical functions that states are expected to perform (provision of public goods, demonstrated compliance with fiduciary norms in management of public funds, maintenance of a sound macroeconomic and business environment, participation in various international fora, coordination of and reporting to donors). Some of the Pacific Island countries only have one or two qualified accountants in the whole country – this inevitably creates binding capacity constraints to implementing public financial management systems that meet international standards, for example. I’d also add another item to your typology of causes of bureaucratic dysfunction. This is the common donor insistence that developing countries adopt the same institutions that operate in the same ways and do the same things as institutions in developed countries, too-often without consideration of the appropriateness of these institutions given available capacity, political-economy context, affordability, or fit with existing, local institutions. Matt Andrews at the Kennedy school has written a lot about this in relation to PFM, and Rodrik also talks about the dangers of “blueprint” approaches when it comes to both policies and institutions. One approach that might help avoid both capacity problems and some political economy issues is ensuring that bureaucratic reform is informed by what can work to achieve the relevant outcome within constraints, rather then attempting to impose hugely ambitious imported models (accompanied by endless capacity-building programs) regardless of their appropriateness.

The obvious impasse between the stated intention of achieving stated national objectives and the ensuring they are met is the curse of many nations. In Melanesia, the root causes of the problem is well known. Firstly, from a cultural perspective, the imperative to meet deadlines and targets can be fluid and flexible when it comes to enforcement. Responsibility and accountability are nebulous terms that often don't translate. They are culturally disconnected from reality. Secondly, when overseas aid and stated expertise is constantly made available, local incentive to achieve results often is ameliorated to the point where any personal responsibility is dissipated and watered down as to be virtually ineffectual. Lastly, unless there is political will and the power to make it happen, it simply won't. The real problem is that no one is really prepared to face reality.

Thanks Tess. I agree entirely with your comment. Thanks for posting it, I think it's a useful partial counter to my post which could be read as being overly pessimistic. The only caveat I'd add is that identifying and supporting normative champions requires an adroitness and soft touch which, through no fault of their own, is often very hard for govt. aid agencies (and even NGOs). Hard, but not impossible though, and definitely worth doing.