Governing Decentralized Innovation

Owen Barder, a development practitioner based in Ethiopia, has some illuminating thoughts on the need for more effective coordination and governance in the field of development. He starts the conversation by questioning why it is necessary.

Why do we need any kind of governance for development cooperation?  It is reasonable to be sceptical about any kind of authority...My inner anarchist would really like development cooperation to be like the internet.  Could development programmes be best left with as little interference as possible?  Why not leave the people and organisations involved, donors and recipients, to solve their own problems, leaving them freedom to innovate, adapt and grow?   Let people form their own networks and own arrangements that work for them?  Given that aid consists of arrangements freely entered into between consenting adults, and between sovereign nation states, why do we need any kind of interference in those relationships?

He then goes on to outline 9 reasons for developmental coordination

(a) Encouraging positive spillovers

There is quite a bit of evidence of complementarity of development activities: an increase in agricultural productivity is more valuable if there is a road to take surplus production to market; children are more likely to stay on at school if there are skilled jobs for them to take when they graduate...

(b) Reducing negative spillovers

Conversely, the activities of one development programme can adversely affect the impact of another. An aid agency or NGO that recruits a specialist from a government department undermines efforts to build capacity and government institutions...The behaviour of some aid agencies can stifle the evolution of accountability of the institutions that other aid agencies are trying to support.  The prices of scarce resources used by aid agencies – offices, houses, and vehicles – can be driven up by the presence of multiple aid agencies. The influencing agenda of one development organisations can dilute and undermine the carefully constructed agenda of another.

(c) Coordination

Some development cooperation is inherently collective – such as designing and implementing a reform programme, or filling a financing gap – and requires donors to work with their partners towards a shared goal. In these areas, the best choice for one agency may depend on the behaviour of others...Complex institutional change programmes requiring sequenced reforms will work best if donors are making their contribution to a coordinated overall programme.

(d) Preventing free-riding

Everyone benefits from a more just, safer world, with prosperous trading partners. But some countries may be tempted to ‘free-ride’ on the generosity of others. For example, a country could give relatively little aid, in the expectation that others will give more. This means it can benefit from economic development without helping to pay for it...

(e) Information sharing and learning

Aid agencies increasingly see themselves as organisations whose value is not administration but knowledge. Project administration is important, but it is also fairly straightforward; more complicated is learning what works, adapting it to different circumstances, influencing policies, sharing ideas, building capacity and shared commitment. This learning and knowledge sharing is needed across aid agencies as well as within them, and it is clearly a public good...

(f) Specialisation

In markets, firms specialise in the activities in which they have a comparative advantage, because this is how they will make most profit; the division of labour is the consequence of this specialisation.  (Note that specialisation drives the division of labour, not the other way round). Some kind of mechanism is needed to push aid agencies into specialising in the countries and activities where their opportunity costs are lowest, sometimes in the face of countervailing political pressures to spread themselves more thinly.

(g) Investing in public goods

Public goods tend to attract too little investment because, by definition, the benefits are widespread, so individual countries and agencies do not have a strong enough incentive to spend money on them. For example, the knowledge created by research and development is a public good, with benefits that spread far beyond the country that paid for it. There is significant global under-investment in development public goods such as reducing climate change, financial stability, norms and standards, evaluation, and knowledge...  

(h) Keep incentives aligned to development objectives (address prisoners’ dilemma)

Donors invest in foreign assistance for a variety of reasons. Some of these are shared global objectives: for example, reducing absolute poverty is a shared moral endeavour. All nations benefit when the world is less unequal, when world trade increases, and the risks of conflict, organised crime, drugs and disease are reduced by economic advancement. But donors are also willing to invest in foreign assistance in pursuit of their own national interest, sometimes to the disadvantage of other countries: for example, to win commercial contracts, gain geopolitical strategic advantage over rivals, or to stake a claim to oil or mineral resources. The aid system is more effective at reducing poverty to the extent that the shared goals are pursued in preference to the competitive interests of donors; but it is often not individually rational for donors to do this.  ...

(i) Protecting long-term goals from short-term pressures

Most well-governed nations have evolved institutions to protect the nation’s long-term interests from immediate political pressures. Examples include written constitutions, an independent judiciary and independent central banks. In development, long-term interests are frequently sacrificed to short-term expediency. For example, imported food aid may reduce hunger immediately, but at the expense of undermining domestic agriculture production in the long run. Donor-managed projects may provide services to the poor with lower risk of corruption and theft, but at the expense of undermining the capacity of the country’s institutions to manage their own services. A good system of governance ensures that long-term needs are not ignored in the interests of short-term pressures. 

While outlining these reasons for more effective governance, he then warns of the pitfalls of bureaucracy and conformity.

But we should not make the mistake of equating governance with government. Too often in development we assume that more effective cooperation comes about by committee.  Because we need better governance, it does not follow that solution to these problems is some kind of world governing body for development cooperation, nor that we need more communiqués, declarations and committees. Depending on the context, governance can mean a combination of social norms, standards and values, regulations, taxes and subsidies, decentralised decision making within the right framework of incentives and accountability, and mechanisms for transparency, information sharing and accountability.

 You have noticed that there is a logical step missing in this argument so far: we may be right that the development system suffers from lack of governance, but it is possible that the cure could be worse than the disease.  What if well-intentioned steps to make the system work better actually make it worse?  We could end up creating a mono-culture of ideas, donor monopolies, intensive bureaucracy and endless harmonisation meetings, so reducing, rather than improving the effectiveness of development cooperation. This is clearly a risk. Just as the solution to failures of private markets is not necessarily nationalisation, so the solution to the governance failures in aid may not be greater bureaucratic control. We have to try to address the problems, but take care not to undermine the scope for innovation, initiative, and especially ownership by the citizens of developing countries themselves.

I find this discussion interesting beyond its application to development aid. In some ways it parallels the recent efforts in the Baha'i community of balancing the need for decentralized innovation and broader governance/coordination. I believe that the Baha'i world can gain a lot from studying the successes and failures of these other efforts, most of which have orders of magnitude more economic resources at their disposal. At the same time, it speaks well of the Baha'i community that these discussions are just beginning to be seriously made in the development aid world, without much time yet for application.  For at least 15 years, the Baha'is have been experimenting and gaining practical knowledge through the institute process, and can now start to offer it as a model for others. This in addition to its spiritual teachings, which of course are inseparable.