Service delivery is the name of the game in development today. We’ve read a lot about the great changes brought by, say, the rise of governance as a development issue, the shift from ’getting the policies right in order to facilitate growth’ to eliminating poverty, the World Bank’s dream since 1990. Thinking about how to deliver better health and education services is perhaps a natural step in this evolution. The 2004 World Development Report “Making Services Work for the Poor” marked a seminal step in this evolution. Its central concepts underpin much of our current thinking on service delivery; especially the salience of accountability relationships between state, citizen and provider, along with the associated principal-agent issues.
Just to be clear: this evolution is a good thing. Public services don’t work in all sorts of ways. Many service delivery problems are knowable and understandable; more knowledge enables those who care to come up with better solutions. But there can always be too much of a good thing.
The development community is working very hard on figuring out how to deliver better services, and how to overcome obstacles that stand in the way. For the World Bank it now comes under the heading of ‘science of delivery’, heavily influenced by the rhetoric of ‘deliverology’. There’s a lively debate at the Bank about some of the more fanciful claims of what can be achieved, as well as its blind spots. ODI’s own Kevin Watkins noted the curious absence of politics from the science of delivery. At ODI, the politics of service delivery has been a long-standing interest.
There is a catch, though. All of this knowledge being generated assumes that somebody in power cares enough to use it. For the development community – agencies, scholars, and activists – the case is clear: the more we know the better we can do our jobs. For reforming coalitions of policymakers and officials within governments, the case is almost as good. A minister would now know better what technologies to employ and how, and could figure out the politics around reforms better (though presumably knowing the politics already, if not the fancy words to describe it, is an entry requirement for most ministerial posts).
But what if nobody who matters cares? Or simply doesn’t care enough to take a risk?
Elite indifference to service delivery outcomes is common in many countries, not just developing ones. Politically savvy aid will be able to make a difference to this, at the margin. It can finance and empower local officials and politicians to a degree. But local power dynamics are what they are for a reason; just figuring out what the obstacles are doesn’t make them go away. Sometimes it just makes it easier to understand the size of the challenge. In response, aid providers can also step in and take a much more active role and deliver services either themselves or through non-state intermediaries, as they frequently do. This gets into a much bigger debate about whether massive direct aid interventions mostly ‘work’ or mostly don’t.
Even if all these interventions did mostly work, and could somehow become at least in part sustainable (a huge if), what sort of a state would that leave behind? A state with deep technical knowledge about frontline services, about supply chains, and eventually with the capacity (though not necessarily the capability) to deliver better services. The state would be supported by donors who have a better sense of the politics, helping them to catch up with the endogenous political actors who don’t need any political analysis, since they’re the ones politicking. Over time, there might be some drift amongst the powers that be towards a service delivery agenda to garner legitimacy. So the hopeful theory goes – such a virtuous circle is far from certain. But even the best-case scenario is a doughnut state: all front, no centre.
While all the attention is on service delivery, nobody seems to look at the core. How governments decide what policies they prefer, how they garner political support, how they administer the statecraft of policy coordination and implementation. These questions are notably absent from the current development debate. The centre of the doughnut is hollow (the term is not mine, I borrowed it from Patrick Dunleavy, who uses it slightly differently in reference to the UK’s Coalition government).
Close inspection of the hollow core would reveal a not so pretty sight. Political science thinking going all the way back to Max Weber suggests that states don’t exist to deliver services based on some form of a social contract – they exist to survive. Core institutions of the modern state are in essence survival mechanisms borne out of the fiercest evolutionary mechanism of them all: competition with other states. Where states, or state-like organizations, exist stably without functioning service delivery machineries, they evidently don’t need them to survive. The institutions and means of control that sustain such states are well documented, and very well covered in the recent literature (see North et alor Acemoglu and Robinson). It is a high form of hubris for donors to believe that they could transform the underlying institutions of statehood by caring deeply about, and providing funding and advice on, service delivery.
The real dilemma of service delivery is that governments that do know what they want, and are good at adapting and surviving (examples from West Bank and Gaza to Rwanda come to mind), might well be en route towards an institutional transformation that is at the core of what development is all about. They might be laying the foundations of sustainable service delivery in the future, even if they are anything but caring for their citizens today. Escaping from capability traps through adaptation is an active process – states themselves can’t be adapted. But today’s donors (and of course citizens themselves) care very deeply about service delivery in the here and now. This drives donors to meddle in the institutional fabric of government and could very well pre-empt the eventual emergence of the kind of state they wish they had (and often assume they already have) as a counterpart. One suspects that citizens can tell the difference between government-led and donor-led service delivery. We simply don’t know whether the rapid establishment of a full-service delivery suite leads to the institutional foundations of a modern state. The graveyard of organograms strongly suggests not always.
Perhaps this concern is overblown, and those virtuous cycles of service delivery and stronger states will come about, with generous support from outside. South Sudan, Afghanistan, Liberia and other countries will provide ample evidence over the coming years. I won’t hold my breath. In the meantime, it is a bit disconcerting just how comparatively little we know about what makes states and governments tick.