DEVOLUTION: A dream deferred?

Log into your member account to listen to this article. Not a member? Join the herd.

Views on devolution tend to be of two types. For some, devolution is the best thing since hotel buffets invented sliced ugali that you can eat with cutlery. It has led to new roads, new dispensaries and greater self-determination. For others, we are well on our way to Sodom and Gomorrah, with corrupt county officials leading the way. Devolution has resulted in wasted resources and incoherent development policy.

So which is it? Are we on the road to hell, paved with good intentions? Or the superhighway to resolving Kenya’s most intractable challenges?

You won’t be surprised that I think the truth is probably somewhere in between. I examine three common statements about counties and offer my view on whether or not we should believe them.

Counties are bringing development closer to the people.
Counties are failing to collect revenue, undermining the potential of devolution to lead to economic transformation.
The national government is trying to kill devolution.

1. Counties are bringing development closer to the people.

Counties have received hundreds of billions of shillings in the last five years. This money has been distributed to them on the basis of a redistributive formula proposed by the Commission on Revenue Allocation (CRA) and approved by Parliament. It would take an incredible amount of corruption and incompetence for this amount of money to have no impact on development. Indeed, there does seem to be evidence that at least some of this money is building new roads, new health facilities, new early child development centres, and so on. Because of its redistributive nature, the formula ensures that these facilities are being constructed, at least partially, in marginalised areas of the country that need them the most.

Kenyans have a tendency to overestimate the role of capital projects in development. This tendency, which, to be fair, can be seen in many countries, leads citizens to expect government to focus its resources primarily on building things, rather than on ensuring that services are actually delivered.

Yes, it is true that there is waste and corruption. But it is not credible to believe that all of the money channeled to counties is going to waste. A more important question, however, is whether the investments that are being made by county governments will actually yield development. To understand this, it is important to distinguish between investment in capital projects and true development, a distinction that is frequently ignored in public debate in Kenya.

Kenyans have a tendency to overestimate the role of capital projects in development. This tendency, which, to be fair, can be seen in many countries, leads citizens to expect government to focus its resources primarily on building things, rather than on ensuring that services are actually delivered.

I believe that devolution has so far delivered and increased the number and the reach of capital projects, which we might call the hardware of development. The more difficult task, however, is to ensure that the software is also delivered. This means planning for and using the recurrent side of the budget: personnel and operational costs that ensure that capital projects deliver.

It also involves redistributing human resources, recruiting and retaining skilled professionals, and demonstrating managerial capacity. In this regard, we have a long way to go. My experience watching county executive committees interact suggests exceedingly low managerial capacity. The decision by incoming county governors in 2017 to indefinitely suspend civil servants working in their counties will have the perverse effect of alienating and driving away the best of these officials while ensuring that those without exit options remain. Therefore, while I believe that we can say that devolution has brought capital closer to the people, we cannot yet say that it has brought development closer.

While I believe that we can say that devolution has brought capital closer to the people, we cannot yet say that it has brought development closer.

There is one more point to be made here, and it concerns distribution. While the national CRA formula endeavors to be equitable, many counties have introduced equal sharing formulas across their wards. Given massive intra-county inequalities, this is not fair. Data from the 2009 census shows that inequalities between counties are much less than inequalities between wards. For example, the county with the best access to improved waste disposal has 15 times the level of access of the county with the least access. But the ward with the highest access to improved waste disposal has more than 500 times the level of access of the ward with the least access.[1] While equal distribution of resources ensures that every ward gets something, which might be better than what existed previously, it does not ensure that those who need the most get the most, a cardinal principle of equity and one that must be observed if we are going to bring development closer to the people for whom it has been furthest away in the past.

2. Counties are failing to collect revenue, undermining the potential of devolution to lead to economic transformation.

From the start of devolution, observers have been complaining about county own revenues falling short. Generally, these complaints have rested on false premises. It is true that some counties failed to collect as much as local authorities in the first year of devolution, but on average, in 2013/14 counties raised more (Ksh 26.3 billion) than local authorities raised in 2012/13 (Ksh 24.5 billion). These figures have continued to rise, until 2016/17, when they declined modestly.

There are several things to note here. First, counties largely have the same revenue-raising powers as local authorities did before them. There is, therefore, no obvious reason why they should suddenly collect much more than these authorities did.

The elephant in the room, however, is that counties lack revenue-raising powers. While it is true that a handful of the economically more advanced and more urbanised counties are under-collecting property tax and probably business permits as well, which are the biggest sources of revenue for most counties, the truth is that most counties are more rural and do not have economic structures that lend themselves easily to greater taxation of this type.

Second, these revenue sources were highly volatile from year to year under local authorities. There is no reason why this should have suddenly changed either.

Third, while many have interpreted poor revenue collection against target by counties as evidence of pilfering, the truth is rather more prosaic: counties are trying to make their revenues look better than they are by deliberately overestimating their collections. In 2016/17, even counties that saw their revenues rise fell short of target in most cases. For instance, Machakos collected only 44 per cent of its target in 2016/17; in 2015/16, it collected 47 per cent. Obviously, this is part of a repeated pattern of inflating revenue estimates. Overestimating local revenue can have several benefits: it can make it appear that the government is able to spend more without creating a deficit, and it can increase the total size of the budget, thereby, decreasing the share that goes to recurrent spending and helping counties to meet legal requirements to limit such expenditure.

The elephant in the room, however, is that counties lack revenue-raising powers. While it is true that a handful of the economically more advanced and more urbanised counties are under-collecting property tax and probably business permits as well, which are the biggest sources of revenue for most counties, the truth is that most counties are more rural and do not have economic structures that lend themselves easily to greater taxation of this type. This leaves counties relying more heavily on user fees for services, such as charges to patients at health facilities. While these kinds of fees can help to fill budget gaps, they can also be regressive and lead to a decline in access to services for the poor. We should be wary of pushing counties too fast and hard into such approaches. It may be smarter to think about things like a surcharge on VAT or income tax that would go to counties. This would require national legislation, but is not unthinkable.

3. The national government is trying to kill devolution.

There has been much hue and cry about the deliberate attempts of national institutions to kill devolution. This is understandable given what happened to earlier attempts to devolve power after independence, when regional governments were scrapped and power was centralised in an Imperial presidency. Nevertheless, we must still ask whether indeed the national government is trying to strangle counties.

What seems more plausible than a concerted effort to kill off the devolved system is that there is a general level of incompetence around how to manage such a system sensibly.

In favour of this hypothesis, consider the following. One piece of evidence is the failure by the Transition Authority to fully address various transition issues, including a full unbundling of county functions in the constitution, which should be carried out by the national government. Who exactly was to blame for this failure is a matter of conjecture, but the national government was certainly not helpful. And this failure has made it difficult to address some rather significant challenges, such as the need to devolve, or at least to reform, various state corporations in the roads, water, and energy sectors. Maintaining these corporations means maintaining control and money at the national level.

Another piece of evidence is that money seems to flow to counties somewhat slower than it flows to other government agencies each year. It is not entirely clear why this is so. One possible reason is that someone is trying to make counties look bad. Another is that counties themselves have struggled to approve budgets and, therefore, have created their own delays in accessing funds. Most have bank balances during the time they are not accessing national transfers, though they may not be able to access these funds against a budget without approval from the Controller of Budget. Counties are also receiving a smaller share of increasing total revenues than the share going to the national government, consistent with the “kill devolution” mentality.

However, much more could be done by a national government intent on killing devolution by starving counties of resources. Some of the claims for more resources for counties are fantastical; there is no basis for claims that as much as 45 to 50 percent of national revenue should go to counties (though we may wish it to be so). To the extent that we can estimate the cost of county functions, they do not reach this level of funding.[2] The fact that counties do not receive this level of financing is, therefore, not evidence of anyone wanting to kill devolution.

Of course, it may be that we want to give more funding to health or early childhood education than we have traditionally spent on these areas, and this could lead to additional funding for counties. However, notice that the logic here is to determine the functions we want to spend money on and then finance them, rather than designating an arbitrary total that we would like counties to receive and then trying to justify that.

The view that counties should massively increase their own revenue collections seems to rest on a misconception of types of revenue sources counties can exploit, and what we know about them.

What seems more plausible than a concerted effort to kill off the devolved system is that there is a general level of incompetence around how to manage such a system sensibly. The decisions by national institutions that have led to escalating wages for health workers and other county officials seem to be less about killing devolution than an inability to cooperate across levels of government to control costs. The failure of the National Treasury to provide proper guidance to counties on, for example, how to implement programme budgeting, is not about wanting counties to fail, but about lack of capacity and vision at the national level. Even the national government has fallen short on these reforms, so it is not well-placed to advise counties.

In conclusion, there is quite a bit of sound and fury, but it does signify progress of sorts. Counties are making the first important steps toward bringing development to the people, but they are more apt at the capital than the operational side of things. The view that counties should massively increase their own revenue collections seems to rest on a misconception of types of revenue sources counties can exploit, and what we know about them.

If we want counties to be less dependent on national transfers, we will have to rethink county revenue sources and give counties additional legal ways of raising funds. And finally, while the national government has not treated counties well, this has more to do with incompetence and lack of coordination than a deliberate attempt to kill devolution (though there may be specific actors with that agenda, including state corporations protecting their own turf).

Friends of devolution would do well to remember that devolution alone is not a substitute for reforming the central government. Too much attention on counties leaves no one minding the store. If we want devolution to succeed, we must also invest in reforming the national government.

References