By the end of the year, Kenya is meant to have revised the way it shares money among the 47 counties. This should be a time of intense public debate about how Kenyans share resources in a way that is fair and constitutional.
The revisions are primarily driven by the requirements of the Constitution, which require periodic reviews of the formula (normally after every five years, but after three years during the transition to devolution).
This first review will focus on improving the accuracy and fairness of the formula designed by Commission on Revenue Allocation (CRA) in 2012. Why discuss revenue sharing at all?
Some believe the only fair way to share funds is to give an equal amount to everyone. If there were Sh235 billion available for counties, each county should get Sh5 billion (235/47). If so, there would not be much to discuss. However, most of us believe that a health clinic that sees 100 patients a week needs more than one with 50 patients, and that the fairest way to share money is to take into consideration differences among people, facilities and counties. Having agreed to look at these differences, we then have to agree how to do so.
In thinking about what is fair in revenue sharing, we normally think of three key principles. The first is called “need,” and it is based on the premise that a county that needs more resources should receive more resources, or that a county with more sick people needs more money for health care.
The second principle is capacity. Here we mean that those counties with more of their own resources should finance more of their own development and receive less from the overall pot. This is equivalent to saying a rich family should pay more for services, while a poor family should receive more help from the state.
Our third principle is effort: we also believe that counties that make more of an effort to finance themselves should not be punished for doing so. We want people to be rewarded when they make efforts to raise money on their own and spend it wisely.
The first formula focused almost exclusively on need. It tried to measure the needs of counties by looking at their population, their poverty levels and their land area. These variables do affect need. A county with more people is likely to need more health services. The problem with these variables is that they are not directly linked to the services that counties provide.
So while population is a key driver of health services, it is better to look at visits to health facilities and the risk of falling ill in different counties to get a better sense of how much counties “need” to provide health services.
This is what South Africa does. Similarly, one could look at other core county services, such as agriculture extension services, and determine costs based not on population, but on the number of farmers.
It is also important to distinguish two types of need that the current approach confuses. One is what we might call ongoing need. This is what it costs objectively to make services run every day for the existing population of a county. Another type of need relates to compensating some counties for marginalisation in the past.
These counties “need” funds to fill gaps in infrastructure, for example, because they have been unfairly treated in the past. Normally, we address these two types of needs differently. The first is dealt with through a formula, driven heavily by population-related factors.
The second is dealt with through targeted transfers for infrastructure, such as a roads grant. In a sense, this is the role of the Equalisation Fund in the Constitution, rather than the revenue sharing formula. But the Equalisation Fund is too small, so we may need a larger Infrastructure Transfer to deal with these backlogs.
Fairness dictates that we look at capacity and effort as well, because there are large differences in the size of county economies and their tax bases. A county like Nairobi can certainly raise more than a county like Wajir, and this should be recognised in some way.
Many countries look at capacity by estimating what all counties could collect if they applied a similar tax to their tax bases. This approach avoids a situation where counties are punished when they collect more taxes.
When it comes to effort, the idea is that two counties with similar economies and tax bases might collect different amounts because one county might make more of an effort to enforce the tax laws, or might use more creative taxes and charges to access the base.
One way of looking at effort is to see how much counties increase their tax collections over time and to reward those counties that have higher increases. Effort can also be about how well counties manage their funds, and whether they follow the law in raising and spending money. CRA has proposed using a Fiscal Responsibility Index for this purpose and invited debate about how to craft it.
At this point, rather than measure fiscal responsibility directly, counties should perhaps be assessed on their systems. Those that are more transparent and facilitate greater scrutiny of their budgets could be rewarded.