Kenya’s Commission on Revenue Allocation released their proposals for revising the revenue sharing formula just minutes before this column was due.

These remarks therefore represent my first take on the recommendations, which we shall discuss in greater detail over the coming weeks.

As expected, the proposed formula is very similar to the current one. The core parameters (97 per cent) remain the same and have only been tweaked. This is consistent with the fact that the Commission must endeavour to comply with the Constitution, which requires stable and predictable revenue streams for counties. 

This is also a likely outcome of the entire exercise because, while people may not like the current formula, it is hard to get governors or senators from 47 counties, each of whom will lose or gain from any change, to agree on anything beyond minor modifications to the status quo.

For this reason, CRA has also stuck to the idea of having a single transfer, though we have long suggested that it would be more sensible and more transparent to have a set of transfers with different formulas, rather than putting everything into one basket.

That said, a few new things are on the table. To accommodate two new parameters, “Development” and “Personal Emoluments,” the CRA is recommending a reduction in the poverty parameter from 20 per cent to 18 per cent.

In general, the poverty parameter is not a great measure of the direct need for services, and we prefer approaches that take into consideration specific areas of marginalisation, such as access to infrastructure.

The development parameter, weighted at one percentage point, does in fact do this, although it incorporates a huge number of factors into a very small weight: Roads, water, literacy, sanitation, electricity and maternal health.

While the weight given to these factors is too small, it is a step forward to incorporate measures like these that relate directly to county functions, rather than poverty.

I would have preferred a larger drop in poverty and population to accommodate a higher weight for these measures, but also would have separated them into those that focus on meeting ongoing service needs, and those that focus on addressing marginalisation. The CRA approach only looks at the latter.

To my surprise, CRA is also recommending a reduction of the fiscal responsibility parameter, which has never been measured, from two to one per cent. In our conversations with people around the country, many wanted to see the weight of this parameter increase.

CRA appears to have bowed to the demands of the governors in reducing it, though it seems likely that the Senate will want to revisit this. The proposed measure for fiscal responsibility awards counties an equal amount to set up their financial management systems, and a varying amount based on the degree to which they finance their budgets from local sources. 

It isn’t clear why an equal share is needed to set up financial management systems, since this should be covered by the basic equal share (see below). The decision to focus on own revenues is consistent with international practice, but the measure — own revenues as a share of the total budget for a county — tends to favour the wealthier counties.

I would have preferred a measure that looks at tax effort measured as the percentage change in own revenues over time, which would reward poor and rich counties alike for their efforts.

The proposal from CRA may reward counties with greater fiscal capacity. It will also be difficult for poor counties to benefit from this parameter because, being poor, the share of their budget from national will always be large.

CRA has decided to keep the weight for the basic equal share at 25 per cent, bowing, I think, to political pressure. In our view, a basic equal share should only cover those things that are identically priced across counties, such as the salaries for governors, deputy governors and some basic running costs.

Everything else varies from county to county based on other factors, such as population, and should be catered for by other parameters.

Counties with smaller populations, smaller land area and few poor people all benefit disproportionately from a high basic equal share, and there are a number of counties in that category that will fight for it to remain high.

The parameter for personal emoluments is designed to help a number of counties, like Nairobi and Mombasa, that have inherited huge staff costs. That is a good idea, but we had proposed that it be dealt with separately from the overall formula through a conditional grant.

This is not only more transparent, it also allows for more frequent adjustments, and for the use of conditions that incentivise a reduction in these wage bills over time. Nevertheless, CRA is to be applauded for recognising this problem and proposing a solution.

Jason Lakin is country director for Kenya of the International Budget Partnership. E-mail: [email protected]