As Grossman and Lewis show in this paper, a lot of decentralization efforts in the developing world have not resulted in greater capacity or control of policy by the devolved units. But there are exceptions, like in Kenya where since early last year 47 subnational units (Counties) have come into existence with a constitutionally mandated (at least 15%) sharing of ordinary revenue between Nairobi and the counties. In the 2014/5 financial year about 32.4% of the most recently audited ordinary revenues (2011/12) will go to the Counties. The revenue sharing is governed by a strict formula (with population, geographic size, and poverty rate weights) in order to limit the discretionary powers of officials at Treasury.  More crucially, the County heads – a governor and a County Assembly – are directly elected, not appointed as is the case in most of the instances of “fake decentralization” noted by Grossman and Lewis.

The quote above from the governor of Mandera County in the northeast of Kenya sums the potential impact of Kenya’s brand of devolution.

Yes, in the interim there will be massive corruption, insufficient absorption, weak capacity for implementation and the like.

But the important point is that Kenya’s new government structure has 47 capitals handling billions of shillings each year. If all else fails, the system will at least create strong politically autonomous regional elites with sufficient power to check Nairobi. And that is a fantastic thing. Already a few governors (including Nairobi, Machakos and Bomet) have broken ranks with their sponsoring parties, evidence that the interests of the national parties and County governments will not always be aligned. And if the last two fiscal years are any indication, the political pressure on the national government to overshoot the 15% minimum requirement will continue to hold. Those perceived to be enemies of devolution will be punished at the polls.

The new system also has another plus: Kenya now has 47 training centres for the job of chief executive. Governors who do well – like the Governor of Machakos – will become very strong contenders for State House in the not so distant future (Also more work for me to study inter-governmental political careers!!)

My biggest concern about the new devolved system of government is its potential impact on the coercive capacity of the Kenyan state (recently the state has been at sixes and sevens in response to rising insecurity and sporadic terror attacks). If there was ever Kenyan exceptionalism in Africa it was on account of its post-colonial inheritance of a strong state. Since for political reasons security cannot be devolved (latent centrifugal tendencies still exist at the periphery), the centre must still guarantee security of life and property. My hope is that the ongoing restructuring of the Provincial Administration will not be as large a swing as to completely gut a system (reviled or not) that helped hold the country together over the last 50 years.