The mechanics of building strong states in 21st century Africa

On models of state-building and their applicability to the African context

Thank you for being a regular reader of An Africanist Perspective. If you haven’t done so yet, please hit subscribe to receive timely updates on new posts along with over 35,000 other subscribers. New regular content is free. Book reviews and the archives are gated.

I: Getting serious about building the African state

My previous post discussed the need to rethink approaches to conflicts in the Sahel, making the case for aggressive technological, operational, and organizational upgrades of Sahelian militaries. A number of people read my arguments therein as a bellicist case for state-building in Africa. They were not. This post lays out my views on how to go about state building in Africa in the 21st century.

This post’s key arguments are as follows. There are multiple models of state-building, depending on the prevailing dominant ideas and nature of the international system. Therefore, Africa’s would-be state-builders shouldn’t fixate on old models (especially the bellicist models) that aren’t particularly suited to this moment in time. Furthermore, state-building ought to be principally about subnational self-government and service delivery. To this end, form should strictly follow function in order to avoid problems of over-developed yet terribly ineffective state structures.

Before proceeding further, please allow me to vent a little. Too often people bring up James Scott and other critiques of statist overreach in conversations about state-building in Africa. This “gotcha card” gets played as a critique of the very idea of strong states. Oddly enough, those most likely to play this card are people who live in or have benefitted immensely from the trappings of high state capacity. They were immunized as kids. All their lives they’ve taken for granted basics like running water, indoor plumbing, safe waste disposal, access to electricity, safe transportation infrastructure, and the like. They attended public schools, or universities heavily subsidized by capable states. They take pavement/sidewalks for granted. They mostly live in working economies that, despite challenges here and there, have historically delivered on the dream of mass prosperity. They mostly don’t have to worry about their physical safety all the time. And above all, their societies afford them personal freedoms and the best opportunities for self-actualization that humanity has ever known.

These are the people who will reflexively invoke Scott’s Zomia in discussions about state-building in Malawi, Sierra Leone, or South Sudan; and think that they are being very clever while at it. Let’s stop that.

Don’t get me wrong. There are lots of valid critiques of state power. But it’s one thing to criticize and (presumably) want to improve upon the state in an intensely-governed context or when states overplay their hand (e.g., the misguided high scientism that Scott critiques), and quite another to extrapolate from that and make a case against state-building in countries like the Central African Republic or Mali.

Scott was a fantastic scholar who offered much-needed caution against blind belief in state power and central planning. In particular, his works on weapons of the weak and everyday acts of resistance to unaccountable domination are a reminder that coercive interventions by public authorities must always scale a high bar; and that resisting tyranny is an honorable imperative. That’s the reasonable read of Scott, who lived in relatively intensely-governed Connecticut — with its 169 incorporated municipalities and nearly $29b state budget for a mere 3.7m people.

A terribly under-governed (and therefore under-served) continent

Much of Africa, on the other hand, is terribly under-governed (a fact that gets ignored due to the obsession with ritual electoralism and regime types). From per capita public spending (about 1/20 of OECD figures in most places), to civil servants per 100,000 residents (lowest of any major region), to intensity of subnational government, the Continent lags much of the rest of the world on the degree of government (and has been for a while — see table below).

African countries have historically had comparatively very small governments. Source: Mkandawire (2015)

Not enough people appreciate just how badly under-governed African countries are (even before we begin to talk about the quality of governance). The government of Liberia (pop 5.6m) has a smaller budget ($1.2b) than my university ($2.1b). Nigeria (pop 220m) all but lacks a coherent system of government and service delivery below its 36 states. A 2024 Supreme Court ruling may finally force Nigeria’s political elites to get serious about the country’s 774 local government areas (LGAs). For years hundreds went without elected officials, with governors at liberty to appoint interim commissioners and appropriate LGA budgets. The average LGA budget ranges from $2m-$6m, mostly cash from petroleum rents since most of these units scarcely raise their own revenue. In contrast, Brazil (pop 214m) has 5,570 municipalities that actually work as units of subnational self-government (with warts and all). The typical mid-sized Brazilian municipality has a budget of between $15m and $25m. A few Brazilian subnational units have budgets that dwarf those of dozens of African states. For the example, city of São Paulo has a budget of $22b, which is almost half of the entire federal budget of Nigeria ($45b).

One of these world regions is not like the others. Source: Our World in Data

Ignorance about the degree of under-government on the Continent yields odd arguments. For example, in Kenya (pop 56m) you often hear otherwise sane people complain that, at 47, the country has too many economically-unviable subnational units (counties). The same people usually bring up the fact that the U.S. has 330m people and only 50 states. This, in their view, is a case for fewer counties. What these people never bother to look up is the sheer intensity of government in the U.S. — with its 3,144 counties and about 19,500 incorporated municipalities (about three quarters of which have about 5000 people).

The point of these comparisons isn’t to suggest that Nigeria and other African countries ought to throw money at more government as an end in itself. Far from it.

Let’s be clear. The point of self-government isn’t just to spend money or create public sector jobs. Beyond a certain threshold of “fixed costs,” the scope and depth of state functions ought to co-evolve with levels of economic and social complexity. This is only possible when societies build the platforms for collective action and growth that they can afford. Indeed, I’d argue that a core driver of the widespread disillusionment with government across the Continent is the challenge of isomorphic mimicry:

[which is] the adoption of the forms of other functional states and organizations which camouflages a persistent lack of function. Second, an inadequate theory of developmental change reinforces a fundamental mismatch between expectations and the actual capacity of prevailing administrative systems to implement even the most routine administrative tasks. This leads to premature load bearing, in which wishful thinking about the pace of progress and unrealistic expectations about the level and rate of improvement of capability lead to stresses and demands on systems that cause capability to weaken (if not collapse).

Thus when most people think government, they imagine corruption, big cars, sinecures for cronies, fat salaries, expensive benchmarking trips, endless summits, and obscene allowances. Not pro-growth service delivery. Not a conscientious attempt at self-government. Not any hint of historical conscientiousness or strategic depth.

I should also note that the goal of state-building ought not be to maximize the number of subnational units.

Consider the example of Vietnam (pop 113m). To start with, Vietnam’s public budget of $118b dwarfs Nigeria’s $45b. Before its current commendable reorganization, the country had 63 provincial-level units and over 10,000 communes. The reorganization yielded 34 provincial-level units and 3,321 commune-level administrative units. Districts were abolished. The reorganization is meant to streamline pro-growth provision of public goods and services. This is in contrast to the careless administrative unit proliferation that we’ve seen throughout the Continent over the last three decades.

Faster and broad-based economic growth is a big part of the answer

It follows from the discussion above that the process of state-building in African countries should proceed on two fronts. First, functions need to be clearly defined on the basis of objective realities; and form must strictly follow realistic functions. Government across the Continent doesn’t have to look like their counterparts in the OECD (despite what the donor-funded governance advisers will tell you).

This is the only way to (1) make sure that government actually delivers by matching functions with capabilities and that policies meet people where they work and live; and (2) keep costs under control.

Second, government must be deliberately designed as a vehicle for collective action towards socio-economic advancement, with specific attention to economic growth and development. Not for distributing rents among elites, or facilitating external exploitation. Relatedly, structures of self-government must be present wherever people live. Organized collective action enables communities to thrive.

The foregrounding subnational governments and service delivery thus far is intentional. Above all else, state-building ought to be about rationalizing self-government and local service delivery. Period.

Of course this process has to culminate at the national level, where states play a critical role in national coordination of economic and social development (including trade), helping societies navigate global affairs, as well as national defense. Notice that the critics of state-building in Africa usually ignore the important role of subnational self-government. This oversight can only be explained by anti-statist ideologies that spring from (well-meaning) anti-colonial sentiments. Yet what the maximalist ideologies end up doing is to erase African agency and histories of state-building and self-government.

We don’t have to pretend that colonialism invented the African state. States existed in Africa well before the 80 odd years of colonial humiliation beginning in the late 19th century. And we know quite a bit about how these worked, including what needed improvement. Rather than waste intellectual firepower boxing shadows, it would be beneficial to be forward-looking and clear-headed about the very urgent need for robust state-building throughout the Continent.

As I’ve argued before, only strong African states will be able to correct the lasting legacies of colonialism and shield African societies from the ills of an increasingly unstable world. And, to be honest, it’s patently absurd that the people who decry historical exploitation of the Continent’s peoples by outsiders tend to espouse the strongest anti-statist views. This when the same people openly admire the impressive state-directed achievements of perceived ideological allies on the global stage.

People shouldn’t be allowed to get away with this sort of ideological confusion.

II: Competing visions of state-building in Africa

Which brings me to the question of which models should guide the process of state-building on the Continent. The practicalities of state-building in Africa present two related sets of questions. First, what models of state-building ought to guide the process? And second, how should the preferred model be applied to the African context at this point in history.

The nature of the international system and ideas largely determine the best models of state-building at any point in time

While the bellicist model has long influenced the discourse on state-building in Africa (and elsewhere), it’s worth recalling that there are multiple other models out there (including some variations of the bellicist model). The 18 books below represent a sample of these models (the list isn’t exhaustive; it’s books I have read over the years and found useful on this topic).

Rather than summarize each book in detail, below I summarize the main takeaways from the lot:

(1) the nature of the international system (which varies across time and space) conditions what kind of state-building strategies will predominate. In addition, the interaction of domestic and foreign factors shape the process of state-building. Therefore, we shouldn’t try to shoehorn specific pet models of state-building in every case or across time.

For example, when norms were more permissive of wars of conquest and border changes, the bellicist model held sway in structuring state-building incentives (supported by specific demographic and geographic realities). In other times, organizations with reach across polities (like the Church) invested in and facilitated the spread of the building blocks of strong states — and therefore shaped what was understood as the ideal forms of state-building. Yet in other times and places, geopolitics (e.g., the Cold War) can facilitate collaborative state-building by neighboring states (in ways that both support and contradict the bellicist model of state-building).

(2) The process of state-building is political, and has real distributional consequences. States are coercive. State-directed collective action invariably requires involuntary compliance. Would-be state-builders aren’t always benevolent, and often stand to disproportionately benefit relative to the rest of society. Concentration of power and authority increases the risk of tyranny by entrenched factional interests. It’s very difficult to come up with complete contracts to constrain ruling elites and public officials. All these things are true.

Therefore, the process of state-building involves constant negotiations over the distribution of social, political, and economic power. Theorists of the state long recognized these features of state-building, and came up with models of understanding the trade-offs involved (see social contract theorists).

(3) Ideas matter a great deal. Which means that narratives about the socio-cultural and moral economy bargains that underpin the “social contracts” that make state-building possible are absolutely important. Ideas help societies answer related questions, such as what’s the point of having a state? What tradeoffs are necessary to build a well-ordered society? What do states and ruling elites owe their citizens? What do citizens owe both society and the state? Who belongs as a citizen?

Unfortunately, most African countries are yet to ask, let alone answer these questions. Instead, they merely go through the motions of mimicking highly functional states.

(5) State-building is an uneven and perpetual process. Therefore, state capacity can vary over time (in both directions) and across space.

Here, three books offer important lessons on African cases. Boone’s book (the most underrated book on state-building in Africa), describes state-building as a negotiated/discursive interaction between national governments and historical legacies of statehood at the subnational level. The outcome of this interaction determines both the form and strength of state administrative/bureaucratic structures. Herbst presents a more static model — that African state capacity is concentrated in capitals/urban areas and radiates in concentric circles. However, I would argue that even Herbst ought to be read as description of spatial variation in state-building. State-building efforts have to start somewhere, and there’s no reason why over time those circles of high state-capacity shouldn’t reach state borders. In other words, Herbst shouldn’t be read a static description of an immutable feature of African states.

McDonnell’s excellent work on islands of bureaucratic excellence also fits into this model of uneven development of state capacity. Again, you have to start somewhere. And rather than wait to build the most effective bureaucratic-administrative state in Ghana, one can start with building a highly effective agency in Ghana and then consciously scale the proven bureaucratic-administrative forms and modes of operation. In other words, islands of excellence need not remain islands forever. Instead, they should be sites for state-building through learning by doing (this works best when forms are strictly aligned with functions as suggested above).

Which models should inform African state-building efforts?

As noted above, ideas (in interaction with politics) matter a great deal for state-building. Which is why it’s worth interrogating which models of state-building people perceive to be most applicable to the African context, and the “schools of thought” behind them. For purposes of this discussion, I identify four schools of thought based on my understanding of the discourse on state-building over the years (which certainly aren’t mutually exclusive). The effort here is transparently designed to advance my preferred school of thought (which has its own weaknesses, obviously). Proponents of other schools of thought should write them up and defend them. Here are the four schools of thought:

(1) Anti-state anarchists: James Scott had a point. Being (coercively) governed by a state isn’t always pleasant or beneficial; which is why throughout history lots of people exited such arrangements and came up with ways to avoid being governed. In Africa, too, with its recent history of ineffective and often terribly autocratic states (especially under rapacious European imperialism and its successor states), has its share of people who are (understandably) invested in not being governed. This school of thought is primarily defined by its symbolic oppositionist politics, rather than any coherent positive case for how to organize African societies in a world full of states.

(2) Global integrationalists: This school of thought conceives of African states as mere extensions of the “international community” and global multilateral institutions. The ideal African state in this rendering is a dutiful believer in the promise of global compacts (like SDGs); and which acts as a perfect agent of donors/philanthropies (the so-called “donor darling”). This school considers any investment in state-building that increases African agency as wasteful, and potentially dangerous as it could yield/entrench autocracy. Unfortunately, it’s likely that the plurality (if not the largest share) of Africans in the professions (including nearly all ruling elites) are primarily disciples of this school of thought.

(3) Unreconstructed Tillians: This school posits that the African state must be strong, for both practical (effective service delivery) and strategic reasons (to avoid future humiliations as happened under colonialism). At the same time, it argues that the best way to achieve this is through a bellicist redrawing of African borders (here’s why this is a terrible idea). This, the argument goes, is the only way to instill historical and strategic consciousness among African political and economic elites. A section of this school goes as far as calling for the establishment of a United States of Africa. It also tends to attract staunch ethnic primordislists who prefer the creation of identity-based states or subnational units.

(4) Historical pragmatists: Related to the Tillians are the historical pragmatists who also believe in strong African and independent African states for efficiency and strategic reasons. However, this school acknowledges that the Tillian model isn’t always optimal across time and space. Specifically, it buys into the ideas from Africa’s independence generation who argued that state-building through emulation, practical experience over time, deliberate nation-building, and gradual intensification of service delivery is possible within Africa’s current borders. This school also believes in Pan-Africanism, but a practical Pan-Africanism anchored on strong states and whose institutional design reflects the variation in state capabilities across the region (in order to lower the costs of Continental collective action).

III: The case for pragmatic state-building

As argued at length above, state-building isn’t just about centralization or empowerment of leviathan. If anything, it should principally be about intensification of self-government and service delivery at the local level. At the same time, the process of state-building requires form to strictly follow functions. There should be no room for expensive isomorphic mimicry that detracts from the basics of context-based policymaking, service delivery, and accountability.

Along these lines, the starting point of African state-building should be at the subnational level — with subnational self-government being first and foremost about development; and focusing on human capital development (health and education) and private sector development for job creation at scale. It is only when form mirrors the objectively required functions that African states will become the platforms for socio-economic advancement that they ought to be.

Two more things deserve emphasis here. First, the process of state-building at both the national and subnational levels must not pander to ethnic chauvinists. That would be a capitulation to some of the most retrogressive elements in African states. Ethnostates are a terrible idea. Period. Second, if done well, urbanization offers a cheat code for accelerated state-building. By this I do not mean megacities. Instead, the focus should be on the more than 7,000 urban centers across the Continent with more than 30,000 residents. With all due respect to Scott, planned urbanization has the potential to boost efficient delivery of public goods and services, in addition to seeding economic dynamism. It is a shame that in most countries on the Continent urbanization continues apace in the most haphazard manner imaginable.