Dotted by pitfalls but functional
Merkeb Negash, Special to Addis Standard
In the past, series of articles published in this magazine on Ethiopia’s developmental statism focused on what a developmental policy oriented Ethiopia can’t perform – that the Ethiopian bureaucracy is far from the Weberian ideal type which demands competent, efficient and autonomous institutions that characterized the East Asian States; that the Ethiopian bureaucracy does not demonstrate the meritocratic selection process, long-run career rewards and corporate coherence that are essential to civil service autonomy. By providing more than ample empirical evidences, these articles dub Ethiopia’s attempt to become a developmental state as “dysfunctional” and predict its inevitable demise if the state is not fixed before it is too late.
At the core of this claim is what Peter Evans called ‘Embedded Autonomy’: a state machinery run by a sophisticated technocratic elite, and sufficiently autonomous institutions to avoid ‘capture’ by rent-seeking business elites. Anything short of this, it is argued, makes the state incapable of effectively intervening in the market and leaves it vulnerable to ‘capture’ and/or ‘patrimonialism.’
My argument is, however, comparing Ethiopia with the ideal East Asian states and demanding overly-ambitious and complex institutions runs the risk of making the pursuit of the best an enemy of the good. Yes, today’s Ethiopia is not South Korea or Taiwan, but is not Mobutu’s Zaire either. It is important to go beyond rhetoric and focus on what the Ethiopian developmental state can perform than what it can’t.
Fix the state!? How dysfunctional is it?
For starters, the mere claim that only a bureaucratically competent Weberian state brings about economic growth and transformation does not hold up. Desirable ends in and of themselves shouldn’t be confused with what is needed to accelerate economic growth and ‘catch-up’. Empirical evidences from South Asia, Latin America and some African counties indicate that a Weberian state is neither a necessity nor even desirable for some developing counties.
Brazil presents a useful example in this regard. The Brazilian state was known as a massive source of jobs populated on the basis of connection rather than competence. Extensive power of political appointment complemented to the lack of meritocratic recruitment. Extending Johnson’s comparison of Japan and the United States, Ben Schneider points out that while Japanese officials appoint only dozens of officials and U.S. officials appoint hundreds, Brazilian politicians appoint thousands (15,000 to 100,000 at a time) to various civil service offices. Based on this observation, Schneider prophesied that “the structure and operation of the Brazilian state should prevent it from fulfilling even minimal government functions.”
If Schneider was right, Brazilians today would have little cause to celebrate. Curiously, however, the Brazilian state has managed to play a major role in fostering both growth and industrialization. From its aggressive provisions of financing for railways and other infrastructure through its direct involvement in high-technology ventures like aircraft manufacture, the Brazilian state played a central role in what has overall been a miraculous industrialization.
Like their Brazilian counterparts, today’s politicians in Ethiopia appoint thousands of officials; institutional recruitment and promotion is based on loyalty and political affiliation than merit and performance, while excessive political intervention that does not allow autonomy characterizes the bureaucratic sphere. Simply put, professional cadres have dominated every institution in the country leaving its ambition to become a successful developmental state dotted by endless pitfalls.
In parallel, however, the economy has been growing at rates only seen in East Asian states. Massive infrastructural development, impressive social service delivery and Grand National projects characterize today’s Ethiopia. To the surprise of naysayers, a remarkably unprecedented socio-economic development has coincided with massive politicization of the bureaucracy.
How is this possible given the nature of the Brazilian state described above and its prototype Ethiopian state? In his ‘Embedded Autonomy’, Peter Evans persuasively argues that the ‘nonbureaucratic elements of bureaucracy’ are as important to state apparatuses as Durkheim’s ‘noncontractual elements of contract’ are to markets. This is because solidarity groups, informal networks or tight-knit party organizations enhance the coherence of the bureaucracy. Their presence provides critical reinforcement for the compliance to organizational norms and sanctions that Weber took for granted.
Truth be told, it is too early to conclude that tight-knit party organization and politicization of the bureaucracy resulted in impressive economic growth in Ethiopia. However, there are ample and visible evidences that demonstrate the functionality of the Ethiopian developmental state, at least ample enough not to be obsessively pessimist to report its ‘dysfunctionalism’, yet.
The tale of two thieves: Suharto and Mobutu
Closely related with the above is the assumption that the Ethiopian state is neo-patrimonial, and neo-patrimonialism of all sorts is bad and growth retarding. But it is an inconclusive argument. Evidences show that Southeast Asian states were no less patrimonial than African states. Suharto’s Indonesia, for example, was not necessarily less patrimonial than Mobutu’s Zaire. Ranking high on the world leaders’ corruption menu, both Suharto and Mobutu ripped off billions of dollars from the economy of their respective nations.
But while Mobutu’s Zaire failed miserably, Suharto’s Indonesia emerged to be one of the fastest growing economies in the world.
With untapped deposits of raw materials estimated to be worth the GDP of the US and Europe combined (USD 24 trillion), today’s Zaire, on the other hand, is an international object of pity and a classic example of mineral curse.
According to a recent research titled “Developmental Patrimonialism?” Questioning the Orthodoxy on Political Governance and Economic Press in Africa” by Kelsall Tim, Asante Kojo and Booth David, the secret behind the difference lies at the way rent-seeking and corruption were organized. In Indonesia, the potential damage done by patrimonialism was limited by the centralization of patronage, which ensured that clientelist pressures and corruption did not spin out of control. General Suharto understood that regime survival and his own personal fortune depended heavily upon long-term economic growth. On the other hand, Mobutu Sese Seko was just “the chairman of the board of impatient profiteers”, to use Frantz Fanon’s words.
Hence, Kelsall and et al argue that Indonesia appears to confirm the idea that “[i]t is not neo-patrimonialism per se that is bad for investment and growth but rather the specific form that neo-patrimonialism often takes, and whether this occurs in an organized or unorganized way.” For that matter, even our ‘African Star’, Botswana, has been characterized by ‘illiberal authoritarianism’ with Fredrik Söderbaum going so far as to describe patrimonialism in Botswana as ‘pathological’. Therefore, the argument that a patrimonial political structure need not be an obstacle to capitalist economic development in the early stages deserves a fair hearing. In certain conditions a strong, centralized, patrimonial structure can be a better way of achieving an acceptable, growth-promoting balance than conventional Good Governance arrangements.
In this regard, Sarah Vaughan and Mesfin Gebremichael in their “Rethinking business and politics in Ethiopia” set out to explain the importance of stable, long-horizon, and centralized and selective growth-enhancing rents in achieving successful ‘developmental patrimonialism’ in Ethiopia . For Sarah and Mesfin, Ethiopia broadly fits within the typology of long-horizon, centralized rent process that can generate dynamic growth.
Another critical research by Oxford professors titled “Africa’s Illiberal State-Builders” categorize Ethiopia as one of Africa’s emerging illiberal state-builders whose aspirations go beyond a short-term resource grab, and use the state to centralize resources and create and/or strengthen a vigorous system of control. Though very critical about what it calls “the gamble of definitely subduing old ethnic and centre-periphery demons through rapid growth and efficient service delivery”, the research paper admits that this emerging model “may indeed materially change Ethiopia beyond recognition”
In addition to the above alternative assumption, the recent global financial crisis has eroded the hitherto strong faith on the so called independent institutions. Many specialists now admit that key Western institutions, once typical examples of good institutions, have been taken hostage by special interests.
The lesson, according to Joseph Stiglitz et al, is that we should be less confident about what we mean by good policies and institutions; and that we should be even more modest in our belief that exact replicas of institutions and policies that may have worked in one context would be as successful in another.
The task ahead
The task of building a developmental state is a trial and error process whose success depends on whether policy makers are capable and committed to correct pervasive errors here and there. Building new institutions and restructuring administrative apparatus is a massive political process. What’s more, while gleaning lessons, Ethiopian policy makers should be vigilant not to replicate mistakes done by others. If Suharto’s Indonesia shows the transformative power of a well managed developmental patrimonialism, Mobutu’s Zaire shows the dangers of clientelism.
As it stands now, “Developmental-patrimonialism” is the modus operandi of state political economy in Ethiopia and is here to stay. Let’s hope that creating more Suhartos and fewer Mobutus is the next task on the to-do-list of politicians in Ethiopia.
The writer is a lecturer at the Department of Governance and Development Studies, Jimma University. He can be reached at firstname.lastname@example.org