Why Every Regime Repeated the Pattern: A Political Economy

Thesis

The Ethiopian constitutional pathology this series has documented is not a story of three different regimes making three different mistakes. It is the story of three regimes — imperial Ethiopia from 1931 to 1974, the Derg from 1974 to 1991, the EPRDF/Prosperity Party order from 1991 to the present — making the same mistake under different ideological vocabularies. The question this article must answer is why. If the cure is known (Article 23), if the institutional architecture for constitutional functioning is achievable, if other comparable countries have constructed it — why has each Ethiopian regime, with substantively different ideological commitments, ended up producing constitutional forms without constitutional substance?

The article’s argument is that the political-economy conditions of the Ethiopian state have, across nine decades, consistently produced the constitutional pathology. The conditions include: a state that has, despite considerable size, been institutionally weak at the level of independent civil and judicial institutions; an agrarian economy in which the principal social classes (peasants, landlords until 1975, urban professionals and merchants of varying scale) have lacked sufficient organisational capacity to demand constitutional constraint; a regional security environment that has provided successive regimes with external threats to justify executive consolidation; an international donor and intelligence environment that has rewarded developmental and security outcomes over institutional ones; and a recurrent pattern in which the constituency that would, in other countries, demand constitutional rule of law has either not existed (1931–1974) or has been suppressed (1975–present) before it could anchor a new institutional equilibrium. Each regime’s individual choices matter — Haile Selassie was not Mengistu, Mengistu was not Meles, Meles was not Abiy — but each operated within the same political-economy configuration, and each produced the same broad constitutional outcome.

The article uses the Acemoglu-Robinson framework explicitly as its analytic spine. The framework asks: what are the political-economy conditions under which institutions become inclusive rather than extractive? Ethiopia’s answer, across three regimes, has been that the conditions have not assembled. The series ends in the next article with the question of whether they can.

The Acemoglu-Robinson framework

The Acemoglu-Robinson framework, set out in Why Nations Fail (2012) and The Narrow Corridor (2019), holds that institutions are inclusive when they distribute power broadly and constrain incumbents, and extractive when they concentrate power and entrench incumbents. The transition from extractive to inclusive institutions, on this framework, requires (a) a broad social coalition with the political organisational capacity to demand inclusion, (b) elite incentives to accept inclusive institutions in exchange for political stability or other gains, (c) institutional infrastructure capable of operationalising the inclusion, and (d) sustained political will to maintain the institutions against pressure to revert to extraction.¹

The framework’s application to specific cases is empirical. The transitions from extractive to inclusive institutions that the framework analyses — England in 1688, the United States in 1776 and after, Germany in 1949, South Korea in 1987, Indonesia after 1998 — all involved configurations of (a) through (d) that produced sustained institutional transformation. The failures to transition — most of post-colonial Africa, much of Latin America before 1980, the former Soviet Union after 1991 — involved configurations in which one or more of the conditions were absent.

The framework is not deterministic. Specific political choices matter; specific leaders matter; specific events can produce institutional breakthroughs that the political-economy conditions would not have predicted. But the political-economy conditions are the structural background that makes choices, leaders, and events more or less likely to produce institutional transformation.

The Ethiopian application of the framework is straightforward in its broad outlines. The conditions (a) through (d) have not been present in any of the three regimes. Each regime has had specific political choices that could have moved in inclusive directions; each has had specific leaders who articulated inclusive visions in various forms; each has had events that could have produced institutional breakthrough. None has been sufficient. The framework predicts that the configuration that produced the pathology must change before the pathology will be addressed.

Condition (a): broad social coalition

In each of the three Ethiopian regimes, the social coalition with the political organisational capacity to demand inclusive institutions has been absent or insufficient.

The imperial period (1931–1974). Ethiopian society in 1931 was overwhelmingly rural (over 90 percent), largely illiterate, organised through local hierarchies of clergy, landlords, and customary authority, and operating in an economy that did not produce significant urban concentrations of professional or commercial classes. The university — Haile Selassie I University, founded in 1961 — produced a small graduate class by the late 1960s and 1970s, and the Ethiopian student movement was the principal constituency demanding institutional reform. The movement was significant but too small relative to the population to anchor a sustained institutional transformation. The 1974 revolution was, in significant part, produced by the student movement and the urban professional classes the imperial period had begun to produce, but the revolution did not have the social weight to determine its outcome, which was captured by the military.²

The Derg period (1974–1991). The Derg’s policies (the 1975 land reform, the urbanisation pressures of the 1980s, the universal literacy campaign of the late 1970s) did produce some expansion of the urban professional class. But the Derg’s political project also suppressed the political organisation of that class: the EPRP destruction in 1976–78, the Meison destruction in 1977–78, the systematic harassment of independent civil society throughout the period. By 1991, Ethiopia had a marginally larger urban professional class than in 1974 but a marginally weaker organisational capacity within that class to demand institutional inclusion.³

The EPRDF/Prosperity period (1991–present). The post-1991 period has produced the largest sustained expansion of the Ethiopian urban professional and commercial class in modern history. The growth of Addis Ababa from approximately 2 million to over 5 million people, the expansion of higher education from approximately 18,000 students in 1991 to over a million by 2020, the growth of commercial sectors (construction, manufacturing, services) — all of these have produced a substantially larger urban middle class than existed in 1991.⁴ The class has political consciousness; it produced the 2005 CUD electoral surge, the 2014–2018 protest mobilisations, and the post-2018 political activism.

But the class has not, by 2026, achieved the organisational capacity required for sustained constitutional demand. The CUD was suppressed in 2005. The 2018 mobilisations produced a leadership change but not institutional reconstruction. The civil society sector remains restricted under the post-2024 amendments to the Civil Society Organisations law. The independent media operates under continuing pressure. The political opposition has been fragmented and intermittently suppressed.

The reasons the class has not consolidated organisationally are several: the constitutional architecture (Article 39 in particular) has incentivised ethnic political organisation over class-based organisation; the urban-rural divide has limited the class’s reach into the rural majority; the political opportunity structure has rewarded individual exit (emigration, accommodation with the regime) over collective demand. The class exists. The political coalition has not assembled.

Condition (b): elite incentives

The elite incentives for accepting inclusive institutions have, in each Ethiopian regime, been weak.

The imperial elite had no reason to accept institutional inclusion that would constrain imperial authority. The imperial period’s elite included the emperor, the imperial family, the great nobles, the church hierarchy, and the senior military and civil-service officials. None of these had structural reason to support constitutional inclusion; each had specific reasons to oppose it.

The Derg elite consisted of the military officers who had taken power in 1974, the Marxist-Leninist civilian intellectuals who served the regime, and the security and military commanders who maintained it. None of these had structural reason to support inclusion that would have limited their authority; the Marxist-Leninist framework provided ideological justification for executive consolidation that aligned with their personal interests.

The EPRDF elite consisted of the TPLF, ANDM, OPDO, and SEPDM senior cadres; the federal bureaucracy that emerged from the post-1991 period; the EFFORT (Endowment Fund for the Rehabilitation of Tigray) and parallel regional economic networks; and the senior military and security officials. The framework Meles articulated — revolutionary democracy — provided ideological justification for executive consolidation. The economic interests of the EPRDF elite — concentrated in the EFFORT, in the federal contracting networks, in the construction and import sectors that benefited from regime favour — were not aligned with constitutional constraint.⁵

The Prosperity Party elite consists of Abiy Ahmed and his immediate circle, the senior figures of the merged ruling party, the federal bureaucracy, and the regional affiliates. The economic interests of this elite have continued the EPRDF-era patterns in modified form, with the addition of the Tigray-related dispossessions and the post-2020 redistribution of economic position from TPLF-aligned to Prosperity-aligned networks. The elite incentives for constitutional inclusion have not, in operation, been stronger than under previous regimes.⁶

In each case, the political coalition demanding inclusion would have had to offer the existing elite something the elite valued more than the position it would lose. The CUD in 2005 offered electoral participation that would have, in principle, included EPRDF figures in the new political settlement, but the offer was not credible at the level the EPRDF leadership would have required. The 2018 transition offered, briefly, the possibility of constitutional renegotiation, but the new leadership (Abiy and the Prosperity Party coalition) had different priorities and the TPLF was unwilling to participate. The current configuration provides no obvious path to elite alignment with inclusion.

Condition (c): institutional infrastructure

Each regime has inherited and added to the institutional infrastructure of executive authority while failing to construct the institutional infrastructure of constraint.

The imperial state had the centralised civil service that Haile Selassie had developed through the 1930s and 1940s, the army that had been built with Western and Soviet assistance, the church-state apparatus that legitimated imperial authority, and the rudimentary regional administrative structure. It did not have a functioning constitutional court, an independent press, an organised opposition, or a civil society sector capable of independent organisation.

The Derg added to the imperial state the more extensive security apparatus (the kebele and woreda mass-organisation structures), the cadre-based party (eventually the WPE), the nationalised economy’s administrative structures, and the Soviet-style ministerial coordination. It did not add functioning constitutional review, judicial independence, or civic infrastructure.

The EPRDF added the federal administrative structure of the regions and zones, the expanded judiciary, the National Electoral Board, the House of Federation, and the Council of Constitutional Inquiry. The institutions exist; they have not been constructed to operate independently of executive control.

The Prosperity Party era has inherited the EPRDF infrastructure and has added the centralised Prosperity Party organisational structure across the federation, the strengthened federal coordination of security forces, and various centralised economic management institutions. The institutional infrastructure has not been retrofitted to include constitutional constraint.

The infrastructure deficit is not absolute. Some institutions function: the National Electoral Board under Birtukan Mideksa’s leadership (2018–present) has produced more credible electoral conduct than its predecessors. The Ethiopian Human Rights Commission has, since the 2018 reforms, operated with greater independence than before. The federal judiciary handles ordinary civil and criminal matters with reasonable professionalism. But the institutions of constitutional constraint — those that would actually constrain the federal executive in politically consequential cases — have not been constructed at the scale required.

Condition (d): sustained political will

The most decisive of the four conditions is sustained political will — the maintenance of institutional construction against pressure to revert to extraction.

Each Ethiopian regime has, at moments, articulated commitments that, if sustained, would have produced institutional transformation. Haile Selassie’s 1955 Constitution articulated the form of constitutional monarchy. The Derg’s 1987 Constitution articulated the form of socialist legality. The EPRDF’s 1995 Constitution articulated the form of multi-ethnic federalism with rights. The Prosperity Party’s National Dialogue Commission articulated, in principle, the framework for constitutional renegotiation. In each case, the articulated commitment has not been sustained at the level required for institutional transformation.

The reasons for the failure of sustained political will are several. The international environment has rewarded developmental outcomes over institutional ones. The regional security environment has produced repeated crises that the executive has used to justify postponing constitutional construction. The internal political dynamics of each regime have produced incentives for executive consolidation against opposition rather than institutional construction with opposition. The personal characteristics of leaders have varied but, in each case, have ultimately favoured the path of least institutional change.

The pattern is that political will for institutional transformation has been intermittent rather than sustained. Each regime has had moments of openness — the 1955 constitution, the 1987 constitution, the 1995 constitution, the 2018 transition — that, if pursued through to institutional construction, could have produced constitutional functioning. None has been pursued through to that point.

The international and donor environment

The political-economy framework must include the international and donor environment as one of the factors shaping Ethiopian elite incentives.

Ethiopia has, since 1991, been one of the largest recipients of international development assistance in Africa, with multilateral institutions (IMF, World Bank, African Development Bank), bilateral donors (United States, European Union member states, United Kingdom, China, Gulf states), and humanitarian agencies all maintaining significant programmes.⁷ The development assistance has produced real outcomes: infrastructure, education, health system expansion, food security improvements. It has also produced political effects: the regime that successfully partnered with donors received resources that consolidated its position, while institutional reform was rewarded inconsistently or in symbolic ways that did not require operational transformation.

The IMF and World Bank’s relationship with Ethiopia has, since 1991, focused on macroeconomic stabilisation, sectoral policy reform, and developmental investment. The institutions have, at various moments, raised concerns about governance and institutional capacity, but the operational relationship has been principally with the executive on developmental matters. The 2025/2026 IMF Article IV consultation, which reported macroeconomic stabilisation including inflation reduction to 11.9 percent, exemplifies the pattern: serious developmental engagement with the executive, limited engagement with constitutional and institutional matters.⁸

The United States, the European Union, and the United Kingdom have, at various moments, expressed concern about specific institutional matters (the 2005 crackdown, the 2009 Anti-Terrorism Proclamation, the Tigray war’s conduct, the post-2023 civil society restrictions) without producing sustained pressure for institutional transformation. The principle has been engagement over isolation, with the assumption that engagement produces gradual institutional change. The empirical record does not support the assumption strongly.

The Trump administration’s 2025 cuts to foreign aid produced significant effects on Ethiopian humanitarian and development funding but have not, as of mid-2026, produced sustained political pressure for institutional reform.⁹ The bilateral relationship has shifted in character without shifting the underlying incentive structure that the donor environment had created.

China’s engagement with Ethiopia has been larger in scale than Western engagement for parts of the post-2000 period and has been explicitly indifferent to internal institutional matters. The Chinese model has provided the Ethiopian executive with an alternative source of developmental financing that does not condition on institutional reform. The Gulf states (UAE, Saudi Arabia) have provided similar non-conditional financing in the post-2018 period. The result is an international environment in which the executive has had multiple sources of resources available without institutional conditions.¹⁰

The regional security environment

The regional security environment has consistently provided successive Ethiopian regimes with external threats that justified executive consolidation against opposition.

The imperial period faced the Italian invasion of 1935–41, the post-war Somali irredentist threat, and the early Eritrean insurgency. The Derg period faced the Somali invasion of 1977–78, the protracted wars with the EPLF and TPLF, and various lower-level regional threats. The EPRDF/Prosperity period has faced the 1998–2000 Eritrean war, the Somali state collapse and al-Shabaab insurgency from 2006, the various boundary disputes with Sudan, the post-2011 Egyptian tensions over the GERD, and the post-2020 deterioration with Eritrea.

In each period, the security environment has provided justification for the centralisation of authority. The justification has been partly accurate — real security threats have existed and required state response — and partly instrumental — the threats have been invoked to justify centralisation that exceeded what the security requirements alone would have produced.

The pattern is that the security environment has provided permanent occasion for postponing institutional construction. There is always a crisis; the crisis always requires executive authority; the institutional construction is always something to be undertaken after the current crisis is resolved. The current crisis is, in mid-2026, the simultaneous conflicts in Amhara, Tigray, and Oromia along with the deteriorating Eritrean relationship. The same pattern that has prevented institutional construction across three regimes is operative in 2026.

The recurring pattern

The political-economy framework predicts that, absent change in the conditions (a)–(d), the Ethiopian constitutional pattern will continue. Each new regime will inherit the institutional infrastructure of executive authority. Each will articulate constitutional commitments. Each will face circumstances — security crises, economic pressures, international engagements — that justify postponing institutional construction. Each will, over time, consolidate executive authority within the inherited framework. Each will produce form without substance.

The 2018 transition tested this prediction. The opening provided unusually favourable conditions for institutional construction: a leadership change driven by mass mobilisation with constitutional content, an internationally celebrated leader with political capital, a regional environment that was briefly favourable (the Eritrean rapprochement), donor support that would have rewarded institutional reform. The prediction was that, absent sustained political will against the structural pressures, the opening would close. The opening closed. The prediction was confirmed.

The 2026 supermajority provides an opportunity to test the prediction again, this time in different conditions. The federal government has reduced internal political friction; the post-Pretoria framework provides nominal stability in Tigray; the National Dialogue Commission process provides institutional cover for constitutional discussion. Whether the supermajority is used for institutional construction or for further executive consolidation is the question that the next eighteen months will answer.

Steelman of the opposing readings

Two opposing readings deserve engagement.

The first is that the political-economy framework is too deterministic, that individual choices matter more than the framework allows, and that Ethiopian leaders could have made different choices that would have produced different outcomes. This has weight. The framework is not deterministic; it predicts probabilities, not certainties. Specific leaders making specific choices have produced specific outcomes that, in cumulative effect, have followed the pattern. The reply is that the framework explains the cumulative effect even if it does not predict each individual choice.

The second is that the political-economy framework underestimates the international and security environment’s responsibility for the Ethiopian outcomes, attributing too much to domestic choices and too little to external constraints. This also has weight. The international environment has rewarded developmental performance over institutional construction. The security environment has provided permanent occasion for executive consolidation. The donor environment has been complicit in extractive institutional outcomes. The reply is that other countries with comparable external pressures (South Africa, Indonesia, Nigeria) produced different institutional outcomes. The external pressures shape but do not determine the domestic choices.

What the article establishes

The Ethiopian constitutional pathology is the product of political-economy conditions that have been consistent across three regimes. The conditions are the absence of a broad social coalition with sufficient organisational capacity to demand inclusion, the absence of elite incentives to accept inclusion, the absence of institutional infrastructure for constitutional constraint, and the absence of sustained political will for institutional construction. The conditions are not unchangeable; they are constructed, and they can be reconstructed. The next article — the closing article of the series — takes up the question of how, and what the closing argument of the series is.

---

References

Footnotes

1. Daron Acemoglu and James A. Robinson, Why Nations Fail (New York: Crown, 2012); The Narrow Corridor (New York: Penguin, 2019). ↩

2. On the imperial-era social structure and the student movement, see Bahru Zewde, The Quest for Socialist Utopia: The Ethiopian Student Movement, c. 1960–1974 (Oxford: James Currey, 2014); John Markakis, Ethiopia: Anatomy of a Traditional Polity (Oxford: Clarendon Press, 1974). ↩

3. On the Derg’s effects on civil society, see Christopher Clapham, Transformation and Continuity in Revolutionary Ethiopia; Babile Tola, To Kill a Generation. ↩

4. On the post-1991 expansion of the Ethiopian middle class, see Christopher Adam and Stefan Dercon, “The Political Economy of Development in Ethiopia,” in The Oxford Handbook of the Ethiopian Economy (2019). ↩

5. On EFFORT and the EPRDF economic networks, see Sarah Vaughan and Mesfin Gebremichael, Rethinking Business and Politics in Ethiopia (London: Africa Power and Politics Programme, 2011). ↩

6. On the Prosperity Party economic networks, see “Abiy Ahmed’s Hegemonistic Prosperity Party,” Borkena, June 2026. ↩

7. On Ethiopia as donor recipient, see World Bank Ethiopia Country Partnership Strategy documents 2008–2026; OECD DAC reports on Ethiopia. ↩

8. IMF Article IV consultations 2023–2025; “Ethiopia’s 2025 Ascent,” ENA English, January 2026. ↩

9. On Trump administration 2025 aid cuts, see Human Rights Watch World Report 2026: Ethiopia. ↩

10. On Chinese and Gulf state engagement, see various 2018–2026 reports; UAE-Ethiopia strategic partnership documentation. ↩