The Asian Developmental Model: The Common Formula

Thesis

The Asian developmental transformations of the twentieth century — Japan from 1868, Japan again from 1945, South Korea from 1961, Taiwan from 1949, Singapore from 1965, China from 1978, with lesser variants in Malaysia and Indonesia — share a common formula. The formula is not a set of specific policies that transfer identically from one country to another; each case had its own particular institutional configuration, its own leaders’ idiosyncrasies, and its own external conditions. The formula is a set of five structural moves that every successful case made in some variant: (1) land reform that broke the political-economic power of the traditional landowning class and produced a mass smallholder base with tenure security; (2) universal primary education under a standardised curriculum in a national language that constructed literate, technically capable, and identity-integrated workforces; (3) meritocratic bureaucracy insulated from patronage that could coordinate the state’s developmental interventions; (4) export-oriented industrialisation supported by state-directed credit allocation, protected domestic markets in early stages, and disciplined firm-level performance; and (5) active state-led construction of a pan-national identity that permitted the workforce and citizenry to operate as one political community.

The article develops each of these five moves in turn, examines the cases in which the move was well-executed, and identifies the Ethiopian analog in each case. The Ethiopian analog exists in each of the five: the Derg’s 1975 land reform was a variant of move (1); the EPRDF’s mass education expansion was a variant of move (2); the various civil-service reform efforts under Meles and Abiy have been variants of move (3); the export-oriented industrialisation strategy of the 2000s and 2010s was a variant of move (4); and the Medemer framework since 2018 has been a variant of move (5). What the Ethiopian variants have shared with the Asian successes at the structural level is real. What they have not shared is the execution quality, the sustained political will, the institutional depth, and — most importantly — the integration of the five moves as parts of one coherent developmental project. The Asian cases treated the five moves as complementary components of a single transformation. Ethiopia has, in operational terms, treated them as separable initiatives that could be pursued or reversed in isolation.

Move 1: Land reform

Land reform is the first move in every successful Asian transformation. The reason is not primarily economic — it is political. A society dominated by a traditional landowning class contains a coalition with structural interest in preserving pre-industrial arrangements: the landowners’ wealth is tied to land, land is tied to peasant labour under conditions the landowners set, and the political system that upholds these arrangements is the system the landowners control. Break the landowning class, and the coalition dissolves; the peasants become smallholders with interest in state investment in agricultural productivity and rural infrastructure; the political system becomes available for redirection toward developmental purposes.

Japan. The Meiji-era reforms did not conduct a redistributive land reform in the twentieth-century sense; they converted the feudal-domain system into a modern land-tax system that transferred wealth from the samurai class to a rising merchant and small-landowner class. The definitive land reform in Japan came under American occupation between 1946 and 1950: the Land Reform Law of 1946 required absentee landlords to sell their land to the state, which then resold it to tenant farmers at fixed prices below market rates. By 1950, tenancy had declined from approximately 46 percent of cultivated land to under 10 percent, and Japan had a mass smallholder rural base that would provide the political stability underwriting the post-war developmental transformation.¹

South Korea. Korean land reform proceeded in two stages. The United States Army Military Government redistributed 240,000 hectares of previously Japanese-owned land in 1948, accounting for approximately 12 percent of cultivated land. The Farmland Reform Act of 1950, enacted by the Syngman Rhee government under American pressure just before the outbreak of the Korean War, redistributed most remaining tenanted land at prices tenant farmers could afford. By the end of the redistribution, land tenancy in South Korea had been effectively eliminated as a large-scale phenomenon. The land Gini coefficient dropped from approximately 0.73 in 1945 to under 0.4 by 1955.² The reform destroyed the political power of the traditional Korean landlord class and produced the mass smallholder rural base that underwrote Park Chung-hee’s later industrialisation programme.

Taiwan. The Kuomintang, having fled to Taiwan in 1949, executed a land reform between 1949 and 1953 that was in some ways more thorough than the Korean or Japanese reforms because the KMT was executing it against a landowning class (the Taiwanese) different from itself (the mainland Chinese émigrés). The reform’s stages — rent reduction (1949), sale of public land (1951), and the “Land to the Tiller” programme (1953) — produced tenancy rates comparable to Japan’s and Korea’s within four years. The KMT’s political interest in weakening the pre-existing Taiwanese elite aligned with the developmental logic of the reform.³

China. The Chinese Communist Party’s land reform between 1946 and 1953 was the most violent and most thorough of the Asian cases. It destroyed the landowning class as a social category, redistributed land to peasant households, and produced the political base for subsequent Party mobilisation. The subsequent collectivisation and communisation (1955–1961) reversed the redistribution in operational terms, but the political effect — the destruction of the pre-industrial landowning coalition — was permanent. When Deng’s 1978 reforms brought back household farming through the Household Responsibility System, the peasants had land-use rights rather than tenancy under landlords.⁴

The Ethiopian variant. The Derg’s Proclamation No. 31 of 1975 nationalised all rural land and distributed use-rights to peasant associations. The reform was, in its immediate effect, one of the most dramatic social transformations in twentieth-century African history. It destroyed the traditional Ethiopian landlord class, transferred effective control of rural land to peasant users, and produced the political base for the Derg’s rural mobilisation. Structurally, this was a version of the Asian move (1).

What made the Ethiopian variant fail to produce the developmental payoff was three specific features. First, the reform did not produce individual smallholder tenure; it produced state ownership with peasant use-rights, subject to state administrative decision. This is examined in “The Hollow Pillar” Article 11 and will not be re-litigated here. Second, the reform was not integrated with the other four moves of the Asian formula; there was no coordinated push into universal education, meritocratic bureaucracy, export-oriented industrialisation, or identity construction to convert the rural transformation into a broader developmental transformation. Third, the reform was subsequently entrenched by the 1995 Constitution’s Article 40, which prohibits private land ownership and thereby prevented the emergence of the propertied middle class that would have produced political demand for the further developmental moves. The Ethiopian land reform’s political effect was retained; its developmental effect was foreclosed by what followed.

Move 2: Universal education

The second move is universal primary education under a standardised curriculum in a national language. The economic function is to produce a workforce with the basic literacy, numeracy, and technical capacity that industrial economies require. The political function is to produce a citizenry that shares a common body of knowledge, a common historical narrative, and a common capacity to move between regions and roles.

Japan. Universal primary education was among the earliest Meiji reforms. The Fundamental Code of Education (Gakusei) of 1872 established the framework for a national school system. By 1900, primary school enrollment for the school-age population had reached over 90 percent; by 1910 it was effectively universal. The curriculum standardised in Japanese language, mathematics, sciences, moral education (built around the emperor and the state), and physical training. The Imperial Rescript on Education of 1890 established the moral-ideological content that would be recited daily in classrooms for the next fifty-five years.⁵

South Korea. Universal primary education was achieved earlier in Korea than the industrialisation itself. By 1960, before Park’s coup, primary school enrollment was over 90 percent; the literacy rate had risen from approximately 22 percent at Japanese-colonial liberation in 1945 to over 70 percent by 1960. This was accomplished under conditions of extreme poverty and only a few years after the devastation of the Korean War. Park’s subsequent development strategy could presuppose a literate workforce; the workforce was there when the industrial policy came.⁶

Taiwan. Comparable to South Korea. By 1960, primary school enrollment was near-universal; secondary education was expanding rapidly. The Taiwanese education system emphasised Mandarin Chinese as the national language, standardised curriculum, and shared historical narrative that constructed a Chinese national identity within the KMT framework — a construction contested by Taiwanese nationalist opinion but operationally effective in producing a workforce that could operate on shared cultural terms.⁷

Singapore. Lee Kuan Yew’s education policy from 1965 was bilingual — English as the working language of the state and economy, mother tongue (Mandarin, Malay, Tamil) as the identity-language of the family and community. The bilingual policy produced a workforce fluent in English (permitting integration into global markets) and a citizenry that retained connection to distinct ethnic cultural traditions (permitting the ethnic-integration policy to function without cultural erasure). Universal primary education was achieved by the early 1970s; secondary education expanded rapidly through the 1970s and 1980s.⁸

China. Universal primary education was among the CCP’s earliest post-1949 achievements, though the quality was uneven and the Cultural Revolution (1966–1976) produced substantial disruption. By 1985, primary school enrollment was over 90 percent nationally. The standardised curriculum, delivered in Mandarin as the national language, produced the workforce capacity that Deng’s reforms could subsequently mobilise.

The Ethiopian variant. The Derg’s National Democratic Revolution literacy campaign of 1979–1984 was the first serious attempt at mass literacy in Ethiopian history. The campaign taught basic Amharic literacy to an estimated 15 million adult Ethiopians and reduced adult illiteracy from over 90 percent to approximately 63 percent by 1984. This was a real achievement.⁹ The EPRDF’s post-1991 expansion of primary education was more dramatic: primary school enrollment rose from approximately 3 million children in 1991 to over 20 million by 2015; higher education enrollment rose from approximately 18,000 students to over one million; new universities were established in every region.

What the Ethiopian education variant lacked, and what distinguishes it from the Asian successes, is the standardised national curriculum in a shared language. The EPRDF’s regional-language policy — appropriate in its recognition of Ethiopian linguistic diversity — produced regional variants of primary education that did not integrate around a shared pan-Ethiopian curriculum. Children educated in Afaan Oromo received one version of Ethiopian history and one set of cultural references; children educated in Tigrinya received another; children in Amharic-speaking regions received a third. The regional variants did not produce cross-regional workforce mobility, did not construct a shared national historical narrative, and did not build the pan-national identity that a proper standardised education would have built. The move was made in structural form; the identity-construction dimension of the move was not delivered.

Move 3: Meritocratic bureaucracy

The third move is the construction of a meritocratic civil service capable of coordinating the state’s developmental interventions without the pathologies of patronage-based recruitment. Every successful Asian developmental state has had a bureaucracy of this kind: Japan’s MITI, Korea’s Economic Planning Board, Taiwan’s Council for Economic Planning and Development, Singapore’s civil service and Economic Development Board, China’s post-1978 state economic apparatus.

Japan. The Meiji-era civil service was constructed from a mix of former samurai retained for their educational qualifications and new recruits selected through competitive examination. By the 1890s, the higher civil service was recruited primarily through examinations administered through the Imperial University system. Post-war Japan retained the pre-war civil service structure with modifications — MacArthur’s American Occupation did not dismantle the bureaucracy — and it was this civil service that coordinated Japan’s post-war economic recovery. The Ministry of International Trade and Industry (MITI) became, from 1949, the coordinating institution for industrial policy.¹⁰

South Korea. The competitive civil-service examination system was implemented from 1949 but was undermined during the Syngman Rhee period by patronage appointments. Park Chung-hee’s post-1961 government re-established the examination system as the primary path into the civil service, insulated the Economic Planning Board from political interference (Park chaired the Board personally and gave its head, the Deputy Prime Minister, cabinet seniority), and produced the meritocratic technocracy that could execute the five-year plans. The ratio of public-official corruption to general-population indictment declined from 0.2 percent in the 1950s to 0.04 percent by the 1980s, an empirical indicator of meritocratic-bureaucracy consolidation.¹¹

Singapore. Lee Kuan Yew’s civil-service reform is the most explicit of the Asian cases. The Singaporean civil service adopted the British colonial administration’s structural framework, retained its examination-based recruitment, established high salaries (linked to private-sector benchmarks), and enforced strict anti-corruption measures through the Corrupt Practices Investigation Bureau established in 1952 and strengthened progressively through the 1960s and 1970s. The Singaporean civil service by the 1980s was widely regarded as among the world’s most professional and least corrupt.

The Ethiopian variant. Ethiopian civil-service reform has been the subject of multiple efforts under multiple regimes. Haile Selassie’s post-1955 modernisation established the framework of a professional civil service, and the Institute of Ethiopian Studies (later part of Addis Ababa University) produced graduates who staffed it. The Derg’s civil service inherited the imperial framework and largely destroyed its professional-independence character through cadre-based promotion and political-loyalty testing. The EPRDF’s Civil Service Reform Programme from 1996 (with subsequent iterations) attempted to reconstruct professional standards, and the Meles-era Public Service Reform introduced performance-based management and pay reforms.¹² The Abiy government from 2018 has continued various civil-service reform initiatives, including anti-corruption prosecutions and structural reorganisations.

What Ethiopia has not produced, across these successive reform efforts, is a civil service that operates with the structural insulation from political interference that the Asian comparators produced. The 2018 Prosperity Party reorganisation, in operational terms, made the civil service more responsive to political direction rather than less; the parallel Party-state structures that operated under the EPRDF have not been dismantled under the Prosperity Party; the Party continues to exercise significant informal authority over civil-service personnel decisions. The move has been attempted; the institutional consolidation has not been achieved.

Move 4: Export-oriented industrialisation

The fourth move is export-oriented industrialisation supported by state-directed credit, sector-selective protection, and firm-level performance discipline. The theoretical framework is what Alice Amsden called “getting the prices wrong on purpose” — deliberately manipulating market signals to direct investment toward sectors the state has identified as strategically important, while enforcing performance standards that prevent the protected sectors from becoming permanently uncompetitive.¹³

Japan. MITI’s coordination of Japanese industrial policy from 1949 through the 1970s is the model case. MITI directed credit through the Japan Development Bank toward sectors identified by the ministry as strategic (steel, shipbuilding, automobiles, electronics), protected these sectors from foreign competition through tariff and non-tariff barriers, but simultaneously required them to export to global markets so that the protection did not become a permanent shelter for uncompetitive firms. The result was Japanese industrial output that could compete with, and eventually surpass, established Western producers in these sectors.¹⁴

South Korea. Park Chung-hee’s five-year plans directed state credit through state-owned banks toward chaebol conglomerates (Hyundai, Samsung, LG, and others) that were required to hit export targets or lose their credit access. The chaebol were not free-market firms; they were state-directed instruments whose growth was underwritten by state credit conditional on export performance. The heavy-and-chemical-industry drive of 1973–1979 pushed South Korean industry from light manufacturing (textiles) into steel, shipbuilding, chemicals, and automobiles. By 1980, South Korean industry could compete with Japanese industry in several sectors.¹⁵

Taiwan. Taiwan’s variant emphasised small and medium-sized enterprises rather than chaebol-style conglomerates. The Council for Economic Planning and Development directed credit and technical assistance toward export sectors; the export processing zones established in the 1960s and 1970s attracted foreign direct investment; the semiconductor industry, developed from the 1980s through TSMC and related firms, became by the 2000s a globally strategic sector. The Taiwanese variant is instructive because it shows that the export-oriented industrialisation move can be executed through different organisational forms — chaebol in Korea, SMEs in Taiwan, government-linked corporations in Singapore.

Singapore. The Economic Development Board (EDB), established in 1961, attracted foreign direct investment through explicit courting of multinational corporations, tax incentives, industrial-estate development (Jurong), and workforce provision through the education system. The Singaporean variant emphasised foreign investment rather than domestic conglomerates, but the structural logic was the same: state coordination of industrial development around export markets.

China. Deng’s post-1978 reforms combined the special economic zones (Shenzhen, Zhuhai, Shantou, Xiamen, and subsequently expanded) with state-directed credit through state-owned banks, township-and-village enterprises, and gradual liberalisation of foreign direct investment. The Chinese variant produced the world’s largest export-manufacturing sector by the 2010s.¹⁶

The Ethiopian variant. The EPRDF’s Growth and Transformation Plans (GTP-I 2010–2015, GTP-II 2015–2020) explicitly modelled the Asian developmental strategy. The plans directed state credit through the state-owned Commercial Bank of Ethiopia and the Development Bank of Ethiopia toward priority sectors (textile and garment, leather, agro-processing, cement, floriculture), protected domestic markets, and attempted to build export capacity. Industrial parks were established at Hawassa, Bole Lemi, Kombolcha, Mekelle, and elsewhere. Export targets were set and, in some sectors, partially achieved (floriculture, textile-garment, coffee).¹⁷

What the Ethiopian variant did not achieve, at the scale of the Asian comparators, was the durable industrial transformation that would have shifted the country’s economic structure from agrarian to industrial. The industrial-park model produced some export capacity but was disrupted by the Tigray war (particularly Mekelle), the Amhara conflict (Kombolcha), and macroeconomic instability (chronic foreign-exchange shortage, hyperinflation reaching 30 percent in 2022 before declining). The chaebol-equivalent — durable domestic industrial conglomerates capable of competing internationally — did not emerge; the EFFORT-related enterprises that were closest to the model operated as ethnic-regional-political networks rather than as export-competitive firms. The move was made; the outcomes have been partial.

Move 5: Identity construction

The fifth move is the active state-led construction of a pan-national identity that permits the workforce and citizenry to operate as one political community. This is the move that “The Hollow Pillar” argued Ethiopia has not made, and that this series will develop in detail across Parts II through IV. Here it is stated at the level of the general formula.

Japan. State Shinto, emperor cult, Imperial Rescript on Education, universal military conscription, standardised curriculum in Japanese language, national ceremonial calendar, standardised national historical narrative built around the imperial line from the Jimmu emperor forward.¹⁸

South Korea. Post-Japanese-colonial reconstruction of Korean national identity around the March First Movement (1919) as founding event, Sejong the Great and the Korean alphabet (Hangul) as cultural anchor, the Korean War as shared trauma, and post-Park industrial modernity as achievement. The Saemaul (New Village) movement of 1970–1979 was, among its other functions, an identity-construction programme.

Taiwan. KMT-era construction of Chinese national identity within the Taiwanese framework (Mandarin as national language, Republic of China historical narrative, Sun Yat-sen’s Three Principles), contested by post-1987 Taiwanese-nationalist identity construction that has produced a distinctive Taiwanese identity separate from mainland Chinese identity by the 2020s.

Singapore. The most explicit and most operationally sustained identity construction of the Asian cases. Bilingual education, national service, HDB housing and ethnic-integration policy, the pledge and national songs, the National Day parade, the specifically Singaporean-English public culture, the constructed Singaporean historical narrative built around 1965 as founding moment. All of this examined in detail in Part IV of this series.

China. Party-directed national identity construction around Chinese civilisation as ancient inheritance, humiliation narrative (1839–1949) as motivating history, socialist modernisation as ongoing project, and Chinese nation as continuing subject. Recent decades have added specifically Chinese-Communist-Party-mediated identity content that has produced tensions with Hong Kong, Taiwan, and international perceptions but has been operationally effective within mainland China.

The Ethiopian variant. The absence. Neither the Derg’s Marxist-revolutionary framework, nor the EPRDF’s ethnic-federalism framework, nor the Prosperity Party’s Medemer framework has produced the operational pan-Ethiopian identity that the state requires. This is examined at length in Article 2 and in Part VI.

What the formula shows

The five moves are complementary. Land reform without education does not produce a workforce capable of industrialisation. Education without meritocratic bureaucracy does not produce a state capable of coordinating industrial development. Bureaucracy without export orientation does not produce durable industrial transformation. Export-oriented industrialisation without identity construction does not produce a political community capable of accepting the distributional consequences of the transformation. Identity construction without economic delivery does not produce durable political consent.

The Asian comparators executed all five moves as parts of one integrated project. Some cases executed all five well; some executed some well and others poorly; but the integrated project was the framework in every case. Ethiopia has executed variants of each of the five moves in isolation, and has not integrated them into one project. The result is that each Ethiopian variant, examined against its Asian counterpart, appears as a partial move whose developmental payoff has been foreclosed by the absence of the other four.

The next article takes up why identity construction, rather than restoration, is the correct frame for the Ethiopian case — the move that the Asian comparators executed and Ethiopia has not.

---

References

Footnotes

1. On Japanese land reform 1946–1950, see Ronald P. Dore, Land Reform in Japan (Oxford, 1959); Al McCoy, Closer Than Brothers (Yale, 1999). ↩

2. On Korean land reform, see Jong-Sung You, “Demystifying the Park Chung-Hee Myth: Land Reform in the Evolution of Korea’s Developmental State,” Journal of Contemporary Asia 47, no. 4 (2017). ↩

3. On Taiwanese land reform, see Yun-han Chu, “The Political Economy of Taiwan’s Mature Industrialization,” in Contending Approaches to the Political Economy of Taiwan (M.E. Sharpe, 1989). ↩

4. On Chinese land reform, see Vivienne Shue, Peasant China in Transition: The Dynamics of Development Toward Socialism, 1949–1956 (California, 1980). ↩

5. On Meiji education, see Herbert Passin, Society and Education in Japan (Columbia, 1965); Ellen Kirschenbaum on the 1890 Rescript. ↩

6. On Korean literacy and education, see Michael J. Seth, Education Fever: Society, Politics, and the Pursuit of Schooling in South Korea (Hawaii, 2002). ↩

7. On Taiwanese education, see Denny Roy, Taiwan: A Political History. ↩

8. On Singapore’s bilingual education policy, see Lee Kuan Yew, From Third World to First: The Singapore Story (HarperCollins, 2000). ↩

9. On the Derg literacy campaign, see UNESCO reports on the Ethiopian literacy campaign 1979–1985. ↩

10. On MITI, see Chalmers Johnson, MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975 (Stanford, 1982). ↩

11. On Korean civil-service reform, see the process-tracing corruption-control paper by Alina Mungiu-Pippidi’s group (Development of Corruption Control in South Korea). ↩

12. On Ethiopian civil-service reform, see Peter Van Arkadie and Christopher Adam analyses; the Public Service Reform Program documents. ↩

13. Alice H. Amsden, Asia’s Next Giant: South Korea and Late Industrialization (Oxford, 1989). ↩

14. Johnson, MITI and the Japanese Miracle. ↩

15. On Korean industrial policy, see Ha-Joon Chang, The Political Economy of Industrial Policy (Macmillan, 1994). ↩

16. On Chinese reform and opening, see Barry Naughton, Growing Out of the Plan: Chinese Economic Reform 1978–1993 (Cambridge, 1995). ↩

17. On Ethiopian industrial policy, see Arkebe Oqubay, Made in Africa: Industrial Policy in Ethiopia (Oxford, 2015). ↩

18. On Meiji identity construction, see Carol Gluck, Japan’s Modern Myths: Ideology in the Late Meiji Period (Princeton, 1985). ↩