Health Policy and Planning, 35, 2020, 1318–1327

doi: 10.1093/heapol/czaa095
Advance Access Publication Date: 10 November 2020
Original Article

The origins of Ethiopia’s primary health care

expansion: The politics of state building and
health system strengthening
Kevin Croke*
Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 677 Huntington Ave,
Boston, MA 02115, USA
*Corresponding author. Department of Global Health and Population, Harvard T.H. Chan School of Public Health, 677
Huntington Ave, Boston, MA 02115, USA. E-mail [email protected]
Accepted on 4 August 2020

Abstract
Ethiopia’s expansion of primary health care over the past 15 years has been hailed as a model in
sub-Saharan Africa. A leader closely associated with the programme, Tedros Adhanom
Gebreyesus, is now Director-General of the World Health Organization, and the global movement
for expansion of primary health care often cites Ethiopia as a model. Starting in 2004, over 30 000
Health Extension Workers were trained and deployed in Ethiopia and over 2500 health centres and
15 000 village-level health posts were constructed. Ethiopia’s reforms are widely attributed to
strong leadership and ‘political will’, but underlying factors that enabled adoption of these policies
and implementation at scale are rarely analysed. This article uses a political economy lens to iden-
tify factors that enabled Ethiopia to surmount the challenges that have caused the failure of similar
primary health programmes in other developing countries. The decision to focus on primary health
care was rooted in the ruling party’s political strategy of prioritizing rural interests, which had
enabled them to govern territory successfully as an insurgency. This wartime rural governance
strategy included a primary healthcare programme, providing a model for the later national pro-
gramme. After taking power, the ruling party created a centralized coalition of regional parties and
prioritized extending state and party structures into rural areas. After a party split in 2001, Prime
Minister Meles Zenawi consolidated power and implemented a ‘developmental state’ strategy. In
the health sector, this included appointment of a series of dynamic Ministers of Health and the mo-
bilization of significant resources for primary health care from donors. The ruling party’s ideology
also emphasized mass participation in development activities, which became a central feature of
health programmes. Attempts to translate this model to different circumstances should consider
the distinctive features of the Ethiopian case, including both the benefits and costs of these
strategies.

Keywords: Primary health care, political economy, Ethiopia

"Let the peasants never be disaffected. Once they are disaffected, it "Even within Ethiopia, many similar projects had been initiated
will be the end of the world. But whatever happens, with the support some years before. The reason for their lack of success, I think, was
of the peasantry, we may stagger, but we would surely make it." the lack of political commitment or political will . . . The govern-
(former Prime Minister Meles Zenawi, quoted in Markakis (2011, ment truly believed that primary health care, including a Health
pp. 248–9)). Extension Program focused on preventive care, should be the centre-
piece of the health system. That is why it succeeded." (former

C The Author(s) 2020. Published by Oxford University Press in association with The London School of Hygiene and Tropical Medicine.
V
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
[email protected] 1318
Health Policy and Planning, 2020, Vol. 35, No. 10 1319

KEY MESSAGES

• This study examines Ethiopia’s primary healthcare reforms. While primary care programmes have failed in many low-income coun-
tries, Ethiopia has scaled up primary care service delivery by building and equipping health centres and health posts, and through im-
plementation of a large-scale community health worker programme.
• Many studies of these reforms highlight ‘political will’ as the central factor. However, in the existing literature it remains unclear how
this political will enabled Ethiopia to overcome the barriers to implementation of primary healthcare programmes at scale that have
hindered progress in other countries.
• This article uses case study methods to identify historical and political factors that contributed to the decisions of political leaders to
prioritize primary health care. It also identifies institutional changes which enabled implementation at scale.
• The decision to emphasize primary health care was rooted in the ruling party’s political strategy of prioritizing rural interests, which
had previously enabled them to govern territory successfully as an insurgency. This wartime rural governance strategy included a pri-
mary healthcare programme featuring community health workers, providing a model for the later national programme. After taking
power nationally, the ruling party also prioritized extending state and party structures into rural areas, enabling service delivery in
these areas. After a split in the ruling party in 2001, Prime Minister Meles Zenawi consolidated power and pursued his developmental
state ideology, including through appointment of a series of highly competent, technocratic Ministers of Health and mobilization of
significant resources from donors. The ruling party’s ideology also emphasized mass popular mobilization, which became a feature of
primary health programmes. This strategy of mass mobilizationhas become more controversial during Ethiopia’s ongoing period of
political transition.

Minister of Health, current WHO Director-General Tedros Globally, there are major evidence gaps regarding the conditions
Adhanom Gebreyesus).1 under which leaders choose to invest in structural, pro-poor health
system reforms, such as broad-based expansion of primary health
care. While there is a large political economy literature on the polit-
ical preconditions for the emergence of a ‘developmental state’
focused on industrialization and economic growth, there is limited
Introduction comparable theory and evidence on the requirements for a health-
Ethiopia’s expansion of primary health care over the past 15 years focused developmental state.2 However on primary care specifically,
has been hailed as a model in sub-Saharan Africa. Since 2004, there is retrospective literature examining the failure of the global
>30 000 Health Extension Workers (HEWs) have been hired and PHC programmes following the 1978 Alma Ata Conference. This
deployed and >2800 health centres and over 15 000 village-level literature suggests that political, institutional, financial and ideo-
health posts were constructed. Ethiopia has seen sharp increases in logical barriers have hindered PHC programmes in developing coun-
access to basic services, although from a low base. For example, just tries. First, in many countries, health budgets have been spent
5% of deliveries took place in health facilities (Central Statistical predominantly on secondary and tertiary curative care in urban
Agency of Ethiopia (CSA), ORC Macro, 2001), but this increased areas, rather than on primary care (World Health Organization,
almost 10-fold to 48% by 2019 (Ethiopia Public Health Institute, 2008). This is largely because political, economic, and even medical
Federal Ministry of Health, ICF, 2019). Based on these achieve- elites prefer this focus on curative services in urban facilities
ments, Ethiopia has been described as a present-day successor to ear- (Packard, 2016). Second, new PHC programmes require significant
lier primary healthcare success stories such as Sri Lanka, Kerala investment, but poor countries are highly budget constrained, even
(India), Costa Rica and China (Balabanova et al., 2013). In 2016, as new technologies and treatments emerge to further strain resour-
the then-Minister of Health stated that 99% of Ethiopians now have ces (Glassman et al., 2018). Donors have not typically emphasized
access to primary health care (PHC) (Admasu, 2016). Momentum or consistently funded primary health care, often preferring to em-
has built for the extension of this model to other countries, especial- phasize verticalized, disease-specific programmes, or else capital
ly across sub-Saharan Africa. investments (such as new clinic construction), technical assistance,
Replicating Ethiopia’s model in other developing countries or drug and commodity purchases, rather than support for recurrent
requires understanding it. yet the origins of this programme and the costs such as community health worker salaries. As a result, even
conditions that enabled this expansion of primary care services are when large-scale PHC programmes have been attempted, many have
not widely studied (Ostebo et al., 2017). This study uses a politically broken down due to gaps in recurrent financing. Third, deployment
informed approach to health system analysis to address this ques- of PHC-focused health workers and services requires significant im-
tion, by analysing Ethiopia’s PHC reforms in the context of the his- plementation capacity, and community health worker programmes
torical and political science literature on Ethiopian politics, political have also often decayed due to limited state capacity for manage-
economy and health system development (Young, 1997; Kloos, ment, oversight, and supervision of the large workforce that is
1998; Barnabas and Zwi, 1997; Lavers, 2019; International required (Perry et al., 2014). Finally, particularly in the post-Alma
Institute for Primary Health Care – Ethiopia, 2019; Berhe, 2020a), Ata era, global ideological currents have not been hospitable to pri-
and by comparing Ethiopia’s experience to the retrospective litera- mary health care. For example, community health worker pro-
ture on global experiences with PHC programmes in developing grammes represent a large expansion of the public sector work
countries (World Health Organization, 2008; McGuire, 2010; force, which fell out of fashion globally shortly after the 1980s
Glassman et al., 2018; Packard, 2016; Rifkin, 2018). (Rifkin, 2018). For others, these programmes are associated with a
1320 Health Policy and Planning, 2020, Vol. 35, No. 10

model of low-quality, basic care by low-skilled providers from from analysis of the secondary literature. Data are presented from
which countries seek to graduate. other sub-Saharan African countries of comparable size to Ethiopia
Yet while these structural factors are considered crucial in the (Nigeria, Tanzania, Democratic Republic of the Congo, and
retrospective literature on PHC in developing countries, they have Kenya).3
been relatively neglected in explanations of the Ethiopian primary
healthcare expansion. Most analyses of the Ethiopian case highlight
‘political will’ as the central variable, yet as is common in global
health research (Fox et al., 2015), they do not analyse the concept of
Results
political will in detail, treating it as an exogenous factor or ‘black Ethiopia’s health system first emerged as Haile Selassie’s imperial re-
box’ concept, which cannot be further analysed. Some analyses gime (1930–74) created Ethiopia’s modern national institutions,
focus on the personal characteristics of senior leaders such as former including the Ministry of Public Health, established in 1947, referral
Prime Minister Meles Zenawi and former Minister of Health Tedros hospitals and medical and health universities such as Addis Ababa
Adhanom Gebreyesus (Minister of Health from 2005 to 2012) University medical school and Gonder Public Health College (1952)
(Donnelly, 2011; Balabanova et al., 2013). While this analysis also (International Institute for Primary Health Care – Ethiopia, 2019,
finds that these individuals were deeply influential, it also seeks to pp. 25–6). These early health structures were mostly in urban
understand the strategies and mechanisms by which these individu- centres and were focused on curative care (Kloos, 1998, p. 509) and,
als and other senior leaders addressed the structural challenges to as a result, reached relatively few Ethiopians. After the 1974 revolu-
PHC implementation that have prevented progress in other settings. tion that overthrew the imperial regime, the avowedly Marxist Derg
The specific choices that leaders make regarding provision of basic regime had ambitious plans for primary health care, including a
health services are driven by their ideas, values and political skill but community health worker programme, but implementation of these
are also shaped by the political constraints and opportunities that plans was undermined by the Derg’s harshly repressive tactics and
they face and are also mediated by the political institutions that disproportionate military spending (Kloos, 1998 ). As a result, when
structure political action. This article analyses the intersection of the Tigrayan People’s Liberation Front (TPLF) overthrew the Derg
structural determinants and individual agency, highlighting aspects and took power in 1991, health services were dramatically under-
of Ethiopia’s historical and political development over the past gen- provided in Ethiopia, even by regional standards. Health was not
eration that spurred leaders to make major investments in PHC, as the immediate priority for the new regime, which first focused on
well as the ideas and institutional choices and innovations which creating a governing coalition incorporating ethnic parties from
enabled implementation. In the discussion section, the limitations of across Ethiopia (the Ethiopia People’s Revolutionary Democratic
this model and the implications for other countries seeking to emu- Front, EPRDF), developing a new constitution and reorganizing the
late Ethiopia’s experience are also considered. federal structure of the country. Despite the lack of major health
programmes at this point, important initial steps in the first decade
of EPRDF rule included the first National Health Policy in 1993 and
the first Health Sector Strategic Plan in 1997. Despite this initial
Methods progress during the first decade of the EPRDF regime, at the time of
This article relies primarily on case study methodologies as used in the first Demographic and Health Survey (DHS) in 2000, just 5% of
political science and related fields (Gerring, 2004). Case studies of deliveries took place in health facilities, compared to 44% in
this type do not seek to test hypotheses or estimate causal relation- Tanzania (1999), 37% in Nigeria (1999), and 40.1% in Kenya
ships statistically, but rather to provide ‘thick description’ of phe- (2003), and just 14.3% of children aged 12–23 months were fully
nomena in question, identify mechanisms of action through close vaccinated, compared to 68% (Tanzania), 51.8% (Kenya) and
observation of causal processes and generate hypotheses for further 16.8% (Nigeria). In the World Health Organization’s (2000) World
exploration. The goal of this approach is to provide specific insights Health Report, Ethiopia’s health system was ranked 180 out of 189
from the Ethiopian case, which can complement findings from in the world.
cross-country and within-country quantitative research designs on Ethiopia began its national expansion of primary health services
the same topic (Willis, 2014). In this framework, Ethiopia is consid- in the second decade of the EPRDF regime, starting circa 2003. The
ered a deviant case (Gerring, 2008), in that its implementation of most well-known component of this was the Health Extension
primary health care has been unusual relative to other low-income Program (HEP), in which over 42 000 HEWs were trained to deliver
countries over the period in question. It is also a theory-motivated basic primary care (Assefa et al., 2019). This health worker pro-
case study: Ethiopia’s primary healthcare policies are analysed with gramme was implemented together with large-scale construction of
reference to a set of specific barriers to PHC expansion, which are primary hospitals and health centres and health posts to house
synthesized from historical and analytical accounts of Packard HEWs. In addition, the Ministry bolstered the higher-level health
(2016), McGuire (2010), Perry et al., (2014), World Health workforce (doctors, health officers and midwives); including by cre-
Organization (2008) and Rifkin (2018). The obstacles to successful ating a cadre of junior clinical staff trained specifically for rural ser-
large-scale primary health programmes described in these accounts vice and then the broader ‘flooding’ policy, i.e. a policy to ensure
are synthesized by the author. Based on this synthesis, I categorize adequate public sector workforce by flooding the market with clinic-
these challenges as distributional, implementation capacity-related, al staff. For the HEW programme, two female secondary graduates,
financial and ideological. How Ethiopian policymakers addressed given 1 year pre-service training, were deployed to each newly con-
each barrier to PHC programme expansion is analysed with refer- structed health post to serve a kebele (5000 people). The pro-
ence to Ethiopia’s political and health system history. In addition, gramme was launched in 2003 in agrarian areas and adapted in
data from Demographic and Health Surveys, National Health following years to pastoralist and then urban areas. Care was largely
Accounts, and international governance metrics such as the World preventive, with 4 major areas of activity (family health, disease pre-
Governance Indicators are analysed to place Ethiopia’s experience vention and control, hygiene and environment and health education
in comparative context and to triangulate against conclusions drawn and communication) and 16 service packages. An important part of
Health Policy and Planning, 2020, Vol. 35, No. 10 1321

each HEW’s responsibility was to train ‘model families’ in each Structural challenges: distributional politics
kebele who would demonstrate good health behaviour to their com- Primary healthcare programmes in developing countries require
munities. After 2010–11, this aspect of the programme was directing scarce resources to poor rural areas. This is often resisted
expanded into the ‘Women’s Development Army’, in which this by political and economic elites, the middle classes and the medical
structure was massively extended, such that a model household was professions, especially when rural citizens are already politically and
identified for every five households (Figure 1). economically marginalized. Underinvestment in primary care had
Over this period of primary healthcare expansion, Ethiopia has characterized Ethiopian health budgets in earlier periods: Kloos
seen important improvements in key health system access measures: (1998) reports that circa 1972, 92% of health expenditures was on
for example, the percentage of women who received antenatal care hospitals.
by a skilled provider increased from 27% in 2000 to 74% in 2019, The EPRDF regime, like the imperial and Derg regimes before it,
while facility delivery increased from 5% to 48% over the same was undemocratic, which could provide some insulation from the
period. The percentage of children receiving all basic vaccinations pressures to channel resources to urban areas that more open
ranged from 14% in 2000 to 43% in 2019. These absolute values as regimes might face. However, many autocratic regimes use their in-
of 2019 were not out of line with other comparator countries, but sulation from popular pressure to favour elite, urban interests. The
the rate of change was relatively rapid; for example as Table 1 EPRDF regime did the opposite, shifting resources to rural areas not
shows, Nigeria achieved a 6 percentage point increase in facility just in health but across multiple sectors. To understand why, it is
births over 15 years, while the same indicator in Tanzania increased necessary to analyse the political and ideological basis of the EPRDF
by 19 percentage points over 17 years, and in the Democratic regime and its historic origins. Political settlements that tax rural
Republic of Congo it increased 11 percentage points over a 12-year production to favour urban residents have been common in sub-
period. Ethiopia’s annual increase in this outcome was roughly twice Saharan Africa in the post-independence era (Bates, 1981), and
as fast as these comparator countries, although from a very low strong incentives for political leaders to extend the reach of the state
base. to remote rural areas have been limited (Herbst, 2000) and at times
How did Ethiopian policymakers address the challenges to pri- resisted by rural populations (Hyden, 1980). Ethiopia under the
mary care that stymied many other countries? In the section, draw- EPRDF sought the opposite strategy, investing in rural priorities
ing the secondary literature and data analysis conducted in 2019– such as agriculture, safety nets and rural services (De Waal, 2013;
20, the four barriers synthesized from the secondary literature on Tadesse, 2015; Lavers 2019; Clapham, 2017) while limiting invest-
primary care (distributional politics, state capacity for implementa- ment in urban areas. This strategy was rooted in the regime’s origins
tion, financing and ideology) are addressed in turn. in the TPLF’s rural insurgency, which grew from a handful of fight-
ers in a remote region in the late 1970s to ultimately defeating the
Derg and taking over the national government in 1991. As a small
guerilla force, the TPLF needed popular support from peasants in
heavily rural Tigray to survive. Multiple accounts of the TPLF ex-
perience from both participants and academic analysts stress that
this focus on peasant well-being was what allowed the TPLF to de-
600

feat other local insurgent groups, and ultimately to defeat the much
larger military force of the Derg regime (Young, 1997; Berhe,
Number of public facilities opened

2020a). The TPLF refined this strategy over the course of their
17 years fighting to control territory in Tigray, focusing their activ-
400

ities during wartime on land reform, rural administration and secur-

ity, popular participation and political education and mobilization.
This political and governance strategy had a well-developed
health component. Barnabas and Zwi (1997 ) describe how over the
200

course of the long guerilla war in Tigray (1975–91) the TPLF devel-
oped progressively more complex health strategies over time in the
territory they controlled. Health programmes started with rudimen-
tary first aid and treatment for combat injuries but gradually shifted
0

1980 1990 2000 2010 2020

opening year of health facility to well-designed public health programmes including a community
health worker programme to provide basic care, a strong emphasis
on prevention (rather than curative medicine) and emphasis on com-
Figure 1 Government health facility openings in Ethiopia by year (source:
National Health Facility Census 2014) munity involvement and local resource mobilization. These health
programmes were an important part of their strategy to govern the

Table 1 Facility deliveries births over time, Ethiopia and comparator countries

Country Start year Percentage of deliveries in health facility End year Total change Change Per year change (percentage points)

Ethiopia 2000 5 2019 48 43 2.3

Nigeria 2003 33 2018 39 6 0.5
Tanzania 1999 44 2016 63 19 1.1
DRC 2007 70 2018 82 12 1.0
Kenya 1993 44 2014 61 17 0.8

Source: Demographic and Health Survey (DHS) and Multiple Indicator Cluster Surveys (MICS) reports, various years.
1322 Health Policy and Planning, 2020, Vol. 35, No. 10

areas that they controlled and win the support of the rural poor. according to international metrics: according to the World Bank’s
These wartime community health experiences were a key reference Government Effectiveness Indicator, Ethiopia’s was ranked 155 out
point for the TPLF leaders when they gained control of the national of 186 countries. This indicator improved sharply in the mid-2000s,
health system (Berhe, 2020b). as Ethiopia shifted discontinuously from an estimated state capacity
When the EPRDF took power nationally in 1991, they needed to equal to Nigeria, to capacity roughly equal to Tanzania (Figure 2).
legitimize their power nationally, given that they represented an eth-
nic minority from a remote part of Ethiopia. The immediate needs
Historical origins
in the early 1990s were post-conflict reconstruction and the consoli-
Despite Ethiopia’s low governance ratings as of 2000, the country
dation of a national political settlement. Focus and budget were
benefitted from a unique heritage of ‘stateness’ due to the continu-
next diverted by the 1998–2000 war with Eritrea. After the war,
ous existence of a centralized state, which had originally been cen-
there was a deep split within the EPRDF. Prime Minister Meles
tred in Axum in northern Ethiopia. From its origins in northern
Zenawi emerged from the split with consolidated power. Prime
Ethiopia, this state gradually extended its control over much of the
Minister Meles increased his personal authority within the regime
central highlands and had reached its current borders by end of the
and made his theory of the developmental state national policy
19th century. The implementation capacity of this early state was
(Clapham, 2018; Berhe, 2020a). The regime invested heavily in their
limited by its feudal organization (Markakis, 2011), but it generated
‘Agricultural Development-led Industrialization’ strategy, rural
enough centralized military power to repulse Italian attempts to col-
roads, electricity, the ‘Productive Safety Net Program’, and other
onize Ethiopia in 1896. Unlike the rest of the continent, therefore,
programmes to target rural dwellers. Major investments in rural
Ethiopia did not have its indigenous political development and state
health care formed an important component of this strategy. By
formation process derailed by colonization. This is the foundation
2016/17, 44% of government recurrent health expenditure targeted
upon which later state-building experience rests.
the largely rural health posts and health centres, compared to 24%
Despite this history, when the EPRDF took power circa 1991,
for public hospitals (Federal Ministry of Health, 2019). These policy
the Ethiopian state was in many ways little better prepared to deliver
choices in turn were rooted in earlier experiences: As former senior
primary care at national scale than other countries in the region,
TPLF and EPRDF official Mulugeta Gebrehiwot Berhe has argued,
which lacked Ethiopia’s ancient state tradition. While the imperial
‘the initial model of governance of the TPLF aligned with the needs
regime built national institutions like the civil service, universities,
and knowledge of the rural poor was the beginning of its later pro-
and public schools and hospitals, these state-building efforts were
poor approach of governance that transcended the liberation war
limited and ultimately halted by the 1974 revolution. The Derg
and shaped its policies of its transition into government’ (Berhe,
regime’s (1974–91) efforts to further strengthen the state were lim-
2020a).
ited by their repressiveness, infighting and massive diversion of
resources into the military. The poor performance of the massive
Structural challenges: state capacity Derg military against lightly armed rural insurgencies and the
Even if distributional politics poses no obstacle to shifting health regime’s inability to address the 1984–85 famine both demonstrate
resources to primary care, a second major challenge is building the low Ethiopian state capacity through the 1980s.
capacity of the state to deliver services in rural areas. Key elements
of state capacity include the ability to extract taxation, coordinate
Insurgent experiences
the activities of multiple social groups and elicit compliance from
While the state that they inherited had been weakened, the TPLF
both citizens and front-line agents of the state (Berwick and
could contribute valuable governing and state-building experience
Christia, 2018). With respect to delivery of primary health, the state
that they had accumulated in Tigray over the course of the insur-
must mobilize resources and then hire, train, deploy, motivate and
gency period. As mentioned above, the TPLF benefitted from suc-
monitor a large front-line workforce conducting complex tasks in
cessful governance of the territory they controlled; a capability
geographically dispersed settings. These are what Andrews et al.
developed out of necessity, since the movement had limited econom-
(2017) refer to as high transaction intensity, high discretion activ-
ic resources and no external sponsor. As described in Barnabas and
ities, which require relatively high levels of state capacity. Circa
Zwi (1997 ), TPLF health leaders iterated, via trial and error, to-
2000, Ethiopia was not known for high levels of state capacity
wards a functional PHC programme. Rather than a process of ‘iso-
morphic mimicry’ where external donors and experts prescribe
0
Government Effectiveness, Estimate

externally generated models for health service delivery (Andrews

et al., 2017), the TPLF went through a difficult but ultimately pro-
-.5

ductive experience of learning how to deliver primary health services

in ways that made sense for the Ethiopian context. Woldemariam
-1

(2018) describes such governing experiences by insurgent groups in

Ethiopia as ‘laboratories of the future’ in which experiments in gov-
-1.5

ernance and service delivery could be carried out.

Yet while they had learned valuable lessons in Tigray, in the im-
mediate post-war phase, the TPLF faced the challenge of scaling ef-
-2

1990 2000 2010 2020

fective governance to all of Ethiopia. This challenge could not be
Year separated from the ever-present ‘national question’ of how
Ethiopia Nigeria Ethiopia’s disparate nationalities and ethnicities could be integrated
Tanzania DRC into a single polity. The TPLF represented ethnic Tigrayans who
Kenya
comprise <10% of the Ethiopian population. Building a state that
could effectively deliver services to maintain support, while also
Figure 2 Government Effectiveness Estimates, Ethiopia and comparator
countries (source: World Governance Indicators) accommodating this diversity, was a complex problem.
Health Policy and Planning, 2020, Vol. 35, No. 10 1323

State-building strategies: centralization, mass mobilization and top- regions, the Capacity Building bureau head emerged as the senior
down accountability and most experienced party figure in the local administration, his
TPLF leaders made several strategic state- and nation-building political status regularly eclipsing that of the wereda chairman’.
choices, which ultimately shaped their ability to implement PHC Having staffed this structure, administrators and new cadres
and other national programmes. The first was a strategy of hybrid such as the "kebele manager" were then managed through top
federalism. They decided to shift from the integrationist model of down accountability mechanisms, including a detailed performance
the imperial and Derg regimes to instead pursue a federal structure management system, which each level of the administrative hier-
in which each region would have their own political parties and re- archy was ranked based on their performance. These ratings mat-
gional governments, operating within regional borders redrawn to tered for promotions and were taken seriously. These state-building
coincide with ethno-linguistic boundaries. Such federal structures reforms, supported by donor funding, strengthened national institu-
are often viewed (as they were by many in Ethiopia) as appropriate tions focused on development implementation. Initial planning for
redress for the historical regional inequities and can strengthen the the national HEW programme dates to this period, and implementa-
representativeness and legitimacy of the political order, but they tion was shaped by these state-building strategies and dynamics.
also imply substantial delegation of powers over policy, budget and
human resources to local political actors, limiting the centre’s ability
Mass mobilization and incorporation
to aggressively implement national programmes (such as Ethiopia’s
This state-building effort was then shaped by electoral develop-
PHC initiatives).
ments. In the period before the 2005 election, the EPRDF, confident
However, the EPRDF’s strategic choices in this area preserved
in their popularity, allowed relatively open competition. They lost
centralized power for national initiatives. The TPLF created a na-
urban areas badly and faced unusually strong opposition showing,
tional coalition of ethnic parties by selecting and training a new set
although the final election results remain disputed. After an initial
of local elites and local parties, instead of by coopting pre-existing
period of direct repression (with several hundred killed in the imme-
elites and institutions. For example, they created the Oromia
diate aftermath by security forces), this threat to the regime spurred
People’s Democratic Organization (OPDO) in Oromia region, ra-
a massive programme of expanding party and state structures at
ther than incorporating the existing Oromo Liberation Front (OLF),
lower administrative levels to incorporate mass participation, coopt
and encouraged the formation of ‘People’s Development
potential opposition and ensure regime support. The party grew
Organization’ parties in other regions, staffed largely by cadres who
from 700 000 members in 2005 to over 5 million in 2010 (Gilkes,
were directly affiliated with the TPLF (Vaughan, 2011; Markakis,
2015). Thousands of new government positions were created in
2011; Lyons, 2019). In this way, despite the construction of a feder-
local government, especially in health, education and agriculture
alist state, EPRDF leaders retained significant direct authority over
(Vaughan, 2011; De Waal, 2015). These new cadres were seen both
local actors. In addition, the EPRDF retained their ‘democratic cen-
as a means of coopting potential opponents, as well as delivering
tralist’ model of decision-making in which EPRDF decisions were
public services in health, agricultural and other sectors, in both ways
taken centrally and then implemented using a hierarchical adminis-
shoring up the regime’s political support (Lavers, 2019).
trative structure (Lyons, 2019, pp. 80–2). Regional governments
These emerging political needs of the regime fit well with the or-
have significant legal autonomy, but de facto autonomy is limited:
ganizational needs of the HEP. Initial evaluations of the programme
they have been governed by EPRDF member parties and also rely
highlighted the need for more fundamental mobilization of com-
heavily on federal budgetary transfers.
munities, building on the model family structure, particularly in
This choice also brought real costs, risking the creation of re-
light of continued high levels of maternal mortality. At this point,
gional parties with limited local legitimacy. These trade-offs were
the idea of the ‘Health Development Army’ emerged (Admasu,
considered acceptable by the architects of the EPRDF regime, who
2016). It combined intensification of the HEP model (desired by
viewed the developmental project as their highest priority.
health policymakers) with the mass mobilization and party expan-
However, unresolved issues of regional representation re-emerged
sion initiatives, which were already underway for political reasons.
sharply in 2015, leading to political conflict, the resignation of then-
In this new model, the HEWs were not the main conduit to citizens
Prime Minister Hailemariam Desalegn and ultimately to the reshap-
but rather they stood at the top of large hierarchy of volunteers; at
ing of national politics under Prime Minister Abiy Ahmed; a process
community level, there was a 1:5 structure, whereby there would be
that continues at the time of writing.
a ‘model household’ for every five households in the country, in line
with the agriculture extension structure (Østebø et al., 2018; Lyons,
2019). This created a unique administrative infrastructure to reach
Building the state and managing it through top down accountability
households with health messages and interventions.The result of this
Once the EPRDF had constructed a centralized decision-making
and the related state- and party-building activities was that, as Prime
structure with reliable local partners, a related state-building chal-
Minister Meles argued, ‘unlike all previous governments, our writ
lenge was staffing and building capacity of lower-level government
runs in every village. That has never happened in the history of
to fulfil Prime Minister Meles Zenawi’s developmental state pro-
Ethiopia’ (Matfess, 2015). Yet the political origins of the structure
gramme. A major state-building effort commenced. After the 2001
meant that the willingness of citizens to participate in 1:5 groups
party split, effective power was shifted from political party structure
voluntarily varied across regions, and according to the dynamics of
to state structures, and senior EPRDF cadres were placed in charge
local politics (Emmenegger Keno and Hagmann 2011).
of ‘superministries’ including the new Ministry of Rural Capacity
Building (Berhe, 2020a), and the EPRDF initiated a series of reforms
aimed at strengthening woreda and kebele-level administration Structural challenges: the role of ideas
(Emmenegger, Keno, and Hagmann 2011). Vaughan (2011) notes Another common challenge for primary healthcare programmes is
that ‘the capacity building of state structures was a strongly political ideological. The 1978 Alma Ata declaration is widely cited in
enterprise and accorded the highest government priority. By early Ethiopian policy documents, suggesting that international ideas
2004 at wereda level in many parts of the four EPRDF-administered about PHC were influential. However, several Ethiopia-specific
1324 Health Policy and Planning, 2020, Vol. 35, No. 10

ideologies and currents of thought were more important, most not- pressures, as governments faced economic crises and cut health
ably former PM Meles Zenawi’s distinctive conceptualization of the budgets. Ethiopia certainly faced financial challenges with respect to
developmental state (De Waal, 2013; Gebregziabher, 2019). Meles’ its PHC investments: The primary healthcare investment programme
ideas focused on the role of the state in fostering economic develop- (including the HEP) required an estimated $1.2 billion in start-up
ment; he was well-known for rejecting advice from donors about the costs over 5 years (Workie and Ramana, 2013). By contrast,
need to limit the size of the state and liberalize Ethiopia’s economy Ethiopia’s domestically financed government health expenditure
(De Waal, 2013). Hiring 30 000-plus public sector HEWs was not circa 2007–08 was 2.5 billion Ethiopian Birr ($267 million).4
on the donor policy agenda, but it fit in well with Meles’ state- Ethiopia’s decision to hire over 30 000 HEWs and build over 2000
centred developmental vision. Even more central to the PHC agenda new health centres and over 15 000 health posts also generated sig-
was the idea that the state’s legitimacy would emerge from develop- nificant recurrent costs. As of 2016, the HEW programme salary
mental progress, rather than by winning free and fair elections. This costs were $31.7 million annually (Wang et al., 2016), equivalent to
developmental progress, in Meles’ view, was only possible through 21% of the health sector recurrent budget. The HEW programme
the actions of an ‘autonomous state’, i.e. one that pursues a broad remains predominantly donor funded, with 74% of total cost cov-
national interest (defined by the EPRDF leaders) instead of a state ered by donors (International Institute for Primary Health Care –
that was captured by rent-seeking individuals or groups. Ethiopia, 2019).
This idea about state autonomy also helps explain why the The Ethiopian government took on important financing respon-
EPRDF’s rural policies were implemented in a programmatic rather sibilities at the same time that it skilfully leveraged large health aid
than clientelist way. The EPRDF’s focus on supporting rural dwell- inflows. Much has been made of the decision by the Ethiopian gov-
ers is not completely unusual, especially in agrarian economies ernment to take financial responsibility for the HEP, by making
where most people live outside cities, but rural transfers are often HEWs full government employees (at woreda level), in contrast to
implicitly used to maintain the ruling political coalition and distrib- the lower cost voluntary HEW model favoured in other countries.
ute rents. The EPRDF’s ideological emphasis on programmatic ra- The government’s assumption of this responsibility was critical; it
ther than predominantly clientelist approaches to rural communities avoided reliance on volunteer labour for this critical cadre, which
flows from this element of Meles’ thinking, as well as from earlier had contributed to the demise of many other community health
TLPF organizational culture and practices, which were instituted worker programmes (Perry et al., 2014).5 At the same time, the gov-
to avoid personalization and clientelist politics (Berhe, 2020a, ernment shifted investment costs onto donors and, to a lesser extent,
pp. 54–5). communities: local communities contributed labour and materials
Other elements of the TPLF’s stated ideology of ‘revolutionary for the construction of health posts and donors contributed to train-
democracy’ (Gebregziabher, 2019) also affected how PHC pro- ing and to financing equipment and supplies for health posts. The
grammes were implemented. One element of this ideology suggests government has also been slow to increase nominal salaries in the
that the state, given the absence of a large rural middle class, should face of sustained inflation, decreasing the real income of HEWs, and
engage directly with the rural population, rather than with local the cost to government, over time.
non-EPRDF elites. In Zenawi’s words, ‘in agricultural areas we do Rapid economic growth also opened up some fiscal space:
not make coalitions with elites: the only coalition we want to make Ethiopia’s GDP grew 9.9% per year from 2003 to 2015, compared
is with the people’ (Vaughan, 2011). In practice, this meant induc- to the regional average of 5%.6 Yet while government health ex-
ing, and at times coercing, rural people to participate in state- penditure grew from $1.90 per capita in 1999/2000 to $3.20 in
organized developmental activities. The Health Development Army 2010–11, this was far surpassed by the increase in donor health aid,
fit well into this ideological framework. which grew from $0.90 to $10.40 per capita over the same period
More directly, the idea of a national community health worker (World Bank, 2016). Development assistance for health grew sharp-
programme was modelled on previous initiatives, in the health sec- ly over this period for many developing countries. However, the
tor as well as from the agriculture sector. A historical account from Ethiopian government’s ability to attract resources for the health
the International Institute for Primary Health Care – Ethiopia argues sector, and to do so on essentially its own terms, was unusually suc-
that ‘the ideation of HEP began from the “Model Family” approach cessful (Teshome and Hoebink, 2018). The country created and
initiated in 1997 by the Tigray Regional Health Bureau’, while enforced aid management processes and institutions such as five
others emphasize the earlier wartime community health worker pro- year Health Sector Development Plans, a donor code of conduct,
gramme, also in Tigray region (Barnabas and Zwi, 1997; Berhe, health aid coordination groups and several large pooled funding
2020b). To a lesser extent, the idea of PHC as the organizing prin- mechanisms that brought aid funds on budget and under govern-
ciple of the health system started in the Derg years, including with ment control. Equally important was Ethiopian policymakers’ skill
focus on prevention and on primary interventions delivered by com- in convincing donors to back the locally developed primary health-
munity workers (Kloos, 1998). Finally, Dr Tedros himself has care programme rather than verticalized programmes exclusively
pointed to Ethiopia’s agricultural extension programme as a model (Bulletin of the World Health Organization, 2009). The Global
that policymakers drew on for the HEW programme (Witter and Fund and GAVI, for example have been criticized for resisting health
Awowsusi, 2017; International Institute for Primary Health Care – systems-focused investment in the heath workforce and facility con-
Ethiopia, 2019). There were thus a number of ideological currents struction, but for Ethiopia, they supported the construction of over
and local policy models that supported primary health care, based 500 of the 2500 health centres that were built. The multi-donor
on centrally-directed community mobilization and implemented by Protection of Basic Services Project funded purchase of drugs, com-
community health workers. modities and equipment for a large proportion of the health posts
and health centres (International Institute for Primary Health Care –
Ethiopia, 2019, p. 84; Lavers, 2019). Teshome and Hoebink (2018)
Structural challenges: financial barriers highlight local ownership of health plans, strong technocratic lead-
In many developing countries in the post-Alma Ata period, large- ership of the Ministry of Health and low corruption as contributors
scale primary healthcare programmes fell apart due to fiscal to the Ethiopian government success in mobilizing aid funds. This
Health Policy and Planning, 2020, Vol. 35, No. 10 1325

highlights how donor contributions to a sector can reflect not just implemented since 2018 by Prime Minister Abiy Ahmed have impli-
aid dependence but can also reflect state capacity, as Ethiopian lead- cations for health programmes. In this climate, the mass mobiliza-
ership’s investment in a more capable state attracted large aid tion components of the PHC programme, such as the Women’s
inflows. Development Army, face increasing criticism: a recent review of the
HEP conducted in 2018–19 states that WDA agents are now ‘mostly
considered as political agents’ and recommends reforms to the struc-
ture (Merq Consultancy PLC, 2019, p. 21). Others have similarly
Discussion and conclusion argued that the project of using party-linked cadres to mobilize soci-
Ethiopia’s primary healthcare expansion is widely praised by leading ety for health and other development activities cannot easily be sep-
global health organizations and leaders and is often seen as a model arated from the imposition of social and political control (Maes
for primary care in low-income countries. External observers and et al., 2015) and that this is unsustainable in a polity undergoing lib-
Ethiopian policymakers alike stress ‘political will’ as the causal fac- eralizing reforms (De Freytas-Tamura, 2017; Fick, 2019).7
tor (Admasu, 2016; International Institute for Primary Health Care Yet even if these elements of the PHC programme are reformed
– Ethiopia, 2019), but few analyses unpack both the historical roots or discarded, the programmatic orientation of public policy priori-
and the specific political and health policy strategies that enabled tizing rural areas, combined with investment in state building and
Ethiopia’s PHC programmes to be implemented at scale (Lavers, implementation, provides a foundation for continued health system
2019; Ostebo et al., 2017). strengthening. Several middle-income countries demonstrate the
This article highlights the ways that a primary healthcare pro- power of such rural-based political coalitions to drive health invest-
gramme based on community health workers and mass incorpor- ment: At roughly, the same time period that Ethiopia was imple-
ation of the population was consistent with the broader political menting its PHC programmes, for example both Thailand and
and developmental strategies of the EPRDF regime. In line with Turkey saw the election of populist leaders with rural political bases
other analyses, this article finds that the characteristics of political who passed and led the implementation of ambitious programmatic
leaders were critical, most notably Prime Minister Meles Zenawi’s health reforms (McCargo and Zarakol, 2012; Harris, 2015; Sparkes
developmental state vision for Ethiopia, and the ability of a series of Bump and Reich, 2015).
Ministers of Health starting Dr Tedros Adhanom and Dr There has also been recent critical analysis of the effectiveness of
Kesetebirhan Admasu. Yet Meles died in 2012, Dr Tedros left the the primary care model, including whether the HEW service pack-
Ministry of Health in the same year, and despite this, progress has ages, largely focused on health education and prevention, remain
continued, with a series of highly regarded and dynamic ministers of the most important priority, as the burden of disease changes over
health and continued ambitious health sector initiatives, such as the time. Globally, there is a renewed emphasis that health system qual-
2015–2020 Health Sector Transformation Plan. The Ethiopian pri- ity must go hand in hand with expanded access (Kruk et al., 2018),
mary healthcare experience cannot be completely reduced to the while in Ethiopia, a recent study found that Ethiopia’s large-scale
leadership of small number of individuals. health facility construction programme increased the utilization of
Rather than exclusively driven by ‘political will’ by an unusually health services but was not associated with reduced neonatal mortal-
dedicated Prime Minister or Minister of Health, Ethiopia’s primary ity (Croke et al., 2020). In recent reviews of the HEP, demand for
healthcare programme is the result of a historical process, enabled higher quality care and better-equipped health posts with more con-
by Ethiopia’s longstanding state traditions and rooted in policy sistent drug supply has been highlighted (International Institute for
experiments carried out by post-war political leaders during their Primary Health Care – Ethiopia, 2019). Programme reviews have
previous experience as a guerilla army in northern Ethiopia. These also highlighted growing dissatisfaction on the part of HEWs, who
experiences also led these leaders to believe strongly in prioritizing remain modestly paid with large populations to serve, continuously
the welfare of rural Ethiopians as the foundation for their political broadening scope of activities, and limited paths for promotion
success. Post-2001, the implementation of PHC at national level within the health system (Merq Consultancy PLC, 2019).
was enabled by a series of state-building strategies. These strategies Yet other Ethiopia’s experiences still contain lessons for other
were shaped by Ethiopia’s top-down model of federalism and the countries, even if the specific historical circumstances obviously can-
ruling party’s ideas about ensuring political stability and develop- not be imitated. One point is that ‘political will’ is powerless with-
ment through service delivery models that incorporated and coopted out functional state institutions. Yet political will can be exerted in
the Ethiopian population. These institutions supported service deliv- support of a state strengthening project. In Ethiopia, the desire to de-
ery but were also used to incorporate and control popular participa- liver services to rural communities was enabled by investment in the
tion, even as other avenues for civic mobilization were closed off. personnel and systems at local level, including meaningful mecha-
These processes were influenced by global heath ideas about pri- nisms of performance management and accountability. It is also not-
mary health care and supported by global funding streams. able that the ideas and political pressures that motivated this state-
However, global ideas and funding were all reshaped in fundamen- building project were locally generated and differed significantly
tal ways by local actors. These reformulated and adapted ideas from global ‘best practice’ institutions. The HEP ran counter to
about primary health care have in turn become influential and are mainstream donor thinking circa 2003, as did the whole range of
widely cited in global forums, including by international organiza- local governance institutions. Yet these heterodox local institutions
tions, which were originally sceptical about Ethiopia’s initiatives. appear to have generated rapid improvements in access to a range of
This is an example of policy diffusion in global health of an ap- rural public services, including primary health care.
proach originating from the global south and gaining broad influ- From a theoretical perspective, this article highlights both the
ence (Edwards, 2020). path dependency of development processes, as well as their contin-
However, this narrative complicates any simple approach to gency and the role of leaders and ideas. A major argument of the art-
translating Ethiopia’s model by other developing countries. First, icle is that Ethiopia’s primary health reforms were deeply shaped by
the political unrest that began in Ethiopia in 2015 has demonstrated the country’s history, including long-term processes of state forma-
the limits to EPRDF’s top down approach. The liberalizing reforms tion and, more importantly, the wartime circumstances that led the
1326 Health Policy and Planning, 2020, Vol. 35, No. 10

TPLF to develop rural-focused policies, including primary health Acknowledgement

care. But the process by which these elements were translated into a This research was supported by a grant from the Harvard T.H. Chan School
sustained national programme was subject to contingency and indi- of Public Health’s Dean’s Fund for Scientific Advancement.
vidual agency. It also highlights an important role for the subjective Ethical approval. This project was approved by the Harvard Longwood
ideas and understandings of political leaders (Lavers, 2018), and Medical Area IRB, proposal IRB19-1743.
political organizations that can diffuse and institutionalize those
ideas. Ethiopia’s innovative health policies were not national prior-
ities, even under EPRDF rule, until after the 2001 ruling party split References
and the resulting consolidation of power under Meles Zenawi. Admasu KB. 2016. Designing a resilient national health system in Ethiopia:
Meles’ distinctive ideas about the developmental state and how to the role of leadership. Health Systems & Reform 2: 182–6.
ensure programmatic rather than rent-seeking policies, and the or- Andrews M, Pritchett L, Woolcock M. 2017. Building State Capability:
ganizational forms and ideological apparatus of the EPRDF, were Evidence, Analysis, Action. New York: Oxford University Press.
central to the implementation of primary healthcare programmes. Assefa Y, Gelaw YA, Hill PS, Taye BW, Van Damme W. 2019. Community
Health Extension Program of Ethiopia, 2003–2018: successes and chal-
This contingency is highlighted again by the current political uncer-
lenges toward universal coverage for primary healthcare services.
tainty in Ethiopia, the resolution of which is likely to affect the fu-
Globalization and Health 15: 24.
ture of health policies and primary health care. Balabanova D, Mills A, Conteh L, et al. 2013. Good health at low cost twenty
The limitations of this article are that it is a single case study and five years on: lessons for the future of health system strengthening. The
that it is largely based on the secondary literature, which may em- Lancet 381: 2118–33.
phasize positive aspects of Ethiopia’s post-1991 trajectory, especial- Barnabas G, Zwi A. 1997. Health policy development in wartime: establishing
ly given the political sensitivities around the regime and lack of the Baito health system in Tigray. Health Policy and Planning 12: 38–49.
freedom of expression in EPRDF-governed Ethiopia. Trends in gov- Berhe MG. 2020a. Laying the Past to Rest: The EPRDF and the Challenges of
ernance and key health service indicators from comparator countries Ethiopian State-Building. London: Hurst Publishers.
are presented, but Ethiopia is a relatively unique country and it is Berhe MG. 2020b. Barefoot Doctors and Pandemics: Ethiopia’s Experience and
COVID-19 in Africa. https://africanarguments.org/2020/04/08/barefoot-doc
difficult to identify ideal comparators. More work is needed to place
tors-and-pandemics-ethiopias-experience-and-covid-19-in-africa/. Last accessed
the Ethiopian case in comparative perspective. Despite these limita-
August 22, 2020.
tions however, the Ethiopian case offers insights about the factors Berwick E, Christia F. 2018. State Capacity Redux: Integrating Classical and
that contribute to the implementation of primary healthcare pro- Experimental Contributions to an Enduring Debate. Annual Review of
grammes at scale in low-income countries, and attempts by other Political Science 21: 71–91.
countries in the region to implement PHC programmes of compar- Bulletin of the World Health Organization. 2009. Ethiopia extends health to
able scale and ambition may benefit from the consideration of this its people 87: 485–564.
experience. Clapham C. 2017. The Horn of Africa: State Formation and Decay. London:
Hurst and Co.
Clapham C. 2018. The Ethiopian developmental state. Third World Quarterly
Endnotes
39: 1151–65.
1. Interview with Dr Tedros, June 19, 2013, https://www.bcg. Central Statistical Agency of Ethiopia (CSA), ORC Macro. 2001. Ethiopia
com/en-us/publications/2013/public-health-health-care-payers- Demographic and Health Survey 2000. Addis Ababa and Calverton, MD:
providers-adhanom-ghebreyesus-transforming-health-care-in- Central Statistical Agency of Ethiopia and ORC Macro.
ethiopia, last accessed September 3, 2020. Croke K, Mengistu A, O’Connell S, Tafere K. 2020. The impact of a health fa-
cility construction campaign on health service utilization and outcomes:
2. One exception is Harris (2015) who theorizes about the ‘devel-
analysis of spatially-linked survey and facility location data in Ethiopia.
opmental capture’ of the Thai state led to health reform.
BMJ Global Health, in press.
Lavers (2019) considers similar issues in the Ethiopian context. De Freytas-Tamura K. 2017. ‘We are everywhere’: how Ethiopia became a land of
3. These countries are the six largest by population in sub- prying eyes. New York Times. https://www.nytimes.com/2017/11/05/world/africa/
Saharan Africa, excluding South Africa. ethiopia-government-surveillance.html. Last accessed August 22, 2020.
4. https://www.hfgproject.org/wp-content/uploads/2014/04/ De Waal A. 2013. The theory and practice of Meles Zenawi. African Affairs
Ethiopia-NHA-Findings-Briefing-Notes.pdf, last accessed 112: 148–55.
August 22, 2020 see table 2. De Waal A. 2015. The Real Politics of the Horn of Africa: Money War and the
5. By contrast, the Health Development Army was all unpaid Business of Power. Malden, MA: Policy Press.
volunteers. Edwards Z. 2020. Postcolonial sociology as a remedy for global diffusion the-
ory. The Sociological Review.
6. Author’s calculations from World Bank data, accessed at
Ethiopia Public Health Institute, Federal Ministry of Health, ICF. 2019.
https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.
Mini-Demographic and Health Survey: Key Indicators. https://dhsprogram.
ZG?locations¼ET, last accessed August 22, 2020. com/pubs/pdf/PR120/PR120.pdf.
7. A related criticism of the EPRDF party-state comes from Berhe Federal Ministry of Health (Ethiopia). 2019. National Health Accounts
(2020a), who argues that the gradual merging of party and 2016/17. http://www.moh.gov.et/ejcc/sites/default/files/2020-01/Ethiopia
state has led to decline in state capacity compared to the ear- %207th%20Health%20Accounts%20Report_2016-17.pdf.
lier, more meritocratic and participatory TPLF model. Fick M. 2019. Ethiopia’s surveillance network crumbles, meaning less fear
and less control. Reuters, December 17, 2019. https://www.reuters.com/art
icle/us-ethiopia-politics-surveillance-insigh/ethiopias-surveillance-network-
crumbles-meaning-less-fear-and-less-control-idUSKBN1YL1C2.
Fox AM, Balarajan Y, Cheng C, Reich MR. 2015. A rapid assessment tool to
measure political commitment to food and nutrition security. Health Policy
Funding and Planning 30: 566–78.
This research was supported by a grant from the Harvard T.H. Chan School Gebregziabher TN. 2019. Ideology and power in TPLF’s Ethiopia: a historic
of Public Healthã s Deanã s Fund for Scientific Advancement. reversal in the making? African Affairs 118: 463–84.
Health Policy and Planning, 2020, Vol. 35, No. 10 1327

Gerring J. 2004. What is a case study and what is it good for? American McGuire J. 2010. Wealth, Health, and Democracy in East Asia and Latin
Political Science Review 98: 341–54. America. New York: Cambridge University Press.
Gerring J. 2008. Case selection for case-study analysis: qualitative and quanti- Merq Consultancy PLC. 2019. National assessment of the Ethiopian Health
tative techniques. In: Box-Steffensmeier JM, Brady HE, and Collier D (eds). Extension Program. http://repository.iifphc.org/handle/123456789/560.
The Oxford Handbook of Political Methodology, pp. 645-684. New York: Østebø M T, Cogburn M D, Mandani A S. 2018. The silencing of political
Oxford University Press. context in health research in Ethiopia: why it should be a concern. Health
Gilkes P. 2015. Elections and Politics in Ethiopia 2005-2010. In: Ficquet E, Policy and Planning 33: 258–70.
Prunier G (eds.) Understanding Contemporary Ethiopia: Monarchy, Packard RM. 2016. A History of Global Health: Interventions into the Lives
Revolution, and the Legacy of Meles Zenawi. New York: Oxford University of Other Peoples. Baltimore: Johns Hopkins University Press.
Press. Perry HB, Zulliger R, Rogers MM. 2014. Community Health Workersin
Glassman A, Keller JM, Lu J. 2018. The Declaration of Alma-Ata at 40: Low-, Middle-, and High-Income Countries: An Overview of Their
Realizing the Promise of Primary Health Care and Avoiding the Pitfalls in History,Recent Evolution, and Current Effectiveness. Annual Review of
Making Vision Reality. Center for Global Development working paper. Public Health 35: 399–421.
https://www.cgdev.org/sites/default/files/Glassman-Declaration-Alma-Ata- Rifkin S. 2018. Alma Ata after 40 years: primary health care and health for
40-final.pdf. all—from consensus to complexity. BMJ Global Health 3: e001188.
Harris J. 2015. “Developmental capture” of the state: explaining Thailand’s Sparkes SP, Bump JB, Reich MR. 2015. Political strategies for health reform in
universal coverage policy. J Health Polit Policy Law 40: 165–93. Turkey: extending veto point theory. Health Systems & Reform 1: 263–75.
Herbst J. 2000. States and Power in Africa: Comparative Lessons in Authority Tadesse M.2015. The Tigay People’s Liberation Front.In: Prunier G, Ficquet E
and Control. Princeton, NJ: Princeton University Press. (eds.) Understanding Contemporary Ethiopia: Monarchy, Revolution, and
International Institute for Primary Health Care – Ethiopia. 2019. the Legacy of Meles Zenawi. New York: Oxford University Press.
Evolution of Health Extension Program in Ethiopia: Past, Present, and Teshome SB, Hoebink P. 2018. Aid, ownership, and coordination in the health
Future. sector in Ethiopia. Development Studies Research 5: S40–S55.
Kloos, H. 1998. Primary health care in Ethiopia under three political systems: Wang H, Tesfaye R, Ramana GNV, Chekagh CT. 2016. Ethiopia Health
community participation in a war-torn society. Social Science and Medicine Extension Program: An Institutionalized Community Approach for
46: 502–22. Universal Health Coverage. Washington, DC: World Bank.
Kruk ME, Gage AD, Arsenault C et al. 2018. High Quality Health Systems in Willis B. 2014. The advantages and limitations of single case study analysis.
the Sustainable Development Goals Era: time for a Revolution. The Lancet E-international relations. https://pdfs.semanticscholar.org/0912/
Global Health 6: e1196–e1252. 0ea06684457c89c56184a46e83d3f7834bd8.pdf?_ga¼2.251106624.
Lavers T. 2018. Taking Ideas Seriously within Political Settlements Analysis. 771413045.1592919866-1807458858.1592919866.
ESID working paper 95. Witter S, Awowsusi A. 2017. Ethiopia and the Health Extension Worker
Lavers T. 2019. Towards Universal Health Coverage in Ethiopia’s ‘develop- Program. Oxford policy management case study. https://learningforaction.
mental state’?: The political drivers of Community-Based Health Insurance. org/wp-content/uploads/Learning-for-action-across-health-systems_
Social Science and Medicine 228: 60–67. Ethiopia-cast-study-1.pdf.
Maes K, Closser S, Vorel E, Tesfaye Y. 2015. A women’s development army: Woldemariam M. 2018. Insurgent Fragmentation in the Horn of Africa:
narratives of community health worker investment and empowerment in Rebellion and Its Discontents. New York: Cambridge University Press.
rural Ethiopia. Studies in Comparative International Development 50: Workie NW, Ramana GMV. 2013. The Health Extension Program in
455–78. Ethiopia. World Bank Universal Health Coverage Series 10. Washington
Markakis, J. 2011. Ethiopia: The Last Two Frontiers. Woodbridge: James DC: WorldBank.
Currey. World Bank. Ethiopia Public Expenditure Review 2016. Washington, DC:
Matfess H. 2015. Rwanda and Ethiopia: developmental authoritarianism and World Bank.
the new politics of African strong men. African Studies Review 58: World Health Organization. 2000. World Health Report: Health Systems:
181–204. Improving Performance. Geneva Switzerland: World Health Organization.
McCargo D, Zarakol A. 2012. Turkey and Thailand: unlikely twins. Journal World Health Organization. 2008. World Health Report: Primary Health Care
of Democracy 23: 71–9. – Now More than Ever. Geneva Switzerland: World Health Organization.