How suitable is a ‘developmental state’ to tackle unemployment, inequality and poverty in South Africa?

Reads 20,053
Philippe Burger, 26 March 2014
The National Development Plan envisions the achievement of a ‘capable and developmental state’. Developmental states are usually associated with high economic growth. Such states in East Asia often are seen as models for SA to emulate. However, given the structure of the SA economy, state and society, a developmental state is not suitable, nor attainable. The concept of a social investment state is a better alternative, but it will need key institutional and policy reforms to work.


The concept of the developmental state is prominent in policy discussions regarding unemployment, inequality and poverty. In ANC, government and allied labour-union circles it is an important guiding concept in debating the role of government in addressing these problems. For example, the 2012 National Development Plan (or NDP) (NPC 2012a) sets out the intention to transform the South African state into a “capable and developmental state able to intervene to correct our historical inequities…” (Minister Manuel, foreword to the NDP).     

Developmental states are usually associated with high economic growth. Japan in the 1950s-80s, South Korea in the 1960s-90s, China since the 1980s and Brazil since 2000 are all examples of developmental states, most of which grew at phenomenal rates. Some of these are seen as models that South Africa should emulate.

Although the NDP uses the term ’developmental state’, the precise meaning that it ascribes to the term is not clear. Often the same is true for the proponents (or the opponents) of the concept.

The academic literature displays a similar lack of consensus on what the term ‘developmental state’ means and whether or not it could be useful in the context of South African policy (see Burger 2014 for a discussion). The existence of several, sometimes quite different, understandings of the developmental state makes the debate vulnerable to confusion and opportunism.

This raises the question of the suitability of the developmental state model (in one or another of its forms) to tackle the developmental problems of unemployment, inequality and poverty in South Africa.[1]

1.        The original, Asian developmental state

Chalmers Johnson coined the term ‘developmental state’ in his seminal book MITI and the Japanese Miracle (1982). At the height of the Cold War and during the tension between the world views of the USA and the USSR, Johnson (1999:32) used the concept of the developmental state as a description of the Japanese state – but especially to contrast it with the binary distinction between the USA’s market-based system of capitalism and the completely state-dominated system of the Soviet Union. A developmental state would be a state where government plays a large, active and significant role in the economy in support of selected, mostly private-sector industries.

Key elements, and requirements, of the ‘developmental state’ as understood by Johnson and which crystallised in the experience of some East Asian economies are (cf. Evans 1989, Baissac 2009, Pekkanen 2004):

i)   a singular focus on economic growth as the prime directive of the economy and society: indeed, growth provides the legitimacy of the system;

ii)   a state-led industrial policy with the government actively supporting selected industries;

iii)  a professional bureaucracy that is highly capable and well connected to industry;

iv)  a labour market and an education system that are subject to the singular focus on growth;

v)   an authoritarian regime intent on maintaining stability so as not to undermine economic growth;

vi)  the establishment of a very high saving rate that can be used for investment and capital deepening; and

vii)  the establishment of an export-led growth model.

The Japanese developmental state has served as a model for several East Asian economies. The most notable of these are South Korea, Taiwan and Singapore and, more recently, China.

2.        Other interpretations of the developmental state

Several authors have recently provided broader interpretations of the developmental state than that of the traditional East-Asian model.

Evans (2010) argues that, unlike the 20th century developmental state that focused on manufacturing, the 21st century developmental state will focus on the tertiary or service sector and be knowledge based. As such, governments should concentrate on human capabilities and therefore increase capability-expanding services.

Chang (2010) expands the definition of the developmental state to include not only the pro-corporate developmental states of the East-Asian model, but also the “left-wing Scandinavian ‘developmentalist welfare state’”. Whereas the legitimacy of the East-Asian developmental state depends on its attainment of high economic growth, the legitimacy of the Scandinavian developmentalist welfare state depends on its reaching its social equity objectives through, inter alia, welfare policies and generally active government intervention.

Bernard and Boucher (2007), whilst not using the term ‘developmental state’, situate the above focus on human development and the Scandinavian model in a refined three-part distinction. The latter is based on what they identify as the ‘trilemma’ that governments typically face: they have to keep inequality under control, maximise economic growth, and ensure fiscal restraint (i.e. maintain fiscal sustainability). Bernard and Boucher (2007:220-6) then distinguish three types of state in terms of how a state attempts to overcome the trilemma:

  ■  the Anglo-Saxon ‘liberal state’;

  ■  the Scandinavian ‘social investment state’; and

  ■  the continental ‘transfer welfare state’.

1) In a liberal state, private investment by individuals is responsible for the care of dependents, education, training and life-long learning, as well as the maintenance and restoration of health. Because the liberal state tolerates more inequality, there are fewer efforts by the government to address the inequality that arises from the (private) provision of these social investment goods. As a result, liberal governments spend less on social goods and therefore need to raise less revenue from taxes. Thus, fiscal sustainability is less likely to be a problem and growth is robust and market-driven. 

2) In contrast to the liberal state, the social investment state is responsible for the care of dependents, education, training and life-long learning, as well as the maintenance and restoration of health. What poor individuals cannot afford in a more liberal state, the government in this model provides to everyone (thereby addressing inequality). Thus, the government in such a social investment state spends more. However, even though there are transfers, the focus is not on transfers. Rather, it is on investment in education, health and the freeing-up of time to work (since government takes care of dependents), which has the effect of improving productivity and international competitiveness. This supports economic growth, enables and allows individuals to attain higher levels of income and thereby generate the tax base needed to finance the publicly provided social investment. Thus, fiscal sustainability is less likely to be a problem.  

3) Compared to a social investment state, the transfer welfare state has a larger component of passive social transfers such as old-age pensions, generous unemployment compensation (particularly for the long-term unemployed) and early-retirement compensation. (Note that, while a social investment state such as Sweden also has sizeable transfers, the greater part of the system is focused on social investment.) While inequality could be reduced significantly in a transfer welfare state, the type of expenditure (being mainly transfers) is unlikely to improve productivity and international competitiveness and thus growth. This undermines the revenue base of the government. Thus, with high expenditure and a limited tax base, transfer welfare states might face more fiscal pressure than either the liberal or the social investment states.

Bernard and Boucher (2007:225) report that Scandinavian social investment countries and Anglo-Saxon liberal countries both typically outperform the (mostly Southern European) transfer welfare countries in terms of economic growth and competitiveness. (The liberal states fare worst in terms of inequality.)

The framework of Bernard and Boucher raises the question of whether the term ‘social-investment state’ might be more accurate, analytically, than the more generic term ‘developmental state’. In highlighting the social and human aspects of development, it describes the focus of such a state better and helps to distinguish it clearly from the liberal and transfer welfare states – not just conceptually, but also in terms of performance and outcomes.

The term ‘social investment state’ also clarifies how the Scandinavian developmentalist welfare state differs from the East-Asian developmental state. The East-Asian version focuses primarily on economic growth (via low wages and exports), with inequality and human development being of much lesser importance. As such, the East Asian developmental state is more like a fourth category alongside the liberal, social-investment and transfer welfare states identified by Bernard and Boucher.

3.        Does/can South Africa fit the East-Asian mould?

It is instructive to assess South Africa in terms of the characteristics required of a typical East Asian developmental state.

i) Deriving legitimacy via a primary focus on economic growth: The SA government as a whole does not have a singular focus on economic growth; rather, its stated intention is to reduce unemployment, poverty and inequality. In addition, the Constitution sets out a series of social rights that need to be realised progressively (though subject to the government’s ability to afford them). The latter means that resources cannot solely be dedicated to pursuing economic growth.

ii) A strong state-led industrial policy in support of selected industries: The government launched the Industrial Policy Action Plan (IPAP) in 2007/8 (DTI 2012). However, the budgeted IPAP expenditure of the Department of Trade and Industry (which manages the plan) for 2015/6 comprises only 0.27% of GDP (National Treasury 2013: 124; 165). An IPAP of this scale does not constitute a major state-led industrial policy.

iii) The labour market: The labour market in South Africa is certainly not subject to policies with a singular focus on economic growth. (In East-Asian countries labour rights are proscribed and limited, which keeps wages relatively low in support of higher exports and growth.)

iv) Authoritarianism to maintain stability and secure growth: East-Asian developmental states usually are rather authoritarian and require the containment of civil liberties in the pursuit of growth (Van Dijk & Croucamp 2007:669). Given the high price paid to attain democracy, it is unlikely that South Africans will accept the containment of their civil liberties in the interest of economic growth.

v) A high saving rate: Compared to the typical pattern in East-Asian developmental states, the saving rate in South Africa remains very low. The same is true of investment.

v) An export-driven economy: Although the stated policy, from the implementation of GEAR to the adoption of the NDP, has been that South Africa should use exports to drive growth, exports have remained muted (and the current account perpetually in a deficit). Given that the cost of South African labour far exceeds that in China, India, Brazil, Russia and Turkey, productivity will need to increase to make us internationally more competitive (NPC 2012b). However, this requires improving skills and education – a main area of weakness.

vii) A capable, professional bureaucracy: A competent civil service is key to the success of an East-Asian type developmental state. However, with its problems and lack of capacity in education, health and local government service delivery, South Africa does not resemble those states at all. Moreover, the extent to which such sectors (e.g. education) resist reform will undermine any possibility of implementing an East-Asian type developmental state.

It seems clear that South Africa does not fit the East-Asian developmental state mould.

4.        Is South Africa a social investment state, or actually a 'transfer welfare state'?

South Africa has an extensive social welfare system that comprises a variety of social grants and transfers (including child grants); it reaches over 16 million out of a population of 52 million people and this expenditure constitutes 3.4% of GDP (National Treasury 2014:89). Thus, at first glance it may appear to fit the description of a transfer welfare state.

However, the country also spends a significant amount on education and health care, with 6.9% and 4.1% of GDP respectively having been budgeted for the 2014/5 fiscal year.[2] Does this mean that South Africa would more accurately be categorised as a social investment state?

The main problem in this respect is the low quality of these social services and hence the limited return that they yield in terms of enabling people to become economically active and productive. The low quality and limited return seriously undermine any prospect of South Africa’s becoming a social investment state proper.

This means that, in effect, South Africa is a transfer welfare state – not by design, but by default.

5.        Conclusion: what is desirable, what is possible?

In tackling its dire social and economic problems, South Africa faces a conundrum:

The East-Asian developmental state, with its singular focus on the growth objective and the accompanying suppression of labour earnings and rights, does not and will not fit South African conditions. Its implementation would be politically unfeasible and probably unconstitutional. 

The transfer welfare state is likely to be fiscally unsustainable in the longer run, while the dominance of social transfers does little to lift the poor out of dependency and into skilled and better-paying jobs.

Though fiscally more sustainable, the liberal state is quite unlikely to be an option. The poor will not have the resources to invest privately in education and health and therefore the liberal state will do little to help them to access skilled and thus better-paying jobs. Its lack of concern with inequality would make it politically unacceptable.

 While the social investment state could potentially endow the poor and the unemployed  with the human capital needed to participate in skilled and better-paying employment, it also requires decisive reforms of public education and health and many other government services, coupled with reforms of the civil service and bureaucracy. All these reforms might be very problematic to secure. (Inter alia, resistance to reform in public sector labour unions would have to be overcome.)

It can be concluded that neither the East-Asian developmental, nor the transfer welfare, nor the liberal state is likely able to address the challenges of unemployment, inequality and poverty in South Africa in a sustainable way.

The preferred model for South Africa seems to be the social investment state, with its emphasis on investment in human development and enabling people to participate in skilled and thus better-paying jobs. However, if the needed efficiency in the public sector and reforms of social policy are not forthcoming, any intention to establish a sustainable social investment state will falter.

This would leave South Africa in the current position of being a transfer welfare state by default – and facing a future of continual fiscal pressures and chronic social and political stress. 


Baissac C. 2009. A South African developmental state? The tripartite alliance plan, its probable fate and a perspective into alternatives. Working paper.

Bernard P & Boucher G. 2007. Institutional competitiveness, social investment, and welfare regimes. Regulation & Government 1: 213-9.

Burger P. 2014. Presidential address, 2013 Biennial Conference of the Economic Society of South Africa, Bloemfontein, September 2013. South African Journal of Economics, June.

Chang H-J. 2010. How to ‘do’ a developmental state: Political, organisational and human resource requirements for the developmental state. In Edigheji O. (ed). Constructing a democratic developmental state in South Africa: Potentials and challenges. HSRC Press.

DTI (Department of Trade and Industry). 2012. Industrial Policy Action Plan 2013/4-2015/6.

Evans P. 2010. Constructing the 21st century developmental state: Potentialities and pitfalls. In Edigheji O. (ed). Constructing a democratic developmental state in South Africa: Potentials and challenges. HSRC Press.

--------- 1989. Predatory, developmental, and other apparatus: A comparative political economy perspective on the Third World state. Sociological Forum 4(4): 561-87.

Johnson C. 1982. MITI and the Japanese miracle: The growth of industrial policy, 1925-1975. Stanford University Press.

--------- 1999. The developmental state: Odyssey of a concept. In Woo-Cumings M. (ed). The developmental state. Cornell University Press.

NPC (National Planning Commission). 2012a. National Development Plan.

--------- 2012b. Diagnostic document: Economy.

National Treasury. 2013 & 2014. Budget Review.

Pekkanen R. 2004. After the developmental state: Civil society in Japan. Journal of East Asian Studies 4:363-388

Van Dijk H.G. & Croucamp P.A. 2007. The social origins of the developmental state: Reflections on South Africa and its local sphere of government. Journal of Public Administration 42(7): 664-75.


[1] This article is based on the author’s Presidential Address at the 2013 Biennial Conference of the Economic Society of South Africa, Bloemfontein, September 2013 (see Burger 2014).

[2] This is calculated by using consolidated National, Provincial and Social Security Expenditure. Source: National Treasury (2014:87; 144 Table 6).

Download article

PDF icon Burger P 2014 Developmental State FINAL.pdf
Post a commentary

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. To comment one must be registered and logged in.

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. Please view "Submitting a commentary" for more information.