Inequality in developed countries and Latin America: coordinated, liberal and hierarchical systems

Abstract

The first half of this article explains the enduring disparities in inequality and welfare states across advanced economies in terms of varieties of capitalism and political systems. Where capitalism is coordinated, as in much of northern Europe, political systems are consensus-based with proportional representation (PR); consensus politics and coordinated capitalism reinforce each other in generating relatively low inequality and strong welfare states. Where capitalism is liberal – the Anglo-Saxon countries – political systems are competitive with majoritarian voting: mutual reinforcement of politics and capitalism generates relatively inegalitarian outcomes and safety-net welfare states. The second half of the article develops a model of hierarchical capitalism to show how this reinforces and is reinforced by a Latin American political system with majoritarian presidents and PR legislatures to generate high and persistent inequality and reduce development options.

Keywords:

• inequality
• electoral system
• varieties of capitalism
• Latin America
• diversified business groups
• multinational corporations (MNCs)
• skills

Acknowledgements

We would like to thank Pablo Beramendi, Marco Fernandez, Peter Hall, Penny Harvey, Torben Iversen, Nicola Lacey, Maxine Molyneux, Diego Sánchez Ancochea, Rosemary Thorp, Guillermo Trejo and an anonymous referee. Ben Schneider thanks the Tinker Foundation for financial support. In particular we thank Andrew Schrank, who still has many disagreements with this paper but whose comments led to substantial rewriting.

Notes

1. Key differences lie in public-sector and gendered provision of social services; universal versus earnings-related is no longer so salient.

2. The numbers in brackets are the numbers of countries with the relevant electoral system which have had more than 50 per cent of government years in the relevant partisan category: thus eight of nine PR countries had a left-of-centre government more than 50 per cent of the time, and eight of eight majoritarian countries had a right-of-centre government more than 50 per cent of the time.

3. Voting participation is higher under PR than in majoritarian systems, and this is sometimes explained by the efficacy of a vote in the two systems – a vote in a majoritarian election is less likely to affect the outcome than with PR. While this is true, from a rational actor perspective the effect is virtually zero in both cases and so should not affect participation. This argument moreover says nothing about why the poor should vote less than the rich; in a rational choice perspective, the rich – to whom time is more valuable than it is to the poor – should vote less than the poor.

4. Ireland is a special case.

5. For background and elaborations, see Schneider (2008a,b,c, 2009) and Schneider and Karcher (2008). These elaborations also address further issues of fit, nuance and intra-regional variation that cannot be accommodated here.

6. Phillips notes that ‘The model is fundamentally neoliberal in character, although its traits are not necessarily consonant with conventional understandings of the Anglo-American variety of capitalism’ and it includes ‘“flexible” labour markets and/or ongoing processes of labour flexibilisation, with high levels of informality in the labour sector and frequently high levels of structural unemployment’ (2004).

7. High-end employers (some MNCs, some large business groups) rely increasingly on a dual strategy where they outsource most unskilled jobs and hire a small core workforce in which they assume they will have to provide most of the necessary training. For the small number of core workers, turnover is low and firms invest heavily in training. However, the net impact of these training policies is to reinforce the low incentives individuals have to invest in their own skills, because high-end employers are less looking to acquire skills in the labour market than seeking workers to whom they can impart the skills they need.

8. One major study on the constraints that kept students from investing as much as they should in education concluded: ‘Thus, in order to increase schooling levels in Latin America, it will be necessary to disseminate information on the economic returns to education, as well as find alternative forms of care for young children so that older siblings do not have to drop out of school’ (Menezes-Filho, 2003, p. 143).

9. In the 1990s, outside Mexico, the service sector generated the vast majority of new jobs in the region, though most of these were low-end jobs in areas like restaurants, hotels and domestic service (Stallings & Peres, 2000).

10. Beyond private education, residential class sorting, where higher income areas have their own public schools which then largely cater to children of parents in that area, adds class barriers and fragments potential employment networks.

11. The relationship is looser than the view of ‘entrenchment’ in Morck et al. (2005), because privileged political access in Latin America does not guarantee survival or stem a steady turnover in the ranks of the top business groups.

12. The rapid expansion of tertiary enrolments in the 2000s in countries like Brazil and Chile may address some of the skill shortage. However, much of this education is part-time and of dubious quality, and may do little to help overcome deficits in networks and cultural capital.

13. Liberalized financial markets further fuel these fears, in two different ways. First, they penalize governments that pursue Keynesian strategies premised on deficit spending which in turn can compromise overall growth. Moreover, they also increase the already well-developed ability of higher-income groups to evade taxation and protect their wealth abroad. Middle-income voters have fewer options for evasion, and hence greater reason to fear that proposals to tax the rich may in practice fall disproportionately on them rather than on the truly rich. In addition, widespread clientelism and patronage (due in part to weak parties, discussed later) promote a suspicion that proposals for redistribution are unlikely to succeed and more likely to end up going for particularistic ends.

14. This is especially true in systems with a small number of large parties (as in Mexico, Argentina and Chile) where each party has presidential ambitions. When not in the presidency, these large parties also have more to gain, with an eye on the next presidential elections, by attacking presidents from opposing parties rather than entering coalitions with them.

15. In the wave of market-oriented reforms, individual labour rights were reduced or made more flexible and union membership declined. However, unions managed to shore up collective or union rights and regulations (Murillo & Schrank, 2005).