Inequality Trends and Their Determinants: Latin America over the Period 1990–2010

By Giovanni Andrea Cornia

For at least the last quarter of the twentieth century, Latin America has suffered from low growth, rising inequality, and frequent financial crises. However, in recent years Latin American countries with different political orientations and economic structures have enjoyed sizeable drops in income inequality. Below, Giovanni Andrea Cornia argues that if left unaddressed, the structural biases of the Latin American economy may block future declines in inequality by retarding the shift to a sustainable and equitable growth path.

Trends in Income Inequality

Initial Conditions

The colonial origins of the high income inequality that has afflicted Latin America for centuries have been well analysed by Engerman and Sokoloff (2005),1 who argue that the high concentration in the distribution of land, assets, and political power inherited from the colonial era led to the development of institutions which perpetuated the privileges of a small agrarian and commercial oligarchy well into the post-Second-World-War period. Prados de la Escosura (2007)2 further argues that during the globalization period of 1870–1914 such a high inequality was exacerbated by improvements in the region’s international terms of trade which raised land yields and the land rental/wage ratio to the benefit of large landowners.

The Widespread Decline of Income Inequality over the Period 2002–10

The last decade experienced a Polanyian reversal of the political, economic, and distributive trends of the 1980s and 1990s. Between 2002 and 2010 inequality fell––albeit to a different extent and with different timing––in all 18 countries analysed with the exception of Nicaragua and Costa Rica. In Honduras and Uruguay inequality rose until 2005 and 2007 but then started declining. The unweighted regional Gini, which had risen by 0.32 Gini points a year during the 1980s and 0.16 during the 1990s, fell by 0.5 points over the period 2002–08, 0.47 in 2009, and 1.93 in 2010. These declines were much more sizeable than the earlier rises and in 2010 the region returned to the pre-liberalization level of inequality. In half of the countries these gains benefited not only the poor but also the middle class (Cornia 2012),3 which is increasingly seen as a key player in promoting long-term growth, political stability, and the pursuit of low inequality via progressive taxation, social expenditure, and labour policies.

[ms-protect-content id=”5662″]

About a third of the last decade’s inequality gains can be attributed to the drop in inequality in 2003–04 following its sharp rise during the 2001–02 crisis. However, inequality continued to decline during the subsequent years, including during the crisis of 2009. The decline involved both medium- and high-income-inequality countries and was observed under regimes reflecting all types of political orientation, though the yearly Gini drop was much stronger under the social democratic left regimes. The decline, involving all types of economies, was a bit faster among the industrial economies.

The last decade experienced a Polanyian reversal of the political, economic, and distributive trends of the 1980s and 1990s.
An appreciation of the exceptionality of the recent distributive improvements of Latin America is offered by a comparison of the inequality trends of other regions. During the broad period of 1980–2000, the majority of Latin American countries experienced an increase in inequality, a trend observed also in most other regions. In contrast, during the years 2000–10 inequality rose less frequently and less sizeably than during the previous two decades. However, only in Latin America was there a clear and generalized improvement.

Proximate and Underlying Causes of the Recent Decline in Inequality

The proximate causes of the recent decline in inequality can be identified by means of a simple framework which accounts for changes in both the factorial and personal distributions of income. If Yi is the total income of household i, yi = Yi /ni the average household income per capita, and ni the number of its members, Yi can be equated to the sum of the products of household i’s endowments of unskilled labour (LF, i.e. the number of unskilled adults), human capital (HC, i.e. the number of adults with at least complete secondary education), physical capital (K), land and other non-renewable assets (L), all of them multiplied by their rates of returns, namely ‘uw’ (unskilled wage), ‘sw’ (skilled wage), ‘rk’ (return on capital), and ‘r’ (rent of land and mines).

The next, more complex step consists of relating the changes in the proximate causes of inequality to their underlying causes. Most changes in the latter reflect exogenous shocks or policy interventions. For instance, a change in ‘uw/r’ due to an increase in land/mining rents may be driven by a rise in commodity prices. In fact, the number of possible underlying causes can be quite high. For convenience, in the following analysis they are clustered in five groups: changes in global economic conditions; changes in the rate of growth of GDP and job creation; changes in exogenous factors such as fertility, dependency, and activity rates; changes in the distribution of educational achievements due to efforts at raising enrolments among the poor; and policy factors. The impact of these underlying causes of inequality is reviewed next.

Underlying Causes of the Decline in Income Inequality over the Period 2002–10

An Improvement in External Conditions

During the last decade, the rapid growth of the Asian economies has entailed a rise in the regional terms-of-trade index from 100 in 2000 to 117 in 2008 while the exports/GDP ratio rose from 27.6 to 35.7 per cent (CEPAL 2010).4 In turn, migrant remittances grew rapidly in Central America, and to a lesser extent in Bolivia, Paraguay, and Ecuador, while the regional ratio of official remittances to GDP climbed from 2.2 per cent in the 1990s to 5.4 per cent in 2007–08 (Cornia 2012).3 Furthermore, between 2002 and 2008 and again in 2010 the region experienced portfolio inflows amounting to 2.4 per cent of the region’s GDP. In contrast, the FDI stock broadly stagnated at 22 per cent of the region’s GDP, after having risen sharply over 1995–2002 (UNCTAD 2009).5

Given the high concentration in the ownership of land and mines prevailing in the region and their high capital- and skill-intensity, the recent gains in terms of trade generated, ceteris paribus, an un-equalizing effect on the functional distribution of income. However, if the mining rents accrue to the state or are taxed and redistributed in a progressive way, a rise in them can generate favourable distributional effects. Yet, the evidence suggests a weak relation between terms of trade and revenue/GDP ratio in Latin America. The only relatively strong correlation (r = 0.63) was found for the eight main commodity exporters for the years 2003–07.

Impact of the Rapid Growth of 2002–08 and 2010 on Income Inequality

In the absence of an economy-wide model, the general equilibrium effects of the mid-2000s boom in commodity exports, remittances, and capital inflows are difficult to trace. Yet, as suggested by the ‘balance of payments constrained growth model’ (Thirlwall 2011),6 these events do relax the foreign exchange constraint to growth and, as a result, may raise employment. Indeed, between the average for the 1990s and 2003–08, the average growth rate of GDP per capita tripled in South America and rose by half a point in Central America. In 2009, it contracted by 2.9 per cent but rebounded to 4.2 per cent in 2010 (CEPAL 2010).4

The evidence suggests a weak relation between terms of trade and revenue/GDP ratio in Latin America.
Exogenous Changes in Dependency and Participation Rates

The recent inequality decline might have been due to an increase in labour supply and the simultaneous drop in dependency rates, both of which could be equalizing. Yet, dependency rates had also fallen in the 1980s and 1990s, a period characterized by rising inequality, and López-Calva and Lustig (2010)7 suggest that their contribution to the recent decline in inequality was of limited importance. Inequality might have fallen also due to a faster rise in the participation rate of the poor. In this regard, López-Calva and Lustig (2010)7 suggest that the surge in activity rates had a small equalizing effect on income inequality in Argentina, Brazil, and Mexico, while the opposite was true in Peru.

An Improvement in the Distribution of Human Capital

An improvement in the distribution of human capital among households due to a rise in secondary and tertiary completion rates starting in the 1990s and accelerating in the 2000s especially among the poor contributed markedly to the recent fall in inequality. The rise in schooling generated two effects: a ‘price effect’ as sw/uw fell due to a higher supply of skilled workers, and a ‘quantity effect’ due to a more equal distribution of human capital. Yet, the price effect can also be explained by a decline in the supply of unskilled labour due to demographic factors, the educational upgrading of formerly uneducated workers, a drop in the demand of skilled workers, a rise in that of unskilled workers, and policy changes.

The Spread of Progressive Regimes and New Policy Approaches

During the last twenty years the region experienced a return to and consolidation of democracy which affected income inequality through the adoption of progressive policies. As suggested by Robinson (2010),8 genuine democracy and greater electoral participation reduce the concentration of power in the hands of the elites and facilitate the transition towards non-clientelistic policies. From the late 1990s the region witnessed a shift in political orientation towards left-of-centre governments due to growing frustration with the disappointing results of the liberal policies which had led to a shrinkage of manufacturing and of the industrial working class, a weakening of the unions, rising unemployment, and an enlargement of the informal sector.

The new regimes within the region differ. Some can be defined as ‘social democratic’, as in the case of Chile, Uruguay, and Brazil. A second group (Argentina and Ecuador) developed left-nationalist platforms, while a third (Venezuela, Bolivia, and since 2007 Nicaragua) adopted a radical-populist approach that also entailed redistribution of assets. All of them have abandoned the notion of a revolutionary break in favour of electoral politics and respect for the institutions of liberal democracy. In a sense, the social democratic model is consistent with the ‘redistribution with growth’ paradigm (Chenery et al. 1974),9 while the radical-populist regimes share some policies with the radical ‘redistribution before growth’ paradigm. The main components of these new models are reviewed hereafter:

acroeconomic Policies

With the exception of Brazil and Venezuela, most countries abandoned the 1980s and 1990s regimes of free floats and fixed pegs, and opted for managed exchange rates aimed at achieving a competitive real exchange rate and limiting its appreciation. Consistent with this approach, central banks accumulated large international reserves and in a few cases introduced capital controls. On several occasions, this policy came under pressure, owing to a surge in export receipts, capital inflows, and remittances (CEPAL 2011).4 Such events have eroded only in part the competitiveness of many countries due to the strong real devaluation of 2001–02, although without capital controls, accumulation of reserves, and central bank interventions, the real appreciation and related un-equalizing asset price inflation would have been stronger.

An improvement in the distribution of human capital among households due to a rise in secondary and tertiary completion rates starting in the 1990s and accelerating in the 2000s especially among the poor contributed markedly to the recent fall in inequality.
To support the new exchange rate policy, most countries avoided the traditional pro-cyclical fiscal and monetary biases of the past. Fiscal deficits were typically reduced below one per cent of GDP and in several cases were turned into surpluses. Furthermore, in line with the shift towards a countercyclical fiscal management, the region recorded a primary surplus of 1–2 per cent between 2004 and 2008 and a balanced budget during the fast growth years of 2006 and 2007 (CEPAL 2011).4 The authorities also attempted to control the money supply, the fall in interest rates, and credit expansion triggered by financial inflows through an accumulation of reserves and sterilization. By 2009, only Argentina and Colombia had introduced capital controls, but this policy became more common in 2010–11. In turn, during the crisis of 2008–09, most governments lowered interest rates and expanded lending by public banks.

Tax Policy

During the 2000s tax policy re-emphasized the role of income tax and reduced exemptions, extended the scope of presumptive taxation, cut excises, introduced indirect taxes on luxuries, and, in some countries, launched a surrogate tax on financial transactions and a selective export tax. As a result, the regional tax/GDP ratio rose by 3.5 points over the period 2003–08 and dropped only 0.35 points during the 2009 recession. The increase in world commodity prices contributed to the surge in tax/GDP ratios in a few countries, although also here the revenue rise began before the commodity boom. The revenue surge also affected inequality indirectly, as it allowed the funding of social transfers and public expenditure on education in a non-inflationary manner, and the elimination of the un-equalizing macro instability of the past.

Trade and External Indebtedness

The free trade policies of the 1990s were not overturned, in part because the new exchange rate regime offered some protection to the tradable sector. In contrast, intra-regional trade grew quickly and so did South–South trade, particularly primary commodity exports to the Asian countries. Governments also attempted to reduce their dependence on foreign borrowing. Short-term stabilization agreements were generally not renewed, while Brazil (in 2005) and Argentina (in 2006) prepaid their outstanding debt to the IMF. Meanwhile the region’s foreign reserves grew from 150 to 550 billion US dollars between 2002 and 2009, and its gross foreign debt declined from 40 to 20.4 per cent of the regional GDP over the same period. These changes were likely to be equalizing, as they helped to reduce vulnerability to macroeconomic shocks.

Labour Market Policies

Labour policies explicitly addressed the problems inherited from the two earlier decades, i.e. unemployment, job informalization, falling unskilled and minimum wages, declining social security coverage, and weakening of institutions for wage negotiations and dispute settlements. Most centre-left governments and a few others decreed hikes in minimum wages which reduced the minimum/average wage ratio. Average wages grew moderately, reflecting recognition that, unless backed by increases in productivity, nominal wage raises may fuel inflation with no effect on real wages.

Rising Social Expenditure and Redistribution

In most countries, public social expenditure started rising in the 1990s but has accelerated its upward trend since the early 2000s (Cornia 2012).3 Most of the increase concerned social security/assistance and education, generated positive redistributive effects, and appears to have become more progressive over time (Barrientos 2011).10

Final Considerations

The chapter has argued that in recent years Latin American countries with different political orientations and economic structures often enjoyed sizeable drops in income inequality that were unparalleled in other regions. The trend towards declining inequality will endure if the recent policy changes are intensified and structural reforms introduced––especially in the poorest countries––to deal with the deep-seated polarization that still affects the region. Yet, the Latin American governments may face formidable hurdles in pursuing this agenda, as shown by the opposition encountered in Bolivia, Honduras, and Ecuador, where interest groups nearly stalled attempts at redistribution. Finally, if unaddressed, the structural biases of the Latin American economy––the lack of an industrial policy, low savings, dependence on foreign capital, continued pressure towards real appreciation and commodity dependence––may well block future declines in inequality by retarding the shift to a long-term sustainable and equitable growth path.

The book Falling Inequality in Latin America: Policy Changes and Lessons will be available in bookshops January 2014.

Reproduced by permission of Oxford University Press and UNU-WIDER. Excerpted from Falling Inequality in Latin America: Policy Changes and Lessons by Giovanni Andrea Cornia. Copyright 2014. All rights reserved.

About the Author

Giovanni Andrea Cornia has taught development economics at the University of Florence since 2000. He has also been a visiting professor/scholar at several other European universities. Prior to that he was the director of UNU-WIDER and chief-economist of UNICEF. His main areas of interest are macroeconomics, inequality, poverty, political economy, child wellbeing, and human development. His forthcoming OUP book (co-edited with Frances Sytewart), ‘Towards Human Development: New Approaches to Macroeconomics and Inequality‘ is expected in April 2014.


1. Migration Model’. IZA Working Paper 808. Bonn: Institute for the Study of Labour. Engerman, S. L. and K. L. Sokoloff (2005). ‘Colonialism, Inequality and Long Run Paths of Development’. NBER Working Paper 11057. Cambridge, MA: National

2. Prados de la Escosura, L. (2007). ‘Inequality and Poverty in Latin America: A Long- Run Exploration’. In K. H. Hatton, K. O’Rourke, and A. M. Taylor (eds), The New Comparative Economic History: Essays in Honour of Jeffrey Williamson. Cambridge, MA: MIT Press

3. Cornia, G. A., Oxford: Oxford University Press (2012). ‘Inequality Trends and their Determinants: Latin America’

4. CEPAL (2006, 2008). ‘Panorama Social de America Latina’. Santiago de Chile: CEPAL. CEPAL (2010, 2011). ‘Balance Preliminar de las Economias de América Latina’.

5. UNCTAD (2009). ‘World Investment Report’. Geneva: UNCTAD

6. Thirlwall, P. (2011). ‘Balance of Payments Constrained Growth Models: History and Overview’. Kent School of Economics Discussion Paper, KDPE 1111. Canterbury: University of Kent

@font-face { font-family: “MS 明朝”; }@font-face { font-family: “MS 明朝”; }@font-face { font-family: “Cambria”; }@font-face { font-family: “CalistoMT”; }@font-face { font-family: “CalistoMT-Italic”; }p.MsoNormal, li.MsoNormal, div.MsoNormal { margin: 0cm 0cm 10pt; font-size: 12pt; font-family: Cambria; }p.BodyTextindentp3, li.BodyTextindentp3, div.BodyTextindentp3 { margin: 0cm 0cm 0.0001pt; text-align: justify; text-indent: 11.35pt; line-height: 120%; font-size: 9.5pt; font-family: CalistoMT; color: black; }.MsoChpDefault { font-size: 10pt; font-family: Cambria; }.MsoPapDefault { margin-bottom: 10pt; }div.WordSection1 { page: WordSection1; }

7. Lopez-Calva L. F. and N. Lustig, Declining Inequality in Latin America: A Decade of Progress?, Brookings Institution Press, 2010

8. Robinson, J. (2010). ‘The Political Economy of Redistributive Policies’. In L. F. López- Calva and N. Lustig (eds), Declining Inequality in Latin America: A Decade of Progress? Washington, DC and New York: Brookings Institution Press and UNDP

9. Chenery, H., M. Ahluwalia, C. Bell, J. Duloy, and R. Jolly,  (1974),
CEPAL, Caribe Santiago de Chile

10. Barrientos 2011, “On the Distributional Implications of Social Protection Reform in  Latin America”, UNU/WIDER Working Paper 2011/69