Kuznets Curve and falling wage share in the time of globalisation

C. P. CHANDRASEKHAR and  JAYATI GHOSH
It is now evident that the period of globalisation has been one of increasing inequality, partly manifested in the decline of labour incomes as shares of the national income in many parts of the world. Are these declines inevitable because of current economic tendencies?

  

The past three decades have witnessed significant increases in inequality of both assets and incomes, particularly within countries. It is also becoming evident that such inequalities are socially and politically damaging, in addition to being economically unjust. Several analysts have linked this to the economic processes unleashed by capitalist globalisation.

But it should be noted that this evidence runs counter to the perception that was widely prevalent among economists in the 20th century, that the long run process of economic growth first enhances and then reduces tendencies for greater inequality.

This stemmed from the argument made by Simon Kuznets in 1955, that inequality would be low at early stages of development, when societies are mostly agricultural and per capita incomes are also low. As industry develops, countries urbanise and economies grow faster, inequalities increase. Then, as countries develop further, the growing political power of lower income groups creates pressures to improve their income share and enables the introduction of broad-based policies for education and social protection.

As a consequence inequality was expected to move along an inverted u-curve over time, increasing as societies develop and then decreasing. This would be reflected not only in the movement of the Gini coefficient (the standard measure of distribution of personal incomes or household consumption) but also in the functional distribution of income: the division of income between (broadly) the remuneration of workers and surplus earnings (consisting of profits, rent and interest).

It was further supposed that in richer societies inequality would be relatively stable and less subject to sudden fluctuations. Indeed, in the developed countries, relatively stable shares of labour in national income had become accepted as a “stylised fact” of economic growth. Therefore concern with the functional distribution of income lagged in academic research.

Increasing inequality

But the processes of the past two decades generated rather different economic tendencies. Increasing inequality has been evident not just in rapidly growing low-income or middle income countries where the Kuznets Curve could still be used to justify it, but also in richer countries that were supposed to display more stable patterns.

Within countries, the functional or primary distribution of income — the division of the national income between workers’ wages and salaries on the one hand and forms of surplus like profits, rent and interest — is the core driver of economic inequality. (Inequality between wage earners can also be significant and has also increased in most countries in the same period.)

This is what drives the secondary income distribution that results from public action. Even the extent to which public policies can affect inequality through fiscal and other measures is obviously crucially dependent upon this primary distribution.

Charts 1 to 6 plot the basic changes in labour shares of income in the world as a whole using data generated by the UN Global Policy Model. Obviously this model only provides approximations of the reality, but these are based on historical data up to 2010 and model projections for the two years thereafter as well as for some intervening years.

The estimates refer to share of income from wages and salaries, and the mixed income of self-employed persons, so they do contain some portion of what could be called profits especially for countries with high proportions of self-employment.

Declining wages

For the world as a whole, as evident from Chart 1, the share of wages and mixed incomes has been coming down continuously since 1980, and the decline has been particularly evident in the high income countries in the G20. Interestingly, the decline is also evident in the developing countries that are part of G20, although the more recent period suggests a stagnant trend around a relatively low share. Indeed, the decline in labour shares in the rich countries is possibly sharper than in the non-G20 countries that are disproportionately lower income countries.

In the developed world, where the income share of wage workers and self-employed persons has expectedly been much higher, the decline in this share has been quite marked especially since 2000. Chart 2 shows this clearly for the United States, and even more so for Japan, whereas the decline is less evident in Canada. It is worth noting that in general, the most significant declines in labour shares are observed in countries where economic strategies have been based on export-led growth, such as Japan and Germany, but they are also evident in other large and rich countries.

This emerges from Chart 3, which shows that within Europe the sharpest falls in labour shares have been in Germany and in the more export-oriented economies of South and East Europe, especially in the period before the Euro Zone crisis.

The processes of globalisation of trade and finance that have generated increasing competitive pressures, and the associated patterns of technological change that have been labour-saving in nature, are generally accepted to explain the reduced bargaining power of workers in all of these societies, and the consequent decline in wage shares of income as well as greater instability of work. However, these shifts cannot be ascribed purely to economic forces, since domestic social and political forces and policies also play important roles. In these developed industrial countries, the pressures of globalisation have not been counteracted by domestic measures that would protect and/or improve the incomes of workers; instead, it has often appeared that policies have been oriented in the opposite direction.

Global capitalism

The pattern in East Asia (Chart 4) confirms that broader forces of global capitalism are only partly responsible for distributional trends within countries. The share of wages and self-employed incomes has fallen quite sharply and dramatically in China and East Asia generally.

The decline has been particularly sharp in the past decade for China and other high income East Asian countries (such as South Korea and Malaysia) and the trend was evident even before the global financial crisis. The emphasis on export orientation clearly played a role in this decline, but it was not the only factor.

Chart 5 shows how the low income countries of Africa showed quite a precipitous fall in the income shares of workers over the 1980s, the period when policies of structural adjustment and fiscal repression in Africa were wreaking enormous damage on employment and living conditions.

The decline continued in the 1990s, though it was less extreme. In North Africa, the pattern has been reversed, however, with rising labour shares in the 1990s and declines thereafter. But in the 2000s, wage shares have improved, reflecting not just the improved external and macroeconomic conditions stemming from the global commodity boom, but also different and less rigid combinations of domestic policies. In South Africa, however, the end of the apartheid regime has not been associated with any improvement in the share of workers’ incomes, which continued to decline and in 2010 did not even reach the levels achieved 30 years earlier.

So what does this suggest about whether in the period of globalisation, falling wage shares and shares of income of self-employed workers in the national income are inevitable? Certainly the tendencies under globalisation towards aggressive global competition and more rapid diffusion of labour-saving technological change, make for powerful forces that reduce the bargaining power of workers everywhere. This means that wage increases do not keep pace with productivity growth in most economies, and so wage shares have declined even when real wages have increased.

However, the interesting point is that this pattern has not been the same everywhere. Some exceptions to the rule in different parts of the world have already been noted. But the region that seems to have bucked the trend most convincingly is Latin America, which has experienced both an improvement in income inequality as measured by Gini coefficients as well as an improvement in labour shares of income in the 2000s.

Bucking the trend

As shown in Chart 6, in recent years the South American region has shown a remarkably different pattern from much of the rest of the world.

The region experienced sharply declining wage shares during the “lost decade” of the 1980s, when policies of structural adjustment and fiscal repression were associated with significant increases in unemployment and informal work and declining power of workers’ unions. In most countries of South America this continued into the 1990s when standard mainstream policies were still very much the norm.

But thereafter Brazil and Argentina have shown that it is possible to have sharp increases in wage share even in a very globalised world — and incidentally this is also true of several other countries in South America (such as Ecuador and Venezuela).

In Argentina, labour income shares plummeted in the 1990s. The recovery in labour income shares after the Argentine crisis of 2001-2002 has been significant and rapid, but it still has not managed to bring the wage share back to levels of the early 1990s. In Brazil, by contrast, the wage share declined more moderately in the 1990s, and thereafter the increase has taken it to levels higher than even before the lost decade.

So how could countries in Latin America buck the global trend for declining labour incomes shares even though this is a heavily globalised region? Some direct factors that are commonly noted are declines in the earnings gap between skilled and unskilled workers, in part due to the expansion of education and increase in government transfers to the poor. But these were possibly secondary factors to the more significant role of social welfare reforms, employment programmes, public spending, as well as taxation on commodity export revenues.

The increase in public spending was itself the result of a shift in the economic policy model in the region that was driven by political changes in many countries, which created more social consensus that the State should serve as the engine of development, provide social welfare and be responsible for public utilities, education, including university education, healthcare and pensions. All of these were reflected in macroeconomic policies, taxation strategies, labour market and social protection policies, and increases in social assistance.

So the conclusion must be that — while globalisation certainly unleashes forces that reduce labour’s relative power and tends to reduce labour shares of national income — this outcome is not inevitable. It can be countered by progressive economic policies that work actively to shift both the growth strategy and current public fiscal policies in favour of workers (including both wage workers and the self-employed). In a world in which economic inequalities are becoming a matter of increasing political concern, this lesson is absolutely critical.