SANTIAGO – Few things excite intellectuals of the old Latin American left like a book on inequality written by a Frenchman. So, predictably, Thomas Piketty’s Capital in the Twenty-First Century has been a big hit. In the two months since the book’s publication in English, many an essay has been penned claiming that the Paris School of Economics professor’s grand oeuvre confirms earlier claims (usually the author’s own) about the perils of inequality in Latin America.
Piketty weaves a grand narrative about the dynamics of capital accumulation in a market economy. In his now-famous formulation, if the rate of return on capital is greater than the rate of growth of the economy, inherited wealth will grow more quickly than wage income, and the owners of capital will receive an increasing share of national output.
No one can deny that the distribution of income is scandalously unequal in Latin America. But it will come as a surprise to Piketty’s boosters (many of whom have yet to read his book) that his theory has little, if anything, to do with the measured dynamics of income distribution in the region.
Piketty’s theory concerns what economists call the functional distribution of income, or the split between providers of labor and owners of capital. But the maldistribution that causes so much unease in Latin America concerns the personal distribution of labor income – that is, the split among wage earners.
That is because the bulk of income-distribution data in Latin America comes from household surveys, which seldom manage to elicit truthful information about how much Piketty’s rentiers, who receive their income as profits, dividends, or interest, actually earn. For example, the results from the 2009 CASEN, Chile’s broad-ranging household survey, suggest that capital income is more equally distributed than labor income.
No one in his right mind should believe this, of course. All this result shows is that owners of stocks and bonds tend to lie to government surveyors.
That, in turn, reveals two bits of information – both grim – about income distribution in Latin America. First, the true personal distribution of income – comprising all income, whether from labor or from capital – is almost surely worse than commonly cited headline figures suggest.
Second, even if all of the capital dynamics that so concern Piketty could be wished away, the distribution of income in Latin America would still be appallingly skewed. And the cure for that maldistribution does not lie just in the sizeable wealth taxes that Piketty advocates.
Why not? Surely, if labor income is skewed, redistributing capital income or assets to the poor can boost overall equality.
A recent paper by economists at the International Monetary Fund, using a new data set covering many countries, is fairly sanguine about the potential for increased redistribution without undermining economic growth. But the paper also offers a reminder that there are limits to how much income the tax system can redistribute.
The authors compare the Gini coefficient (a commonly used 100-point index of inequality, with zero signifying perfect equality and 100 indicating perfect inequality) before and after government taxes and transfers. They show that only a handful of countries redistribute enough to make more than a ten-point difference in the coefficient, and that redistribution-fueled swings of more than 13 points do tend to have an adverse effect on growth.
Of course, real-life redistributions, at least in Latin America, are much smaller than that. The tax reform recently proposed by President Michelle Bachelet’s administration in Chile involves raising an additional 3% of GDP. Even if none of that money leaks and all is redistributed to poorer Chileans, the reform is unlikely to lower the Gini score by more than three points.
The problem is that Chile’s after-tax Gini coefficient is approximately 50 (Brazil, Colombia, and Peru have similar figures), while those of the advanced countries are mostly in the low 30’s or even the high 20’s. Turning Chile and some of its neighbors into countries with OECD levels of equality will require a great deal more than tax reform.
Put differently: if a society’s initial playing field is very uneven, that society will remain quite unequal even after a sizeable fiscal redistribution. The policy focus, therefore, must also be on what Yale University political scientist Jacob Hacker calls “pre-distribution”: changing the market-determined structure of wage incomes.
Three main tools are available for improving the pre-distribution of income. First, education reform – with a strong emphasis on technical training – would endow low-income citizens with new skills that they could supply in the labor market. Second, targeted industrial policy would create the demand for those workers and their newly acquired skills. And, third, modernization of labor markets would lead to more efficient matching of workers’ skills and firms’ special needs in a context of growing heterogeneity.
These policies are not substitutes but complements: they must all be undertaken at the same time. Doing so is not easy. In Latin America, center-left political leaders, concerned as they are with economic and social justice, will have to fashion their own country-specific approaches. There is no ready-made recipe awaiting them in a French economist’s magnum opus.
Read more from "Piketty's Charge"