The great global inequality debate7 min read . Updated: 19 Dec 2014, 02:13 AM IST
Trade and technology have both played a role in widening income inequalities
Trade and technology have both played a role in widening income inequalities
The economically uninitiated might believe that raging debates around the causes and consequences of inequality date just to the recent past and spring from the discussion around the recent mega-hit from an academic economist, Capital in the Twenty-First Century by Thomas Piketty.
Those who have read the great historical classics of political economy will already know that this isn’t true, as Piketty’s title evokes Capital, the magnum opus of Karl Marx, who prophesied a falling rate of profit amidst the implosion of the capitalist system and the excesses of a market economy. More generally, discussions around inequality have been central to much of mainstream economics and political science even before Marx, and will no doubt continue well past Piketty.
Yet, with all of its resonance in public debate and discourse, and the celebrity status that Piketty has achieved, one could plausibly argue that, in terms of impact on policy outcomes, a previous generation’s debate on inequality has had greater impact, and may well be more relevant in the future.
The early 1990s were a period of malaise in the US, with fears around what was then called “deindustrialization"—that the US, and other advanced economies, would become largely service-oriented, with most people flipping hamburgers while “good jobs" associated with factory work had migrated to lower-wage developing and emerging economies.
The debate around deindustrialization might have been confined to academic seminars—indeed, I remember attending one when I was a student at Columbia—except for the fact that it soon gained political salience with the approach of the 1992 presidential election and the debate around the proposed North American Free Trade Agreement (Nafta) which was at the centre of it.
Critics of the incumbent president George H.W. Bush argued that, during his administration and in previous years, the US had given up valuable concessions to trading partners through successive rounds of multilateral trade negotiations, and that the process of what we today call globalization (principally through liberalized trade, but also foreign investment, and the proliferation of multinational enterprises, outsourcing, and the like) was jeopardizing the wages of American workers—and, in particular, of unskilled workers—and thereby exacerbating the inequality in wages between skilled and unskilled workers.
The “trade and wages" debate found its way into the pages of newspaper op-eds and even into the presidential debate itself, in which, facing off against president Bush and his chief rival, Arkansas governor Bill Clinton, independent candidate Ross Perot spoke famously of the “giant sucking sound" as Nafta would lead to the displacement of unskilled jobs to Mexico.
While Perot—who also warned ominously of a “race to the bottom" pitting Americans against poorly paid and non-unionized foreign workers—might have been a populist fear monger, there was a certain ring of academic credibility to at least some of his arguments drawing a causal relationship between increasing economic integration through international trade and falling unskilled wages. Further, for what it is worth, many believe that Clinton defeated Bush on the strength of voters’ dissatisfaction with the economy—despite the latter riding high in the polls a year earlier, after the First Gulf War.
The central edifice of neoclassical trade theory is the Heckscher-Ohlin-Samuelson (HOS) or “factor proportions" model, and one of its supporting pillars is the Stolper-Samuelson theorem (SST). The SST essentially says that trade protection in the form of tariffs and quantitative restrictions benefits the “scarce" input in the production process, while free trade benefits the “abundant" input—where scarcity and abundance are defined in terms of relative or proportional factor supplies.
In the context of the American political debate, SST would thus imply that protection has benefited unskilled workers, while trade liberalization would benefit skilled workers—the entirely reasonable presumption being that the US is relatively abundant in skilled labour and relatively scarce in unskilled labour compared with its trading partners.
The empirical fact that trade liberalization was followed by a stagnation of unskilled wages, a rise in skilled wages, and a corresponding sharp rise in wage inequality, seemed, on the surface, to bear out SST and validate the concerns of those who blamed international trade for the changing pattern of wages.
Yet, as Jagdish Bhagwati and I argued at a conference at the American Enterprise Institute in 1993—a paper subsequently published in a volume (now out of print) edited by Bhagwati and Marvin Kosters, Trade and Wages: Leveling Wages Down? (AEI Press, Washington, DC, 1994)—blaming free trade may be premature, or, at best, incomplete.
SST posits a conventional channel through which the freeing up of international trade may have an impact on the structure of wages: working through the relative prices of exports and imports, known as the terms of trade. By contrast, Bhagwati and I suggested that trade may work more indirectly to flatten the growth in earnings of unskilled workers and, thus, contribute to widening wage inequality.
The indirect mechanism that we had in mind works through what Bhagwati and I termed “kaleidoscopic comparative advantage". By this, we meant that, thanks to globalization, margins of comparative advantage had narrowed, and so even a small shift in costs might induce a large churning in the structure of production and, hence, the pattern of employment.
In a world of greater churn—in which labour turnover, in particular, is increasing—we conjectured that unskilled workers, who do not have the benefit of a university education and rely on on-the-job training to increase productivity, might suffer proportionately more from increased job turnover than skilled workers, who tend to embody general, rather than job-specific, human capital.
Thus, increasing labour turnover—quite apart from any change in the terms of trade—could potentially be an important, indirect, mechanism through which trade might adversely affect unskilled workers.
A further element to the “Bhagwati-Dehejia hypothesis", as it has come to be known in the literature, conjectures that unskilled workers may experience longer spells of unemployment than skilled workers—perhaps due to lesser intensity of job search or as an additional side effect of degraded job-specific human capital—which will accentuate the differential impact of trade-induced job turnover on the relative wages of skilled and unskilled workers.
A separate argument that we made in the same paper posited that some, although likely not all, of the changing wage structure might be due to the nature of technological change—in particular, that new technologies, such as information technology which was just starting to become widespread in the early 1990s, tend to be biased in favour of skilled workers and against unskilled workers. Thus, a skilled worker (such as a university professor) equipped with a personal computer could replace a pool of unskilled typists, who formerly would type up, say, lecture notes and correspondence, making the latter redundant, and thereby putting downward pressure on unskilled wages.
Interestingly, the thesis that technical change is skill-biased, which appeared in a big way in the academic literature in the 1990s, bears a strong family resemblance to a much earlier literature arguing that capital and skilled labour are complementary in the production process, so that increases in the capital stock are good for skilled workers, and vice-versa—but will have deleterious effects on unskilled workers. The classic reference is a 1969 research paper by the late Zvi Griliches, a brilliant labour economist who many believe was robbed by death of a Nobel Prize.
Reflecting on the shifting contours of the trade versus technology debate, I now feel that—just perhaps—some of us were too sanguine in downplaying the role of international trade and playing up the role of technology in helping to explain wage inequality—although we were certainly right, I believe, in pointing to the role of technology at a time when the doomsayers were pointing the finger of blame largely, if not entirely, at international trade.
In the 20 years since that original debate, most observers, even staunch free traders, would now concede that trade is one of several important culprits in explaining wage and other inequalities. I, too, would agree with this observation, and suggest that trade and technology bear roughly equal culpability in explaining the persistence of inequality in the structure of wages.
That, of course, does not mean that protectionism is the answer, since a nation, in aggregate, benefits from free trade. The challenge is for policymakers to design appropriate adjustment mechanisms to temper the impact of free trade on those adversely affected—unskilled workers in the case of the US and other advanced economies.
Economics Express runs weekly, and features interesting reads from the world of economics and finance.