Skip to content

The Role of Technology in Political Economy: Part 2


Yochai Benkler (@YBenkler) is the Berkman Professor of Entrepreneurial Legal Studies at Harvard Law School and the faculty director of the Program on Law and Political Economy at Harvard Law School.

This post is part of a symposium on the political economy of technology. Read the entire series here.


Yesterday, I outlined the ways in which the dominant “skills-biased technical change” and “winner-take-all economics” explanations of inequality share an idealized view of both markets and technology as natural and necessary. Today I’ll write about the most influential criticism of these dominant stories that have been developed by labor economists. These focus on the central role that institutional choices played in shaping bargaining power, and through it, the ability of the managerial class and shareholders to cause stagnating wages for the median worker and the great extraction by the 1%. Larry Mishel of the Economic Policy Institute, with various co-authors, played a central role for over twenty years in pushing back on the SBTC narrative. Richard Freeman early emphasized the central role of unions, work later extended and deepened by David Card and collaborators. The pro-labor economists’ story is that policy choices as diverse as minimum wage erosion (particularly for women), deregulation, monetary policy, trade, immigration, as well as legal and political attacks on unions and unionization combined to weaken labor’s negotiating power and enable managers and shareholders to extract an ever-growing share of productivity growth, leaving labor running as fast as it can just to stay in place, at best.


In this framework, widely used in left-leaning labor economics, technology is absent as an explanatory dimension. Its role is, at most, to divert attention from the political origins of the shape of inequality in society. If they are right, then those of us spending our lives thinking about how technology shapes social relations are spinning our wheels and wasting energy—at least insofar as we are trying to explain the future or work or the rise and response to inequality.

Unlike the left-oriented labor economists, mainstream new institutional economics does offer an institutional explanation, which can readily be integrated with technology, but leaves substantially less room for egalitarian institutional reform because it is limited by the constraints of market efficiency and the need to sustain innovation to secure growth as opposed to stagnation.


This was clearest in the work of the law and economics movement of the 1970s and 1980s, but even in its most sophisticated contemporary versions, Acemoglu and Robinson, for example, maintain that Nordic social democracies only stay at the technological frontier by latching on to the US-market-incentive-driven innovation coattails.

This intellectual state of affairs leaves those of us working on technology and inequality with an intellectual challenge. Our closest allies in economics, labor economists who do focus on institutions, leave little room for technology to play any role. Those economists who do pay a lot of attention to technology, tend to treat it as natural and necessary, not itself the product of politics and institutions, and largely as a constraint on the ambition of pursuing an egalitarian economic program.

Many of us searching for new answers under the umbrella term of “political economy” tend to look for a revival of Karl Polanyi’s Great Transformation. Polanyi actually had a relatively limited and deterministic role for technology (pp 78-80 of the 2001 Beacon 2d ed.): the development of large expensive machines created a need for a much more fluid labor force than was feasible under existing institutions in the 17th and 18th century, and it was this need to feed the machine that drove the politics and institutional change necessary to turn labor, land, and money into the fictional commodities they had to be in order to have the flexibility to feed the machine.  Enclosure, the Poor Laws, and so forth were necessary to disembed the masses of peasants from their patrimonial feudal social relations and convert them into the urban commodified labor force necessary to feed the new machines; and land had to be commodified in order to destabilize those older social relations and convert uses to feed the machines the raw materials they demanded.  But these very same institutional frameworks, and the decades-long battles over them, also slowed down the process and created friction that enabled these large population-level shocks in social relations to be absorbed and adapted so as to make the great transformation sustainable as a social-political settlement.  Hence the “double movement” of institutions—to both enable the conversion of rural, late-feudal society into market society governed by the commodified exchange of the three fictional commodities—land, labor, and money—and to slow down that transition to a rate that would not result in collapse.


None of these treatments incorporate the insights that Winner’s canonical Do Artifacts Have Politics? emphasized: that technology develops as a function of institutional choices; that it is the subject of politics and the site of politics; and that it makes a difference. His classic examples—Cyrus McCormick’s introduction of pneumatic molding machines worked by unskilled workers in order to break the iron molders’ union, and Robert Moses’s use of low bridges on Long Island parkways to achieve de-facto racial segregation on the beaches—were at the foundation of much of the politics of technology of the past quarter century. Lessig’s “code is law” provided the cri de coeur for one of the most vibrant grassroots activism efforts of its time. The free software movement, net neutrality, open standards, spectrum commons, privacy by design, the copyright wars over DRM and ISP liability; all these have been repeated battles in a quarter century that reflect a belief that technology is anything but exogenous; that it is the outcome of politics, law, and the ethical guidelines followed by technology developers. No one in this movement was a Luddite. If anything, most of us were technophiles.

But the driving assumption was that (a) contrary to both SBTC and its neighbors and Polanyi, technology was very much a function of institutions; (b) cutting edge innovation did not have to follow one narrow “most-efficiency creating” path, but that there was meaningful choice in how innovation progressed; and (c) contrary to the primary explanations of inequality as a function of institutions (deunionization; erosion of minimum wage, etc.), technology had a significant independent role in structuring social relations in the economy, such that winning battles over the dominant designs of the technology could be independently more powerful at structuring social relations than winning political battles or institutional changes that directly regulate those social relations. In its most ambitious version, it could mean that winning political battles over free software or open source hardware could make people better able to live independent lives than winning political battles over labor or employment law. The past decade has led me to be more skeptical of this stronger claim on behalf of technology than I’ve been in the past; but it would be a mistake to abandon the idea that technology and the politics of technological design are of real significance in interaction with other systems through which we exert power over each other and that shape our social relations. Tomorrow, I’ll try to flesh out how this dynamic works to complement the purely institutional pro-labor critique of the rise of economic inequality.