Directed Technical Change

Directed technical change is the principle that innovation does not fall from the sky equally on all production methods. Instead, profit incentives and market size push R&D effort toward technologies that economise on expensive factors. When labour is scarce and wages are high, firms invest in machines that replace workers. When skilled workers command premium pay, they invest in tools that amplify skill. The direction of innovation is endogenous—shaped by relative prices—not exogenous luck.

For much of the 20th century, economists treated technological progress as a black box. Innovation happened at some constant or exogenously accelerating rate, and economists modelled its consequences. This framing was convenient but divorced from reality. Firms do not randomly fund R&D; they fund projects they expect to be profitable. A high-wage country has an incentive to invent labour-saving machines; a low-wage country, labour-using ones. A country awash in skilled workers has an incentive to invent skill-complementary tools.

The direction of technical change therefore reflects the economic environment. This observation, formalised by economists including Daron Acemoglu, transforms how we understand growth divergence and inequality. Innovation is not a gift bestowed equally. It is endogenous—determined by the same forces that drive capital allocation and trade. This reframing explains why some countries remain stuck in low-skill, labour-intensive production: the market size for skill-biased tools is small, so firms do not invest in them, so the tools never materialise, and the poverty equilibrium persists.

Consider an example. In the 19th century, mechanical looms and steam engines emerged to replace unskilled textile workers in Britain and the United States, where wages were high and labour scarce. These technologies were worthless in India, where labour was abundant and wages were low; Indian textile firms stuck with hand looms, which were cheaper to run. The same technological frontier did not advance for all industries equally; it advanced fastest where profits were highest, which was where factor costs made new tools most valuable.

Skill-biased technical change is the modern manifestation. Since the 1980s, computerisation has disproportionately raised wages for college-educated workers and compressed wages for high-school graduates, widening inequality. This was not inevitable. The computer could have been designed to augment routine work and displace managers; instead, it was designed to augment analytical and abstract work. Why? Because the market for computer-aided analytic tools is vast and lucrative—financial services, pharmaceuticals, software itself—whilst the market for computer-aided routine labour-replacement is smaller and faces political and union resistance. Firms pour R&D into tools that complement skills because profit margins there are higher.

Scale matters enormously. A large country with millions of college-educated workers can support a thriving ecosystem of skill-biased tool development; smaller countries cannot, because the addressable market is smaller and development costs are high. This creates path dependence. Once an innovation direction is established—say, skill-biased tools in the United States or Germany—more researchers are trained in that tradition, more venture capital flows to those startups, more universities focus on supporting those fields. The direction becomes self-reinforcing, and countries that start in different directions pull further apart in their productive capabilities.

For poor countries, directed technical change poses a challenge. They cannot easily absorb or adapt skill-biased innovations designed for wealthy labour markets because the underlying skill base is weak. A farmer in rural Bangladesh cannot benefit from advanced precision-agriculture software if she is illiterate. The technology is not “unsuitable” intrinsically; it is unsuitable given her constraints. Conversely, the kind of labour-using, low-skill agricultural technology that would be profitable for her—perhaps improved hand tools or low-cost irrigation techniques—is not being developed anywhere, because no rich country has a market for it. The profitable innovation frontier has moved past the technology frontier that would actually help her.

This implies that growth and inequality can be influenced by directing R&D. Policies that expand markets for labour-augmenting technologies—raising skills broadly, subsidising training for routine workers—can steer innovation in that direction. Countries that have done this, like Scandinavian nations through generous lifelong-learning subsidies, have managed to moderate the inequality side effects of skill-biased change, even as they have reaped the productivity benefits. Conversely, countries that have neglected broad-based skill development have seen their labour-saving-innovation treadmill accelerate, stranding low-skilled workers with fewer and fewer productive roles.

The interaction between directed change and distance to the frontier is also crucial. A leader country developing the latest skill-biased tools may not be the best place for a poor country to import from. The poor country may benefit more from technologies closer to its own factor endowments—labour-using, low-cost, adapted to its climate and infrastructure constraints. But if R&D investment has moved so far toward skill-biased tools that this intermediate technology is no longer being developed anywhere, the poor country is stuck. It either adopts the skill-biased technology prematurely and unemployment rises, or it sticks with old methods and productivity lags.

Policy implications follow. Governments and multilaterals can encourage directed technical change toward broader applications by funding research into labour-augmenting or labour-saving low-cost technologies, subsidising training to make certain labour supplies more abundant (and thus less profitable to replace), or protecting nascent sectors that serve low-income markets. These are not market distortions so much as corrections for the market failure that underdeveloped poor countries face: the private incentives for innovation have moved away from them.

See also