A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Developing countries typically adopt existing frontier technologies (cell phones, solar panels, improved crops) faster than inventing new ones. However, adoption is uneven due to credit constraints, lack of information, and weak incentives. Technology alone is insufficient; complementary skills, infrastructure, and institutions determine impact.
Think of developing countries as latecomers to a technology race that richer countries started decades earlier. The remarkable feature of this position is that the hard work of invention has already been done — the knowledge exists, the technology works, and the price has often fallen dramatically as production scaled up. Mobile phones, solar panels, drought-resistant seed varieties, and oral rehydration therapy are all examples of frontier technologies developed elsewhere that can dramatically improve productivity and welfare in low-income contexts. This is the technology adoption gap: the distance between what a country uses and what is available at the global frontier.
But adoption is rarely automatic, even when a technology is clearly beneficial. Three barriers repeat across development contexts. First, credit constraints: a smallholder farmer who can see that a solar irrigation pump would triple her yields may still be unable to adopt it if she cannot access credit to cover the upfront cost. Second, information failures: people may not know that a better technology exists, how to use it, or whether it has worked for others like them — social learning and demonstration effects matter enormously here. Third, coordination and complementarity problems: a mobile phone network is valuable only if others have phones; improved seeds need fertilizer, which requires a functioning market; a factory adopting automated equipment needs workers who can operate it. These barriers can trap communities at inferior technologies even when better ones are available and affordable in principle.
The lesson is that technology is not self-deploying — it arrives embedded in a context that either enables or blocks its impact. You saw this logic in structural transformation: economies that successfully industrialized did so not merely by importing machines but by building the complementary capabilities around them — training, infrastructure, supply chains, and regulatory frameworks. The same logic applies at a more granular level. Agricultural modernization research, from your soft prerequisite, shows that the Green Revolution succeeded where fertilizer markets, irrigation, and credit access were in place, and failed where they were absent. The technology was the same; the institutional context determined outcomes.
This means that understanding technology adoption in developing countries requires thinking simultaneously about supply (what technologies are available and at what cost), demand-side barriers (credit, information, risk aversion), and the systemic conditions that determine whether adoption translates into sustained productivity gains. Diffusion — the spread of an innovation through a population over time — typically follows an S-shaped path: slow initial uptake while early adopters learn and signal success, rapid spread as social learning and cost declines reduce barriers, and eventual saturation. Policies targeting early-stage barriers — subsidies, demonstration programs, information campaigns — can accelerate diffusion at the point where it is most elastic to intervention.
A final implication is the leapfrogging hypothesis: because developing countries lack legacy infrastructure, they can sometimes skip intermediate technologies and adopt the latest generation directly. Mobile banking (M-Pesa in Kenya) succeeded in part because traditional banking infrastructure was weak — there were fewer entrenched systems to displace. This is not guaranteed, but it illustrates that the absence of old technology is sometimes an adoption advantage, not just a development deficit.