James Robinson on What It Looks Like When a State is Failing

James Robinson defines national failure primarily as economic failure caused by institutions that channel people’s effort either toward innovation, entrepreneurship and creativity or toward a narrow group’s interests. He argues that Venezuela’s decline reflects extractive institutions, which concentrate opportunity and weaken broad-based incentives. Explaining the country’s trajectory therefore requires examining the politics of Chávez and Chavismo alongside Venezuela’s earlier political history. Robinson says Venezuela under Rodriguez remains extractive, but the discussion ends before he explains in detail how path dependence shaped its course.

Robinson cautions that institutional performance should not be judged only by income per capita, which could make the argument seem to infer poor institutions from poor economic outcomes. He says other indicators, including life expectancy, infant mortality and night lights, can also help assess how a society is functioning. More broadly, institutions reflect a political equilibrium: explaining how they emerge is different from explaining how to change them. Countries that move toward more inclusive systems follow distinct paths, and outside powers may avoid making conditions worse but cannot simply solve Venezuela’s problems.

He traces Latin America’s economic divergence from North America to colonial societies built on coercion, slavery, hierarchy and extractive institutions. The US South also had extractive institutions and remained poorer, less urbanized and less industrialized than the rest of the country until relatively recently; Robinson says the North played the decisive role in establishing the constitutional framework. In Virginia, early Jamestown colonists tried to imitate Spanish extraction of Indigenous people, but there was no Indigenous population there to exploit. Two thirds of the colonists starved after neglecting crop production, and Virginia later sought settlers from Britain and developed more inclusive political and economic institutions, including a legislative assembly with universal male suffrage.

Robinson argues that democracy is generally associated with more inclusive public policy, greater education spending and faster economic growth, but it can also be corrupt, clientelistic or captured. He sees inequality’s political effects as a central danger, pointing to the hundreds of millions of dollars spent on US elections. Trade unions in Western Europe, including Scandinavia, helped build and sustain more inclusive institutions; he warns that de-unionization could briefly raise GDP while increasing alienation and contributing to institutional breakdown over time. Industrial policy, he says, can create competitive industries, but it requires strategy and planning. South Korea’s shipbuilding industry grew from scratch despite Hyundai having built no rowing boat in the late 1960s; fair policy also depends on accountability and transparent contract allocation.

Robinson describes the United States as facing serious institutional problems and a failure of inclusion. He cites extremely low trust and says people without college degrees have no higher income than they did 50 years ago. He sees Trump as a symptom of public discontent, not necessarily a solution, and says Biden did not understand the underlying problems. Robinson urges attention to whether the coming US election is properly conducted and whether anyone tries to undermine its perceived legitimacy. He notes little evidence that US elections are improperly conducted, while claims that they are fake or manipulated can damage institutions.

China, in Robinson’s account, made its economic institutions more inclusive from 1978 onward by allowing entrepreneurship and business formation, while its political institutions remain unreformed and extractive. He expects that arrangement eventually to “go off the road,” though he does not know how; China’s meritocratic tradition and social mobility are advantages. He also says concerns about AI’s effects on jobs, incomes and living standards are justified. Computer diffusion and the displacement of relatively unskilled workers contributed substantially to worsening outcomes for people without college education, and AI may bring large productivity gains. How society distributes those gains, he stresses, will be critical.