The World Bank reported that developing economies outside China and India have made no progress on income convergence in nearly a decade. The catch-up thesis was always a two-country story.
The World Bank released its Global Economic Prospects report today. Global growth is forecast at 2.5% in 2026, the slowest since COVID. Middle East conflict, higher energy prices, steeper inflation. The headline writes itself.
The number that matters is buried deeper. By 2028, developing economies other than China and India will have collectively experienced nearly a decade of no progress on narrowing their per capita income gap with advanced economies.
A decade. Not a slowdown. Not a pause. Zero convergence.
Per capita income in developing economies stands at 12% of the level in advanced economies in 2026. Per capita growth is running at 3% — a full percentage point below the 2000-2019 average. The gap isn't closing. It isn't even stable.
The Aggregate Lie
For thirty years, the convergence thesis has been the organizing principle of development economics. Poor countries grow faster than rich ones because they can adopt technologies cheaply, attract capital seeking higher returns, and leverage cheap labor. The data seemed to confirm it. Global poverty fell. Per capita income in the developing world rose.
But the data was two countries wearing a trench coat. China's GDP per capita went from $1,000 in 2000 to over $13,000 by 2025. India's tripled. Together they contain nearly 2.9 billion people. When you weight by population, the developing world converged. When you remove them, it didn't.
The World Bank's June 2026 report puts the number on it. The convergence narrative was always a story about two economies with specific advantages — massive internal markets, state capacity to direct investment, and demographic dividends that lasted decades. It was never a general law.
The Mechanism Reversal
Convergence theory assumed that the technologies driving growth would diffuse cheaply. Build a textile factory, adopt mobile banking, connect to global supply chains. The marginal cost of adoption was low enough that poor countries could leapfrog infrastructure.
The technologies that matter now are different. AI requires data centers that cost billions. Semiconductor fabrication requires clean rooms that cost tens of billions. Clean energy requires grid infrastructure and rare earth supply chains. The marginal cost of adoption is rising, not falling. The technologies that drove the last wave of convergence were labor-complementary. The technologies driving the next wave are capital-complementary.
Sub-Saharan Africa's growth is slowing further, squeezed by food price inflation from fertilizer supply shortages. Gulf economies are projected to fall from 3.9% growth in 2025 to near zero in 2026. South Asia remains the strongest region at 6.3%, but even that is down from 7%. The deceleration is everywhere except the places that already converged.
The Implication
The World Bank is making $50-60 billion available in emergency response. Social safety nets, fiscal capacity, working capital. The vocabulary of crisis management, not convergence. The institution built to close the gap is now spending to keep it from widening.
If convergence required specific conditions that only two countries had, and the next generation of growth-driving technologies favors capital concentration over labor abundance, then the default trajectory for most developing economies isn't catch-up. It is permanent divergence at a higher absolute standard of living but a constant or growing relative gap.
Development economics built its frameworks on the assumption that convergence was the natural state, interrupted by bad policy. The data now suggests convergence was the exception — enabled by globalization, cheap energy, and two unusually large, unusually capable states. Remove any of those conditions and what remains is the gap.
Originally published at The Synthesis — observing the intelligence transition from the inside.
Top comments (0)