The Traditional Development Ladder
In 1953, South Korea's GDP per capita was $67 – among the lowest in the world, with little infrastructure and no natural resources. Thirty percent of the country couldn’t read. Permanent aid dependency seemed to be a strong possibility – but Park Chung-hee's government made a novel bet. South Korea would compete globally, starting with the simplest products.
Korean women sewed garments in cramped factories for $3 a day, made wigs from human hair, assembled basic electronics. By the 1970s, those garment supervisors became shipyard managers; electronics technicians became automotive engineers. Korea built entirely new and value-adding industries from scratch: steel mills without any iron, petroleum refineries despite importing 100% of their crude oil.
Each rung enabled the next: textiles in the 1960s led to heavy industry in the 1970s, which spurred the development of an electronics industry in the 1980s.
In 1983, Samsung invested $4 billion in semiconductor manufacturing when Japan dominated the market and Korea had zero market share. It looked ambitious at the time, but it worked. By 1992, Samsung led the global memory chip market. By 2020, Korea’s GDP per capita reached $33,000.
South Korea’s GDP per capita from 1960 to 2025. Source: macrotrends.net.
Though South Korea had other critically necessary benefits, such as strong pro-capitalist economic institutions and strategic American support, its ascent depended primarily on a long window of export growth – driven by labor arbitrage.
This development ladder – moving from largely agrarian, low-skill economies, to building manufacturing, to developing services and technology – has been the most reliable development path at scale for developing countries. And it is a pathway that appears to be at risk from this latest wave of AI automation.
Successful pathways to development
In the past 60 years, there have actually been few reliable pathways through the development ladder for low-income countries. I’ll list a few below:
Export-led development: Exemplified by East Asian countries such as South Korea, Japan, China, and Taiwan, the strategy described above has lifted probably over a billion people out of poverty in the past half-century. It has succeeded by transferring capital, technology, and learning-by-doing to massive, previously unskilled labor forces over decades, building via political stability, education, and re-investment in critical infrastructure.
Exporting labor and remittances: Countries such as Egypt, El Salvador, and Mexico receive between 5-20% of their GDP from remittances. While this strategy can supplement economies, it creates external dependencies and brain drains, rather than long-term development or industrial capabilities.
A rough approximation based on % of due to remittances suggests that perhaps 100 - 200 million people worldwide are supported by this strategy.
Resource extraction: With competent governance and favorable geographies, the discovery of natural resources has been a massive windfall for perhaps one to two dozen countries.
The UAE has leveraged its oil revenues to create economic diversification; Chile nationalized copper and enforced fiscal discipline via a copper stabilization fund.
On the other hand, many countries such as the DRC have suffered from the resource curse, which has led to endless cycles of corruption, extraction, and instability.
This strategy has been both fraught with risks for nations with weak institutions and is not easily reproducible.
Tourism: Many countries such as Thailand, Costa Rica, and Greece have successfully leveraged global tourism for economic growth. While tourism supports hundreds of millions of jobs, it offers shallow development roots: it is concentrated in specific regions, is vulnerable to shocks and seasonal fluctuations, and rarely generates technological or industrial spillover effects.
Many other economic strategies have broadly failed to secure long-term growth, or backfired spectacularly:
Import substitution industrialization: Tariffs to protect domestic industries and build local manufacturing (as exemplified by Argentina).
Significant sums of foreign aid: External aid often fails to translate to economic development, for a myriad of reasons (e.g. Lebanon).
Government-led borrowing & debt-fueled growth: Reliance on sovereign borrowing to finance development and stimulus often leads to unsustainable debt cycles and fiscal crises (e.g. Brazil in the 1980s, or Greece in the 2010s).
In short – there actually aren’t many paths for sustainable development for nations without comparative advantages in tourism, geography, or resources. The export-driven development ladder may be the only proven and reproducible strategy that we know works. Billions of lives have been improved by it – and many more billions in emerging countries like India, Bangladesh, and Indonesia have already placed their bets accordingly.
Cheap labor has been the key advantage
Of course, there are dozens of determinants for the success of countries leveraging the export-led development ladder. Crucial requirements have historically included strong political and economic institutions, investment capital, effective education systems, or access to major waterways. Developing countries benefit in comparison to their developed counterparts from reduced land costs, minimal regulatory burdens, and occasionally, generous tax incentives.
The foundation of the export-led strategy, however, is based on one key factor: labor arbitrage. A Bangladeshi garment worker earning $3 a day, competing against an American earning $150, creates a 50:1 wage differential that makes this ladder self-evident even after accounting for shipping costs, quality variance, and supply chain complexity. The entire strategy is optimized around one core advantage: abundant and cheap human labor.
Crucially for development, this system generates organic technology transfers through what development economists call "learning-by-doing" spillovers. Foreign direct inadvertently creates capability-building. Korea's semiconductor industry emerged directly from workers and engineers who learned electronics assembly in the 1970s, then applied those capabilities to progressively more complex manufacturing.
AI could remove the bottom rungs of the development ladder
AI-enabled automation could fundamentally break this equation by reducing opportunities for global labor arbitrage across the board. Specifically, it could make it challenging for new countries to enter export-led manufacturing, while simultaneously automating the emerging digital services market, leaving no proven pathway for late developers.
This is likely to happen in three ways:
First, it could continue to raise the capital requirements for competitive manufacturing to levels late-arriving countries cannot afford.
Export-led industrialization has succeeded historically because bottom-rung manufacturing required minimal upfront capital. Countries could enter with minimal technology and remain competitive through lower wages. The model was self-financing: profitability at each stage funded progression to the next level of development.
In the past, the technology gap between new entrants and incumbents was narrow enough that wage differentials could compensate. Today, manufacturing demands partial automation even in entry-level industries. Globally competitive garment factories require automated cutting systems and complicated inventory management. Modern electronics assembly increasingly relies on automated pick-and-place machines and optical inspection systems.
Late arrivals could have challenges mobilizing the capital required for a single competitive facility, much less build out complementary infrastructure. The technology gap is widening faster than these countries can accumulate capital, effectively pulling up the ladder while they're climbing.
Second, it could accelerate the reduction of employment intensity and knowledge transfer of manufacturing.
The transformative power of export-led development came from systematic learning-by-doing at massive scale. A 1970s electronics facility employing 5,000 workers generated 5,000 learning trajectories: workers learning precision manufacturing, supervisors learning quality systems, engineers learning process optimization. These workers would turnover regularly and disperse these skills widely, by starting businesses, joining competitors, and training others.
Technological automation accelerated by powerful AI is likely to continue to reduce employment intensity dramatically: perhaps by 2-5x in the coming decades. This would break the traditional spillover mechanism: 500 learning trajectories instead of 5,000 means an order of magnitude less knowledge diffusion.
Human labor may also start to concentrate on skills that are more specialized and less transferable. When next-generation factories operate at 90% automation, the skills required (e.g. AI system management, advanced robotics) may not readily transfer to other contexts the way an assembly worker's procedural knowledge did. The ladder's rungs may no longer connect smoothly upward due to less fungible skills – further reducing economy-wide spillovers.
This trend is already emerging globally. Foxconn cut its Kunshan workforce from 110,000 to 50,000 while maintaining output; Adidas recently prototyped an automated footwear facility that employed 160 workers versus 1,200-1,500 for conventional production. Factories are employing fewer workers and transferring less knowledge, weakening the mechanism that made export-led industrialization transformative.
Countries early in the ladder often will not be able to afford modern facilities, and even when they can, those facilities won't generate the broad-based knowledge transfers that made manufacturing historically transformative.
Third, it could eliminate cognitive services as the most generalizable alternative modern development route.
The obvious irony of this latest wave of AI automation is that cognitive services are being automated faster and more completely than manual labor – precisely when many developing countries were betting on digital services as their primary development pathway. Many contemporary developing countries, especially those in Sub-Saharan Africa, have explicitly positioned themselves around digital services, fintech, and IT rather than attempting to replicate East Asia's manufacturing-led model.
Kenya is an excellent example. It is launching a national business process outsourcing (BPO) policy seeking to bring in a million jobs over five years, and has been pouring resources into call centers, content moderation, and data labelling. Rwanda has explicit intentions to become the “Singapore of Africa”. Nigeria has started to build a tech hub in Lagos home to multiple unicorns.
This was a rational response for the 21st-century: why invest decades in infrastructure when you can leverage educated populations, mobile connectivity, and English proficiency? Digital services offer 2-4x higher wages than manufacturing, better working conditions, and lower infrastructure requirements. Many modern countries have plans to potentially leapfrog directly from agriculture to knowledge work, or to pursue diversification strategies to bypass manufacturing.
Unfortunately, this strategy appeared most promising before 2023. Today, those countries are beginning to discover that the digital services pathway could be closing even faster than manufacturing.
Manufacturing automation still may operate on multi-decade timelines due to capital costs and the marginal cost of robotics. On the other hand, many forms of cognitive service automation could operate on 5-10 year timelines because of near-zero marginal costs – deployment costs essentially nothing beyond API calls.
Most call centers are likely facing complete automation within the next decade. Many tech leaders are predicting the advent of drop-in remote workers by 2027. If these predictions hold true, digital export strategies for developing countries will be devastated.
Developed countries – and particularly the US and China – are already capturing the vast majority of data center development, due to strengths in capital deployment, energy, and regulation. This would leave little to no financial incentives to offshore AI cognitive labor to developing countries.
Source: Microsoft AI Diffusion Report
As a result, countries betting on digital services could have even less time to establish themselves and climb upward than countries pursuing manufacturing: a total upheaval of decades of economic planning.
Note that the claims above don't predict massive reshoring of existing manufacturing from Asia back to wealthy countries. Sunk costs, existing supplier networks, and cheaper inputs will likely protect incumbent industries. China can now automate garment manufacturing just as effectively as the US can. Countries already industrializing might face headwinds, but not catastrophe.
The crisis is for countries that are still on the lowest rungs. Most of Sub-Saharan Africa, parts of South Asia, and other late developers face a closing door. They can't afford the capital-intensive entry requirements modern manufacturing demands. They can't absorb enough workers into digital services before automation re-shores those jobs to wealthy countries. They may lack consistently viable economic pathways that will work at the scale needed.
Previous industrial transitions created opportunities for developing countries to move up the ladder even as wealthy countries dominated the top rungs. AI could erode the bottom and middle rungs before developing countries can climb them – and we don’t have proven alternative strategies.
Thanks to Yolanda Laanquist, Danny Buerkli, Anna Yelizarova, and Ankit Mishra for their excellent feedback on this article.
