How Industrial Policy Saved South Korea in the 70s & 80s
South Korea's targeted heavy industry intervention from 1973-1979 doubled output in selected sectors and created lasting competitive advantages that persisted decades after the policy ended
Nathan Lane's study "Manufacturing Revolutions: Industrial Policy and Industrialization in South Korea" in The Quarterly Journal of Economics provides the first rigorous empirical evidence that purposeful industrial policy can successfully transform an economy's manufacturing base. Using modern econometric techniques on newly digitized data spanning two decades, Lane's findings challenge conventional economic wisdom that governments can't effectively pick winners — demonstrating that temporary interventions can shift entire industries toward advanced markets when implemented with clear objectives and strong state capacity.
By the numbers:
Direct industrial effects:
128% increase in output for targeted vs non-targeted industries (1973-1986)
109% rise in real intermediate outlays in HCI sectors
81.4% jump in total investment for treated industries
63.8% employment growth in targeted sectors
17.2% boost in labor productivity (value added per worker)
9.55% relative price reduction in HCI products
29.7% increase in number of plants in HCI markets
4.3% higher total factor productivity in HCI plants during 1980s
Export transformation:
13% greater increase in revealed comparative advantage vs non-HCI exports
10.6 percentage point rise in probability of achieving comparative advantage
4.92% increase in share of manufacturing exports
50% of total exports from HCI by 1983 (surpassing 1980 target)
11-13% higher probability of export success vs peer countries
Policy instruments:
5 percentage points - interest rate discount on policy loans
50% of all domestic credit through subsidized channels
100% import duty exemptions for HCI producers
80% average tariff exemption rate for "key industries"
6 sectors targeted: steel, nonferrous metals, shipbuilding, machinery, electronics, petrochemicals
Learning and spillover effects:
5.3% unit cost reduction per doubling of cumulative output in HCI
4.4% output increase for every 1% rise in forward linkages to HCI
0.46% price decrease for every 1% rise in forward linkages
0% significant effect through backward linkages to suppliers
Note: Forward linkages measure how much an industry supplies to others as inputs; backward linkages measure how much an industry purchases from suppliers
Study scope:
278 five-digit industries analyzed
1967-1986 study period with 1973 intervention
320 input-output sectors for linkage analysis
91,094 plants in post-1979 microdata
4 major industrial classification revisions harmonized
The backstory
Nixon's 1969 announcement of U.S. military withdrawal from Asia triggered an existential crisis for South Korea. The country faced an industrially superior North Korea without the promise of American protection. North Korea had pursued aggressive military industrialization throughout the 1960s, while South Korea lacked even a basic domestic arms industry. Without U.S. troops, military planners calculated South Korean forces couldn't withstand a North Korean invasion.
President Park Chung-hee responded with the Heavy and Chemical Industry (HCI) drive in January 1973, targeting six strategic sectors: steel, nonferrous metals, shipbuilding, machinery, electronics, and petrochemicals. The choice wasn't random. Korean planners recognized they couldn't leap directly to advanced weapons manufacturing, but they could produce the quality inputs that weapons required. Steel and metals would supply crucial defense components, electronics would enable modern weaponry systems, and machinery would support precision military production.
Foreign skeptics thought Korea was delusional. The World Bank blocked Korea's proposed integrated steel mill in 1969, declaring the country had "no comparative advantage in the production of steel." The IMF and USAID rejected financing for proto-HCI schemes throughout the early 1970s. The U.S. Export-Import Bank joined European lenders in refusing support for heavy industry ventures. These rejections continued right up until Park's autocratic self-coup in late 1972, which finally gave him the political power to push through the controversial program.
How the intervention worked
The government's primary weapon was directed credit, channeling massive resources through both state development banks and commercial lenders. Half of all domestic credit in Korea consisted of subsidized "policy loans" that carried interest rates 5 percentage points below market rates. The Korea Development Bank led the charge, with lending to machinery and intermediate sectors exploding after 1973. Commercial banks were essentially forced to participate, with policy loans automatically rediscounted by the central bank at preferential rates.
Tax policy reinforced this credit bias. Effective marginal tax rates for heavy and chemical industries dropped dramatically after 1973, when new laws gave strategic industries a menu of incentives: five-year tax holidays, 8% investment tax credits, or 100% special depreciation allowances. The divergence was stark and lasted until fiscal reforms in the early 1980s finally closed the gap between strategic and non-strategic sectors.
The trade policy surprise: Despite claims that Korea's intervention was protectionist, the data reveals something different. HCI producers received exemptions from up to 100% of import duties, and average output protection was actually lower for targeted industries than for non-targeted ones during the policy period. The focus remained firmly on export competitiveness rather than import substitution. Input tariffs faced by heavy industry fell significantly, as planners understood these sectors needed access to foreign technology and materials.
What actually happened: The results were massive and persistent. Real output in targeted industries increased 128% relative to non-targeted sectors. Labor productivity jumped 17.2%, while employment in HCI sectors rose 63.8%. Perhaps most surprisingly, output prices in targeted industries fell 9.55% relative to other sectors — the opposite of what critics predicted would happen under protection.
The manufacturing landscape shifted dramatically. The share of total manufacturing output and employment moved decisively toward heavy and chemical industries, and this reallocation proved durable. The number of plants operating in HCI markets increased nearly 30%, indicating genuine industrial expansion rather than mere capacity stretching. Investment surged 81.4%, while intermediate outlays — the materials and components needed for production — jumped 109% in real terms.
Learning effects proved crucial: The data reveals powerful learning-by-doing forces in targeted industries. Each doubling of cumulative output reduced unit costs by 5.3% in HCI sectors — significantly stronger than the learning effects in non-targeted industries. Plant-level data from the 1980s shows HCI establishments maintained 4.3% higher total factor productivity (TFP) — a measure of efficiency that accounts for all inputs — even after policies ended. These weren't just individual plant improvements; industry-wide spillovers amplified the learning effects, suggesting knowledge diffused across firms within targeted sectors.
The productivity gains took time to fully materialize. Industry-level total factor productivity differences only became statistically significant after 1979, and the upward trajectory continued through the mid-1980s. This delayed emergence aligns with infant industry theory, which predicts that productivity improvements require sustained production experience.
Export transformation followed a different timeline: While output and employment effects appeared quickly, export competitiveness emerged more gradually. During the early years (1973-1976), gains in revealed comparative advantage were modest. The middle period (1977-1979) saw acceleration in export share growth, but the full realization of export competitiveness came in the post-policy period (1980-1986).
By the numbers, targeted industries saw their revealed comparative advantage — a measure of how much a country specializes in exporting specific products relative to the world average — increase 13% more than other manufacturing exports. The probability of achieving comparative advantage (RCA > 1, meaning Korea exported proportionally more of that product than the global average) jumped by 10.6 percentage points. Korea's original target — having heavy and chemical products constitute 50% of exports by 1980 — was actually surpassed by 1983. Cross-country comparisons using triple-difference estimates confirm these patterns weren't global trends; Korean HCI sectors dramatically outperformed the same industries in peer countries.
Downstream industries reaped unexpected rewards: The policy's effects rippled through Korea's production network — the web of connections showing which industries buy from and sell to each other. Industries that relied heavily on inputs from targeted sectors (forward linkages from HCI's perspective) experienced significant benefits. For every 1% increase in an industry's dependence on HCI inputs, downstream sectors saw 4.4% higher output and 0.46% lower output prices. These weren't temporary gains — the benefits persisted and even strengthened after 1979.
The mechanism was straightforward: as HCI sectors expanded and became more productive, they supplied cheaper, higher-quality intermediate inputs (materials and components used to make other products) to downstream users. Material outlays expanded differentially for heavy users of HCI products, and these downstream industries eventually developed their own export advantages. Remarkably, downstream export competitiveness mainly emerged after the policy ended, suggesting these spillovers took time to translate into international competitiveness.
Upstream effects disappointed: In contrast to the strong downstream benefits, spillovers to upstream suppliers were limited and statistically insignificant. Industries selling inputs to HCI sectors (backward linkages from HCI's perspective, like steel suppliers selling to machinery makers) didn't experience differential growth in output, employment, or productivity. This asymmetry likely reflects the deliberate choice to target relatively upstream industries — there simply weren't many domestic suppliers further up the value chain to benefit from increased HCI demand.
The crowding-out that wasn't: Critics often assume industrial policy must harm non-favored sectors, but Korea's experience suggests otherwise. Investment in non-targeted industries continued growing throughout the HCI period. Commercial banks maintained substantial lending to light industry, and Japanese lenders filled financing gaps for non-HCI sectors. Capital-intensive industries outside the HCI umbrella showed no decline in investment rates. The absolute growth of non-targeted sectors continued; they simply grew more slowly than their HCI counterparts.
Political economy mattered: The policy's success hinged on unusual political circumstances. The existential security threat from North Korea focused policy on coherent objectives rather than political favoritism. Park's autocratic control from 1972-1979 ensured consistent implementation without legislative interference. His assassination in October 1979 triggered immediate liberalization — special loan rates were eliminated, tax advantages phased out, and trade policy shifted toward broader liberalization.
This abrupt ending actually strengthens the empirical findings. The policy's termination was driven by political assassination rather than economic factors, providing a clean break for analysis. Moreover, the persistence of benefits after 1979 demonstrates that temporary interventions can create lasting structural change.
About the Study
The study employs difference-in-differences design to establish causality. The study comparing how targeted and non-targeted industries evolved differently after 1973, along with analyzing 278 five-digit industries over two decades, with pre-trends from 1967-1972 statistically indistinguishable from zero. Double-robust estimators that combine propensity score weighting with regression adjustment confirm the magnitude of effects. Cross-country triple-difference estimates (comparing Korea's industry differences to the same differences in other countries) validate that Korean HCI sectors performed exceptionally compared to global peers.
The data construction required extensive work. The research digitized and harmonized Korea's Mining and Manufacturing Surveys across four major industrial classification revisions, hand-matched legislative acts to industry codes, and combined this with 1970 input-output tables for network analysis. This creates one of the most comprehensive databases of industrial development during a major policy intervention.
Bottomline
The World Bank declared in 1969 that Korea had "no comparative advantage in steel production" and blocked their proposed steel mill. The IMF rejected HCI financing. USAID turned them down. European lenders refused support. Every major international institution advised against pursuing heavy industry.
Korea implemented the policy anyway, driven by the existential threat from North Korea. Nixon's 1969 announcement of U.S. military withdrawal from Asia left Korea facing an industrially superior North Korea without American protection. Korean military planners calculated they couldn't withstand a North Korean invasion without domestic defense industrial capabilities.
The results documented in this study followed: 128% output growth in targeted sectors, 17.2% productivity gains, and transformation into a major exporter of heavy and chemical products within a decade. Without this intervention, Korea would have remained focused on light manufacturing exports. The chaebols that became global powerhouses - Samsung, Hyundai, LG - built their capabilities through the heavy industry drive.
China under Deng Xiaoping followed a similar path, learning from the heavy handed mistakes of Mao and the successes of Japan and South Korea, implementing targeted industrial policies that transformed China from an agricultural economy into the world's manufacturing center. Like Korea, China ignored conventional wisdom about comparative advantage and successfully built industries through strategic state intervention.
The study demonstrates that strategic industrial policy can create comparative advantage when implemented correctly, which isn’t easy by any means. Korea succeeded by targeting sectors with strong learning effects and forward linkages, providing temporary rather than permanent support, and maintaining focus on export competitiveness. The policy addressed specific market failures in credit allocation and allowed infant industries to mature.
The evidence shows that countries need not accept their current position in the global economy. With clear objectives, strong state capacity, and appropriate sector selection, industrial policy can reshape a nation's manufacturing base and development trajectory. Both Korea and China prove this transformation works in practice.