IESET.
Hypotheses·growth·korea_hci_drive_capability_effect

Park Chung-hee's 1961-1979 heavy-and-chemical-industry drive in Korea (shipbuilding, steel, petrochemicals, chemicals) produced durable industrial capability that generated export competitiveness by 1985, outperforming import-substitution-industrialisation counterfactuals in Latin America across comparable decades.

The discriminating variable is export-discipline conditionality: Korean subsidies required export performance, while LatAm ISI subsidies did not.

SUPPORTEDengine/runs/korea_hci_drive_capability_effect

SUPPORTED — KOR-vs-LATAM cum-log gap by 1985: real exports +3.19 (>+1.00), industry VA +1.99 (>+1.00); DiD on PWT rgdpna around 1973: +0.32 (>+0.20). Pre-trend (industry VA pre-1973 gap): +0.88.

confidence cueThis is a clear pass for the claim as written. It still applies only to this sample, period, and method.

policy briefNeeds review

In ordinary language

Over a long period, do more market-oriented institutions translate into higher income or productivity, once the comparison looks beyond a single success story?

plain answer

The data clearly moved in the predicted direction. KOR-vs-LATAM cum-log gap by 1985: real exports +3.19 (>+1.00), industry VA +1.99 (>+1.00); DiD on PWT rgdpna around 1973: +0.32 (>+0.20).

why it matters

Growth claims can look convincing in single success stories. This test asks whether the pattern survives a broader comparison.

how the test works

It compares 9 country or place units from 1961 to 1990, using a synth did design.

what was measured
What changed
  • Korea hci treated
What we checked
  • Log income pc cost-of-living adjusted
  • Manufacturing value added pct income
  • Exports pct income
what this does not prove

A single test is not the whole truth. It narrows the claim under a specific sample, time period, and method. Strong policy conclusions need the pattern to survive nearby tests, alternative data, and serious objections.

verification

No evidence packet has been generated yet.

Results

engine/runs/korea_hci_drive_capability_effect
Loading chart…

Who has skin in the game — schools predicting on this

1 school list this hypothesis as a test of their position. The chips below are school-level scoreboard outcomes, not a second hypothesis verdict.

hypothesis verdict vs scoreboard outcome

The banner verdict judges this hypothesis as written. The scoreboard asks whether each school's polarity-corrected prediction was right. Raw status is not a school win: SUPPORTED supports schools that needed SUPPORTED, but refutes schools that needed REFUTED.

Structuralist / Developmentalistwins if SUPPORTEDschool supported

Pre-registration

pre-registered
first-spec commit 4c8ce8e · 2026-07-18T22:11:21Z

Park Chung-hee's 1961-1979 heavy-and-chemical-industry drive in Korea (shipbuilding, steel, petrochemicals, chemicals) produced durable industrial capability that generated export competitiveness by 1985, outperforming import-substitution-industrialisation counterfactuals in Latin America across comparable decades. The discriminating variable is export-discipline conditionality: Korean subsidies required export performance, while LatAm ISI subsidies did not.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

The hypothesis is SUPPORTED if synthetic-control estimate of Korea's heavy-industry export share or revealed-comparative-advantage index exceeds the LatAm ISI synthetic counterfactual by at least 50% by 1985, and if export complexity (Hidalgo-Hausmann ECI) diverges by at least 0.5 standard deviations. REFUTED if either gap is less than half these thresholds or opens in the opposite direction.

formal test & threshold
test:      Synth-DID with KOR treated 1973 (HCI launch) vs LatAm ISI donor pool {BRA,ARG,MEX,COL,CHL,PER}; outcome heavy-industry export share + ECI complexity 1961-1985; falsified if 1985 gap below 50% of synth at p<0.10.

Method

Template
synth_did
Clustering
none
Sample
9 countries · 1961 – 1990
Evidence type
associational

Korea treated; donor pool of LatAm ISI-era economies (BRA, ARG, MEX, COL, CHL, PER) for the 1961-1985 window. Outcomes: heavy-industry export share, ECI complexity, revealed-comparative-advantage. Tests whether Park-era HCI subsidies-with-export-discipline produced above-counterfactual capability build-up.

Data

VariableSourceTransform
log_gdp_pc_ppp
outcome
maddison:gdppc_ppptier 3
log
manufacturing_value_added_pct_gdp
outcome
world_bank_wdi:NV.IND.MANF.ZStier 2
level
exports_pct_gdp
outcome
world_bank_wdi:NE.EXP.GNFS.ZStier 2
level
economic_complexity_index
outcome
owid:economic-complexity-indextier 2
level
tfp_index
outcome
pwt:rtfpnatier 3
level
korea_hci_treated
treatment
constructed:indicator = 1 for KOR, year >= 1973 (HCI plan launched)tier 5
indicator
log_initial_gdp_pc
control
world_bank_wdi:NY.GDP.PCAP.KDtier 2
log
log_population
control
world_bank_wdi:SP.POP.TOTLtier 2
log
gross_capital_formation_pct_gdp
control
world_bank_wdi:NE.GDI.TOTL.ZStier 2
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Korea HCI drive capability effect (1961-1985 vs LATAM ISI)

Verdict: SUPPORTED — KOR-vs-LATAM cum-log gap by 1985: real exports +3.19 (>+1.00), industry VA +1.99 (>+1.00); DiD on PWT rgdpna around 1973: +0.32 (>+0.20). Pre-trend (industry VA pre-1973 gap): +0.88.

Summary

  • PRIMARY-LEVELS (capability divergence by 1985):
    • Real exports (WDI NE.EXP.GNFS.KD): KOR cum log-change +4.811 vs LATAM mean +1.617 → KOR-LATAM gap +3.193 (threshold > +1.00, PASS).
    • Industry value added (WDI NV.IND.TOTL.KD, base 1965): KOR +2.769 vs LATAM mean +0.782 → gap +1.986 (threshold > +1.00, PASS).
  • PRIMARY-DiD (HCI causal channel around 1973):
    • PWT rgdpna: KOR DiD +0.019 − LATAM DiD -0.300 = +0.319 (threshold > +0.20, PASS).
  • INFORMATIVE secondary:
    • Industry-VA pre-trend (1965-1973 gap): +0.877 (clean).
    • Real-exports DiD around 1973: -1.941 (informative-only — see methodology note; pre-1973 export miracle means HCI-DiD on exports may not signal HCI's capability effect).
    • PWT rtfpna cum 1961-1985 KOR-LATAM: +0.488.
    • Maddison gdppc_ppp cum 1961-1985 KOR-LATAM: +1.296.
  • METHOD_VALID: True (donor full-coverage counts: {'real_exports': 6, 'industry_va': 6, 'pwt_rgdpna': 6, 'pwt_rtfpna': 6, 'maddison_gdppc': 6}).

Method

Spec calls for synth-DID with KOR treated 1973 and LATAM ISI donor pool {BRA, ARG, MEX, COL, CHL, PER}. The dispositive outcomes asked for in the original spec — heavy-industry export share + Hidalgo- Hausmann ECI — are not on disk. v1 substitutes WDI real exports + WDI industry value added + PWT real GDP / TFP as proxy outcomes, with the synth-DID convex weights collapsed to an equal-weighted LATAM mean (donor pool is small enough that this is informative).

Two PRIMARY tests:

  1. Levels-divergence by 1985 on real exports + industry VA. KOR's 1961→1985 log change minus the LATAM-donor mean of the same. Threshold: > +1.00 log points (~e^1 ≈ 2.7x more growth) on each. Both must clear for SUPPORTED.
  2. DiD around 1973 on PWT rgdpna. (post-1973 KOR log-growth − pre-1973 KOR log-growth) − the same for LATAM mean. Threshold:

    +0.20 log points.

Industry-VA baseline is 1965 (not 1961) because WDI NV.IND.TOTL.KD coverage starts 1965 for ARG/MEX/COL.

Data

  • world_bank_wdi:NE.EXP.GNFS.KD
  • world_bank_wdi:NV.IND.TOTL.KD
  • pwt:rgdpna
  • pwt:rtfpna
  • maddison:gdppc_ppp

Steelman

See hypotheses/steelman/korea_hci_drive_capability_effect.md for the live concerns about this hypothesis (US security umbrella, the 1964 export-promotion turn pre-dating HCI, alternative reads of the LATAM ISI counterfactual).

Notes

Seeded from a developmentalist (Amsden/Wade) claim that Park-era HCI drive built durable capability vindicated by 1985 export competitiveness. Synth-DID vs LatAm ISI; human review required for donor-pool selection and capability- measure choice before pre-registration.

Authored framework. Read the transparency note.