IESET.
Hypotheses·growth·ethiopia_developmental_state_growth_effect

Pre-2010 Ethiopian developmental-state strategy (state-led infrastructure, industrial parks, export-credit allocation) produced the fastest sustained African growth rates, consistent with late-developer industrial-policy theory.

PARTIALengine/runs/ethiopia_developmental_state_growth_effect

partial — Ethiopia grew faster than the peer median (+20.6 log-points) and ranked #3/12, but did not meet both primary thresholds (rank #1 AND gap >= 20 log-points). Synthetic counterfactual gap at 2010 = -7.3 log-points.

confidence cueThe result is useful, but not decisive. Treat it as a clue, not a settled conclusion.

policy briefMixed or noisy

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 evidence is suggestive but not decisive. Ethiopia grew faster than the peer median (+20.6 log-points) and ranked #3/12, but did not meet both primary thresholds (rank #1 AND gap >= 20 log-points).

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 12 country or place units from 1991 to 2019, using a synth did design.

what was measured
What changed
  • Ethiopia treated
What we checked
  • Log income pc cost-of-living adjusted
  • Gross capital formation 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/ethiopia_developmental_state_growth_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 SUPPORTEDpartial/no-call

Pre-registration

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

Pre-2010 Ethiopian developmental-state strategy (state-led infrastructure, industrial parks, export-credit allocation) produced the fastest sustained African growth rates, consistent with late-developer industrial-policy theory.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

PRIMARY (dispositive): the hypothesis is SUPPORTED if BOTH (P1) Ethiopia ranks #1 in cumulative log-GDP-per-capita (PPP) growth 1995-2010 within the SSA peer pool {KEN, TZA, UGA, RWA, SEN, GHA, MOZ, MWI, ZMB, MDG, BDI} AND (P2) Ethiopia's gap to the peer-pool median cumulative log-growth over 1995-2010 is at least +0.20 log-points (~22% extra per-capita PPP growth versus the median peer over the 15-year window). PARTIAL if Ethiopia is in the top-3 OR its gap to median is positive but not both primaries hold. REFUTED if Ethiopia is outside the top-3 OR its gap to peer median is negative. INFORMATIVE: synthetic-control-lite (Dirichlet sample search on pre-treatment 1991-1994 levels of log GDPpc PPP) counterfactual gap at 2010 is reported but does not gate the verdict. METHOD_VALID: WDI NY.GDP.PCAP.PP.KD must be available 1991-2010 for ETH and at least 9 of 11 named peers. METHOD_VALID failure -> inconclusive.

formal test & threshold
test:      ssa_peer_panel_cumulative_log_growth_rank_and_gap
threshold: PRIMARY: rank_eth(1995-2010 cum log GDPpc PPP growth) == 1 AND eth_cum_log_growth - peer_median_cum_log_growth >= 0.20

Method

Template
synth_did
Clustering
none
Sample
12 countries · 1991 – 2019
Evidence type
associational

Ethiopia treated; donor pool of comparable Sub-Saharan African economies (KEN, TZA, UGA, RWA, SEN, GHA, MOZ). Outcome: real GDP per capita 1991- 2010. Tests whether Ethiopia's developmental-state strategy produced above-counterfactual growth.

Data

VariableSourceTransform
log_gdp_pc_ppp
outcome
world_bank_wdi:NY.GDP.PCAP.PP.KDtier 2
log
gross_capital_formation_pct_gdp
outcome
world_bank_wdi:NE.GDI.TOTL.ZStier 2
level
ethiopia_treated
treatment
constructed:indicator = 1 for ETH, year >= 1995 (post-EPRDF developmental-state strategy)tier 5
indicator
log_population
control
world_bank_wdi:SP.POP.TOTLtier 2
log
trade_openness
control
world_bank_wdi:NE.TRD.GNFS.ZStier 2
level
rule_of_law
control
wgi:RL.ESTtier 4
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Ethiopia developmental-state growth effect (1995-2010)

Verdict: partial — Ethiopia grew faster than the peer median (+20.6 log-points) and ranked #3/12, but did not meet both primary thresholds (rank #1 AND gap >= 20 log-points). Synthetic counterfactual gap at 2010 = -7.3 log-points.

Summary

  • Ethiopia cumulative log-GDP-per-capita (PPP) growth 1995-2010: +59.7 log-points (~82% in levels).
  • SSA peer-pool median cumulative growth: +39.1 log-points; Ethiopia gap to median: +20.6 (threshold 20).
  • Ethiopia rank: #3 of 12 (threshold #1).
  • Synthetic-control-lite counterfactual gap at 2010: -7.3 log-points.

Ranking

  1. MOZ: +84.6 log-points
  2. RWA: +64.7 log-points
  3. ETH: +59.7 log-points
  4. UGA: +55.5 log-points
  5. TZA: +45.4 log-points
  6. ZMB: +43.2 log-points
  7. GHA: +39.1 log-points
  8. MWI: +27.7 log-points
  9. SEN: +18.3 log-points
  10. KEN: +7.9 log-points
  11. MDG: -1.6 log-points
  12. BDI: -17.8 log-points

Method

Two dispositive primary tests, then synthetic-control-lite as an informative robustness check:

  1. Cumulative log-difference of WDI NY.GDP.PCAP.PP.KD from 1995 to 2010, computed for Ethiopia and each peer; Ethiopia must rank #1 (P1) AND its gap to the peer median must be at least 0.20 log-points (P2). The 0.20 threshold corresponds to roughly 22% extra per-capita PPP growth over 15 years versus the median peer — the magnitude a fair reader of "fastest sustained African growth" would expect to see for the Ethiopian developmental state to count as the binding cause.
  2. Synthetic-control-lite using a Dirichlet sample search to weight the 11 SSA donor countries on pre-treatment (1991-1994) levels of log GDP per capita PPP. Counterfactual log GDPpc at 2010 reported.

Donor weights are not meant to compete with Abadie-Diamond-Hainmueller's quadratic programming SC; the panel is small and the donor pool is the spec's pre-specified peer pool, so the synth gap is reported as informative, not dispositive.

Mechanism diagnostics (gross capital formation share and trade openness, both as % of GDP) are reported as 1995->2010 changes for context but do not enter the verdict.

Data

  • world_bank_wdi:NY.GDP.PCAP.PP.KD (primary outcome)
  • world_bank_wdi:NE.GDI.TOTL.ZS (gross capital formation, % of GDP)
  • world_bank_wdi:NE.TRD.GNFS.ZS (trade, % of GDP)

Notes

Seeded from a developmentalist position-claim that Ethiopia's pre-2010 EPRDF developmental-state strategy yielded the fastest sustained African growth. Synthetic control vs SSA peer pool; human review required on donor-pool selection and aid-flow confounders.

Authored framework. Read the transparency note.