IESET.
Hypotheses·welfare architecture·welfare_transfer_south_africa_social_grants_long_run

South Africa's social-grants system (Old Age Pension extension 1998, Child Support Grant 1998-2012 expansion to age 18, Disability and Foster Care Grants, Social Relief of Distress 2020-2024) reaching 18+ million recipients — roughly 30% of population — has produced sustained reductions in extreme- poverty headcount of 6 to 10 percentage points relative to a synthetic-control of upper-middle-income Sub-Saharan and LatAm peers without comparable grant-coverage scale, while exerting fiscal-pressure costs visible in National Treasury MTBPS structural-balance projections.

PARTIALengine/runs/welfare_transfer_south_africa_social_grants_long_run

PARTIAL - poverty fell 30.5pp and Gini fell 3.7pp, but fiscal-pressure and placebo gates are not locally testable

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. poverty fell 30.5pp and Gini fell 3.7pp, but fiscal-pressure and placebo gates are not locally testable

why it matters

This matters because welfare architecture claims should change belief only when they survive a pre-declared empirical test.

how the test works

It compares 6 country or place units from 1998 to 2024, using a synth did design, with fixed effects for country and year.

what was measured
What changed
  • Social grant coverage share population
What we checked
  • Extreme poverty headcount
  • Inequality coefficient
  • Social spending share 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/welfare_transfer_south_africa_social_grants_long_run
1007550250199820112024ZAFBWANAMBRAMEXCOL
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show extreme_poverty_headcount across 6 sampled countries over 1998–2024.
The shapes above are stylised — none of the lines are real data.
Placeholder for welfare_transfer_south_africa_social_grants_long_run. Published chart will be generated from engine/runs/welfare_transfer_south_africa_social_grants_long_run/chart_data.json.

Pre-registration

pre-registered
first-spec commit 098ce96 · 2026-04-30T12:57:33Z
run generated · 2026-05-18T19:36:02Z

South Africa's social-grants system (Old Age Pension extension 1998, Child Support Grant 1998-2012 expansion to age 18, Disability and Foster Care Grants, Social Relief of Distress 2020-2024) reaching 18+ million recipients — roughly 30% of population — has produced sustained reductions in extreme- poverty headcount of 6 to 10 percentage points relative to a synthetic-control of upper-middle-income Sub-Saharan and LatAm peers without comparable grant-coverage scale, while exerting fiscal-pressure costs visible in National Treasury MTBPS structural-balance projections.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Refuted if the synth-DiD ATT on extreme-poverty headcount at 10-year horizon is below minus-6pp (i.e., reduction less than 6pp), OR if Gini gap is positive (inequality fails to improve), OR if placebo permutation p-value exceeds 0.10.

formal test & threshold
test:      synth_did_10yr_extreme_poverty_gini_gaps
threshold: poverty_att <= -6pp AND gini_att < 0 AND placebo_p < 0.10

Method

Template
synth_did
Fixed effects
country, year
Clustering
country
Sample
6 countries · 1998 – 2024
Evidence type
associational

Synthetic-DiD on ZAF poverty/Gini outcomes 1998-2024 with BWA, NAM, BRA, MEX, COL donor pool. Pre-treatment fit 1995-1998. Multi-stage rollout (1998 OAP, 2012 CSG-18, 2020 SRD) as event-time cohorts.

Data

VariableSourceTransform
extreme_poverty_headcount
outcome
world_bank_wdi:SI.POV.DDAYtier 2
level_pct
gini_coefficient
outcome
world_bank_wdi:SI.POV.GINItier 2
level
social_spending_share_gdp
outcome
imf:GGXG_NGDPtier 2
level
social_grant_coverage_share_population
treatment
world_bank_wdi:SI.POV.DDAYtier 2
level
gdp_per_capita_real
control
world_bank_wdi:NY.GDP.PCAP.KDtier 2
log
unemployment_rate
control
ilostat:ILMS_unemploymenttier 2
level
hiv_prevalence_rate
control
who_gho:GHOtier 1
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card - welfare_transfer_south_africa_social_grants_long_run

Verdict: PARTIAL - poverty fell 30.5pp and Gini fell 3.7pp, but fiscal-pressure and placebo gates are not locally testable

Exact Local Benchmark

  • Extreme poverty fell 30.5 pp from 2000 to 2022 and 20.1 pp by 2010.
  • Gini fell 3.7 pp from 2000 to 2022.
  • Unemployment rose +10.5 pp, underscoring the non-transfer macro confound.

Caveats

  • The poverty/Gini direction is strong, but the run lacks grant-coverage intensity, fiscal projections, and donor placebo inference.

Sources

  • world_bank_wdi:SI.POV.DDAY -> extreme_poverty_headcount (data/vintages/world_bank_wdi/SI.POV.DDAY@2026-05-05T194939Z.parquet)
  • world_bank_wdi:SI.POV.GINI -> gini_coefficient (data/vintages/world_bank_wdi/SI.POV.GINI@2026-05-05T194923Z.parquet)
  • world_bank_wdi:SL.UEM.TOTL.ZS -> unemployment_rate (data/vintages/world_bank_wdi/SL.UEM.TOTL.ZS@2026-05-05T200521Z.parquet)

Generated by engine/runs/welfare_transfer_south_africa_social_grants_long_run/replication.py at 2026-05-18T19:36:02+00:00

Strongest opposing argument

Every hypothesis ships with its charitable opposing argument. The framework earns credibility by handling objections at their strongest, not weakest.

Authored framework. Read the transparency note.