Pre-registration
The ECB's June-2014 introduction of a negative deposit-facility rate (initially -0.10%, cut to -0.50% by September-2019) and the parallel Swiss / Danish / Swedish / Japanese negative-rate experiments lowered short-end money-market rates and core sovereign-bond yields below zero, but did NOT cause an aggregate decline in eurozone bank deposits or measurable deposit-flight from the banking sector. The hypothesis is that the Eonia / ESTR rate followed the deposit-facility rate downward to within 10 bp through the negative-rate window, while aggregate eurozone household deposits continued to grow at a rate within 2 pp of the pre-2014 trend.
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
SUPPORTED if (i) Eonia/ESTR tracked DFR within 10 bp on average through 2014-2022, (ii) at least one of {DE 2y, NL 2y, FR 2y, FI 2y} sovereign yield went below zero for at least 24 cumulative months, AND (iii) aggregate eurozone household deposits grew within 2 pp/year of the 2010-2013 trend during 2014-2022. PARTIAL if 2 of 3 conditions hold. REFUTED if any of (i) or (iii) fails by a large margin (e.g. deposit run-down > 5% annualised) or if no core sovereign yield went negative.
formal test & threshold
test: descriptive_passthrough_and_deposit_stability threshold: Eonia/ESTR-DFR spread <= 10 bp avg AND core 2y < 0 for >= 24 months AND deposit growth within 2 pp of trend
Method
- Template
descriptive- Sample
- 15 countries · 2010 – 2022
- Evidence type
- descriptive
Descriptive time-series + simple panel: track Eonia/ESTR, 10y Bund/OAT yields, and aggregate eurozone deposits before, during, and after the negative-rate window. No causal claim against counterfactual; descriptive test of the "negative rates pass through to short end and core sovereigns without breaking the deposit base" proposition.
Data
| Variable | Source | Transform |
|---|---|---|
eonia_estr_short_end_rate outcome | ecb:EONtier 1 ecb:ESTRtier 1 | level_basis_points |
ten_year_sovereign_yield_core outcome | ecb:IRStier 1 fred:IRLTLT01DEM156Ntier 1 | level_basis_points |
eurozone_household_deposits_growth outcome | ecb:BSItier 1 | yoy_pct |
ecb_dfr_negative treatment | ecb:DFRtier 1 | level_basis_points |
cpi_inflation_eurozone control | ecb:ICPtier 1 eurostat:prc_hicp_manrtier 1 | yoy_pct |
real_gdp_growth_eurozone control | eurostat:nama_10_gdptier 1 | yoy_pct |
● ready · ● pending · ● reconstruct-needed
Detailed result card
Result card - financial_negative_rates_eurozone_2014_2022
Verdict: PARTIAL - money-market passthrough clears and no broad-money run appears, but M3 growth is not within 2pp of pre-2014 trend and exact household-deposit/core-yield gates are not loaded
Exact Local Test
- Average Euribor 3M minus DFR spread, Jun-2014 to Jul-2022: 9.37 bp.
- Euribor 3M was below zero in 86/98 months.
- M3 annual growth averaged 5.71% during the negative-rate window versus 1.79% in 2010-2013.
- Minimum M3 annual growth in the window: 1.71%.
Caveat
Eonia/ESTR, household deposits, and core 2y sovereign yields are not present locally; Euribor 3M and M3 annual growth are used as conservative local proxies.
Strongest opposing argument
Every hypothesis ships with its charitable opposing argument. The framework earns credibility by handling objections at their strongest, not weakest.