Monetary tightening reduces income growth most at the top
BOC Paper

Monetary tightening reduces income growth most at the top

Monetary tightening compresses Canadian household income inequality by reducing higher earners’ income growth more, Bank of Canada research finds. Tax records from 1982–2019 show that labor income drives the persistent differences across income groups.

Higher earners absorb larger losses

Yaz Terajima of the Bank of Canada (BoC) and Carolyn A. Wilkins of Princeton University estimate that a 100-basis-point tightening shock lowers average market-income growth by 0.903 percentage points after one year.

The decline peaks at 2.207 percentage points in year four.

After one year, the highest income decile loses 0.741 percentage points more than the fifth decile; the lowest decile's difference is not statistically significant.

Wages, salaries and commissions drive the persistent gradient across income groups, while capital-gains responses fluctuate across horizons.

The authors conclude that tightening counters cyclical income inequality by restraining gains among high-income households.

Their research does not represent the BoC's official position.

Capital gains shape the cycle

Statistics Canada's Longitudinal Administrative Databank samples 20 percent of tax filers.

The authors use 1982–2019 records and panel local projections, studying market income before taxes and transfers under conventional monetary policy.

Income-growth skewness rises with GDP growth, driven by both low- and high-income households and especially capital gains; variance is not counter-cyclical.

An aggregate regression finds that tightening reduces skewness after one year, but not statistically significantly at later horizons.

The caveat is material

The authors note that the aggregate skewness result loses statistical significance when they add lagged controls.

That exercise has only 32–36 annual observations across horizons.

The tax-filer sample also limits the findings to intensive-margin income dynamics and excludes quantitative easing and tightening.

Report an error