## Figure 2.9. Bank Equity Returns and Household Debt

## Source details

**Canonical URL:** [Figure 2.9. Bank Equity Returns and Household Debt](https://www.imf.org/-/media/files/publications/gfsr/2017/october/chapter-2/pdf-data/figure2-9.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/gfsr/2017/october/chapter-2/pdf-data/figure2-9.pdf.md)
- [Structured JSON version](/-/media/files/publications/gfsr/2017/october/chapter-2/pdf-data/figure2-9.pdf.json)

---

### 1. Banking Sector Abnormal Returns (Regression coefficients)
- Regression setup:
  - Dependent variable: future bank equity risk-adjusted abnormal returns, one to five years ahead.
  - Independent variable: past three-year changes in the household debt-to-GDP ratio.
- Time horizons reported: One year ahead; Two years ahead; Three years ahead; Four years ahead; Five years ahead.
- Visual scale: vertical axis spans from –8 through 12.
- Statistical significance:
  - Solid bars indicate responses that are statistically significant using 95 percent confidence intervals.

### 2. Bank Equity Crash Risk (Marginal effects)
- Estimation reported:
  - Marginal effect of the change in the household debt ratio (normalized by the standard deviation) on the probability of equity crashes in the next one to five years.
- Definition of equity crash:
  - Bank equity crashes are defined as annual bank equity returns lower than one standard deviation below the mean, as in Cheng, Raina, and Xiong 2014; and Baron and Xiong 2017.
- Time horizons reported: One year ahead; Two years ahead; Three years ahead; Four years ahead; Five years ahead.
- Statistical significance:
  - Solid bars indicate responses that are statistically significant using 95 percent confidence intervals.

*Source: IMF staff calculations.*

---


_Source: https://www.imf.org/-/media/files/publications/gfsr/2017/october/chapter-2/pdf-data/figure2-9.pdf_