financial-metrics
ROE is net profit divided by book equity. It is the residual accounting return to shareholders and rises automatically with leverage — which is why it is a poor standalone quality metric in M&A.
ROE = Net profit / Book equity
Average equity in the denominator is cleaner than year-end equity after a recap. Damodaran’s return-measures paper is explicit: ROE is not a measure of operating quality once leverage and non-operating items enter.
DuPont decomposition (teaching form):
ROE = Net margin × Asset turnover × Equity multiplier
That is net profit margin × ROA ingredients × leverage.
Illustration only. Same operations, different gearing.
| Conservative | Levered | |
|---|---|---|
| Net profit | €5.0m | €4.2m (after more interest) |
| Book equity | €40m | €18m |
| ROE | 12.5% | 23.3% |
| ROIC | 11% | 11% |
The levered company looks like a better equity story on ROE and a worse credit story on debt-to-EBITDA. An LBO is, among other things, a machine for raising ROE if it works — and destroying equity if interest cover fails.
High ROE with high ROIC and modest leverage is genuine. High ROE with low ROIC is financial engineering or a shrinking equity base (buybacks, losses already written off). Negative equity makes ROE meaningless.
Banking (regulatory capital is the point); owner-managed statutory analysis; screening — then immediately open ROIC.
Book equity is an accounting residual, not market equity value. IFRS OCI items, hybrids, and NCI complicate the denominator. One-off gains juice ROE.
Dazur shows ROE only beside ROIC and leverage. A teaser that leads with ROE in a geared mid-market file is usually selling the recap, not the customers.