How do I evaluate earnings persistence for forecasting RNOA?

Question

AcadiFi · Accepted Answer

Earnings persistence measures how much of current period performance carries forward, essential for forecasting and valuation. High ROE/RNOA is mean-reverting unless supported by durable competitive advantages.

Orchard Wireless analysis:
- Current RNOA 24%, 5-year avg 19.6%, industry 12%
- Premium over industry: 7.6 percentage points (avg) or 12 points (current)
- Volatility: std dev ~3.5% → economic exposure, not stable moat

Persistence drivers (support high RNOA):
1. Patents/IP (wireless standards, 5G/6G portfolio)
2. Network effects / switching costs
3. Regulatory barriers (spectrum licenses)
4. Scale economies in capex-heavy industry
5. Brand loyalty (consumer vs enterprise)

Mean reversion pressures:
1. New entrants attracted by high returns
2. Technology shifts (unlicensed spectrum, satellite alternatives)
3. Regulatory rate cuts / price caps
4. Customer consolidation reducing pricing power

```mermaid
flowchart TD
    A[RNOA 24%] --> B[Sustainable Portion]
    A --> C[Transitory Portion]
    B --> D[Moats: IP, Scale]
    C --> E[Competitive Erosion]
    D --> F[Terminal 14-16%]
    E --> F
```

Quantitative persistence test:
- Regress RNOA_t+1 on RNOA_t: slope near 1 = high persistence; near 0 = full reversion
- Industry studies find RNOA persistence ~0.6-0.7 year-to-year
- Abnormal returns fade toward industry median in 5-10 years

Forecast framework for Orchard:
- Year 1-2: 20-22% (some persistence, near recent levels)
- Year 3-5: gradual fade 18% → 15%
- Years 6-10: fade to 12-14% (industry + moat premium)
- Terminal: 12% (industry level) or 14% if moat sustainable

Valuation impact: every 1pt of persistent RNOA premium adds significant terminal value. A fade assumption from 22% to 14% over 10 years vs 22% to 12% represents millions in valuation difference. Test sensitivity explicitly in the model.

Red flag: if management asserts permanent RNOA premium without identifying specific moats, apply faster fade and haircut valuations. Persistence assumptions are where analysts most often overpay.

How do I evaluate earnings persistence for forecasting RNOA?

Master Level II with our CFA Course

Related Questions

Practice Questions