Why does the curriculum emphasize getting the 'long-run level of returns right' above precision?

Question

AcadiFi · Accepted Answer

This is one of the most important lessons in the entire CME curriculum, and it has a real-world cautionary tale behind it.

**Why Level > Precision:**

Consider a foundation that must distribute 5% of assets annually. If the true expected portfolio return is 6%, the foundation can barely cover distributions plus expenses and inflation. But if the foundation projects 10% returns (too high by 4%), it will:
- Underfund contributions ("we don't need to add capital — the portfolio will grow")
- Set unrealistic spending goals
- Take on inappropriate risk
- Eventually face a painful reckoning when returns don't materialize

Conversely, projecting too-low returns might lead to excessive conservatism — forgoing growth and underspending relative to the mission.

The point is: a projection that's off by 4% on the level is catastrophic for planning, even if the relative ranking of asset classes is correct.

**The Late-1990s Lesson:**

Before the tech bubble, it was standard practice to simply extrapolate historical returns. US equities had returned ~15% annually during the 1990s bull market. Many institutions projected double-digit portfolio returns indefinitely based on this extrapolation.

What happened:
- Pension funds calculated that their existing assets would cover obligations — so they reduced contributions
- Foundations increased spending rates based on inflated return projections
- Endowments took on excessive risk, believing high returns were the 'normal' state

When the tech bubble burst (2000-2002), portfolios lost 30-40%. The return projections that had driven planning for years proved wildly optimistic. Pension funds were suddenly massively underfunded. Foundations had overspent.

**The Post-Bubble Shift:**

Since then, most institutions have abandoned naive extrapolation in favor of forward-looking methods:
- **Building-blocks approach** (risk-free rate + risk premiums)
- **Grinold-Kroner model** (dividend yield + earnings growth + repricing)
- **Equilibrium models** (anchored to economic fundamentals)

These methods typically produce much lower return projections — often 6-8% for balanced portfolios vs. the 10-12% that extrapolation suggested. The lower projections are more realistic, even if imprecise.

**Consistency Over Precision:**

The curriculum also emphasizes that cross-sectional and intertemporal consistency matter more than precision for any single asset class:

- **Cross-sectional inconsistency** (e.g., projecting strong equity returns alongside falling bond yields) leads to portfolios that unknowingly embed contradictory bets
- **Intertemporal inconsistency** (e.g., short-term and long-term projections that don't connect through a plausible path) distorts the link between portfolio decisions and investment horizon

Consistent projections produce well-balanced portfolios even if every individual estimate is somewhat off. Inconsistent projections produce structurally flawed portfolios.

**Exam Application:** Be ready to identify when a vignette's CME process prioritizes precision over consistency, or when extrapolation is being used instead of forward-looking methods.

For more on CME challenges, explore our comprehensive article on the CME framework.

Why does the curriculum emphasize getting the 'long-run level of returns right' above precision?

Master Level III with our CFA Course

Related Questions

Related Articles

Practice Questions