What are the main sources of tracking error in smart beta ETFs, and how should they differ from traditional market-cap-weighted index ETFs?

Question

AcadiFi · Accepted Answer

Smart beta ETFs systematically deviate from market-cap weighting by following alternative index construction rules (value, momentum, low volatility, etc.). This inherent deviation generates structural tracking error relative to cap-weighted benchmarks, but additional implementation-driven tracking error can indicate problems.

**Sources of Tracking Error**

**Worked Example — Pinnacle Low Volatility ETF:**

The Pinnacle Low Vol ETF tracks the Sentinel Low Volatility 200 Index. Over the past 12 months, the ETF returned 9.2% versus the index return of 9.8% and the S&P 500 return of 14.1%.

| Tracking Error Source | vs. Underlying Index | vs. S&P 500 |
|---|---|---|
| Strategy deviation (intended) | 0% | 480 bps (by design) |
| Rebalancing timing | 15 bps | 15 bps |
| Transaction costs | 22 bps | 22 bps |
| Cash drag | 8 bps | 8 bps |
| Sampling (not full replication) | 12 bps | 12 bps |
| Securities lending offset | -5 bps | -5 bps |
| **Total implementation TE** | **52 bps** | — |
| **Total TE vs. broad market** | — | **532 bps** |

**Structural vs. Implementation Tracking Error:**

1. **Structural (intended)** — the deviation from cap-weighted returns that IS the product's purpose. A momentum ETF should have high tracking error versus the S&P 500 — that is the active bet investors are paying for.

2. **Implementation (unintended)** — the deviation from the ETF's own stated index. This is what investors should minimize. Sources include:
   - *Rebalancing costs* — smart beta indices rebalance more frequently (monthly or quarterly) versus annual for most cap-weighted indices
   - *Turnover* — factor-tilted indices generate 30-80% annual turnover versus 3-5% for cap-weighted indices
   - *Liquidity impact* — equal-weight or fundamental-weight schemes overweight smaller stocks with wider bid-ask spreads
   - *Cash drag* — higher inflows/outflows require holding more cash than cap-weighted funds
   - *Reconstitution front-running* — predictable rebalancing dates allow arbitrageurs to trade ahead

**Analyst Evaluation Framework:**

- Compare the ETF's tracking error against its own index, not a cap-weighted benchmark
- Acceptable implementation tracking error for smart beta: 30-80 bps annually
- Traditional cap-weighted ETF implementation tracking error: 5-15 bps annually
- Higher is acceptable only if offset by outperformance (positive tracking difference)

See our CFA Level II Equity module for ETF evaluation frameworks.

What are the main sources of tracking error in smart beta ETFs, and how should they differ from traditional market-cap-weighted index ETFs?

Master Level II with our CFA Course

Related Questions

Practice Questions