Breadth Divergence at Index High
A large-cap equity index prints a new all-time high. The equal-weight version of the same index is 3.5% below its high; the advance-decline line made a lower high; the percentage of stocks above their 50-day moving averages has fallen from 78% to 54%; the 52-week new-high minus new-low reading has narrowed for five weeks. Small caps are underperforming large caps by 6% over three months. Credit spreads are near cycle tights and the VIX is 14. The manager's benchmark is the cap-weighted index, but the portfolio has a 12% overweight to small caps and a 5% overweight to industrials. The fund's tracking error is 4.5% and its risk budget is nearly fully used. The PM must decide whether the breadth divergence is a distribution signal or healthy rotation, and whether to cut the small-cap overweight before quarter-end.
1. Interpret the breadth divergence. Is it a distribution signal, healthy rotation, or insufficient evidence? Use the specific indicators and explain the limits of breadth analysis.
Read the worked answer
The divergence is a warning, not a standalone sell signal. The cap-weighted index at a new high while the equal-weight index is 3.5% below its high shows that leadership is narrow and concentrated in larger stocks. The lower high in the advance-decline line and the drop in the percentage above the 50-day MA from 78% to 54% confirm that participation has weakened. Five weeks of narrowing new highs minus new lows is consistent with fading upside momentum. However, breadth is a diffusion measure, not a timing tool; it can diverge for months in strong bull markets. Low VIX (14) and tight credit spreads do not confirm stress; they argue against an imminent credit-driven breakdown. Healthy rotation would show small caps and cyclicals taking leadership after a pause, not just a narrow index. The evidence is 'narrowing leadership with no credit confirmation,' which justifies risk reduction, not a wholesale bearish call.
2. Recommend whether to cut the 12% small-cap overweight and 5% industrials overweight before quarter-end. Address tracking error and the fully used risk budget.
Read the worked answer
Reduce, but do not eliminate, the small-cap overweight. With tracking error at 4.5% and the risk budget nearly full, the portfolio has little room for a factor drawdown. Cut the small-cap overweight from 12% to 6-8% and the industrials overweight from 5% to 2-3%, funding a mix of benchmark-weight large caps and cash. This lowers active risk while retaining some cyclical exposure in case breadth improves. Do not cut solely because of the breadth divergence; use it as a trigger to trim the most correlated, highest-beta positions. If the equal-weight index breaks below its 50-day MA and the A/D line makes a lower low, cut to benchmark weight. If small-cap relative strength stabilizes and the percentage above the 50-day MA recovers above 60%, restore part of the overweight. This is a risk-budget decision, not a directional bet.
3. What confirmation would distinguish a late-cycle narrowing from a healthy rotation, and how would you implement the decision without overreacting to one month of breadth data?
Read the worked answer
Look for confirmation across independent dimensions. Late-cycle narrowing would be confirmed by: the equal-weight index making lower highs while the cap-weighted index rises; new 52-week lows expanding; credit spreads widening from tight levels; defensive sectors outperforming cyclicals; and the VIX term structure inverting. Healthy rotation would show: small-cap relative strength bottoming, the A/D line repairing, the percentage above the 50-day MA stabilizing above 50%, and credit spreads stable. Implementation: use a two-step rule. First, trim overweights by half now because the risk budget is full. Second, require at least two of the four confirmation signals (breadth repair, small-cap relative strength, credit spreads, volatility term structure) before restoring risk. This avoids overreacting to a single month while respecting the possibility that narrowing breadth persists.