Protective Asset Allocation (PAA)
Developed by Keller & Keuning · Breadth Momentum · Med Risk
Protective Asset Allocation was published by Wouter Keller and Jan Willem Keuning in their 2016 SSRN paper (#2759734), introducing the first graduated crash protection mechanism in tactical allocation research. Prior to PAA, tactical strategies used binary switching — fully invested or fully defensive — which created sharp transitions prone to whipsaw losses during choppy markets. Keller's innovation was recognizing that the breadth of market weakness across a diversified universe could serve as a continuous gauge of systemic risk, allowing the defensive allocation to scale proportionally rather than triggering all at once.
The breadth-based approach drew on concepts from market internal analysis — the intuition that the percentage of assets in uptrends reveals more about overall market health than any single asset's performance. When most assets across different categories are trending positively, conditions are broadly favorable. When breadth deteriorates — when more assets fall below their moving averages — the probability of a meaningful downturn increases proportionally. Keller formalized this intuition into a quantitative framework where the count of assets above their moving averages directly determines the portfolio's equity-to-bond ratio.
PAA monitors twelve diverse assets spanning US equities, tech, small caps, European and Japanese stocks, emerging markets, real estate, commodities, gold, high-yield bonds, investment-grade bonds, and long-term Treasuries. This breadth across asset classes, geographies, and sectors provides a comprehensive reading of global market health that is difficult to manipulate through any single market event. The graduated response — more bonds as breadth declines, fewer as breadth improves — produces smooth transitions that avoid the whipsaw costs inherent in binary switching strategies.
The strategy's risk-on allocation is directed toward the top three assets ranked by their price-to-moving-average ratio, concentrating capital in the strongest current trends. This creates a dual-layer decision system: breadth determines how much to invest, and momentum determines where to invest — two complementary decisions operating on different information about market conditions.
How It Works
Breadth Measurement
Each month, all twelve assets in the universe are evaluated against their own twelve-month simple moving averages. The strategy counts how many are trading above their respective trend lines, producing a breadth score between zero and twelve. This score captures the overall health of the investment landscape across asset classes, geographies, and sectors simultaneously.
A breadth score of twelve means every asset in the universe is in an uptrend — conditions are broadly favorable across all segments of the global market. A score of six means half the universe has broken below trend, suggesting meaningful and widespread deterioration. A score near zero indicates a severe bear market affecting virtually all asset classes. The breadth score provides a natural, continuous measure of market stress that degrades gracefully as conditions worsen.
The Crash Protection Fraction
The breadth score is converted into a crash protection fraction that determines what percentage of the portfolio shifts to intermediate-term bonds (IEF). The relationship is inverse and proportional: when all twelve assets are above trend, the bond allocation is minimal. As more assets fall below their moving averages, the bond allocation increases smoothly and continuously.
This proportional scaling is PAA's defining innovation. Rather than a binary switch that creates a discrete jump from fully invested to fully defensive, the protection fraction adjusts gradually as conditions evolve. During a slow-developing bear market, the bond allocation increases progressively as successive assets break their trend lines — one by one, the portfolio becomes more defensive without any single dramatic reallocation event. The sensitivity of the protection can be calibrated through a threshold parameter, allowing customization for different risk tolerance levels.
Momentum-Ranked Asset Selection
The portion of the portfolio that remains invested in risk assets is allocated equally across the top three assets ranked by their price-to-moving-average ratio. This ratio measures how far each asset has risen above its own trend line, functioning as a normalized momentum metric that can be meaningfully compared across asset classes with different volatility characteristics.
This momentum ranking ensures the risk-on allocation is concentrated in the assets showing the strongest trends relative to their own history. A stock ETF trading at 110% of its moving average and a bond ETF trading at 103% of its moving average are both in uptrends, but the stock ETF's stronger relative position earns it a higher ranking and a larger share of the risk-on allocation. The equal weighting among the top three provides some diversification while maintaining a meaningful tilt toward the strongest performers.
Explore Protective Asset Allocation (PAA)
See the full backtest across 18 years of market data, or run your own what-if scenarios by adjusting all research parameters.