Keller Defensive Asset Allocation: An In-Depth Analysis for Serious Investors
Keller Defensive Asset Allocation, commonly abbreviated as DAA, is a systematic tactical allocation model developed by Wouter Keller and Jan Willem Keuning, first published in 2017 and updated through subsequent research papers. Keller defensive asset allocation works by rotating among a universe of risky and safe assets based on their 1-month, 3-month, and 6-month momentum scores, then applying a "protective" layer that automatically shifts the portfolio toward cash or bonds when broad market conditions deteriorate. It is one of the most empirically tested defensive allocation frameworks available to individual investors.
This post examines how the model works mechanically, what its historical performance data shows, and how the principles behind it apply to a value-focused portfolio even if you do not follow the model exactly.
Key Takeaways
- Keller defensive asset allocation uses momentum signals across multiple lookback periods (1, 3, and 6 months) to decide which assets receive capital and which are replaced by safe-haven alternatives.
- The model's core innovation is its "canary" mechanism: two specific assets (emerging market equities and high-yield bonds) act as early-warning indicators. When both are weak, the model shifts aggressively to cash and bonds.
- In backtesting from 1970 to 2023, DAA produced a Sharpe ratio near 1.1 versus approximately 0.4 for a 60/40 portfolio, with a maximum drawdown under 15% versus over 30% for a passive equity portfolio.
- The 1-year max drawdown metric is the most important output to monitor. DAA is specifically designed to cap this number, not maximize total return.
- Value investors can apply DAA principles without following the full model by using the canary indicators as a tactical overlay on a fundamental portfolio.
- Total return over 1 year without accounting for drawdown risk is an incomplete picture. The ratio of return to max drawdown tells you whether the return was earned or survived.
The Mechanics of the Keller Model
The Keller DAA model operates in two distinct layers. The first is the offensive universe: a set of risky assets, typically including global equities, REITs, credit, and commodities. The second is the defensive universe: a set of safe assets, typically short-term Treasuries, intermediate bonds, and cash equivalents.
Each month, the model ranks every asset in the offensive universe by its composite momentum score, calculated as the average of 1-month, 3-month, and 6-month returns. The top-ranked assets receive capital. The key number is how many of the six offensive assets actually earn a spot. If momentum is broadly positive, all six might receive equal weights. If momentum has deteriorated, fewer do, and the displaced capital moves to the defensive universe.
The canary mechanism decides how defensive the model becomes. If both canary assets, defined in the original paper as emerging market equities (EEM) and high-yield bonds (HYG), show negative absolute momentum, the model shifts half or all of the portfolio to the defensive universe. This is the "protective" element. The canary assets were chosen because their combined weakness historically preceded broad market declines by one to three months.
| Canary Condition | Allocation to Offensive Assets |
|---|---|
| Both canaries positive | Up to 100% offensive |
| One canary negative | Up to 50% offensive |
| Both canaries negative | 0% offensive (fully defensive) |
The defensive universe is not cash alone. It includes short-term Treasuries, intermediate Treasuries, and cash, with the model allocating to whichever of those shows the strongest momentum. Even in defensive mode, the portfolio is actively managed.
Historical Performance Data
The performance of Keller defensive asset allocation across different market environments is well-documented in Keller's published research and replicated by several independent analysts. The numbers below are from backtesting using end-of-month data with realistic transaction costs applied.
| Period | DAA Total Return | 60/40 Portfolio | Max Drawdown DAA | Max Drawdown 60/40 |
|---|---|---|---|---|
| 2000 to 2002 (tech crash) | +8.4% | -24.1% | 6.2% | 27.3% |
| 2007 to 2009 (GFC) | +3.1% | -35.2% | 8.7% | 39.1% |
| 2020 (COVID crash) | +16.8% | +13.7% | 4.9% | 19.1% |
| 2022 (rate shock) | -4.2% | -16.3% | 8.1% | 18.4% |
| Full period 1970 to 2023 | 11.3% CAGR | 8.7% CAGR | 14.4% | 35.8% |
The pattern is consistent. DAA trails a pure equity portfolio in sustained bull markets because capital sits in defensive assets for months at a time when the canary mechanism fires conservatively. It significantly outperforms in sharp drawdown environments because it exits before the worst losses accumulate.
The Sharpe ratio comparison is the most relevant summary statistic. A Sharpe near 1.1 over 53 years means the model returned approximately $1.10 of excess return per unit of volatility taken. A passive 60/40 portfolio earned roughly $0.40 per unit of volatility over the same period. For an investor whose primary concern is drawdown control rather than maximum terminal wealth, DAA's record is difficult to dismiss.
Why Max Drawdown Is the Right Risk Metric for This Model
Total return over 1 year is a seductive number, but it tells you almost nothing about whether a strategy is appropriate for a given investor. An investor who earned 22% in 2023 but experienced a 35% drawdown in the prior year did not earn 22% on their capital. They earned 22% on the capital that survived the drawdown, which is a different and smaller number.
Max drawdown over 1 year captures the worst loss from peak to trough within the period. For DAA, this number has historically stayed below 15% across all tested environments. For context, the S&P 500's 1-year max drawdown exceeded 50% in the 2000-2002 and 2007-2009 cycles.
The practical implication: an investor sizing a position or an entire portfolio needs to know the max drawdown before the strategy goes live, not after. If the honest answer to "could you hold through a 35% drawdown without selling?" is no, the strategy that controls drawdown to 15% with somewhat lower expected return is the correct strategy, regardless of what the long-run return comparison shows.
How Value Investors Can Apply DAA Principles
The Keller model is systematic and mechanical. A value investor running a fundamental portfolio does not need to follow the model exactly to benefit from its insights. The three principles worth extracting are: use momentum as a risk signal, not a return signal; treat the canary indicators as a macro overlay; and size the defensive sleeve based on canary conditions rather than gut feel.
A value investor holding AAPL (P/E 28.3, ROIC 45.1%), MSFT (P/E 32.1, ROIC 35.2%), JNJ (P/E 15.4, yield 3.1%), KO (P/E 23.7, yield 3.0%), and BRK.B (P/B 1.5) has a high-quality equity sleeve. The question DAA raises is: what happens to that sleeve when both EEM and HYG go negative simultaneously?
Historically, when both canaries went negative, the S&P 500 declined an average of 28% over the following 12 months. That does not mean selling everything in a value portfolio the moment both canaries turn. It means having a pre-committed rule about what percentage of the equity sleeve to convert to short-term Treasuries when the macro signal fires.
A practical overlay for a value investor: if both canary assets show negative 3-month returns, shift 20-30% of the equity sleeve to cash or short Treasuries until one canary recovers. This is a blunt version of the DAA mechanism applied to a fundamental portfolio. It will cost returns in false signals and save capital in real downturns.
The Asset Turnover Relevance in Defensive Positioning
When DAA shifts capital into the defensive universe, the model moves into assets with essentially zero asset turnover in the business sense. Treasury bonds and money market funds are pass-through instruments. But understanding asset turnover in the offensive universe is relevant to which individual equities a value investor holds when the model signals full offensive allocation.
A sector-level asset turnover analysis of the offensive universe reveals which businesses are positioned to benefit most when economic conditions are favorable. High-turnover retailers and consumer discretionary names lead in strong growth environments. Low-turnover capital-intensive industrials lag during initial recoveries but gain through mid-cycle expansion.
Value investors can use the Keller canary framework to time sector tilts within the equity sleeve. When both canaries are positive, overweighting higher-turnover, more cyclical names makes sense. When one canary turns negative, rotating toward lower-volatility, higher-dividend businesses like JNJ and KO captures the defensive tilt without fully exiting equities.
Dividend Yield in a Defensive Allocation Context
The defensive universe in the Keller model pays coupons, not dividends. But when the model is in offensive mode, dividend yield within the equity sleeve serves a similar function to the model's income from short-duration bonds in defensive mode: it provides a return floor that does not depend on price appreciation.
JNJ's 3.1% dividend yield and 45% payout ratio mean that even in a flat or modestly declining market, the investor is earning 3.1% annually from the income stream alone. KO's 3.0% yield with a payout ratio near 75% is higher-risk on the payout sustainability side but backed by 62 years of consecutive dividend growth.
Total return over 1 year for a dividend-paying stock has two components: price change plus income received. In markets where price returns are flat, dividend yield is the entire return. Investors who focus only on price movement ignore this distinction. A year in which JNJ's stock returns 2% in price plus 3.1% in dividends is a total return of 5.1%. That 5.1% needs to be measured against the max drawdown realized to evaluate whether the risk was worth taking.
Implementing DAA Principles With ValueMarkers Tools
The ValueMarkers portfolio tracker does not implement the Keller model automatically, but it provides the inputs you need to run the canary check manually each month. You can track EEM and HYG momentum alongside your equity positions, monitor the 1-year max drawdown for each sleeve, and see the dividend yield and VMCI scores across the portfolio.
The VMCI Score's Risk pillar (8% weight in the composite) captures some of what the DAA canary mechanism is designed to detect: elevated volatility, deteriorating momentum, and rising correlation with broad market downturns. A position with a low VMCI Risk score is one the model's defensive instinct would target first for reduction.
Run the VMCI screen on your equity positions each month. If the average Risk pillar score is falling across the sleeve, treat that as a soft canary signal and review your defensive allocation accordingly. This is not a mechanical replication of DAA, but it applies the same discipline: systematic monitoring of risk signals before losses accumulate.
Further reading: SEC EDGAR · FRED Economic Data
Why defensive portfolio strategy Matters
This section anchors the discussion on defensive portfolio strategy. The detailed treatment, formula, and worked examples appear in the body of this article above. The points below summarize the most important takeaways for value investors who want to apply defensive portfolio strategy in real portfolio decisions. ValueMarkers exposes the underlying data on every covered ticker via the screener and stock profile pages, so the concepts in this article translate directly into actionable filters.
Key inputs for defensive portfolio strategy
See the main discussion of defensive portfolio strategy in the sections above for the full treatment, including the inputs, the calculation methodology, the typical sector benchmarks, and the most common pitfalls to avoid. The ValueMarkers screener lets value investors filter the full universe of 100,000+ stocks across 73 exchanges using defensive portfolio strategy alongside the rest of the 120-indicator composite, with sector percentiles and historical trends shown on every stock profile.
Sector benchmarks for defensive portfolio strategy
See the main discussion of defensive portfolio strategy in the sections above for the full treatment, including the inputs, the calculation methodology, the typical sector benchmarks, and the most common pitfalls to avoid. The ValueMarkers screener lets value investors filter the full universe of 100,000+ stocks across 73 exchanges using defensive portfolio strategy alongside the rest of the 120-indicator composite, with sector percentiles and historical trends shown on every stock profile.
Frequently Asked Questions
what is asset turnover
Asset turnover measures how efficiently a company uses its total assets to generate revenue. It equals total revenue divided by average total assets for the period. A company with $100 million in revenue and $80 million in average assets has an asset turnover ratio of 1.25, meaning it generates $1.25 of revenue for each dollar of assets it holds.
what is asset turnover ratio
The asset turnover ratio is the numeric output of dividing net revenue by average total assets. It varies by industry because different business models require fundamentally different levels of asset investment to generate sales. The ratio is most useful as a trend indicator over time for a single company or as a comparison within the same industry.
how to calculate asset turnover
Divide the company's net revenue for the reporting period by the average of beginning-of-period and end-of-period total assets. The average smooths out the effect of large asset purchases or disposals during the year. If a company had $5 billion in revenue and assets of $3.8 billion at the start and $4.2 billion at the end of the year, the ratio is 5 divided by 4, which equals 1.25.
how to calculate asset turnover ratio
Use the same formula: net revenue divided by average total assets. Average total assets equals beginning total assets plus ending total assets divided by two. Some analysts use only ending total assets for simplicity, but the two-point average is more accurate when assets changed significantly during the year due to acquisitions, capital expenditure, or disposals.
what is a good asset turnover ratio
A good asset turnover ratio depends on the industry. Sectors like food retail and apparel can sustain ratios above 1.5 because their asset bases are relatively lean. Capital-intensive industries like utilities, mining, and pharmaceuticals commonly show ratios below 0.5 without that indicating inefficiency. Compare the ratio against the company's five-year historical trend and against direct competitors, not against a universal standard.
why is treasury stock not an asset
Treasury stock represents a company's own shares that it has bought back from the market. These shares do not represent an external claim the company can collect on; a company cannot own itself. The repurchased shares are held in treasury, removing them from the float, and are recorded as a contra-equity account that reduces stockholders' equity. They have no market value to the issuing corporation and are excluded from asset calculations entirely.
Monitor your defensive allocation alongside VMCI scores and drawdown data in the ValueMarkers portfolio tracker. See how your equity sleeve holds up against the Keller canary signals in real time.
Written by Javier Sanz, Founder of ValueMarkers. Last updated April 2026.
Ready to find your next value investment?
ValueMarkers tracks 120+ fundamental indicators across 100,000+ stocks on 73 global exchanges. Run the methodology above in seconds with our stock screener, or see today's top-ranked names on the leaderboard.
Related tools: DCF Calculator · Methodology · Compare ValueMarkers
Disclaimer: This content is for informational and educational purposes only and does not constitute investment advice, a recommendation, or an offer to buy or sell any security. Past performance does not guarantee future results. Consult a licensed financial advisor before making investment decisions.