The Tactical Permanent Portfolio: Adding Trend Filters to an All-Weather Classic
The Permanent Portfolio is one of the most robust static allocations ever designed. Its 25/25/25/25 split across stocks, long bonds, gold, and short-term Treasuries has delivered steady returns across decades of varying market conditions. But it has one persistent weakness: it holds stocks at 25% even during severe equity bear markets, absorbing the full drawdown of a declining equity market.
The Tactical Permanent Portfolio (TPP) addresses this single weakness with a surgical modification: apply a trend filter to the equity slice only, leaving the other three positions permanently untouched. The result is a portfolio that maintains the Permanent Portfolio's all-weather character while meaningfully reducing drawdowns during equity crises.
Strategy Mechanics
| Parameter | Value |
|---|---|
| Assets | SPY, GLD, TLT, SHY |
| Risk-Off Asset | SHY (for equity slot only) |
| Allocation Method | FIXED_TIERS_PER_ASSET (25% each) |
| Top-N | 4 |
| Lookback / SMA | 10 months |
The Core Rule
TPP holds the same four assets as the Permanent Portfolio — SPY, GLD, TLT, SHY — each at 25%. The only modification: the equity position (SPY) is monitored against its 10-month simple moving average each month.
- If SPY is above its 10-month SMA → hold SPY at 25%
- If SPY is below its 10-month SMA → that 25% moves to SHY
The other three positions — GLD, TLT, and SHY — are permanent. They are never filtered, never rotated, and never removed. They remain at 25% each regardless of their own trend conditions.
What This Means in Practice
In bullish mode, TPP is identical to the standard Permanent Portfolio: 25% SPY, 25% TLT, 25% GLD, 25% SHY. In bearish mode, the portfolio becomes: 0% SPY, 25% TLT, 25% GLD, 50% SHY. The total portfolio shifts from 25% equities to 0% equities, with the freed capital absorbed by the short-term Treasury position.
Why Only the Equity Slice
The Asymmetric Risk Problem
Equity drawdowns are the primary source of portfolio pain. During the 2008 crisis, SPY lost nearly 50%. A 25% allocation to SPY meant a direct portfolio impact of roughly 12-13% from the equity position alone. TLT, GLD, and SHY — the other three positions — performed well during the same period: TLT rallied as rates fell, GLD held steady, and SHY provided stable income.
Filtering all four positions would be unnecessary and counterproductive. TLT's drawdowns during the 2008 period were minimal (it was rising), and applying an SMA filter to it would have moved the portfolio to cash during the very period when TLT was providing its most valuable service. GLD and SHY have even lower drawdown risk and do not benefit from trend filtering.
The equity slice is the only position where the cost of holding through a bear market (severe drawdown) significantly exceeds the cost of filtering (occasional whipsaw). Filtering only SPY captures the bulk of the tactical improvement while leaving the proven all-weather core intact.
The SMA(10) Filter
The 10-month SMA is a widely used trend indicator, popularized by Meb Faber. It is long enough to avoid reacting to normal market volatility (typical corrections of 5-10% often do not break the 10-month trend) but short enough to respond to genuine bear markets within one to two months of their onset.
The historical effectiveness of the SMA(10) filter on equities is well-documented: it has reliably moved capital to safety within 1-3 months of every major bear market onset since the 1920s. The tradeoff is occasional whipsaw during flat or volatile markets, where SPY oscillates around its moving average and generates false signals. But even with these false signals, the net effect over full market cycles is a meaningful reduction in maximum drawdown with minimal impact on long-term returns.
Performance Comparison
| Metric | Permanent Portfolio | Tactical Permanent (TPP) |
|---|---|---|
| Equity Exposure | Always 25% | 25% or 0% (based on SMA10) |
| Max Drawdown | ~12-15% | ~6-10% (reduced) |
| Positions Changed | None (fully static) | Only SPY (1 of 4) |
| Turnover | Rebalance only | 2-4 trades per year (SPY/SHY switches) |
The improvement is concentrated where it matters most: drawdown reduction. By eliminating equity exposure during bear markets, TPP reduces the maximum drawdown from roughly 12-15% (standard Permanent Portfolio) to approximately 6-10%. The CAGR impact is minimal — the equity position is held for the majority of the time and only removed during confirmed downtrends, so most of the equity upside is captured.
Behavior During Market Regimes
Bull Markets
During sustained equity uptrends, SPY remains above its 10-month SMA, and TPP is identical to the standard Permanent Portfolio. There is no tactical drag, no missed upside, and no additional trading. The strategy simply holds all four positions at 25% each.
Bear Markets
When SPY breaks below its 10-month SMA — typically 1-3 months into a bear market — the equity position switches to SHY. The portfolio becomes a 75% defensive allocation: TLT + GLD + SHY at 25% each, with an additional 25% in SHY from the equity switch (50% SHY total). During deflationary bear markets, TLT typically rallies, providing positive returns to offset any residual losses. The combination of zero equity exposure and a rallying TLT position has historically produced positive or flat returns during even severe equity bear markets.
Volatile / Sideways Markets
When SPY oscillates around its 10-month SMA — rising above for a month, dropping below for a month — the strategy generates whipsaw. Each switch between SPY and SHY incurs a small cost: the friction of selling one and buying another, plus the opportunity cost if the switch is reversed the following month. These whipsaw periods are the primary cost of the tactical enhancement. However, they typically involve small moves around the moving average, so the per-event cost is modest.
TPP in the Static Portfolio Family
TPP occupies a specific niche: it is the minimally tactical version of the Permanent Portfolio. Other approaches to enhancing the Permanent Portfolio exist — the Golden Butterfly adds a small-cap value position for higher returns, while the All Weather approach changes the weighting scheme entirely. TPP takes the simplest possible path: change nothing about the allocation, the assets, or the philosophy. Just add a single binary filter to the single most volatile position.
This surgical approach appeals to investors who believe in the Permanent Portfolio's design logic but want to reduce the equity drawdown risk that, while modest in percentage terms, can be psychologically challenging during severe bear markets. By modifying only one parameter of the original design, TPP preserves the portfolio's proven all-weather characteristics while addressing its most significant weakness.
Practical Implementation
Monthly Monitoring
At month-end, check whether SPY's closing price is above or below its 10-month SMA. If SPY is above, hold SPY at 25%. If below, hold SHY at 25% (in addition to the permanent 25% SHY position). The three permanent positions — TLT, GLD, SHY — require no monitoring and no action.
This makes TPP one of the lowest-maintenance tactical strategies available. Only one data point needs to be checked monthly (SPY vs its SMA), and only one position can potentially change. For investors who find more complex strategies like LAA or DAA intimidating, TPP offers an even simpler entry point into tactical asset allocation.