Learn tactical asset allocation

Why do this? Markets aren't always fairly priced. When stocks are expensive (high P/E ratios, euphoric sentiment), tilting toward bonds or cash can reduce risk. When stocks are cheap (recession fears, high dividend yields), overweighting them can boost returns. TAA tries to harvest these mispricings without abandoning your core plan.

What Tactical Asset Allocation Actually Means

Key distinction from strategic allocation:

• Strategic: "I'm 35, moderate risk tolerance → 70/30 stocks/bonds, rebalance back quarterly, hold for 20 years"
• Tactical: "Market P/E is in the 90th percentile historically, recession indicators rising → shift to 60/40 for the next 6 months"

Key distinction from market timing:

• Market timing: "I'm selling ALL stocks today, going 100% cash, will buy back at the exact bottom"
• TA: "I'm reducing stocks from 70% to 60%, adding to bonds and REITs, keeping diversification, and I'll reassess in 3 months"

TA is bounded, systematic, and reversible—not all-or-nothing bets.

Why TA Can Work (And Why It's Hard)

Theoretical Case

Derivation from first principles:

1. Portfolio return $R_p = \sum w_i R_i$ (weighted sum of asset returns)
2. Strategic portfolio return: $R_p^{\text{strat}} = \sum w_i^{\text{strat}} R_i$
3. TAA portfolio return: $R_p^{\text{TAA}} = \sum w_i^{\text{TAA}} R_i$
4. Excess return (alpha) = $R_p^{\text{TAA}} - R_p^{\text{strat}} = \sum (w_i^{\text{TAA}} - w_i^{\text{strat}}) R_i = \sum \Delta w_i R_i$
5. Taking expectations: $E[\text{Alpha}] = \sum \Delta w_i \cdot E[R_i]$

Why this step? We're decomposing the outperformance into how much we overweight ($\Delta w_i$) times the expected return of that asset. This shows TA only works if you overweight the assets that will outperform.

The Practical Challenge

To profit, you need:

1. Predictive signals that actually forecast 3-12 month returns (valuation, momentum, macro indicators)
2. Low transaction costs (if you trade too much, fees eat alpha)
3. Discipline to reverse tilts when signals flip (don't let a tactical tilt become permanent)

Most retail investors fail at #1 (signals are noise) or #3 (overweight stocks in euphoria, freeze in panic—exactly backward).

Common TA Signals & Their Logic

Logic: When stocks are expensive, future 10-year returns tend to be lower. Reduce equity allocation temporarily.

Action:

• Strategic allocation: 70% stocks, 30% bonds
• Tactical tilt: → 60% stocks, 35% bonds, 5% commodities
• Time horizon: 12 months, reassess quarterly

Why this step?

• CAPE = Price / 10-year average earnings. High CAPE → you're paying a lot per dollar of earnings.
• Historically, starting CAPE predicts ~40% of variance in subsequent 10-year returns (not next month—this is a slow signal).
• You're not abandoning stocks (still 60%!), just trimming froth and adding dry powder.

Calculation example:

If historical median = 17, you're paying 1.9× the normal multiple. Expected 10-year return might be 3-4% instead of 8-9%.

Logic: Assets showing strong recent performance tend to continue for 3-6 more months (momentum persistence).

Action:

• Strategic: 50% US stocks, 30% bonds, 10% EM, 10% REITs
• Tactical tilt: → 45% US, 25% bonds, 20% EM, 10% REITs (double EM exposure)

Why this step?

• Momentum works because of underreaction (news diffuses slowly) and herding (trend followers pile in).
• The 6-month window is empirical—shorter windows are noise, longer windows start to mean-revert.
• You're riding the wave, but keeping a stop-loss rule (if EM reverses next month, you'll trim).

Calculation (EM index went from 1000 six months ago to 1180 today):

Rank all assets by this metric, overweight top 2-3.

Logic: Inverted curve predicts recession ~12-18 months out. PMI < 50 = manufacturing contraction. Both → reduce equity risk.

Action:

• Strategic: 60% stocks, 40% bonds
• Tactical tilt: → 45% stocks, 50% bonds, 5% gold

Why this step?

• Inverted curve → bond market expects Fed to cut rates in the future (only happens in slowdowns).
• Equities average -30% in recessions; bonds rally as rates fall; gold is safe haven.
• You're hedging recession risk, not predicting the exact timing (impossible).

Calculation:

How to Implement TAA (The Process)

Step 1: Define Your Signal Framework

Pick 2-4 signals you'll monitor (don't use 20—overfit). Examples:

• Valuation: CAPE ratio, P/B ratio, dividend yield vs. historical percentile
• Momentum: 6-month or 12-month total return ranking
• Macro: Unemployment rate, yield curve, manufacturing data
• Sentiment: VIX level, put/call ratio (contrarian)

Where $S_j$ is the $j$-th signal (e.g., CAPE), $\mu_j$ is its historical mean, $\sigma_j$ is its std dev, and $M$ is the number of signals.

Why this step? Raw signals are in different units (CAPE is a ratio, momentum is a %, yield curve is basis points). Z-scores standardize them so you can compare and average.

Example:

• CAPE z-score = +1.8 (expensive)
• Momentum z-score = -0.5 (weak)
• Yield curve z-score = +0.2 (slightly steep)
• Average = (+1.8 - 0.5 + 0.2) / 3 = +0.5 → neutral-to-slightly-bearish, maybe tilt -5% equities

Step 2: Set Tilt Limits

Rule: Never deviate more than ±15% from strategic weights. Example:

• Strategic: 70% stocks → TA range is 55-85%
• Strategic: 30% bonds → TAA range is 15-45%

Why? Larger deviations turn TA into gambling. If your strategic allocation is 70/30, you already decided that's your risk tolerance—don't become 100/0 tactically and violate your own plan.

Step 3: Rebalance Calendar

Review signals monthly or quarterly. If signal flips, gradually adjust back over 1-2 months (don't whipsaw with every headline).

Tax consideration: In taxable accounts, avoid triggering short-term capital gains. Use new contributions or tax-loss harvesting to implement tilts.

Step 4: Track and Learn

Log every tactical decision: date, signal values, tilt made, outcome. After 3-5 years, calculate:

If alpha is negative or < 0.5% after costs, abandon TA and just rebalance back to strategic.

Common Mistakes (And Why They Feel Right)

Why it feels right: You want to protect gains and avoid losses. Cash feels safe.

The steel-man: You're recognizing risk, which is valid. Markets can stay overvalued for years (1998-2000 dot-com bubble). Staying invested in euphoria is psychologically hard.

The fix: TA is partial tilts, not binary switches. Go from 70% stocks to 55%, not 0%. The key insight is that you don't know the direction in advance—so keep exposure and stay diversified. Also set a time-bound review (e.g., "I'll reassess in 6 months") so cash doesn't become permanent.

Math of the mistake (why all-or-nothing is a bad bet): Suppose you're forced to choose ahead of time and consider two scenarios—market rises 10% then falls 20% (Scenario A), OR market falls 20% then rises 10% (Scenario B). Assume cash earns 0% for simplicity.

Honest reading: In these two particular sequences, cash "wins." That is exactly the point of the steel-man: cash can win, and so can 100% stocks, depending on the path you can't predict. Betting everything on one guess is a coin flip. The 60/40 tilt is the robust middle: it never suffers the full −12% crash and never gets left fully behind in a rally. TAA prefers the bounded tilt not because it always beats cash, but because it reduces the range of outcomes when your forecast is uncertain.

Why it feels right: Momentum! Plus FOMO—you want those gains.

The steel-man: Momentum does exist over 6-12 months. Assets that have done well often continue.

The fix: Check valuation before chasing momentum. If EM rallied 25% but P/E ratios went from 12 to 20, you're buying expensive. Better TA: overweight EM if momentum is strong AND valuation is reasonable (P/E still < historical median). Combine signals.

Example: In 2021, ARK Innovation ETF (ARKK) had 3-year returns of 300%+. Momentum chasers piled in at the top. By 2022, it fell 70% as valuations normalized. A valuation check (stocks trading at 15× sales!) would have stopped you.

Why it feels right: You want to stay ahead of the market. Daily news feels urgent.

The steel-man: Markets do react to news (Fed announcements, earnings surprises). Being reactive shows diligence.

The fix: TA signals must be low-frequency (monthly or quarterly). Daily noise has zero predictive power—it's just randomness. Use a rule: "I only adjust if my composite signal changes by >1 standard deviation, AND it's been at least 30 days since the last change."

Why daily moves are noise: Daily stock return std dev ≈ 1.2%. Signal-to-noise ratio for 1-day prediction ≈ 0.05. For 6-month prediction ≈ 0.4 (still weak, but ~8× better). Transaction costs eat any edge at high frequency.

Active Recall Questions

"Okay, so you know how we said you should own some stocks and some bonds and never change it? Well, tactical allocation is like this: most of the time, you stick to your plan. But if you notice something unusual—like everyone is going crazy buying stocks and prices are super high—you might sell a little bit of your stocks and buy more bonds, just to be safer. It's like if you're riding a bike: you mostly stay on the path, but if you see a big rock, you steer around it a little, then get back on the path. You're not abandoning the path (that would be market timing), just making small adjustments when there's a good reason."

#flashcards/stock-market

Connections

• 5.6.01-Learn-strategic-asset-allocation: TA is temporary deviations from your strategic baseline
• 5.5.03-Apply-rebalancing-strategies: Tactical tilts create drift; you rebalance back to strategic targets periodically
• 4.2.01-Understand-market-cycles: TAA tries to exploit where you are in the cycle (expansion → overweight stocks, recession → underweight)
• 3.4.02-Calculate-PE-ratio: CAPE (cyclically adjusted P/E) is a key TA valuation signal
• 5.3.02-Design-diversified-portfolio: TAA maintains diversification across tilts—never goes 100% into one asset
• 6.1.03-Manage-behavioral-biases: TAA discipline fights recency bias (chasing winners) and panic selling (timing the market)

Concept Map

Hinglish (regional understanding)