Tactical asset allocation: shifting weights on valuation

Strategic asset allocation is the split you decided on and hold. Tactical asset allocation is deliberately deviating from it — leaning towards equity when it looks cheap, away when it looks expensive. It is the most intellectually appealing idea in portfolio management and the one with the worst realised track record among the people who try it.

Both of those things are true at once, and the reason is worth understanding before you decide whether to attempt it.

The idea, stated fairly

Valuation has some predictive power over long horizons. Markets bought at high multiples have historically delivered lower subsequent ten-year returns than markets bought at low ones. That relationship is real and reasonably well documented.

So the reasoning goes: if valuation predicts long-run returns, why hold a fixed 60/40 through both extremes? Lean to 70/30 when the market is cheap and 50/50 when it is expensive, and you should improve on the static mix.

The logic is sound. The implementation is where it dies.

Why it fails in practice

Valuation predicts long horizons, not next year. The relationship between today's multiple and the next ten years' return is meaningful; between today's multiple and the next twelve months' return it is close to noise. Tactical allocation asks you to act on a signal calibrated to a decade using a decision you will review in a quarter.

Expensive markets stay expensive for years. A disciplined investor who reduced equity at "expensive" in 2015, or 2017, or 2021, spent multiple years watching the market get more expensive still, underperforming visibly, before any of those calls would have paid. Most people do not last.

You have to be right twice. Getting out is half the trade. The harder half is getting back in, which by construction must happen when the news is at its worst and the case for waiting is at its most persuasive. This is where most tactical programmes actually break — the exit was fine, the re-entry never happened.

Costs and tax are real. Every tactical move in a taxable Indian account realises capital gains: 20% on equity units under a year, 12.5% above the ₹1,25,000 exemption thereafter. A strategy that adds 1% a year gross and triggers two taxable switches has added nothing.

And it is a doorway. Most portfolios that adopt "tactical" allocation drift into ordinary market timing within two years, because the framework legitimises acting on a view, and views arrive more often than valuation extremes do.

If you are going to do it anyway

Some rules that separate a disciplined tactical process from a rationalised hunch:

1. Write the rule before you need it. A specific, mechanical trigger: "Equity target moves to 70% when the market's trailing P/E falls below the 20th percentile of its ten-year range, and to 50% above the 80th percentile." Anything you can only articulate in the moment is a hunch.

2. Cap the deviation. ±10 percentage points from the strategic weight is plenty. Beyond that you have replaced your strategy rather than tilted it.

3. Use a slow signal. Valuation percentiles over ten years, or long-term earnings-based measures. Anything responding to a single month is a momentum signal wearing a valuation costume.

4. Move gradually. Shift in tranches over months, via a systematic transfer plan, rather than in one decision on one day.

5. Execute the tax-cheapest way, in this order: redirect new SIP money to the underweight asset; use tax-sheltered or internally-rebalanced holdings; sell last. This is the same ladder as ordinary rebalancing and it matters more here because tactical moves are more frequent.

6. Keep a written log. Date, trigger value, action, and the reasoning. Read it a year later. Nothing improves calibration faster, and nothing exposes a drifting process sooner.

The alternative most people should take

Let a fund do it, inside a wrapper, without the tax.

A balanced advantage or dynamic asset allocation fund runs a published, valuation-driven model that moves equity exposure between wide bands — and crucially, the switches happen inside the scheme, so they are not taxable events for you. That single structural advantage often exceeds whatever alpha a retail investor's own tactical process would generate.

The trade-offs, stated honestly: the model is the AMC's and not yours, the equity exposure will not be what you would have chosen at any given moment, and these funds typically use derivatives to maintain the 65% equity threshold for tax purposes — a detail covered in that guide.

Or: do nothing tactical at all. A strategic allocation with a ±5 point rebalancing band already does a mild, mechanical version of "buy low, sell high" without requiring any forecast. For most investors that captures most of the available benefit at none of the behavioural cost.

Pitfalls to avoid

• Confusing tactical allocation with market timing. The difference is a written rule and a capped deviation. Without both, they are the same thing.
• Acting on a twelve-month view with a ten-year signal.
• Going to cash. A "temporary" 100% cash position is the position people hold for years.
• Ignoring re-entry. Define the buy-back trigger at the same moment you define the sell trigger, or you will not have one when it matters.
• Forgetting the tax bill. Two switches a year can consume the entire edge.
• Judging it over a year. If the process is valuation-based, its evaluation window is a full cycle — and by then, most people have abandoned it.

Key takeaway

Valuation genuinely predicts long-run returns, which is why tactical allocation sounds compelling — and it predicts almost nothing about the next year, which is why it so rarely works in practice. You must be right twice, expensive markets stay expensive for years, and every move in a taxable account gives back part of the edge. If you attempt it, write the trigger down in advance, cap the deviation at about ten percentage points, move in tranches and log every decision. For most investors the better answers are a balanced advantage fund, where the switching happens inside the scheme and is not taxed to you, or simply a strategic allocation with a rebalancing band — which already buys low and sells high without needing anyone to forecast anything.