Rebalancing Strategies

Why drift matters

Suppose you set 60/40 in 2013 and never touched it. A decade of strong equity returns leaves you near 80/20 by 2023: a portfolio with roughly a third more drawdown exposure than the one you chose, arrived at by inaction rather than decision. The 2008 version of this story is the painful one: investors who had let equity drift upward through the 2003–2007 run met the crisis with materially more risk than their plan specified.

That is the honest case for rebalancing, and it is entirely about staying at the risk level you signed up for (Chapter 19's euro-denominated worst case only means something if the allocation behind it is still true). Any return benefit is secondary and conditional.

The methods, honestly compared

Calendar rebalancing. Rebalance on a fixed date regardless of drift: annually, semi-annually, quarterly. Simple, automatable, and immune to market-watching. The research consensus (Vanguard's work is the standard reference) finds annual to be the sweet spot: more frequent rebalancing raises costs and taxes without improving risk control materially.

Threshold (band) rebalancing. Rebalance whenever an asset drifts more than X percentage points from target, commonly 5. Slightly better risk control than pure calendar, at the price of requiring monitoring, which is precisely the activity Chapter 64 says harms investors.

Combined: annual check with bands. Look once a year; act only if something is outside its band. This captures nearly all the benefit of both, requires exactly one decision point per year, and is what this book recommends.

Cash-flow rebalancing. Direct new contributions (or, in the spending phase, take withdrawals) from whatever is furthest from target. Costs nothing, triggers no tax, and for anyone still saving it frequently does the whole job by itself. This should be your first tool, with the annual check as backstop.

Never rebalancing. A legitimate position for someone who genuinely intends to hold equities forever and treats the drift as their true preference. But then the written policy should say 100% equities, so the plan and the portfolio agree.

The "rebalancing bonus": what's actually true

The claim that rebalancing adds return is oversold. The accurate version:

• In mean-reverting, similar-return markets, systematically selling the strong and buying the weak can add a small amount: typically measured in tenths of a percent, not whole points.
• In persistently trending markets, rebalancing costs return. A rebalanced portfolio during a long equity bull run underperforms an unrebalanced one, by construction: you kept selling the thing that kept winning.
• Across long histories with equities out-returning bonds, rebalancing has usually reduced return slightly and reduced risk more.

So the honest framing: you rebalance to control risk and accept a small expected return cost for it, occasionally offset by the diversification effect. Any source promising rebalancing as free return is selling something.

Costs, and how to avoid most of them

Three costs, in order of size for most investors:

1. Tax. In a taxable account, selling to rebalance realises gains. This dominates everything else and is why cash-flow rebalancing exists.
2. Trading costs. Spreads and commissions (Chapters 56–57); small per trade, real if you rebalance often.
3. Attention. Every rebalancing occasion is a chance to make an unplanned decision.

The cost-minimising hierarchy, in order:

• Wrappers first. Rebalance inside ISAs, PEAs, 401(k)s and pensions where selling is tax-free (Chapters 28–29). Many investors can hold their entire allocation adjustment inside sheltered accounts while leaving taxable holdings untouched.
• Cash flows second. New contributions, dividends and (later) withdrawals, directed at the underweight asset.
• Sell last, and if you must, in the most tax-efficient lots (Chapter 52's specific identification), pairing with loss harvesting where available.
• Multi-asset funds solve it entirely. A one-fund 60/40 rebalances internally, invisibly and with no tax event for you (Chapter 27). For anyone who dislikes this whole chapter, that is the answer.

Rebalancing in a crash

The hardest part of the method, and the reason it is written down in advance. A rebalancing rule in a bear market instructs you to sell the asset that held up and buy the one that just fell 35%. It feels like catching a knife. It is also the entire mechanism by which a written plan buys low without anyone predicting anything (Chapter 26's Investor A).

Two guardrails make it survivable:

• Bands prevent hair-trigger action. With a 5-point band you rebalance a handful of times per decade, not weekly during turmoil.
• The decision was made in calm. Chapter 19's policy document, written when nothing was happening, is what authorises the March purchase. That is the whole reason it exists on paper.

One honest caveat: in a genuinely prolonged decline, rebalancing repeatedly into a falling asset can compound losses before it eventually pays. Bands, an annual cadence and a sane allocation are the protection; the alternative, abandoning the rule mid-crisis, is the failure mode this book worries about far more.

A worked example

Marta, 70/30, €300,000, still contributing €1,000 monthly.

• January check. Equities have run: 76/24. Drift is 6 points, outside her 5-point band.
• Step one, cash flow. Restoring balance needs about €18,000 shifted. Her next 12 months of contributions come to €12,000, so she redirects all of it to bonds, closing two-thirds of the gap for free.
• Step two, sheltered accounts. The remaining €6,000 she rebalances inside her pension, where selling is tax-free. Taxable holdings untouched, no tax event, no realised gains.
• Step three, satellites. Her Chapter 61 healthcare satellite has grown from 5% to 6.4%; she trims it back to 5% in the same session, in the same sheltered account.
• Total elapsed time: twenty minutes, once a year. Cost: near zero.

The following March, markets fall 25%. Her allocation is where she chose it to be, her written worst case still describes her actual portfolio, and the January routine, not a forecast, is what made both true.