Rebalancing Your Portfolio

Left alone, a portfolio quietly becomes riskier than the one you chose. Rebalancing is the boring maintenance task that puts it back. This guide covers why drift happens, how often to act, bands versus the calendar, how to rebalance with new money instead of sales, and what it actually costs in taxes.

By Karanvir Singh, founder of Quantix Invest

What drift is and why it matters

You pick an allocation, say 70% stocks and 30% bonds, because that mix matches the risk you are willing and able to carry. Then the market goes to work. Stocks and bonds do not return the same amount in any given year, so the weights move away from your target. That movement is drift.

Drift is not random noise that averages out. It has a direction: over long stretches, the higher-returning asset takes over. A portfolio left alone for a decade of good equity markets does not stay at 70/30. It becomes 80/20, then 85/15. Nobody decides this. It just happens, and it happens most aggressively right before the moments when you would most want the ballast.

The result is a portfolio whose risk profile no longer matches the person who owns it. You signed up for a mix that would lose maybe 30% in a bad bear market. Without noticing, you now own a mix that would lose 40%. If a 40% loss would make you sell at the bottom, the drift has done real damage, and it did it silently.

There is a second effect that people find harder to accept: rebalancing forces you to sell what has done well and buy what has done badly. That feels wrong in the moment, every single time. Selling the winner feels like cutting flowers to water weeds. But the alternative is letting whatever just ran up become an ever larger share of your net worth, which is exactly the behavior that concentrates people into the top of bubbles.

Worked example: ten years of neglect

Start with $100,000 at a 70/30 target: $70,000 in a total stock market fund, $30,000 in a total bond market fund. Assume, purely for the arithmetic, that stocks compound at 10% per year and bonds at 3% per year for ten years. Nothing is added and nothing is sold.

Stocks after 10 years: $70,000 x 1.10^10. Since 1.10^10 = 2.594, that is $181,580.

Bonds after 10 years: $30,000 x 1.03^10. Since 1.03^10 = 1.344, that is $40,320.

Total: $221,900. Stock weight = 181,580 / 221,900 = 81.8%. Bond weight = 18.2%.

The investor who wanted 70/30 now owns roughly 82/18 without ever placing a trade. Now suppose a bear market hits and stocks fall 45% while bonds gain 5%.

(The 70/30 column assumes the same $221,900 total, rebalanced back to target the day before the crash: $155,330 stocks and $66,570 bonds.)

Six percentage points of extra drawdown, about $13,000 on this portfolio, is the price of the drift. Whether that matters depends entirely on whether it changes your behavior. If you hold either way, the drifted portfolio recovers faster too. If a 36% loss makes you capitulate and a 30% loss does not, the drift cost you the entire investing plan.

Rebalancing is risk control, not a return booster

You will read claims about a "rebalancing bonus," the idea that systematically selling high and buying low adds return by itself. Be careful with this. The bonus is real only in a specific situation: when you are combining assets with similar long-run returns and low correlation to each other, so their zigzags genuinely offset. Rebalancing between two volatile, similarly performing assets can add a small amount of compound return relative to buy-and-hold.

Stocks and bonds are not that case. Stocks have materially higher expected returns. Rebalancing from stocks into bonds means repeatedly moving money from the higher-returning asset to the lower-returning one. Over long horizons, the rebalanced 70/30 portfolio typically ends up with less money than the drifted one, not more. What it has instead is a smoother ride and a risk level that stayed where you put it.

That is a fair trade, and it is worth being honest about: you are buying risk control, and paying for it with a bit of expected return. The historical studies that show rebalancing "adding" return usually cover periods that ended in a stock drawdown, or measure risk-adjusted rather than absolute return.

Calendar, bands, or both

There are two ways to decide when to act.

Calendar rebalancing. Check on a schedule: annually, semiannually, or quarterly, and restore the target regardless of how far it has drifted. Simple, easy to automate, easy to remember. Its weakness is that it is blind: it may trade when nothing has moved, and it will ignore a violent 20% move that happens two weeks after your annual check.

Threshold (band) rebalancing. Set tolerance bands around each target, and act only when an asset breaks out of its band. Responsive to what actually happens, which is the point. Its weakness is that it requires you to look, which most people do not do consistently, and in a volatile year it can trigger several times.

The research on this, going back to Vanguard's well-known work on rebalancing frequency, has a consistent punchline: the choice between reasonable rules matters much less than having a rule at all. Annual, semiannual, and 5% band strategies produced broadly similar risk-adjusted results in long historical tests, while never rebalancing produced a clearly different and riskier portfolio. More frequent rebalancing did not reliably improve outcomes and did increase costs and taxes.

The practical answer for most people is the hybrid: check on a schedule, trade only if a band is breached. You get the responsiveness of bands with the discipline of a calendar, and you avoid trading for the sake of a date.

The 5/25 rule

A widely used band rule, popularized by Larry Swedroe, handles both large and small positions sensibly:

• The 5 rule: for any asset class that is 20% or more of the portfolio, rebalance when it drifts 5 percentage points (absolute) from target.
• The 25 rule: for any asset class smaller than 20%, rebalance when it drifts 25% relative to its target.

Why two rules? Because an absolute 5-point band is meaningless for a 5% sleeve, which could triple before it triggers. And a 25% relative band is far too twitchy for a 60% holding, which would trigger on a 15-point move that is arguably fine.

Note that the 25% relative rule on a 5% sleeve triggers on a move of only 1.25 percentage points, which is a very small dollar amount. Many people sensibly add a minimum trade size, say "ignore any rebalance worth under $500," so they are not placing trivial trades and generating paperwork.

Rebalancing with new contributions

The best rebalancing trade is the one you never have to place. If you are still adding money, you can steer the portfolio back toward target simply by directing new contributions to whatever is underweight. This is sometimes called cash flow rebalancing, and while you are accumulating it does most of the work for free.

The advantages are large and unglamorous:

• No realized gains, so nothing taxable happens in a taxable account.
• No sales, so no bid-ask spreads, no wash sale complications, no trade tickets.
• It buys the laggard automatically, which is the behavior you want and the one you find hardest to do deliberately.

The same logic runs in reverse in retirement: if you are withdrawing, take the withdrawal from whatever is overweight. A retiree drawing 4% a year has a meaningful rebalancing lever built into normal spending, and in a year when stocks are down, that lever means selling bonds for living expenses and leaving equities alone to recover.

Dividends and interest are the third lever. Turning off automatic reinvestment in a taxable account and instead sweeping distributions to cash, then deploying that cash to the underweight asset, is a small, entirely free rebalancing engine. In tax-advantaged accounts, automatic reinvestment is fine and simpler, since you can rebalance there without tax consequences anyway.

Worked example: rebalancing without selling

Target is 70/30. Current balances after a strong stock year:

• Stocks: $88,000
• Bonds: $32,000
• Total: $120,000, which is 73.3% stocks and 26.7% bonds

Stocks are 3.3 points over target. Under a 5-point band this does not even trigger, but suppose you want to fix it anyway with the $10,000 you are about to contribute.

Step 1. Compute the post-contribution total. $120,000 + $10,000 = $130,000.

Step 2. Compute the target dollars. Stocks 70% of $130,000 = $91,000. Bonds 30% = $39,000.

Step 3. Compute the gaps. Stocks need $91,000 minus $88,000 = $3,000. Bonds need $39,000 minus $32,000 = $7,000.

Step 4. Check that it fits. $3,000 + $7,000 = $10,000, exactly the contribution. Direct the money accordingly and the portfolio lands precisely on 70/30 with zero sales and zero tax.

Now the case where the contribution is not enough. Same starting balances, but stocks have run further:

• Stocks: $98,000, Bonds: $32,000, Total: $130,000, so 75.4% stocks. This breaches a 5-point band.
• Contribution: $10,000. New total $140,000. Target bonds = $42,000, so bonds need $10,000. Target stocks = $98,000, so stocks need $0.

Here the entire contribution goes to bonds and the portfolio still lands exactly at target. If stocks had run to $110,000 instead, the target bond dollars after a $10,000 contribution would be 30% of $152,000 = $45,600, a gap of $13,600, which is more than the contribution. You would put all $10,000 into bonds and either accept being modestly overweight stocks until the next contribution, or sell $3,600 of stock (preferably inside a tax-advantaged account) to close the rest.

Tax-aware rebalancing

In an IRA or 401(k), rebalancing is free of tax consequences. Sell anything, buy anything, no tax event. This is why the first rule of tax-aware rebalancing is: do your rebalancing inside tax-advantaged accounts whenever the assets are there to do it with.

Think of your accounts as one portfolio with different tax wrappers. If your target is 70/30 overall and your 401(k) holds enough bonds, you can restore the whole household allocation by trading only inside the 401(k), leaving the taxable brokerage account untouched. Nothing is realized, and the overall weights are correct.

When you must trade in a taxable account, the ranking of least to most painful:

1. Direct new contributions and dividends to the underweight asset. No tax.
2. Trade in tax-advantaged accounts to offset the taxable drift. No tax.
3. Sell lots with losses or minimal gains. Use specific lot identification, not average cost, and sell the highest-cost-basis shares. A loss can offset other gains and up to $3,000 of ordinary income per year.
4. Sell long-term gains (held over a year), taxed at 0%, 15%, or 20% federal depending on income, plus the 3.8% net investment income tax at higher incomes and any state tax.
5. Sell short-term gains (held a year or less), taxed at ordinary income rates. Avoid if at all possible; waiting a few weeks to cross the one-year line is often worth far more than the precision of hitting your target on a particular date.

Worked tax comparison. You need to move $20,000 from stocks to bonds, and the shares carry a $6,000 embedded gain. In the 24% federal bracket, with a 15% long-term capital gains rate and a 5% state tax:

• Sold as a long-term gain: $6,000 x (15% + 5%) = $1,200 in tax.
• Sold as a short-term gain: $6,000 x (24% + 5%) = $1,740 in tax.
• Rebalanced inside the 401(k) instead: $0.

That $1,200 is roughly 1% of a $120,000 portfolio, a real cost against a risk benefit that is worth perhaps a few tenths of a point per year. It is a genuine reason to prefer wider bands, less frequent checks, and cash flow methods in taxable accounts, and to be relaxed about small drift there.

Two smaller mechanics worth knowing. The wash sale rule disallows a loss if you buy a "substantially identical" security within 30 days before or after the sale, and it applies across all your accounts including IRAs, so switch to a genuinely different index if you are harvesting a loss while rebalancing. And charitable giving is a quiet rebalancing tool: donating appreciated shares held over a year to a donor-advised fund or charity lets you shed an overweight position, skip the capital gains entirely, and take a deduction if you itemize.

Rebalancing while your target itself changes

Most people do not keep the same target forever. A glide path moves the allocation more conservative with age, and that raises a question: are you rebalancing to the old target or the new one?

The clean answer is to update the target on a schedule, typically once a year, and then rebalance to the new number. If you are moving from 70/30 to 68/32 this year, do not run two separate operations. Compute the target dollars at 68/32 and trade to those, using contributions first.

An age-based glide has a pleasant side effect during bull markets: because your target equity share is falling while equities are rising, ordinary contributions do a lot of the de-risking for you. A target date fund automates all of this internally, which is the honest argument for owning one: it is a rebalancing and glide path service wrapped in a single ticker, and it never gets bored or scared. The cost is less control and, in a taxable account, no ability to place bonds and stocks in the tax-optimal locations.

Mechanics: order of operations

A repeatable annual routine, done in maybe twenty minutes:

1. Pick a date and put it on the calendar. Any date. Many people use a birthday or the first weekend of a chosen month. Avoid tying it to market events, since "I will rebalance when things settle down" means never.
2. List every account and every holding with its current value, in one spreadsheet. Household level, not account level. This step alone catches forgotten old 401(k)s and duplicate funds.
3. Compute current weights and compare to target. Note which asset classes are outside their bands.
4. If nothing is outside a band, stop. Doing nothing is a valid and common outcome. Close the spreadsheet.
5. Compute target dollars using the total including any contribution you are making now.
6. Fill gaps in the cheapest order: new money first, tax-advantaged trades second, taxable sales last and only if needed.
7. Place the trades. Use market orders for large, liquid index funds during regular hours, or limit orders if you prefer control. Mutual funds trade once daily at the closing net asset value, so a same-day stock-to-bond swap inside a fund family settles cleanly; ETF trades are instant but require you to watch the spread.
8. Write down what you did and why, one paragraph. Next year's version of you will want the record, and it discourages improvised changes.

One caution on frequency: rebalancing more often is not better. Daily or monthly rebalancing raises costs and taxes and, in trending markets, cuts winners too early. Studies of frequency have repeatedly found no reliable advantage to checking more than a few times a year for a long-term portfolio.

Common mistakes

Never rebalancing at all. The most common one by far. The portfolio slowly becomes an equity portfolio, and the owner discovers it during the crash rather than before.

Rebalancing account by account instead of household-wide. If your 401(k) is all bonds and your brokerage is all stocks, each account looks wildly off target while the household is exactly right. Manage the total.

Turning rebalancing into market timing. "I will rebalance into stocks when the market bottoms" is not rebalancing, it is forecasting. The rule works because it is mechanical. The moment you add discretion, you have replaced a system that works with a judgment that mostly does not.

Rebalancing too often. Quarterly or monthly rebalancing multiplies costs and taxes for no demonstrated benefit. Annual with bands is plenty.

Ignoring taxes in a taxable account. Short-term gains and average-cost accounting can turn a routine maintenance task into an expensive one. Specific lot identification is a setting you should turn on once and forget.

Counting your emergency fund or house as part of the allocation, inconsistently. Decide once whether cash reserves and real estate are in or out of the target, write it down, and be consistent. Otherwise your "allocation" changes meaning every time you look at it.

Abandoning the rule at the worst moment. The one time rebalancing genuinely feels impossible is deep in a bear market, when the rule says to buy more stocks with money from your shrinking bond pile. That is precisely the trade that has historically mattered most, and precisely the one people skip. If you know you will freeze, automate it or own a fund that does it for you.

This guide is educational material, not individualized financial or tax advice. Your bracket, your accounts, and your time horizon change the right answer.

Related guides: Understanding Risk, Behavioral Investing, Bonds and Fixed Income, Index Funds, Mutual Funds, and ETFs