Portfolio Rebalancing — When and How to Rebalance

Photo: Alesia Kozik / Pexels
You pick a target asset allocation — say, 60% stocks, 40% bonds — because it matches your risk tolerance and goals. A few years go by. Stocks rip; bonds drift sideways. You check your account and you're now 75% stocks, 25% bonds. Nobody made that decision on purpose. The market made it for you, one green day at a time, and now your portfolio is carrying more risk than you originally signed up for.
Educational only — not financial advice. Investing involves risk of loss; past performance does not guarantee future results. This article explains mechanisms and published research, not personalized recommendations. Consult a licensed financial advisor before making investment decisions.
Rebalancing is the discipline of selling what's overweight and buying what's underweight to return a portfolio to its target mix. On paper, it's a spreadsheet exercise — check some percentages, place a few trades. In practice, it's one of the only tools in investing that forces you to do the psychologically hard thing (sell what's been winning, buy what's been losing) on a schedule, instead of on a whim. This piece walks through why drift happens, the three main rebalancing methods, what Vanguard's and other institutions' research actually found when they tested them against decades of market data, how taxes change the math, and where the "just don't rebalance" argument has a point.
Why portfolios drift in the first place
Asset classes don't grow at the same rate. Stocks are more volatile than bonds, which means over any stretch where stocks outperform, their share of the portfolio grows disproportionately, not just their dollar value. A portfolio that starts 60% stocks and 40% bonds doesn't need a crash to become risky. It just needs stocks to keep winning for long enough, quietly, without anyone touching a thing.
This matters because your original 60/40 split wasn't arbitrary. It was calibrated (ideally) to a drawdown you could tolerate without panic-selling. Modern portfolio theory, formalized by Harry Markowitz in his 1952 paper "Portfolio Selection" in the Journal of Finance, treats a portfolio's risk as a function of how its assets are weighted and how they move relative to each other. Change the weights and you change the risk profile, even if you never traded a single underlying holding. If diversification math is new to you, why diversification reduces risk walks through the covariance mechanics that make an 80/20 portfolio behave very differently from a 60/40 one in a downturn.
There are three distinct reasons to fight that drift, and it helps to rank them, because people often reach for the third one (returns) when the first one (risk) is the actual job:
- Risk control. Allocation drift is unintended risk. A 60/40 portfolio that drifts to 80/20 has a materially different drawdown profile in a bear market — deeper losses, longer recovery, more temptation to sell at the bottom.
- Forced discipline. Rebalancing makes you sell high and buy low on a schedule, working against the emotional grain that usually pushes investors to do the opposite (chase what's already gone up, dump what's already gone down).
- A modest return contribution. Academic work on the "rebalancing bonus" or "volatility pumping" effect — going back to research popularized by William Bernstein and studied formally by Erb, Harvey, and Viskanta in commodity-futures contexts — suggests rebalancing can add a small amount of return over long periods by systematically buying dips and trimming rallies between negatively or weakly correlated assets. The effect is real but modest; most institutional research puts it in the tenths of a percentage point per year, not multiple points.
What rebalancing does not reliably do is transform a mediocre portfolio into a great one. The long-run expected return of a 60/40 portfolio is roughly the weighted average of its components' expected returns, regardless of how aggressively you rebalance it. The job is risk management first, with a small, secondary return tailwind. Investors who rebalance expecting it to meaningfully outperform a static allocation are usually disappointed; investors who rebalance to stay inside their risk budget usually aren't.
The three main rebalancing methods
There is no single "correct" rebalancing rule. What exists is a set of trade-offs between responsiveness, simplicity, trading costs, and taxes. Three approaches cover almost every real-world implementation, from a DIY brokerage account to a $1 trillion pension fund.
1. Calendar rebalancing
Rebalance on a fixed schedule — monthly, quarterly, semi-annually, or annually — regardless of how far the portfolio has drifted from target at that moment.
Calendar rebalancing is the simplest rule to follow because it requires no monitoring between check-ins. Set a reminder for the first week of January, look at your allocation, trade back to target, done. The weakness is that it's blind to actual market conditions: it might trigger a trade when drift is trivial (say, one percentage point off target, not worth the transaction cost or tax hit), or it might miss a sharp move that happens mid-quarter and reverses before the next scheduled check.
Pros: simple, automatable, no ongoing monitoring required, easy to explain to a spouse or co-investor. Cons: can trigger unnecessary small trades, can miss large intra-period swings entirely.
2. Threshold (band) rebalancing
Rebalance whenever an asset class deviates from its target by more than a defined band, checked continuously or on a regular monitoring cadence (weekly, monthly). Two common flavors exist:
- Absolute bands — for example, "rebalance if any asset class drifts more than 5 percentage points from target." For a 60% stock target, a 5-point absolute band triggers at 55% or 65%.
- Relative bands — for example, "rebalance if any asset class drifts more than 25% of its own target weight." For that same 60% stock target, a 25% relative band triggers at 45% or 75% — a much wider gap in absolute terms. Relative bands matter more for small allocations: a 5% target allocation to emerging markets moving to 5.5% is a trivial 0.5-point absolute drift, but it's a 10% relative move, which a relative-band rule would catch and an absolute-band rule would ignore.
Pros: responsive to actual market moves, avoids rebalancing when drift is immaterial, catches large swings that calendar rebalancing might miss between check-ins. Cons: requires ongoing monitoring (or automation to do that monitoring for you), can trigger a cluster of small trades during volatile stretches if bands are set too tight.
3. Hybrid (calendar + threshold)
Check the portfolio on a schedule — commonly annually — and rebalance only if a band has been breached at that check-in. This is the approach Vanguard's research group has repeatedly recommended, and it's effectively what most target-date funds and robo-advisors do under the hood: periodic review, conditional action.
The logic is straightforward. Checking too often invites overtrading; checking too rarely lets risk drift for too long unmonitored. An annual check with, say, a 5-percentage-point band captures nearly all of the risk-control benefit of constant monitoring while keeping the number of actual trades — and the taxes and costs that come with them — close to the minimum necessary.
How often should you rebalance? (AEO)
For most individual investors, checking once a year and rebalancing only if an asset class has drifted more than about 5 percentage points from target captures nearly all the benefit of more frequent rebalancing at a fraction of the trading and tax cost, according to Vanguard's long-run backtests.
What the research actually says
The most-cited institutional research on this question comes from Vanguard, which has run this analysis multiple times over more than a decade — first in a 2010 paper, and again in an October 2022 study titled "Rational Rebalancing: An Analytical Approach to Multiasset Portfolio Rebalancing Decisions," published by Vanguard's Investment Strategy Group (source PDF). Vanguard also published a companion piece, "The Rebalancing Edge: Optimizing Target-Date Fund Rebalancing Through Threshold-Based Strategies" (source PDF), which applies the same framework specifically to the target-date funds tens of millions of 401(k) participants already hold.
The core findings, distilled:
- Very frequent rebalancing (monthly or quarterly) did not improve risk-adjusted outcomes compared with less frequent rebalancing, once realistic transaction costs were included in the simulation.
- Extremely infrequent rebalancing (every two years or longer) also underperformed a moderate schedule, because it let risk drift too far between checks.
- The efficient zone sits in between — annual to semi-annual checks, often paired with a threshold band, rather than either extreme.
- Transaction costs rise during volatile periods, which is precisely when a naive "always rebalance immediately" rule tends to want to trade the most. Vanguard's model explicitly weighs the utility of staying on target against the drag of trading into a wide bid-ask spread or a stressed market.
- Tax- and cost-sensitivity should be built into the rule itself, not bolted on afterward — meaning the "right" frequency is genuinely different for a taxable account than for a 401(k) or IRA holding the same assets.
An earlier, frequently cited Vanguard paper by Colleen Jaconetti, Francis Kinniry, and Yan Zilbering, "Best Practices for Portfolio Rebalancing," reached a compatible conclusion using a different methodology: it found that wider bands (around 5%) captured nearly all of the risk-control benefit that narrower bands (around 1%) provided, while generating meaningfully fewer trades — because a 1% band gets breached constantly in ordinary market noise, not just in genuine allocation drift.
None of this research claims rebalancing is a free lunch. The consistent message across both papers is that the quality of the rule matters less than avoiding the two failure modes at either extreme: trading so often that costs eat the benefit, or trading so rarely that risk drifts uncontrolled for years.
Table 1. Rebalancing methods compared
| Strategy | Typical trades per year | Best suited for | Tax efficiency | Monitoring required |
|---|---|---|---|---|
| Calendar — annual | 1–2 | Set-and-forget investors, simple 2–3 fund portfolios | Good | None between check-ins |
| Calendar — quarterly | 4 | Investors who want a fixed routine, active monitors | Worse (more turnover, more short-term-gain risk) | None between check-ins |
| Threshold, 5% absolute | 0–4, varies with market volatility | Risk-focused investors comfortable checking periodically | Good (trades only fire when needed) | Periodic (monthly/quarterly check) |
| Threshold, 25% relative | 0–6, varies | Portfolios with several small allocations (e.g., sector or factor tilts) | Medium | Periodic |
| Hybrid — annual + 5% band | 0–2 | Most individual investors | Best among active approaches | Annual |
| No rebalancing | 0 | Long horizon, tax-sensitive, comfortable with drifting risk | Best on taxes — but risk drifts unchecked | None |
Trade counts are illustrative ranges drawn from Vanguard's published backtests and common industry implementation; actual counts depend on your specific asset mix, band width, and market conditions in a given year.
Tax-aware rebalancing: the moves that actually save money
In a tax-advantaged account — a 401(k), traditional or Roth IRA, or HSA — rebalancing is essentially free from a tax standpoint. Sell whatever's overweight, buy whatever's underweight, no capital gains event triggers. This is why financial planners routinely tell clients to do the bulk of their rebalancing inside these accounts first, before touching a taxable brokerage account at all.
In a taxable account, selling an appreciated position realizes a capital gain, and the IRS taxes it based on how long you held the asset. For 2026, long-term capital gains (positions held more than a year) are taxed at 0%, 15%, or 20% depending on taxable income — for single filers, the 0% bracket runs up to roughly $49,450 of taxable income and the 20% bracket kicks in above roughly $545,500, with married-filing-jointly thresholds roughly double those figures, according to IRS-derived figures reported by CNBC and Kiplinger (CNBC, Kiplinger). Short-term gains, on positions held a year or less, are taxed as ordinary income, which is why reflexive quarterly rebalancing in a taxable account can be considerably more expensive than it looks on a spreadsheet.
The practical tax-aware moves, roughly in order of how often planners recommend them:
- Rebalance with new contributions first. Direct fresh paycheck deposits, bonus money, or any new cash into whatever's currently underweight. This nudges the allocation back toward target without selling anything, so there's no capital gains event at all.
- Rebalance inside tax-advantaged accounts before touching taxable ones. If your overall household allocation spans a 401(k), an IRA, and a taxable brokerage account, do as much of the actual buying and selling as possible inside the accounts where trades don't trigger tax.
- Use dividends and interest as rebalancing fuel. Set distributions to pay out as cash rather than automatically reinvesting, then use that cash to top up whatever's underweight at your next check-in.
- Tax-loss harvest while you're at it. If a piece of the portfolio is sitting at a loss, selling it both moves you toward target and generates a realized loss you can use to offset gains elsewhere — subject to the IRS wash-sale rule, which disallows the loss if you buy a "substantially identical" security within 30 days before or after the sale (a 61-day window in total). The disallowed loss isn't gone forever; it gets added to the cost basis of the replacement shares, per IRC §1091 (IRS code text).
- For a heavily appreciated taxable position, consider "never sell, only buy." If a taxable account holds a large embedded capital gain, the tax cost of trimming it back to target can outweigh the risk-control benefit. Many planners will let that position ride and instead direct all new money — and all rebalancing activity in tax-advantaged accounts — toward correcting the overall household allocation, leaving the appreciated taxable position untouched.
This is also the space where robo-advisors have built an actual product edge. Wealthfront runs tax-loss harvesting on a daily basis rather than the quarterly or annual cadence typical of a DIY investor, and above a $100,000 taxable balance it offers direct indexing — buying the individual stocks that make up an index like the S&P 500 instead of a single ETF share, which multiplies the number of individual lots available to harvest for losses (Wealthfront, Wealthfront support). Betterment offers comparable automated tax-loss harvesting on its standard accounts. The mechanism behind both is the same wash-sale-aware harvesting described above, just automated and run far more frequently than most individuals would bother doing by hand.
The "don't rebalance at all" argument
There's a minority position among long-horizon investors, and it has real logic behind it: let your winners run. If a stock-heavy allocation drifts upward over decades because stocks outperform bonds, and stocks have historically outperformed bonds over long horizons, then a portfolio that never rebalances ends up with a larger allocation to the asset class that produced the higher return — and, mechanically, a higher terminal balance in the scenarios where that outperformance continues.
Two things complicate that logic. First, "stocks have historically outperformed bonds" is a statement about the past, not a guarantee about your specific investing horizon; there have been multi-decade stretches (Japan from 1990 onward is the standard example) where a domestic stock-heavy allocation badly lagged more balanced alternatives. Second, and more directly relevant to most people's actual behavior: an unrebalanced portfolio has taken on more risk than the investor originally signed up for, and the investor most likely to abandon a long-term plan is the one who gets surprised by a drawdown they weren't prepared for. A 60/40 investor who let the portfolio drift to 85/15 over a bull run isn't just carrying more risk on paper — they're carrying more risk than their own stated tolerance, discovered at the worst possible moment, in the middle of the next bear market.
The compromise most practitioners land on, including target-date fund managers, is not "never rebalance" but "rebalance with wider bands, and let the bands widen further as the investor's time horizon shortens and their capacity for volatility declines." A 25-year-old with decades until retirement can tolerate a wider band and more drift than a 60-year-old five years from drawing down the account, because the younger investor has time to recover from a bad sequence of returns and the older one may not.
Rebalancing during volatility: the counterintuitive part
In a sharp, fast crash — the March 2020 COVID-19 drawdown is the standard reference point, when the S&P 500 fell roughly 34% in about a month before recovering most of that loss within the same year — mechanical rebalancing tells you to do something that feels wrong in the moment: sell bonds (which held up or gained) and buy stocks (which had just cratered). That's exactly backward from what fear tells most people to do.
This is precisely where the discipline pays off, and precisely where most individual investors fail to execute. Investors who mechanically rebalanced into that 2020 drawdown — buying equities near the bottom because their bands had been breached — captured a larger share of the subsequent V-shaped recovery than investors who froze, held cash on the sidelines, or waited for "things to feel safe again" before re-entering. The mechanism isn't prediction; nobody rebalancing in March 2020 knew the recovery would be V-shaped rather than prolonged. The mechanism is that a pre-committed rule removes the need to predict anything at all — you're following a decision made calmly, months or years earlier, instead of one made under acute stress.
This is also the strongest practical argument for automated rebalancing. A human staring at a 30% portfolio decline has to overcome a real, evolutionarily wired impulse to protect what's left rather than buy more of what just fell. A robo-advisor or target-date fund doesn't have that impulse. It executes the rule.
Who does this automatically, and how
Not every investor wants to run their own rebalancing calendar. Several categories of product handle it as part of what you're paying for:
- Target-date funds (offered by Vanguard, Fidelity, Schwab, and most 401(k) providers) automatically rebalance internally and also shift the overall stock/bond mix over time along a preset "glidepath" as the target retirement year approaches. This is the mechanism Vanguard's target-date-fund-specific rebalancing research directly studied.
- Robo-advisors like Wealthfront and Betterment monitor allocations continuously (effectively a tight threshold-band approach) and rebalance automatically, often bundling in daily tax-loss harvesting as described above.
- Full-service digital advisory products, such as Vanguard Digital Advisor or Schwab Intelligent Portfolios, apply similar threshold-based logic with additional human-advisor access at higher tiers.
- Brokerage "rebalance to target" features. Several major brokers now offer a built-in tool that compares current holdings to a saved target allocation and generates the trades needed to close the gap in one click, without requiring a full robo-advisor subscription.
If you're holding a diversified ETF portfolio built from a handful of low-cost, broad-market funds, a target-date fund or robo-advisor is usually the cheapest, most disciplined path to staying rebalanced, because the fund is doing the work you'd otherwise have to remember to do yourself. Understanding how the underlying index funds track their benchmarks also clarifies why rebalancing a two- or three-fund portfolio is mechanically so much simpler than rebalancing a portfolio of individual stocks: you're managing three or four percentages instead of dozens of individual position weights.
Costs to watch, even when a trade seems "free"
Zero-commission trading removed the most visible cost of rebalancing, but it didn't remove all of them. Before executing a rebalance, it's worth checking:
- Bid-ask spreads. Minimal for major, highly liquid ETFs (often a cent or two on funds tracking the S&P 500 or total market indexes), but wider and more costly for niche sector funds, small-cap funds, or thinly traded international ETFs.
- Commissions. Mostly $0 at major US brokers now, but not universal — always check, especially for mutual funds versus ETFs, and especially outside the US.
- Capital gains taxes. The dominant cost in a taxable account, as covered above, and the reason the "right" rebalancing frequency genuinely differs between a taxable and a tax-advantaged account holding the identical target allocation.
- Mutual fund redemption fees. Some mutual funds (though rarely ETFs) charge a short-term redemption fee, often 1% or 2%, if you sell within 30 to 90 days of purchase, specifically to discourage the kind of frequent in-and-out trading that quarterly or monthly rebalancing can resemble.
- Minimum trade sizes and lot fragmentation. Frequent small rebalances in a taxable account can leave you with many small, oddly sized tax lots, which complicates future tax-loss harvesting and record-keeping even if each individual trade was cheap.
Common mistakes people make with rebalancing
A few patterns show up repeatedly among investors who try to implement rebalancing without a clear rule:
Rebalancing on every market headline. A rule that gets overridden by "well, this time feels different" isn't really a rule — it's a suggestion. The entire behavioral value of rebalancing comes from following a pre-committed band or schedule regardless of what the news cycle is saying that week.
Rebalancing individual stocks instead of asset classes. Selling your best-performing individual stock because it's grown to "too much" of your portfolio conflates rebalancing with concentration management, which is a related but different problem, and it doesn't have the same clean theoretical justification that rebalancing across asset classes does. It also tends to generate more tax drag for less risk-control benefit.
Ignoring account location. Rebalancing the taxable account first, out of habit, when the same correction could have been made tax-free inside a 401(k) or IRA holding overlapping assets, is one of the most common and most avoidable costs investors incur.
Setting bands too tight. A 1% or 2% band gets breached by ordinary daily market noise, not genuine drift, which is exactly the overtrading problem Vanguard's research flagged. Wider bands, in the 5% range, capture the meaningful drift while ignoring the noise.
Forgetting new contributions exist. Investors who are still actively saving often reach straight for selling appreciated assets to rebalance, when directing new contributions toward the underweight asset class would accomplish the same thing without triggering any tax at all.
Frequently asked questions
Q: How often should I rebalance? A: For most investors, checking once a year and rebalancing only if an allocation has drifted more than about 5 percentage points from target captures nearly all the risk-control benefit that more frequent rebalancing provides, according to Vanguard's published backtests, at a fraction of the trading and tax cost.
Q: Should I rebalance during a market crash? A: If the crash pushes an allocation past your threshold band, yes — mechanical rebalancing into a sharp downturn has historically been one of the more reliable sources of long-run risk-adjusted benefit, precisely because it forces buying while prices are depressed. It requires real discipline, which is why many investors outsource the decision to an automated product rather than trust themselves to execute it manually under stress.
Q: What if I'm 100% stocks? A: You can still rebalance between stock sub-allocations — US versus international, large-cap versus small-cap, growth versus value, or individual sectors. The same drift-and-correct logic applies within an all-equity portfolio, just among stock categories instead of between stocks and bonds.
Q: Is rebalancing necessary if I already hold a target-date fund? A: No. Target-date funds rebalance internally and shift their overall risk level along a glidepath automatically as the target year approaches. That automation is a significant part of what you're paying the fund's expense ratio for.
Q: Does rebalancing reduce returns during a bull market? A: Often, slightly, yes — you're trimming an asset class that's currently winning to buy one that isn't. But the same rule that trims gains in a bull market is also the rule that limits losses in the following bear market. Evaluated across a full market cycle rather than a single up-leg, the trade-off tends to favor rebalancing on a risk-adjusted basis, per Vanguard's multi-decade research.
Q: Should I rebalance individual stocks or just broad asset classes? A: Broad asset classes and fund-level allocations, not individual stocks. Trimming a single winning stock because it's grown "too large" is a concentration-risk decision, not a rebalancing one, and it typically generates more tax drag without a comparably clean theoretical justification.
Q: What's the wash-sale rule and how does it affect rebalancing? A: The IRS wash-sale rule, codified at IRC §1091, disallows a tax loss if you sell a security at a loss and buy a "substantially identical" one within 30 days before or after the sale, creating a 61-day window in total. It matters during rebalancing because selling a fund at a loss to correct your allocation, then immediately buying back a near-identical fund, can forfeit the tax benefit of that loss unless you wait out the window or swap into a genuinely different fund.
Q: Do robo-advisors rebalance better than doing it manually? A: They rebalance more consistently and, in the case of providers like Wealthfront, with more granular daily tax-loss harvesting than most individual investors manage by hand. Whether that translates into a materially better outcome depends heavily on account size, tax bracket, and how disciplined the investor would otherwise have been.
Q: Is there a "best" rebalancing threshold for everyone? A: No single number is universally optimal — it depends on account type (taxable versus tax-advantaged), the number of asset classes involved, and trading costs. Vanguard's research points to roughly 5-percentage-point absolute bands checked annually as a reasonable default for a simple multi-asset portfolio, but a portfolio with many small allocations may need relative bands instead.
Q: Can rebalancing ever increase my risk instead of reducing it? A: Rarely, and mostly through implementation mistakes — for example, rebalancing so frequently in a taxable account that the tax drag outweighs the risk-control benefit, or setting bands so tight that ordinary market noise triggers constant trading. Used as intended, on a reasonable schedule with a sensible band, rebalancing reduces unintended risk rather than adding it.
The takeaway
Rebalancing isn't a return-boosting strategy in disguise. It's a risk-management rule that keeps a portfolio anchored to the allocation an investor actually chose, rather than the allocation the market happened to leave them with after a few good or bad years. The research consensus, most thoroughly documented in Vanguard's rebalancing studies, points toward a specific, fairly narrow sweet spot: check roughly once a year, act only past a meaningful threshold like 5 percentage points, and do as much of that rebalancing as possible inside tax-advantaged accounts before touching a taxable one. Everything more complicated than that — constant monitoring, tight bands, reflexive quarterly trading — tends to add cost and tax drag without adding meaningfully better outcomes. The unglamorous version of the rule is also, according to the data, close to the best one.
Read next:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

What Is FIRE? Financial Independence Retire Early
What is the FIRE movement? Financial Independence Retire Early explained — 4% rule, savings rates, leanFIRE vs fatFIRE, math, and criticisms.

Value vs Growth Investing — Which Style Wins?
Value vs growth investing compared: definitions, the Fama-French value premium, why value lagged before 2021, 2022's reversal, value traps, and style tilts.

Asset Allocation Explained — Portfolio Building
Asset allocation is how you split a portfolio across stocks, bonds, and cash. See what research shows, the main model portfolios, and how to rebalance.

Index Funds vs Active Funds — Which Wins?
Index funds vs active funds: S&P SPIVA data on 20-year results, fee gaps, the Sharpe arithmetic, and where active management can still make sense.

Stocks vs Bonds — Investing Basics Explained
Stocks vs bonds explained — risk, return, diversification, index funds, and what a century of market data say about long-term investing.

ETFs Explained — Index Funds Made Simple
How ETFs and index funds work: creation and redemption, fees compounding over 30 years, taxes, popular funds, and leveraged ETF traps, with 2025 ICI data.