What Is Portfolio Rebalancing? Why Use New Contributions

Rebalancing is a return trip to your own plan. You pick a target mix of asset classes, markets move, and your actual mix changes. Rebalancing closes the gap.

The SEC's investor education site explains it this way: over time some investments grow faster than others, which can pull your holdings out of line with your goals and change the risk of your portfolio. Vanguard adds that the point is to manage risk, not to maximize returns, and that rebalancing "isn't about market-timing."

The target itself is a personal decision. The SEC says the best allocation changes over your life, depending on your time horizon and your risk tolerance. Rebalancing doesn't choose that target for you. It just keeps you near it.

Assets earn different returns, so the weights change even if you never place a trade. Two examples:

• The SEC describes a portfolio that starts at 60% stocks and rises to 80% after market gains.
• Vanguard describes a 70/30 stock and bond portfolio that drifts to 76% stocks and 24% bonds. Its example trigger for acting is a gap of 5 percentage points.

Drift can also build for years. In a Vanguard note on target-date funds, $100 invested in U.S. stocks ten years earlier had grown to $334 by the end of 2024, compared with $160 for non-U.S. stocks. Past performance doesn't predict the next decade, but the numbers show how far two parts of a portfolio can move apart.

There are two paths:

New money can arrive all at once, such as a bonus or cash sitting in your account. Or it can arrive over time, through your regular contributions: you send more of each contribution to the underweight categories until the mix is back. If you invest every month, that's the second case, and it's the focus of this post.

The same idea works for money your portfolio produces. Vanguard suggests sending dividends and interest to the underweight asset classes. When you need to withdraw money, take it from the overweight ones first. Bernstein, an asset manager, calls this "soft rebalancing" and describes it as a "free" way to rebalance, because the portfolio moves back toward its target "without requiring additional trade."

Because it avoids the costs of selling. Fidelity notes that rebalancing in a brokerage account can trigger tax consequences if you sell part of a position. It also warns that rebalancing too often can bring unnecessary costs in a taxable account, or when you pay transaction fees. The SEC and FINRA suggest checking whether a sale will trigger capital gains or other taxes before you sell.

Contributions skip that step. Fidelity says rebalancing with new contributions might bring a portfolio back on track without selling any position, which can help avoid triggering taxes in a brokerage account.

There is a limit. Contributions work gradually. If your drift is large and your contributions are small, closing the gap can take a long time, as the example below shows. Bernstein adds that rebalancing only part of the way back can also be efficient in taxable accounts.

Selling can also make sense for reasons that have nothing to do with rebalancing, such as a holding that has grown far past its target, a change in your view of it, or a need for cash. Contributions are the default and the cheaper route, not an absolute rule.

Worked example: sell or contribute? (illustrative)

This is a made-up portfolio to show the arithmetic. It's not a forecast and not a recommendation.

Target: 60% stocks, 40% bonds. After a strong year for stocks, the portfolio holds:

• Stocks: $33,000
• Bonds: $17,000
• Total: $50,000, so stocks are $33,000 ÷ $50,000 = 66% and bonds are 34%.

Option A: sell. Move $3,000 from stocks to bonds. Stocks become $30,000 (60%) and bonds become $20,000 (40%). You sold $3,000 of an appreciated holding, which may create a taxable gain in a taxable account.

Option B: contribute $5,000 and put all of it in bonds. Bonds become $17,000 + $5,000 = $22,000. The total is $55,000. Stocks are $33,000 ÷ $55,000 = 60%, and bonds are 40%. Nothing was sold.

Option C: contribute only $2,000, all to bonds. Bonds become $19,000, the total is $52,000, and stocks are $33,000 ÷ $52,000 = 63.5%. Better than 66%, but not yet 60%.

Option D: contribute $2,000 and split it 60/40 as usual. Stocks get $1,200 and bonds get $800. Stocks become $34,200 out of $52,000, or 65.8%. The mix barely moves, because the new money kept feeding the part that was already too big.

Option D is the lesson that matters most. New money only rebalances the portfolio if you steer it to the underweight side instead of splitting it by the target percentages.

There's no single best answer, but most guidance agrees on the basics.

• The SEC mentions two common approaches: a calendar (every six or 12 months) or a threshold (when an asset class moves more than a set percentage). It adds that rebalancing tends to work best when done relatively infrequently.
• Fidelity lists three: calendar ("like once a year"), threshold ("like anytime a position deviates by 5 percentage points or more from your target") and a hybrid of the two. It warns against tinkering with the portfolio too much.
• Vanguard says that for many investors an annual rebalance is optimal. If that doesn't fit your schedule, pick one that's easy to follow and stick with it.
• Bernstein's 2026 analysis argues that acting when weights leave a band gives a better signal than a fixed calendar.

The fact that guidance splits between calendar and band suggests either one is reasonable. What everyone agrees on is not doing it constantly.

It doesn't promise higher returns. Its job is to manage risk, not to maximize returns. It doesn't replace choosing a sensible target. And if you own a target-date fund, you may not need to do it yourself, because the fund's adviser handles rebalancing over time.

When you hold several funds and add money monthly, the work is the arithmetic: how far each holding is from its target, and how to split the next contribution so the most underweight holdings get filled first. A spreadsheet can do that. Allocta is a contribution calculator that does this buy-only math and never suggests selling, so you can check the result against your own plan.