Resetting portfolio exposures periodically can help minimize the pain of taxes.

A common issue facing investors is whether, after enjoying gains in stock holdings, they should do some trimming and put the proceeds in bonds and other assets. Investors are often quite reluctant for a couple reasons: the “loss” to taxes and the pain that goes with it. It’s very unpleasant to pay taxes, so why pay more than you absolutely have to?

The problem is that a decision not to pay taxes at a given time may have investment consequences later on. Specifically, selling is part of the reallocation process through which to limit volatility and move portfolios back into balance. Over time, neglecting it can gradually elevate risk, which may worsen results in a market downturn.

As a result, despite the pain, we generally favor regular, tax-aware reallocation of portfolios. The bright side is that this gradual process can limit current outlays (although the long-term aggregate may be comparable) and reinforce an understanding of taxes as a carrying cost for your portfolio that typically gets imposed, one way or another.

Establishing Your Framework

The approach you take to rebalance your portfolio may vary. It may be a function of time (for example, annually) or of the portfolio’s deviation from your preferred long-term allocation (how much you let particular asset classes shift from the strategic baseline). The rebalancing may combine these two ideas. How frequently you reallocate may also depend on transaction costs (including time spent) and comfort level as you move further from your original target.

When talking with clients, we like to remind them of a couple of points:

First, taxes are paid on realized gains. So, rather than thinking of them as out-of-pocket losses, consider them a haircut to those gains. The display below illustrates this point. It assumes that a stock advances a moderate 5%, while generating a 2% yield, over the course of just over a year before being sold. The long-term capital gains and qualified dividend taxes imposed (federal only) do not even reach the level of the dividend payout in this case.1

A ‘Haircut’ From Portfolio Gains

Hypothetical Return and Tax Consequences

Portfolio Value ($)

Source: Bloomberg, Neuberger Berman. For illustrative purposes only. See assumptions and disclosures in footnote 1

Second, it’s useful to consider taxes as the price of risk mitigation against worse-than-expected portfolio volatility. The longer you go without a rebalancing, the greater the risk that your portfolio will drift further away from your baseline allocation. As noted, this changes the risk profile of your portfolio, often without your realizing it—particularly in cases of substantial return differences between asset classes.

To illustrate, we simulated a hypothetical portfolio with an initial value of $1 million, allocated 60% to equities (the S&P 500) and 40% to bonds (the Bloomberg Barclays U.S. Aggregate Index). We then modeled the returns, dividends and bond income each year to mirror actual index results from 2009 – 2018. In Scenario 1, we assumed a buy-and-hold strategy was maintained for 10 years, followed by a single, taxable rebalancing at the end of the period. In Scenario 2, the portfolio was rebalanced back to its original 60/40 mix after each year. In both cases, we assumed that the investor withdrew all dividends and coupon payments annually for income—a common scenario for retirees. The result? Over the 10-year period, which saw exceptional equity performance, the Scenario 1 portfolio’s asset allocation shifted to 81% stocks and just 19% bonds.2 In other words, it moved from a moderate to near-aggressive investment profile, drastically increasing its vulnerability to market volatility.

Indeed, such an investment mix would not have felt particularly good during last December’s sharp equities downdraft. And although markets have since enjoyed a substantial rebound, what if we had seen another period like 2008? In that year, stocks fell by 37% (total return); and while the hypothetical targeted 60/40 portfolio would have lost 24%, an 81/19 portfolio would have declined by 33%.

Keep in mind that investors usually pick a strategic asset allocation for valid reasons. A less aggressive investor profile, for example, may indicate the approach of retirement, or potential outlays for education. Participation in 80% of a market drawdown may hinder such goals, particularly if stocks underperform and the investor’s time horizon is not one that can tolerate prolonged (multiyear) recovery periods. An initial moderate or moderately conservative allocation may also reflect discomfort with a higher level of risk. The panic sometimes associated with market declines may prompt such an investor to “pull the plug” on stocks at exactly the wrong time. To us, it may be well worth the “cost” of regular tax payments to keep an investment program on track and help avoid emotional decisions during market declines.

On the Other Hand…

To play devil’s advocate, however, there are certain advantages to not paying your taxes as you go: First, assuming that you ordinarily pay taxes from noninvestment accounts, you will have somewhat more money at your disposal if you postpone the outlays, which could support your lifestyle or contribute to other ongoing needs. Second, if the taxes are to come from your investment accounts, by deferring them you will avoid removing assets that might otherwise contribute to potential compounded returns. Third, there’s always a chance that the asset class from which you plan to rebalance will outperform in subsequent years.

Let’s look at points two and three in more detail. The table below lays out the results of the 10-year simulation discussed earlier. As you can see, the Scenario 1 portfolio (single rebalance after 10 years) outperformed the Scenario 2 (annually rebalanced) portfolio, for two reasons. First, deferred taxes allowed more assets to stay in the portfolio and earn a return. Second, the drift to an 81% equities/19% bonds allocation created momentum for further outperformance: in a decade where equities saw 13% annualized returns, aggressive portfolios would have been disproportionately rewarded.

‘Buy-and-Hold’ vs. Rebalancing: A Hypothetical Comparison

  Scenario 1
(Single Rebalance After 10 Years)
Scenario 2
(Rebalanced Annually)
Starting portfolio value $1,000,000 $1,000,000 $0
Ending portfolio value $1,986,156 $1,789,902 $196,254
Annualized return 7.1% 6.0% 1.1%
Capital gains taxes paid $123,760 $113,109 $10,651
Total taxes paid
(dividends, bond coupons, capital gains)
$236,940 $233,501 $3,439

Source: Neuberger Berman. See assumptions and disclosures in footnote 2.

We would note, however, that the next decade may be unlikely to provide similarly compelling results, which could reduce the advantage to portfolios heavily weighted in stocks. Moreover, it’s possible that there will be episodes of significant (even if temporary) market weakness. Looking again at our hypothetical scenario, if we assume that the buy-and-hold investor did not rebalance after the 10-year period, and then saw declines equal to those suffered in 2008, the buy-and-hold portfolio would have experienced a loss of $687,458 versus $437,430 on the rebalanced portfolio, a difference of about $250,000 that would have more than offset the roughly $200,000 advantage previously enjoyed by the buy-and-hold portfolio.

With all this in mind, a fundamental question is whether the difference in return potential tied to retaining a more aggressive portfolio is worth the risk, especially if your investment horizon is too short to achieve recovery later on.


Circling back to where we started, it’s clear that much of the reluctance to pay taxes on portfolios is tied to aversion to incurring outflows. However, taxes aren’t optional. Generally, the only escape from capital gains taxes on realized gains comes if you hold your investments until death, at which point there is a step-up in tax cost basis that benefits your heirs. In other cases, you have some control over when you pay taxes but, in general, eventually they will come due.

Looking again at our simulation, the choice of “paying as you go” translates into annual capital gains outlays to Uncle Sam of about $11,300 over the 10-year period. However, a decision to pay at the end results in a one-time cost of about $124,000, amounting to “tearing off the Band-Aid” versus gently pulling at it over time. Many who have already accumulated gains may need to be more aggressive in reallocating from stocks, but they can still take a gradual approach to make it easier to tolerate. Along the way, tax-loss harvesting can often mitigate tax costs by offsetting realized gains. All told, the tax question is about choice, with a clear understanding of the risks of holding onto appreciated assets too long based on tax considerations.

‘Tactical Tilts’ and Taxes

Many investors establish a strategic, or long-term, asset allocation to guide portfolio structure over time. However, where it is feasible, we also favor the use of tactical tilts, or shorter-term over- and under-weightings of different asset classes to capitalize on current market conditions and/or limit potential risks. In the context of this article, it’s worth considering how taxes may affect tilting decisions, and in what circumstances it can make sense to move forward with such portfolio shifts.

Return potential is at the core of such an assessment. The asset to which you are considering reallocating must not simply outperform the one you would be trimming, but do so by enough to offset the tax bill you incur on any gains you realize. Another element is confidence: How certain are you that the new asset can actually outperform (or the existing asset underperform) by the amount you assume?

The table below may help to clarify the issues. For those who hold “Asset A,” it shows how much outperformance “Asset B” must achieve over the next year to make up for the tax cost (paid from the portfolio) of the sale of Asset A. The lower the cost basis on Asset A as a percentage of its current price (see vertical axis), the higher the tax on the sale, and the better the substitute, Asset B, must perform.

Interestingly, the longer you wait, the higher the cost of a tactical shift may become, due to an increasing gap between the cost basis of Asset A and its market price. Thus, it may be harder to find compelling substitutes for current assets—something that could ultimately hurt portfolio performance. For example, hypothetically, let’s say you believe that Asset A could generate an 8% return next year, while Asset B could generate a 12% return. If the cost basis in Asset A were 90% of today’s price, a return on Asset B above 10.2% would make the trade worthwhile. But if your cost basis were 30%, your return on B would need to be at least 29.6%. In the latter case, all things equal, you would continue to hold Asset A to your detriment relative to a portfolio including Asset B.3

When Does Reallocation Overcome Tax Cost?

Cost Basis, % of Current Price Hypothetical Return: Asset Class A
  0% 2% 4% 6% 8% 10% 12% 14%
Breakeven Annualized Return: Asset Class B
100 0.0% 2.0% 4.0%6.0% 8.0%10.0% 12.0% 14.0%
90 2.4% 4.5% 6.5% 8.6% 10.6% 12.7% 14.7% 16.8%
80 5.0% 7.1% 9.2% 11.3% 13.4%15.5%17.6%19.7%
70 7.7% 9.8% 12.0%14.2%16.3%18.5%20.6%22.8%
60 10.5% 12.7% 14.9%17.2%19.4%21.6%23.8%26.0%
50 13.5% 15.8% 18.0%20.3%22.6% 24.9% 27.1% 29.4%
40 16.7% 19.0% 21.3%23.7%26.0% 28.3% 30.7%33.0%
30 20.0% 22.4% 24.8%27.2%29.6%32.0%34.4%36.8%
20 23.5% 26.0% 28.5%30.9%33.4% 35.9% 38.3%40.8%
10 27.3% 29.8% 32.3%34.9%37.4% 40.0%42.5%45.1%
0 31.2% 33.9% 36.5%39.1% 41.7%44.4%47.0%49.6%

Source: Neuberger Berman. For illustrative purposes only. See assumptions and disclosures in footnote 3.