THE MUCH-DEBATED 4% rule—which I wrote about back in July—is a popular way to think about portfolio withdrawals in retirement. But it isn’t the only way. Another approach, called the bucket system, is also worth understanding. Below is some background.
What is the bucket system? As its name suggests, an investor divides his or her portfolio into multiple containers. Each container, or bucket, is then assigned a different role.
The most popular implementation of the bucket system involves three containers: The first is earmarked for a year or two of spending and is held entirely in cash. The second is earmarked for another five to 10 years of expenses. Because of the longer time horizon, this can be invested with somewhat more risk—in high-quality bonds, for example. Finally, the third bucket is earmarked for expenses beyond 10 years. Because of that, it can be invested entirely in stocks.
What’s the purpose of all this? In constructing a portfolio, retirees face a fundamental dilemma: If a portfolio is too conservative, meaning it holds too little in stocks, inflation can erode its buying power over time. On the other hand, if it’s too aggressive, with too much in stocks, there’s the risk of a large, sudden loss.
The bucket approach is designed to help balance these risks. The first two buckets—the more conservative ones—won’t provide much long-term growth, but they protect against short-term stock market risk. That, in turn, allows the investor to take lots of risk with the third bucket. Result: The bucket system helps an investor thread the needle between too much and too little risk.
The bucket strategy is a lot like the asset-liability matching system used by insurance companies. To cover future claims, insurers also set aside funds using buckets. As I’ve noted before, this has contributed to making insurance companies very durable.
What are the benefits? The bucket system can help retirees sleep at night. This is especially true when the market drops. Retirees who have, say, a year of spending money in cash, plus another seven years in bonds, will have a much greater ability to tune out the bad financial news because they know that they can survive a multi-year stock market downturn.
Buckets can be particularly helpful in those early retirement years, when sequence-of-returns risk is greatest. Because funds are earmarked so clearly, new retirees need not worry about a nightmare scenario in which they retire on Monday and the stock market drops on Tuesday. With buckets, that kind of scenario might be unnerving, but it wouldn’t derail a new retiree’s plans.
What are the disadvantages? Critics cite two drawbacks. First, the system can be complicated to manage. If you search online for illustrations of the bucket strategy, you’ll see they involve a fair amount of maintenance. Interest and dividends need to be moved periodically from bucket Nos. 2 and 3 to bucket No. 1. A retiree also needs to monitor the market to determine which bucket to draw from and which to refill. During periods of extreme market volatility, like last year, the bucket system can send retirees scrambling as they work to maintain their buckets.
Another potential issue with the bucket system: Some studies have found that bucket portfolios, on average, might underperform. Why? A 2014 paper identified a key weakness in bucket portfolios: They don’t enforce rebalancing. Specifically, there’s nothing in the bucket system that would lead an investor to buy more stocks when the market is down. By contrast, a traditional investor, following rebalancing rules, would buy more stocks when the market drops.
Take last year. When the market fell more than 30%, a bucket investor would have simply been withdrawing from his or her cash bucket, but not necessarily adding to stocks. A traditional investor, on the other hand, would have been rebalancing and adding to stocks when they were on sale. That would have resulted in a performance boost as the market recovered. More recently, another study came to the same conclusion—that bucket portfolios may have a performance disadvantage. The difference isn’t dramatic, but it’s measurable.
On balance, is the bucket system a good idea? Yes and no. If you’re trying to decide on an asset allocation, I think the bucket idea provides an excellent place to begin. Suppose you’re heading into retirement with a $4 million portfolio and plan to withdraw $100,000 a year for expenses. Under the bucket system, you might set aside eight years of spending, or $800,000, in bonds and cash. That would translate to an asset allocation of 20% in bonds and cash ($800,000 ÷ $4 million), and hence 80% in stocks.
Would that be the right asset allocation? In my view, that would be aggressive for someone entering retirement, but it isn’t totally unreasonable. It is, at a minimum, a good starting point for discussion—and better, I think, than starting with an arbitrary allocation, such as 50% stocks-50% bonds, which doesn’t have any relationship to the investor’s spending needs. That’s why, in setting asset allocations, I always start with a calculation like this. In that sense, buckets are a useful idea.
Still, I’m not sure I’d 100% follow the bucket system. That’s because the ongoing maintenance of buckets takes work. Instead of managing a single portfolio, a bucket investor effectively needs to manage three. It’s much simpler, in my opinion, to manage a portfolio the traditional way—with assigned percentages for each asset class. In addition, I see a lot of value in rebalancing and agree with the research cited above, which calls this out as a weakness of the bucket system.
For these reasons, I don’t recommend strict adherence to the bucket system. But I don’t want to be too critical. Conceptually, I see a lot of value in the way the bucket system uses an investor’s spending needs as the all-important yardstick. For risk management, I see no better way to assess a portfolio than to measure an investor’s withdrawal needs against a portfolio’s holdings of bonds and cash.
The upshot: I recommend a hybrid approach. Use bucket calculations to help settle on an asset allocation, but then translate those calculations into percentages that you use to monitor and rebalance your portfolio. That way, you can enjoy the peace-of-mind benefit of buckets, while also benefitting from the simplicity and potential performance benefit of a traditionally managed portfolio.
Adam M. Grossman is the founder of Mayport, a fixed-fee wealth management firm. Sign up for Adam’s Daily Ideas email, follow him on Twitter @AdamMGrossman and check out his earlier articles.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
The buckets are also important to tax planning. If you have an IRA, keep the cash and bonds there. It’s taxed as ordinary income. If you have a ROTH, keep the high growth stocks there. If no Roth, then in a discount brokerage. You’ll minimize the taxes on the IRA, and pay no tax on the ROTH (which will make up the bulk of your gains), or at least, low dividend and cap gains rates on stocks in the brokerage..
@Elizabeth Adams It sounds okay to me, but I wonder if it might be possible for you to run through the cash in your portfolio before the equity portion recovers to the point of your net worth clawing its way back to the 90% point. It’s worked easily in the last 5 years, but you may want to check more infamous periods of investing history (20s, 00s, for example).
I retired 5 years ago at 51. I put about 7 years of my spending in cash and multiple small CD’s*. The rest of my money is invested in low cost index funds or ETF’s and residential real estate (my primary home and a winter home in the desert to be sold/downsized if needed).
Here’s my rule: So long as my net worth is within 90% of it’s all time high, I sell index funds to fund my life. When my net worth falls below that 90% threshold, I’ll use cash and CD’s to fund my life.
Two of the five years, the market seemed frothy in January so I sold index funds for my entire year’s spending. This allowed me to avoid a couple downturns so I’ve lived entirely on my stock portfolio during my retirement.
This seems like the application of the bucket system, but with the sale rule, I’m selling stocks when they’re high and not when they’re low. I’m curious to get others’ thoughts on my method.
*Multiple, small cd’s so the penalty for early withdrawal is minimal if I need to break the CD. I don’t believe bonds fairly compensate me for interest rate risk now so I don’t have any bonds.
I rebalance once a year at the end of December if my portfolio is more than 5% off it’s target. In the 2020 covid panic the S&P recovered its February-March losses by August so there was no need to rebalance by buying more stocks.
Another really good column from Adam Grossman.
There’s a defined difference between a threat and a risk:
If someone were to classify running out of money during retirement as a potential harm, it allows an interesting conversation to begin about both the likeliness of it happening and the factors (threat agents) that can cause it to happen.
Of all the possible threat agents, there’s one that seems to me more likely than any other: Spending that’s unmonitored. That’s because unmonitored spending is just a wink away from uncontrolled spending.
Yes! Learn about decumulation methods. (Thank you, Adam Grossman!) Choose the right one for your circumstance. Just be aware that methodologies of decumulating come with NO guarantee against running out of money in retirement.
If you want to read more about the bucket system, along with many real life examples, hop over to the Morningstar site and read about them in some columns written by Christine Benz.
I second the usefulness of Christine Benz articles on Morningstar. It was her talking about buckets and risk capacity that I came to a similar hybrid approach. Use the cash flow timeline from the bucket approach, adjust for your risk capacity, tweak for your risk tolerance based on historical returns over rolling time periods to reach my AA. Then just use a 3-fund portfolio with 5/25 rebalancing bands. Simple to maintain.
Thanks, the hybrid system you describe seems practical and easy to implement. One question for some might be that the RMD requirement sets a minimum amount to account for in the first bucket. The market has been generous lately! Do you know how the IRS treats the possibility of funding a fixed annuity with retirement 401K/403b/IRA funds in order to reduce the totals and help lower the RMD? Perhaps the annuity could then contribute towards the immediate bucket, leaving more rebalancing flexibility in the investment buckets?
The question you asked…
Please ask a tax advisor who knows the specifics of your situation…
And after a good conversation about federal taxes, then you can ask about state tax consequences.
I’m thinking having these buckets in qualified plans presents a problem dealing with RMDs or at least more complexity.