My “Bucket Strategy” for Retirement Spending
I frequently get questions about the “bucket strategy” I’m using for spending my assets in retirement. I prefer not to use the term “bucket” because my strategy differs from bucket methods in important ways. In fact, my retirement spending more resembles single-portfolio strategies.
My decumulation approach involves holding 5 years’ worth of my annual spending in short-term fixed income and the rest in stocks (described in more detail in my post Cushioned Retirement Investing). Each year, I sell enough stock to replenish the fixed-income allocation.
My annual spending each year is calculated from my age and current portfolio value. As I get older, I spend a slightly higher percentage of my remaining portfolio (see the spreadsheet in my Cushioned Retirement Investing post for the exact percentages). If stocks perform well, my annual spending will rise, and if they perform poorly, my spending goes down.
Because my annual spending changes from year to year, I have to calculate how much stock to sell each year. I take my new annual spending, multiply by 5, and subtract the current value of my fixed-income investments. This is essentially a form of rebalancing that has me selling fewer stocks when they’re down and selling more when stocks are up. In rare cases when stocks crash, I might have more fixed income than 5 times my new annual spending, and I actually buy some stocks instead of selling.
Important Differences from Bucket Strategies
No active decision on which bucket to spend from
Many bucket strategies involve making an active decision about whether to spend from stocks or fixed income in a given year. This only makes sense if you believe you have the ability to time the market. I don’t. Most people (maybe all) who believe they can time the market are wrong.
No hard switches between spending from stocks and fixed income
It’s possible to devise a bucket strategy that is entirely rules-based. A simple example would be to spend for the year entirely from fixed income if stocks are more than 20% below their peak level. The idea would be to ride out stock market declines by spending from fixed income until stocks rebound. My strategy has no hard switches between spending from stocks and spending from fixed income.
Other differences from typical decumulation strategies
Predetermined spending changes based on portfolio size
Decumulation strategies often have retirees set a spending level and keep it fixed (possibly with inflation adjustment each year). If stocks perform poorly, these retirees then have to decide when to cut their spending. My strategy automatically adjusts spending level based on portfolio size. This has the advantage of guaranteeing I won’t run out of money. However, it becomes a disadvantage for any retiree who can’t or won’t spend less when stocks disappoint.
Fixed-income allocation adjusts automatically with age
Many decumulation strategies are vague about when to increase the allocation to fixed income investments in a retiree’s portfolio. My strategy uses a fixed rule to increase the percentage of fixed income each year.
No long-term bonds or corporate bonds
I get enough volatility from my heavy stock allocation. I prefer not to be open to shocks in bond markets. So, my portfolio holds no corporate bonds at all, and no government bonds of duration more than 5 years. In fact, I currently have only GICs and savings accounts.
Conclusion
My decumulation strategy is close to single-portfolio methods than it is to bucket strategies. It’s impossible to know whether my strategy will work out any better than anyone else’s. But I prefer to remove as much of my own decision-making as possible. Faced with a decision, it’s very easy to just do nothing. I have a spreadsheet that tells me what to do and when to do it.
My decumulation approach involves holding 5 years’ worth of my annual spending in short-term fixed income and the rest in stocks (described in more detail in my post Cushioned Retirement Investing). Each year, I sell enough stock to replenish the fixed-income allocation.
My annual spending each year is calculated from my age and current portfolio value. As I get older, I spend a slightly higher percentage of my remaining portfolio (see the spreadsheet in my Cushioned Retirement Investing post for the exact percentages). If stocks perform well, my annual spending will rise, and if they perform poorly, my spending goes down.
Because my annual spending changes from year to year, I have to calculate how much stock to sell each year. I take my new annual spending, multiply by 5, and subtract the current value of my fixed-income investments. This is essentially a form of rebalancing that has me selling fewer stocks when they’re down and selling more when stocks are up. In rare cases when stocks crash, I might have more fixed income than 5 times my new annual spending, and I actually buy some stocks instead of selling.
Important Differences from Bucket Strategies
No active decision on which bucket to spend from
Many bucket strategies involve making an active decision about whether to spend from stocks or fixed income in a given year. This only makes sense if you believe you have the ability to time the market. I don’t. Most people (maybe all) who believe they can time the market are wrong.
No hard switches between spending from stocks and fixed income
It’s possible to devise a bucket strategy that is entirely rules-based. A simple example would be to spend for the year entirely from fixed income if stocks are more than 20% below their peak level. The idea would be to ride out stock market declines by spending from fixed income until stocks rebound. My strategy has no hard switches between spending from stocks and spending from fixed income.
Other differences from typical decumulation strategies
Predetermined spending changes based on portfolio size
Decumulation strategies often have retirees set a spending level and keep it fixed (possibly with inflation adjustment each year). If stocks perform poorly, these retirees then have to decide when to cut their spending. My strategy automatically adjusts spending level based on portfolio size. This has the advantage of guaranteeing I won’t run out of money. However, it becomes a disadvantage for any retiree who can’t or won’t spend less when stocks disappoint.
Fixed-income allocation adjusts automatically with age
Many decumulation strategies are vague about when to increase the allocation to fixed income investments in a retiree’s portfolio. My strategy uses a fixed rule to increase the percentage of fixed income each year.
No long-term bonds or corporate bonds
I get enough volatility from my heavy stock allocation. I prefer not to be open to shocks in bond markets. So, my portfolio holds no corporate bonds at all, and no government bonds of duration more than 5 years. In fact, I currently have only GICs and savings accounts.
Conclusion
My decumulation strategy is close to single-portfolio methods than it is to bucket strategies. It’s impossible to know whether my strategy will work out any better than anyone else’s. But I prefer to remove as much of my own decision-making as possible. Faced with a decision, it’s very easy to just do nothing. I have a spreadsheet that tells me what to do and when to do it.
I like the rules based spending, and I like the annual reset based on portfolio value. I assume you also have CPP, OAS, and pension as a base amount that covers the basics? Will you delay collecting? If so, do you have a plan for a bridge to when you do start collecting?
ReplyDelete@Garth: Good questions. My wife and I will both get CPP and OAS. We plan to delay both for both of us until we're 70. My spreadsheet takes into account these payments and plans for reduced spending from our portfolio after age 70 to compensate. So, we are effectively bridging.
Delete@Garth: A little more detail. What I do is I take the future CPP and OAS payments, compute a present value, and add that to our portfolio for the purposes of computing a safe spending level.
DeleteDo you do all your calculations in today's dollars? I find it so much easier to relate to.
Delete@Garth: Yes, all calculations are in today's dollars. The easy way to do this is to use real returns rather than nominal returns. But, you have to be careful to calculate taxes on nominal returns.
DeleteAlthough I am still about 15 years out from retirement, I plan on a similar strategy. I have constructed a 5 year GIC ladder with 20% maturing each year and the rest in equities. The percentage of investable assets in GICs will increase as I reach my target amount and retirement date.
ReplyDeleteOne item that I wonder about is a buffer over the 5 year ladder in case of a prolonged downturn, perhaps a short-term bond fund that fits between the risk/return profile of GICs and equities. If I could find longer-term GICs with better return rates, that could fit the bill as well although they don't seem to be offered with my discount broker.
Your thoughts?
JB
@JB: CDIC won't insure GICs over 5 years, so they are rare. You can find government bonds that mature in longer than 5 years. It's easier to just buy a bond fund if you're happy not exactly matching the duration profile of a ladder. Whatever you settle on, the important thing is to stick to a plan rather than change paths every time stocks soar or crash.
DeleteEffective 2020 April 30, CDIC "Extended coverage of eligible deposits with terms greater than 5 years".
DeleteWill you continue to keep five years' of sending/expenses in cash once you & your wife begin receiving OAS and CPP?
ReplyDelete@Deborah S.: Good point. I plan to reduce the cash component to the amount we need to withdraw from the portfolio over 5 years. So, when we reach 70, we'll need less cash due to receiving CPP and OAS. However, 70 is a log way off for us right now.
Delete