Asset allocation over the lifecycle

Asset allocation over the lifecycle examines how investors can vary their asset mix over time, in the context of long term investing, typically when investing for retirement, and then during retirement, over several decades. The "need, ability, and willingness to take risk" framework discussed in the main asset allocation article lets you arrive at an initial asset allocation. But how should that asset allocation evolve over time? Should it be constant, e.g. a "balanced portfolio" with a 60-40 split between equities and fixed income (mostly bonds), during both accumulation (pre-retirement) and decumulation (after retirement)? Or should the portfolio asset mix evolve, in a planned fashion, over time?

Strategies examined in this article include a constant allocation over time; declining equity glidepaths; and rising equity glidepaths. A glidepath is a pre-decided trajectory of asset allocation over time. This article best fits with total return decumulation strategies in which the retiree wants to continue to manage a portfolio and be exposed to stock markets after retirement, as opposed to other Retirement income strategies and styles.

All age-based guidelines are predicated on the assumption that an individual's circumstances mirror the general population's. Individuals with different retirement ages (earlier or later), asset levels, or needs for the money (e.g. saving for a goal other than retirement) would be well-advised to consider what circumstances make their situation different and adjust their asset allocation accordingly.

Constant allocation

The simplest strategy is a constant (or "static") asset allocation over the lifecycle. An investor could buy a single balanced mutual fund or asset allocation ETF and maintain that exposure for several decades, in a very long-term buy and hold (but rebalance) attitude. Forsyth and Vetzal compared a constant allocation during the accumulation phase to various "deterministic" (predetermined glidepath) strategies and concluded that:

Consistent with the theory, optimal deterministic strategies offer virtually no improvement compared to constant proportion strategies.

Again looking only at the accumulation phase, Poterba et al. conclude that:

...the distribution of retirement wealth associated with typical lifecycle investment strategies is similar to that from age-invariant asset allocation strategies that set the equity share of the portfolio equal to the average equity share in the lifecycle strategies.

Some Financial Wisdom Forum members have a "VBAL and chill" mantra (VBAL is a popular asset allocation ETF with 60% equities and 40% fixed income). Other members like 80% equities and 20% fixed income ("VGRO and chill").

There is also discussion on the forum about a "100% equities for life" strategy, although Asness explains that "a diversified portfolio provides more return per unit of risk", where diversification generally means bonds. Swedroe notes that many investors weigh "the pain of losses more heavily than the benefits of gains" and that investors may not be able to stand the pain or large drawdowns associated with the all-equity strategy and stay the course. Finally, McQuarrie, using over 200 years of data, show that "over multi-decade periods, sometimes stocks outperformed bonds, sometimes bonds outperformed stocks and sometimes they performed about the same... sometimes there is an equity premium, sometimes not".

Regardless of the specific asset mix chosen initially, a potential problem with a constant allocation over decades, is that the need, ability, and willingness to take risk of an investor tends to change over time. Dan Bortolotti writes:

I’m a big fan of simple solutions, but you can take that too far. The idea of a "set it and forget it" investment strategy is appealing, and it can work well for a while. Eventually, however, you need to switch off the autopilot...

Declining equity glidepaths

Declining equity glidepaths represent the conventional wisdom. As you age, your reliance on your investments will likely increase and thus it is expected that your focus adjusts less towards growth (and its inherent variability) and more towards stability (and its inherent preservation of capital). This concept is best graphically represented by a human capital graph:

Asset Allocation over Time

When you are 20, you have little financial wealth but an entire earning lifetime ahead of you, and more flexibility in how long and how hard you decide to work. You can afford to have an aggressive asset allocation, although you may not have the stomach for it. As you age, your financial wealth increases but your remaining earning years decrease and your asset allocation becomes more conservative. You need to protect your assets because you have no way of replacing them.

This is illustrated by the well-known rule of thumb of "your age in bonds" (fixed income), with the rest of the portfolio in stocks (equities). A 20-year-old investor would have 20% bonds, a 40-year-old would have 40% bonds, and so on. This can be rephrased as the "100 minus your age" rule, which yields the proportion of equities. The 20-year-old investor would have 100-20 = 80% equities (and the rest in fixed income). Variations include "110 minus age" or "120 minus age" rules, with a start at 100% equities for the 20-year-old investor, in the latter. The first two rules are illustrated by the blue and orange lines in the following graph:

Declining equity glidepaths can be completely automated with lifecycle or "target date" funds. You pick the fund with the year closest to your anticipated retirement date, and the manager adjusts the asset allocation over time. In Canada, these funds are mostly available in some defined contribution pension plans, where they make up nearly 20% of assets; in the US they are also available in ETF format, open to all investors. In general, target date funds start with very agressive allocations and maintain them until the investor is about 40 years-old, before declining in equity percentage at least until the assumed retirement age of 65 (see the two green curves in the graph above).

The simplest way to implement a declining equity glidepath with exchange-traded funds (ETFs) in Canada would be to use two funds, one for stocks (one of the "all-equity" asset allocation ETFs) and one for Canadian bonds, rebalance manually to the target allocation every year or so, and adjust the asset mix every five years using the chosen rule (e.g., "age in bonds" or "110 minus age in stocks"). A slightly more complex, equivalent strategy can be realized with a simple index portfolio of 3-5 ETFs (or index mutual funds), again changing the target asset mix every 5 years.

Declining equities then stable after retirement

Should the allocation to equities really continue to decline when portfolio withdrawals start, as the age-based rules imply? After retirement, human capital has been depleted and is not a factor anymore. The proportion of equities could therefore stabilize to a final constant asset mix ("glide to" retirement, as opposed "glide through"). Further reducing the equities over time to very low levels, although possibly tempting for psychological reasons, would reduce the expected return of the portfolio and might not allow sustainable withdrawals (see discussion in Total return decumulation).

Some target-date funds indeed have a constant asset allocation beyond a certain age. A US example from Vanguard is shown by the grey line in the graph within the previous section.

Blanchett studied a number of asset allocation strategies during retirement and concluded that:

...based on the research conducted for this paper, as well as other qualitative and practical considerations, the optimal allocation for most retirees is likely a (constant allocation with a) balanced portfolio, such as a 60 percent equity and 40 percent fixed income/cash allocation.

A constant allocation to 60% stocks during retirement has also been suggested by Estrada because it produced lower failure rates, better downside protection and higher expected bequests over 30 year historical periods, relative to dynamic strategies examined.

Rising equity glidepaths

A seldom-discussed strategy is a rising equity glidepath. The "rising" part often concerns only the post-retirement period: the proportion of equities would reach a minimum at retirement, to protect the investor from sequence-of-returns risk, and rise again thereafter. If bad returns occur early in retirement and good returns later, the portfolio might be prematurely depleted by withdrawals, even if the average returns throughout the retirement period are not especially low. Higher fixed income allocations, especially at the beginning of retirement, can help attenuate this risk; the stock allocation can climb back higher later. The lifetime trajectory of the proportion of equities would be U-shaped: "the stock allocation is the lowest at the point when lifestyle spending goals are most vulnerable to absolute losses in wealth (the retirement transition itself), but greater in both the earliest years and also the latest".

In a subsequent study, Blanchett found that rising equity glidepaths during retirement were generally inferior to other strategies in terms of success rates in simulations. However this may have been influenced by their investment returns assumptions. Waggle and Aggrawal support rising equity glidepaths during retirement and write:

One of the biggest threats to a portfolio’s success at providing lifetime income is a big loss in value early, rather than later, in retirement. Lower up-front allocations to stock would potentially leave a retiree less susceptible to this sequence risk. The increasing steps upward in the stock allocation would come if the portfolio successfully weathers the early retirement years and the safety cushion of the overall portfolio increases. As the retirement time horizon shortens and the desired income stream from the portfolio becomes more certain, the focus of the portfolio gradually shifts. Instead of focusing on guaranteeing the retiree’s income needs, the portfolio management could emphasize growth for the future heirs, which is where the higher allocations to stock come into play.

Rising equity glidepaths throughout the lifecycle (pre- and post-retirement) have also been studied (see and references therein).

Conclusions

There is no consensus in the academic literature, or on the Financial Wisdom Forum, on a single "best" trajectory for asset allocation over the life cycle. What is ideal for one investor may not be the best for another one, depending on asset levels, risk tolerance, amount of guaranteed income (Old Age Security, CPP/QPP, employer pensions, annuities, ...), planned withdrawal method, bequest motives, etc.

An important point is the behavioural factors that are at play: an asset allocation strategy considered 'optimal' in a simulation will not work in real life if the investor can’t stick to it over several decades.

Alternative approaches

As mentioned in the lead section, this article fits with total return decumulation strategies and conventional retirement planning. The assumption is that the investor manages a portfolio of stocks and bonds (+/- cash) both before and after retirement, and intends to cover the "four Ls" that way: the goals of longevity, lifestyle, liquidity and legacy. The question becomes what is the best withdrawal method and asset allocation (this article) to do this, while minimizing the odds of running out of money during retirement.

There are alternative approaches, as covered in Retirement income strategies and styles. This includes safety-first retirement planning, in which the retiree is typically more preoccupied by longevity goals – maintaining spending power no matter how long the retiree lives, especially for essential expenses, like housing, food, healthcare, etc. – than by lifestyle goals which include both essential and discretionary expenses. In the safety-first approach, no probability of failure is acceptable, so essential income sources must be guaranteed or protected through liability matching, for example by delaying QPP/CPP, or by using annuities, bond ladders, and so on. Stocks will not be relied upon to fund essential expenses. However, discretionary spending (see dual-budget model) and legacy goals can be funded using an "aspirational portfolio" which could be stock-heavy since essential expenses have already been secured.

External links

• Bogleheads wiki: Glide paths