Safety-first retirement planning
Safety-first retirement planning is an alternative to conventional retirement planning methods. For the safety-first school[1], you only have one opportunity to experience retirement, so even a small probability of complete portfolio depletion is not acceptable.[2][3]
Safety-first planning is typically contrasted with "probability-based" approaches.[2][4] Advocates of the safety-first school argue that "probability-based" approaches can lead to big swings in spending (consumption disruption, as opposed to consumption smoothing), and that high expected returns do not necessarily result in high actual returns.
Retirees face several risks, including:[5]
- Longevity risk: you don’t know how long you will live and while it is great to live longer, it is also costly.
- Sequence-of-returns risk: Retiring at the start of a bear market is very dangerous because your wealth can be depleted quite rapidly and your ability to return to the workforce may be limited.
- Inflation risk: retirees face the risk that inflation will erode the purchasing power of their savings as they progress through retirement.
The goal of safety-first investing is to guarantee that minimum income needs will be met during retirement. So the dual budget model, which distinguishes essential (non-discretionary) spending from discretionary or "preferred" spending, is used for planning.
Essential spending should be covered by matching strategies that rely on fixed income or annuity products.[1][6][7] This means giving up on some upside potential, and giving up on wealth maximization,[4] on the portion of the portfolio needed to provide the income floor.
"Preferred" spending, i.e. aspirational goals, are met with a "aspirational" or "growth" portfolio, which is invested in risky assets, or a mix of safe and risky assets.
Using these components together should decrease longevity risk, sequence-of-returns risk, and inflation risk.
Who would choose this style?
Murguia and Pfau[8][9] have developed a four quadrant matrix of retirement income styles. The safety-first approach described in this article fits within the lower-left corner, called income protection:
Investors in the lower-left quadrant tend to be more preoccupied than average with longevity goals, and less preoccupied than average with lifestyle goals.[9] In other words, having a safe source of income to cover essential expenses for life is the most important consideration, even if this might mean giving up on some ‘upside’ (potential portfolio growth allowing a higher income in the future). Investors in this quadrant are potentially willing to commit to strategies that guarantee a lifetime income, at the price of decreased flexibility (less optionality).
Direct access to the four strategy groups:
Retirement spending
The needed or desired spending level is a key factor in retirement plans. Conventional approaches to retirement planning use an income replacement rate approach, or a single retirement budget. The single budget method mixes needs (essential consumption) and wants (discretionary consumption). It gives a fixed target income, generally indexed for inflation, which can be linked with a withdrawal method that provides such a constant real income. The dual budget method is more flexible and can be linked with withdrawal methods that provide variable income.
Dual budget models of retirement spending incorporate two total spending estimates.[1][10] The first or Essential budget represents the lowest level of retirement spending that can be accepted. The second or Preferred budget represents a higher level of retirement spending that is actually desired. The retiree’s spending in any year is assumed to fall within the range bounded by these two budgets.
Liability matching
This section describes the strategy to implement the essential spending component of your budget.
Liability matching (or "income protection") strategies allows an investor planning for retirement to meet specific financial targets with near certainty.[4] Matching strategies only work with fixed income and insurance products such as annuities. There must be a known maturity date for the asset so the timing of the asset and liability can be matched.[11]
Matching strategies are very safe, but offer little or no upside potential.[7] An important reason that matching strategies are safer is because typically they use hedging to mitigate risk. By hedging risk you are protected against both non-market and market (systemic) risk.
In the context of retirement, the 'liability' that needs matching is a retirement income floor (the "Essential budget" above), indexed for inflation.[1]
Guaranteed income streams such as Old Age Security (OAS), Canada Pension Plan (CPP), Québec Pension Plan (QPP), and defined benefit pensions (which are not all inflation indexed), if applicable, contribute to this income floor, but for many Canadians that will not be enough. The missing portion to reach the income floor can be obtained by:
- delaying OAS and/or CPP/QPP, up to age 70, to increase the amounts to be received;
- with a liability matching portfolio, that will consist entirely of fixed income products;
- with life annuities;
- a combination of methods.
Aspirational portfolio
Once the retirement income floor has been secured, the rest of the portfolio can be used to fund the second part of the dual retirement budget, i.e. the preferred budget or "aspirational income goal". Depending on how much money was accumulated, and what the aspirational goals are, there may not be enough left to completely reach the aspirational goals, since the focus is on first funding the essential expenses ("safety-first"). The size of the "aspirational portfolio" is therefore determined by how much is left, and the income actually generated from it will depend on asset allocation and the chosen withdrawal method.
The aspirational portfolio (or "growth portfolio") will typically contain a mixture of stocks and bonds, with risk primarily managed through diversification.[12] Going to 100% stocks is fine if the investor has the stomach for it, since an inflation-indexed income floor has been secured with the liability matching approach. Otherwise, risk tolerance is used to come up with an asset allocation.
Many withdrawal strategies are available to guide how much to take out of the aspirational portfolio every year during retirement. Flexible strategies that adjust withdrawals to market performance do a better job than rigid rules at avoiding premature portfolio depletion. Strategies that take the investor's age into account, sometimes called actuarial approaches, allow efficient spending without leaving unexpectedly large bequests. Variable percentage withdrawal (VPW) is an example of a strategy that combines flexibility with age dependency, and also takes the asset allocation into account.
Examples
Jack and Jill are married, are both aged 30, want to retire at 65, and their money must last to age 95 or later. With the single budget model, their target retirement income is $60k in today’s dollars.
Using the dual budget method, they decide that $50k of gross retirement income would be an acceptable floor. Suppose that OAS will be $7k per person, CPP will be $13k per person, and there are no workplace pensions. So Jack and Jill are looking at a guaranteed inflation-indexed retirement income of $40k (gross) from OAS + CPP, if they claim both at age 65. They needs to generate another $10k of safe income to meet the minimum floor of $50k.
Beyond that, their aspirational goal is to generate a further $10k of income per year to reach the original $60k target, but this extra income will be allowed to fluctuate from year to year.
Deferring CPP/QPP
The easiest strategy to reach the income floor is to defer CPP to age 70. The CPP amount would be 42% larger[13], so our hypothetical $13k at age 65 would become $18.5k per person by waiting until age 70. If OAS is taken at age 65 but CPP is deferred to age 70, Jack and Jill would together get $51k of guaranteed income starting at age 70, which slightly exceeds their minimum floor. I.e., an extra $11k of safe income has been obtained.
- If they had deferred OAS too, the OAS amount would be 36% larger[14], so a hypothetical $7k at age 65 would have become $9.5k at age 70, for a total of $56k of guaranteed inflation-indexed retirement income.
How much would deferring CPP cost? Starting at age 65, Jack and Jill need to extract $36.9k per year (two times CPP at age 70) from their RRIFs, over five years, to bridge the gap to age 70. Suppose that the real return on GICs and HISAs is zero, for simplicity, and that these are the chosen vehicles for the CPP deferral strategy. Therefore, the amount needed is simply five times $36.9k or $184.6k.
Compare this strategy with not deferring CPP and leaving the money in a RRIF. With a 4% 'safe' withdrawal rate, it takes $250k of capital to generate $10k of yearly income, and there is no guarantee that this income will last to the end. Deferring CPP requires less capital, and produces a guaranteed inflation-indexed income for life. What’s the catch? If one spouse dies, the other inherits the RRIF, whereas CPP offers a survivor pension, but it’s not advantageous.
The question of how the $184.6k should be accumulated, and invested, before retirement, is not that obvious. Hardcore adherents to the safety-first school might answer that the capital should be accumulated progressively over 35 years, and invested entirely in fixed income, preferably inflation-indexed bonds.
Less safety-obsessed investors might want to accumulate the $184.6k over the last 15-20 years before retirement instead, and again put it all in fixed income. This would, in effect, mean investing mostly in stocks from age 30 to age 45 or 50 (to give the aspirational portfolio a head start), and then shifting to more fixed income over time as the CPP-deferral "mental account" takes priority when adding funds to the portfolio.
Liability matching portfolio or annuities
For the sake of illustration, suppose that the OAS/CPP/QPP deferral strategy is not available. Jack and Jill need a liability matching portfolio that will generate $10k of inflation-indexed income for 30 years (from age 65 to 95). Assuming a real return of zero for simplicity, the liability matching portfolio therefore needs to be $300k at retirement, in today’s dollars (30 times $10k). They have 35 years to accumulate this, so they can set aside about $8.6k a year to reach this target, and invest it all in safe assets such as real return bonds (RRBs). This is a very expensive approach compared to deferring CPP, and longevity risk is not addressed, unless an annuity is purchased (typically between the ages of 65 and 75).
If the annuity route is chosen from the start, the amount of capital needed can be less. For example, using Feb. 2018 quotes, it would cost Jack and Jill $247.3k of registered funds to buy an indexed (2%) joint life annuity that pays $10k of yearly income at age 65.
Young investors
The liability matching approach (with a complementary aspirational portfolio) seems well suited for investors approaching retirement, and retirees. However, from a life-cycle perspective, it can seem overly defensive for a 25-30 year old investor, who still has a lot of human capital (future work income).
For a young person with a stable, "bond-like" job in particular,[15] a stock-heavy portfolio would seem acceptable at least until age 40-45. If stocks have high returns during this period, great! The investor can switch to safer investments. If stocks don't do well in the first decades, there is still time to convert human capital into financial capital before retirement.
This approach -- take risk with stocks for young investors, then switch to a "two portfolio" (liability matching portfolio + aspirational portfolio) mode later -- is recommended by W. Bernstein in his book "The Ages of the Investor: A Critical Look at Life-cycle Investing".[16]
See also
References
- ^ a b c d Bodie Z, Taqqu R (2011) Risk Less and Prosper: Your Guide to Safer Investing, Wiley, 196 p. ISNB 978-1-118-01430-1.
- ^ a b Pfau W (2014) 2 Schools of Thought on Retirement Income, Journal of Financial Planning, April 2014 issue
- ^ Post by longinvest in Financial Wisdom Forum, topic "Probabilistic approaches to investing", March 14, 2018.
- ^ a b c Pfau W, Cooper J (2014) The Yin and Yang of Retirement Income Philosophies, report for Challenger Limited, 28 p.
- ^ Pfau WD (2018) An Overview of Retirement Income Planning, Journal of Financial Counseling and Planning 29:114-120, available at SSRN
- ^ Summary of Bodie & Taqqu (2011) by member Quebec in Financial Wisdom Forum, November 12, 2016.
- ^ a b Merton RC (2014) The Crisis in Retirement Planning. Harvard Business Review, July–August 2014 Issue
- ^ Murguia A, Pfau WD (2021a) A Model Approach to Selecting a Personalized Retirement Income Strategy, working paper available on SSRN
- ^ a b Murguia A, Pfau WD (2021b) Selecting a personalized retirement income strategy. Retirement Management Journal 10:46-58, available on SSRN
- ^ Paul Merriman, Live it Up Without Outliving Your Money, revised and updated edition (John Wiley & Sons, Inc., 2008), 206 p.
- ^ Bodie (2006) A Note on Economic Principles and Financial Literacy. Networks Financial Institute Policy Brief No. 2006-PB-07. Available at SSRN, viewed March 31, 2018.
- ^ Westmacott G, Daley S (2015) The design and depletion of retirement portfolios. PWL Capital report, 27 p.
- ^ Tim Cestnick, How to time the collection of CPP and OAS, The Globe and Mail, September 25, 2015, viewed April 1, 2018.
- ^ Canada.ca, Deferring your Old Age Security pension, viewed April 1, 2018.
- ^ M.A. Milevsky, Is your client a bond or a stock, Advisor's Edge, November 2003
- ^ Pfau review of "The Ages of the Investor: A Critical Look at Life-cycle Investing" by William J. Bernstein, 2012 (ISBN 1478227133)
External links
- Financial Wisdom Forum, Risking Less and Prospering using Real Return Bonds?
- Bogleheads wiki: Life-cycle finance
- Bogleheads wiki: Risk transfer techniques
- Rational Reminder podcast: Safety-First: A Sensible Approach to Retirement Income Planning with Wade Pfau, March 2020
- Pfau WD (2017) Retirement Income Showdown: Risk Pooling vs. Risk Premium. Journal of Financial Planning, February 2017, also available on SSRN.