Are you feeling lucky? That’s how the debate between the two schools of retirement income often shapes up.
Do you keep your retirement assets in the stock market, where you’ll have more flexibility and better odds for long-term financial growth, along with a chance of failure? Or do you go the safe, secure route and annuitize your assets for a reliable, though possibly mediocre, retirement paycheck?
Even if you don’t want to take sides in this debate, it’s helpful to understand it. These two schools, and the retirement income strategies that belong to each, define your options for retirement income.
Just don’t believe that you are required to choose one or the other. The reality is that you will probably need to mix the two philosophies in the right portions for your personal situation. Let’s begin with some definitions….
Probability vs. Safety
The probability-based retirement income philosophy generally involves managing an investment portfolio using some type of withdrawal strategy. It could be a safe withdrawal rate, or a variable withdrawal rate. The advantages of this approach are its flexibility and its potential upside. Your capital remains under your control at all times. You don’t hand it over to somebody else to manage. And, the odds are in your favor over the long term. As the chart here from J.P. Morgan shows, the average long-term return on stocks has been more than 10%. And, in the last 60+ years, the stock market has never gone for any 20 year period without turning at least a 6% profit, annualized.
The disadvantages of this school are significant as well. Most serious but least understood is “sequence of returns risk.” Because you aren’t simply managing a static portfolio, but must also withdraw to live off it, you are subject to a special mathematics of loss. I’ll explain this in more detail below, but essentially you must pay the mathematical piper in the form of reduced performance. The next disadvantage is that by relying on a lump sum portfolio for an indefinite period — your lifespan — you are taking on longevity risk. The longer you live, the greater the chance that you’ll run out of money, regardless of market returns. Finally, if you are managing the money yourself, then cognitive decline could affect your ability to manage your own financial affairs.
By contrast, the safety-first retirement income philosophy generally involves purchasing an annuity or bond ladder to “lock in” retirement income. The advantages are security and predictability. You might sleep easier at night knowing you aren’t subject to stock market volatility, at least if you have confidence in insurance company and bond ratings. And your available retirement spending will be known (and fixed) in advance. You won’t need to adjust your lifestyle unexpectedly based on future market moves.
But the disadvantages of this school are significant as well. For me, the single point of failure looms large: Rather than piggybacking on the self-interest of thousands of companies in the broad stock market, you are trusting a single insurance company to look after your interests. For decades. Financial planner and commentator Michael Kitces points out that insurance companies do fail, and even AAA-rated bonds have measurable default rates. Inflation protection is another question mark hanging over the safety-first approach. It’s either unavailable or very expensive. So just how “secure” and “predictable” is your fixed income stream? Add to these negatives the lack of flexibility in managing your money, and no possibility of an upside. You can’t generally tap annuitized assets for emergency or legacy purposes. And, if has always been the case, the stock market outperforms other assets over the course of your retirement, you will have to watch the party from the sidelines.
Two Styles of Planning
Choosing between the probability-based and safety-first styles of generating retirement income is not just an investing decision. It also leads to two different analyses of retirement expenses, and two different styles of financial planning:
In the probability-based philosophy, you, or a financial planner, lump all your expenses together, essential and discretionary, then compute the probability of meeting them over the course of your retirement, using your portfolio of assets. Failure is defined as running out of money before running out of life. A failure probability in the neighborhood of 10% is often considered acceptable. That’s one chance in ten. So failure is entirely possible, in theory, with this approach. Some people are uncomfortable with that. None of us would get on an airplane with those kinds of odds. But, in reality, complete financial failure is unlikely if the analysis shows a relatively low probability. Why? Because most sensible people are going to modify their lifestyle if the numbers start heading in the wrong direction. So the failure rate associated with a probability-based analysis is more a metric for the likelihood and severity of having to adjust your lifestyle, than it is a prediction for the odds of eating cat food later in retirement.
In the safety-first philosophy, you, or a financial planner, match guaranteed income to essential expenses. So, if you have non-discretionary expenses that aren’t covered by a pension or Social Security, you purchase an annuity, or possibly a bond ladder, to cover that need. In theory, assuming you have enough assets, “failure” is impossible with a safety-first approach. That’s because you’ve assigned guaranteed income sources for all your needs. In reality, there are holes: those needs won’t be fully met unless that income is inflation-adjusted — which can be difficult to achieve in today’s world. And you are banking that your insurer, and its guarantee association if any, will remain solvent for decades.
The Most Important Distinction: Risk Management
The financial industry, whether insurance agents or investment advisors, would like you to believe the retirement income debate comes down to security (safety-first) versus upside (probability-based). You are asked to choose between playing it safe, or maximizing your retirement lifestyle. In fact, framing the debate that way is a distraction.
Michael Kitces points out, “the real distinction is whether (market and longevity) risk is transferred or retained, and if retained how those risks are managed or avoided.”
In other words, the real difference isn’t between playing it safe and gambling. There is risk either way. Insurers speak of “risk transfer.” In reality, the risk never disappears. It’s a question of who has the job of monitoring and managing it. So the real difference between the retirement income schools lies in who manages the risk — you and your portfolio manager, or an insurance company. And, it should come as no surprise that if you choose to have somebody else manage the risk, you’ll have to pay them to do that job.
As with asset allocation, the most important factor in choosing a retirement income strategy may be understanding which suits your temperament best. Who should manage the risk? You or somebody else? Do-it-yourselfers will likely favor the probability-based approach, and those without financial experience will likely favor safety-first.
Doing it Yourself: Paying for Volatility
Admittedly, do-it-yourselfers have the harder job. If you take that on, you’ll need some appreciation for the risk/return of stocks over the long term. I’ve already referenced a study showing that in the last 60+ years, the stock market has never gone for any 20-year period without turning at least a 6% profit, annualized. But experts can look at the same statistics and come to different conclusions. Some argue that individuals can’t rely on averages, because we only get a single chance at retirement. They’ll go on to point out that we are in unprecedented market conditions with low interest rates and high valuations. And they’ll say we can’t count on historical averages going forward. This is scary stuff, that might send more than a few do-it-yourselfers into the safety-first camp, myself included, at least later in retirement.
Do-it-yourselfers also need a healthy respect for “sequence of returns” risk. As mentioned above, this is the mathematical reality that reduces the returns on stocks in a distribution portfolio. In essence, a portfolio that you must withdraw from continually in retirement performs differently from one you can leave alone to accumulate. The problem stems from the impact of those regular withdrawals when the market is down. They rob you of principal that would have provided future earnings. Jim Otar, in his comprehensive book on retirement planning, Unveiling The Retirement Myth
, calls this the “time value of fluctuations” (TVF). He offers a formula, based on historical data, to compute its value as a function of the time span. For 20 years, his TVF is about 2.2%. For 30 years it’s about 1.6%. So, when living off a portfolio in retirement, at a minimum, you must reduce your expected average long-term stock market returns by at least a percentage point or two, to account for sequence of returns risk.
This volatility penalty sounds like a mark against long-term stock holding, and it is. But it’s not a knockout punch, compared to the alternative — annuities backed by a bond portfolio. Bonds, being less risky, are nearly certain to underperform stocks over the long term. And generally bonds underperform stocks by much more than the costs associated with sequence of returns risk.
Read next: How Much Should I Hold in Company Stock?
Finding Your Mix
Supposedly, if you can’t handle a small chance of running low in retirement, then you belong to the “safety-first” camp. If you can roll the dice and still sleep at night, then you are in the “probability-based” school. But that’s a simplistic formula.
SPONSORED FINANCIAL CONTENT
These two categories are useful tools for retirement planning. But neither represents the reality of how retirees live their lives. You don’t blindly commit your financial success to a single throw of the dice early in retirement. Rather, retirement finances are an ongoing, iterative process. If you’ve decided that your retirement lifestyle will be more a function of market returns than insurance company payments, then you’re certainly going to adjust that lifestyle if the market doesn’t behave as expected! This is our strategy. We are currently treading water in the probability-based pool, and I expect that to continue for another ten years. Then we’ll likely pull the trigger on a partial safety-first solution in our 60’s, by purchasing annuities with a portion of our assets.
Why are we doing it this way? Generally speaking, the safety-first approach is irreversible. Once you buy an annuity, you own that decision for life. The probability-based approach is more flexible. Yes, you must be prepared to cut spending if needed, but you can also take advantage of any upside that appears. In the first half-decade of our retirement, our assets have done a little better than expected, and we have chosen to spend a little more. We have the health to enjoy that spending, and the ability to cut back or generate more income if needed. In later years, our expenses will probably fall naturally. If not, we may need to cut back. We are OK with that. Had we already chosen safety-first and annuitized, we would have locked in a more restricted lifestyle that could have left us regretting our 50’s.
“Probability-based” and “safety-first” label the extremes in retirement planning. As in other areas of life, you’d do well to avoid those extremes and find the compromise path that is right for you. Rather than picking sides, many retirees will be best served by diversifying their strategies to create a hybrid retirement income solution. Understand your own temperament and finances, then choose your own best mix and timing.
Are you feeling lucky? It was a great line for Clint Eastwood. But it’s the wrong question for retirement planning!
Darrow Kirkpatrick is a software engineer and author who lived frugally, invested successfully, and retired in 2011 at age 50. He writes regularly about saving, investing and retiring on his blog CanIRetireYet.com.
More From Can I Retire Yet: