The Sustainable Withdrawal Rates Fantasy

Sustainable Withdrawal Rates (SWR) is a strategy for spending personal savings in retirement. It is sometimes referred to as the 4% Rule and it’s intended to create your own homemade fixed annuity without having to hand your principal over to an insurance company. Here’s how it would work.

You retire with a million dollars and invest it in a portfolio of stocks and bonds. You withdraw $45,000 (4.5%) to pay for the first year of retirement.  The second year, you withdraw $45,000 plus an additional amount that would cover the previous year’s rate of inflation.  SWR proponents say you can keep doing this for at least thirty years and not run out of money. 

In fact, the studies seem to show that you could end up with a lot of money.  Maybe even more than you started with.  And there’s “only” a 5% chance that you would go broke before you die.  Does that sound wonderful or what?

The strategy is based on statistical studies said to show that a retiree can invest his savings in a portfolio of stocks and bonds, withdraw 4.5% of the initial portfolio value[1], and continue to withdraw that same dollar amount annually with a 95% probability of funding at least thirty years of retirement.

But, they don’t show that at all.

Imagine you asked me how safe it would be to walk across the highway in front of my house.  I tell you that I watched 100 people try to cross in the last five years and “only” five of them were run over.  You might conclude that you have about a 95% chance of crossing safely.

Now imagine that I add, as you walk away, that I only watched people cross the highway who refused to check for oncoming traffic and ignored any cars they saw coming toward them after they began crossing.  I can’t tell you anything about the results for people who tried to avoid getting hit, I say, because I didn’t count them.

You would probably think that I was a little nutty, and then wonder why my statistics would be meaningful to you in any way.  You, after all, look both ways before you cross a road and if a car approaches while you’re still crossing, you damned well try to get out of its way. If we measured the fatality rates of pedestrians whose highway-crossing strategy was to dart into traffic without looking, common sense tells us to expect that they would be struck more often than pedestrians who would do everything possible to avoid danger. Otherwise, none of us would bother looking.

The SWR studies have a similar flaw.  They assume that a retiree would continue spending the same amount every year even when it would be obvious that they were about to go broke.  Consequently, they don’t predict the number of retirees who would go broke using this strategy. They show the probability of successfully funding at least 30 years for a retiree who would refuse to reduce spending to avoid going broke.

If I could clearly see that I would go broke if I didn’t cut back my spending, I would cut back my spending.  How about you? 

The logic of the SWR studies is flawed by what statisticians call the “unrepresentative sample fallacy of inductive reasoning”. A mouthful to be sure, it means that the sample we measure doesn’t represent the population for which we want to generalize.

An extreme example would be measuring the heights of players on an NBA basketball team and concluding that the average American is 6’ 6” tall and can dunk a basketball. Measuring the bankruptcy rate for people who would do nothing to avoid going broke and asserting that the rest of us share the same odds is another extreme example.

How to Improve Your Chances of Dying Broke

We would expect the probability of financial ruin for retirees who would try to avoid bankruptcy to be lower than for retirees who would simply continue spending until they
were broke and bankruptcy rates bear this out.

A 95% probability of successfully funding retirement in the SWR studies is a different way of saying there is a 5% chance of failure, and failure in these studies is defined as depleting one’s savings before retirement ends. That’s a 5% chance that the SWR retiree ends up bankrupt during retirement. Broke. Penniless. Ruined. Or, as Ray Charles so elegantly put it, busted.

While the SWR strategy predicts about a 5% probability of financial ruin with 4.5% withdrawals, according to Elizabeth Warren, a Harvard expert on personal bankruptcy, the bankruptcy rate for Americans aged 65 and older was recently about 0.43%[2]. That’s about 1/12^th of the rate of financial ruin promised by the SWR strategy for this age group.

Let me say that in a different way. The SWR studies predict that retirees who employ the strategy are about 12 times more likely to go broke than the typical American of retirement age.

The Yale Study 

A number of studies refer to a 1973 Harvard University study to determine how much their managers could safely withdraw from their endowment fund without eroding the principal. Though I have been unable to locate a copy of that oft-quoted study, it is reported that the answer was determined to be about 4%.  (The 2008 Yale Endowment report does state their spending policy is limited to 4.5% to 6% annually.)

Some argue that this supports the SWR findings, but this seems a better argument that 4% is not a safe withdrawal rate for retirees.

If the Harvard endowment, with its unequalled investment management talent, lower costs, non-profit tax status, and broader range of investment opportunities earns only enough to support 4% annual withdrawals, should individual retirees expect to sustain the same level of spending? Perhaps, but only if they are able to match or exceed the investment returns of the Harvard endowment. Most individual investors and money managers don’t even match average  market returns over the long term.

The Harvard endowment can also solicit new contributions from donors, an option I suspect is rarely available to retirees.

One would expect the Harvard endowment investments to outperform those of the typical retiree; otherwise, some extremely well paid investment managers would be fired. The consequences of major investment losses would also seem to be less severe for Harvard, since a retiree would be left penniless with few prospects of avoiding welfare, while Harvard can always go out and solicit donations from wealthy alumni. And, of course, the Harvard Endowment’s assets provide a larger margin of error— retirees with $26B in assets can probably feel good about spending $1.2B (4%) annually with little fear of bankruptcy.  Those of us with significantly smaller portfolios need to be more careful.

With greater risk and lower expected market returns, we should expect retirees to be able to spend somewhat less than Harvard’s 4%.

The 2007-2008 Stock Market’s Rebuttal of SWR

In the 2007-2008 market crash, many retirement savings accounts dropped nearly 50% in value. The financial press began to call 401K accounts “201Ks”. Workers who had planned to retire soon were forced to delay those plans, some for several years. People who had already retired and had planned to live off 4.5% withdrawals from their stock and bond portfolios were forced to reduce their standard of living or to try to re-enter the workforce. The shortcomings of the SWR strategy became brutally obvious.

SWR advocates have recently tried to explain to countless retirees whose SWR-based retirement plans were obliterated by the 2008 market downturn that the 4% Rule was never more than a rule of thumb.  Rules of thumb can be handy guidelines when we don’t have a lot at stake.  “Show up twenty minutes before the movie starts and you’ll probably get a ticket” works pretty well, while “fasten the engine to the wing with a handful of the biggest bolts you can find”, not so much.

Financial plans that fail in retirement can mean destitution with few options for recovery.  Retirees deserve better than rules of thumb.

A Few Other Rebuttals

In a paper entitled “The 4% Rule-- At What Price?”, written by Nobel Laureate, William Sharpe and others, the authors showed that the SWR strategy is much more expensive than its alternatives. In its conclusion the authors state, “Despite its ubiquity, it is time to replace the 4% rule with approaches better grounded in fundamental economic analysis.”

David Babbel of The University of Pennsylvania’s Wharton Business School refers to these strategies as fantasies.

Or, consider Dallas News finance columnist, Scott Burns.  Burns, who was an avid supporter of the SWR strategy for years as a columnist, recently co-authored a book with Laurence J. Kotlikoff entitled Spend ‘til the End in which he now refers to the 4% Rule of Thumb as a “Rule of Dumb”.

Sustainable Withdrawal Rates strategies, though wildly popular in the financial press and among many financial planners prior to the 2007-2008 market crash, are voodoo economics and misapplied statistics.  Unfortunately for many people who neared retirement in 2006, and many others who had recently retired and expected to fund their golden years with the SWR strategy, the books, advertisements and web calculators that promote the strategy were quite effective.

Don’t fall for it.

---

1² There are several “Sustainable Withdrawal Rate” studies and each calculates slightly different probabilities, but all project about 4% to 4.5% safe withdrawal rates for thirty years.

[2] Generations of Struggle, D. Thorne, E. Warren and T. Sullivan, AARP, June, 2008.

It's A New World

When I decided to retire a few years ago, I read enough books and papers on retirement planning to fill a small library.  I also talked to several financial planners.  I got some good advice, a little great advice, and a lot of very bad advice.

I read about Sustainable Withdrawal Rates, for example, a body of work that suggests retirees can invest their nest egg in stocks, withdraw 4½% of their total savings the year they retire, and continue to withdraw that same dollar amount annually with a 90% - 95% probability of safely funding at least thirty years of retirement.

Not only did I read books and papers about “SWR”, I built my own Monte Carlo simulation software.  In doing so, I realized that the statistical inferences are flawed and that, even if the statistics were correct, the strategy is not “safe”.  

Retirees who employ the SWR strategy are promised a rate of bankruptcy of “just” 5% - 10%, but the bankruptcy rate for households in the United States over the age of 65 is currently less than half a percent.  If your goal is to increase your chances of going bankrupt by more than an order of magnitude, SWR may be just the ticket.

It is easy to understand why the financial services industry likes SWR, though; its premise is that retirees will buy lots of stocks and mutual funds.

Next, I began to read academic papers written by economists.  Surely, they would be objective. I found one that explained the virtues of fixed annuities, contracts sold by insurance companies that promise to provide income for life, no matter how long that life might be.  Then I discovered that the academic who wrote them moonlights as a contractor for the insurance industry.  Maybe not entirely objective after all.

I considered hiring several financial planners and even hired one for a brief period.  Again, I heard some good advice and a lot of what appeared to be just plain wrong.

A highly regarded financial planner in the Washington, DC area gave me a lot of good advice regarding stock investments, but some terrible advice on mortgages.

When I asked if I should pay off my mortgage of several hundred thousand dollars, he replied, “I don’t see any reason why you should.”

 

Over the next few years, my taxes shrank dramatically, leaving little tax advantage to holding a mortgage.  Then the stock market tanked and that several hundred thousand dollars I had invested in stocks instead of paying off my mortgage began to evaporate.  Meanwhile, I was paying a bank nearly 7% a year in mortgage interest for the privilege of losing the principal in the stock market.

There are lots of people who want to help you plan retirement, but most want to sell you something, like mutual funds, stocks, or annuities.  It’s difficult to get an objective opinion on fixed annuities from someone who makes their living selling mutual funds.  Like asking a barber if you need a haircut.

Ultimately, I took the drastic step of becoming a financial planner myself, primarily to plan my own retirement, but also to help others.  We Baby Boomers started serious planning for retirement way too late.  We failed to understand the challenge of funding perhaps three decades worth of Golden Years and, in my experience, we find it far too difficult to find competent, unbiased retirement advice.

It isn’t surprising that we missed the memo.  During our lifetime, life expectancies have risen dramatically and company pensions have been replaced by IRA’s and 401Ks that shifted all of the financial risk of retirement from our employers to us.  We are the first generation to face potentially long periods of retirement without company provided pensions.  It’s a new world.

I hope to remedy this as best I can by providing unbiased, logical financial planning advice.  My goal is to provide information that is useful, easily understood and as enjoyable to read as retirement planning advice can be.

Retirement is a Risky Business

Funding a retirement that could last for decades is a very risky undertaking, not quite like Pilgrims planning to leave England to spend the rest of their lives in the unknown America, but not totally unlike that, either. They had to plan to provide for themselves for a very long time, they were heavily dependent upon themselves and their families, the risks were substantial, and turning back was extremely difficult if not impossible.

There are risks that we can’t foresee.  We didn't know until the 1980's, for example, that we need to protect ourselves against HIV and we didn't know until recently that some of the largest financial services companies in the U.S. could fail.

There are also risks we know about, like the risk of inflation, which we can prepare for. It's difficult to plan for unforeseeable risks, but there are things we can do to protect ourselves against the risks we can envision.  

Many households who had saved enough money to retire comfortably or were on track to do so saw those dreams vaporized by the stock market crash of 2007-2008.  Many others, people who thought they were quite wealthy, learned that Bernie Madoff had absconded with their dreams. As these people can tell you, hanging onto your retirement savings by carefully planning to mitigate risk is as important as saving enough money in the first place.

Inflation Risk

Imagine you retired in 1980 with a company pension (they were more common then) that had no cost of living adjustment (also common). Those monthly checks that seemed so generous in the days of Ronald Reagan have been battered by inflation. Each dollar of those pension checks that purchased $1 worth of goods back then buys only 38 cents worth of goods in 2010.

Careful planning to mitigate inflation risk can help you avoid a similar predicament at the end of your retirement thirty years from now. Inflation-protected Treasury bonds (TIPs), stocks, Social Security benefits and some fixed annuities offer inflation protection.

Longevity Risk

Financial planners call the risk that you will live a long life and outlive your savings “longevity risk”. A 60-year old man today has a 20% probability of reaching age 95 and a 60-year old woman has a 30% chance. However, there is a 40% chance that at least one member of a married couple of the same age will live until 95. That's a lot of years to fund if you’re lucky enough to live a long life.

We can ensure against longevity risk by assuming in our retirement plan that we will live to age 95. That means spending less of our savings each year in case we do live that long. Fixed annuities are insurance policies that protect against longevity risk, since they promise to pay a certain annual income no matter how long we live. Social Security benefits are an annuity with inflation protection that helps address longevity risk.

Foreclosure Risk

Owning our home at any stage of life entails the risk that we might not be able to keep our mortgage payments current. Ultimately, the mortgagor will foreclose and take our home. Foreclosure can be both more likely and more devastating after we retire because without a job we are less capable of recovering from an economic crisis.

Buying a smaller home after we retire can reduce the size of our mortgage and make foreclosure less likely.  We can also protect ourselves from foreclosure risk by renting instead of owning and eliminating the mortgage altogether.   Paying off most or the entire mortgage before we retire reduces foreclosure risk and often increases our standard of living. 

Retiree Fraud Risk

Retirees have long been the targets of fraud.  Many have lost their entire savings to shysters and Bernie Madoff recently showed that the wealthy and "financially sophisticated" are not immune. In fact, they are more attractive targets.

The best way to protect against retirement fraud is to educate yourself.  A good way to start is by reading Ken Fisher's book, How to Smell a Rat: The Five Signs of Financial Fraud.

Accumulation Risk

There is a risk that we won't save and invest enough money while we are working to fund retirement without accepting a lower standard of living than we enjoyed during our working lives.

An economic technique called "consumption smoothing" helps calculate the amount we should save before retirement to avoid a decline in our standard of living after retirement begins.

“Countdown” Risk

As we approach retirement, the risk increases that we won’t have time to recover from a market setback.

John was 63 years old in 2005 and had accumulated $500,000 dollars in his 401K account.  His son, Jacob, was 40 and had saved $150,000 in his retirement savings accounts. Both had all of their retirement funds invested in stocks.  That was perhaps an OK strategy for Jacob; for John, not so much.

John looked forward to retiring at age 65 with $25,000 a year in social security benefits and another $25,000¹ to spend annually from his stock earnings. He and his wife figured $50,000 a year would support the retirement they wanted.

By 2008, the market crash had destroyed nearly half the value of both portfolios.  Jacob had 25 more years of earning and investment returns to rebuild his savings before he retired, but John did not.

John knew that he would still receive the same social security benefits and realized he had dodged a bullet when recent attempts to privatize social security had failed— the last thing he needed was to have his social security benefits invested in the market, too. But, his stock portfolio was now valued at only $250,000 and would only support $12,500 of annual spending. He would either have to reduce his standard of living after retiring or work several more years to replace the $250,000 he lost in the market crash.

The most effective way to reduce market risk is to own more bonds and less stock. At about age 55, we should reduce our exposure to stocks. By that age, we should have saved most of the money we will need to retire and the focus then changes from making lots of money in the stock market to keeping what we have already accumulated.  

Early Retirement Market Risk

The first decade of retirement is another time we should reduce our exposure to stocks. At that point, we may still have more than two decades of retirement to fund and we can ill afford to lose our savings. The aforementioned pre-retirees who lost half their savings in the 2007-2008 market crash had to delay retirement by several years, but retirees who had left the workplace after 2000 found themselves in even worse shape. Having left their jobs already, they faced an even more difficult task of recovering financially.

Health Care Cost Risk

Perhaps the biggest risk we face in retirement is the cost of health care.  Health care costs are growing at a rate much higher than the general rate of inflation.

According to a recent study by Anthony Webb and Natalia Zhivan at the Center for Retirement Research at Boston College, long term care is by far the greatest portion of health care cost risk that we face.

According to the study, at age 65, a typical married couple free of chronic disease can expect to spend nearly $200,000 on remaining lifetime health care costs excluding nursing home care. There is a 5% probability that these costs will exceed $300,000.

The prospect of needing nursing home care worsens an already bad situation dramatically. Including nursing home care, the average cost of remaining lifetime health care costs at age 65 increases from $200,000 to $260,000, with a 5% probability of costs exceeding $570,000.

According to Webb and Zhivan, less than 15 percent of households approaching retirement have accumulated that much in total financial assets.  In other words, health care costs after age 65 could conceivably consume more than most households have saved to pay for their entire retirement.

Clearly, the greatest risk to retirement security regarding health care costs is nursing home care. Medicare benefits do not pay for most nursing home care costs. Long Term Care insurance is available but it is expensive and its cost-effectiveness for many families is debatable.  

Social Security Risk

Social Security has been under attack by fiscal conservatives since its inception in the 1940’s. Though they constantly argue it is about to "go broke", moderate changes like a two-year increase in the retirement age would leave these entitlements solvent for 75 more years.  So far, these attacks have been successfully repelled. Nonetheless, it would be unwise to overlook the risk that benefits might be reduced or delayed in the future.

These are some of the major foreseeable risks to successfully funding a retirement that could last three decades or more.  After retiring, we no longer have work income to help recover from financial setbacks and the longer we are retired, the more difficult it becomes to reenter the workforce.

You might not save enough for retirement, then again you might save more than you should.  If the stock market turns against you in the last decade of employment or the first decade of retirement, your plans for the Golden Years could be toast.  Social security benefits could be reduced, you'll be a fraud target, inflation will eat away at your purchasing power.  You or your spouse might need long term care that could cost more than you saved for all of retirement in the first place. If the risks don't scare the hell out of you, then you don’t properly understand the situation. 

On the other hand, nearly all of these risks can be mitigated or even eliminated if we plan retirement properly.  Avoiding these financial setbacks is as important as saving enough money for retirement in the first place.

---

¹ About 5% of $500,000 savings.

The Biggest Retirement Planning Mistakes

Talk to retirees about their financial situation and some of the worst retirement planning mistakes tend to pop up over and over again.  Here is my list of the big ones to avoid, but the biggest mistake is not to plan at all.

1. Not Planning

Retirement planning may not be fun and the initial results may be discouraging, but without a plan your odds of optimally or even adequately funding retirement are greatly reduced.  With a plan, you can make the most of your economic resources, whatever they may be, avoid pitfalls and perhaps fix some of the problems you uncover in the planning process.

Find a fee-only planner who will work with you to develop a retirement plan but won’t try to sell you financial products he or she will profit from.  If you can’t convince yourself that a planner is worth the cost, use my website to develop your own plan.  As a friend of mine likes to say, “Bad breath is better than no breath at all.”  That’s true of retirement plans, too.

2. Using Life Expectancy for Planning

Composer-pianist Eubie Blake famously said, “If I'd known I was going to live this long, I would have taken better care of myself.”  Blake was born in 1887, when life expectancy for males was about 59 years.  He lived to age 96.  If Eubie had planned to fund his retirement only until his life expectancy, he would have been broke for the last 37 years of his life.

According to the Social Security Administration’s website, the life expectancy for a room full of 100 randomly selected men born in 1940 is now about 83 years.  About 50 of them will live longer than age 83 years. Around 25 will live past age 90, and 10 or so will live past the age of 95.  

Don’t ignore longevity risk. Plan on living a long time, just in case you do.

3. Believing in Sustainable Withdrawal Rates

A widely-touted retirement funding strategy says that you can invest your retirement savings in stocks,withdraw 4.5% of your nest egg on the day you retire and withdraw that same amount every year thereafter, adjusted for inflation, and safely fund at least 30 years of retirement.  This is voodoo finance.

If you don’t want to take my word for it, ask anyone who retired using this strategy and had their standard of living cut in half by the 2007-2008 stock market crash.  Or, ask Nobel Laureate, William F. Sharpe, who debunked the strategy in his paper, “The 4% Rule—At What Price?”  

Or, you can ask retired Dallas News finance columnist, Scott Burns.  Burns, who was an avid supporter of the "SWR" strategy for years as a columnist, recently co-authored a book with Laurence J. Kotlikoff entitled Spend ‘til the End in which he now refers to the 4% Rule of Thumb as a “Rule of Dumb”.

If this is the best your financial planner can come up with, find a new planner.

4. Ignoring Inflation

Even mild inflation averaging 3% a year will more than halve your income over thirty years.  If you have a pension or annuity that has no cost of living adjustment, you need to plan for its inevitable and constant decrease in purchasing power throughout your retirement.

5. Drawing Social Security Benefits Before You Need Them

The longer you wait until age 70, the greater your social security retirement benefits will be.  If you can’t retire without them, then by all means take them when you need them.

But if you can delay your benefits by working a little longer or by paying for the early years of retirement from your savings, you may be able to improve your standard of living significantly.

6. Not Planning for Long Term Care

Medical expenses can destroy retirement savings, but the cost of long-term care, should you or your spouse need it, can be devastating.  Medical costs could easily total more than $250,000, but long-term care could cost more than half a million dollars.  Medicare helps with health care costs, but it doesn’t cover long term care.

7. Carrying Too Much Debt

Debt is a huge burden once the paychecks stop.  Try to pay off all consumer debt before you retire and consider paying down or paying off your mortgage.

8. Trusting Your Retirement Planning to a Salesman

Did you ever buy a car and drive off the lot thinking, “That nice salesman only cares about me driving away with a great car at the best possible price”?

Me, neither.

If your financial planner makes money by selling financial products, how can he or she have your best interests at heart?  If he sells stocks, for example, you can bet stocks will be a big part of the “right plan” for you – maybe even too big a part. 

As the saying goes, when you have a hammer everything looks like a nail.  Find a fee-only planner with an entire box full of tools and no incentive to pick the wrong one.

9. Saving Too Little (Or Too Much) for Retirement

Ideally, we would like to maintain about the same standard of living after retiring that we enjoyed before. We don’t want to scrimp through our careers in order to live a lavish retirement, or live beyond our means and then retire as paupers.

Consumption-smoothing software like ESPlanner.com can help determine the amount of pre-retirement saving that is required to achieve a similar standard of living after we retire.

10. Establishing the Wrong Priorities

Saving for retirement is important, but it isn’t always the top priority.  People with a lot of expensive consumer debt would be better off retiring that debt before retiring themselves.

Retirement savings may well be more important, however, than paying for your kid’s college.  Your children have a lot more time to repay college loans than you have to save for retirement.  Besides, what kid wants you to pay for their college tuition if it means having to move in with them when you’re 80?

Young people right out of college would be better off maintaining a decent standard of living, avoiding consumer debt and building an emergency fund than saving for retirement.

Not everyone has the same priorities, but understanding yours and putting them in perspective is critical to a successful financial plan.

This is by no means an exhaustive list of mistakes people – even professional financial planners – make when planning retirement.  Use it as a checklist when reviewing your retirement plan to make sure you have covered the obvious. 

How Long Will Retirement Last?

Perhaps the biggest challenge in planning retirement is dealing with the uncertainty of how long we will live.  Economists and financial planners refer to this as longevity and the possibility that we will outlive our savings as longevity risk.

We could spend more every year of retirement if we knew we would only live to age 75 than we could spend if we knew we would live to age 95.  Of course, we don't know.  In fact, most of us literally have no earthly idea when we will die, so we need to choose a way to plan for that uncertainty.

It's a very important decision.  No one wants to find themselves bankrupt in old age after it's too late to reenter the workforce.
 
There are two basic ways to fund retirement: with annuities or with rationing.

Annuities are contracts an insurance company will sell you that promise to pay you a certain amount monthly or annually for the rest of your life, no matter how long you live.  Social Security benefits are an annuity, too, except they are "sold" by the U.S. government instead of an insurance company.

Here is an example from immediateannuities.com. A 65-year old man living in North Carolina with his 65-year old wife could pay $100,000 today to an insurance company for a fixed annuity that would pay them in return about $538 a month for the rest of their lives, no matter how long that might be.

Annuities insure against longevity risk. They eliminate the risk of outliving your wealth. If all your retirement income came from annuities, your annual income would be the same no matter how long you lived. Of course, the people who buy these annuities and live to 95 will get much more for their $100,000 than people who buy the same annuity and only live to 70.  Those who live to 95 will get 360 monthly payments of $538 instead of just 60 payments.

A lot of people can't seem to get past that. They think, ". . . but what if I die before I get really old and the insurance company doesn't have to pay back that entire $100,000?" On the other hand, people who buy these annuities don't have to worry about outliving their savings.

The alternative to annuities is rationing. We can save money and instead of paying it to an insurance company for an annuity, we can invest it and ration it to ourselves.

Rationing, unlike annuitizing, is totally dependent on how long we believe we will live.  Imagine you were stuck on a desert island waiting for rescue and only had the food you brought with you. You would eat less each day if you expected to be marooned for a week than you would eat if you knew you might be stranded for a month.

Here's a point, though, that many of us miss when we think about retirement.  We have to ration for the longest time we believe we might be marooned on that island, not the average.

Suppose we knew that people marooned on islands are rescued  in one week on average, but some people have been stranded as long as a month.  We would want to ration our food to last 30 days, not 7.  If people rationed for the average duration, half of them would run out of food before being rescued.

Similarly, planning for retirement based on life expectancy— the average age people like us will attain— is too risky.  It leaves a 50% chance that we will outlive our savings.

Which leads us to the crux of the matter. We need to plan to live to a ripe old age, say 90 or 95, just like we would need to ration food on that island for the worst case, 30 days, because running out of money in old age and running out of food are both intolerable outcomes.

The Key Points

• If we buy fixed annuities, we are guaranteed a paycheck for life.  The risk is that we might die not long after purchasing the annuity and not receive many payments from the insurance company.

• If we don't purchase fixed annuities, we can invest our savings and ration them.  The problem with this approach is that we don't know how long we will live.  

• To avoid the possibility of outliving our savings with the rationing approach, we have to plan on a very long life, 95 years for example, and that means spending a lot less each year than we could spend if we assumed we would only reach our life expectancy.

• Annuities generally provide more annual income than the rationing approach, but the rationing approach improves the odds that we will leave an inheritance.

• And most important of all, a retirement plan has to assume that we will live much longer than our actual life expectancy, because we just might.