How to Develop a Cohesive Retirement Plan

A Discussion of the Critical Issues

Contrary to popular belief investment management is not a subject based primarily on opinions. Investment management is a science with a developed methodology, a set of investigative tools, standards by which it may be evaluated, and a large body of literature developed by theorists and practitioners from around the world. Indeed, its validation as a legitimate scientific discipline reached the same level as that of other more established sciences in 1990 when three of its practitioners were awarded Nobel prizes.1 On the practical side investment managers must distill a lot of information, simplify it, and make it understandable to investors. Specifically, they must be able to help you answer three basic questions about your retirement portfolio: "How much do you need to save to meet your retirement objective?;" "What level of risk can you afford to take?;" and, "What is the probability that you will live out your years without exhausting your investment nest egg?" As mundane as these questions may sound, finding acceptable answers requires quite a lot of serious work. In the search for answers facts must replace opinions. This paper presents the rationale and the method for answering these questions with a much higher degree of certainty than was possible in the past.

Retirement planning is everyone's biggest investment responsibility. In the aggregate we are all living longer and retiring earlier, therefore we need to accumulate ever larger sums of money to sustain us during retirement. According to the IRS for every two retirees age 65 today, the odds are 50/50 that one will live to age 90.2 This means that most of us need to plan for a retirement period far beyond that of our parents and grandparents. It also means that we need to carefully manage our retirement funds for a long time after retirement. There are several other issues to consider that argue for more precision in retirement planning. As we grow older our skills become less and less marketable in the job market. After age 70 or so, many of us will become virtually unemployable, thus we must rely exclusively on our financial assets at the same time that we are depleting them. We must therefore find ways to reduce the margin of error in our planning. If we should miscalculate living expense increases, or realize large unexpected medical expenses--or see a prolonged downturn in the financial markets--we could run out of money long before we run out of time. And going broke during retirement could be catastrophic. Because we could not recover losses, it is the clearly the most serious investment risk we face. Affluent retirees, of course, can meet financial contingencies with funds from many different sources: from family businesses, inheritances, pensions, Keogh accounts, company retirement plans, real estate, social security, deferred compensation, guaranteed investment contracts, fixed annuities, and cash. They will likely be more concerned about how to arrange and distribute their estate after they die than about how to manage their assets during retirement. By contrast, most of us must rely on a company retirement plan, social security, and our own savings to fund our retirement. Retirement planning for the great majority is therefore a lifelong activity. With these issues and background in mind, let us now walk through the steps in developing a cohesive retirement plan.

The Accumulation Stage: How Much is Enough?

This piece of the retirement puzzle is relatively straight forward in terms of its arithmetic. The next table shows several examples in which the preliminary retirement accumulation is established.

These calculations are easy enough to make with any business calculator. Like any projection, however, they are not a prediction of what will happen; rather, they show what is possible and potential. The purpose of such projections is to find the monthly contribution and rate of return that will let you reach your retirement objective on your timetable. Once you are satisfied with each component, then you can set into motion a specific plan to make it happen.

Caveats. In order to determine if any of the preceding calculations is sufficient to fund a retirement, several more calculations are needed. First, we must determine the present value of the future accumulations. Thinking in terms of future dollar values can be deceptive. Reducing them to current dollars (stripping away expected inflation) simplifies this task by allowing us to think of future dollars in terms of today's purchasing power. Again, any standard business calculator can be used to make these calculations:

Present Values of Future Accumulations

Notice that each future accumulation is approximately equal when inflation is stripped away. Whether any given nestegg is sufficient to fund a retirement, however, still depends on other factors: present income and standard of living, needed income after retirement, and life expectancy. With estimates of each, let's look at the depletion period during retirement. In the following example, the vehicle that guarantees future income is the fixed annuity.

Retirement Income Calculations

Assumptions: Pre-retirement income = $100,000.
Needed retirement income = $80,000.
Funding vehicle = fixed annuity.
Planning period = 30 years.
Ending Accumulation = $0.0

* Social security estimates use 1997 data in which the primary beneficiary receives the maximum individual benefit, and the spouse receives 50 percent of the primary amount.

Discussion. The first noteworthy observation about these data is this: Anyone with an annual income of approximately $100,000 must think of retirement dollar accumulations in excess of $700,000. Indeed, nearly $750,000 is needed to fund an annual income of $80,000 per year, assuming the recipient and his/her spouse also draw the maximum social security benefit. The second observation is that none of the retirement accounts shown will fully fund a retirement plan under the assumptions given. All must rely on personal savings to augment incomes. In the case of the first account, an additional personal accumulation of approximately $200,000 would be needed to make up the shortfall. Because we are still in the planning stage, we are able to alter many of the assumptions. As long as we stay within the bounds of what is possible, we could continue to make changes until we achieve a favorable outcome. For example, we could assume higher savings in either a company or personal retirement plan; we could assume a higher annual rate of return during the accumulation period; or we could assume a higher return during retirement. Higher rates of return, however, involve higher levels of risk. And that opens a second area of discussion. Let's now consider the level of risk someone may need to take in order to accumulate the future dollar totals shown.

Risk Taking as it applies to the financial markets has both practical and psychological dimensions. The practical dimension is concerned with ability to take risks, while the psychological dimension is concerned with willingness. Ability relates to facts: your age, time to retirement, income, living expenses, debts, and the value of your current holdings. Whether or not you are psychologically disposed to risk taking, you must first have the wherewithal to do so.

Willingness deals more with preferences, safety, and comfort levels. This dimension is probably overemphasized in measures of risk tolerance. Questionnaires designed to tap the psychology of risk typically pose games and other hypothetical situations in which there is no chance of a loss and no bad consequences, and ask you to respond to them. The problem with this approach is that during retirement, your chances of loss are real and with irrevocable consequences. You should not, threfore, give hypothetical game playing more credence than your own experience. A further problem with measures of willingness is that the author assumes that willingness is unchangeable. Although a risk tolerance questionnaire might suggest an aversion to financial risk, the need to accumulate a sum certain in a fixed time period might also change that person's willingness to take additional risk. Willingness therefore has a rational as well as an emotional component.

By contrast we believe that needs are the most important consideration to investment risk taking. Someone who is 45 years old, who currently has $50,000 in retirement assets, the ability to save $250 each month, and who needs to accumulate $660,000 by age 65 must also be willing to take the risks necessary to achieve the needed future accumulation. Although investment returns are neither fixed nor guaranteed, time and compounding are both. Other elements of risk aside, 11 percent is the minimum average annual return that will let $50,000 plus $250 per month grow to $660,000 in 20 years. The arithmetic is indisputable. Anyone who is not willing to assume the risks associated with market returns of 11 percent must consider a lower standard of living during retirement. These facts should supersede risk tolerance scores in planning efforts.

Our own risk tolerance questionnaire avoids hypothetical situations in favor of direct measures of ability and willingness. Although we believe it is well grounded in theory and fact, it has yet to be validated by the tools of science. With additional data we intend to answer the three important scientific questions about it: "Does it measure what it purports to measure?" "Does it produce consistent results over time and sampling content?" and "Is it useful for a specific purpose?"

The final issue that should be resolved before your actual retirement is this: How much can you safely withdraw from your savings during retirement without going broke before you die?

Withdrawal Stage

Once you are retired with ample funds, the common belief is that you can now relax your investment management tasks. That could be your biggest mistake. In spite of your best calculations, you still face the possibility of going broke. How to prevent going broke during retirement is a crucial piece of everyone's retirement puzzle. Should it happen, your self esteem and self reliance would be shattered, you might need to rely on relatives, welfare, or worse--all at a time when there would be little you could do to change your circumstances. The purpose of this section is to help you avoid becoming such a victim.

The Problem. Picture this scenario: You are age 78, in bad health, and flat broke. You retired 13 years earlier with $800,000 in your retirement account. Your financial planner told you that you could expect 9 percent annual returns, and that you could draw down 6 percent per year indefinitely with no possibility of running out of cash. But there you stand, broke and retired--with no hope of financial recovery.

What could make this picture happen? Less than you may think. A combination of chance factors could do it. You could spend more than expected, the financial markets could slide into a prolonged decline, inflation could rise faster than expected, or some combination of the three could happen. Moreover, if you relied on constant investment returns and expenses through it all, and you kept the same investment mix and monthly withdrawal amount, you would have helped your own disaster along.

Suffice to say risk management neither begins nor ends with retirement. Your financial future at any age is too important to be left to chance or opinion. It requires a detailed analysis of your circumstances in a new way. The balance of this paper contains new risk analysis information that is not generally available to even the most astute banker, broker, or financial planner.

The Solution. Solving the probability of going broke during retirement begins by calculating a few statistics from your present assets. Specifically, you need to know your portfolio's overall risk and rate of return. Then you need to determine the range of inflation that is most likely to occur, and the amount you will need to withdraw from your assets to meet your expenses. These calculations are relatively routine and require no special skills. The next set of calculations, however, are not routine and require considerable skills.

Monte Carlo Simulation. The term "Monte Carlo" was introduced during World War II as a code name for simulation of problems associated with development of the atomic bomb. A few algorithms were used to generate random samples from different kinds of probability distributions. In so doing, a whole range of possible outcomes could be determined in advance. Then by altering the inputs, the probability of achieving the desired outcome was increased. In addition, the most sensitive inputs could be identified so that resources could be used to control them without wasting these same resources on other less important factors.

The same techniques can be used to control investment risk, but now there are more probability distributions, more sampling techniques, and much faster calculation speeds than were possible 55 years ago.

Finding the probability of a total loss of your investment funds means that we must simulate a large number of possible and potential futures. Specifically, we must identify the range of futures that could happen given your personal circumstances. In some, your assets will rise in value, in others they will hold their value, while in still others, they may fall. We can't pick our own future, but we can protect ourselves against the worst of the bunch. The purpose of simulation is to find out what the odds are of going broke, then reduce those odds to very low levels. We do that by creating a mix of assets in your portfolio such that the combination of risk level and withdrawal rate is most likely to keep your balance positive or growing indefinitely given the market's volatility and the probable inflation rate.

Simulation requires that we make assumptions about future market risks and rates of return, inflation, and expenses. Then we test our assumptions, alter them if the results are unacceptable, and continue the procedure until we find an acceptable solution. As cumbersome as all this sounds, the simulations can be done quickly with today's computers, appropriate software, and experience with the procedures.

Simulation Example

Age: 65.
Needed Income: $48,000 initially, increasing annually with inflation.
Type of Portfolio: Balanced.
Portfolio Balance: $800,000.
Expected Annual Rate of Return: 9%.
Volatility (standard deviation): +/- 10%.
Inflation Range: 2.5 to 4.5%.
Projection Period: 30 years.

Simulation:
Sampling: Monte Carlo.
Number of iterations: 1,000.
Output Distribution: Normal.

Results:
Probability of exhausting all funds before 30 years: 39%.

Discussion. This example is typical for an affluent retiree. It also seems reasonable that this account should grow over a 30 year time period. But will it? Our simulation shows there is a 61 percent chance that this person will be able to draw down the desired income from his/her investment accumulation over the next 30 years without going broke. But there is also a 39 percent chance that he/she will go broke. The $800,000 question is this: Are you (the retiree) willing to risk a 39 percent chance of going broke? Assuming you are not so willing, then we can do several things: First, we can change the mix of assets to reduce the volatility, and rerun the simulation. Then we can optimize the risk/return tradeoff so that this portfolio is efficient, and rerun the simulation. Then we would alter the income level and repeat the procedures until we have achieved an acceptable probability that you will realize a satisfactory income level during all your retirement years.

Concluding

Uncertainty about the financial markets coupled with greater longevity and higher expectations about our standards of living have given rise to the need for most of us to plan our retirement years with greater care than was necessary in the past. In particular, we all need to start early, build up higher levels of wealth, and become better risk managers.

Above all else, going broke during retirement is now preventable. There are no easy answers or short cuts to assuring that end, however. There is a unique answer for everyone based on personal circumstances. Simulation is the key to thoroughly analyzing and solving this issue.

Without this kind of analysis most retirees will still survive without having to confront an unhappy financial ending. Unfortunately, others will not. Even large amounts of accumulated wealth can be exhausted if not properly managed. A good risk manager today must have the means for helping you overcome this final investment risk.
______________________
1 Profs. William Sharpe, Merton Miller, and Harry Markowitz received the Nobel Prize in 1990.
2 IRS publication 590.