Beyond Greed and Fear
Page 20
To what extent do you agree or disagree with the following three questions? Answer each question on a six-point scale, where 0 denotes unsure, 1 denotes completely disagree, and 5 denotes completely agree.
1. I am on track with my retirement planning, and I am confident of a comfortable retirement.
2. I know what my living costs will be after retirement.
3. I know what my pension and Social Security benefits will be in retirement.
The Roths’ financial planner posed these questions to them. Ira’s responses were 2, 2, and 1. In effect, he disagreed with all three statements. On the other hand, Jeannie responded with 4, 3, and 4: She agreed somewhat or was neutral. In this respect, Ira and Jeannie straddle the typical case. The mean responses to these questions are 3.5, 2.3, and 2.0 respectively.2
Look at the declining pattern in these responses. Despite feeling somewhat on track, most people know neither what their needs will be nor what their means will be once they reach retirement. These responses are consistent with the findings of a Wall Street Journal/NBC News poll conducted in December 1997 by the polling firms of Peter D. Hart and Robert M. Teeter.3 They found that although 42 percent of the 2,013 people they surveyed were confident that they will have enough money to live in retirement, 57 percent did not know how much they need to save in order to reach their retirement goal. But of the 55 percent who are saving for old age, 26 percent had accumulated no more than $10,000.
The Hart-Teeter results are similar to findings in the general academic literature. Consider the personal financial assets of the median family headed by someone between the ages of 55 to 64. James Poterba, Steven Venti, and David Wise (1996) report that in 1991 that family had personal financial assets amounting to only $8,300. Almost 20 percent of such families had no financial assets at all, let alone enough to fund a comfortable retirement.
Self-Control: Facing Temptation Along the Road to Retirement
Why do so many people lack the self-control necessary to set aside adequate financial resources for retirement? In articles published in the 1980s, Richard Thaler and I (Thaler and Shefrin 1981; Shefrin and Thaler 1988) suggested that a major factor is insufficient self-control. The needs of the present make themselves felt through emotion. Those needs have a strong voice and clamor for immediate attention. In contrast, the needs of the future have a much weaker voice, expressing themselves more through thought. Most people feel the urge to satisfy their immediate needs, but they only think about satisfying their future needs.
The temptation to satisfy present needs is everywhere. Tantalizing stimuli generate impulses that call for a gratification response. Saving for retirement while at the same time facing temptation requires inordinate amounts of willpower. Most people do not have that much will-power. So what should they do? It’s actually what they should not do: They should not face temptation at the same time that they save. They should not place themselves in situations where they make the thin voice of retirement saving compete for a portion of take-home pay with the shouts for food, entertainment, and vacation travel. The voice for retirement saving will get drowned out. What is required is a system like that used in golf. Retirement saving needs some form of handicap.
The right type of handicap for retirement savings involves money “coming right off the top,” so that retirement needs do not compete directly with those for food, entertainment, and travel. Retirement saving is more likely to get accomplished through a regimented routine rather than through discretion. That is a crucial feature of 401(k) plans and 403(b) plans: Funds are automatically deducted from gross income.
Investors benefit because these retirement savings programs are both tax deferred and framed in terms of an employer contribution as well as an employee contribution. If it takes the stimulus of a reduction in one’s current tax bill to generate the emotion required to induce participation in the company’s 401(k) plan, then fine—whatever works.
Some people also use IRAs to save. And we learn one very important lesson about 401(k) plans and IRAs from Poterba, Venti, and Wise (1996). IRAs and 401(k) plans are the major vehicle through which people increased their retirement saving between 1984 and 1991. These authors use the Survey of Income and Program Participation (SIPP) to track the composition of household assets from 1984 through 1991. The households in SIPP contributed to a 401(k) plan or an IRA, or to both. For families with a 401(k) plan but no IRA, the median value of their total financial assets rose from $8,566 to $9,808 from 1987 to 1991. However, the portion of their assets outside their 401(k) plan fell slightly, from $2,587 to $2,498.
A similar pattern held for those households who were not eligible for a 401(k) plan but did contribute to an IRA. Total financial assets rose from $20,686 in 1984 to $27,094 in 1991, but most of the increase occurred in their IRA accounts. Non-IRA assets rose only slightly, from $13,098 to $13,355. Note that IRA contributors tended to be wealthier than non-IRA contributors.
Mental Accounting: Saving Along the Road to Retirement
Not all saving takes place in 401(k) plans and IRAs. But saving rarely holds its own in the battle for regular take-home pay. Those who do manage to save additional funds do so by dividing their money into different mental accounts.
Here is an example to help get the point across. Imagine that you will receive a special bonus over and above your regular compensation. This bonus will be paid out of the ordinary cycle; it will be paid monthly over the course of a year and will increase your after-tax take-home pay by $500 a month for the next twelve months, for a total of $6,000. Think about how your family’s consumption might change as a result of this special bonus. Specifically, by how much would you expect your family’s monthly consumption to increase during the next year? Notice that you are being asked what you honestly think would happen, which may or may not be the same thing as what you would ideally like to have happen.
Now suppose that your $6,000 (after-tax) bonus will be paid to you in one lump sum. Think about whether you would treat the bonus money any differently now, especially with respect to your expenditures in the month immediately following receipt of the bonus.
What happens if the $6,000 is not a bonus but an inheritance? Imagine that a distant relative has died and left you a small inheritance with an after-tax value of $6,000. You will not receive the money for five years. During that time the money will be held in a unit investment trust (UIT), where it will be invested in conservative interest-bearing securities. At the end of five years, you will receive the $6,000 plus interest. By how much would you expect your family’s consumption to rise over the next twelve months because of this gift?
Given these three possibilities, how do people tend to respond? Thaler and I posed these questions in a survey of MBA students in 1987. I repeated them ten years later and obtained essentially the same results. I also surveyed a group of professionals who work in the financial services industry and found that they respond in the same way. Figure 11-1 depicts the results for 204 individuals surveyed between 1996 and 1998.
Figure 11-1 Monthly Consumption Spending as a Function of How the Bonus Is Received
How much of a $6,000 windfall people decide to save depends on the form in which they receive the money. The spikes at the left indicate that most people are inclined to save the entire $6,000 when it arrives as an inheritance in the future, but they consume it quickly when it arrives as part of regular income.
Begin with the three spikes at the left. These spikes depict the percentage of respondents who plan to save the entire bonus. Notice that this percentage is lowest when the bonus arrives in their regular take home pay, and highest when it arrives as an inheritance. At the right of the figure, notice something else: Some people tend to spend a sizable chunk of their bonus when it arrives as a lump sum.
Figure 11-1 shows that people treat the $6,000 bonus differently, depending on the form in which they receive the money. Part of this difference is due to the way that they allocate the money internally. If the bonus ar
rives as part of their regular paycheck, they tend to allocate it to their current income account. This account gets used to fund regular expenses. If it arrives in a lump sum, then people tend to react in one of two extreme ways. Either they place it off limits at once, by allocating it to a wealth account; or they treat themselves to a large purchase of some sort or another, placing the remainder into a wealth account.
Finally, if they perceive the money as arriving at some point in the future, then it enters their future income account. This account is usually way off limits to consumption. Almost 90 percent of the sample save the entire amount. Many of these folks hold large liquid assets. They could draw these down in anticipation of the time when the inheritance will arrive as cash in hand. But most don’t. They use mental accounting to help them to both save money and safeguard their past savings from being accessed too early.
This example serves to illustrate how people engage in self-discipline. They use crude rules, based upon associations they find helpful. One of the common themes discussed in this volume is that people tend to categorize. They use mental accounts to help themselves simplify complex problems. In this case, they build on that mental accounting structure by developing rules that help them deal with their impulses.
Not everyone is fortunate enough to get a bonus. But most of us pay income taxes. And many use the Internal Revenue Service (IRS) to provide themselves with a simulated bonus. How? We overwithhold during the year, and receive a sizable check from the IRS at tax-filing time. In my own survey of investors, I find the average planned income tax refund to be between $800 and $900. For some, this is just precaution. They are somewhat uncertain about what their income will be, and prefer not to have an unplanned expenditure on April 15.
Others consciously use overwithholding as a self-control device, knowing that they will save the refund, whereas they would spend the money if it appeared in their regular paycheck. One investor in my survey received a $25,000 income tax refund in the previous year, and was planning for a $12,500 refund the next. That $25,000 could be earning a return! Oh, and what does this particular investor do for a living? He’s a financial planner. I think he understands the concept of foregone interest. But he is willing to pay the cost in order to discipline himself.
Myopic Loss Aversion: Taking Risk Along the Road to Retirement
There are at least two different ways that myopia can plague investors saving for retirement. We have already encountered the first way: insufficient attention to retirement because it seems so far off. The second way concerns investors’ attitudes toward facing risk. Many investors choose portfolios that are too conservative for the goals to which they aspire. The question is why.
Here is an example that enables us to get to the heart of the issue. Consider a lottery whose outcome is determined by the toss of a fair coin. If the outcome of the coin toss is heads, you win $2,000. If the outcome of the coin toss is tails, you lose $1,000. If you had the opportunity to play this stylized lottery exactly once, would you?
Losing $1,000 can hurt, even when you stand an even chance to make $2,000. Less than 40 percent of those I survey indicate that they would take this bet. Perhaps the size of the stakes has something to do with it. What happens if we go with lower stakes—say, $100 instead of $1,000 and $200 instead of $2,000? It does make a difference. Now 60 percent say that they would take a chance. But that still leaves 40 percent on the sidelines, fearful of losing $100.
Would a repeated bet make accepting the gamble more attractive? Suppose the offer includes playing the lottery exactly twice, keeping the stakes at $100 and $200. Up to 70 percent would go for the two-shot gamble.
Finally, what would happen if instead of a two-shot gamble, we offered the opportunity to play the lottery 100 times in a row? The proportion of takers in my sample climbs from 70 percent to 90 percent.
Notice what has happened to people’s tolerance for risk as we went through all four possibilities. In each case, respondents became more tolerant of taking the risk. In particular, tolerance for risk increased as the number of repetitions of the gamble went up. Why? And what is the general principle at work here?
Shlomo, Benartzi and Thaler (1995) suggest that people respond to the preceding questions as they do because of myopic risk aversion—or, to put it more accurately, myopic loss aversion. Those that turn down the first bet do so because the $1,000 loss looms larger than the $2,000 gain. Similarly, for those reluctant to take the same bet with the smaller stakes, the risk of losing $100 outweighs that of winning $200.
People evaluate one-shot gambles in isolation from other decision problems and record the possible outcomes for each of these gambles in a separate mental account. But playing the smaller stakes gamble multiple times offers a form of time diversification, where people feel that the law of averages is on their side. Now they only need a head to be tossed on one of the two rounds, and they will end up registering a gain. Their chance of incurring a loss is still 25 percent. But the probability they will gain has risen from 1 in 2, to 3 in 4. Of course, the magnitude of a loss, if it occurs, has doubled from $100 to $200. But the maximum gain has also doubled, from $200 to $400.
There are two separate and important issues involved in the above example. First, many people cannot tolerate the risk inherent in this gamble. Now what drives their low tolerance for risk in this situation? One thing is their low tolerance for loss, their fear of loss. Given that the odds of a loss are even, those who turn down the opportunity to play are telling us that they fear the pain of a $100 loss more than the pleasure of a gain of twice the magnitude.
Second, those that turn down the one-shot gamble but accept the multiple-shot risk display myopic loss aversion. That is, they frame every gamble they face in isolation from all others. Yet people face small gambles of one sort or another all the time. Most such gambles are independent of one another. In this sense, people continually experience the advantage of the law of averages for small gambles. But people don’t frame that experience in an integrated way. Instead, they frame every gamble as a one-shot deal rather than as one more turn on the multiround gamble road of life.
The offer of a two-shot version of the lottery induces people to frame the gamble in multiround terms. In effect, this stretches the perceived time horizon of the problem, making people less myopic. Moving from two rounds to one hundred rounds makes people less myopic still—and less averse to single-round losses.
Benartzi and Thaler (forthcoming) suggest that many investors hold overly conservative portfolios. Why? Because they overfocus on the potential for short-term losses. Myopic loss aversion leads investors to hold too little in equities and too much in fixed-income securities. Horizon is critical. When looked at in months, stocks have made money 62 percent of the time since 1926, with the average loss having been almost as large as the average gain. But when looked at in five-year segments, stocks made money 90 percent of the time, and the average loss was only 63 percent the size of the average gain.
Loss aversion is perhaps the major element that drives investors’ tolerance for risk. Now, loss is partly perception, and therefore is subject to framing. The expression “If someone hands you a lemon, make lemonade” concerns the reframing of a loss. Investors perceive a loss as having to sell something for a lower amount than they bought it for. People fear having to realize what they regard as losses.
Kenneth Fisher and Meir Statman (1999a) discuss the fact that an important element in time diversification is the ability to postpone, or avoid altogether, the realization of a perceived loss. This is one reason why some investors favor Treasury securities over stocks. Holders of Treasurys need never recognize a perceived loss on their bond purchases as long as they are willing to hold those Treasurys to maturity.
That is why some financial planners recommend the use of “ladders”—purchasing a series of Treasurys of different maturities, and continually replacing those that mature. Although the market value of a laddered portfolio may fluctuate over time,
and investors may even find that the value of their portfolio dips below the purchase price, they need not focus on this fact. For them, having a ladder enables them to frame their outcomes as being one gain after another rather than a capital loss. After all, the capital loss reflects an opportunity cost for them, not an out-of-pocket cost. And as Thaler pointed out in his classic (1991) article, out-of-pocket costs loom much larger than opportunity costs.
Myopic loss aversion leads investors to choose portfolios that are overly conservative. But myopia is not the only factor operating on portfolio selection, and its effect can be cancelled by excessive optimism. Excessively optimistic investors may select portfolios that are unduly aggressive. Benartzi and Thaler conducted their study in the early 1990s, before the dramatic surge in stocks that began in 1995. As I discussed in chapter 5, those investors who have experienced a bear market are considerably less optimistic than those who have not.
Dollar-Cost Averaging: Acting Prudently Along the Road to Retirement
Financial planners are big on dollar-cost averaging, a practice closely related to myopic loss aversion and time diversification. Here is how financial planner Ginita Wall (1995) describes the practice in her book The Way to Invest.
Dollar-cost averaging is a fancy term for investing money over time. In dollar-cost averaging, you invest a set amount each month, acquiring more shares when the price is low and fewer shares when the price is higher. The result is a lower cost per share than if you bought a set number of shares each month. …
Dollar-cost averaging has an added advantage, especially with an automatic investment plan, when you invest a set amount each month. It helps keep you from panicking when markets go down. You can be delighted that the market has dropped and you are able to acquire so many more shares that month, rather than become frightened and pull all your money out of fund just when the market reaches bottom. …