Article

Can Annuity Purchase Intentions Be Influenced?

The findings in this study are applicable to not only purchase of annuities, but all financial products as well. The more effectively we as professional retirement counselors frame the education and guidance we give consumers and employees, the more receptive, desirous and confident consumers and employees will be to take steps to improve their retirement security. – the Editor

For many baby boomers their retirement savings are solely in defined contribution and IRA accounts where they personally assume the full effect of market and interest rate volatility and inflation on the value of their savings and income during retirement. There is a high likelihood they’ll need to annuitize a portion of their savings if they’re not willing to postpone Social Security long enough to provide a more solid lifetime retirement income foundation.

One of the most important roles of a retirement counselor is to identify and evaluate the money behaviors and financial/retirement literacy of their clients and incorporate this information into the retirement planning and income management recommendations that might best fit their clients’ needs for retirement security. You’re probably already aware of the underlying, overall negative bias toward the word “annuity” that prevents consumers, and many advisors for that matter, from utilizing insured retirement income products that might be highly appropriate for people who are at real risk of running out of money before they run out of retirement. Several studies have evaluated the “annuity puzzle” over the past decade where rational reasons fall short of explaining why consumers indicate they prefer receiving income for the rest of their life from their savings, but are less inclined to purchase an “annuity.”² The findings of “Can Annuity Purchase Decisions Be Influenced,” an experiment by the Society of Actuaries, sheds light on how pre-retirees, retirees, and women are inclined to use an annuity in their retirement income plan, and how practitioners might better frame their discussions with clients when educating them about the role annuities can play in securing their retirement income plan.

Research Design

The study was designed to assess if the 1) intention and 2) likelihood of purchasing annuities from current savings might be influenced (the study did not measure whether annuities were actually purchased). The following four approaches of presenting information were evaluated.

• Very simply and neutrally stating factual information related to retirement ages of Americans, life expectancy and suitable use of annuities in generating retirement income.
  
• Explaining how behavioral biases can negatively influence anticipated retirement age, life expectancy, and annuity purchase decisions.
  
• Providing both factual and behavioral bias information (items 1 and 2 above).
  
• Presenting an anecdotal scenario that negatively framed the effects of not having enough lifetime income.
  

For all four approaches, survey respondents³ were presented with the following hypothetical situation and a photograph of “Bill” to make the scenario more meaningful: “Bill Marks is a typical American employee. He has a full-time job and is in good health. At what age would you expect Bill to retire from his primary career, assuming he is a typical American employee?”

Women were presented the same scenario for a hypothetical woman (Sarah Jones). Other questions included:

• Until what age would you expect John (Pam) to live, assuming he (she) is a typical American man (woman)?
  
• How likely do you think Dave (Judy) would be to buy a life income annuity?
  

An explanation of the four approaches

Survey respondents were either presented with one of the four information approaches, or not presented with any information (the control group). Survey respondents presented with factual information were told:

• that eight out of ten retirees retired before age 65, and nearly three out of ten retired before the age of 55
  
• of same-gender life expectancy and probability of living to age 90
  
• of a very basic definition of a life income annuity, and that “research has shown that purchasing a life income annuity makes sense for most Americans because it protects them from ever running out of money when they live longer than they had planned to live.”
  

To test the influence of behavioral biases, respondents were told that certain behavioral tendencies such as optimism and anchoring (over-reliance on a specific piece of information to govern decision-making) can affect how people choose their expected retirement age. They were provided with an explanation of these biases and a description of how they might negatively impact the way people respond.

In the combination approach, survey respondents received both the factual and the behavioral bias information.

For the anecdotal approach, the following scenario was posed: “Dave (Judy) did not buy a life income annuity to protect himself from running out of money. Instead, he decided to put his retirement savings in an account with the bank, thinking that would be the safest thing to do. Things were fine for the first 10 years of his retirement when he was able to maintain a comfortable lifestyle, but over time, his account dwindled. Eventually, he couldn’t afford to live in his home, and he was quickly running out of money. He wasn’t sure where to turn. He could move in with his son, but he knew it would put a strain on his son’s family life. In the end, that was the only thing he could do, and it was a very difficult situation for all involved. Had Dave purchased a life income annuity, he would have never run out of money and could have continued to pay for some, if not all or most, of his expenses.”

All respondents were then asked to answer “How likely are you to buy a life income annuity?” and “What do you think you (or your spouse/partner) will do with some or all of your retirement savings?”

• Invest or save it in accounts from which you withdraw as you like
  
• Invest or save it in accounts that provide a series of regular payments
  
• Spend it or use it to pay down debt
  
• Buy a life income annuity that provides guaranteed monthly payments for as long as you live.
  

These final questions were designed to compare both the effect of the different information approaches on purchase intentions, as well as how the approaches might have influenced their attitude toward annuities. 

Which approaches proved most effective?

One of the study’s most important findings is there was a difference in purchase intensions as a result of the approaches depending on whether the person was a pre-retiree or a retiree. The window of opportunity to influence annuity purchase behavior may end upon retirement.

Pre-retirees were three times as likely to report they would buy an annuity with either the factual information or the anecdotal approach that negatively framed the effects of not using annuities, as compared to the control group. The combination of factual and behavioral bias approach was the third most effective approach with pre-retirees. The behavioral bias approach by itself was least effective with pre-retirees, though it was more effective than providing no additional information.

Interestingly, none of the approaches were more effective with retirees than another, which suggest their opinions may be ensconced upon retirement.

Other findings:

•  People who reported themselves as being more familiar with annuities had higher intentions of purchasing an annuity.
  
• Women retirees were nearly 11% more likely to intend to purchase an annuity.
  
• Approximately only one-third of respondents answered all five literacy questions correctly. In addition:
  
• Consistent with prior research, over half of respondents incorrectly answered a question about the relationship between interest rates and bond prices.
  
• Subjectively, retirees rated their financial knowledge slightly more highly pre-retirees did. The objective results, however, showed no significant difference in knowledge between the pre-retiree and retiree groups.
  

So how might retirement counselors apply this information?

There was not a discussion of trade-offs and downsides to annuity purchase within the information approaches presented to respondents, so while the experiment indicated that people acted differently with different information, it did not necessarily provide a good indication of what people will do.

Here are some thoughts to keep in mind when working with clients who should prudently consider annuitizing a portion of their savings at some point in time because they are at risk of running out of money toward the end of their retirement:

• Many people are over-confident about their investing skills. It might be worthwhile incorporating a financial literacy quiz or an exercise within the course of your discussions like the one used in this study to benchmark client knowledge before delving into retirement income solutions. Combining this with identifying how much control the client wants to have in making income decisions should narrow the field of potential income solutions.
  
• This research also found that people who were in good health were significantly less likely to be open purchasing an annuity. If their health assessment is correct, however, these individuals are likely to live longer lives and are therefore more at risk of running out of money. A basic financial management approach used by corporations and pension funds is to match the duration of their assets to their liabilities. Help the client understand the importance of matching their lifetime income (their assets) to their essential expenses (their liabilities).
  
• Be aware of gender differences, particularly within your retired client population. In this study, retired women were appropriately more inclined to purchase lifetime income solutions as they are at greatest risk of outliving their savings. However, a husband might not want to give up control of assets. If dealing with both a husband and wife, make sure you have three plans: one for if he dies first, one if she dies first, and one if they both live longer than expected.
  

¹ Society of Actuaries’ Post Retirement Need and Risks Committee, 2002 Retirement Risk Survey
² Sponsored by the Society of Actuaries (SOA) Pension Section August, 2011. Prepared by Jodi DiCenzo, Suzanne Shu, Liat Hadar, and Cory Rieth of Behavioral Research Associates
³The research employed a five-condition (four treatment conditions and a control), between-subject design. Respondents were randomly selected from an online panel (Market Tools, Inc.) and screened for age (45 to 75) and the existence of retirement assets before they were randomly allocated to one of the five experimental conditions. The online survey was conducted January 4-5, 2011. The respondent group included 505 pre-retirees and 504 retirees.

Betty Meredith, CFA^®, CFP^®, CRC^®, is Director of Education and Research for the International Foundation for Retirement Education (InFRE), whose purpose is to increase the retirement readiness of the American worker through the accredited Certified Retirement Counselor® (CRC^®) professional certification and continuing education. Betty oversees incorporation of research findings and best practices into InFRE’s certification study and professional continuing education programs in order to help retirement advisors and counselors more effectively meet the retirement preparedness and income management needs of clients and employees. She has almost 30 years experience in the financial services industry.