Complexity as a Barrier to Annuitization 1 Jeffrey R. Brown, Illinois & NBER Erzo F.P. Luttmer, Dartmouth & NBER Arie Kapteyn, USC & NBER Olivia S. Mitchell, Wharton & NBER GWU/FRB Financial Literacy Seminar March 2013
Do Consumers Know How to Value Annuities? Attempts to reconcile optimizing behavior with consumers reluctance to annuitize are uncomfortable and often leave other puzzles 2 We have binders full of papers documenting decision-making biases, low financial literacy, and poor financial decisions in nearly every other domain of household finance Why should decision to annuitize be any different?
Annuitization is a Complex Decision 3 Requires two types of decision-making with which individuals often have difficulty 1. Inter-temporal decision-making 2. Decision-making under uncertainty Inability to learn from own experience or that of others If anything, it is the opposite of other insurance decisions with which they are more familiar! (e.g., life insurance) Perhaps consumers simply do not know how to value annuities?
What We Will Show 1. Enormous heterogeneity in stated preferences, including many valuations that are at odds with any plausible model of optimizing behavior 2. The price people are willing to pay to buy annuities is much lower than the price at which they are willing to sell annuities. AND, Buy versus Sell prices are negatively correlated. Not driven by endowment effects or liquidity constraints Stronger negative correlation among less financially sophisticated 3. Valuations are sensitive to framing and anchoring 4. Overall, it is difficult for rational model to explain much of the variation in annuity valuations, especially for those who are less financially sophisticated 4
Methods and Data American Life Panel Internet survey of US age 18+ Two waves two weeks apart 5 Ask about willingness to exchange Social Security annuity and a lump-sum Branching answers Four different ways of eliciting response (more in a moment) Rich data on factors that should correlate with how people value the Social Security annuity
CV or EV Two Dimensions to Elicitation 6 Compensating Variation (CV) = evaluate at the current level of utility (one option is status quo) Equivalent Variation (EV) = evaluate at the new level of utility (both options are change from status quo) Buy or Sell Sell: Give up Social Security benefits in exchange for receiving an additional lump sum Buy: Receive more Social Security benefits in exchange for paying a lump sum
CV-Sell 7 In this question, we are going to ask you to make a choice between two money amounts. Please click on the option that you would prefer. Suppose Social Security gave you a choice between: (1) Receiving a Social Security benefit of $1600 per month OR Estimated SSB for individual (2) Receiving a Social Security benefit of $1500 per month and receiving a one-time payment of $20,000 one year from now Trade $100 per month for $20k lump sum
Screen shot: CV-Sell 8
Four Versions: Randomized 9 Note that LS received later of retirement age or one year from now (to avoid responses being driven by hyperbolic discounting) If shift CV version by X, you get the EV question!
EV-Buy 10
But irrational tails Reasonable median value 11
Big drop in valuation when asked to buy CV 12
Similar Patterns for EV Versions 13
Correlations of Annuity Valuation Measures 14
So when faced with a complex decision that they may not fully understand: There is information in responses: EV/CV answers are positively correlated even though questions asked two weeks apart But, Sell price is much greater than Buy price Buy low, sell high heuristic Buy and Sell responses are negatively correlated! People who need to be compensated a lot to give up annuity are willing to pay very little to buy one Financially literate individuals give more coherent answers across Buy and Sell questions 15
Other Sources of Experimental Variation 16 To test if they are taking survey seriously Order of versions of trade-off question shown Show lump-sum amount first or second These do not matter, suggesting that at least people are trying! To see if their preferences poorly defined Initial lump-sum amount presented (anchoring) 10% increase in start value increases final value by 3.5% Effect larger for less financially sophisticated Size of change in Social Security monthly benefit People give larger responses when prior question was larger
17
Can We Explain Variation in Annuity Valuation? According to life-cycle model, valuation should vary based on mortality, risk aversion, bequests, etc. We examine: 18 Annuity Equivalent Wealth (AEW) accounts for variation in mortality rates, risk aversion, etc. Actuarial value as alternative Age, age squared, female, married, race, ethnicity, Education, income, home ownership, financial wealth Self-reported health, kids, risk aversion, return expectations Confidence SS will pay (political risk)
19
Actuarial v. Utility Valuation Key finding 1: There is an approximately one-for-one relation between the actuarial value of the annuity (which varies by birth year, gender, and age at start of annuity) and subjective / stated valuation Key finding 2: With or without conditioning on actuarial value, there is no significant relation between life-cycle measure of utility gains from annuitization (AEW) and stated preferences Table 5 shows robustness Overall: consistent with use of simplifying heuristics 20
But, Overall Explanatory Power is Low R-squared in most specifications are low In general, can not explain more than 10% of the variation 21 Distance from regression line (root MSE) is substantially greater for less financially sophisticated individuals Financial literacy index EV Buy and Sell coherence in CV regression Educational attainment (</> Bachelor s degree)
22 Higher for Less Sophisticated Low
Conclusions Evidence is consistent with the hypothesis that people do not have well-defined preferences over annuitization 23 Thus, the observed lack of annuitization may not reflect a preference for lump-sum Revealed preference logic is highly questionable Important for evaluating policy options Asymmetry in Buy versus Sell careful not to use one to infer the other
Thank You! 24 QUESTIONS?
Fin Lit Questions 25 Q 3.2.1 [INT_RATE_LITERACY] Interest rate literacy [$100 2% ms5_l001] Suppose you had $100 in a savings account and the interest rate was 2% per year. After 5 years, how much do you think you would have in the account if you left the money to grow? (1) More than $102 (2) Exactly $102 (3) Less than $102 (4) Don t Know Q 3.2.2 [INFLATION] Inflation Literacy (ms5_l003) Imagine that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, would you be able to buy more than, exactly the same as, or less than today with the money in this account? (1) More than today (2) Exactly the same as today (3) Less than today (4) Don t Know Q 3.2.3 [SAFER] Stock literacy [ms5_p002] [Linked questions ms5_randomnum] Please tell us whether this statement is true or false? Buying a [If ORDER_STOCK=1, insert single company stock ; else insert stock mutual fund ] usually provides a safer return than a [If ORDER_STOCK=1, insert stock mutual fund ; else insert single company stock ]. (1) True (2) False (3) Don't know