What is the Socially Optimal Level of Economic Freedom? The Case of Retirement Savings and Pensions

Transcription

1 What is the Socially Level of Economic? The Case of Retirement and Pensions David Laibson Robert I. Goldman Professor of Economics Harvard University October 30, 2012 Three theories of freedom 1. is an instrumental means to better choices. If I strongly prefer apples to oranges, then I am better off if I am free to choose. This is the approach taken in economics. 2. has a direct impact on well-being. Knowing I am free raises my subjective well-being. 3. is a higher order principle. is intrinsically good, whether or not it raises subjective well-being. 2 1

2 Why not give people maximal freedom? Externalities: My choices may hurt you. If I don t save, you may feel a moral obligation to bail me out when I get old. Internalities: My choices may hurt myself. I make mistakes I have a self-control problem leads to undersaving. 3 Why not give people maximal freedom? Externalities: My choices may hurt you. If I don t save, you may feel a moral obligation to bail me out when I get old. Internalities: My choices may hurt myself. I make mistakes I have a self-control problem leads to undersaving. 4 2

3 Thirsty subjects 1 sip now 2 sips in 5 minutes 60% 40% 1 sip in 20 minutes 2 sips in 25 minutes 30% 70% McClure, Ericson, Laibson, Loewenstein and Cohen (2007) Choosing fruit vs. chocolate Choosing Today Eating Next Week Time If you were deciding today, would you choose fruit or chocolate for next week? Read and van Leeuwen (1998) 3

4 Patient choices for the future: Choosing Today Eating Next Week Time Today, subjects typically choose fruit for next week. 74% choose fruit Impatient choices for today: Choosing and Eating Simultaneously Time If you were deciding today, would you choose fruit or chocolate for today? 4

5 Impatient choices for today: Choosing and Eating Simultaneously Time 70% choose chocolate Present bias Immediate events get full weight. Everything else gets half weight. Phelps and Pollak (1968), Laibson (1997) 5

6 Procrastination Exercise has effort cost 6 Delayed health benefit of 8 Exercise Today: -6 + ½ [8] = -2 Exercise Tomorrow: 0 + ½ [-6 + 8] = 1 Akerlof (1991), O Donoghue and Rabin (1999) Joining a Gym Cost of membership: $75 per month Number of visits: 4 Cost per visit: $19 Cost of pay per visit : $10 Della Vigna and Malmendier (2006) 6

7 What else are we planning to do tomorrow? Invest in financial capital (e.g., cut credit card debt, refinance mortgage, obtain better insurance rates, join 401(k) plan). Invest in human capital (education, skill acquisition) Invest in social capital (children, family, friends). Invest in health capital (exercise, nutrition, smoking, substance abuse) 13 Neural foundations Stay on your diet (Executive function in analytic cortex dlpfc) I want a donut (impulsivity in dopamine reward system) McClure, Laibson, Loewenstein, Cohen (2004) McClure, Ericson, Laibson, Loewenstein, Cohen (2007) Hare, Camerer, Rangel (2009) Figner, Knoch, Johnson, Krosch, Lisanby, Fehr, Weber (2010) Albrecht, Volz, Sutter, Laibson, von Cramon (2011) 7

8 Saving intentions vs. saving behavior Out of every 100 surveyed employees 68 self-report saving too little 24 plan to raise savings rate in next 2 months 3 actually follow through Choi, Laibson, Madrian, Metrick (2002) The modern system of retirement accounts in the US $11 trillion in balances Leakage: before retirement Loans that are not repaid Unpenalized withdrawals 1 st home, education, health, Penalized withdrawals (10% penalty) 16 8

9 Assumptions for simulation (all real 2011 dollars) 6.5% guaranteed return 2% inflation rate 6% 401(k) saving rate 100% employer match No leakage Start working at age 22 First job: $35,000 Start saving at age 22 1% real wage growth 50% SS replacement 4% rule in retirement At retirement: 103% replacement ratio $719,275 financial assets (+ house + SS) 17 Laibson 2012 A little more realism Replacement Financial Ratio Assets Original scenario 1.03 $ 719, % balance leakage 0.72 $ 302,322 40% don t have access 0.64 $ 198,021 Match rate is $ 152,672 Net return is 5.5% 0.59 $ 123,115 20% with access don t participate 0.57 $ 101,174 Start saving at age $ 86,732 SS replacement rate lower 0.51 $ 86, Laibson

10 65-74 year old households In 2007, median holding of financial assets is $68,100 HRS: In 2008, the median holding of financial assets is $12,500 among 1-person households HRS: In 2008, the median holding of financial assets is $111,600 among 2-person households 19 Source: Survey of Consumer Finances What is the out-of-pocket cost of retirement medical care? Premiums Medigap Co-payments $240,000 And add in long-term care on top of that 20 Source: Fidelity 10

11 Summary so far: Left to their own devices, people don t save enough. Next step: The tradeoff between freedom and savings adequacy. 21 US Denmark Singapore 11

12 US Better Denmark Worse Singapore 1 Better Indifference curves: Combinations of and that offer a fixed level of social well-being 3 Worse 2 12

13 Observations: Maximizing freedom is not a good guide to finding the optimal policy. Giving up freedom is sometimes welfare improving. Obtaining optimal savings is not a good guide to finding the optimal savings policy. Moving away from the optimal savings rate is sometimes welfare improving. With bowed indifference curves, compromise policies will always do better than extreme policies. Convex combinations are always better Better 2 Worse 1 13

14 Institutions that increase (household) savings and decrease freedom Social security (compulsory savings) Defined benefit pension schemes (penalize non-savers) Matching contributions (penalize non-savers) Restricted investment menus 27 Better 1 Welfare increasing reduction in freedom 2 Worse 14

15 US: the bastion of libertarianism No prominent US politician favors eliminating the mandatory savings feature in Social Security Even in the US, we are willing to sacrifice freedom to improve household savings 29 Is it always a tradeoff? vs. Is it always the case that we get more of one and less of the other? No. preserving nudges ( libertarian paternalism Sunstein and Thaler). expanding nudges

16 1 2 preserving nudge Better Worse What do these look like? Default enrollment ( opt out ) Target date funds Active choice enrollment 32 16

17 Opt-in enrollment Opt-out enrollment (auto-enrollment) PROCRASTINATION UNDESIRED BEHAVIOR: Non-participation DESIRED BEHAVIOR: participation START HERE Madrian and Shea 2001 Active Choice PROCRASTINATION UNDESIRED BEHAVIOR: Must choose for oneself Non-participation DESIRED BEHAVIOR: participation START HERE Carroll, Choi, Laibson, Madrian, and Metrick (2009) 17

18 Improving 401(k) participation Opt-in enrollment 40% 50% Active choice (requirement to choose) 70% Opt-out 90% 0% 20% 40% 60% 80% 100% Participation Rate (1 year tenure) Home Delivery Utilization for All Drug Classes (including ineligible classes) 5 0 Active Choice Program Beshears, Choi, Madrian, Laibson, Sakong (2011) 18

19 How does the US public feel about freedom preserving interventions? Opt-out rates under automatic enrollment are typically only 5%-10% (opt-out rates rarely exceed 20%) Under automatic enrollment employers report no complaints in 401(k) plans In surveys, 97% of employees in auto-enrollment firms report that they approve of auto-enrollment. Even among employees who opt out of automatic enrollment, 79% report that they approve of autoenrollment The US government has adopted automatic enrollment for its own employees. 37 Are these kinds of interventions really freedom preserving? If these interventions help people more easily obtain what they want, the interventions are freedom increasing. If the interventions make it more costly for people to obtain what they want, the interventions are freedom decreasing. If they create anti-paternalistic resentment, they are freedom decreasing. All of these freedom-changing effects are likely to be small though not negligible

20 Better preserving nudge Some types of libertarian paternalism are in fact slightly freedom-reducing Worse What about freedom increasing choice architecture? These are relatively rare, but they do exist. Simplification/transparency interventions 40 20

21 Quick enrollment PROCRASTINATION UNDESIRED BEHAVIOR: Non-participation DESIRED BEHAVIOR: participation START HERE Beshears, Choi, Laibson, Madrian, and Metrick (2008) Quick enrollment PROCRASTINATION UNDESIRED BEHAVIOR: Non-participation DESIRED BEHAVIOR: participation START HERE Beshears, Choi, Laibson, Madrian, and Metrick (2008) 21

22 Fraction Ever Participating in Plan Simplified enrollment raises participation Beshears, Choi, Laibson, Madrian, Metrick (2008) 50% 40% % 20% % 0% Time since baseline (months) increasing pro-saving policy 22

23 Improving 401(k) participation Opt-in enrollment 40% Quick Enrollment ( check a box ) Active choice (requirement to choose) 50% 70% Opt-out 90% 0% 20% 40% 60% 80% 100% Participation Rate (1 year tenure) New categories of choice architecture: voluntary freedom reduction This is called commitment or self-commitment Are people willing to restrict their own (future) choices? 46 23

24 How to design a commitment contract Participants divide $$$ between: account (22% interest) Goal account (22% interest) withdrawal restriction Beshears, Choi, Madrian, Laibson, Sakong (2011) Initial investment in goal account Goal Account 10% penalty 35% 65% Account Goal account 20% penalty 43% 57% Account Goal account No withdrawal 56% 44% Account 24

25 Now participant can use both a 10% penalty account AND a no withdrawal account 16.2% allocated to 10% Penalty account 33.9% allocated to no withdrawal account Beshears, Choi, Laibson, Madrian, Sakong (2012) Explore self-commitments? Are they a policy-making sweet spot? Are these restraints acceptable, since they are selfgenerated? The policy maker only needs to foster a legal/regulatory environment in which selfcommitment is practical

26 Better increasing(?) Self-commitment Worse Policy menu increasing pro-saving policies increasing self-commitment policies preserving pro-saving policies preserving pro-saving policies reducing pro-saving policies with good tradeoff properties Small loss of freedom + large pro-saving impact Valuable targeting (see asymmetric paternalism ) reducing pro-saving policies with poor tradeoff properties Large loss of freedom + small pro-saving impact Random or even perverse targeting 52 26

27 An example of perverse targeting? Matches: saver puts in a dollar and the government/firm matches a dollar. Has a small effect on contributions/participation Primarily benefits upper-income households (who are disproportionately likely to participate and tend not to have a savings problem to begin with) 53 Better Portfolio of policies 0: status quo 1: with a weak SS system 2: with auto-enrollment in DC 3: with active choice 4: with self-commitment 5: with quick enrollment Worse 27

28 Better 0 Worse S 5 Signifies the point a policy maker would choose were she to have access only to Social Security Better 0 Worse S 5 Signifies the point a policy maker would choose were she to have access only to Social Security 28

29 Policy Possibility Frontier: the portfolio of policies that achieves the maximal level of freedom for any given level of savings. Indifference curve Policy Possibility Frontier 29

30 Socially Level of Economic U D 30

31 U U Policy menu increasing pro-saving policies increasing self-commitment policies preserving pro-saving policies preserving pro-saving policies reducing pro-saving policies with good tradeoff properties Small loss of freedom + large pro-saving impact Valuable targeting (see asymmetric paternalism ) reducing pro-saving policies with poor tradeoff properties Large loss of freedom + small pro-saving impact Random or even perverse targeting 63 31

32 Big picture People tend to save too little for retirement. How can we help people align their good savings intentions with their actions? Attain the policy possibility frontier: the portfolio of policies that achieve the maximal level of freedom for any given level of savings. A policy portfolio that combines freedom-reducing, freedom preserving, and freedom increasing policies achieves the policy possibility frontier