Lecture 5: Behavioral Public Economics

Transcription

1 Lecture 5: Behavioral Public Economics Johannes Spinnewijn London School of Economics Lecture Notes for Ec426 1 / 20

2 Outline Beyond Revealed Preferences Positive Behavioral Public Economics Behavioral Frictions in a Market Equilibrium 2 / 20

3 Beyond Revealed Preferences Neo-classical approach to public economics: choices reveal preferences! No reason to distinguish between a model of choice and a model of well-being This changes in a world with humans (rather than homo economicus ) who are subject to bounded rationality, biased beliefs, frictions,... Behavioral public economics introduces insights from behavioral economics into the analysis and design of public policies. 3 / 20

4 Behavioral Welfare Economics: Two Approaches Approach #1: Build a positive model of deviations from rationality Ex: hyperbolic discounting, bounded rationality, reference dependence Then calculate optimal policy within such models Approach #2: Choice-based welfare analysis (Bernheim and Rangel 2009) Do not specify a positive model to rationalize behavior Instead map directly from observed choices to statements about welfare 4 / 20

5 Behavioral Welfare Economics: Two Approaches Consider three different health plans with different copays: L, M, H and corresponding variation in premiums We have data from two environments: 1 On red paper, H > M > L 2 On blue paper, M > H > L Approach 1: build a model of why color affects choice and use it to predict which choice reveals true experienced utility Approach 2: Yields bounds on optimal policy L cannot be optimal given available data irrespective of positive Optimal copay bounded between M and H 5 / 20

6 Behavioral Public Economics: Two Approaches Approach 2 requires no theory of choice to make welfare statements ( RP paradigm). Hence, it avoids paternalism, but may be inconclusive Compromises: Identify choice environments which are not welfare-relevant to tighten bounds Libertarian Paternalism: focus on soft policies (e.g., default options) which do not affect rational individuals 6 / 20

7 Positive Behavioral Public Economics Revisit our welfare analysis in the presence of biases/frictions; Agent s behavior ˆx (p) maximizes Û (x, p) rather than U (x, p). The latter determines welfare. Optimal policy maximizes W (p) = U (ˆx (p), p) + λb (ˆx (p), p). The marginal impact of a policy change equals dw dp = U p + U x ˆx (p) + λ [ B p + B x ˆx (p) ] 7 / 20

8 Positive Behavioral Public Economics Three important changes due to agent s biases/frictions: 1 Impact of policy & behavior on U rather than on Û 2 Welfare impact evaluated for ˆx (p) rather than x (p) 3 Behavioral response captured by ˆx (p) rather than x (p) 8 / 20

9 Illustration: Biased Beliefs in the Baily Model Source: Spinnewijn (2014) 9 / 20

10 Illustration: Biased Beliefs in the Baily Model Mueller and Spinnewijn (in progress) 10 / 20

11 Illustration: Biased Beliefs in the Baily Model Spinnewijn (2014): agent s maximize their perceived expected utility, while their welfare depends on their "true" expected utility. Impact of a change in UI benefits: 1 Should be evaluated using π rather than ˆπ: optimistic job seekers ( ˆπ > π) underestimate the value of UI benefits 2 Welfare impact is different given ˆx (p) rather than x (p) envelope result no longer holds; e.g., inducing more search implies a first-order change in "true" expected utility consumption smoothing benefits will be different; e.g., optimists may save too little for unemployment and thus benefit more from UI second effect is captured by Baily formula, first effect is not! 3 Biased beliefs change behavioral response to policy: this may call for alternative policies as well (e.g., providing information) 11 / 20

12 Identifying Behavioral Biases/Frictions Recent papers provide behavioral evidence for biases/frictions in response to public policy: Chetty, Kroft & Looney (2009): tax payers are more responsive to salient taxes included in the posted price Kleven & Waseem (2013): tax notches (jumps in average tax rates) - individuals lose money by earning more Chetty, Friedman & Saez (2013): information frictions across neighbourhoods affects response to Earned Income Tax Credit Pragmatic view (Chetty s Ely lecture, 2015): evidence helps predicting behavioral responses and designing new policies for specific goals. Challenge is to integrate this into welfare analysis: four different models of smoking/addiction with very different policy recommendations / 20

13 Becker and Murphy 1988 Show that addictive goods can be modeled in perfectly rational framework Dynamic model with habit formation current consumption of the addictive good decreases long-run utility but increases marginal utility of consumption tomorrow rationally choose to become addicted (i.e., current consumption increases tomorrow s consumption) and to go cold turkey Implication: no reason for special taxes on these goods; set taxes according to Ramsey rules. 13 / 20

14 Gruber and Koszegi 2004 Hyperbolic discounting preferences for smokers U 0 = u (c 0 ) + β( δ t u(c t )) with β < 1. t 1 U 1 = u(c 1 ) + β( δ t u(c t )) t 2 Planner maximizes U 0 with β = 1 (true utility) Individuals overconsume c: fail to take full account of harm to future selves. Implication: taxes help by reducing demand and thus can partly correct the internality. (Calibration implies corrective tax should be very large) 14 / 20

15 O Donoghue and Rabin 2006 Studies optimal "sin" taxes in a model with two types of consumers: rational and those who overconsume (e.g., because of self-control problems) Can be thought of as a hybrid of Becker and Gruber-Koszegi models Implication: irrationality among a few consumers leads to substantial role for corrective taxation/subsides for rational individuals, excess burden due to taxation is second-order (Harberger triangle) for irrational individuals, welfare gains from correction of internality is first-order (Harberger trapezoid) therefore always optimal to have a positive tax; calibrations suggest fairly large corrective taxes 15 / 20

16 Bernheim and Rangel 2004 Model of cue-triggered addiction. Two selves: cognitive self with rational preferences visceral brain triggered by random cues in which addictive good is consumed at any cost. Probability of trigger increases with past consumption levels Ideal policy: only allow rational consumption, eliminate consumption in hot mode. For example, regulated dispensation must place orders in advance. Implication: corrective taxation is counter-productive by only distorting consumption in rational state, not visceral state 16 / 20

17 Biases in Market Equilibrium Old arguments that behavioral biases are unimportant in market equilibrium: "competitive forces prevent firms from exploiting biased individuals" "if costs of biases are high, individuals will eventually learn" New argument: biases may interact with other ineffi ciencies in market equilibrium Unclear whether relaxing biases improves welfare when accounting for market responses 17 / 20

18 Illustration: Biased Beliefs in Einav&Finkelstein Source: Spinnewijn (2014) 18 / 20

19 Illustration: Biased Beliefs in Einav&Finkelstein Source: Spinnewijn (2014) 19 / 20

20 Illustration: Biased Beliefs in Einav&Finkelstein Adverse selection leads to under-insurance; frictions may aggravate or mitigate this ineffi ciency depending on their type. Presence of frictions reduces selection based on risk types (i.e., cost curves are flattened relative to the demand curve) removing frictions will increase an individual s welfare for given prices however, prices will increase as more risky types buy insurance impact of reducing frictions on welfare may be negative (e.g., Handel 2013, Handel, Kolstad & Spinnewijn 2015) 20 / 20