Download - - Competitive equilibrium under Adverse Selection …darp.lse.ac.uk/pdf/EC426/EC426_Classes/EC426_Class03_Q2.pdf · Competitive equilibrium under Adverse Selection and Moral Hazard

Transcript

Competitive equilibrium under Adverse Selection and Moral HazardIs there a role for government intervention?

Eduard K. Bohm / Morten N. Støstad

EC 426 - Public EconomicsLondon School of Economics

16 October 2017

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 1 / 21

Content

1 Moral hazard

2 Adverse selection

3 Government intervention

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 2 / 21

Moral hazard

Moral hazard

1 Moral hazard

2 Adverse selection

3 Government intervention

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 3 / 21

Moral hazard

Definition

Moral hazard

“The term “moral hazard” originated in the insurance literature. Itsmodern use in economics is understood—by economists—to describeloss-increasing behavior that arises under insurance.”

Rowell and Connelly (2012), A History of the Term ‘Moral Hazard’

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 4 / 21

Moral hazard

Moral Hazard – Theory I

• Consumer desires insurance

• Insurance leads to consumer being willing to take more risk

• Contract cannot specify consumer behavior

• Consumer takes more risk than is socially optimal

• Information asymmetry: Insurance contract cannot specify consumerbehavior

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 5 / 21

Moral hazard

Figure: Schmieder et. al. (2012)E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 6 / 21

Moral hazard

Moral Hazard - Theory II

Krueger (2002) on unemployment benefits:

“More generous benefits will provide greater protection against risk, butwould likely generate larger distortionary effects. For example, generousUnemployment Insurance benefits insure workers against the earningslosses that accompany job loss, but also induce some workers to searchless intensively for a new job.”

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 7 / 21

Moral hazard

Moral Hazard – UI Empirics I

• Gruber (1997): Benefit levels are too high (in the US)

• Parsons (1980) and Hurd and Boskin (1984): Benefit levels explainlower labor force participation

• Krueger (2010): Job search is inversely related to the generosity ofunemployment benefits

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 8 / 21

Moral hazard

Moral Hazard - UI Empirics II

• Card and Riddell (1993, 1997), Riddell and Sharpe (1998), Riddell(1999) examines the timing of UI benefits increasing in Canadarelative to the US, and finds no robust relation to relativeunemployment rates

• Belot and Van Ours (2001, 2004): Uses variation of UI programsacross OECD countries to show that social insurance programs reduceunemployment rates

• Social cohesion, inequality. . .

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 9 / 21

Moral hazard

Moral Hazard - UI Empirics III

Sjoberg et al (2010):

“However, no clear consensus over the empirical evidence seems to havebeen reached. Nickell et al. (2005) and Nunziata (2002) found a highlysignificant positive correlation between unemployment benefits and theunemployment rate, Baker et al. (2004) found no clear relationship. Belotand van Ours (2001, 2004) have even suggested that the generosity ofunemployment benefits may be negatively correlated with unemployment.”

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 10 / 21

Adverse selection

Adverse selection

1 Moral hazard

2 Adverse selection

3 Government intervention

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 11 / 21

Adverse selection

Definition

Adverse selection

“Situations where one side of the market can’t observe the ‘type’ orquality of the goods on other side of the market. For this reason it issometimes called a hidden information problem.”

H. Varian, Intermediate Microeconomics

“The fact that insured individuals know more about their risk level thandoes the insurer might cause those most likely to have the adverse outcometo select insurance, leading insurers to lose money if they offer insurance.”

J. Gruber, Public Finance and Public Policy

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 12 / 21

Adverse selection

Adverse selection - Modelling I

Nobel Prize in 2001 for Akerlof, Spence and Stiglitz awarded for work inthis area.

• “The Market for Lemons” (Akerlof, 1970), asymmetric information onsupply side of the market.

• Insurance markets are covered in Rothschild and Stiglitz (1976) in amodel with price-quantity contracts offered.

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 13 / 21

Adverse selection

The Market for Lemons

Q

pS(p)D(p)

Q∗

p∗

Q

pS(p)D(p)

Demand depends on expected quality in the market. Inefficient outcome(left) and market breakdown (right). Based on Akerlof (1970).

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 14 / 21

Adverse selection

Adverse selection - Modelling II

A more general and graphical model for insurance markets presented byEinav and Finkelstein (2011).

• Competitive insurance firm offering single contract

• Private information on policyholder, knows own risk

• Falling marginal cost representing adverse selection.

• Demand reflects risk aversion of consumers

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 15 / 21

Adverse selection

Textbook model (EF)

Textbook model in Einav, Finkelstein (2011). Falling marginal cost curvereflects self selection, efficiency is not achieved.

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 16 / 21

Adverse selection

Adverse selection - Empirics

Summarized briefly in Chetty, Finkelstein (2013), extensively in Cohen,Siegelmann (2009)

• Search for coverage-risk correlation

• Strong evidence in annuity markets, health insurance (Harvard)

• Policyholders’ use of information, advantageous selection

• Pending work: markets with social insurance

Outcome (Theory and empirics)

Leads to an inefficient market equilibrium or to a market breakdown

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 17 / 21

Government intervention

Government intervention

1 Moral hazard

2 Adverse selection

3 Government intervention

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 18 / 21

Government intervention

General result

Government intervention

In general, there is a role for the government in markets with adverseselection, but not in those with moral hazard.

• In the former, government can prevent people from exiting the marketand enable an efficient outcome

• In the latter, it has no relative advantage when compared to a privateagent

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 19 / 21

Government intervention

Objections

Adverse selection

Intervention counterproductive?

• AS market still efficient

• Gov. may be source

• Opposed effects

• Additional inefficiencies

Moral hazard

Intervention useful?

• Government to “change theproblem”

• Work programs, fire safetyadvertising

• Government-supplied insurancecan reduce (or increase?)incentives to cheat programs

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 20 / 21

Government intervention

Thanks

Thank you for the attention!

E. Bohm / M. Støstad (LSE) Adverse Selection / Moral Hazard 16 October 2017 21 / 21