Interventioner och empiriska metoder Flashcards
Government interventions in markets
Laissez-faire - no interventions
Efficiency
To correct for externalities, asymmetric information, and imperfect competition (neoclassic economics)
To correct for internalities (behavioral economics - not an issue for homo economics)
(e.g., time inconsistency and bounded rationality)
Equity
To re-distribute income and wealth
To introduce social insurance
Equity is about fairness while equality is about equal shares
Policy
To achieve policy goals
To finance public expenditures
Policy instruments
Regulation
Ban - e.g. age limits, emission standards
Restriction - e.g. max noise level, max levels of emissions from cars, quotas (e.g., tradable permits)
Fiscal (economic incentives)
Tax
Tax on pollution (market failure) - corrective tax (decreases inefficiency)
Tax on a product - distorting tax (increases inefficiency)
Lump-sum tax (e.g. per capita tax) does not change relative prices
Subsidize
Information (e.g., labeling on products, health information
campaigns)
Behavioral interventions - Nudge
Policy instruments require
Implementation
Monitoring and punishment (if regulation or fiscal)
Penalty (monetary or time spent in prison) if caught
The guilt of committing a crime
The shame of being caught for committing a crime
Behavioral responses
How are individuals expected to respond to the change in policy?
Do we expect any unintended consequences of the policy?
What is a nudge?
A nudge, as we will use the term, is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Putting fruit at eye level counts as a nudge. Banning
junk food does not.” (Thaler and Sunstein, 2008, p.6)
Key features of a nudge
Choice architecture - physical, social, and psychological aspects of the choice contexts
Alters people’s behavior in a predictable way - behavioral economic theory
The objective is better decisions by individuals
There is no neutral way that choices can be presented, that is there is no neutral choice architecture.
A choice situation has an (implicit) default, that is what happens if no choice is made.
Food shopping online: basket is full of the items from the last purchase or empty basket
Why do people make poor decisions?
Inherent as humans (e.g., time-inconsistent behavior)
Poorly informed and inattention
Inherent in decision-making (dual process theory)
Why do we need to nudge people?
People sometimes make bad decisions (bad for themselves):
Habits, complex situations and complex information, and time-inconsistent behavior
People sometimes make decisions that are bad for others (e.g., negative externalities)
What nudges exist?
Default rules (e.g., automatic enrollment (status quo))
Simplification
Use of social norms (feedback)
Increase in ease and convenience (e.g., healthy food visible (change reference point)) – change in physical environment and presentation of information
Disclosure (e.g., energy use)
Warnings (e.g., tobacco)
Precommitment strategies
Reminders
Informing people of the nature and consequences of their own past choices (e.g., energy costs)
Can nudges replace traditional policies?
In most cases no, but might complement when traditional policies are not optimal.
Sometimes very cheap to implement
Liberatarian paternalism
Thaler and Sunstein argue that nudging should be used by governments to improve people’s lives.
With nudging freedom of choice prevails (libertarian).
But nudging still suggests what people should do (paternalism)
Dual-process theory
There are two ways to make a decision (Kahneman, 2003 - Note: He was not first with this idea)
System 1: Automatic, unconscious process: quick and intuitive (nudge e.g., default)
System 2: Explicit, reflective process: slow and rational (nudge e.g., information about calories)
Decision process
System 2 requires higher effort and/or higher cognitive capacity
Decisions, when System 1 is used, are more susceptible to factors such as framing, norms, and emotions.
Habit formation
New behaviors can be automatic (System 1) through habit formation
Habit formation takes time
When are nudges particularly useful?
Benefits now but costs later
Infrequent decisions
No feed back directly
Outcomes are unclear
Research design
Theory on human behavior
Define what aspect of choice architecture to change
Predict the effect and how to measure it
Randomly assign users to treatment and control group
Observe outcomes and compare outcomes in both groups, especially
long-run effects and spillover on other behavior
Default option
A pre-set option that someone will receive if not actively making another choice
A default option does stop anyone from making another choice
”Overall, the finding of large default effects is one of the most robust results in the applied economics literature for the last ten years” (DellaVigna, 2009,p332)
Cost-benefit analysis
A cost-benefit analysis is often applied to evaluate projects in the public sector
The idea is to measure social welfare changes in monetary units
Costs (tangible and intangible) (we make a simplification in this course and assume tangible costs equal price.)
Benefits (tangible and intangible)
Other issues to consider
Discounting (different timing of costs and benefits)
Risk
Intangible benefits are often the hardest to estimate, for example, how much does an individual value cleaner air, better health, or shorter commuting time?
These values are often measured by using the stated preference method
(see lecture on environmental economics).
The change in utilities is translated into monetary terms, for example by asking how much an individual is a maximum willingness to pay for an improvement (that is indifferent between the state before improvement and state after improvement but with less money)
Distributional effects
Little weight to the poor if using the willingness to pay to estimate benefits
Hick-Kaldor improvement - when gainers hypothetically can compensate the losers (compare Pareto improvement)
Why do we use data?
Test predictions from a theoretical model (deduction)
Are people selfish? Test: Dictator game (give an endowment to a subject and ask how much she would like to give to another (Typically anonymous) subject.
Use data to improve a theoretical model (induction)
We observe that people give money to charities. How can utility function be adjusted to capture this?
Analyze the effect of a policy
If we correctly can model people’s behavior, theory can indicate the the direction of a change but not the size of it.
Where to find data?
Previous studies (E.g., elasticities)
Natural experiments
Field, lab-in-the-field, or lab experiments
Surveys
