key exam revision cards (First half) Flashcards
what are the common statistical tools used for natural experiments?
1) Instrument variables (IV) - there exists a variable (Z) that determines X but does not determine Y (while X is endogenous to Y.
use changes in Z to estimate X, then use this idealized x to show how it affects Y
2) Difference in differences (DiD) - compare the time series of treated units before and after treatment with the time series of untreated units.
3) Regression Discontinuity design (RDD) - compare statistical units close to a cutoff as they should be very similar in terms of background characteristics.
ideally landing below or above the cutoff involves uncertainty and cant be easily targeted.
What are some challenges of public provision of public goods?
1) crowding out: As the government provides more public goods, the private sector will provide less, offsetting any effects (under reasonable assumptions). In practice full crowding out is rare but partial crowding out is often observed.
2) Measurement: How do we measure costs and benefits of public goods. often hard for governments.
3) What are the preference of the population for these public goods and how do they transform through aggregation? requires perfect knowledge.
How does Peltzman (1973) show crowding out of the education and what is his solution?
Free public education often crowds out private education provision. This arises because the government provides some fixed level of educational quality, higher- quality education then needs to be purchased from private schools. Attending private schools removed free public education entitlement.
He proposes solving this issue through educational vouchers: give parents credit for cost of tuition at any school.
How does Duflo (2000) try to understand the returns to education?
Duflo (2000) uses a school construction project in Indonesia between 1973 and 1978 where more than 61000 primary schools were built. Enrollment rates increased from 69% to 83% . Results show that construction of schools did indeed increase primary school enrollment when compared to secondary school enrollment (which was the control).
What are vertical and horizontal equity with respect to taxes?
Vertical equity says that those with more resources should pay higher taxes than those with less.
Horizontal equity says that those with same resources should pay the same amount of taxes.
Why are tax credits a better tool to ensure vertical equity than tax deductions?
Tax deductions are a set amount deduction from taxable income so deduction value rises with higher incomes (higher tax rates), Credits on the other hand are constant across incomes and so are progressive.
Why is statutory burden not equal economic burden?
Who pays the tax is not necessarily the one whose resources get diminished
WHat are the three rules of tax incidence?
1) statutory burden is not equal economic burden
2) the market side on which the tax is levied is irrelevant for distribution of tax burden
3) the more inelastic side bears a larger share of the burden
How did Chetty et al. (2009) use a triple difference approach to in their research design to help din the affect of sales tax in product purchase?
Triple difference approach:
1) Products: Cosmetics, deodorants, and hair care accessories vs other products in same aisle.
2) Store: one large store in Northern California vs two other stores with similar demographics
3) Time period: 3 weeks vs calendar year plus 6 weeks
What did chetty et al. (2009) dins about the effectiveness of sales tax compare to excise tax?
excise tax: included in price
sales tax: added at the register
they found a stronger reaction to increases in excise tax than sales tax
What is the formula to calculate deadweight loss of taxation?
DWL = 1/2 * t * dX
where t is the tax
and dX is the change in quantity due to taxation
When is a commodity tax optimal?
A commodity tax is optimal when the tax rate across goods is chosen to minimize the deadweight loss (conditional on raising a given amount of government revenue).
What are two important optimal tax rules considered in the slides and what are their implications?
two tax rules; Ramsey rule and Inverse elasticity rule
Ramsey rule implication: the government should set taxes across commodities so that the ratio of marginal deadweight loss to marginal revenue is equal across commodities.
if lambda is large: higher value for government spending: we should have higher taxes.
Inverse elasticity implication: set taxes proportional to the inverse elasticity of demand for that good
- tax elastic goods less and inelastic goods more
What methods did Eissa (1995) use to answer the question: How do earners respond to income tax changes?
She considered the tax reform act (TRA) of 1986 where 14 income tax brackets were condensed to only two.
She focused on married women (second earners) who are usually more respondent to changes in earning opportunities
Eissa employed a DID setup to compare:
- Very high earning women (99th percentile) with
moderately earning women 70th percentile
- Before and after TRA1986
justification for using 70th percentile: women with moderately high earning saw a much lower reduction in their marginal tax rate
control group necessary because labor supply of women changes over time.
Results: women with very high earning increase their labor supply. A 10% rise in after-tax wages yields an 8 % greater labor supply.
What is Earned income tax credit (EITC) and what methods did Chetty et al. (2013) use to find the impacts of EITC on labor supply?
The EITC is the largest anti- poverty program in the US.
idea: Provide tax refunds to low income households with children - when you file your taxes and earn very little, you will pay no taxes and also get a check from the government.
results: Individuals that bunch in high knowledge areas change wage earnings sharply to obtain larger EITC refunds relative to those in low knowledge areas.
