Reverse EC325 LT Readings

Question 1

Studies VAT enforcement in Chile through sending random threat letters.
Result 1: Additional monitoring improves compliance and increases tax paid.
Result 2: Enforcement effect is transmitted upstream to the suppliers of firms in the treatment group.

Answer 1

Pomeranz (2015)

Question 2

2 Strategies:

1. An experimental Diff-in-diff set up where authors posted tax inclusive price tags on shelf for a subset of products in a selected store, then tracked quantity sold by product
2. Diff-in-diff design, comparing the elasticities of alcohol consumption across states where sales tax and excise tax is levied.

2 findings:

1. Found 8% reduction in demand from posting post-tax price labels compared to pre-tax price labels. -> tax salience affects people’s demand
2. Found price elasticities of alcohol is greater for imposition of excise tax than sales tax -> tax salience affects people’s demand

Conclusion: Salience matters in terms of demand and elasticities!

Answer 2

Chetty et al. (2009)

Question 3

Paper used tax relief programmes on gasoline at state levels in 2000 to study pass-through of gas tax.

Main findings:

1. Substantial pass-through (70-100%) to consumers
2. Much smaller pass-through near borders, where demand is more price elastic (cross-border substitution)

Answer 3

Doyle and Samphantharak (2007)

Question 4

Paper used the 1986 Tax Reform Act as a natural experiment to identify the labour supply responsiveness of married women to changes in tax rate.

Paper used diff-in-diff and analyzed the response of married women at or above the 99^th percentile of the income distribution, using married women from the 75^th percentile as the control group.

Finding: The increase in total labour supply of married women at the top of the income distribution implied an elasticity with respect to the after-tax wage of approximately 0.8, which is highly elastic. However, the high elasticity was accompanied by large standard errors, so effects were not statistically significant.

Also, parallel trends assumption may not be satisfied because the treatment group started from a lower level of labour supply, so the control group may be less able to absorb an upward trend in female labour supply.

Answer 4

Eissa (1995)

Question 5

Use a dif-in-dif to estimate the effect of the EITC expansion in TRA86 on labor force participation and hours worked for single mothers.

Treatment group: single mothers

Control groups: Single women without children

Found EITC (a transfer system) expansion caused great increase in labour force participation by single mothers

Answer 5

Eissa & Liebman (1996)

Question 6

Paper used the closing of disability screening offices that increased travel costs for disability benefit recipients to study the differential effects of increased ordeals on disability scheme take-up for different types of disabled individuals.

Found that increased ordeals reduced more people who were deserving of disability benefits than those who did not taking up disability benefits. -> indicating utility cost of ordeals in real life is likely not that much higher than the non-deserving

Answer 6

Deshpande and Li (2019)

Question 7

In lecture, 3 reasons were proposed for why eligible individuals don’t take up social security programs.
1. Welfare stigma
2. Imperfect information
3. Transaction costs related to taking up programmes (i.e. ordeals) – Currie (2004) finds evidence for.
This paper offers a review of recent literature regarding the take up of social programs in the U.S. and U.K.
Paper finds take up is enhanced by default enrolment and lowered by administrative barriers, suggesting lowering transaction costs (ordeals) improves programme take-up.

Answer 7

Currie (2004)

Question 8

Found strong migration effect of foreign football players into spain following ‘Beckham Law’ that gave foreign top income earners an effective tax cut.

Answer 8

Kleven, Landais, Saez (2011)

Question 9

Studied the effect of Danish tax cut on top income earners’ immigration, and found

• Migration effect of Danish tax scheme was highly significant, and top earners demonstrated particularly high elasticity.
• Particularly large elasticity on entertainment & sports professionals.

Answer 9

Kleven et al. (2014)

Question 10

Using Danish non-EU migrant welfare policy change to study the ‘welfare magnet’ theory.
Found that Denmark reducing welfare scheme for immigrants reduced number of migrants by 5000 each year. (Most reduction came from family re-union migration & asylum seeking)

Answer 10

Agersnap et al (2019)

Question 11

Estimate ETI Using Reagan 1986 tax reform as natural experiment, and found that the 1986 tax cut had very high effect on the top 1% income share -> evidence for very high ETI, and trickle down effect (later overturned by other literature)

Answer 11

Feldstein (1995)

Question 12

Studied the effect Thatcher tax cuts on the top income share of population in UK.
Finding: The ETI using full-time series regression between 1978 and 2003 is around 0.3 – not that large.

Answer 12

Brewer, Saez and Shepard (2010)

Question 13

Paper used Diff-in-diff design to study the effects of Danish tax reforms on labour supply responses.

Finding: behavioural responses to tax cuts are clear, but low ETI. (recently only 0.05-0.15 for reforms)

Why is Danish elasticities smaller than that of U.S.

1. True Danish ETI is smaller
   
   1. Possibly due to smaller avoidance & evasion responses due to broad tax bases and strong enforcement.
2. US ETI is over-estimated

Answer 13

Kleven and Schultz (2014)

Question 14

Result: If income effects explain all demand differences across rich and poor, then the optimal tax system uses only income taxes, and no differentiated commodity taxes.

Answer 14

Atkinson-Stiglitz (1976)

Question 15

Illustrates a stylized model on tax evasion – predicts that most people will want to have some degree of tax evasion.
Reality suggests otherwise, and (not in this reading) possibilities are:
- Third-party information
- Misperception
- Psychology and culture

Answer 15

Allingham-Sandmo (1972)

Question 16

Through sending random threat letters on auditing, authors used field experiment with Danish tax collection agency to verify the below results about tax evasion:

1. Tax evasion rate is 0 for third-party reported income, and evasion rate is large for self-reported income.
2. Increased enforcement has no effect on tax evasion for 3^rd party income, but is significant for self-reported income
3. Evasion effect of increased tax rates is 0 for 3^rd-party reported income, and is possibly significant for self-reported income.

Answer 16

Kleven et al. (2011)

Question 17

How to evade taxes using offshore account:
Own a company, open a sham corporation in Cayman Islands, and sham corporation sets up a Swiss bank account.
The legit US company that you own purchase sham services from the sham corporation, and pay the money to Swiss bank account. Then invest and generate returns with the Swiss account without having to pay U.S. taxes.

Answer 17

Zucman (2014)

Question 18

Shows revenues and structures of tax varies across countries.
Findings of differences in tax structure in DEVELOPING countries:
- Personal income taxes as a share of revenue are very low.
- Capital income taxes are more important than labour income taxes
- Differentiated taxes on luxuries and necessities is common.
- Tariffs constitute an important revenue source

Answer 18

Gordon & Li (2009)

Question 19

Through studying tax take and tax structure in different countries, authors found 3 stylized facts about tax:

• The positive relationship between development and tax take across countries are driven entirely by modern taxes.
• Evolution of tax take within countries over time is driven entirely by the expansion of modern taxes (S-Shape?)
• Tax take is positively related to firm size and negatively related to self-employment within and across countries.

Kleven, Keiner and Saez (2015) theory:

development-> Firms’ size and complexity increase -> third party tax enforcement becomes more effective -> effects: 1. Tax level increases 2. Tax structure changes from traditional to modern taxes

Answer 19

Kleven, Kreiner and Saez (2015)

Question 20

• When will in-kind transfers (instead of cash transfers) can improve targeting efficiency:
  1. If differences in demand across rich and poor are due only to income effects, then the in-kind transfer cannot improve targeting efficiency
  2. If, conditional on income, demand is higher for low-ability individuals, then in-kind transfers can improve targeting efficiency.

-Ordeals can also be a mechanism that improves target efficiency and separates imposters and the true low-ability type.

Answer 20

Nichols & Zeckhauser (1982)