Johannesen and Zucman (2014)

Question 1

What are the data limitations of Johannesen and Zucman (2014)

Answer 1

1. do not know what fraction of deposits are used to evade taxes;
2. know residence of immediate owner not ultimate beneficial owner;
3. only observe bank deposits – Swiss data shows 75% of assets held by foreigners in Switzerland are in equities and bonds

Also, critiques of the treaty:

treaties do little to reduce bank secrecy because

1. they do not impose automatic information exchange
2. do not cover many country pairs

Question 2

What data did Johannesen and Zucman (2014) use?

Answer 2

Use quarterly Bank for International Settlements (BIS) data on bank deposit holdings by “nonbanks” at the bilateral level, e.g. deposits held by French resident nonbanks in Swiss banks

Question 3

What was Johannesen and Zucman (2014) about?

Answer 3

Johannesen and Zucman evaluate the effectiveness of the new information exchange treaties between non-havens and haven countries in deterring tax evasion

Question 4

What was the identifying assumption?

How did Johannesen and Zucman (2014) test for the identifying assumption?

Answer 4

Parallel Trends: Causal interpretation of estimates requires assuming that in absence of treaties deposits in “treaty” and “no treaty” pairs would have grown similarly

Test this assumption by examining whether treaty signature associated with pre-treaty deposit growth rates. (i.e. study whether pre-treaty growth in deposits between country pairs could predict the signing of treaties. → found no evidence that deposit growth pre-treaty could predict the signing of treaty, so parallel trend assumption is satisfied.

Question 5

What were the findings of Johannesen and Zucman (2014)?

Answer 5

1. our results suggest that tax evaders shifted deposits to havens not covered by a treaty with their home country. The crackdown thus caused a relocation of deposits at the benefit of the least compliant havens. We discuss the policy implications of these findings. – because of relocation instead of repatriation, the paper does not find evidence on tax law compliance (i.e actually paying taxes)
2. We also observe negative correlation between number of treaties signed and growth-rate of deposits between 2007 and 2011, indicating the efficacy of the treaties
3. Shows positive relationship between a treaty not being signed with a country and deposit flows from that country. – evidence of deposit shifting.
4. Also finds deposit held through sham corporations decrease when treaty is signed between haven and non-haven countries.