Is Tax Amnesty a Good Tax Policy? by Luitel Hari S.;

Author:Luitel, Hari S.; [Luitel, Hari S.]
Language: eng
Format: epub
Publisher: Lexington Books
Published: 2014-08-15T00:00:00+00:00

Notes

1. Ideally I wanted to exactly replicate their results and contacted one of the original authors. But I was not able to obtain the data. It will become clear at the discussion section (pages 38–40) that for my review purpose, the exact replication of Alm and Beck’s (1993) results is not necessary.

2. Alm and Beck (1993) do not clearly articulate their null and alternate hypotheses. Below provide them for readers’ easier comprehension:

Null hypothesis: an amnesty does not have any effect on tax revenues (H0: bj = 0)

Alternate hypothesis: an amnesty has an effect on tax revenues (H1: bj ≠ 0)

Alm and Beck (1993) accepted a false null hypothesis of no effect of a tax amnesty on tax revenues where it should have been rejected. This is an example of a Type II decision error.

3. Surprisingly, Alm and Beck’s claim could not be independently verified by other researchers (see Mikesell and Ross 2012, table 2).

4. Due to one missing observation, these numbers do not add up to 111. The information for the amnesty North Dakota run between October 1, 2003, and January 31, 2004, was not available.

5. A statistical analysis is internally valid if the statistical inferences about causal effects are valid for the population being studied. The analysis is externally valid if its inferences and conclusions can be generalized from the population and setting studied to other population and settings (Stock and Watson 2003).

6. Sample selection bias is not an inconsequential issue. It often gets ignored for a variety of reasons and, on a side note, we take this opportunity to remind readers of the historical statistical blunder involving sample selection bias in the US. In the general presidential election of 1936, the then Literary Digest magazine embarked on predicting the winner of the election in which the Republican governor of Kansas, Alfred M. Landon, was contending with the then incumbent, President Franklin Delano Roosevelt. According to the New York Times article on October 30, 1936, the Literary Digest predicted that Governor Landon would win by an electoral college vote of 370 to 161, would carry 32 of the 48 states, and would lead President Roosevelt about four to three in their share of the popular vote. When the actual votes were tallied, as Arthur Krock reported in another New Your Times article on November 5, 1936, contrary to the Literary Digest’s prediction, President Roosevelt won the election by the largest popular and electoral majority—a margin of approximately 11,000,000 plurality of all votes cast, and 523 votes in the electoral college. As a result of the Literary Digest forecasting error, the magazine went bankrupt soon after the 1936 election.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.