3
55% said that “rich people” would “benefit most” from it.
4
A Harris Poll in June 2003 found that
50% thought the 2003 tax cut was “a good thing,” while 42% said it would help “the rich” a lot
and only 11% said it would help “the middle class” a lot.
5
An even more recent survey in which
respondents were reminded that “President Bush and Congress have made two major cuts in
federal income tax rates” found that 54% of the public approved of those cuts, while only 37%
disapproved.
6
My aim in this paper is to explore the bases of this remarkable public support for tax cuts.
The primary data for my analysis come from the 2002 National Election Study survey.
7
The
survey included 1,511 respondents interviewed by telephone in the six weeks before the 2002
midterm election; 1,346 of these respondents (89%) were reinterviewed in the month after the
election.
8
The respondents answered a series of questions about their perceptions of economic
4
“Do you favor or oppose George W. Bush’s $1.6 trillion tax cut for the country over the next 10
years?” Favor, 51% -- Oppose, 37% -- Don’t know, 12%. “From what you’ve heard so far, who do you
think would benefit most from George W. Bush’s tax cut plan: rich people, poor people, or middle-
income people?” Rich people, 55% -- Poor people, 4% -- Middle-income people, 26% -- Other, Don’t
know, 13%.
5
“The Congress passed and the President has signed a new tax cut. Overall do you think this tax cut
was a good or bad thing?” Good thing, 50% -- Bad thing, 35% -- Not sure, 15%. “Do you think that the
tax cut will help the rich a lot, some, only a little, or not at all?” A lot, 42% -- Some, 30% -- Only a little,
12% -- Not at all, 6% -- Not sure, 10%. “Do you think that the tax cut will help the middle class a lot,
some, only a little, or not at all?” A lot, 11% -- Some, 39% -- Only a little, 31% -- Not at all, 11% -- Not
sure, 8%.
6
“In recent years, President Bush and Congress have made two major cuts in federal income tax rates.
Do you approve or disapprove of these tax cuts?” Approve, 54% -- Disapprove, 37% -- Don’t know, 9%.
7
The data, codebooks, and more detailed information about the study design are publicly available from
the NES website: www.umich.edu/~nes.
8
Some of the questions analyzed here were included in the pre-election survey, some appeared in the
post-election survey, and some were randomized (asked of half the respondents in the pre-election wave
and of the other half in the post-election wave). The NES staff produced two sets of sampling weights,
one for the pre-election data (V020101) and the other for the post-election data (V020102). Most of my
analyses include data from the post-election survey (whether as dependent variables, explanatory
Convention