CLASS POLITICS AND PARTISAN CHANGE – Bartels:
Historically real income growth has been much stronger under democratic presidents
(especially true for the poorer side of spectrum), however the postwar period has seen
more conservative presidents than democratic (5 out of the 7 last presidencies were
o Why is that?
Maybe voters care more about other things than economy, maybe values
are more important to them?
• “Voters in a two party system must sometimes sacrifice secondary
concerns in order to support candidates who share their views on
the issues they care more about” –
2004: supposedly important role of the “values voters” in the victory of
G.W. Bush (especially opposition to gay marriage)
The goal of this text is to clarify the role of class, “moral values” and economic issues in
In search of the Working class:
A) Class politics and partisan change
Some scholars argue that the civil right era provoked (particularly on the part of lower
class whites) racial hostility. They suggest the basis of New Deal voting patterns have
given way to a new cleavage structure in which conservative ideology and cultural issues
brought large numbers of working class white in the Republican camp.
o BUT others argue that the migration for democrat votes to republican ones since
the 1950s come entirely from middle and upperincome voters. (Stonecash)
Along the same lines authors such as Nolan McCarthy or Keith Pole state
that income has been an increasingly strong factor of prediction for
o However: The 2004 elections saw an affluence of lower income citizens towards
the republican side (G. W. Bush won by 23 points the vote of the working class).
Who’s right? Have the 2004 elections marked a change in American
This contrast in opinions can be explained by the lack of one
understanding of the working class. Indeed, there are different notions as to
who counts as ‘working class’.
Defining working class:
o Stonecash: compares low middle and highincome voters
o Brooks: includes all white voters without a college degree
Problem with his definition is it is quite misleading to define people
without a college degree as “poorer folks”.
• 2004: 2/3 of white voters fit in that category. When asked most of
them define themselves as “middle class”. White voters without
college degrees whose incomes fell at bottom third of income distribution represented 2 percentage points of Bush’s margin
victory in 2004.
Disclaimer: This article limits its analysis to white voters explains Bartels this obviously causes
a distorted picture of the American party system however those distortion are particularly
relevant for an analysis of classrelated cleavages given the persistent correlation between race
and economic status in American Society. Nevertheless the distinctiveness of white political
behaviour and focus on whites in existing scholarly and popular literature make this limitation
expedient for his purpose.
When focusing on white people without a college degree, “working class” voters indeed
tend to be more republican
o Among the half century, noncollege degree voters votes have declined in the
Democratic party, but voters with college degree’s votes have increased.
The changing pattern of voting behaviour represented in the above document shouldn’t be
taken as sole evidence of a partisan realignment among white working class cultural
o 1 shift in voting behaviour is both larger and more consistent among college
graduates: seems odd to portray “working class” as the vanguard of whatever
political development these changes are supposed to reflect. o 2 Bear in mind, the make up of the two educational groups marched markedly
over the past half century. (More people have diplomas than in 1950s)
o 3 Lack of correspondence between formal education and concrete economic
circumstances makes educational attainment problematic as a basis for identifying
the “working class”.
Frank: usage of family income (absolute income) as class definer.
Bartels explain then that he categorizes people’s class status on the basis of their family income
relative to the overall distribution of incomes at any given time.
Low income/working class: bottom third of income distribution
Middle income/class: middle third of income distribution
High income: top third of income distribution
B) Has the white working class abandoned the Democratic Party?
(See figure 3.2) White presidential Vote by Income class 9522004
o This figure suggests the exact opposite of 3.1. Lowincome voters tend to vote
more form democrats.
o Shows that over the past halfcentury, economic status has become more important
not less important in structuring the presidential voting behaviour of white
Americans. Moreover the general trend in support for democratic presidential
candidates among whites in the bottom third of the income distribution has been
(See figure 3.3) White party identification by Income Class, 19522004
o Income gap increased.
o Overall decrease in democrat party identification
o This is entirely attributable to the demise of the Solid South as a bastion of
(See figure 3.4) Party identification by region (low income whites) 19522004
o South vs. Non South
o Decline of 46% in net democratic identification in the South.
o No particular trend of decline in nonSouth. In 2004 democrats outnumbered
republican by the same 10 points as in 1952.
o Erosion of democratic identification among lowincome whites is a phenomenon
only seen in the south.
o Democratic attachment between poor whites and more affluent whites appears
clearly both in South and in the rest of the country.
(See table 3.1) Trends in White democratic party identification by region and