| Voter Suppression as an Historical Phenomena
This fear was most openly expressed and acted on when the South rolled back the advances of Reconstruction, and disenfranchised millions of black voters in the late 19th Century. But this action was actually part of a much broader and long-lasting tendency, as described by historian Alex Keyssar in the introduction the first (2001) edition of his book, The Right to Vote: The Contested History of Democracy in the United State:
There were, in fact, four distinctive periods, or "long swings," in the history of the right to vote in the United States. The first was a pre- and early industrial era during which the right to vote expanded: this period lasted from the signifing of the Constitution until roughly 1850, when the transformation of the class structure wrought by the Industrial Revolution was well under way. The second period, stretching from the 1850s until roughly World War I, was characterized both by a narrowing of voting rights and by a mushrooming upper and middle-class antagonism to universal suffrage. The third era, lasting until the 1960s, was countoured differently in the South than in the North, but throughout the nation was marked by relatively little change in the formal breadth of the franchise; in the North this period also was distinguished by state-sponsored efforts to mitigate the significance and power of an unavoidably growing electorate. The fourth and last period, inaugurated by teh success of the civil rights movement in the South, witnessed the abolition almost all remained restrictions on the right to vote. During each of these periods, the right to vote was contested; at times, the breadth of suffrage was a major political issue; at stake always was the integration (or lack of integration) of the poor and working people into the polity.
It was not conservatives alone who drove the contraction of the right to vote in the second period. They had plenty of help from predominantly middle-class progressives who embraced a "good government" ideology that generally held the right to "good government" trumped the right to vote. But this also can be seen as evidence of the dominance of conservative views, influencing others across the political spectrum, that defined this era. I spoke with Keyssar on Wednesday about the problems involved in defining and measuring voter suppression, and he mentioned a couple of historic examples from this era:
"[T]here are some remarkable numbers out there. I think [author/historian] Morgan Kousser has them. Where, like in Louisiana the numbers went form like 1.2 million to six thousand.
And there's even this one anecdotal thing that's in the original version of the book about registration rules, where I gave this example from Pittsburgh, where I had the minutes of the Pittsburgh voter, new voter registration commission... where they celebrate the fact that with the new registration law that had been passed there were 50,000 fewer votes.
Thus, it's very clear that there was a widespread intentional effort to suppress voting that was particularly intense and racial in the South, but was clearly much broader than that. Once in place, however, these practices were perpetuated institutionally, for the most part. It's not necessary to find individual villains at work, election after election, from the late 19th century down to today, to simply look at the historical patterns put in place back then, see how much those patterns have been changed by subsequent struggles, see how much they remain intact, and draw conclusions about the object effect of voter suppression tactics, strategies and institutional arrangements over time.
It's also worth noting that a great deal of the northern activity during this period involved suppressing the votes of the urban poor and working class, more for purposes of state politics than national. Although Keyssar doesn't deal with this, another practice reinforcing the same political goals of containing urban voting power was that of drawing--or refusing to re-draw--legislative districts, so that rural interests and suburban elites were significantly over-represented compared to urban voters. This situation only worsened over time, until the Warren Court finally intervened with landmark cases such as Wesberry v. Sanders, Reynolds v. Sims and Baker v. Carr. While these didn't deal with voter suppression in terms of suppressing voters from participating at the polls, they most certainly dealt with suppressing voters' political power--they very purpose for which people vote in the first place.
Internalized Voter Suppression
Another important--but often overlooked--aspect of institutionalized voter suppression, in the broadest sense, is internalized political discouragement. A number of studies over the years have shown considerable difference between voters and non-voters in terms of political views as well as demographics. One example, which I've cited & quoted several times here at Open Left (most notable, "It's The Democracy, Stupid!") is the 2006 survey and report by the Public Policy Institute of California, "California's Exclusive Electorate". From PPIC's press release:
California's electorate is significantly whiter, older, wealthier, and more educated than the population at large. "As its population has become more diverse, its voters have become less representative of that population," the report, notes. "And the difference between voters and nonvoters is especially stark in attitudes toward government's role; elected officials; and many social issues, policies, and programs."
For example:
* Governor Schwarzenegger's reelection chances would plummet. In May 2006, non-voters disapproved far more sharply (61-21 percent) than likely voters (48-42 percent).
* The $3 billion affordable housing bond (Prop 1C) could easily pass: 80 percent of nonvoters support it, versus 49 percent of likely voters in a May poll.
* California could have bigger government and higher taxes: Nonvoters prefer higher taxes/more services to lower taxes/fewer services by a 66-26 percent margin, compared to 49-44 percent among voters.
In "It's the Democracy, Stupid!" I also quoted from Project Vote's report on the 2006 elections. "Representational Bias in the 2006 Electorate" by Douglas R. Hess. A couple of its findings:
- If all eligible minorities had voted at the rate of non-Hispanic Whites, more than 7.5 million additional Americans would have participated in the 2006 elections.
- Americans with household income in the top 20 percent of the population (over $100,000/ year) were 1.75 times more likely to vote than those with income in the lowest 20 percent (under $25,000/year) in 2006.
It's not surprising that there's a self-reinforcing tendency: those excluded from power tend to feel powerless, and this feeling of powerlessness tends to suppress their votes. This is reflected in data gathered for more than half a century by the American National Election Study:
Voter Suppression--Historical Data
While intentional voter suppression consciously engaged in is particularly morally repugnant, and deserves specialized attention and analysis, in the end it does not matter so much why large sub-populations of the American electorate are suppressed from voting as much as it matters that they are suppressed, and that our democracy is disfigured and rendered dysfunctional as a result.
To get a big-picture view of how much voting participation has declined over time, here's a chart of presidential voting, using data from Wikipedia. From 1856 to 1896, the percentage of eligible voters voting in presidential elections ranged from about 71% to 81%. It was near the high in 1896, when it began a decline that it has never recovered from:
While this decline is not what people immediately think of (if they think of anything) when they hear the words "voter suppression", it is a decline that's not reflected in the history of most other countries--a decline that's primarily race- and class-based, that sets the US apart from all other advanced industrial democracies, and that undoubtedly has electoral and policy outcomes. We can, in short, consider it an enduring historical feature of American democracy, a built-in level of voter suppression that permanently (at least thus far) skews our electorate significantly to the right, as well as toward over-representation of more affluent white voters. How many suppressed voters does this amount to? Taking 1896 as the base, the decline in voting percentage in 2008 amounts to roughly 38 million voters (an 18.3% decline times an eligible electorate of just over 208.3 million voters).
Comparing this to 8 fraud cases per election, we come up with a ratio of 4.75 miilion to 1. That's roughly 160 times the conservative victimology ratio of 29,000 to 1 that I got for Christian martyrdom and for MoveOn.org's financial dominance of corporate America.
Voter Suppression--Cross-State Evidence
Another large-scale approach to analyzing mass voter suppression from all sources, including institutional, trans-generational effects, is to compare voter participation levels between states. Mid-term elections typically have much lower levels of voter turnout. The last one, 2006, has been analyzed by Project Vote in the report I quoted from above, "Representational Bias in the 2006 Electorate". For our purposes here, I looked at the broadest level of analysis, presented in "Table 1a: Registration and Voting by State, 2006 CPS and used three different voting-level benchmarks to measure gaps in voting participation. First, I tabulated and totaled the gaps between each state's level of voting, and that of Minnesota, the state with the highest turnout level. That was the highest benchmark I used. The lowest benchmark I used was the average level of voting in the country as a whole. For each state that fell below the average, I totaled how many voters it fell short of the national average--roughly the level of voting for the media, 25th state. In between, I used a benchmark midway between these two--that of the 13th state (actually, a 3-way tie for 11th). It's notable that the total figure for the highest benchmark is roughly 35.4 million missing votes--very near to the 38 million figure gained by comparing the 2010 election to 1896 levels. It's as if voting levels in Minnesota today are relatively close to their historical levels, while those in other states have declined by a widely varying range. How true this might be would take a good deal of looking into, but the big picture results are compatible with that impression. More importantly, it means that if all states today did as well as Minnesota, we would now be voting in numbers roughly equal to 1896. Even at the lowest benchmark, the figure for missing voters is over 5 million voters, while the mid-level benchmark shows 15.45 million missing:
With 8 fraud cases per election cycle, that gives us conservative victimology ratios of 4.42 million to 1, 1.93 million to 1, and 633 thousand to 1. All considerably higher than the 29,000 to 1 ratios I got in my first two examples.
Project Vote did a similar analysis of the 2008 election, which came out just last month. Because Obama's candidacy drove a noticeable surge in black participation, I've chosen to use the 2006 figures above for purposes of this diary.
Suppressing Voter Registration: Non-Compliance and Under-Compliacnce With the National Voter Registration Act
The so-called "motor-voter" bill, actually called the National Voter Registration Act, was intended to increase registration not just by providing registration forms and asking people if they wanted to register in motor vehicle agencies. It also included registration in social service agencies serving low-income individuals. There was fierce resistance to this in many quarters, but there was even broader lack of compliance and effort, as reported in the 2005 report, "Ten Years Later: A Promise Unfulfilled: The National Voter Registration Act in Public Assistance Agencies, 1995-2005"-a joint report from ACORN, Demos and Project Vote.
While the initial implementation left much to be desired, and is hardly an ideal benchmark, the fall-off in registration rates at public assistance offices is so dramatic that it can serve as an admittedly understated measure of voter suppression by deliberate state action. The following is based on a chart from the report, eliminating states for which there were significant amounts of missing date. Since there were, however, similar patterns of decreased registrations at public assistance agencies, we can be confident that restricting the list of states to those with the most complete date significantly understates the amount of voter suppression:
The first registration gap is obtained by using the overall growth rate for registrations from all sources. The second is obtained by using the growth rate for Tennessee, which had the highest growth rate for registrations from public assistance offices. Since they cover 2-year cycles, deriving conservative victimology ratios requires us to use a two-year figure for voter fraud as well--16 cases. The results are 89,360 to 1 and 98,909 to 1, respectively--roughly 3 times the 29,000 range for the first two conservative victimology ratios I derived.
Felony Disenfranchisement
Another form of state action that suppresses the vote is the disenfranchisement of felons, which falls disproportionately on blacks and low-income individuals of all races. It should be noted that Japan, which has a dramatically lower crime rate than the US, or even most European countries, has a very pro-active approach to reintegrating criminals back into society. The idea of keeping them apart from civic life in any way is regarded as frankly counter-productive. Encouraging ex-felons to vote is seen as an important aspect of getting them to identify with society as a whole, rather than continuing to identify in opposition to it. Needless to say, this view is virtually never discussed in mainstream American political discourse. Instead, we have massive disenfranchisement of felons--even after they have otherwise completely repaid their debt to society.
A 1998 report from the Sentencing Project and Human Rights Watch, "LOSING THE VOTE: The Impact of Felony Disenfranchisement Laws in the United States" gave a state-by-state breakdown of felony disenfranchisement by state and by race within state. These ranged from 0% in states without such laws--Utah, plus the New England states of Maine, Vermont, and Massachusettes--up to Alabams's total disenfranchisement rate of 7.5%, and 31.5% for black men.
As can be seen above, the national totals were 3,892,400 total disenfranchised, a 2.0% rate, while the number of black men disenfranchised was 1,367,100, a 13.1% rate. Had the rate been race neutral at the state level, an additional 1,158,382 black men would have been allowed to vote. Had the rate been race neutral at the aggregate level for the entire US, the number would have been slightly lower--an additional 1,113,529. For any one election cycle, this translates into a conservative victimology ratio of 144,797 to 1 and 139,191 to 1, respectively. These are between four and five times the 29,000 to 1 ratio in the first two examples I examined in earlier diaries.
What's more, while 36 states--more than 2/3rds--do not disenfranchise ex-felons who have completed their sentence, including probation or parole, the remaining states are so intensely punitive that the total number of ex-felons who were disenfranchised at the time--1,391,000--was more the number of disenfranchised prisoners (1,032,300), probationers (1.016,000) or parolees (452,600):
Furthermore, looking just at the states that disenfranchise ex-felons, they constitute a 55.7% majority of those deprived of the right to vote.
Can these numbers make a difference in elections? You bet they can. In Florida, ex-felons comprised 67.52% of all those disenfranchised--436,900, far more than the 537-vote margin that "elected" George Bush in 2000. But that proportion of ex-felons isn't the highest in the land. Wyoming (73.76%), New Mexico (77.71%), Alabama (80.26%), Virginia (80.28%), and Mississippi (85.85%) were all higher.
If we just consider ex-felons--those who have legally repaid their debt to society according to every other standard--the ratio of votes suppressed to cases of voter fraud is 173,875 to 1--a conservative victimology ratio roughly six times the earlier figure of 29,000 to 1.
Specific, Intentional Voter Suppression
The most morally repugnant form of voter suppression is that which is undertaken specifically to win a particular election. Going forward, I intend to take a much closer look at this. But its necessarily a good deal harder to quantify, since it manifests unevenly, episodically, and in forms that are often hard to quantify. Part of the reason for this is that the forms of voter suppression I've already discussed above do such a good job of voter suppression as a background condition of American democracy that there's usually little or no need to engage in specific "retail level" voter suppression tactics. This doesn't mean they are rare, however. After all, my whole point is that the underlying dynamic is based on exaggerated, irrational fear. Indeed, the GOP has engaged in voter suppression efforts aimed at minorities since at least the 1950s, as described in "Caging Democracy: A 50-Year History of Partisan Challenges to Minority Voters", a 2007 Project Vote report by by Teresa James, J.D.
Just to provide a rough example of the scales involved, I'll briefly discuss one notorious example, the intentionally flawed voter purge felon list used by the State of Florida in 2000, which included an incredibly high number of non-felons. For those not familiar with it, a concise summary can be found at the Sourcewatch.org Election Protection Wiki entry on the Florida voter roll purge. Database Technologies (DBT) was the company responsible, however, they were micro-managed in how to create the list by Florida Secretary of State Katherine Harris. Sourcewatch notes:
DBT subsequently tried to defend their lists by claiming they were 85% accurate. [8] But that would still mean that well over 10,000 mostly minority, poor, and Democratic Floridians were illegally disenfranchised - more than twenty times Bush's margin of victory in the state.
Because we're just talking about the State of Florida(2000 pop: 15,982,378) not the entire US (2000 pop: 281,421,906), the annual voter fraud rate is just over .45, rather than 8. Using the conservative figure of 10,000 votes suppressed, this gives us a conservative victimology ratio of 22,010 to 1. This is roughly 24% lower than the 29,000 to 1 ratio I found in my first two examples.
Summary
The subject of voter suppression is a great deal more complex than the first two subjects I examined to look at conservative victimology ratios, and this diary only begins to scratch the surface of the subject. Nonetheless, it provides a wide range of evidence that is broadly consistent with the initial evidence on conservative victimology ratios. The lowest ratio reported in this diary was the same order of magnitude as the earlier ratios, although it was 24% lower. Summarizing all the findings I've described above, we have:
For historical declines in voting participation in presidential elections: A conservative victimology ratio of 4.75 million to 1.
For contemporary cross-state ratios of voter participation rates, depending on the benchmarks used: Conservative victimology ratios of 4.42 million to 1; 1.93 million to 1; and 633 thousand to 1.
For resistance to implementing the National Voter Registration Act, depending on the benchmark used: Conservative victimology ratios of 89,360 to 1 and 98,909 to 1.
For black felony disenfranchisement, depending on the calculation method (state level vs. aggregate US): Conservative victimology ratios of 144,797 to 1 and 139,191 to 1.
For ex-felon disenfranchisement: A conservative victimology ratio of 173,875 to 1.
For the faulty Florida felon purge list: A conservative victimology ratio of 22,010 to 1, resulting in the fraudulent election of George W. Bush as President.
This is the initial statistical background against which conservative victomology claims about voter fraud should be judged by any objective observer. |
STATE BLOGS
