At the NAACP, Attorney General Eric Holder said Tuesday he opposes a new photo ID requirement in Texas elections because:

"the State of Texas filed against the Justice Department under Section 5 of the Voting Rights Act seeking approval of its proposed voter ID law. Now, after close review the department found that this law would be harmful to minority voters ... Many of those without IDs would have to travel great distances to get them and some would struggle to pay for the documents they might need to obtain them. We call those poll taxes."

Do voter ID laws adversely affect minorities?

  • Please avoid original research and speculation in the answers. This site is not a place where opinions are more interesting than facts.
    – Sklivvz
    Sep 2, 2012 at 23:32
  • 2
    wow, seriously, voting using a utility bill as ID?
    – vartec
    Sep 3, 2012 at 11:57
  • 1
    the only reason for voter ID laws to cause minorities to be unable to vote is if minorities can't get a photo ID. The only reason they might not get one is if they're not citizens and therefore not allowed to vote. Thus saying you don't want such laws because they prevent people voting effectively says you want to enable election fraud!
    – jwenting
    Oct 12, 2012 at 6:25
  • @jwenting You still have to register to vote in your district though. When you show up to vote your name has to be on that list of registered voters in the district, so the only real fraud that can occur is if somebody shows up and claims to be a different registered voter in that district, and he/she knows that this person is not voting or has not voted yet. You can only vote that one day and the districts are small enough that it would be overly burdensome to pull this off for a handful of votes. Voter fraud of this kind is just not worth it and facts prove this. Oct 12, 2012 at 11:50
  • @maple_shaft yet each election there are many votes going to dead people, fake and stolen SSNs, etc. etc. And yes, it is worth it. That's shown time and again. Be a little industrious as an organisation and you can arrange for say half a dozen people to each cast several hundred votes. Organise a few hundred illegals and give each a fraudulently obtained voter registration card, then send them each to every polling booth in a county, and you've thousands upon thousands of fraudulent votes.
    – jwenting
    Oct 12, 2012 at 14:15

2 Answers 2


It seems to depend on who you ask, and they will depend on different interpretations of the same data gathered since HAVA went into effect.

tl;dr Photo ID does not seem to lead to minority voter suppression, but it does lead to drops in lower income voter participation.

If you ask John Lott or the Federalist Society, voter fraud is a known problem, with increasing persecutions and dire consequences, while increasing voter ID requirements actually seems to produce more voters (and campaign finance laws reduce voter turnout). From John Lott:

The results provide some evidence of vote fraud and that regulations that prevent fraud can actually increase the voter participation rate. It is hard to see any evidence that voting regulations differentially harm either minorities, the elderly, or the poor. While this study examines a broad range of voting regulations, it is still too early to evaluate any possible impact of mandatory photo IDs on U.S. elections. What can be said is that the non-photo ID regulations that are already in place have not had the negative impacts that opponents predicted. The evidence provided here also found that campaign finance regulations generally reduced voter turnout.

So, according to this one (conservative) commentator, non-photo ID laws do not affect turnout, while the jury is still out on whether or not photo ID laws do affect turnout.
Lott's arguments seem to mimic those of @user1873; namely, that participation increased, so therefore there could not have been a suppression of the vote. Later in the paper, he states:

How did these laws impacted voter participation rates? As a first crude measure, I only considered states that had changed their laws over time to compare how the participation rates changed when the laws changed. Obviously this simple comparison ignores that many other factors are changing, but it at least compares only the same states over time.

He goes on to provide a number of other metrics, but they all seem to rely on the assumption that the introduction of a voter ID law should result in a lower number of participating voters. To me, this seems like a shaky assumption, when such a small percentage of potentially eligible voters is participating in any given election. If only 50% (or fewer) of people are voting, then increasing to 54% does not mean that some people were disenfranchised, it just means that more people decided to go vote. People could still have been disenfranchised, and not reflected in final vote tallies.

Richard Atkinson argued in 2007 that the logic in Lott's article is incorrect:

This paper finds that photo ID requirements fail to fulfill their primary purpose (the prevention of fraud); in fact, photo ID requirements decrease legitimate voter turnout (and therefore may increase the impact of fraud)

Unfortunately, it's paywalled, like many other legal documents on the subject, which limits my ability to present the results here.

Meanwhile, Alvarez et al noted similar effects, going further into the data than Lott:

Looking first at trends in the aggregate data, we find no evidence that voter identification requirements reduce participation. Using individual-level data from the Current Population Survey across these elections, however, we find that the strictest forms of voter identification requirements - combination requirements of presenting an identification card and positively matching one's signature with a signature either on file or on the identification card, as well as requirements to show picture identification - have a negative impact on the participation of registered voters relative to the weakest requirement, stating one's name. We also find evidence that the stricter voter identification requirements depress turnout to a greater extent for less educated and lower income populations, for both minorities and non-minorities.

Figure 5 of the paper shows:

voter suppression by ID

(X-Axis categories are: State Name, Sign Name, Signature Match, ID Requested, ID Required, ID Required + Signature Match, Photo ID Requested, Photo ID Required)

Given that this second paper acknowledges the incompleteness of the aggregate data in describing voter behavior, and then proceeds to address that lack in a closer examination of individual behavior, I believe they have produced a more robust argument that stringent voter ID laws lead to voter suppression.

Even so, they proceed to Figure 6, which shows that there's no correlation between race and voter suppression, and suggest that it is actually whites who are more disenfranchised than non-whites:

whites v non-whites

Note, however, the very large error bars for the minority voters-- this gap is probably why the problem is so difficult to resolve.

Even so, the authors claim that it is not race that determines real suppression, but income. They continue, with a later figure:

income based suppression

Note that people with $50k or less in income have a ~72% chance to vote with just stating their name, and ~66% chance to vote if required to present photo ID, compared to 85% to 83% reduction for those making $750k. To me, that's a very significant effect, and clearly demonstrates that photo ID requirements will reduce turnout of lower-income voters.

  • 2
    The linear trend line in Figure 5, when the X axis is not linear, but a bunch of categories, is meaningless, and detracts from my trust in the rest of the statistics presented in the graph.
    – Oddthinking
    Sep 12, 2012 at 0:15
  • @Oddthinking-- the X axis is 'linear' in terms of increasing restrictions placed on voters, but I agree, that's a difficult thing to linearize. Hopefully, my edits with Figure 9 provide a better breakdown of voting turnout vs income.
    – mmr
    Sep 12, 2012 at 0:18
  • 4
    Linear has a special meaning in maths. The difference between ID Required and Signature Match is not exactly twice the difference between ID required and ID Requested. Drawing a linear trend line is an inappropriate measure. I once saw an (unpublished) table, which included the (meaningless) average of the survey respondent's postcodes - this is making the same category of mistake.
    – Oddthinking
    Sep 12, 2012 at 0:24
  • 1
    @Oddthinking-- I'm aware of what 'linear' means, and I agree, it's probably incorrect to try to compare these results. Ignoring the trend line does yield some valid conclusions (to my mind)-- hence the inclusion of Figure 9, which provides a more comprehensive breakdown. I can provide more portions of the paper if you think that's appropriate, but that does seem like overkill.
    – mmr
    Sep 12, 2012 at 0:30
  • 2
    Given that race and income are strongly correlated in the US, I think it's disingenuous to say that race isn't a factor here as well. Oct 12, 2012 at 13:57


This hasn't been true in Georgia, which has a similar Voter ID law to Texas.

Georgia Secretary of State Brian Kemp has the data to back it up.

The 2008 elections were the largest in Georgia’s history, featuring record turnout among minority voters with the photo ID requirement in place. The following figures represent voter turnout statistics among Hispanic/Latino, African-American and White voters from the 2004 and 2008 General Elections.

 Voter Demographic |2004 Total Votes Cast |2008 Total Votes Cast |% Increase
 Hispanic/Latino.  | 18,000.              | 43,000               | 140%
 African-American. |834,000.              | 1.2 million.         | 42%
 White.            |2.3 million.          |2.5 million.          |   8%

The slides above were presented by Kemp at the Civitas Conservative Leadership Conference. He notes:

He reported that in the four years since the law was implemented, Georgians cast tens of millions of votes in 40 state and federal elections and hundreds of city and county elections, and the law has withstood challenges in four courts. He also said that the opponents of photo ID have failed to find one individual who has been harmed due to the requirement.

Even when opponents can find plaintiffs for their cases, when they lose the case, the plaintiff turns around and gets a VoterID the next day.

  • 1
    This proves little more than that all voter turnouts have increased. That higher turnout may be only a reflection of the media blitz from all parties. And a larger increase by minorities likely reflects the VERY low turnout in previous elections, not how easy it is to vote now. Those who voted may have done so DESPITE the obstacles placed in their way. The crux seems to be what fraction of people have been effectively dissuaded from voting by these laws?
    – user3344
    Sep 1, 2012 at 15:15
  • 3
    A previous election is a rather ordinary control. What would the attendance have been had the laws not been in place? Couldn't other factors have played a bigger part in mobilising the vote; e.g. the first presidential candidate without white skin?
    – Oddthinking
    Sep 1, 2012 at 15:18
  • 3
    @user1873 - Right now you have a sample size of one in each category (experimental and control), and a zillion uncontrolled variables floating around. Trying to conclude anything beyond "voter ID laws do not stop all voters from voting" is exceedingly optimistic. How, for instance, do you know that the Hispanic vote wouldn't have increased by 200% if not for that law?
    – Rex Kerr
    Sep 2, 2012 at 20:25
  • 1
    @user1873 - By "know" I mean "not basing one's conclusion on a single data point that could be affected by a whole bunch of extraneous variables including that a Black candidate was running for President for the first time". This couldn't possibly have made a massive difference could it? No, surely not! Please ignore nytimes.com/2009/05/01/us/politics/01census.html
    – Rex Kerr
    Sep 2, 2012 at 22:00
  • 3
    @user1873 - looks like good research, but IMO the critical data point is that "opponents of photo ID have failed to find one individual who has been harmed due to the requirement." I'm pretty sure that there were significant changes between the '04 and '08 elections, e.g. demographics of who voted, population, etc. In addition, the requirement for photo ID may have reduced the number of dead, ineligible, and non-existent people who voted. The fact that opponents of photo ID (who would be MOST likely to try to find "victims") were unsuccessful strongly suggests that there weren't any.
    – Crispy
    Sep 11, 2012 at 21:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .