Warning: don’t believe the American Presidential election turnout myth
Written by Mark Pack on 10th November 2008 – 10:20 amYou’ve seen the stories: massively effective political machines, registering and then mobilising people in unprecedented numbers, leading to big early voting, huge queues at the polls on the day and colossal turnout.
One problem though: it’s a myth. For all the numerous reports, headlines and footage we’ve had telling us the story of turnout soaring up, up and away, the truth is starting to emerge that, actually, turnout was pretty much the same as four years ago. Overall it looks as if turnout will be 61-62%, only slightly up on 2004’s 61% turnout. 2004 was a relatively high turnout year, so increasing turnout slightly on that is nothing to be ashamed of – but neither is it the story the media has been selling.
What conclusions can we draw from this?
First, it is a reminder of how stories can become the accepted wisdom in the media (both online as well as offline) and continue to run even when the evidence doesn’t stack up. There were understandable reasons for predictions of high turnout to have been made prior to the election, but right from the first batch of complete state voting figures, the true turnout picture has started to emerge. The myth though is continuing to linger.
Second, it is far from clear that the Obama GOTV operation was nearly as good as it has been painted in the initial wave of victory euphoria. Where was the surge in turnout? Granted that its impact may have been muted in the overall figures by a lower turnout amongst Republican supporters, less motivated to vote this time than they were in 2004. But overall turnout only inching up suggests, at the very least, that the case hasn’t been made that – despite it looking very impressive – the Obama GOTV operation really delivered lots of extra votes.
Third, it is a rather tired cliché that shots of long lines of people queuing up to vote in the US equal dramatic turnout. The truth is, they don’t tell us anything. A random selection of long queues from across a hugely varied country with a population six times the size of the UK doesn’t say anything about overall turnout levels.
Indeed, if you take a step back it is not even clear why such pictures are used to consistently across different American elections to tell a positive story about people turning out to vote. They could just as well accompany stories of embarrassment and shame at the state of American democracy: why can’t the elections be organised so that people don’t have long queues (as plenty of other countries manage)? And do the long queues in some areas reflect partisan attempts by the politicians in charge of electoral administration details to put off their opponents from voting?
Some of the most moving footage I’ve seen of elections has been of huge queues at the first genuinely democratic elections in a country – particularly in South Africa after the fall of apartheid. But American is no newcomer to organising elections. It’s simply convention that long queues are painted as being good, rather than bad.
Fourth, let’s not beat ourselves up too much about the state of our own democracy here in the UK. Turnout in the 2008 Presidential election will come out at around 5% higher than in our last general election, when calculated on a like-for-like basis.* Five points or so lower, yes, but then in 2005 the British election result was widely seen as a foregone conclusion, there was no Obama like leader to inspire voters and it was generally seen as a poor turnout. And yet only five or so points lower than the already much lauded turnout in this year’s US election.
Fifth, for those looking to technology to raise turnout, the ongoing failure to provide a secure, reliable and not ridiculously expensive technological solution to make it easier for US troops overseas to vote should be a warning about the limitations of online voting technology in practice.
There was much hugely impressive about the way campaigns were managed and run in the US this year. But we shouldn’t blind ourselves to the limitations to what they achieved, or neglect the fact that much of it was due to the huge sums available for spending: around $650 million in total was brought in by just the Obama campaign.
UPDATE: As it has come up in a couple of comments, it’s worth stressing that the number of US adults this year is several million higher than it was in 2004. So just because there were more voters for a particular candidate or in a particular state than four years ago doesn’t necessarily tell us anything other than that the population has grown. That’s certainly not to say that all changes can be put down to population changes – but it means that you can’t just assume “higher numbers = someone has done something right”.
* Turnout in the US is measured as a percentage of the qualifying adult population, rather than as a percentage of those of the electoral register; in the UK the latter measure is used, but estimates of the non-registration rate in the UK from the Electoral Commission and ONS can be used to adjust the UK figure to an estimated equivalent of the US figure. The 2005 general election turnout would have been around 56% on this basis.
Tags: barack obama
Posted in LDVUSA, News 20 Comments »
10th November 2008 at 10:23 am
Interesting post. Im afraid I was suckered into believing the ‘record turnout’ tag line as well. It just shows, when you dig a little deeper !
10th November 2008 at 10:31 am
A bit soon to conclude that the Democrat GOTV operation didn’t work.
Do we have figures on the relative turn-out of the Democrat and Republican vote – particularly in swing states?
10th November 2008 at 10:39 am
In the USA, you actually have to register to vote. Here, the vast majority of people are automatically registered.
Here, the registration form is a trivial name, address and declaration – there, you are asked much more.
Here, you just register as a voter. There, you need to register as a Republican or Democrat – or independent, but this is rarer.
Compare US turnout of registered voters with UK turnout of the same, they look pretty good.
10th November 2008 at 10:44 am
Looking at county-by-county results in some states (e.g. Florida) it does seem thar Obama out-performed Kerry while McCain did little worse than Bush. This suggests Democrat turnout was crucial in those states.
Other states (e.g. Indiana) seem to have more of an even swing with Obama gaining the same number of votes that McCain lost.
Irritatingly, most sources seem not to post turnout figures in the way that they woudl routinely do in a UK election.
10th November 2008 at 10:45 am
Martin: agree we shouldn’t rush to conclude that it didn’t work. But even more reason not to rush to conclude that it did work. There’s some more detailed evidence in the piece that my post links to – suggests a bit of a turnout edge for the Democrats, though that raises the question of what share of the credit goes to message and candidate (=enthusiasm) and what to the GOTV machine.
10th November 2008 at 10:49 am
Looking at this quote:
it appears that we need to take into account the success (or otherwise) in getting people registered.
The article mentions a 4-6 million voter increase. If that was all on the 65 million Democrat vote, it’s a pretty major impact.
10th November 2008 at 10:53 am
PS The US’s adult population is significantly larger than four years ago, which will have driven up the simple ballot total numbers regardless of what the campaigns did.
10th November 2008 at 11:19 am
This has been on my mind for a while. We have a higher turnout in the UK but we don’t have the queues. Presumably we have more polling stations.
However I can see that comparisons are complicated by issues to do with registration.
10th November 2008 at 11:24 am
The queueing in the US is not because of high demand, it is because it takes ages to vote over there! And that is because there are usually several, and often dozens, of extra votes on various local and state level propositions (like California’s Prop 8 banning gay marriage). Most people really dont know which way to vote on a lot of these things which are often financially based.
Finally, Obama scored the greatest number of votes for any president ever. It’s a bit mean spirited to say this doesnt represent a good GOTV operation!
10th November 2008 at 11:40 am
Mark: but isn’t the US adult population also at an all time high? Saying Obama did better than anyone previously because he’s got more votes than them is in danger of saying he’s better than any, say, Canadian leader because none of them have ever got as many votes as him whilst overlooking Canada’s much smaller population.
10th November 2008 at 11:52 am
Good article Mark – There will undoubtedly be lessons to learn from the Obama campaign but they need to be founded in reality. The turnout point was raised by a few analysts on the night in that whilst higher turnout was expected after the surge in early voting it didn’t really appear
I doubt that any of them will alter the fundamentals of election campaigning which some of the wilder commentators are suggesting.
There needs to be some care in analysing the results as well. For example looking at Indiana or Georgia may not be that instructive as the Democrat campaign there will have been at a higher level than in previous years.
10th November 2008 at 12:27 pm
Excellent article
10th November 2008 at 2:08 pm
I don’t know whether overall turnout was high, but early voting turnout was was 25% of total ballots. And, early voters disproportionately favoured Obama.
http://xcurmudgeon.blogspot.com/2008/10/early-voting-exit-polls-and-demographic.html
So, he got his vote out early. That helped make it less likely that late negative campaigning would be effective.
Oh, and if his overall turnout is up, and the Democrats share of the vote is up, then the Democrats turnout is up, no? But, I think the big deal in this campaign was voter involvement through the web – as evidenced by the huge number of small donations Obama got. After all, who’s going to donate $25 then vote for the other guy?
10th November 2008 at 4:44 pm
excellent article Mark.
is it true that the 62% turnout was of registered voters, or was it of all eligble adults? If the former then it would suggest that voter turnout was lower than the UK where a far higher percentage of the adult population is registered to vote.
One other issue is surely the electoral college system.
The three most populous states are California (safe Democrat), Texas, (safe Rep) and New York, (safe Democrat).
America has just the same focus on swing voters in swing districts in swing states as the British political system has on swing voters in swing constituencies.
Were the presidential election decided by the popular vote I am sure turnout would be higher because Republican voters in California and Democratic voters in Texas would have a greater incentive to vote as their vote would have counted just as much as those key ‘undecideds’ in the likes of Ohio, North Carolina and Virginia who decided the election. Plus Gore would have won in 2000!
Of course the electoral college system is part of the constitution and part of the wider system of checks and balances between large and small states. So there is almost no chance of it changing, but I have no doubt that it lowers turnout as millions of Americans have little meaningful say in the election of their president.
10th November 2008 at 5:47 pm
“The queueing in the US is not because of high demand, it is because it takes ages to vote over there! And that is because there are usually several, and often dozens, of extra votes on various local and state level propositions…”
What an appalling system, then! To keep people waiting for hours in queues must massively favour the retired against the working, the childless against those with young children to look after. Not surprising the turnout was on the low side!
Without an exceptional candidate, bags of money, and perfect timing alongside the financial mayhem, the Democrats would have lost in 2008. If they don’t want to lose in 2012, maybe they shoud reform an electoral system which is biased aginst them.
10th November 2008 at 9:35 pm
Paul: eligible adults.
10th November 2008 at 9:40 pm
I’m not sure how much the queueing was the exception rather than the rule. Polling stations without a queue don’t really make good TV shots
10th November 2008 at 11:17 pm
Mark is right that we shouldn’t just acept the hype nor take pictures of long queues as evidence of higher turnout.
On the face of it thouhg there is evidence that the Obama campaign, at least, was successful at increasing the motivation of his supporters to vote, and particu;ar;y in the states that mattered.
For example, it was reported today that the turnout of hispanic voters (strongly pro-Obama) doubled in Colorado.
Overall it is clearly not the case that overall turnout is particularly better than in the US, but there is still much we can learn from the Obama campaign.
11th November 2008 at 12:28 pm
> Geoffrey Payne Says:
> 10th November 2008 at 11:19 am
> This has been on my mind for a while. We
> have a higher turnout in the UK but we
> don’t have the queues.
Actually, we do get queues in this country, just less often. For example, several Cambridge polling stations with high student electorates had queues at the weirdest times in the 2005 General Election.
12th November 2008 at 10:25 am
Mark P, Paul, I read ‘eligible voters’ to mean those who were registered, and so could have voted if they’d wanted to.
This quote, for instance, in the linked article: “In Pennsylvania, 5,851,730 voters cast ballots with 99.8 percent of votes counted — a rise of nearly 690,000 voters over 2004, according its secretary of state. But due to higher registration, the percent of eligible voters who cast ballots dropped from 68.96 in 2004 to 66.8 this year.”
If my interpretation is right, then this discussion is irrelevant and voter registration increases are the more important statistic.