Did Wealthy Voters Create A "Reverse Bradley" Effect?
The day after election day, the head of the polling division in my firm showed me what at the time I thought was a startling statistic: exit polls indicated that those making in excess of $200,000 a year voted for Barack Obama 52 to 46 percent. Like this story from Slate, there are now other indications that this was, in fact, the case.
I was surprised because for much of the past year I made close to 500 presentations around the country almost exclusively to people who were earning $200,000 (or much more) and it didn't take long for them to tell me that there was no chance they would vote for Obama.
This almost blood oath not to vote for Obama got much more vocal whenever the election focus turned to the Obama plan to increase taxes on those making over $250,000. No matter where I was in the more than 70 cities and surrounded areas I visited across the U.S.), I was repeatedly told that this showed that Obama didn't understand the markets, the economy, the importance of wealth creation, etc. A number of people offered to bet big sums that he wouldn't win.
But he did...and it appears that many of the people I spoke to, the ones who seemed to be so opposed to Obama, ended up voting for him. The data indicates that a majority of them did.
It's not hard to conjure up a possible reason for why this happened: while this is the group whose taxes would increase under Obama, it's also the group whose wealth suffered the most as the financial markets took a beating the past few years.
I'm not suggesting that these people have suffered more than others who have lost their jobs or homes or who have to make substantial changes in how they live because of economic conditions. But those at the top of the income scale are the ones who have lost a great deal on paper and those loses may have pushed them to decide that Obama, or not another Republican, was the better choice for president.
What makes the support of Obama of those earning $200,000 or more most interesting is that they were so insistent that they wouldn't do it. We were all watching for the Bradley Effect to show up this election. We were wondering if some white voters were telling pollsters they would vote for Obama even though they had no intention of actually doing so.
But while there is little or no indication of any Bradley Effect in the 2008 election, there might have been something none of us saw coming: a "Reverse Bradley Effect." Many wealthy voters who told pollsters they weren't voting for Obama actually pulled the lever for him when they got into the voting booth.
This has to be conjecture because there's no way to test the theory. My strong suspicion is that, if asked, this group of voters probably still wouldn't admit that they voted for Obama.
er, what?
..so why did this same group admit it to the exit pollsters?
My bet is that your audience skewed systematically, either much higher on the income scale (as your post itself suggests), much more (R) on the partisan scale, or (this is my bet), both.
I don't think the Bradley Effect exists for elections, either by race or income. Primaries. maybe.
Non-random sample
The so-called Bradley effect may be due to bad polling during the Bradley race and not to a real effect.
Wealthy people in the coastal states tend to be liberal and vote for Democrats. Wealthy people in the red states tend to be conservative and vote for Republicans.
In more rural states with a lot of small cities, Republican policies can do OK. However, when it comes to big cities and delivering the services that are required to run big cities, low tax, no spending Republican policies are a disaster. This is why the Republicans are losing by big margins in the urban areas. As areas increase their urban character, they trend more Democratic. Mayor Bloomberg and Gov. Schwarzennegger are far more "liberal" than the rest of the GOP. They would not be electable with low tax-no spend policies that did not deliver the needed services to their cities.
Paying taxes discovered to be
Paying taxes discovered to be less painful than losing money.
My suspicion...
is that your sample--the kinds of people who make $200k AND would self-select to hear you present about finance AND would choose to speak to you one-on-one--probably includes those most committed to the Republican orthodoxy.
As for the Bradley Effect, there's a great paper by Dan Hopkins, a Harvard Postdoc, titled "No More Wilder Effect, Never a Whitman Effect." (Interested readers should be able to find it on the web. I downloaded it from Daily Kos.) Hopkins looked at all US Senate and gubernatorial elections from 89-96 (and then additionally at the 2008 Dem. primary) that featured a black vs white pairing.
He finds that the Bradley effect (or Wilder effect--a reference to former Virgina Gov. Douglas Wilder) disappeared in the mid 90s. Further, Hopkins distinguishes between the Bradley effect and a mere "frontrunner" effect. That is, front running candidates--regardless of race--often underperform on election day relative to opinion polls. Hopkins argues that this is mostly attributable to undecideds who state a preference for the most popular candidate to pollsters but who still remain likely to switch.