Americans’ views of the U.S. economy are increasingly colored by politics, making consumer sentiment a less reliable gauge of the outlook for the spending that drives growth.
Self-identified Democrats and Republicans surveyed for the Bloomberg Comfort Index are showing the least agreement on the trajectory of the economy since records began in 1990. Since Barack Obama took office in 2009, the likelihood that the two groups’ views will move in the same direction is less than half what it was under George W. Bush, and less than a third of that when Democrat Bill Clinton was in the White House.
Growing political polarization helps explain the trend, according to analysts such as Steve Jarding at Harvard University. As partisan gaps widen on issues from the overhaul of health-care policy to restructuring tax brackets, it has become harder for consumers to separate their political views from their assessment of the economy. What’s more, auto sales last month were the strongest since November 2007 even as consumer confidence declined in two surveys, showing that less optimism doesn’t always translate into weaker spending.
“America for 30 years has been just drowned in information on polarizing single issues,” said Jarding, a professor at Harvard’s Kennedy School of Government in Cambridge, Massachusetts, and a former Democratic campaign manager. “It’s more difficult just to gauge the sense of the public” on economic conditions, he said.
As a result, Wall Street analysts say they rely less on such consumer surveys, including those from the Conference Board and Thomson Reuters/University of Michigan as well as the Bloomberg index, to project changes in household purchases that make up about 70 percent of growth.
Unrelated to Spending
“We don’t base too much of our forecast on consumer sentiment, no matter which measure, because they’re responding to things in the news in some sense that are unrelated to consumer spending,” said Daniel Silver, an economist at JPMorgan Chase & Co. in New York.
Analysts use the concept of correlation to measure the strength of the relationship between two groups, with zero indicating no discernible trend and 1 showing they are moving in tandem. The correlation between the confidence of Democrats and that of Republicans is 0.25 since Obama started his first term. During Bush’s two terms it was 0.55 and under Clinton it was 0.95, showing consumers of different political persuasions had just about the same views on the direction of the economy in 1993 to 2001.
The Thomson Reuters/University of Michigan gauge of sentiment plunged 9.8 points in December amid negotiations in Washington over tax increases and federal spending cuts. Consumer spending increased 0.1 percent that month, “not a huge drop-off” and illustrating politics may obscure Americans’ actual purchasing habits, Silver said. The Conference Board’s measure dropped to a four-month low at the end of last year.
The Michigan index dropped to a four-month low in August, and the Bloomberg comfort gauge has declined four straight weeks since the beginning of last month. Meanwhile, sales of cars and light trucks accelerated to a 16 million annualized pace in August, according to data from Ward’s Automotive Group.
“Consumers don’t always do what they say, and there are many times when those two things move in opposite directions,” said Jonathan Basile, an economist at Credit Suisse in New York. “Sometimes people will spend to make themselves feel better when things are bad.”
If respondents were rating the economy as they do their pocketbooks, the results would be different. Republicans, who tend to be more affluent, historically have assigned a higher grade to the economy, said Gary Langer, president of the New York-based Langer Research Associates, which compiles the Bloomberg index.
Until the week ended Sept. 1, self-identified Democratic voters had been more confident overall than Republicans for an unprecedented 75 straight weeks. Democrats’ views of their personal finances haven’t followed suit, exceeding Republicans’ in only seven weeks during the same period, according to the Bloomberg Comfort data.
This has occurred even as Republicans lost ground to Democrats in the higher-income categories since 2008, Gallup data shows.
Among those taking home more than $75,000 a year, Republicans outnumber Democrats by 2.1 percentage points, down from 4.4 points in 2008, according to Gallup.
At the same time, the Democratic advantage among lower-income voters has narrowed. The difference between the two major parties among those making less than $20,000 a year is 18 percentage points, down from 25 points in 2008.
Even as the value of the headline confidence numbers diminishes, some components of the surveys, such as a gauge of personal finances, can provide stronger hints of economic growth, Basile said.
Maury Harris, the New York-based chief economist for UBS Securities LLC, said he assesses expectations components in the Michigan and Conference Board confidence gauges to make consumer spending forecasts. Harris is the top forecaster of the Conference Board’s headline index for the past two years, according to data compiled by Bloomberg.
The Conference Board’s gauge of present conditions measures consumers’ views of current business and employment health, as well as their outlook for the next six months. The Michigan survey also includes measures of current conditions, used to assess how Americans perceive their financial situation, and expectations.
“Confidence matters, but income’s very important” in forecasting consumer spending, Harris said. “There’s hardly anything in what I do that’s a definitive, dominant variable. We know that conceptually confidence has to be important.”
The present situation component of the Conference Board’s survey is a better predictor for employment data than its overall figure, according to New York Federal Reserve research that analyzes data dating back to 1977. That survey component shows a 0.89 correlation with the unemployment rate two months into the future.
The proliferation of political surveys, as well as questions within non-political surveys that ask respondents’ affiliations, tend to “muddy the water” on data intended to show economic perceptions, said Ken Goldstein of the Conference Board, which does not ask respondents their political identifications for its monthly confidence survey.
Rebecca McBee, a research area specialist for the Michigan index, said in an e-mail that the survey doesn’t ask respondents for their political affiliation. McBee referred questions on partisan identification in confidence surveys to Richard Curtin, the survey director. Calls and e-mails to Curtin weren’t returned.
The increasingly “hyper-partisan” political landscape has caused consumers to more frequently take cues from voices on television who express strong feelings on issues that are then absorbed by viewers who repeat the ideas, a trend that’s unlikely to recede anytime soon, said Dan Schnur, director of the Jesse M. Unruh Institute of Politics at the University of Southern California in Los Angeles.
“As our political leaders retreat further and further away from midfield into their respective ideological end zones, at a certain point, voters are going to begin to push back, but it’s hard to see that pushback happening in the immediate future,” and probably not before the 2016 presidential election, said Schnur, a former Republican consultant.