Politics is distorting economic data

Unlock the US Election Countdown newsletter for free

The stories that matter on money and politics in the race for the White House

The writer is chief economist at UBS Global Wealth Management

Traditionally, consumers’ views on the economy have been taken as a leading indicator for political trends. If consumers are happy, incumbent politicians generally do well. If consumers are unhappy, then “it’s the economy, stupid” and incumbent politicians swiftly swell the ranks of the unemployed. But in the Alice in Wonderland world of today, everything is turned upside down; politics is leading (and distorting) economics.

Economic data is extremely reliant on survey evidence. Most official data is presented with authority as an absolute measurement of economic activity, but the reality is that this authority is constructed on the extremely dubious foundations of asking people how they feel. Fewer and fewer people can be persuaded to fill in any kind of survey nowadays. Falling participation rates mean that those who do answer surveys are (by definition) odd. Something peculiar has to motivate someone to fill in a survey form.

One such motive is politics. Political partisanship takes people away from objectivity and into a world of fantasy. If someone is going to bother to answer a survey from political motives, they are unlikely to take the time and effort to objectively research their answers. Politically inspired survey respondents reply with their gut instincts.

In the US, the Michigan consumer sentiment survey has shown a significant partisan bias. At the moment, there is a Democrat in the White House, and so Democrats will tell pollsters that all is for the best in this best of all possible worlds. Republicans, contemplating the Biden White House, tell pollsters that the economy is mired in the worst of times. Four years ago those positions were reversed. Likewise, the position four years before that was back to Democrat optimism and Republican despondency under the Obama administration. This partisanship is a relatively new thing, however. Before the Obama presidency, the evidence of political bias in survey responses was far more muted.

Michigan sentiment rose higher in August and September, suggesting US households were more optimistic about the economy. However, the details showed increasing pessimism among Republicans. Only Democrats actually told pollsters that they were more optimistic. It cannot be a coincidence that Republican pessimism and resurgent Democrat optimism coincided with President Joe Biden withdrawing from the race and vice-president Kamala Harris taking the nomination. This fact alone did not alter the current economic situation, but it did change the polarised political filter through which everything in the US is viewed at present.

This polarisation extends beyond the headlines of surveys. For example, Republicans are much more likely than Democrats to say they think inflation is high. Over the past three years, there have been some significant differences in regional inflation that might correlate with different political perceptions. However, regional differences have become more muted of late and do not justify the extent of the partisan perception gap. This particular political bias is all the more troubling, as US Federal Reserve chair Jay Powell has previously cited inflation expectations as a motive for policy changes.

It should not be thought that this is solely a problem with consumers. Business sentiment data can just as easily be influenced by the political climate. Whenever an economist has had a bad week and is in need of some light relief, they can always turn to the US Dallas Fed manufacturing sentiment survey. The comments section of this report is often hilarious — naked partisan political views litter the remarks made by survey respondents. It is simply not plausible to expect that such bias does not extend to the supposedly objective answers in the data section of the report.

In increasingly polarised societies, where world views are shaped by the partisan nature of the media we consume, survey evidence is less likely to capture economic realities. It is certainly true that in the US and elsewhere, people seem inclined to say one thing and do the reverse. The repeated pessimism in business sentiment data coincides with stable or improving business output. Consumer despondency in surveys has been accompanied by robust, rising overall spending.

If political partisanship is polluting survey results, then economists and investors need to increasingly challenge the conclusions of survey-based evidence. In the absence of impartial opinions, we need to emphasise the data that is sourced from observable, objective facts.