Americans say the economy is bad but keep spending. Don’t mistake the midterm polls for a forecast

Despite consumer confidence hitting a low since 2014, people are still spending money as reported in August. Kari Butcher, Burson's Chair for Intelligence & Insights for the Americas, warns that business leaders commit the same error with political polls. They treat the opinions expressed in polls as a reliable prediction of future actions.

Business leaders are making important decisions based on this flawed assumption. When consumer confidence drops and a retailer cuts inventory, a majority supports a social issue, and a CEO believes speaking out will boost customer loyalty, the poll may be correct. However, the business conclusion can still be wrong. Polls capture what people say at a specific moment.

They cannot accurately predict how that sentiment will turn into purchases, loyalty, employee behavior, or investor reactions. Mistaking this distinction can lead a company to back off from demand while it remains strong, enter a debate that customers did not ask for, or confuse the discomfort of switching for genuine loyalty. Midterm elections add another layer of complexity.

The economy is the top issue for voters, with 84% of registered voters considering it very important to their vote. Polls are prevalent, but business leaders must differentiate between what people say in a poll and what they will actually do. Even though a poll indicating a Democratic majority (52% to 42%) is informative about preference, it does not reveal what customers will do next.

It is essential to discern the decision the poll can provide sufficient evidence for. For instance, households managed to maintain spending even as consumer confidence fell, by saving less (saving rate fell to 4.1%). Although sentiment reveals deep pessimism, spending shows that people are continuing their habits. Both perspectives are partially accurate because people can feel pessimistic while spending nonetheless.

When considering political events, such as the Michigan Democratic Senate primary where Abdul El-Sayed won with only a 15-point lead, the difference in assumptions about voter turnout can significantly impact the outcome. A single poll is merely a snapshot, and trends across multiple polls offer more meaningful insights. Companies often fall into the same trap, and the evidence presented here is sourced from Burson's own research.

In the Burson Pulse survey, 61% of Americans cited economic anxiety as making them more cautious and attentive to companies' actions. Four months later, national data revealed that households are saving less. Although these two figures are from different months, they demonstrate the risk of assuming a stated intention as a forecast.

Burson's Pulse study tracks confidence over time, and 48% of Americans reported losing confidence gradually, while 52% attributed it to a single event. Confidence can deteriorate silently, potentially before it is reflected in a headline or a metric, making an overall score insufficient to reveal eroding confidence until significant damage has occurred.

Polls still provide valuable insights, but leaders should not rely on a single source to answer all their questions. Instead, they should combine stated opinions with actual behaviors like search queries, shares, purchases, and actions. Additionally, leaders should question what decision the poll evidence supports. With the midterms approaching, understanding the limits of polls is crucial.

They can inform us about which candidates are ahead, but they cannot predict what customers, employees, or investors will do in response. To make well-informed decisions, leaders must seek further evidence before acting on any poll.