Ripples of economic misreporting

American economic statistics are among the world’s best, and their constant improvement is a vital aid to a better performing economy. Media reporting of economic data, however, is sometimes misleading, creating misperceptions that perversely affect people’s confidence in the economy.

Reporting of data can be biased by the use of overblown or alarmist language, whether to give added life to a story or attract an audience from the competition. A message can be heightened or submerged by its placement and emphasis and by its brevity or repetition. Ideological biases and opinions can creep in, or reporters may not realize how they interpret data is based on a particular and possibly controversial economic theory. Headlines may not be consistent with the reporter’s text, or reporters may cross the line and become commentators.

Errors of omission occur when the media fail to mention that key economic indicators, such as economic growth data and payroll jobs, are subject to short-term revision. Economic growth data for a given quarter are reported three times within three months, and payroll employment numbers for a given month are revised twice in the two months after the initial estimate. Though the final improved estimate sometimes paints a different picture from the initial estimate, the public is often not forewarned of that possibility.

Reporters’ knowledge of economics can be faulty. For example, employment sometimes rises sharply as unemployment also rises. This is sometimes interpreted as both good and bad news. But informed reporters would know unemployment as officially defined can only rise along with sharply rising employment when labor force participation increases. They would note the job market improvement encouraged people to enter the work force, many of whom got jobs. Unemployment actually did decline, only the decline was among the uncounted jobless outside the labor force.

Reporters are not always aware of, or they choose to ignore, the sampling error in some economic data. An economic statistic can change from one period to the next, but the change may prove significant. Yet reporters and headlines sometimes announce the number went up or down, implying either good or bad news. This happens with the jobs numbers and when the unemployment rate ticks up or down, despite the official interpretation of no change.

There are two national employment series drawn from difference sources, business payrolls and household surveys, which sometimes tell different stories. Because the sample size is larger for the payroll data, which makes monthly changes less “noisy,” the media often go to the extreme and ignore the household employment numbers altogether. The two series have sometimes shown divergent trends, and ignoring one can cost the public potentially important information.

There is also selectivity bias in the reporting of price data. The Labor Department reports two major monthly series on consumer inflation, the chained and unchained price indexes.

The latter is given precedence in the department’s press releases, so the media usually report only that series. However, economists know the former series is superior because it allows for short-term shifts in consumer purchases in response to changes in relative prices. The unchained index does not, and so overstates consumer inflation.

Fortunately people don’t rely on just the media to form their views about the condition of the economy. They look to their own experiences and what’s happening around them.

On the other hand, individuals have their own biases and preferences and so gravitate toward particular media that filter information in the same way they do.

Philosophies and viewpoints have their place, but it helps when they don’t get tangled with the facts. The media have an obligation to keep them separate in their economic reporting, to avoid spin and maintain accuracy.

Misreporting economic data is not merely offensive to economists, it may affect people’s behavior. What people think about the state of the economy today affects their expectations of tomorrow. If led to believe the economy is better or worse than an informed objective assessment of the data would indicate, their confidence is unduly raised or lowered. Research has established that connection.

How the news media affect consumer perceptions of the economy is import enough the Federal Reserve has invested in its research. Fed economists Mark Doms and Norman Morin analyzed how the tone and volume of economic reporting affect consumer sentiment on the economy. They said, “There are periods when reporting on the economy has not been consistent with actual economic events. … As a consequence, there are times during which consumer sentiment is driven away from what economic fundamentals would suggest.”

In a similar study, a team of Dutch economists analyzed the effect of newspaper spin bias in economic reporting on consumer confidence in the Netherlands. They found the perceptions of economic news created by spin, as distinct from the data themselves, had significant short-term impact on consumer confidence.

Not all dimensions of economic misreporting have been studied, and some would be hard to quantify. However, it seems likely the impact on confidence would become magnified or more persistent when misreporting is blatant or systematic. Unfortunately, some media have that bent.

Story Continues →