Is it bad when economic data gets revised down?

Unfortunately, much of the economic data we get regularly is based on surveys, which means it comes with a margin of error. And even after it’s published, the data is often revised as more information comes in.

For example, the July new home sales report showed sales fell 10.5% during the month. But the Census Bureau also noted that the margin of error at the 90% confidence interval was ±14.0%, which means sales could have actually climbed by 3.5%.

With the July jobs report, the Bureau of Labor Statistics revised the tally for May and June lower by 103,000 jobs.

These revisions usually aren’t a big deal. But growing concerns about data quality and the politicization of revisions have made this wonky topic something that more people are thinking about.

In particular, downward revisions to economic activity metrics seem to get a lot of attention.

On that, I have two thoughts for stock market investors.

First, if there’s any doubt about the weekly, monthly, or quarterly economic data, then just wait no more than three months for quarterly earnings announcements.

Earnings announcements are about as reliable as they get. Between the income statement, cash flow statement, and balance sheet, you get a complete accounting of every dollar that moved for the thousands of publicly traded companies in the U.S.

Unless a company commits fraud or experiences a major bookkeeping failure, these numbers do not get revised.

Of course, publicly traded companies don’t cover the entire economy. But for stock market investors, these quarterly results and the management commentary that accompanies them collectively help us understand just how good or bad the economy was for business during those periods. It’s the No. 1 reason why I look forward to earnings season.

Second, negative revisions to past economic data can be a bullish sign.

If job creation data is revised down, it means businesses sold the same amount of product to a weaker consumer base than previously thought. In other words, they were more effective at selling than we realized. If GDP figures are lowered, meanwhile, then it means businesses were more resilient than previously assumed.

Remember, the reported quarterly financial results from publicly traded companies almost never get revised. So negative revisions to economic data just mean they delivered the same earnings in a more challenging environment than previously understood. It would be a different story if economic data revisions fed into quarterly earnings calculations, but they don’t.

The big picture

All things considered, the major economic data published by public and private U.S. agencies are helpful and accurate enough. And TKer will continue to report on the ones that matter.

From the perspective of a stock market investor, I wouldn’t lose too much sleep over revisions, especially when they are applied to data more than a few months old. Revisions to old data may matter for the history books, and politicians may use it in their rhetoric. But they won’t change what public companies already reported, and they’re unlikely to have a material effect on business plans moving forward.

If we do get a revision that’s worth talking about, I’ll be sure to address it on TKer.

Related from TKer

• Mathematical context can totally change the story
• Economic data quality deteriorates — 2 thoughts for investors
• A note about Trump firing the BLS commissioner
• The No. 1 reason why I look forward to earnings season
• A brief 9-point guide to how everything affects earnings