Economic data get revised all the time, and in many cases, revisions are part of the plan all along. Take, for example, gross domestic product (GDP) – the gold standard for how fast an economy is (or isn’t) growing. When the Bureau of Economic Analysis releases its first estimate of GDP a mere few weeks after the close of a given quarter, it’s already working on the first and second revisions, which tend to follow like clockwork: one revision each for the next two months. Other economic data follow a similar report-first-and-update-later framework, meaning that the first look at a given number with economic relevance is almost never the last word. 

The most common reason for updating originally reported figures is the availability of new or more complete data. After all, collecting, analyzing, and organizing the results of the millions upon millions of individual transactions that occur every day in a large, complex economy like ours and compiling it into something as digestible as a single figure is a messy and time-consuming business. So when a number like fourth-quarter GDP growth gets revised from 3.3% to 3.2% like it did last week,¹ it doesn’t capture too much attention.

But something interesting seems to be happening these days: As current data rolls in, previous data is being revised downward with mildly troubling regularity. Examples from last week included durable goods orders for December: They were originally reported as roughly flat, but after review were revised to -0.3%. That’s a fairly sizable revision.² Excluding transportation – a figure thought to provide a better representation of the pace of new business at the nation’s factories given the lumpiness of orders for things like jetliners, helicopter gunships and train locomotives that can distort the headline figure – the revision was even bigger and more eye-opening: that number fell from +0.5% to -0.1% after further review.

And it wasn’t just GDP and factory orders data, either: Other downward revisions from last week’s list included a big downward reset of January’s first read on consumer confidence,³ December’s first read of pending home sales,⁴ and the Census Bureau’s estimate of inventories at the retail level for the last month of 2023.⁵ (In fact, one of the few data points not revised lower last week was home price appreciation, which actually moved higher after the previous months’ data were more fully analyzed.⁶ Arguably, that’s the one data point from last week where a reversal of the call on the field might have been welcomed by hometown fans.)

I want to call attention to the consistency these revisions seem to display last week: They were almost all negative. Statistically, if the issue behind the tendency to revise is better and more complete data, you’d expect the revisions to be more or less randomly distributed, with roughly as many positive revisions as negative ones. The fact that they were so universally negative this time around – together with the fact that last week’s negative revisions covered a very broad swath of U.S. economic activity spanning factories, businesses, consumers, and homebuyers – might simply be a coincidence and doesn’t suggest a conspiracy afoot in any case, but it also might hint at the idea that all these recent revisions contain more signal than noise. 

It’s also probably no coincidence that the current data for most of last week’s revised-down indicators was also pretty weak in the here-and-now. So grouping each of those observations together, after further review, we might find that the economy is eroding just a little bit faster than the initial results suggested.

Also last week: 

• PC-maker Dell joining the chorus of endorsing the revenue- and profit-juicing characteristics of AI during its earnings call  

• A mildly surprising uptick in personal income (driven, for better or worse, by an increase in government transfers⁷)  

• A  handful of regional Fed manufacturing reports, which were mostly weak and suggested more margin compression might lie ahead for goods-producers^8,9,10

• The PMI surveys disagreeing to some extent about the future direction of the manufacturing sector (S&P Global’s survey saw mild and renewed expansion of factory activity,¹¹ while the Institute of Supply Management’s survey saw the opposite¹²)

For the most part, last week’s economic headlines were either as-expected or more-of-the-same. Even last week’s biggest potential controversies – fears that the passage of another congressional deadline might spur a government shutdown or that a re-acceleration of inflation tucked inside the income-and-outlays report might re-energize the Fed’s rate-raising instincts – both passed without a shout.

What to watch this week 

This week’s main event is non-farm payrolls – aka the “employment situation summary” – a figure that has been notoriously subject to revision in recent months. That matters, because it’s hard to find an area of the economy more relevant to the narrative (or home to more disagreement) right now than the state of the U.S. labor market. Not only would a material change in the moderately positive tone of recent payrolls reports catch the market’s attention, but so would a big negative (or positive, for that matter) revision of past months’ data.

Without a doubt, Friday’s payroll data is the hot button of the labor-related data we’ll get this week. But it’s not the only one: We also get JOLTS (and its headline figure, the number of current job openings across the U.S. economy, due on Wednesday), ADP payrolls/pay insights, and Challenger layoffs (both expected Thursday).

Each of these matter as much as they always do, but I’d watch Challenger, Gray & Christmas’ layoff report for the cleanest signal so far of where the market for American labor is headed. That’s because the pace of layoff announcements seems to have picked up in recent weeks but has yet to show up in broadly scoped statistics like the unemployment rate. The natural lag between layoff announcements and their impact on unemployment rates and other data like weekly claims is nothing new, but it still suggests trouble ahead unless the pace of job cut announcements relents soon. Seasonal impacts – namely, the tendency of downsizing to be front-end loaded into the spring months by firms inclined to make cuts – might make that somewhat less likely at the margin, providing yet another reason to pay close attention. 

Next is Federal Reserve Chairman Jerome Powell’s two-day trip to Capitol Hill to brief the House Financial Services Committee (Wednesday) and Senate Committee on Banking, Housing and Urban Affairs (Thursday). In the past, this two-day sojourn has been an opportunity for the Chair to reset policy in a very visible forum. I wouldn’t expect any such fireworks this week for the simple reason that recently, Fed speakers have been mostly aligned that rate cuts aren’t likely in the very near future, unless a sudden turn in the data requires it.