Lately, economic worries have most seeing trouble in almost every data release. Last week was no exception. Yet two data points published last week—which, at first blush, seemingly support slowdown fears—actually have some interesting twists when you take a closer look. Here is a quick rundown.
Eurozone GDP Slowing Has a Silver Lining Most Missed
While most media focused on the ECB’s new “stimulus” (which is nothing of the sort, in our view), they seemingly overlooked the revised Q4 GDP data, published the same day. This report didn’t change headline growth—it was still a slow 0.2% q/q (0.9% annualized)—but it added detail lacking in the preliminary release.[i] The upshot: Consumer spending, business investment and trade all added to growth. Sum them up and you get 0.4% q/q GDP growth. Adding in government spending bumps it to 0.5%. So what dragged growth down?
Inventory change. In Q4, it detracted a big 0.4 percentage point from growth.[ii] Businesses slashed inventories across the eurozone in Q4, apparently led by German auto manufacturers trying to clear the queue. That makes sense when you consider new car sales and production wreaked havoc on data in late 2018. This follows Q3, when rising imports (a sign of strong domestic demand) detracted 0.5 percentage point from headline growth.
Those of you who read MarketMinder regularly likely know our views of GDP—it isn’t synonymous with “the economy.” It includes several facets disconnected from actual economic activity, particularly as public companies experience it. Inventory change and imports are two. Imports, for example, aren’t zero sum to the likes of Amazon, Walmart and myriad other retailers. But GDP features net trade (exports minus imports) to try to isolate national output. It aims to offset consumption figures, which include imports. However, offsetting them implies they have zero economic value. Again, Walmart is likely to disagree. Another quirk: The very same inventory change that blurred eurozone Q4 growth.
GDP treats rising inventories as good and falling as bad. Hence, eurozone inventories’ decline subtracted from growth. But is this actually bad? It depends. Rising inventories could be awful if it means retailers are struggling to sell stuff. Falling may mean fast sales happened! But both depictions are backward looking. The reason we don’t think Germany’s falling inventories are bad: They seem tied to a one-off factor, reinforcing our view the weakness there is fleeting. Bare shelves may mean restocking is necessary, boosting future growth.
If you want to go ahead and forward this discussion to ECB head Mario Draghi, be our guests. After all, the very same day these data hit, he rolled out what he dubbed “stimulus.” We don’t see it as such, owing to the fact it basically extends maturities on existing low-cost loans to banks. But it sure seems unnecessary when you actually look at the data.
About That Disappointing Jobs Report
At 5:30 AM Pacific Standard[iii] Time Friday, the US Bureau of Labor Statistics published February’s Employment Situation report, revealing hiring and unemployment data for the month. Media harps on these reports as if they offer special insight on the economy and stocks’ direction. So when February’s report showed payrolls rose just 20,000—a thud after January’s 311,000—many saw it as another warning of a coming slowdown.[iv] Thing is, jobs data lag. Little about this hiring dip is likely to carry material forward implications. Moreover, this report featured lots of contradictory quirks.
One such quirk: The headline unemployment rate (what stat fans call the U3) fell from 4.0% to 3.8% after rising last month.[v] That may seem confusing, considering last month’s hiring spree and this month’s tepid pace. But it isn’t. The government shutdown seems like the culprit.
The government’s Employment Situation Report yields both the hiring and unemployment rate data. But the data are actually derived from two separate surveys. A household survey yields the unemployment rate data. Meanwhile, a survey of businesses yields the hiring information. The household survey was taken during the government shutdown, and furloughed federal workers are technically considered temporarily unemployed, inflating January’s unemployment rate. That reversed in February, when the shutdown ended. Meanwhile, the business (technically, “establishment”) survey was unaffected. It doesn’t consider temporary furloughed workers as unemployed. Hence, February’s small number of hires wasn’t a function of the shutdown. But the downtick in the unemployment very likely was.
Now, there is one other way the unemployment rate can fall despite slow hiring: Folks can leave the workforce (the denominator of the rate’s calculation). Americans who haven’t sought work in the prior four weeks are considered “discouraged” and removed from the workforce. However, that didn’t happen in February.
The U6 unemployment rate—which includes folks outside the workforce and thus not captured in the headline unemployment rate—fell 0.8 percentage point, by far its largest monthly drop.[vi] This also suggests the slow pace of hiring may be revised up. It reminds us of August 2011, when an initial report showed zero—literally zero—hires.[vii] Coming amid post-downgrade volatility, investors panicked it was a sign of bad things to come. The next month, though, this zero was revised to 57,000 hires. October’s release revised this up again—to 104,000. Maybe that doesn’t happen this time, but we suspect it might.
Regardless, as noted earlier, jobs data lag the economy. Stocks lead it. You cannot forecast stocks’ direction based on hiring or unemployment or whatever. These quirks—while interesting to us—are basically noise for investors.Now, these two quirks likely won’t dispel all the slowdown worries out there. But taken in concert, we think they are a nice reminder to dig into data before you draw conclusions.
[i] Source: Eurostat, as of 3/11/2019. Q4 2018 eurozone real GDP growth rate.
[ii] Ibid. Eurozone inventory change contribution to growth at quarter-over-quarter rate, Q4 2018.
[iii] Ugh. Changing the clocks. Sorry for any flashbacks.
[iv] Source: US Bureau of Labor Statistics, as of 3/11/2019. Total nonfarm payroll change, January and February 2019.
[vii] The original press release is available here: https://fraser.stlouisfed.org/title/144/item/494853.
If you would like to contact the editors responsible for this article, please click here.
*The content contained in this article represents only the opinions and viewpoints of the Fisher Investments editorial staff.