Q3 GDP Report, First Revision: 3.5% Growth Is Terrible

Summary

• The 3.5% headline number is strong, but the underlying details are most certainly not.
• Most of the revisions were in the wrong direction. Some are very large.
• 88% of the growth this quarter came from household services and non-farm inventories.
• The trade picture is ugly and even exports may be starting to be a problem.
• Government spending is becoming an increasing portion of the GDP growth story, led by state and local budgets and the Federal defense budget.

Monday, December 3, 2018

California, USA

I’m Not Just Trolling You

Yes, the title is trolling you. Historically, 3.5% is pretty good. Headline numbers are great for looking at historical trends, but often tell you very little about what happened in the quarter. This is especially true with GDP which is a single number that is the sum of many variables, which themselves are sums of other variables and so on.

Need convincing? If you want one takeaway from this report, try this on: without changes to private inventories, this is a 1.2% print, not 3.5% — inventories accounting for a stunning 65% of growth in the quarter. It better be a merry Christmas, or there’s going to be some overstocked warehouses.¹

Never trust the headline number.

A Note About Data Revisions

When a data set is revised, it generally means the preliminary numbers come with very high error intervals. The worst offender here is the monthly jobs number, where error intervals and revisions are huge. The summed absolute revisions for 2017 was 368,000 jobs out of total 2.1M in the initial reports, for 17% worth of absolute revisions over the year. For the first 8 months of 2018 where we have full revisions, it is 241,000 out of 1.6M in the first reports, for 15% worth of revisions. Of those 20 months, 10 had >10% revisions, 5 had >30% revisions, and September, 2017 had a whopping 215% of revisions. None of this is unusual. Preliminary numbers are not reliable.

There are many things that cause revisions, and without being an economist at the BLS or BEA, it’s impossible to know what they are — every reporting period is different and is filled with non-recurring items. But it’s always been my (mostly unsupported²) theory that the size and direction of revisions tell you about trends at the end of the reporting period that didn’t show up in the first report. As such, I always pay a lot of attention to the size and direction of revisions as well as the current number, as I believe they give us a better idea about the following reporting period. As we will see, there are some significant revisions in this GDP report and they are mostly not comforting.

3.5% in Context

If you read my report on the preliminary data, much of this will be familiar to you, but the numbers have changed, and in some cases dramatically

The Q2 print was truly impressive with strength in all the places you would want to see it, whereas this one, er, not so much. I like to look first at the broad categories, as a percentage of the total dollar gain in GDP. This will give us a good idea of what drove that 3.5% headline number. Good old Table 1.5.5 is our data source here (calculations mine, data nominal).

Even though revisions have taken it down a bit, that Personal Consumption number looks pretty good and matches with the sky-high Consumer Confidence numbers we’ve seen. If every other category were flat, GDP would have grown at 2.45% just from the US consumer. The same number for Q2 is 2.57% and 1.52% in Q3 2017. But digging down farther, we find some worrying trends, which will get to below.

That investment number is not a misprint and the revisions make it more of an eye-popper, but as we will see when we drill down further, it's actually a little limp, with companies filling up inventories in advance of Christmas and tariffs.

That trade number is, um, well it's something, isn't it? But it's also offset by companies filling up inventories in advance of Christmas and tariffs. But worryingly, exports were a net drag on growth.

The growth in government spending continues unabated and is becoming a larger and larger part of the GDP growth story over time. It's pretty much the opposite of the beginning of this cycle, where reductions in state budgets and the Federal sequester kept aggregate government spending low. In any event, a surge in state and local spending and the growth in the defense budget are the main drivers here.

Drilling Down: Personal Consumption

Even though it’s off considerably from the first report, the headline on Personal Consumption looks great. But what did I warn you about the headline?

The subcategory numbers in the first report were not great, but they have gotten worse here. The Goods subcategory was revised well down, and the splits are fairly depressing.

All Durables took a big hit in the revisions, and only the Recreational Goods subcategory still stands out. In Non-Durables, Food and Beverage remains a standout, despite a downward revision, but that sucking sound you hear is Energy inflation.

Services also had big downward revisions, but Household Services still dominate the report, accounting for almost half of GDP growth, and is the first key driver of growth in the report. Health Care and Housing continue to blot out the sun. Like we saw in Goods, Recreation and Restaurants/Hotels were strong. This was a standout summer for vacationers, and is one of the few places we see upward revisions where we’d like.

Drilling Down: Investment

So, that headline Gross Private Domestic Investment number is pretty darn cool. And it got even better with the revisions! Did I mention: DON’T TRUST THE HEADLINE? Scroll to the bottom of this table, you will see where the problem lies. I'll wait.

Non-Farm Inventories accounted for 40% of all growth in the quarter, and this is the second key driver after Household Services. In total, 63% of the quarter's growth is accounted for by housing, health care and nonfarm inventories, and that number is 88% when you add in the rest of household services. This is not a picture of strong growth.

However, the inventory picture is offset somewhat by imports of goods, which I will cover in the next section. But there is finally some good news in revisions, mixed is with some more bad news.

• Fixed Investment growth looks less terrible.
• Investment in residential and non-residential structures were a net negative on growth in the first report, now showing slightly positive.
• Equipment investment, while still well below 2017, now looks robust with many large positive revisions in the subcategories.
• Investment in IP took a nose-dive in the revisions
• Farmers are getting destroyed, but you knew that already.

Drilling Down: Trade

Like with investment, the headline number is an eye-opener, but the splits give some further insight. It is also worse than it looks.

After a quarter where net exports accounted for 24% of GDP growth, this quarter was a dramatic turnaround as tariffs went into effect, and inventories piled up in advance of those tariffs and in preparation for Christmas.

The dramatic rise in imports shaved a full percent off of growth, while non-farm inventories added 1.4% to growth. To some extent, these are linked and offset each other (we can assume many of these imports went into inventories), but it's hard to tell what the extent is until the next couple of quarters.

Additionally, as other countries feel the pain of Trump's neo-Mercantilism, and engage in retaliation and boycotts of US companies, exports are falling off a cliff, with goods exports providing a drag on GDP growth to the tune of 53 bps. This is what happens when you try and cut the rest of the world off at the knees. It does not make you any taller.

Drilling Down: Government

It's not just this quarter -- the growth of government spending is becoming a large part of GDP growth, accounting for 18% of the growth in Q3 despite substantial downward revisions in State and Local spending. This was primarily driven by large increases in state and local budgets and the Federal defense budget.

State and local budgets by themselves added 39 bps to this quarter's GDP growth, and it's worrying that growth is becoming dependent on state budgets, even with the substantial downward revision. Reductions in state budgets in 2009-2011 was a major drag on the recovery, and that may happen again in the next recession given this trend.

The other thing of note is the explosion of the Federal defense budget, which is driving the deficit on the spending side, only becoming larger with the revisions.

GDP Bullets

Too long? Didn't read? Just plain bored? Here's some handy bullets:

• The 3.5% headline number is strong, but the underlying details are most certainly not.
• Most of the revisions were in the wrong direction. Some are very large.
• 88% of the growth this quarter came from household services and non-farm inventories.
• The inventories number is offset somewhat by the surge in imports to get ahead of tariffs and stock up for Christmas. We won't know how much for a couple of quarters.
• The trade picture is ugly and even exports may be starting to be a problem.
• Health care and housing continue to blot out the sun, accounting for almost half the growth in household services and nearly a quarter of GDP growth in Q3.
• Government spending is becoming an increasing portion of the GDP growth story, led by state and local budgets and the Federal defense budget.
• The investment picture is not as bleak as the first report, but still bad.

Conclusions: See You At The Bottom

Wikimedia Commons

Repeating My Caveat: These are preliminary numbers with one revision left, and additionally, it's only one quarter.

Q3 was a disappointing quarter despite the high headline number. Even with upward revisions, investment is still limp. Consumption is favoring services over goods. The trade picture is obviously changing rapidly, and not for the better. And governments are making up too much of the growth picture.

Watching the progression of inventories moving forward will be key here — will the Christmas season empty out these inventories or will trade wars keep them high? Hopefully some of these negative trends will reverse in Q4 or turn out to be errors when the revised report arrives around Christmas.

How does this effect my Market Outlook? Not much really. These were trends I thought we would see, but not so early, and not so drastic. When I read the blowout Q2 report three months ago, it felt like the top of the economic cycle, and this report certainly does not change that opinion.

There's no reason to panic, but we just ticked down to DefCon 4. See you at the bottom. Wheeeeee!

Thanks for reading. Comments? Questions? Insults? Have at it.

Endnotes

¹ The rest of the tables are nominal, while this one is inflation adjusted. Data is taken from Table 1.1.2

² My best evidence is that the direction and size of revisions is highly correlated to the following reporting period. Also, you know, common sense.

Analyst’s Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.