Click image
for larger version
After
annual revisions spanning back through 2014, the
Bureau of
Economic Analysis (BEA) pegged its advance (first) estimate of 2Q2019 U.S. gross
domestic product (GDP) at a seasonally adjusted and annualized rate (SAAR) of +2.05%
(1.9%
expected),
down 1.05 percentage points (PP) from 1Q2019’s +3.10% (previously 3.12%).
On
a year-over-year (YoY) basis, which should eliminate any residual seasonality
distortions present in quarter-over-quarter (QoQ) comparisons, GDP in 2Q2019 was
2.29% higher than in 2Q2018; that growth rate was slightly slower (-0.36PP) than
1Q2019’s +2.65% relative to 1Q2018.
Two
groupings of GDP components -- personal consumption expenditures (PCE) and government
consumption expenditures (GCE) -- contributed to 2Q growth. Private domestic
investment (PDI) and net exports (NetX) detracted from growth. This report reversed
several trends. For example:
PCE
– After decelerating for three consecutive quarters, consumer spending “came roaring
back” in 2Q, contributing 2.85PP -- the most since 4Q2017 -- to the headline
number. Spending on motor vehicles and RVs dominated this category.
PDI
– Drop-offs in the value of private inventories and spending on nonresidential
structures flipped private domestic investment into contraction after three
quarters of expansion. Residential investment spending also exerted a minor
drag on PDI.
NetX
– A drop in exports and rise in imports pulled net exports “into the
red.”
GCE
– Spending at both the federal and state/local levels boosted this category’s
contribution to the headline.
The
BEA’s real final sales of domestic product growth, which excludes the effect of
inventories, rose to +2.91%, up 0.34PP from 1Q2019.
Click image
for larger version
“We
are not quite sure what to make of this new report,” wrote Consumer Metric
Institute’s
Rick
Davis, “primarily because it singularly reverses a number of trends that
were very noticeable over the course of the past year. With that in mind, and
at face value, the key takeaways from this report for 2Q2019 are as follows:
--
The much lamented demise of the consumer sector seems to have been premature.
Combined spending on goods and services provided more growth (+2.84%) than the
net headline number. The improvement in disposable income and the decrease in
savings have likely fueled this spending surge -- reflecting improving
household sentiment.
--
Commercial spending on fixed investment, which had materially supported the
headline number for the past year, slid into contraction for the first time
since 2015.
--
Inventories did their "thing" -- flipping sharply into negative
territory. This is mean reversion at its very best. But arguably it is the flip
side of the improvement in consumer spending. And it brings to mind that
inventory draw-downs are the ultimate mixed message -- demonstrating caution on
the part of inventory holders while simultaneously offering an encouraging
long-range future to manufacturers.
--
Government spending soared, with all of the increase in Federal non-defense
spending. [Federal spending, of which non-defense was indeed the lion’s share, was
actually 60% of the total.] This is likely related to a time-shifted spending
from the extended "shutdown" -- although the "shutdown"
itself (as expected) resulted in no material reduction in spending.
--
Foreign trade also flipped, dropping the headline by -0.64pp after adding
+0.72pp in the prior quarter, a -1.36pp quarter-to-quarter swing.
--
The BEA's deflator is now substantially higher than the CPI-U reported by the
BLS, resulting in a materially more pessimistic growth rate than might
otherwise have been reported -- reversing yet another trend.
--
The 22 quarters of historic revisions were, as a whole, relatively benign --
averaging an upward +0.02PP per quarter. However the immediately preceding four
quarters took a beating, with 4Q2018 dropping by a material -1.07pp.
"Over
the years we have come to expect that the revision process will reduce the
growth reported in the relatively recent past. That raises the obvious
question: Is the BEA's data collection process naturally biased to optimism on
a quarter by quarter basis? Or is it just good bureaucratic policy to bury some
of the negative stuff in the revisions that nobody really looks at?
"It
is plausible that the BEA's survey based approach introduces a short-term
survivor bias in their reports -- a phenomenon also observed in employment
data. Non-responding survey participants are assumed to still be operating, and
their prior responses are simply carried forward. Eventually the dead entities
get weeded out, but not before the earlier assumptions about them creates a
short term survivor bias.
"We
certainly hope that the bias is procedural, and not bureaucratic policy," Davis concluded. "And if
it is procedural, we might point out that this is the 21st century -- with even
tradition bound Major League Baseball embracing instant replays and experimenting
with robots calling balls and strikes. Surely we should expect the BEA to be
doing much better."
The foregoing comments represent the
general economic views and analysis of Delphi Advisors, and are provided solely
for the purpose of information, instruction and discourse. They do not
constitute a solicitation or recommendation regarding any investment.
