by Rick Davis
In their first revision of their estimate of the fourth quarter 2011 GDP, the Bureau of Economic Analysis (BEA) found that the annualized rate of economic growth was 2.98%, up about a quarter of a percent from their “advance” estimate of 2.75% for the fourth quarter and more than a percent higher than their “final” estimate of 1.81% for the third quarter.
This revision to the prior month’s report does not reflect actual changes in the economy, but rather another month’s improvement in the BEA’s understanding of what was happening in the prior quarter. But such revisions continue to tell us much about the timeliness of the BEA’s data collection processes, and the quality of the data that we can expect from their earliest reports.
Perhaps the most remarkable aspect of this report is that, contrary to recent “first revisions,” the changes reflected in the new set of numbers are relatively minor. Expenditures for consumer goods are now reportedly slightly weaker, but there is a somewhat more-than-offsetting increase in expenditures for consumer services. Similarly, investments for fixed assets are reported to be slightly higher, but there is a largely offsetting decrease in the rate of inventory growth. The contraction rate of government expenditures moderated by a statistically insignificant 0.04%, while exports weakened and imports strengthened by similarly insignificant amounts.
However, the most significant revision in this report involves the overall “deflater” that the BEA used to correct the “nominal” data into “real” numbers: annualized inflation during 4Q-2011 is now reported by the BEA to have been 0.86%. This is over twice the lower rate used in the “advance” estimate (0.39%), and more notably it now even more at variance with the dis-inflationary numbers reported by the Bureau of Labor Statistics (BLS) for the CPI-U during the quarter (a -2.13% annualized “inflation” rate). For those inclined to believe that the BEA merely publishes political spin or engineered “happy” numbers this particular deflater is either a curious dilemma or intriguing twist, since it substantially understates what could have been reported using something much closer to the BLS deflation rate. In fact, using a “deflater” close to the CPI-U would effectively double the headline number — which might raise plausibility issues of an entirely different sort.
Among the notable items in the report:
— The contribution to the annualized growth rate for consumer expenditures for goods was revised down slightly to 1.17%, which is still up 0.84% from the 0.33% rate reported for the third quarter of 2011.
— On the other hand, consumer services were revised modestly upward to 0.35%, gaining a quarter of a percent from the 0.10% in the “advance” report.
— The growth rate contribution from private fixed investments was also revised up slightly to 0.53% (from 0.41% previously), but this remains about 1% lower than the 1.52% annualized rate reported for the third quarter.
— The contribution from inventories (1.88% annualized) is only slightly less than the 1.94% reported in the “advance” estimate and it still represents in its own right a +3.23% improvement in the headline number compared to the third quarter. As mentioned last month the substantial swing could be the result of either real changes in inventory levels or phantom data artifacts as changing “deflaters” puffed up the calculated “real” inventory valuations.
— The reported drag on GDP growth from contracting expenditures by governments at all levels was not meaningfully changed at -0.89% (slightly less than the -0.93% drag previously reported), remaining -0.87% worse than during the third quarter of 2011.
— The annualized contribution to the growth rate from exports was revised down slightly to 0.59% (from 0.64% in the “advance” report).
— Imports are now reported to be removing -0.65% from the growth rate of the overall economy, only modestly less than the -0.75% previously reported but still a significant deterioration from the -0.21% rate reported for the third quarter.
— The annualized growth rate of “real final sales of domestic product” was revised upward about a third of a percent to 1.10%, but it is still substantially below the +3.16% reported for the third quarter. If this number is accepted at face value (and not as a consequence of
“deflaters” playing havoc with inventory valuations) it still indicates a much weaker economy than is conveyed in the headline number.
— Real per-capita disposable income is now reported to have grown at a miserable 0.59% annualized rate during 4Q-2011.
The Numbers (as Revised)
As a quick reminder, the classic definition of the GDP can be summarized with the following equation:
GDP = private consumption + gross private investment + government spending + (exports – imports)
or, as it is commonly expressed in algebraic shorthand:
GDP = C + I + G + (X-M)
In the new report the values for that equation (total dollars, percentage of the total GDP, and contribution to the final percentage growth number) are as follows:
GDP Components Table
The quarter-to-quarter changes in the contributions that various components make to the overall GDP can be best understood from the table below, which breaks out the component contributions in more detail and over time. In the table we have split the “C” component into goods and services, split the “I” component into fixed investment and inventories, separated exports from imports, added a line for the BEA’s “Real Finals Sales of Domestic Product” and listed the quarters in columns with the most current to the left:
Quarterly Changes in % Contributions to GDP
The headline number of 2.98% has finally made it to the bottom of the economic growth range (3%-4%) that we might historically expect 11 quarters into a “recovery.” That said, nearly two thirds of the “growth” comes from increasing inventory valuations — while “real final sales” remains sluggish at a 1.10% annualized growth rate and “real” per-capita disposable income is growing at only about half that rate. Other observations include:
— The line item changes relative to the “advance” report are relatively modest and generally not statistically significant. This would seem to indicate that the BEA got the “advance” guesstimate nearly dead-on right. However, on other occasions (especially 1Q-2011), it was the second revision that produced the greatest revisions (and a significant and unexpected downside revision to the headline number) — perhaps as a consequence of the more material data arriving so late that the BEA essentially had nothing new to report in the first revision.
— The headline number still masks relative weakness in the growth rates for consumer goods, consumer services and fixed private investments .
— Governments at all levels are still reported to be contracting their expenditures. This is creating the single greatest headwind for the headline number.
— The happy headline number meets general expectations, even while the key “deflater” used to create those numbers remains curious. We had been critical during most of 2011 of low quarterly deflaters that arguably artificially puffed up the headline number. In this case one could read the absolute reverse into the relatively high deflaters used (at least when compared to dis-inflationary fourth quarter data from the BLS) — which arguably could have restrained this set of headline numbers into something plausible.
Last month we expressed misgivings about the headline number masking general weakness within the detailed line items. Since this revision does not materially revise anything, our misgivings are still in place. Expenditures for consumer goods, consumer services and fixed private investments remain weak by any “recovery” standards. This headline number continues to consist mainly of a substantial inventory growth that is unsustainable — and which (of necessity) must reverse at some point in time.
And recent consumer debt data indicates that the already relatively weak growth rates for consumer expenditures were likely to have been at least partially drawn forward from the first quarter of 2012. When coupled with the current surge in at-pump gas prices we expect that the already anemic “real” per-capita discretionary spending numbers will weaken even further into summer.
And lastly, we can not stress enough the curious dilemma presented by the “deflater” used by the BEA for 4Q-2011. The number (+0.86%) is more than 3% higher than the dis-inflationary rate for the 4Q-2011 CPI-U reported by the BLS. We understand that they measure different parts of the economy for different purposes, but a 3% difference in measured inflation rates is still significant enough to raise questions. But what is particularly strange about the deflater is that a lower deflater more in line with the dis-inflationary numbers published by the BLS would significantly boost the headline number to far above the “modest recovery” range currently being reported — in fact something closer to a highly implausible (if not embarrassing) 6% growth rate that (if real) would require significant monetary and fiscal policy reversals.
At the risk of seeming to be the perpetual curmudgeon we are not about to sell the farm to buy this set of data. It is at best a set of mixed signals masking consumer and investment weakness behind a happy headline number consisting mostly of inventory building. At worst it is held captive by inflation assumptions at material variance with sister agencies — that just happen to generate plausible numbers.