Goldman thinks we may be underestimating US GDP by a full percentage point, double the measurement error observed in 2005.
That’s according to a lengthy piece timestamped Sunday evening from the desk of Spencer Hill.
The note is a dozen pages long, and begins with the bank weighing the relative merits of time traveling back to 1991. “Traveling back in time to witness the collapse of the Soviet Union and the birth of grunge music would be exciting in many ways, but 1991 telecom technology would not be one of them”, Goldman writes, illustrating the point by way of a Radio Shack ad from that year which shows that “it would require over 10 devices and $3,000 dollars to replicate even the most basic functions of today’s smartphones”.
(Goldman)
In light of the Radio Shack observation, the bank thinks it’s pretty remarkable that, as you can see in the right pane, “tech-oriented consumption growth… is rising at its slowest pace in the post-war period”.
Goldman says “measurement error” in rapidly evolving industries is likely to blame. The bank cites a rather glaring discrepancy between smartphone sales and nominal spending on telephone equipment. “We estimate that even after upward revisions to spending and downward revisions to inflation, smartphones continue to represent between $100bn and $225bn of missing real consumption”, the bank says, on the way to suggesting that while an economic census conducted once every half-decade might be sufficient to keep apprised of “most mature consumption categories”, that lag between surveys is woefully inadequate for other industries.
After noting that the three most recent revisions added 0.05pp, +0.2pp and +0.1pp to trailing 10-year growth, Goldman says that in their view, “price measurement issues go well beyond smartphones with healthcare and business ICT among the more problematic categories”. Specifically, Hill says that although some of the measurement problems for hardware and software have been resolved, “quality adjustment in the healthcare sector remains in its infancy, as new and improved medical procedures are not incorporated into price and productivity statistics”. The table below shows that methodological improvements notwithstanding, CPI is probably overstated by 0.4pp. That number doubles if you include “‘free’ digital goods”.
(Goldman)
Moving along, Goldman notes that if they are correct “that technological change is not fully reflected in the real output statistics, we might expect easier-to-measure profits and incomes to be regularly above those justified by the pace of production and expenditure”. As it turns out, that is the case. Domestic income has risen more quickly than GDP since around 1995, in contrast to history.
And so, Goldman derives the following formula for nominal GDP mismeasurement:
Nominal GDP Bias = (GDI – GDP) + Misclassified Foreign Profits + Free “New Economy” Products
You’ll note that there are two additional terms there, one of which is meant to capture the relocation of domestically generated profits to tax havens. As Goldman writes, “profits of US subsidiaries located in these countries have grown dramatically faster than those of US firms and establishments residing domestically or in other geographies”.
(Goldman)
Ultimately, when you combine the GDI-GDP gap and the foreign profit shifting, Hill says you end up with nominal GDP growth that’s understated by between 0.10-0.15pp/year since 1994.
And that’s not the end of it – not by a long shot.
You’ll note that Goldman’s formula contains a third term: Free “New Economy” Products. Suffice to say they’ve done the homework on this, and while someone out there is in a position to debate the point, that someone isn’t me and it probably isn’t you either. Here is a quick excerpt that captures the gist of it:
Since the 2017 publication of Bartering for ‘Free’ Information that suggested just under 0.1pp of missing annual growth—a result which at the time we viewed as conservative–several much larger estimates have been proposed. Byrne and Corrado (2019) for example estimate a GDP-growth boost from consumer telecom worth 0.5pp per year. They construct models of consumer utility using granular telecom price data, household time use, and an assumption of complementarity between hardware and network use. Separately, the “willingness to accept” approach of Brynjolfsson et al. (2019) (how much would you need to be compensated to give up Facebook for a month?) suggests a much larger impact. Smartphone cameras alone are found to generate 0.6pp of annual growth in broadly measured consumer welfare (known as “GDP-B”, where the “B” stands for “Benefits”).
Given all of this, how does one decide “what counts” when measuring GDP? Goldman has a Venn diagram for that which comes with something of a caveat.
“The advent of digital and online activities has blurred the lines between the economic and household spheres”, the bank says, adding that “much of what we do on Facebook is a substitute for traditional household activities that are not in the economic realm to begin with”.
In other words, “willingness to pay” analysis runs the risk of producing, to quote Hill, “implausibly large estimates [as] the migration of household leisure time from the offline realm to the digital sphere does not necessarily mean the economy is growing any faster”.
(Goldman)
What does it all mean? Or, more to the point, just how far off are we in terms of measuring economic output. Well, according to Goldman’s efforts to roll up all of the concerns touched on above and account for all the caveats, about 1% – off that is.
“In our central estimate, we estimate that the pace of annual real GDP growth is understated by around 1.0pp of GDP, up from 0.5pp in 2005 and 0.3pp in 1995”, the bank says. Importantly (considering this is something of a hot-button issue), Hill also writes that the analysis “suggests nearly half of the slowdown in measured productivity growth since the financial crisis can be explained by measurement issues”.
(Goldman)
It goes without saying that there’s a lot of ambiguity in all of this, which means the results are, to quote the bank, subject to a “large” degree of “uncertainty”.
But hey, that’s fine. Because after all, “uncertainty” is just what’s needed to get this economy firing on all cylinders with preemptive rate cuts.
(That concluding sentence is something of a non-sequitur, but readers will agree the punchline was just too good to pass up.)
good article. it “makes sense”, but that doesn’t mean it is correct and could be impossible to prove. but i do agree that updates need to be made more often since the rate of change is ever increasing in technology.