I’ll be brief, because first and foremost, the recent uptick in productivity growth that I’m about to show you may be statistical noise. These are jumpy data. But in case this sticks, I did want to lay down a marker and tout some potential implications.
This morning’s revised productivity report has output per hour up a rousing 3 percent in Q3. That’s an annualized, quarterly rate, and OTE’ers know I like to filter out some of the noise by looking at year-over-year changes.
So, the table below shows the recent acceleration in year-over-year changes in the key variables.
Since productivity growth equals output minus hours growth, we can decompose the increase. It’s all about faster output growth; hours of work have slowed a touch.
Now, there could be a bit of the hurricane season in there, as that hurts employment but boosts (gross) output. Also, as we close in on full employment, employment and hours growth will naturally slow.
What’s a bit disconcerting here–with even bigger caveats re data noise in this series; I’m pretty skeptical of this wage result–are the unit labor costs, which measure compensation growth relative to productivity growth. We typically expect pay to rise along with productivity growth, at least at the average (if not so much the median), especially as the job market tightens. And, as Dean Baker and I point out here, you see real wage pressures in some series. But compensation slows of late in this series. And slower compensation amidst faster productivity growth drives down unit labor costs.
That dynamic–productivity rising faster than comp–also drives down labor’s share of national income, as the next figure shows (the BLS labor share data is more pessimistic than other series; part of the decline is due to imputations of self-employed earnings, but all measures show a similar trend).
Source: BLS
Towards the end of the figure, if you squint you can maybe see the beginning of a trend reversal around 2015, but the series has since turned down again. It’s an unsettling picture of where most income growth has gone in recent decades.
At any rate, if this productivity acceleration sticks, and that’s a big “IF,” here are some implications:
–The Fed has even less reason to raise rates, as faster productivity growth can pay for non-inflationary wage gains.
–Score one for the full-employment-multiplier theory a number of us like to tout. As the job market tightens, firms must find new efficiencies to maintain profits margins as production costs rise (another reason the wage result above looks fishy to me).
–Get ready for a lot of ridiculous claims that the tax cut and MAGA are responsible for the acceleration.
Manufacturing Labor Productivity fell 4.4% as referenced in the same report. That is especially significant and troublesome. Despite manufacturing employing only 8.8 percent of the workforce, the sector’s impact on the rest of the economy is disproportionately large. Interesting report here, notes gross output of manufacturing exceeds 1/3 of GDP (meaning sales instead of just intermediate value add)
http://www.epi.org/publication/the-manufacturing-footprint-and-the-importance-of-u-s-manufacturing-jobs/
K, then how did the service sector do it, increase by more than 3% (cause manufacturing was dropping) without pushing wages. Comments as follows:
1) The record of the last 37 years shows business is very adept at denying workers any benefit from productivity gains. Thus we see more of the same. It would be surprising to see a change in this 30 year trend.
2) Business currently runs under capacity, 1 to 10% according to projections of growth that were given before, during, and after the last recession. Fixed costs are fixed, until capacity is reached productivity gains are simple arithmetic.
3) The labor pool’s reserve army is also like having extra capacity, and the million prime age participation gap (the historic low figure currently 81.6%) means it’s still a buyers market.
4) Half the workforce is off the clock, salaried, exempt from any requirements for 40 hour weeks or overtime pay. Put in an extra 10 minutes a day for half the workforce and you’ve got an additional .5% productivity increase that appears like magic and doesn’t show up as extra hours worked. (uhm, I think this would happen depending on how hours measured)
5) Some tech changes could show up, 3 million call center workers, new software cuts a minute call down to 57 seconds, a 5 percent increase in productivity.
6) A new app or new version is now priced 3% higher, and doesn’t count as inflation because it’s new or improved. No new workers hired to make this, but productivity increases.
7) Less is more. The management consultant does more work by spending less time on each engagement, but produces the same number or reports per client, charges each client the same, but takes on 3% more clients. The big take away is a service economy is hard to measure.
8) Hidden cost of apparent gains. Dean Baker had something about in a service economy customer wait time goes unmeasured. Cut your work staff or handle increased workload, by 1% or 3% and queues increase. In the lulls, the staff catches up, but there is still longer wait times and poorer service for consumers. Again, service economy productivity is hard to measure.
These points all are directed at showing ways the productivity increase doesn’t reflect a healthy economy.
GE announced is cutting 12K jobs which will increase productivity. But the core problem with jobs and wages is two fold. First, the GDP with manufacturing companies such as GE is suffering from the inability to sustain growth in jobs by creating new demand with radical innovation rather than suffering from the impact of radical innovation in industries such as energy which is shifting demand away from steam and gas turbine fossil fueled electric power generation – what has been a major source of GE revenue. GE’s investment in digital services in the Industrial Internet has paid off yet. But, the root cause is shared by economics not being able to provide new theory as guidance for policy to sustain growth by managing radical innovation. There are consequences by focusing on theory that explains only equilibrium. Second, a gap in financial accounting that determines and justifies compensation paid to labor has ignored intangible capital which includes the contribution of knowledge although most business investment in the GDP and the valuation of businesses since about 1985 has come from investments in intangible capital rather than tangible capital. Businesses have exploited the gap to limit wage growth as decoupled from productivity and economics hasn’t been able to provide new theory as guidance for policy to properly compensate workers by considering the contribution of their intangible capital.
According to the World Intellectual Property Report 2017: Intangible capital in global value chains – intangibles contribute more than twice as much in value than do tangibles in manufacturing industries. And in service industries, the ratio is much higher. Intangibles account for around one-third of the production value of a purchased product – or some 5.9 trillion United States dollars in 2014 – across 19 manufacturing industries.
Obviously, IP for drugs raises the price considerably. The intangible including R&D, branding and distribution add more value than the production process (including the investment in tangible capital) and twice as much value as the overall investment in tangible capital. The GOP tax cut proposal ( and supply-side economics) incorrectly ignore intangibles and focus all investment in tangible capital as the major driver of GDP growth.
Dude, the Fed is already behind and is trying to use the rise to “catch up”. The key is debt financing and its cycle. I suspect interest rates are playing catchup.
Real hourly compensation for non farm business is down -1.1 compared to a year ago. This is in the same report “Productivity and Costs Third Quarter 2017, Revised.” which this blog post above links to, https://www.bls.gov/news.release/pdf/prod2.pdf That hourly compensation number appears to be a serious problem. Am I missing something?
Also to find the figure -.1 compared to 1.5% on productivity growth, as shown in the table given in the blog post, you can look here:
https://www.bls.gov/news.release/prod2.t01.htm
It’s in the second table down, first column, sixth row, 2016 III (third quarter) in a table named “Percent change from corresponding quarter of previous year” The figure represents how much productivity increase the past 12 months. A bit further down you’ll note a table marked Indexes 2009=100, meaning the current 108.8 indicates 8.8 percent cumulative increase, or about .5 percent average per year since Great Recession. Even if the improved current 1.5% rate of growth continues, that rate is historically low.
But still on that page I couldn’t help noticing the figure for 2016 productivity increase, 0. How was that not front page news? I had to google considerably to confirm. A cynical interpretation would be media bias buried the story.
http://www.zerohedge.com/news/2017-08-09/productivity-growth-rebounds-q2-thanks-slump-unit-labor-cost-growth
2016 Was Just Revised Down To The Worst Year For US Productivity Since 1982
“Annual average productivity change for 2016 revised to 0.1 percent decline from 0.2 percent increase, marking first annual decrease since 1982;”
https://qz.com/946675/us-productivity-growth-was-negative-in-2016-and-economists-arent-sure-why/
US productivity growth is negative and economists aren’t sure why
“According to the US Bureau of Labor Statistics, for the first time since the global financial crisis, multifactor productivity growth was negative in 2016. ”
Also very relevant to productivity and not discussed enough, the dramatic fall in R &D
https://www.nytimes.com/2017/06/29/opinion/made-in-america-the-bad-news-and-the-good-bad-news.html
Scroll to bottom to see R&D dropping by 2/3 from the 1960s, and 1/2 from the 1990s
Chart is labeled “Spending on Infrastructure and R.&D. Declining”
(also the Math scores chart is misleading because comparing similar socio economic groups leaves the U.S. with similar scores, our children are poorer, not dumber)
Also interesting links on productivity found:
https://www.bls.gov/opub/btn/volume-6/below-trend-the-us-productivity-slowdown-since-the-great-recession.htm
https://www.bls.gov/opub/mlr/2016/article/measuring-quarterly-labor-productivity-by-industry.htm
Would be Interesting to know more about what causes productivity measurements to jump so much from quarter to quarter.
This is a replacement for the previous, and note although the post says there are 3 comments to view, only 2 show to readers because any reply to an unpublished comment does not appear.