INDUSTRIAL PRODUCTION
Industrial production is clearly slowing as total output and manufacturing output fell 0.2% in September– the difference between total and manufacturing output is utilities and mining and mining includes oil and gas production. The slowdown was widespread as output of oil dirlling and the information technology sectors both fell. About the only increase was an expansion of auto and light truck production from 7.75 to 7.84 million units. As the chart below shows after rebounding strongly in the early stage of the recovery industrial output is starting to closely followed the pattern displayed in weak recoveries.
My estimate of monthly productivity growth in manufacturing is also weakening sharply as the smoothed year over growth has slowed to 3.4% versus a peak rate of 9.0% in January of this year. Moreover, the smoothed three month growth rate and the unsmoothed monthly increase of only 0.1% in September implies that the slowing of productivity growth is not quickly reversing. Weaker productivity growth is bad for profits growth that historically leads income and employment growth. Weak productivity could imply that manufacturing employment growth was improving. However, the index of manufacturing aggregate hours worked peaked in May at 78.0 as compared to its September reading of 77.3. So on balance, prospects for continued manufacturing employment growth do not look encouraging.
Spencer: Are the declines in prodcution and the lower rates of productivity a cause and effect phenomena? In other words, does productivity have a more likely opportunity to increase when production capacity is at or near peak?
Productivity is very much a product of accelerating or slowing growth. When growth is accelerating is when productivity growth is strongest and when output growth is slowing is when productivity tends to be the weakest. Historically, productivity growth generally peaks around the end of a recession or very early in a recovery and slow as the cycle proceeds. Productivity growth use to be a great leading indicator –leading real GDP growth be about two quarters.
A number of high-frequency economic indicators have improved a bit in September, suggesting that the June-August period is, for now, the worst of the slowing in economic activity. Though IP jumped in July, it has just put in the slowest 3-month gain in about a year. Assuming it shows the same pattern as other indicators, we would see improvement in Q4. The dollar might help with that.
Problem is, there are a number of other problems that seem likely to derail the bounce. ARRA spending continues to fade, and the overall fiscal stance is becoming more contractionary. For some folks – those who just leave their exemptions as they were – it is probably already too late to avoid a “tax hike” in the form of new tax tables. Congress won’t get back to the bush tax cut extension in time to avoid it.
Thanks. Looking at your first chart, it appears the current cycle is a mirror of the 5 mild recession periods. (Since the trough.) What do unemployment levels look like during those periods compared to current levels?
spencer – “About the only increase was an expansion of auto and light truck production from 7.75 to 7.84 million units.”
The growth in vehicle production is in light, medium, and heavy duty trucks, not cars. As shown in Table 3: http://www.federalreserve.gov/releases/G17/Current/table3.htm
Table 3
Motor Vehicle Assemblies
Millions of units, seasonally adjusted annual rate
………..Apr. May June July Aug. Sept.
Autos 2.75 3.05 2.83 2.76 2.74 2.66
Trucks 4.46 4.78 4.84 5.93 5.16 5.33
Light 4.34 4.65 4.70 5.78 5.00 5.18
Medium and heavy .12 .13 .14 .15 .15 .16
Memo:
Autos and light trucks 7.09 7.70 7.54 8.54 7.75 7.84
No MG, the productivity estimates are my calculations as I said in the note. The offical data is quarterly and is not released yet..
I derive them by dividing the industrial production etimate by the hours worked data from the BLS employment report.
It is not exactly the same as the official estimates of productivity, but it is close enough to give a very good first approximation of what the official productivity data will look like when it is released.
If you look at the final line of the table you reproduced you will see the memo item, production of autos and light trucks was 7.75 and 7.84. That is the number I used in the note, not the number that included heavy trucks.
It looks like unemployment rates are rising at a much faster rate than in any recessionary period post WWII: A Historical Look at the Labor Market During Recessions – Economic Letter, Jan. 2010 – FRB Dallas