Last time: productivity and the limits to capital deepening. This time: the miracle of the Great Depression.
The Great Depression was the most technologically-innovative period in the history of the United States.
You think that sounds ridiculous, but here’s a good intuitive illustration of the fact: Capital and labor inputs fell dramatically during the contraction. They did not fully recover until after mid-1942, as the US geared up for war. No matter how capital and labor inputs are weighted, labor and capital inputs were only as high in 1941 as they had been in 1929.
But real output was between 33 and 40 per cent higher. That means that productivity improved somewhere on the order of 2.3 to 2.8 per cent per year over that twelve-year period. Those are productivity growth rates unmatched in the twentieth-century United States – and unmatched in US history.
How did it happen? Alexander Field’s “Technological Change and U.S. Productivity Growth in the Interwar Years” has some answers.
One way of figuring out what drove aggregate productivity gains is by breaking them down by sector. So what sectors drove productivity growth from 1929 to 1941?
There were two big productivity movements in the twentieth century. The first was in manufacturing, which began in the 1880s but which was most rapid during the 1920s. The second was improvements in transportation, which began in the 1920s, but which was most rapid during the 1930s.
Sectoral Contributions to TFP Growth within the Private Nonfarm Economy, United States, 1929–1941
i. Transport and public utilities
Transport and public utilities show the highest rates of productivity growth (4.5 per cent), but their smaller share of national income (9.2 per cent) means that they account for a smaller share of aggregate productivity growth (23.8 per cent).
Transportation productivity numbers reflect large efficiency gains in all sub-sectors. The biggest gains were in trucking and warehousing (35 per cent of total sectoral gains) and railroads (28 per cent). Electrical utilities made up 24 per cent of the sectoral growth.
The most rapid productivity gains were in trucking and warehousing (13.6 per cent per year) and in air transportation (13.7 per cent).
Now, some part of those gains should reflect increased public investment – in roads, bridges, tunnels, and subsidized airmail, if nothing else. How much do we reduce aggregate productivity growth if we account for that public investment as capital input?
A bit, but not much. The 1930s remain the most technologically-productive decade of the twentieth century by a fair margin.
ii. Manufacturing
Manufacturing productivity growth about halved from the 1920s to the 1930s, from 5.12 per cent to 2.76 per cent. But productivity growth during the 1920s was much more unbalanced: more than four-fifths of productivity growth during the 1920s could be attributed to manufacturing alone, compared to about 50 per cent during the 1930s, and about 27 per cent between 1948 and 1973.
The 1930s was still the second-best period for manufacturing productivity growth in the twentieth century.
Growth of TFP, Labor, and Capital Productivity in Manufacturing, 1889–2006 (percent per year)
Note, for future reference, that during World War II, the highest period of government and equipment investment in the history of the US economy, productivity growth in manufacturing was negative. Although much government investment in machine tools and equipment in the aluminum, synthetic rubber, aircraft engines, and aviation fuel refining sectors was sold off to the private sector, the capital stock was not well-fitted to peacetime production.
The sub-sectoral breakdown in manufacturing suggests that productivity gains of the 1930s were fairly broad-based. The breakdown for 1929–41 and 1941–48 is not available here, but comparing 1929–37 and 1929–48 suggests some of the effects of war production. Data through 1937 suggest that progress was particularly high in beverages, tobacco, textiles, paper, rubber, leather, electric machinery, chemicals, instruments, and petroleum and coal products.
The productivity growth of 1995–2000, by contrast, reflects productivity gains concentrated in manufacturing, manufacturing productivity gains concentrated in durables, and durables productivity gains concentrated in equipment – productivity gains, that is, concentrated in two sub-sectors of a single sector.
iii. Wholesale and retail trade
Field doesn’t actually have a particularly good answer to why productivity in trade grew as rapidly as it did in this period. His earlier papers, suggesting stagnant productivity in services in this period relative to manufacturing and agriculture, pointed to the negative impact of the personal automobile when explaining the low productivity growth in the 1920s. But that does not do much to account for the high productivity growth in the 1930s.
He suggests here the success of improved retailing by large firms – A&P supermarkets, Woolworth’s, others – which took advantage of the long-distance telephone network and the flexible distribution associated with improved telecommunication and the surface road network to serve the suburban middle classes. But that is more of an ad hoc answer than anything else.
iv. The Driver: Disembodied Technological Change
As Field points out, there are still two questions here: (a) Why did productivity improvements halve from the 1920s to the 1930s, and (b) Why didn’t they fall more?
The answer to the first question is that the gains from using small electric motors, and from reconfiguring factory floor plans from the multi-story layouts suited to steam power to single-floor layouts possible with electrification were partially exhausted by the end of the 1920s. But not wholly exhausted.
In 1933 Cadillac consolidated its production of drive trains from four floors to one, leaving the other floors free. In 1934 Packard cut half of its floor space requirements, freeing a whole building. Similar improvements were made by Westinghouse and Western Electric.
Why didn’t they fall more? A few answers: a trend to larger capacity equipment and fixed installations; the use of new materials, including plastics and alloy steels; investments in instrumentation that were both capital- and labor-saving; better chemical processes for extracting minerals and processing agricultural materials; and advances that increased thermal efficiency.
The big story here is that these improvements happened without increasing capital inputs. That means that they must reflect ‘disembodied’ technical change, rather than improvements embodied in new equipment.
Some examples: the percentage of sugar extracted from beets in refining, and the percentage of metal extracted in mining, improved substantially thanks to new chemical processes. New chemical processes lowered capital depreciation rates – increasing the life of wooden railroad ties from 8 to 20 years, reducing the time to paint a car from three weeks to a few hours (quick-drying lacquers), reducing oxidization on railroad cars (stainless steel), and otherwise lengthening the life of tools and moving parts (chrome plating).
The bigger improvements came from thermal efficiency. In 1941, electricity output was 87 per cent higher than it had been in 1929 – and labor productivity was 126 per cent higher, and capital productivity 73 per cent higher. In part, those gains reflected larger plants and bigger boilers, for which kilowatt hour costs were lower.
But they also reflect improved techniques. Topping techniques used the exhaust steam from high-pressure boilers to heat lower-pressure boilers, increasing capacity on existing stations by 40 to 90 per cent with no increase in fuel or labor costs, and little cost in upgrading capital. Other techniques included low-cost, high-return investments in insulation, or configurations that captured exhaust gas from stacks and used them to preheat air for combustion, to preheat materials, or generate steam.
The continued trend to larger units in equipment and fixed installations helped efficiency too. Industrial locomotives were 11.4 tons in the 1930s, versus 7.4 in the 1920s. The average capacity of a power shovel rose from 1.73 cubic yards in the early 1920s to 3.28 cubic yards in the mid-1930s. Capital and operating costs per unit of output fell here, as they did in electric power units and mills.
New materials reduced costs and depreciation. Plastics replaced wood and metal parts, saving in fuel, fabrication, and capital costs. Tungsten carbide blades wore out less rapidly and did not needs as much downtime. Carbon steel blades had to be removed and resharpened after 60 feet when cutting phenol resins. Carbon alloy blades cut 10,000 feet without refitting. It now took a single day to do what once took a month.
And finally, a focus on better instrumentation in what little capital investment there was helped improve process control, reducing downtime and maintenance costs. In 1920s, cracking units in petroleum refining had to be cleaned every four to five days. Instrumentation reduced this to one to two months. Hand-controlled boilers needed re-bricking every three months. Instrument controls never needed to be re-bricked at all.
v. Conclusion
The result was higher rates of productivity improvement in the 1930s than in any other decade in the twentieth century United States – and thus any decade ever in the history of the US. That technological change cleared the way for the high growth of the war and postwar period, once all the factors of production were fully employed.
Next time: the R&D boom of the 1930s; the myth of wartime productivity; and the limits of postwar growth.
floresapriles likes this
rageofthedogstar likes this
information-catalysis reblogged this from xhxhxhx
walpurgisnacht-nightly likes this
prophecyformula likes this
ogingat likes this
xhxhxhx posted this