‘Actions speak louder than words’ is a crucial rule in economics as in every sphere. It therefore damages economic policy making if myths on the development of the US economy, the world’s largest, are believed instead of the real driving forces of the US economy being studied.
A myth proclaimed regarding US economic development is that it is due to ‘individual entrepreneurship’ and uniquely US ‘creativity’. Economic facts show an entirely different reality. US economic growth is overwhelmingly driven by massive accumulation of capital inputs, labour and intermediate products. As Professor Dale W. Jorgenson of Harvard University, the foremost statistical expert on US growth, whose work is now the basis for the official economic accounting methods of the OECD, UN and US, succinctly summarised regarding US post-World War II economic development: ‘the driving force behind the massive expansion of the U.S. economy… was a vast mobilisation of capital and labour resources… Capital and labor inputs combined contributed… more than three-fourths of the growth of output. By contrast, advances in the level of productivity contributed… less than one fourth.’1
Direct measurement unequivocally shows that the pattern of US growth based on huge mobilisation of capital, labour and intermediate products remained unchanged for over six decades from 1948-2014 while indirect data shows that this was also the US pattern of development prior to World War II. The fundamental pattern of development of the US economy therefore remained constant throughout its historical rise, continued during the entire post-World War II period, and persists to the present day.
The aim of this article is therefore to present the data on the real forces of US economic growth using the most modern statistical methods. As the US is the world’s most advanced economy this naturally casts light on what will be the driving forces of an economy’s growth as it becomes more developed.
TFP, Capital and Labour
To show the contrast between the reality of US economic growth and some mythologies spread concerning it Figure 1 shows the percentage contributions to US GDP growth from 1989-2014 calculated according to the most modern statistical methods approved by the OECD, UN and US. This analysis is carried out in the ‘growth accounting’ categories of Solow – capital, labour and Total Factor Productivity (TFP). ‘TFP’ measures increases in output which are not due to inputs of capital and labour but to other forces such as economies of scale, technology, entrepreneurship, managerial improvements etc. TFP therefore includes, but is not confined to, the forces of ‘individual entrepreneurship’, ‘creativity’ etc. of the US economy.
Over the entire period 1989-2014 US average annual GDP growth was 2.4%. Of this growth 76% was due to increased inputs of capital and labour and only 24% to increases in TFP. The contribution to US economic growth of increases in inputs of capital and labour was therefore three times as great as TFP growth.
Of these ‘Solow factors’ of production by far the most quantitatively important was capital investment which accounted for the majority, 51%, of US growth. The second most important input was labour – accounting for 25% of US growth. The least important factor was TFP – accounting for only 24% of US growth.
It should also be noted that the annual increase in US TFP was small – 0.6% a year. Therefore, if US GDP growth had depended wholly or primarily on TFP US economic development would have been extremely slow. In contrast US capital accumulation accounted for 1.2% GDP growth a year. The role of capital investment in the development of the US economy was therefore by itself more than twice as great as that of TFP.
Nor was there a significant shift in the pattern of US economy over time. Studying the earlier period post-war 1948-1979 period Jorgenson, Gollop and Fraumeni’s monumental study of the US economy found TFP accounted for 24% of US growth, labour 31%, and capital 46%. Therefore, in the post-war period US economic growth became somewhat more capital intensive, with the role played by labour falling, but there was no increase in TFP’s role. Inputs of capital and labour contributed three times as much to US economic growth as TFP in the entire post-World War II period.
Nor is the US economy unusual in showing the decisive role of capital investment among ‘Solow factors’ of economic growth. The most recent comprehensive survey of countries representing more than 90% of world GDP, by Vu Minh Khuong, using OECD/UN/US approved statistical methods, concluded: ‘The share of capital input in GDP growth exceeded 50 per cent for the world and its sub-samples.’
This factual data establishes that among ‘Solow factors’ of growth the development of the US economy is primarily driven by capital investment - TFP accounts for less than a quarter of growth. However, even this data substantially overstates the role of TFP in US economic growth – i.e. factual reality is even further away from myths regarding the US economy.
While fixed capital investment is the most important ‘Solow factor’ of production factually it is only the second most important driver of economic growth. The most important factor is the growth of ‘intermediate products’ – the outputs of one industry used as inputs into another (for example the output of the microprocessor industry is an input into computers, production of motor engines an input into the automobile industry etc.).
Solow did not include intermediate inputs in his original formulation of macro-economic growth accounting. At the level of the economy as a whole the sum of all inputs and outputs necessarily balance. But if growth in individual industries is studied each economic sector has inputs from others and rigorous growth accounting requires these inputs must be measured.
The factual results of such studies are striking and of great economic importance. Growth of intermediate products, by measuring inputs of one sector into another, measures the increasing interconnectedness of the economy – put in theoretical economic terms it measures how division of labour is increasing.
This issue is not only of factual but theoretical importance. The first sentence of the first chapter of the founding work of economics, Adam Smith’s The Wealth of Nations, states unequivocally: ‘The greatest improvement in the productive powers of labour, and the greater part of the skill, dexterity, and judgment with which it is directed, or applied, seem to have been the effect of the division of labour.’2 The rest of Adam Smith’s magnum opus founding economics flows from this conception.
Marx, for somewhat different reasons, termed this economic process first comprehensively analysed by Smith as ‘socialisation of labour’ but these reasons were unconnected to present issues. For current purposes the processes described by Smith and Marx, of division/socialisation of labour may be regarded as the same.
Whether the analysis of Smith/Marx is correct is a factual question. If the Smith/Marx analysis is correct then economic development should see increasing division/socialisation of labour which would be reflected in a rise in intermediate products.
It is therefore a result of the greatest importance that modern economic statistics establish unequivocally that Smith and Marx were factually correct including for the US. The growth of intermediate products, reflecting increasing division of labour, is even more rapid than that of fixed capital investment.
In the case of the US Jorgenson, Ho and Stiroh found analysing individual economic sectors that the growth of intermediate inputs was greater than that of all ‘Solow factors’ (capital, labour, and TFP) combined.3 Analysing the US economy as a whole Figure 2 shows that the average percentage contribution to growth in economic sectors was intermediate products 52%, capital 24%, labour 15%, TFP 9%. This data is particularly striking as it covers the period 1977-2000 when the US ICT boom was at its peak – and it might have been supposed that the contribution of TFP to US growth would be particularly high.
Analysing in more detail the 41 sectors of the non-government non-household US economy, Jorgenson, Ho and Stiroh found: ‘In terms of medians for these 41 industries, the typical contribution from intermediate inputs was 1.2 percentage points, compared to 0.5 percentage points from capital and 0.3 from labor inputs … This analysis shows that intermediate inputs play a critical role in explaining economic growth… Investment in tangible assets is the second most important source of growth of output across U.S. industries.’4 In addition to the small role played in US economic growth by TFP they also noted: ‘The results show that the quantity components (capital stock and labor hours) dominated the quality components (capital quality and labor quality) as sources of economic growth.’5 In addition to the small role played in US economic growth by TFP they also noted: ‘investment in tangible assets is the most important source of economic growth in the G7 nations. The contribution of capital inputs exceeds that of total factor productivity for all countries for all periods.’6
This trend, once again, was not a change from earlier periods of US growth. Analysing the US economy in the earlier post war period of 1948-79 Jorgenson, Gollop and Fraumeni found: ‘the contribution of intermediate input is by far the most significant source of growth in output. The contribution of intermediate input alone exceeds the rate of productivity growth for thirty six of the forty five industries for which we have a measure of intermediate input… the predominant contributions to output growth are those of intermediate, capital and labour inputs. By far the most important contribution is that of intermediate input.'7
It should be noted that this process in the US is typical for other economies - including China. Regarding rapidly growing Asian economies:
- For South Korea, Hak K. Pyo, Keun-Hee Rhee and Bongchan Ha found regarding material intermediate inputs: ‘The relative magnitude of contribution to output growth is in the order of: material, capital, labour, TFP then energy.’8
- For Taiwan Province of China, analysing 26 sectors in 1981-99, Chi-Yuan Liang found regarding intermediate material inputs: ‘Material input is the biggest contributor to output growth in all sectors during 1981-99, except… seven’.9
- For mainland China, Ren and Sun found that in the period 1981-2000, subdivided into 1984-88, 1988-94 and 1994-2000: ‘Intermediate input growth is the primary source of output growth in most industries.’10
The above data is for domestic division of labour. It clearly demonstrates that use of intermediate products is the most powerful factor on the economy’s ‘supply side’. However, division of labour in a modern economy extends not only domestically but internationally – the phenomenon of ‘globalisation’. This drives modern economic structure. In particular:
· Intermediate products constitute the largest part of international trade – accounting for approximately 40% of all goods trade.
· Trade in intermediate products is concentrated in advanced economies and East Asia - economies which have intermediate products as the most rapidly growing part of national production.
· International trade has expanded rapidly as a percentage of world GDP.
That international trade is the expression of division of labour of course explains the well-established finding that international economic ‘openness’ is positively correlated with economy growth.
Intermediate products and capital
Finally, it should be noted that although intermediate products and capital are traditionally treated separately nevertheless from a fundamental accounting point of view this is not correct. As Charles Jones noted in the comprehensive study of intermediate products published by the US National Bureau of Economic Research: ‘intermediate goods are just another form of capital, albeit one that depreciates fully in production.’11
Intermediate products are inputs used up in a single production cycle whereas fixed investment is capital used up (depreciated) across several production cycles. The distinction in Western growth accounting between two different forms of capital (intermediate products and fixed investment) is therefore reflected in Marx’s terminology in the distinction between ‘circulating capital’ and ‘fixed capital’ – although which terminology is preferred is not crucial for present purposes. What is crucial is that the role of different forms of capital, i.e. intermediate products/circulating capital and fixed investment/fixed capital, is the overwhelming force driving US economic growth. Taking the two together 76% of US sectoral output growth is due to fixed and circulating capital, 15% due to labour, and only 9% due TFP.
Historical rise of the US economy
Regarding earlier periods for comprehensive US data prior to World War II does not exist for the role of intermediate products - that is during the period of the rise of the US economy to global dominance. However, it does for fixed capital. This confirms that capital accumulation played an overwhelming role in the US rise to world supremacy. As Angus Maddison, former head of statistics of the OECD, noted on the means by which the US overtook the former world leader the UK to become the dominant economic superpower: ‘The rate of US domestic investment was nearly twice the UK level for the sixty-year period 1890-1950. Its level of capital stock per person employed was twice as high as that of the UK in 1890, and its overwhelming advantage in this respect over all other countries continued until the early 1980s.’12
Given it is clear US economic development is primarily driven by accumulation of intermediate products and fixed investment how could arguments be advanced for a myth that US economic growth was primarily propelled by TFP?
The first reason is explained by one of the most classically false statistical methods – taking anecdotes or individual examples instead of analysing the overall situation. In 83% of US economic sectors, an overwhelming majority, TFP is not the main source of economic growth. However, this means that in 17%, a small minority, of US economic sectors TFP is the main source of growth. A ‘myth’ is created by taking an example from the small minority of industries in which TFP is the main source of growth, accurately pointing out that TFP is the main source of growth in it, ignoring the overall numerical context, and then falsely claiming this example shows that TFP is the main source of economic growth. This method, statistically that of using ‘anecdotes’ rather than analysing the overall situation, is invariably a dirty method – as in this case.
The second reason is the use of out of date statistical methods which have now been officially replaced by the OECD, UN and US. Solow’s original formulation of growth accounting was, of course, on an entirely different level of seriousness to the methods of ‘anecdote’. Solow’s basic algebra for growth accounting survives – additions augment but do not overturn its algebraic framework. But the arithmetic, as opposed to the algebra, originally used by Solow was wrong for clear reasons.
· Solow did not include intermediate products. As factually intermediate products are the most important source of growth this naturally invalidates Solow’s numerical conclusions.
· Solow did not control for changes in the quality of capital and labour inputs – a parallel would be attempting to measure real wages without taking into account inflation. This means, to take an example, one hour of labour by a South Korean peasant in 1953, who was possibly illiterate, is counted the same as one hour of labour in 2015 by a South Korean engineer who has a PhD – evident nonsense. Once changes in the quality of capital and labour are taken into account the role of TFP shrinks drastically.
In summary the claim that TFP was the prime source of growth was simply based on wrong methods of measurement which have now been officially changed by international statistical agencies – for details see Jorgenson’s Introduction to the The Economics of Productivity.
Reality versus myth in US economic growth
The implications of the real trends in the US economy are clear. The US is the world’s most developed economy. It continues to develop overwhelmingly on the basis of huge mobilisation of intermediate products and fixed investment and not on the basis of TFP growth. This is in line with the pattern of development of the global economy.
The laws of economics are objective. The US is no more capable of violating them than any other country. Attempting to make US economic development dependent on TFP would condemn the US to extremely slow economic growth. The US is not stupid enough to attempt to do so. While the US proclaims in words the myth that it grows on the basis of TFP it actually develops through massive accumulation of intermediate products and fixed investment – i.e. circulating and fixed capital. This process is what in reality made the US the world’s largest and most advanced economy and it continues to power US growth.
This where the principle ‘do what I do not as I say’ is vital. The US uses one myth for propaganda but actually does something entirely different. Those outside the US who wish to develop to the US level should of course learn from US reality not US propaganda.
The wise Chinese sayings says ‘seek truth from facts,’ not ‘seek truth from myths.’
* * *
This is an edited version of an article which originally appeared in Chinese at Sina Finance.
1. Jorgenson, D. W., Gollop, F. M., & Fraumeni, B. M. (1987). Productivity and US Economic Growth. New York: toExcel p1-2.
2. Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations Volume 1 (1981 ed.). Indianapolis: Liberty Edition p22.
3. Jorgenson, D. W., Ho, M. S., Samuels, J. D., & Stiroh, K. J. (2007, September). Industry Origins of the American Productivity Resurgence. Economic Systems Research, 19(3), 229-252.
4. Jorgenson, D. W., Mun, S. H., & Stiroh, K. J. (2005). Productivity Vol 3 - Information Technology and the American Growth Resurgence. Cambridge, Massachusetts: MIT Press p305.
5. Jorgenson, D. W., Mun, S. H., & Stiroh, K. J. (2005). Productivity Vol 3 - Information Technology and the American Growth Resurgence. Cambridge, Massachusetts: MIT Press p311.
6. Jorgenson, D. W., Mun, S. H., & Stiroh, K. J. (2005). Productivity Vol 3 - Information Technology and the American Growth Resurgence. Cambridge, Massachusetts: MIT Press p84.
7. Jorgenson, D. W., Gollop, F. M., & Fraumeni, B. M. (1987). Productivity and US Economic Growth. New York: toExcel p200.
8. Pyo, H. K., Rhee, K.-H., & Ha, B. (2007). ‘Growth Accounting and Productivity Analysis by 33 Industrial Sectors in Korea (1984-2002)'. In D. W. Jorgenson, M. Kuroda, & K. Motohashi (Eds.), Productivity in Asia: Economic Growth and Competitiveness (pp. 113-145). Cheltenham: Edward Elgar.
9. Liang, C.-Y. (2007). ‘Industry-Wide Total Factor Productivity and Output Growth in Taiwan, 1981-1999’. In D. W. Jorgenson, M. Kuroda, & K. Motohashi (Eds.), Productivity in Asia: Economic Growth and Competitiveness (pp. 146-184). Cheltenham: Edward Elgar.
10. Ren, R., & Sun, L. l. (2007). Total Factor Productivity Growth in Chinese industries, 1981-2000. In D. W. Jorgenson, M. Kuroda, & K. Motohashi (Eds.), Productivity in Asia, Economic Growth and Competitiveness (pp. 76-112). Cheltenham: Edward Elgar.
11. Jones, C. I. (2008). Intermediate Goods and Weak Links: A Theory of Economic Development (Version 2.5). U.C. Berkeley and NBER p2.
12. Maddison, A. (1991). Dynamic Forces in Capitalist Development. Oxford, UK: Oxford University Press p40.