THE EARLY DAYS
The study of I-O has always been about folks trying to demystify the world (Kurz, 2021). So with that, we present to you a brief history of I-O with special thanks to all the hard work of economist and I-O historian Heinz D. Kurz.
On the main entrance to the city of Babylon, sat the Ishtar gate. It was known to have shown a summary, by year, of their economic performance in terms of barley. They would note on clay tablets the total output of barley needed as well as the total input of barley required to produce it. This is now called the “corn model” and is still referenced to this day. We look at how much corn we need to make big corn that we need (Kurz, 2011).
The Physiocrats were a group of mostly French men. Their title comes from the Greek words phýsis, meaning “nature,” and kràtos, meaning “power.” Their belief was that the entire strength of the economy came from agriculture. They had some good ideas, but their big thing in economics was that they maintained that wealth only comes from the gifts of nature, from the fact that “one plants one seed in the spring and obtains two in the fall”. Literally they were saying that only agriculture matters (Science Direct, 2021).
William Petty (1623-1687) was the next (after the corn model) to look at agricultural surplus in terms of corn output less the necessary corn input. He also included wages and land rents in his work. We know that Leontief and those that followed him acknowledge that I-O is concerned with ‘directly observable’ magnitudes and relationships, but this idea was actually first put forth by Petty. Also, Petty starts to think about production as a circular flow (Kurz & Salvadori, 2010; Kurz, 2011).
Francois Quesnay (1694-1774) is considered the leader of the physiocrats and followed the work of Adam Smith. Francois starts to describe the economy in a further analytical way. He was big on that idea that agriculture can generate a surplus and worked on that circular flow idea. His Economic Table outlined the process of production, distribution and expenditures, with the circulation of commodities and money. In his view there were three classes: the “proprietary class” (landowners), the “productive class” (ag, mining, forestry), the “sterile class” (manufacturing). Both of the working classes are required as there are flows between them: agriculture gets inputs from industry, and industry gets inputs from agriculture. Manufacturing, not so important here. The original Tableau contains a summary account of national production in France, distribution and consumption during a given year, in the mid 18th century (Kurz & Salvadori, 2006; Kurz & Salvadori, 2010).
Richard Cantillon (1698-1734) brought in the idea that the quantity of land and labor mattered - akin to what we would now call productivity (Kurz & Salvadori, 2010).
Achille-Nicolas Isnard 1749-1803 comes along and says that the physiocrats are nutty and not only agriculture is productive. He pointed out that our buddy Francois even had industry, not just agriculture, in his model, so his stuff didn’t make sense internally. So, those guys were therefore wrong in asserting that manufacturing is generally not productive. (Isnard, 1781, p. 38-39; Kurz & Salvadori, 2010).
Robert Torrens (1787-1864) starts talking about what we now call profit. That would be what is left after everything needed for production is paid out of output (Kurz, 2010). Karl Marx (1818-1883) comes in and contrasts Quesnay’s Tableau and states that labor, regardless of sector, is productive and generates a surplus (Kurz & Salvadori, 2010).
Vladimir Karpovich Dmitriev (1868-1913) comes in and puts together a “flow-input point-output framework” based on Ricardo’s theory on value and distribution. He is getting much closer to what we know as I-O in the current state (Kurz & Salvadori, 2010).
Ladislaus von Bortkiewicz (1868-1931) chimes in and hones the concept of production as a circular flow and that of a surplus. He stated that commodities are produced through a finite stream of labor inputs, and production is linear and circular (Kurz & Salvadori, 2010; Kurz, 2021).
Georg von Charasoff (1877-1931) was a forgotten figure until some re-reading of his work in 2008! He worked on some streamlining and reformulating of the early concepts of I-O (Kurz & Salvadori, 2010).
Father Maurice Potron (1872–1942) invented the idea of the I-O table in his search for “economic equilibrium.” He studied problems of ‘fairness’, and the setting of ‘fair prices’ and ‘fair wages’. He explained this through the price of a two-kilo loaf of bread. The production is the sum of certain fractions of the prices of the buildings, of the manufacturing equipment, the raw materials, and the labor. He looked at this type of production in terms of coefficients across premises, equipment, raw materials, manpower, and maintenance (Abraham-Frois & Lendjel, 2006; Bjerkholt & Kurz, 2006). So basically, Father Potron set up an I-O table 20 years before Leontief. The plot thickens.
Now, onto the one and only Wassily Wassilyovich Leontief (1906-1999). Leontief was born August 5, 1906 in Munich, Germany to Russian parents; his dad taught economics at the University of St Petersburg and his mom was an art historian. Like most children in Germany back then, Leontief had his initial education in the local gym. But during the February Revolution in 1917 (start of the Russian Revolution), his family lost their property and had to move. From 1917 to 1919, Wassily studied at home; after that he attended the 27th Soviet Union Labor School, from where he graduated in 1921, receiving his diploma at age 15 (Rose & Miernyk, 1989).
In 1921, he entered the University of Petersburg, newly renamed as University of Leningrad, to study philosophy and sociology, but that didn’t last long after he found economics. When he entered the university, he began to take interest in his country’s socio-political environment. He wasn’t a big fan of the communists and their limitations on freedom so he caught their attention (Rose & Miernyk, 1989).
He was first arrested at age 15, being caught while nailing anticommunist posters on the wall of a military barrack. For several days, he was placed under solitary confinement. This didn’t stop him, however, so he was arrested a couple of more times (Rose & Miernyk, 1989).
In 1924, he received his Learned Economist degree, equivalent to a masters (Rose & Miernyk, 1989).
In 1925, a growth was detected on his neck, which the doctors diagnosed as sarcoma. He then applied for permission to travel to Germany. Since the authorities thought that he would die, they allowed him to leave. Good news though, the growth wasn’t malignant. So, he entered the University of Berlin, working toward his PhD. In 1928, Leontief submitted his dissertation entitled, ‘Die Wirtschaft als Kreislauf’ (The Economy as Circular Flow), earning his PhD in 1929 (Bollard, 2020; Rose & Miernyk, 1989).
He started his career at the Institute for the World Economy while he was completing his dissertation and then spent a year assisting the Chinese government with transportation planning (Bollard, 2020; Rose & Miernyk, 1989).
In 1928 he published a part of his thesis entitled `Die Wirtschaft als Kreislauf ’. In it, he proposes a two-sectoral input-output system that was designed to describe the production, distribution, and consumption aspects of an economy as a single process (Bollard, 2020; Rose & Miernyk, 1989).
He stated that his early contributions were rooted in the classical tradition but that I-O as developed in the 1930s and 1940s was `an adaptation of the neo-classical theory of general equilibrium to the empirical study of the quantitative interdependence between interrelated economic activities.’ He postulated a natural or material point of view insisting that data should come from `directly observable basic structural relationships.’ The main idea is that all industries, households, and government in the economy are connected through buy-sell relationships. By modeling changes to these relationships, we can see the ripple effect throughout the economy. An I-O matrix shows the way dollars move between industries and institutions as they operate.
In 1931, he moved to the US, where he joined the National Bureau of Economic Research. Here he began his research on the US economy, which had just entered the Great Depression. He started working at Harvard in 1932. They gave him a grant of $2,000 as well as a research assistant. With that, he began constructing a table covering 42 industries for the years 1919 and 1929. It was terribly tedious work and the figures took him months to compile as everything was done with pencil and paper. In 1935, he got a computer! This was a first for a social scientist. It was, however, not an electronic computer, but a large-scale mechanical computing machine (Bollard, 2020; Kurz & Salvadori, 2010; Rose & Miernyk, 1989).
As WWII broke out, he was appointed a consultant to the Office of Strategic Services, helping the US government to plan for better industrial production, which he did while still teaching (Bollard, 2020; Rose & Miernyk, 1989).
In 1941, he published ‘Structure of the American Economy, 1919-1929’. Thereafter, he continued developing his theory, working to find out its various applications. He then got the Mark I in 1943; a super fancy computer for the time. He received grants from the Ford and Rockefeller Foundations and the Air Force. For this work, he also received a 650-punch-card computer from IBM, the Mark II (Bollard, 2020; Kurz & Salvadori, 2010; Rose & Miernyk, 1989).
In 1949, armed with his fancy computer, he split the US economy into 500 sectors; each with its own linear equation. Here again, he’s one of the first to do giant calculations on a computer. The seminal work on regional and interregional models was published by Isard (1951) and Leontief (1953). Leontief's interregional system has been called a balanced model, while Isard's is known as a pure interregional model. The former is constructed by disaggregating a national table into a set of component regions. A pure interregional model is constructed by developing a set of regional tables (Bollard, 2020; Kurz & Salvadori, 2010; Rose & Miernyk, 1989).
In 1971, he criticized mainstream economics for building complicated models without engaging in empirical research (Kurz, 2021).
In 1973, Leontief received the Nobel Prize in economics "for the development of the input-output method and for its application to important economic problems" (Bollard, 2020). Per the Nobel Prize committee: “Professor Leontief is the sole and unchallenged creator of the input-output technique” (Nobel Prize, 1973).
In 1975, he left Harvard University, disgruntled that too often teachers did not teach and researchers did not research. He then joined the New York University, where he taught both graduate and undergraduate classes. Concurrently, he continued with his research work, producing seminal works like ‘Essays in Economics, II’ and ‘The Future of the World Economy’, in 1977. In 1991, he retired from his position at NYU but continued teaching and co-publishing important papers (Kurz & Salvadori, 2010; Rose & Miernyk, 1989).
The great news is that research on I-O didn’t stop with Leonteif. Robert Remak (1888-1942) published ‘Kann die Volkswirtschaftslehre eine exakte Wissenschaft werden?’ (Can economics become an exact science?), generalizing the concept of I-O to the n-commodity case, n > 2 (Kurz & Salvadori, 2010; Remak, 1929). Remember, Leontief’s work only had those 2 industries at first.
Now we have this guy Piero Sraffa (1898–1983). He’s working in Italy at the same time as Remak in Germany and Leontief is working on his dissertation in Germany. Based on archives, both Sraffa and Leontief put forth eerily similar concepts around foundations for I-O. But no one paid attention to poor Piero, perhaps because he hung out with socialists and communists (Bjerkholt & Kurz, 2006).
Now Ragnar Frisch (1895-1973) published in 1934 in Econometrica, a journal he founded, about how to end the depression. What was his idea? An “analytical structure that exhibits remarkable similarities with Leontief’s closed model.” Funny though, no one really listened to him. Frisch claimed throughout his career that he invented I-O. There is no evidence that Ragnar and Wassily ever had a beer together. Kinda odd to think that they were this close in their thinking but not at all connected (Bjerkholt & Kurz, 2006; Kurz & Salvadori, 2010).
So they started off pretty close in their work. But again, Frisch was looking at how to end the depression, Wassily was looking at “practical problems.” So Frisch gets the dismal title of working in “depression economics.” So after WWII, what does Frisch do? He starts using Leontief’s work. He published with it, lectured on it, and continued to claim it. He even went so far as to criticize Wassily specifically. Wassily declined to answer (Bjerkholt & Kurz, 2006; Kurz & Salvadori, 2010).
Early I-O models gave zero income to the unemployed. Later models explicitly included taxes and transfer payments in impact calculations. These range from specific models dealing with the role of unemployment compensation by Oosterhaven (1983) and Batey, Madden & Weeks (1987) to the more general treatment by Golladay & Haveman (1976) (Rose & Miernyk, 1989).
Richard Stone (1913-1991) comes on the scene. In 1961 he developed the first Social Accounting Matrix (SAM). This was furthered by work done with his 3rd wife as well as by a few other groups of researchers (Rose & Miernyk, 1989).
Walter Isard (1919-2010), not to be confused with Achille-Nicolas Isnard (1749-1803), was the guy that sort of invented the field of regional sciences. He worked with Leontief at Harvard on interregional models. He published quite a bit with him, especially on the idea of how to look at I-O on a smaller scale than national models. Also, he was really into looking at regions as they relate to an economy, not necessarily tied to political boundaries (Isard, 1951)
Karen Rosel Polenske (1937-) was educated at Harvard and did her dissertation with Leontief using I-O. She then left Harvard to take a job at the technical school down the street. That’s what my old boss called his alma mater MIT. She was the first to incorporate gravity model estimations for trade flows of commodities, which led her to construct the first MRIO tables. While she didn’t create the first ideas around interregional I-O tables, she was the first one to actually do it in the US in 1970. She split the US into over 44 regions and 78 industries. We will come to find out later that the Japanese did it in 1960 and 1965. Polenske even starts to talk about including environmental externalities in I-O (Polenske, 1970; Polenske & Rockler, 2014). She’s kinda cool; especially when she said things like “In the context of US regional modeling, a multiregional county-based modeling system from the Minnesota IMPLAN Group is now widely used for regional impact analysis” (Polenske & Rockler, 2014).
Lastly on our trip through the (brief) history of I-O, we have Dr. Jennifer Thorvaldson. She is currently serving as the Chief Economist with IMPLAN, a position she has held since 2013, after joining the company in 2010. She studied at Colorado State University and is completely in charge of the data creation process for the company while continuing to innovate in the field of I-O through additions to the model like environmental and occupational satellite accounts (Thorvaldson & Squibb, 2017; Thorvaldson & Olson. 2014).
Abraham-Frois, G. & Lendjel, E. (2006). Father Potron’s Early Contributions to Input–Output Analysis. Economic Systems Research, 18 (4), 357-372.
Batey, P.W.J., Madden, M. & Weeks, M.J. (1987). Household income and expenditure in extended input-output models, Journal of Regional Science, 27, pp. 341-56.
Bjerkholt, O. & Kurz, H.D. (2006). Introduction: the History of Input–Output Analysis, Leontief’s Path and Alternative Tracks. Economic Systems Research, 18 (4), 331-333.
Bollard, A. (2020). Economists at War: How a Handful of Economists Helped Win and Lose the World Wars. New York: Oxford University Press.
Golladay, F. & Haveman, R. (1976). Regional and distributional effects of a negative income tax, American Economic Review, 66, pp. 629-41.
Isnard, A. (1781). Traite Des Richesses. France: Kessinger Publishing, LLC.
Isard, W. (1951). Interregional and Regional Input-Output Analysis: A Model of a Space-Economy. The Review of Economics and Statistics, 33 (4), 318-328.
Kurz, H.D. & Salvadori, N. (2006). Input–Output Analysis from a Wider Perspective: a Comparison of the Early Works of Leontief and Sraffa. Economic Systems Research, 18 (4), 373-390.
Kurz, H.D. & Salvadori, N. (2010). 'Classical' Roots of Input-Output Analysis: A Short Account of its Long Prehistory. Economic Systems Research, 12 (2), 153-179.
Kurz, H.D. (2011). Who is Going to Kiss Sleeping Beauty? On the ‘Classical’ Analytical Origins and Perspectives of Input–Output Analysis. Review of Political Economy, 23 (1), 25-47.
Kurz, H.D. (2021, March 8). Historical Roots and Theoretical Background of Input-Output Analysis. Retrieved from https://www.iioa.org/isioa/online_module_1.html
Nobel Prize. (18 October 1973). Press Release: The Prize In Economic Sciences In Memory Of Alfred Nobel To The Father Of Input-Output Analysis. Retrieved from https://www.nobelprize.org/prizes/economic-sciences/1973/press-release/
Oosterhaven, J. (1983) Evaluating land reclamation plans for Northern Friesland, Papers of the Regional Science Association, 52, pp. 25-137
Polenske, K. (1970). An Empirical Test of Interregional Input-Output Models: Estimation of 1963 Japanese Production. The American Economic Review, 60 (2), 76-82.
Polenske, K.R. & Rockler, N.O. (2014). Ideal or Not Ideal Interregional Input– Output Accounts and Model. International Regional Science Review, 37 (1), 66-77.
Remak, R. (1929). Kann die Volkswirtschaftslehre eine exakte Wissenschaft werden? (Can economics become an exact science?) Zeitschrift Jahrbücher für Nationalökonomie und Statistik (Journal Yearbooks for Economics and Statistics), 131, (1), 703. DOI: 10.1515/jbnst-1929-0148
Rose, A. & Miernyk, W. (1989). Input–Output Analysis: The First Fifty Years. Economic Systems Research, 1 (2), 229-272.
Science Direct. (2021). Physiocracy. Retrieved from https://www.sciencedirect.com/topics/economics-econometrics-and-finance/physiocracy
Thorvaldson, J. and J. Squibb III. 2017. An Expanded Look into the Role of Economic Diversity on Unemployment. Journal of Regional Analysis and Policy, 47(2):137-153.
Thorvaldson, J. and D. Olson. 2014. Incorporating New U.S. and Canadian Transport Cost Data into IMPLAN’s Trade Flow Model. 2014 MCRSA Conference Proceedings: Madison, WI: June 3-5.
Written August 30, 2023