Current State of Productivity Analysis: A look at the results of the Seventh World KLEMS Conference

MIYAGAWA Tsutomu
Faculty Fellow, RIETI

Overview of the KLEMS Conference

In October 2022, the Seventh World KLEMS Conference was held at the University of Manchester in the United Kingdom. The conference was mostly held in person because of the prevailing consensus in the United States and Europe that the COVID-19 pandemic has subsided. KLEMS stands for capital (K), labor (L), energy (E), materials (intermediate inputs) other than energy and service (M) and service (S). A database created in order to measure productivity based on data concerning these inputs of production as well as outputs and value added is known as a KLEMS database. The JIP Database, which is provided by the Research Institute of Economy, Trade and Industry, is the Japanese version of the KLEMS database.

The World KLEMS Conference is an opportunity for researchers preparing and analyzing national and regional KLEMS databases to get together and make presentations. The first conference was held in 2010 at Harvard University, the alma mater of Professor Dale Jorgenson, the creator of the original KLEMS database. Since then, the conference has been held every two years in principle, with the venue alternating between the United States, Europe, and other regions, including Asia. The previous conference was scheduled to be held in Washington, D.C., in 2020 but was postponed and held in March 2021 in an online format due to the expansion of the COVID-19 pandemic. As a result, the latest conference was the first in-person gathering to be held after an extended hiatus. For people who were unable to attend the conference for reasons such as health problems or lack of visa availability, online participation was permitted for some sessions.

Memorial Session for Professor Jorgenson

Usually, the World KLEMS Conference is a two-day event, but this time, a special session was held on the day before the opening of the main conference. This session commemorated the achievements of Professor Dale W. Jorgenson of Harvard University, who passed away in June 2022. The news of his death came as a shock for me all the more because I had taken part in the planning of the latest conference and had been looking forward to seeing him in October for the first time in a long time.

At the memorial session, first, Professor Bart van Ark of the University of Manchester, who hosted the conference, delivered a speech praising Professor Jorgenson’s achievements, followed by eulogies by researchers representing various regions. As the representative of the Asia region, Professor Kyoji Fukao of the Institute of Economic Research, Hitotsubashi University, expressed his appreciation for Professor Jorgenson having supported the development of the KLEMS databases in Asian countries such as China, South Korea and India as well as Japan’s JIP Database.

It was back in the early 2000s, when the first version of the JIP Database was completed, that the opportunity came for us to cultivate a relationship that later contributed to this Japanese database’s presence as part of the global alliance of KLEMS databases. At that time, Professor Jorgenson, who was visiting Japan, invited us to attend the kick-off meeting for the EUKLEMS, held in Helsinki. This occasion started the exchange between researchers preparing KLEMS databases, such as EUKLEMS and Korea Industrial Productivity (KIP) Database, and their Japanese counterparts involved with the JIP Database. Despite his advanced age, Professor Jorgenson never failed to travel all the way from the United States to participate in the Asia KLEMS Conference, held in the interim year between the biennial World KLEMS Conference.

As will be mentioned later, for developed countries, a common theme of productivity decline is starting to emerge. It is very unfortunate for us to have lost Professor Jorgenson at a time like this, but the productivity database to which he dedicated himself is an important legacy, and I would like to honor his achievements by making good use of it. I offer my sincere condolences for his death.

Investment and Productivity

After the memorial session came the two-day conference. As I served as a program committee member, a session chairman, and a debate participant, I was unable to attend all sessions. With this in mind, below, I will focus particularly on investment and productivity as well as the economic impact of COVID-19 in my overview of the state of research by researchers in various countries on themes related to productivity.

“Investment and productivity” was the theme of the keynote speech made by Professor Jonathan Haskel of the Imperial College London. Professor Haskel argued that the ongoing productivity decline in developed countries is attributable mainly to capital accumulation problems, including a lack of investment in intangible assets, which previously served as an engine of productivity growth. In the case of the United Kingdom in particular, the decline in intangible investment has become pronounced since the country’s departure from the EU (Brexit).

A paper written by François Lafond of Oxford University, who made a presentation in the EUKLEMS session, supported Professor Haskel’s argument. However, his presentation focused on the sluggishness of capital accumulation while examining productivity-lowering factors from a broader perspective. Among them are those factors that were discussed in the 2010s, such as measurement error, the decline of business dynamism due to increased market concentration, and the competitive effects of trade. According to his study which determined which of those factors lowered productivity in five developed countries (Japan, the United States, the United Kingdom, France and Germany), the sluggishness of investment was the main factor in the four countries other than France. On the other hand, the decline in business dynamism was the main factor in the United Kingdom and the United States. These points were supported by the presentation by Javier Miranda and others based on work with micro data. However, in Japan’s case, it was reported that despite the absence of a decline in business dynamism, not only was investment sluggish, the country experienced declines in the spillover effects from intangible assets and trade’s effect of stimulating competition and raising productivity in terms of resource allocation. The paper by Lafond et al. is scheduled to be published in the Journal of Economic Literature and is expected to be appreciated as a comprehensive survey of the results of productivity analyses conducted since the survey by Professor Chad Syverson of the University of Chicago.

The impact of intangible assets on productivity, which was examined in Lafond’s paper, was characterized as a major factor of the productivity decline in Japan, but in other countries, it was not considered to be a significant factor. That was corroborated by the presentation by Professor Van Ark of the University of Manchester. Namely, in response to the argument that the impact of intangible assets may also have declined amid the sluggishness of investments in general, Professor Van Ark, in his presentation, observed that intangible asset investment, including investment in R&D and branding, has maintained a productivity-raising effect even after the global financial crisis.

Economic Impact of the COVID-19 Pandemic

At the latest conference, four presentations were made with respect to how the COVID-19 pandemic has impacted the economy. Presentations concerning COVID-19 were also made at the previous World KLEMS Conference, held in March 2021. Compared with the 2021 conference, the researchers making presentations during this session included both researchers from developed countries such as the United States and European countries and also from Asian countries such as India and South Korea. Additionally, analyses using industry-by-industry data were provided on this occasion.

In the South Korean presentation, it was reported, based on the estimation of seasonal GDP data adjusted for the operation rate, that the main factor behind the decline in GDP during the pandemic has been a fall in capital accumulation and total factor productivity (TFP). Meanwhile, based on the Indian presentation, their industries were divided into two groups—those that depend on face-to-face contact (contact-dependent industries) and those that do not (contactless industries)—and it was shown, based on simulations, that contactless industries with significant environmental footprints (e.g., primary metals, transportation machinery, electric power, and gas) sustained smaller injury, but that those industries were also affected by contact-dependent industries which were more seriously affected, through the process of transactions.

There were two presentations by researchers from the United States. One of them dealt with the theme of how to treat the massive subsidies provided during the pandemic when measuring productivity. The main point of the report was that while GDP includes the value of taxes on production and trade minus subsidies, the corresponding cost is excluded on the expenditure side, which represents an asymmetric calculation method that must be corrected.

Lastly, Matthew Russell of the Bureau of Labor Statistics (BLS), alone among the researchers who made presentations at the latest conference, prepared a KLEMS-type database for the period until 2020 and, based on it, compared industry-specific characteristics observed during the pandemic period with the characteristics observed at the time of the global financial crisis (2007-2009). According to the analysis by Mr. Russell and his colleagues, a broader range of industries suffered damage during the current pandemic than at the time of the global financial crisis, and as a result, overall economic activity declined more steeply. In addition, in contact-dependent industries, such as transportation and accommodation, significant supply constraint occurred as many laid-off workers did not return to the labor market during the phase of recovery from the pandemic. Regarding the low labor participation ratio, which was also pointed out in Professor Haskel’s speech, Mr. Russell did not present any causal findings. On the other hand, regarding capital investment, the amount of investment in the United States apparently has not fallen as steeply as in Japan, thanks mainly to investment in digitalization, possibly because of the effects of the spread of remote work.

Future Outlook

Finally, I would like to mention the future direction of analysis using the KLEMS database. With respect to the impact of COVID-19 since 2020, the usefulness of industry-by-industry databases has been recognized anew in that the intensity of the damage received varies across industries depending on the characteristics of goods and services. Therefore, if data covering the period until 2022 is released, when most restrictions on activity due to infection risk were lifted, more detailed analysis of the economic impact is expected to be conducted.

On the other hand, as the recovery from the pandemic has advanced, supply constraints regarding inputs of production factors, such as labor and raw materials, have arisen, resulting in accelerating inflation in various countries. While attention is focusing on the macroeconomic inflationary trend at present, changes in industrial structures are presumed to be occurring as a result of differences in the input structure between industries. At the next World KLEMS Conference, the impact of inflation due to supply constraints is likely to be a topic of discussion.

While the latest conference was meant, in a sense, as a eulogy for Professor Jorgenson, as the founder of World KLEMS, I also had the sense that expertise on productivity analysis is being passed down from generation to generation, as I saw young researchers make ambitious presentations at many sessions. Since the beginning of the 21st century, major economic and social shocks have occurred. I hope that younger generations will explore new perspectives of analysis and achieve results that enable us to better prepare for future economic shocks.

December 6, 2022

January 10, 2023