The First Asia KLEMS Conference "Productivity Growth in Asia"

World KLEMS and Asia KLEMS (Summary)

Speaker Dale W. JORGENSON
Samuel W. Morris University Professor, Harvard University
Materials Presentation [PDF:1.7MB]


Prof. Dale W. JORGENSON's PhotoProfessor Dale JORGENSON

As previously mentioned by Dean M. Kawai, the focus of economic growth has shifted to Asia. In The Rise of Developing Asia and the New Economic Order, a paper I have written with Professor K. M. Vu from the National University of Singapore, the relative ranking of different countries around the world shows that the rise of Asia is upon us. This is a very auspicious time to consider these issues on the occasion of the first Asia KLEMS conference.

The Asian Development Bank Institute (ADBI) has played a very important role in research on Asian economies and its sponsorship of the conference is a great step for KLEMS-type research around the world. Also, as a co-sponsor, the Research Institute of Economy, Trade and Industry (RIETI) has a long history of research in productivity in Asia going back to the beginning of the institute. Productivity and sources of growth are the central focus of RIETI.

RIETI is the most significant think tank in Asia dealing with economic issues. For example, it published Productivity in Asia: Economic Growth and Competitiveness in 2007. This involved productivity comparisons, mainly comparisons of sources of economic growth in major Asian countries. RIETI is also sponsoring the construction of the Japan Industrial Productivity Database (JIP).

The Hi-Stat Project at Hitotsubashi University similarly has a long history of research on economic growth, going back to the famous studies of the 1950s of the growth rate of the Japanese economy. At that time, Japan was the most rapidly growing economy in Asia. The period of income doubling created a template later followed by the Asian tigers (Korea, Taiwan, Singapore and Hong Kong).

For Developing Asia as well, beginning in India in 1991 and in China in 1978, economic development was patterned directly after the Japanese growth strategies of the 1950s, 1960s and early 1970s and the successful implementation of this strategy by the Asian tigers. Needless to say, growth in Singapore with 5 million people involves different issues than in India or China. Nonetheless, this history helps in understanding the emerging Asian model that informs all the research today.

The relationship between World KLEMS and Asia KLEMS also has a very long history. I think it is fair to say that the basic methodology was crystalized in Information Technology and the American Growth Resurgence, a book I co-authored with Mun S. Ho and Kevin J. Stiroh in 2005. This book was a result of work done on the impact of information technology (IT) on U.S. economic growth during the 1990s and early 2000s. Our book laid out a detailed methodology that has become a roadmap for subsequent research in this area.

OECD report on standards for measuring productivity

The first of the challenges I will identify is to assemble the data on economic growth. Especially in Japan, but also in Korea and the other Asian tigers—there has been a long tradition of research on sources of growth in Asia. Around 2000 this work took a critical turn toward international comparability that is central to the focus of the World KLEMS Project.

In 2000 the Organisation for Economic Co-operation and Development (OECD), which involves amongst others the United States, Japan and Korea, convened a group of distinguished economists and statisticians to formulate a set of standards for productivity measurement around the world. This was an era of "many flowers bloom" and there were many different ways of measuring productivity and thus no basic standard approach that led to firm conclusions about the sources of economic growth in any country.

International comparisons involving comparisons within Asia and similar comparisons in other regions were also impossible because everybody had their own way of doing things. The OECD proposed to meet this challenge by formulating a set of standards that could be used around the world. This set of standards resulted in the OECD Manual—Measuring Productivity: Measurement of Aggregate and Industry-Level Productivity Growth, which was a committee report.

In the OECD style the report was written by a single individual, namely Paul Schreyer, now the Deputy Director of the Statistical Directorate of the OECD. Schreyer took upon himself in 2001 the task of writing up the consensus of the Statistical Committee. What emerged is a new set of standards for measuring productivity.

What do these standards involve? First and most important, they involved a view of capital that treated capital in terms of a service flow instead of a stock. In other words, they formulated standards for measuring capital that said that it is not enough to find out what the stocks of capital are—which was then the prevailing view up to that point—but that it is important to convert those stocks into a flow of services.

This is an important issue because the service flow from a given capital stock differs enormously among different types of assets, in particular between information technology (IT) assets and other types of assets. The elements that make up the service flow are (1) the rate of return, which is uniform across the different assets; (2) depreciation, which differs enormously from one type of asset to another; and finally (3) the decline in the price of assets, or the negative of the rate of increase of the price of assets. These three elements make up the components of the service flow. Although the idea had been present in economics going back into the 19th century history of capital theory, this concrete approach of capital services was featured in the publication by the OECD committee and Paul Schreyer's manual.

The year 2001 was already very late in the history of productivity measurement which had been very well developed much earlier in the 1950s and 1960s. However, the OECD report was a real watershed in defining the role of capital and made it possible to begin to focus sources of growth research on the impact of capital accumulation and how exactly that affects economic growth. This had been vague and imprecise until the OECD took a stand about how capital impacts economic growth.

So the history of the subject started anew in 2001 and that led to the challenge then to implement these ideas for many countries. This challenge was taken up in the United States, Japan, Korea and people in Europe involved later in the European Union's EU KLEMS Project, and the OECD then began to organize meetings to deliberate about these issues of capital measurement, but at a very fundamental level.

Research on economic growth at industry level

What were the other elements in the OECD report? These included the importance of having consistency between measurement of productivity and the sources of growth at the industry level and at the aggregate level. Many of the consequences of research on economic growth are focused on issues at the aggregate level, for example the issue of the role of capital formation. That is something that starts from the top down, looking at the overall growth of the economy, the role of capital, the role of labor, the involvement of quality and so on, and all these things are then formulated at the aggregate level.

However, most of the detail about the sources of growth has a very specific industry dimension; for example, the dimension related to the production and use of information technology equipment and software. That is obviously something that is very specific to a group of industries. Thus, the OECD took upon itself the task of reconciling aggregate productivity measurements focused on broad policy issues with industry-level productivity measurement that would be consistent and could be integrated into an overall aggregate picture.

At the aggregate level, the sources of economic growth were capital and labor; but at the industry level, industries have inputs beyond the primary factors of production. In particular, they have inputs of energy, materials and services as examples. Furthermore, for most industries, these inputs of intermediate goods like sources of energy, materials and services are dominant at the level of particular industry groups.

The OECD said that the best way to proceed is to think of these intermediate inputs in addition to capital and labor as part of the system of inter-industry accounts. This made it possible to draw on decades of research around the world in Japan, Korea, Taiwan, Canada, the United States, Australia, and in many European countries going back to 1947 or even earlier. The results from this research work could be used by the OECD, provided that input-output data could be assembled in a form of consistent time series in both current and constant prices.

Why consistent time series? It is necessary in order to look at the growth of the intermediate inputs to have the same kind of consistent time series in current and constant prices that would be used at the aggregate level to measure capital and labor services. While it was a radical idea at the time, researchers had already devoted a good deal of attention to it and it became one of the central themes in the development of the JIP Database in Japan, and subsequently in Korea and work in Taiwan.

Fortunately, the OECD was able to agree that there should be a framework that incorporated the intermediate goods. This put extremely difficult demands on the government statisticians, because they were used to assembling input-output tables and liked to have as much detail as possible, but were not used to the idea of making these tables consistent over time. Therefore, the OECD posed a huge challenge to countries that were already developing input-output tables to translate those into a form that could be used in the sources of economic growth analysis at the industry level.

Labor input quality measurement

The third accomplishment of the 2001 effort was the idea of incorporating quality dimensions into the measurement of labor input. The great tradition of research in productivity analysis is based on the idea of measuring employment or hours worked, and this initiative of incorporating quality dimensions into a labor input played a very important role in labor economics where there was a whole human capital school that developed in understanding relative rewards in different kinds of labor.

However, the human capital school of research on labor input never got to the point of thinking about labor input along with capital and intermediate goods as input in the production. Thus, the OECD took the step of saying that labor had to be thought of as very heterogeneous and that therefore it is necessary to aggregate over different kinds of labor and to reflect in weights of the aggregation process the relative quality.

How should this quality be measured? This is where the human capital school had a great deal to contribute because in the famous Mincer equation, formulated by Jacob Mincer, the focus of the model was on the reward to human capital as measured by labor compensation, which includes the wages earned, supplements paid in the form of pension, health benefits, etc., and for many emerging economies, the rewards to labor that are earned as part of self-employment and informal activities. In the developments that followed where the technology began to be transferred toward emerging economies, the issue of essentially using these concepts of human capital for informal activities immediately came to the fore, and that was part of the OECD program.

Thus, capital, labor, energy, materials and services (KLEMS) were all formulated explicitly as part of an international consensus among economic statisticians in a committee convened by the OECD resulting in a report that established a handbook for carrying out productivity research. This handbook embodied ideas that were already present in earlier work and databases around the world, but by establishing a set of common standards, the OECD challenged each of the research groups that had already been active in this area to bring their own measurement standards into conformity with the international standards.

International standards for measurement are very well established in the international statistical community. For many years going back to the middle of the last century, the OECD and other international organizations like the International Monetary Fund (IMF), World Bank and the Statistical Office of the European Commission (Eurostat) have been working to develop a System of National Accounts (SNA) which is standard around the world. Although this has not focused on issues having to do with productivity measurement, it illustrates the idea that the international community of statisticians is familiar with the idea of standardization so as to adapt methods and concepts to the local situation.

Measurements are not identical in different countries, as I learned in research comparing Japan and the United States. It is necessary to drill down and deal with the peculiarities of each of these relatively advanced economies. This holds much more forcefully when thinking about moving from industrialized economies with well developed statistical systems to emerging economies. That is why the role of international standards comes to play a very significant role in this whole area.

Revision of the SNA

After the OECD effort was completed, an effort was launched to revise the SNA. This was of great interest to the statistical community but did not receive any headlines in the Wall Street Journal, Nikkei or Financial Times. It was nonetheless a very important event because the SNA involving the United Nations, World Bank, IMF includes not only advanced industrialized economies like the OECD economies but all of the economies of the world, down to the smallest island nations. The UN assembled more than a hundred countries in an effort to work out a new set of standards that would be appropriate for the new environment in the new century. This effort lasted for perhaps a decade and a momentous development took place in a very unusual location. That location was Canberra, the capital of Australia.

What does Canberra have to do with this? A number of committees were appointed by the SNA to consider issues related to the revision of the SNA and these were named after the cities where they were convened. Hence, a committee was convened in Canberra consisting of about 150 people and most of the meetings were held in Canberra with the purpose of deliberating about issues having to do with the measurement of capital.

Many days, weeks, months and years later, the Canberra Group emerged with its recommendations. These were to measure capital in a way that can be integrated with accounts for the sources of economic growth and to make this an integral part of the SNA. The OECD had already established standards, so if anybody wanted to find out how to measure productivity in an OECD country, they could refer to the OECD Manual published in 2001. By moving this to an agenda involving the standardization of methods and measurement practices around the world, the United Nations encompassed all of the issues that arise in dealing with the emerging economies. This was very ambitious.

How could this be accomplished? First, this had to be sold to the management. The way the SNA is organized is that there are various city groups that report to a central commission that passes (or fails to pass) particular recommendations. In March 2007, the United Nations Statistical Commission, the governing body for the revision of the SNA, endorsed the report of the Canberra Group, saying that it was possible to standardize for every country in the world the measurement of capital according to the OECD methodology as part of a world data system, the SNA.

This did not hit the headlines either. It escaped the attention of the financial press, but it was a momentous development, because it meant that the same standards could be applied to India, Myanmar, Japan and the United States as well as every country around the world. And these standards were precisely the ones that had been agreed upon earlier by the OECD which represented a completely different constituency. The people who were deliberating about this were not ignorant of the fact that for many countries this would be an extremely demanding undertaking, which is precisely why this is such a significant development.

As the process was finalized, the OECD and the other organizations that developed the SNA produced a two-volume document called the 2008 SNA, the 2008 revision of the System of National Accounts. In chapter 20 of volume 2, there was a complete description of how to measure capital within the framework of the SNA. In 1993, only 15 years earlier, a similar UN panel had said that it was impossible to measure capital at all and therefore progress in understanding the sources of economic growth during the most dynamic period of economic growth in world history was halted for 15 years as a result of a set of decisions by a group of economic statisticians.

This illustrates the great power held by the statisticians in halting the flow of information that would have been essential to understanding world developments at such a critical time. The Asian crisis passed, the recovery occurred, and the economic and financial crisis went forward, all in total ignorance of the role of capital in a form that could be compared across countries. Fortunately, the authors of the 2008 SNA took this as a challenge and resolved the issue in favor of the OECD methodology. The OECD followed up by writing another handbook on the measurement of capital, spelling out exactly how this is to be done: how capital services are defined, how it is related to capital stock, the role of the deflators, the role of depreciation, the role of the rate of return, and so on. And all of this was summarized in chapter 20 of volume 2 of the SNA, an official document involving the endorsement of the UN Statistical Commission and literally more than a hundred nations around the world.

World KLEMS Consortium

This set the stage for World KLEMS. So what is World KLEMS about? What is the purpose of World KLEMS? The purpose is nothing more or less than to try to implement these ideas, not only in advanced countries in the OECD, but around the world including the emerging economies of Asia. On August 19-20, 2010, one year after the publication of the 2008 revision of the SNA, a meeting took place at Harvard University to form a World KLEMS Consortium to implement the OECD approach for countries around the world.

At the First World KLEMS Conference 40 countries were represented. In addition to the members of the OECD which had been involved in the earlier research, there were representatives from India, China, Latin America (including Chile, Brazil, Argentina and Mexico), Turkey, Taiwan and Indonesia, among altogether 40 countries making up more than 90% of world gross domestic product (GDP).

It should be noted that the representatives at this meeting included not only advanced economies with well developed statistical systems but emerging economies. The World KLEMS Consortium took on the challenge to developing methods for compiling and analyzing data in a form that could be used to understand the sources of economic growth in countries at every level around the world. The Consortium focused on those that are the most significant economically, but ultimately with the goal to encompass the world economy in as much detail as possible.

The World KLEMS Consortium was put forward as an approach to implementing the ideas that had been first developed in the OECD report, then endorsed by the UN SNA and finally incorporated in the 2008 revision of the SNA. The OECD capital measurement manual was a very important document as well. All of these efforts succeeded in establishing a common methodology and a common set of standards that could be applied in the same way as national accounting had been applied in a standard format around the world 40-50 years earlier.

This was the setting for the formation of the World KLEMS Consortium, which greatly broadens the community of people who are involved. It is no longer a matter of simply compiling or collecting the data. Data can be collected on the population, economic units or informal sector, but there may not be an accounting system in place. To find out what is needed is a matter of research in which economic statisticians must collaborate with people who specialize in economic measurement within the community of research economists.

Application of the KLEMS-type methodology

The idea was to turn this into a practical reality—to take a set of abstract ideas with a set of methodological dicta and implement it at the operating level where an economic statistician has to produce a report every month, quarter or year. It would obviously require an enormous research effort to adapt these ideas to the conditions, for example, in China or India. Both of these countries are very promising alternatives because they are among the economies with the best developed statistical systems. Therefore, although they are not ideally adapted to the application of the KLEMS-type methodology, they contain all of the elements.

What are these elements? They are capital, labor and intermediate goods in the form of input-output tables. Input-output research is very well established in China. It is something that goes back to the era of a planned economy and has developed since then through the earnest efforts of many research scholars to be implemented within the framework of the Chinese system of national accounts. There are official input-output tables which can be compared over time and form the basis for a KLEMS-type dataset.

The situation is even better in India, which has a long tradition of empirical research in economics with one the best developed systems of national accounts in the world, a very elaborate system of population statistics, as well as household and economic survey statistics. Although this effort is quite ambitious to begin with, it is in fact feasible. However, it takes talent and it takes the kind of people, for example, who have assembled to launch the Asia KLEMS Project.

Role of statisticians

To take these ideas of the World KLEMS and that have been endorsed by the international statistical community and to turn them into reality and economic research that will push this down to the level of international comparable statistics around the world, it is going to take a mode of organization which is extremely well established. In the United States and Japan and all of the countries that are participating in the Asia KLEMS Project, there are established relationships between economic statisticians and research economists to integrate and use common standards to adapt these abstract ideas having their origins in economic measurement to the specific conditions that are confronted by countries at every level of development.

To meet this challenge of compiling data, it is necessary to involve academic researchers and researchers of think tanks along with economic statisticians. These are the people who collect, compile and report the data. The job will not be finished until the research community has completed the task of providing something that can be reported on a regular basis as part of the national accounts.

The setting is as follows: In 1993, the UN said that it was impossible to report measures of capital as part of the national accounts. In 2007, thanks to the work of the Canberra Group and the OECD, the UN changed completely and took the opposite point of view that it is possible in fact to do productivity measurement using the KLEMS model within the SNA. This is important because it is not mainly a story about research; it is mainly a story about measurement, and measurements must be reported on a regular basis.

And if data are reported on a regular basis, the standards must be the same. If the results are going to be understood by anybody outside the country making the report, they must conform to the international standards of the SNA. Never was there a better time to undertake a project like the World KLEMS Consortium as ambitious as it sounds, involving 40 countries around the world, than now. Fortunately, at the World KLEMS meeting that took place in August 2010, the participants grasped the opportunity and took upon themselves the role of seeing to it that there would be a thoroughgoing implementation of these very basic ideas in economic measurement that would make it possible to begin to understand the momentous changes that are taking place as a result of growth in the world economy and growth in the Asian region, which is becoming central to growth in the world economy and growth in the most important countries of Asia.

Asia KLEMS conference

That sets the stage then for today's meeting, mainly beginning to think of how all this can be done in Asia. Why is that important? First, economic growth and the focus of growth in the world economy is shifting very rapidly to Asia where all the action is taking place and where the understanding is necessary for issues related to the world economy. Second, Asia is a marvelous laboratory for understanding the importance of these measurement issues and how to resolve them in different settings.

Japan is the country the most sophisticated highly decentralized economic statistical system in the world; Korea has an extremely sophisticated system of economic measurement to which enormous resources have been devoted over decades. Paired with these countries and the other Four Tigers, the greatest challenges for economic measurement emerge. These include the challenge of understanding economic growth in China and in India. This is not going to be an easy challenge to meet.

Reading the literature on the sources of growth in China and India, two countries which are so central to the future of the world economy, the same enormous heterogeneity and differences methodologies and approaches that have characterized productivity research around the world until the OECD manifesto of 2001 are seen. With that heterogeneity, of course, it is impossible to draw conclusions, either for an individual country, for the region or for specific groupings of countries.

Without standardization, it is impossible to do measurement. That is the situation that economics finds itself in today. The same applies to measurements in physics or engineering. With the opportunity created by the careful thought of the economic statisticians and being implemented in the World KLEMS Consortium, it really is an opportunity for transformation in economics in a very fundamental way.

The mode of organization is very clear. It is going to be necessary to have the statistical agencies collaborate with the research community in each individual country; that is, the people who have been studying the measurement issues at the ground level where all the specifics and challenges emerge. Another challenge is to identify the informal sector which is essentially "the rest" or the economy that is not capture by conventional measurement methods. This does not mean that statisticians in China or India or other emerging economies have not attempted to deal with the issues, but it does mean that this involves a completely separate set of challenges that must be met as part of this overall undertaking. This is an additional reason why this meeting on Asia KLEMS is so significant.

Role of the regional organizations

That essentially sets the stage for discussing future work and the role of the regional organizations that are now emerging as part of the World KLEMS Consortium, which itself is very recent, dating back only to 2010. It should be noted that the second World KLEMS conference will be held at Harvard University on August 9-10, 2012. Discussion will cover the regional organizations in Asia, the corresponding organization in Latin America, and most importantly the organization in Europe, which has led the way to implementation within Europe.

In terms of the role of the specific regional organizations, after my book with Ho and Stiroh on the U.S. growth resurgence was published in 2005, a group of European economists led by Bart van Ark and others decided that the next step was to try to implement these ideas for Europe. The European economy had been carefully scrutinized in an international conference of all the members of the European Union in Lisbon, Portugal, in 2000. In the Lisbon Agenda, the Europeans formulated the objective that Europe should by 2010 become the world's most dynamic knowledge-based economy.

This goal was originally set in 2000, at a time when the U.S. economy had grown at faster rates than ever before for about 10 years and the European economies flowed relative to their very dramatic growth of the postwar period. They were very specific about what that meant, i.e. the use of information technology and knowledge-based methods for developing technology such as scientific and non-scientific research and development. All of these things were to be developed in Europe at a state that would make Europe comparable and surpass the United States.

At the World KLEMS Consortium meeting in 2010, it was possible to look backward and see what happened. By around 2005, five years after the famous Lisbon Agenda, Europeans began to develop doubts about things not having changed five years into it. The U.S. economy was growing faster than ever and Europe was languishing. What happened to the goal of becoming the world's most dynamic knowledge-based economy?

A series of reports began to emerge from various sources and people in Brussels at the European Union began to realize that they did not know the reasons for the slowdown in Europe and the acceleration of economic growth in the United States. They received a quick lesson from colleagues at the OECD and the prescription was very simple: It is necessary to carry out a KLEMS-type project for Europe in order to understand the differences. It is necessary to understand where the growth comes from before it is possible to talk about a knowledge-based economy.

EU KLEMS project

Thus, beginning in 2005, the Europeans began the implementation of KLEMS for as many European economies as possible. While one might think that this is easy when dealing with advanced economies such as United Kingdom, Germany, France and Italy and so on, there are a lot of economies in Europe that are not so advanced such as the new Central and Eastern European members, countries that had had no contact with the international statistical community until the fall of the Berlin Wall in 1989. These countries were in a very primitive state, statistically speaking.

Nonetheless, 25 EU economies were eventually included in this project which resulted in an initial release on 15 March 2007. This was the first and most successful project so far implementing the KLEMS approach designed by the OECD across a very heterogeneous group of countries in different stages of development, statistical sophistication and research sophistication. Still, this initial release provoked not a storm of controversy but rather a welcoming on the part of statisticians and measurement economists, not only in Europe but around the world.

Even people who were skeptical about the KLEMS approach stated that this was a new era and something that will be the definitive treatment of the European economy for the decades to come. The European countries continued to cooperate and by the middle of 2008, just before the financial crisis, the EU KLEMS Project was completed. This project was reported in a book edited by Marcel Timmer and his colleagues, which was the first international comparison among such a broad range of economies involving 25 of the 27 EU members.

As Professor Pyo remarked, as part of the EU KLEMS Project, the community of people already involved in this type of research in the United States, Canada, Japan and Korea were invited to join as "adjunct members" of the EU KLEMS Project with an idea that if the EU KLEMS Project was successful, it could lead to a broader effort involving other countries. By the time the project was completed, this approach was well known among people in the measurement community not only in Europe, but in the United States, Canada, Japan and Korea and all of these separate efforts began to focus on the same basic methodological approach, mainly implementing the standards agreed upon by the OECD statisticians and finally implemented in the SNA. The book describing the EU KLEMS Project was published in 2010 and volume 2 has gone to press with contributions from members of the EU community who did individual country studies with a full report of a successful implementation of an international comparison project involving KLEMS-type research.

PPP and World I-O database

Exactly what was accomplished with the EU KLEMS Project? It was not merely a comparison of growth of the 25 countries. The ambition was much greater. The people who designed the EU KLEMS Project had in mind not merely the idea to compare growth rates across countries, but also levels. It was necessary to have measures comparing, for example, the level of capital in Italy with the level of capital in France, etc. Therefore, an effort was made to integrate the measurements for the different countries. This is the fundamental concept of purchasing power parity (PPP).

When one begins to compare sources of growth across countries, or outputs across countries, it is necessary to make use of the concept of PPP, a common set of prices that enable the translation of results from different countries into results from other countries. This is worked out in some detail for the United States and Japan and in a lot of detail for the EU economies and is an integral part of the design of Asia KLEMS. The conclusion then is that there are standards in place and an accumulated well of experience now in implementing an approach to KLEMS that makes it possible to standardize measurements across countries, and most importantly to do inter-country comparisons.

This leads to the third and final capstone of this research, namely the world input-output database. The idea is that in addition to making comparisons across countries, it is going to be possible to look at the way in which these countries are integrated through international trade. This will essentially unify the world statistical system and provide a unified picture of the world economy involving not only the individual countries and comparisons across countries, but the very important processes of international coordination that are involved in international trade. This is obviously a very ambitious undertaking and has implications for Asia KLEMS.

Asia KLEMS and the challenge ahead

Asia KLEMS is a project that was launched at the initiative of Professor Pyo who was fortunate to enlist RIETI and ADBI in co-sponsoring this effort with support from Hitotsubashi University based on the basic idea that the EU KLEMS model involving integration of comparison of methods across countries adapted to specific situations could be extended to Asia. The motivation is easily understood. Asia is where the action is. Research in economic growth has to be done on Asia and it is an area which has already been carefully "prospected" by investigators in Japan, Korea and Taiwan in earlier work in productivity. What has been missing so far is the use of common standards. Fortunately, because of the work of the UN, the OECD and of the EU KLEMS Project, these common standards are in place.

What remains to be done? It is necessary to understand how these standards can be applied to the conditions in Asia. This reveals problems of a totally different order and totally different set of challenges than have been seen before. How is it going to be possible to apply these ideas that have been mainly developed and implemented for advanced economies to a region like Asia where there are indeed advanced economies—as advanced as any in the world—along with very large and very important economies that are emerging, that have statistical systems still under development, but are the core of the emerging pattern of economic growth in the world. How is that going to be possible?

Professor Pyo has succeeded in enlisting an outstanding group of collaborators from Japan, India, China, Singapore, Malaysia and so on. This is something that is now beginning to take real form, where confidence is building that the approach that has already been tested so thoroughly over the last decade since the beginning of the new era of productivity measurement can be successfully applied in Asia. Difficult issues will certainly emerge, but groups are already forming that are going to solve these problems. Very well established groups already exist in China and in India, with additional projects being launched in Asia that are going to make it possible to meet these challenges.

Asia KLEMS will have an independent identity from World KLEMS. The problems that are going to emerge are not going to be solved by any kind of top-down dictation of standards and so on from any kind of international organization. Those are already in place and agreed upon, but the implementation is something that has to be done on the ground. That is the purpose of the Asia KLEMS conference.

This is an extremely challenging undertaking and is something that has audience waiting for it. It is not the kind of audience that was waiting for the result of the EU KLEMS Project. In that case, it was a story about a failure; a failure of Europe to keep up with the development of technology which then looked so impressive in the United States. That was the challenge there.

The challenge here in Asia is completely different, namely precisely the reverse. As Asia is the center of gravity with world economic growth moving to Asia, the prospect is not only that China will overtake Japan, which happened last year in terms of exchange rates, but it will also overtake the United States by around 2018 as the world's largest economy. The United States has been the world's largest economy for a hundred years, so this is a historic development, but is not limited to China. India will overtake Japan. This should not be shocking. It is something that is going to happen and it will happen within the next decade.

Developing Asia, excluding Japan, will overtake the famous G-7, the industrialized countries that have led the world since the middle of the last century. This will also happen within the next decade. There will be a new world order emerging as this research for Asia KLEMS is undertaken. Therefore, rather than looking at a divergence in which deficiencies in the EU countries are now becoming increasingly apparent, the project will look at the emergence of the "Asian century," and it is something that economists around the world will be required to study.

There will be many voices that will be part of this debate over the rise of Asia. Goldman Sachs and all investment bankers have their own view on this topic. What role is there for a statisticians and economists in this debate? The numbers that the people are going to be referring to as this debate unfolds are the numbers produced by this group. The same is going to be true around the world as part of the World KLEMS Consortium.

The first Asia KLEMS meeting has something to contribute that is unique. The researchers assembled have the knowledge, standards and the experience that will make it possible for people in the economic statistical community who are reporting the growth rates to extend their reach and build up a new model for measurement of economic growth. This will lead to an understanding of the sources of economic growth that will apply to Asia and around the world as this Asian century unfolds.