This month's featured article
Adapting to Globalization's "Second Unbundling"
Richard BaldwinPolicy Director, CEPR
Professor of International Economics, Graduate Institute of International Studies, Geneva
November 26, 2007: CEPR-RIETI International Seminar - Summary
Presentation by Professor Richard Baldwin
In recent years, globalization has taken on new characteristics, different from the past. The old characteristics of globalization can be thought of as the "first unbundling," which is defined as the separation of production and consumption that was primarily caused by a rapid decrease in trade costs for goods. The first unbundling, seen roughly from 1870 to the present, is the experience that motivated the traditional view of globalization, what might be called the "old paradigm" (basically standard trade theories such as Ricardian and Heckscher-Ohlin models). The second unbundling, observed since around 1985 is defined as the unbundling of factories and offices, which means that not only goods but also tasks are traded; to understand this, it is useful to introduce the "new paradigm." The principal cause of the second unbundling is a drastic fall in trade costs for goods, people, and especially for ideas. The trade costs related to people have fallen the least, which is why the second unbundling (also called fragmentation) has happened mainly on a regional rather than global basis. Although transportation costs are much lower than before, the time of managers and/or skilled workers is even more valuable. This high cost of moving people has meant that the second unbundling has a very strong regional dimension. For example, Japanese manufacturers began offshoring labor-intensive production stages to nearby East Asian nations around 1985. Despite Japan's increase in offshoring from 1987-1997, aggregate industrial employment in Japan did not fall, yet in the U.S. and EU it was stagnant or falling. This suggests that the offshoring of labor-intensive jobs made Japanese corporations more competitive than U.S. and European firms and helped create jobs in Japan.
Professors Blinder, Grossman, and Rossi-Hansberg have been conducting theoretical studies on the new paradigm, yet there is quite little research based on statistical data. In the case of the old paradigm, sectors, firms, and skill groups are the proper level of analysis. However, analyzing the new paradigm requires statistics on a task level because tasks are traded. Winners and losers can be determined according to tasks since global competition now takes place at the level of tasks, rather than between firms, sectors, or skill groups. The major difference between the old paradigm and the new paradigm is that individuals doing a certain task face direct global competition.
Such a difference has significant policy implications. The first noteworthy point is unpredictability. It is very difficult to predict the winners and losers in the new-paradigm globalization. Without the understanding of "glue," which bundles production into firms and offices, it is not possible to predict which tasks will be offshored and which will not. Second, unlike the previous globalization where trade cost, tariffs, and trade barriers fell gradually, the current globalization can proceed suddenly. For example, the finance department of a firm, instead of e-mailing the reports from its office building, can e-mail them from another building (outsource a task) or another country (offshore a task). These changes happen quite quickly. Third, the new-paradigm globalization addresses the need of taking individuality into account. Previously, all workers in the factory in the declining industry were suffering and could take collective action, but currently, it is not everybody in the factory who loses a job but instead the individual performing a certain task. Therefore, policies for individuals, not for sectors, firms, or regions, are needed. But given the unpredictability, policy-makers should be more cautious when they try to identify between winners and losers. Since information society jobs are likely to be subject to dramatic changes in trade cost, they have a high possibility of being offshored. Van Welsum and Reif (2005) and Van Welsum and Vickory (2006) classify "offshore-able" jobs as those characterized by four features:
2. output that is IT-transmittable
3. tasks that are codifiable
4. tasks that require little face-to-face interaction
This suggests that promotion of flexibility and continuing education (learning to learn) is needed in considering policies.
In sum, the gains from trade for a newly tradable task are exactly the same as for trade in goods. But the pains from trade may be less correlated with the skill level, job experience, and sector than they were in the past. Accordingly, social policy is even more necessary politically. For example, in order to maintain political support for continued immigration, it is necessary to change the support system so that workers realize that the pains and gains will be shared more widely. Protection of workers, not employment, is increasingly important. Policies to facilitate acceptance of globalization, ease the pains of adjustment, and exploit new opportunities will be needed at the worker level.
After the presentation by Professor Baldwin, participants discussed the impact on policy-making. First, regarding the predictability mentioned, Professor Tomiura asked if collecting data on task-level information would help alleviate this problem. Professor Baldwin responded that as part of the labor market statistics, the U.S. and some European countries have gathered considerable data including a very fine level of job descriptions, which enables understanding of tasks workers are performing by sector. But the data connected with firms are not yet sufficient and need to be further developed. Professor Wakasugi asked what educational system is needed for responding to the increasing trade in tasks. Professor Baldwin was cautious in proposing a specific education policy as he is not an education expert, but he suggested that it would be necessary to develop a system which helps redeploy assets in response to shocks caused by international competition. Offshoring is often associated with reduction in employment within a country, but it does not necessarily mean that low-skilled jobs shift from high-wage countries to low-wage countries. The reverse is also occurring. According to the analysis on service offshoring by Amiti and Wei (2005), the U.S. is a net exporter of business service to India. Thus, it would be worthwhile thinking about what kind of jobs you can attract and trying to get the economy ready to receive tasks that may be offshored from other countries to Japan.
Offshore Outsourcing and R&D: New Evidence of Japanese Manufacturing Firms
Presentation by Professor Wakasugi and Professor Tomiura
Trade costs have declined markedly since the 1980s, owing to innovation in information technology (IT). As a consequence, international fragmentation has increased rapidly. Attempts to measure corporate offshoring operations, such as offshore procurement, outsourcing, and offshore research and development (R&D) activities, have hitherto relied primarily on input-output tables and sector-level data on trade in parts and components. The RIETI Study Group led by Professors Wakasugi and Tomiura conducted a questionnaire survey that was the first research on offshoring to analyze at the task level. The questionnaire was sent to approximately 14,000 manufacturers in Japan in January 2007; nearly 40% responded. (Refer to the RIETI Discussion Paper by Ito, Tomiura and Wakasugi, DP-07-E-060, for a summary of the research and its findings.) In this survey, they defined contracting-out to suppliers overseas with a contract specifying tasks or providing specifications as "offshore outsourcing." They further divided the offshore outsourcing into three types for the purposes of this research: intra-firm transactions with the firm's own subsidiaries, arm's length with subsidiaries of other Japanese firms, and arm's length with foreign-owned firms. They classified outsourced tasks in the production category into the production of jigs or dies, production of parts, components or other intermediates, and the final assembly or processing of final products. They grouped tasks in the services category into R&D, information services, customer support, and legal, accounting, and financial services. For the study, they also divided outsourcing destinations into regions, including China, ASEAN countries, other Asian countries, and the United States and Europe. Since the survey was conducted at a single point in time, they included retrospective questions about the respondent's offshoring practices five years earlier, to identify any changes over time. They prepared separate, detailed survey items for R&D activities, based on the view that, in many cases, the specifications and tasks for such activities are not fixed in advance in contracts. Survey items included questions about the motivations of R&D, types of R&D facilities, factors behind the choice of sites, and relations with the headquarters, and operating regions. Finally, they asked each respondent to evaluate the enforcement of intellectual property rights (IPR) in foreign countries on a scale of one to five, based on the assumption that the level of IPR protection overseas is one factor that influences corporate offshoring decisions. As a result, they were able to develop an evaluation index for IPR systems that covers 56 countries.
The first thing they learned from their survey findings on outsourcing was that only a limited number (21%) of firms engage in offshore outsourcing. They also found that virtually no firm engages in offshore outsourcing without outsourcing their tasks to firms in Japan. According to answers given to our questions on the current state and the situation five years ago, the number of firms that engage in offshore outsourcing has increased over the five-year period. The breakdown of offshore outsourcing tasks among survey respondents showed more than 70% of such tasks were related to production (production of jigs or dies, production of parts, components or other intermediates, and final assembly or processing of final products), though this finding reflects partly to the fact that the survey was conducted exclusively on manufacturers. By region, China accounted for more than 50% of offshore outsourcing. When figures for ASEAN counties are added to the China figure, nearly 80% of offshore outsourcing was concentrated in East Asia. By outsourcee, offshore subsidiaries in intra-firm transactions accounted for approximately 40% of offshore outsourcing. Asked about important factors in offshore outsourcing, many firms cited low cost for outsourcing of production tasks and deregulation for the outsourcing of professional services. The survey findings also suggested that traditional ties with Japanese suppliers affect decisions on outsourcing to subsidiaries and Japanese firms overseas, and that acquiring information about prospective outsourcees is important for outsourcing to foreign firms.
Only a very small number of firms, just 209 (or about 4% of all respondents), indicated they conduct offshore R&D. Approximately two-thirds of these firms undertook R&D primarily within their own plant sites overseas. Approximately 70% of respondents conducting offshore R&D took an integrated approach to managing the activities together with R&D done at headquarters. The percentage of respondents giving this answer was somewhat higher among those operating in China and ASEAN countries. Asked about factors for selecting locations, more than half of the survey respondents cited accessibility to local markets as the primary factor. The second largest group of respondents mentioned agglomeration of local firms and research institutes. The number of firms citing preferential tax treatment was extremely small. Asked about the motivation for undertaking R&D offshore, the highest percentage of survey respondents, 40%, picked support for local production and sales from the multiple choice answers provided. By region, many firms carrying out R&D in the U.S. and Europe cited collaboration with local firms and research institutions as their motive. Meanwhile, many respondents undertaking such activities in Asia chose low R&D costs as their answer. As for the relationship between offshore outsourcing and offshore R&D, survey respondents performing offshore R&D were more likely to use offshore outsourcing as well. This finding suggested a complementary relationship between the offshoring operations. Finally, an evaluation of foreign IPR systems by firms in the survey showed that the actual legal enforcement is weaker than the level of enacted legal provisions in many developing countries.
Based on the survey findings reported above, the Japanese researchers discussed future research themes and other issues with Professor Baldwin, who noted the possibility of a close relationship between offshoring and corporate size. They replied that answers varied widely among respondents in this survey, which primarily covered relatively large firms, and that the size effect on offshoring decisions may well exist even among large firms. Professor Baldwin also suggested the potential for research findings to differ markedly among respondents in an identical survey conducted on service providers. He also pointed out that the way in which offshoring relates to the evaluation of IPR protection as a national attribute that affects offshoring decisions is another interesting area for analysis. In the closing stages of the discussion, Professor Baldwin and the Japanese researchers exchanged opinions with participants about the future outlook for offshoring. The discussion closed with a shared awareness that continuing in-depth research of this type down to the task level is important for planning policies, and that cumulative analysis based on further research of this type is needed.
Fellow titles and links in the text are as of the date of publication.
For questions or comments regarding RIETI Report, please contact the editor.*If the "Send by mailer" button does not work, please copy the address into your email "send to" field and connect the prefix and the suffix of the address with an "@", sending it normally.
RIETI Report is published bi-weekly.