Most trade statistics, including those found earlier in this report, represent trade between countries on a “gross” basis. Under this approach, both the exporting country and the importing country simply record the total value of a good that has been traded. Statistics produced using this method are both the easiest to find and the easiest to understand, and they do accurately reflect total imports and exports between nations for many purposes.
However, such statistics do not fully account for global supply chains—in particular, for the way slices of value—goods or services—are added at each step of increasingly international manufacturing processes. These inputs, also known as “intermediates,” may be anything from a circuit board added to a product, to the further processing of a half-assembled car, to the work of a testing service ensuring that a food ingredient meets national standards. It is not uncommon for one intermediate to be added in Country A, a second in Country B, and a third in Country C.
Because goods often cross multiple borders, intermediates may be counted several times when the gross method is used to calculate global trade flows. Analysts have recognized that attributing the entire export value to the last exporting country does not provide people with information on the source of value in global trade. In contrast, “value-added” international trade statistics reflect the value added at each step of the supply chain across national borders. The use of value-added statistics has become more routine for analyzing trade flows, although challenges remain—including a lack of data broken out at an intermediate input level by industry or country, as well as difficulties getting timely data. A brief introduction to the concept and its uses is presented below.
Definition of Value Added and Findings from Its Use in Trade Data
"Value added" can be defined at both the microeconomic and the macroeconomic levels. At the microeconomic level, it is defined as “the amount by which the value of a good . . . increases at a specific step in a production process.” At the macroeconomic level, in the context of measuring exports, it is defined as “the value of national work performed (i.e., the contribution of all national factors of production) in a country’s exports.” Export values calculated using the value-added method differ from official reported export values, which do not distinguish between “national” (domestic) inputs into a product and foreign inputs into that product. For example, one microeconomic analysis attempted to calculate the percentage of contributions from Japan, the United States, and the Republic of Korea (Korea) contained in an Apple iPod music player sold in the United States after being assembled in China. The analysis estimated that at least 82 percent of the factory cost originated in Japan, the United States, and Korea, with China’s contribution estimated at only 4 percent of the value. This breakdown would be reflected in trade statistics that are based on value-added calculations. Gross trade statistics, however, show 100 percent of the iPod’s import value originating in China.
One breakdown of the (macroeconomic) elements that make up a country’s gross exports identifies three types of value added: (1) domestic (national) value added that will never reenter the domestic market, (2) domestic (national) value added that will eventually reenter in the form of imports, and (3) foreign value added that is incorporated into domestic exports. In contrast, a country’s value-added export figures capture only the domestic value added that does not reenter the domestic market. Under this definition, value-added imports are only the part of a country’s imports that is foreign value added. The difference between gross and value-added measures of trade is that only the latter can reflect the actual supply chain processes in which goods cross national boundaries several times during a multinational production process.
Measuring trade in terms of value added leads to several discoveries. First, it reveals export values for some countries, including China, to be smaller than reported, because the value of imported inputs, which is included in official export figures, is now excluded. In addition, it shows which countries account for the final demand for the production of each country (for use in consumption, investment, and government expenditures), because the value of production is traced through various stages across countries to its ultimate destination. (It is often useful to distinguish such “final destination” countries from countries that import a good only to incorporate it into a product which they will then export.) Finally, the value-added approach provides more information on the contribution of various sectors to production for export.
Data and Data Sources for Value Added
A primary source for value-added trade data is the Trade in Value Added (TiVA) database, the product of a joint initiative of the Organisation for Economic Co-operation and Development (OECD) and the World Trade Organization (WTO). As of May 2013, it had gathered data on more than 57 countries and economic entities and zones, such as the European Union (EU) and the Association of Southeast Asian Nations, and on 18 economic activities, such as mining and electrical equipment. TiVA takes its underlying data from highly aggregated input-output tables issued by the participating countries and economic entities; these tables reflect how much of the value of the industry’s output is represented by inputs purchased from another industry. The World Input-Output Database (WIOD) is another project that publishes global and national input-output tables using a set format. Studies of high-visibility bilateral trade relationships have also benefited from using input-output tables made available by individual markets, including the United States, China, Japan, and the EU and certain EU members.
The Need for Value-Added Data and Challenges in its Use
Measuring trade on a value-added basis provides valuable insights to the sometimes intricate workings of global supply chains. This approach makes it easier to describe and analyze the various inputs and factors that result in products generated by the contemporary global supply chain. However, it is important to understand the limits on the underlying trade data when applying a value-added analysis. For example, while the TiVA database contains the most recently available data, most country’s data are from 2011, thus yielding a lag of several years in the timeliness of the raw data and a still longer lag in the timeliness of data analysis. The gap widens further if analysis involves national input-output tables from countries with an even longer data lag. As a result, conclusions drawn from the data, and even ways in which the data can be analyzed, can be dated by anywhere from 3 to10 years, depending on the framework and the purpose of the analysis. For goods with dynamic supply chains, such a lag may result in analytic conclusions that no longer reflect current trade trends or policy initiatives.
In addition, difficulties exist in gathering value-added statistics. These include the reluctance of certain sources to release data that may be commercially sensitive; the lack of a common statistical framework; data assumptions that may understate the effects of international engagement on a domestic economy; and challenges in distinguishing between inputs and final goods. A recent paper proposes a framework for addressing some of the data issues systematically, but considerable work remains to be done.
Nevertheless, even analysis of the data currently available yields benefits in terms of a clearer understanding of bilateral trade. For example, a country’s global trade balance is the same whether it is considered in light of gross trade statistics or value-added ones. Measures of bilateral trade balances, though, can differ greatly, depending on the method used to calculate them. For example, the U.S. bilateral trade deficit with China is smaller when measured in trade in value added than when measured in gross trade, because China buys so many of the inputs for its exports from other countries. However, U.S. bilateral trade deficits with many of those other countries are larger (or U.S. trade surpluses with them are smaller) when measured in trade in value added, since many U.S. imports from China incorporate the value of inputs originating in these countries. Given the attention routinely accorded the U.S.-China trade deficit, a more detailed understanding of that deficit would enhance any analysis involving it.
Applying the value-added approach to trade in services might also suggest different levels of trade flows for goods and services trade than traditional measures show. An analysis using the TiVA database finds that trade in a particular good may have a sizable component of trade in related services affecting the data. Measuring trade in manufactured goods using the value-added method, the U.S. trade deficit in manufactured goods is significantly lower than the gross measure, although the overall U.S. trade deficit (in goods and services together) remains unchanged. The lower calculated trade deficit in value added for manufactured goods results from foreign services that contribute to trade in manufactured goods, such as transportation, distribution, and finance services, which were previously categorized as manufactured imports but are now service sector imports in value-added terms. At the same time, the study found that the value created by services in the United States and incorporated directly or indirectly as inputs represented 27 percent of the total domestic value added in U.S. gross manufactured exports in 2009. For certain industrial classifications, such as wood and paper; food, beverages, and tobacco; and transport equipment, it is estimated that “more than one-third of the domestic value-added in exports comes from the services sector.”
Another analysis from a different perspective cautions against the use of value-added trade data when analyzing U.S. trade with China. The analysis contends that those data do not account in a timely fashion for the rapidly increasing domestic content in China’s exports and may be affected by China’s currency policies.