Presentation Transcript

1. Gross Exports Accounting and the Global Value Chain

   Zhi Wang United States International Trade Commission Based on “Tracing value-added and double counting in gross exports” by Koopmam, Wang and Wei, forthcoming in AER; and “Decomposing Gross Exports into Value Added and Double Counting at Bilateral and Sector Level” working in progress by Wang, Wei and Zhu The views expressed in this presentation are solely of the presenter. It is not meant to represent in anyway the official views of the USITC and its Commissioners
2. Presentation Outline Motivation Standard/official trade statistics are misleading in the presence of trade in intermediate goods Need a framework to bridge the gap in official trade data (in gross terms) and national accounts statistics (in value added terms) Need to have a full accounting of intermediate trade, the magnitude and source of the double counting, and the relation of double counting with a country’s position and participation in global value chain Overview of existing concepts and methods in trade in value-added and vertical specialization literature, their major shortcomings Decomposition of value-added production (forward looking) Decomposition of final demand (backward looking) Decomposition of gross trade flows (beyond Leontief insight)
3. Presentation Outline Use two-country case to illustrate the core ideas of gross trade accounting method and derivation of the decomposition formula Country's gross exports = exports of value added + domestic value added that returns home+ foreign value added+ double counting due to trade in intermediate goods Why we think the new accounting framework can reshape our understanding of global trade References: KWW, forthcoming in AER Decomposing gross exports at the country level WWZ, working in progress Decomposing gross exports at the bilateral and sector level
4. Objectives To establish a precise relation between value-added measures of trade and official trade statistics, thus creating a gross trade accounting method that is fully consistent with the SNA standard; To fully account gross intermediate trade quantifying double counting at both sector and bilateral levels and investigate what the source and structure of the double counting imply for cross country production sharing and a country’s position and participation in global value Chains; To provide a unified accounting framework that nests measures of vertical specialization, trade in value-added and other IO based decomposition methods in the literature as special cases; To create a conceptual framework that helps policymakers and the public to understand what official trade statistics really mean.
5. Measuring Domestic and Foreign Content in Exports: Original HIY measure of vertical specialization A country can participate in vertical specialization in two ways - Two measures of “vertical specialization” uses imported intermediate inputs to produce exports; VS measures the value of imported contents embodied in a country’s exports exports intermediate goods that are used as inputs by another country to produce goods for exports; VS1 measures intermediate exports sent indirectly through other countries Two key assumptions by standard HIY measure: the intensity in the use of imported inputs is the same for goods produced for exports and for domestic sales. Not true for processing trade in a large number of developing countries. all imported intermediate inputs are 100% foreign sourced Will underestimate domestic value-added share in exports, particularly for developed countries since their imports often embodied a large share of its own value-added.
6. Newer Measures of Trade in Value-added Value-added exports: Value-added produced in source country but absorbed in destination country (Johnson and Noguera, 2012) ; VAX Ratio : Ratio of value-added exports to gross exports as summary measure of value-added content of trade; Value-added is a "net" concept, double counting is not allowed; Vertical specialization measures intermediate goods cross border more than once, involving values show up in more than one country’s gross exports, so have to include double-counted portions in gross trade statistics. More border crossings of intermediate goods (more double counting) means larger difference between these two measures.
7. Shortcomings of existing measures and IIO based decomposition method Measures of trade in value-added and measures of vertical specialization are not equal to each other but are often used interchangeably in the literature. Existing measures all proposed as stand-alone indicators. No common framework that provides a unified way to account and illustrate their relationships. All measures proposed so far cannot identify all components in gross exports. Many studies ignore double counted components in official trade statistics by only focusing on trade in value-added. To understand shortcomings in existing IIO based accounting method, first need to understand Leontief insights.
8. What is Leontief insights When $1 export is produced, a first round of value-added is generated: Direct domestic value-added induced by the $1 exports. To produce that export, intermediate inputs have to be used. The production of these intermediate inputs also generates value-added. This is the indirect domestic value added induced by the $1 exports. Such a process to generate indirect value-added can be traced throughout the economy, as intermediate inputs are used to produce other intermediate inputs. The total domestic value-added induced by the $1 exports is equal to the sum of direct and all indirect domestic value-added generated from the $1 exports production process.
9. Production and trade in a two-country world All output is used as intermediate or final goods at home or abroad (1) In block matrix notations (2) Rearrange (3) Asr: N by N bock matrix of direct input/output coefficients Bsr: Nby N bock Leontief inverse matrix, measure the amount of gross output in s required for a one-unit increase in final demand at country r Ys: N by 1 vector, global use of country s’ final goods
10. Direct Domestic Value added Coefficient = 1 – intermediate input share from both countries Define as a 2N by 2N diagonal matrix with each country’s direct value-added coefficients by sector along the diagonal; Define as 2N by 2N final demand matrix with each country’s total final demand by sector along the diagonal;
11. Decomposition of value-added and final goods production based on Leontief insights (1) When N=2 (4) Decompose GDP by sector; forward looking linkage Decompose final goods by VA source; backward looking Linkage
12. Decomposition of value-added and final goods production based on Leontief insights (2) Sum up the matrix along the row, accounts how each country's domestic value-added originated in a particular sector is used by the sector and all its down stream countries/sectors. Sum up the matrix along the columns, gives the country/sector sources of value-added in each country's final demand, accounts all upstream countries/sectors’ value-added contribution to a specific country/sector’s final goods output. Thus, supply side perspective (sum across columns) decomposes how each country's GDP by industry are used, directly or indirectly to satisfy domestic or foreign final demand, while the user side perspective (sum across rows), decomposes a country/sector's final goods and services into its original country/sector sources.
13. Decomposition of value-added and final goods production based on Leontief insights (3) In the electronic sector, the supply side perspective would include the value added created by production factors employed at the electronics sector and incorporated into gross exports of electronics itself (direct domestic value-added exports), as well as exports of computers, consumer appliances, and automobiles (indirect domestic value-added exports). The user side perspective would include value added in the electronics sector itself (direct domestic value-added exports) as well as value added in inputs from all other upstream sectors/countries (such as glass from country A, rubber from country B, transportation and design from the home country) used to produce electronics to satisfy domestic or foreign final demand (indirect domestic value-added in exports and foreign value-added in exports).
14. The two methods give the same value-added trade estimates at aggregate level, but different at sector level There are two type value-added trade at sector level can be estimate from ICIO tables: The forward looking estimates give each producing sector's value-added absorbed by each destination country, regardless which sector’s goods in the destination country. It is consistent with factor content method in the trade literature. The backward looking estimates give each source country's value-added embodied in its sector gross export flows, regardless value-added creating sectors in the source country, absorbed by each destination country in term of the source country's gross exports. It is consistent with GVC case studies in the literature. At the country aggregate bilateral level, these two value-added trade flow estimates are exactly the same, but at sector level they are quite different.
15. The two methods give the same VA trade flows at aggregate, but different at sector level due to different way to measure indirect VA exports
16. Gross exports (E) Value-added exports (VT) Foreign Value-added (FV) (1) DV in direct final goods exports (2) DV in interme-diates exports absorbed by direct importers (4) DV in intermediates that returns via final imports (5) DV in interme-diates that returns via Interme-diate imports (3) DV in intermediates re-exported to third countries IV Gross Exports Accounting: Concepts based on but beyond Leontief insights DVA in exports Domestic VA in intermediate exports returns home Pure double counting (9) double counted interme-diate exports produced abroad (6) double counted interme-diateexports produced at home (8) FV in interme-diate goods export (7) FV in final goods exports Domestic content (DC)
17. The conceptual framework of KWW A mathematical framework to decompose a country’s total gross exports into nine value-added and double counted components. The nine components can be grouped into four buckets: (1) value added exports; (2) domestic value added that is first exported but eventually returned home, it is not part of a country's exports of value-added, but account for part of the country's GDP; (3) foreign value added used in the production of a country’s exports and eventually absorbed by other countries; (4) “pure double counted terms” arising from intermediate goods being traded back and forth that cross border multiple times. It occurs only when there is two way intermediate goods trade.
18. KWW nests existing measures as special cases Value-added exports equals (1) + (2) +(3), divided by gross exports is the VAX ratio defined by Johnson and Noguera. GDP in exports (1) + (2) + (3) + (4) +(5). Domestic content in a country's exports equals (1) + (2) + (3) + (4) +(5)+(6). (7)+(8)+(9) is labeled as VS by HIY (2001) (3) + (4)+(5)+(6) is part of VS1 labeled by HIY (2001) (4) are also labeled as VS1* by Daudin et al (2011). (4) through (9) involve value-added cross national borders at least twice, and are the sources of multiple counting in official trade statistics
19. Conceptual Difference between Domestic Value-added in Exports, Value-added Exports and Domestic Content in Exports Domestic VA share in a country’s exports and the VAX ratio are, in general, not equal to each other. The are related but different concept. DVA share only looks where the value added is originated regardless where it is ultimately absorbed. While a country’s “value added exports” refers to a subset of “domestic value added in a country’s exports” that is ultimately absorbed abroad. “Domestic content of a country’s exports” is even broader than “ DVA in a country’s exports.” It is the sum of the first five terms in the conceptual graph, and the fifth term reflect pure double counting of value added that originate in the home country. Symmetrically, “the foreign content of a country’s exports” is the sum of the last three terms and the last term reflect pure double counting of value added that originates in foreign countries.
20. Connections and Distinctions between Trade in Value added and Vertical Specialization Defined in such a way, the foreign content of a country’s exports in KWW is an extension of VS measure proposed by HIY (2001) in multi-country setting without HIY’s restrictive assumption there is no two-way trade in intermediates, but still keep the property of sum of the domestic and foreign contents of a country’s exports is equal to that country’s total gross exports the same as HIY. Gross exports accounting provides additional information on the structure of double counting in gross trade statistics (in addition to the total amount of double counting).
21. Applications of Leontief insights in trade in Value-added and GVC Decomposition (1) Decompose final goods to its country/sector value-added sources based on Leontief insights (backward looking linkage, sum matrix across rows). Timmer at al (2013) did a very nice job to conduct such a decomposition in the maximum dimension that WIOD data allow. (Fby GN by GN). For example, it can tell how much contribution a unskilled worker employed in Chinese steel industry made to a car produced in Germany, or how much contribution a US skill worker in electronic industry made to a computer consumed by a Chinese consumer.
22. Applications of Leontief insights in trade in Value-added and GVC Decomposition (2) Decompose GDP by industry to where and how it is used based on Leontief insights (forward looking linkage, sum matrix across columns). Johnson and Noguera (2012 JIE) did an incomplete job, only measured sector value-added absorbed by foreign countries, did not take GDP absorbed in domestic market into account, so cannot reorganize the conceptual difference between value-added exports and GDP in exports. Most importantly, their sector level VAX ratio did not defined correctly, they compute value-added exports by forward looking linkage, which give the amount of domestic value-added created in the source country’s production sector absorbed in the destination country, but in the source country's sector gross exports, there are also domestic value-added from other domestic sectors.
23. VA exports/GDP and DVA in exports/GDP ratioThe difference is DVA returned to the source country
24. VA exports/GDP and DVA in exports/GDP ratioThe difference is DVA returned to the source country
25. VA exports to gross exports ratio- Electrical and optical equipment (WIOD sector14)
26. VA exports to gross exports ratioDifference between J&N and WWZ
27. VA exports to gross exports ratioDifference between N&J and WWZ Value distribution of N&J sector VAX ratio on WIOD
28. Applications of Leontief insights in trade in Value-added and GVC Decomposition (3) Although there is difference between through and rough jobs that Timmer et al. and J&N have done, but both of them are applying Leontief insights, while our work on decomposition of intermediate goods trade is go beyond Leontief insights. In Leontief’s time, intermediate goods trade are not so important as today, so he did not consider how it should be decomposed into value-added and double counted components as we did now.
29. Useful information from the structure of double counting in a country’s gross exports Estimating double counted terms and their relative importance in addition to value-added exports helps to understand how acountry participates in global value chains. A country’s exports of value-added is smaller than its gross exports due to: a) the use of imported inputs to produce exports; b) a part of its DVA in exports may return home after being processed abroad; c) Double counting due to two way intermediate goods trade. Two countries can have identical value added to gross exports ratios but very different structure of (a), (b) and (c).
30. , Numerical Example: Two countries, one sector, and one of the countries does not export intermediate good
31. Numerical Example: Two countries, one sector, and one of the countries does not export intermediate good The Leontief inverse and share of value-added by source matrix Gross exports decomposition matrix BY The value-added production and trade matrix FV share DV share VA exports
32. Gross exports accounting results for the numerical example Both countries has identical VAX ratio, but the reasons why value added exports smaller than gross exports are different. For USA, due to some of its own value added that is initially exported returns home after being used as an intermediate input in CHN to produce final goods exports back to USA. For CHN, due to its production for exports uses FV: intermediate goods from the USA which embeds USA’s value added. the return home VA (23.3) is a true value added for USA ’s national account, part of its GDP, but it is a double counting in official trade statistics and in China’s point of view;
33. Advanced economies Asia NICs Accounting of Gross Exports--Actual data, 2004 About half of double counting in U.S. exports from its own value-added returns home via imports (12.4% over 25.4%); Almost all of the double counting in Chinese and Mexican processing exports comes from imported foreign contents (56.6/56.9; 63.4/63.7); Highlight U.S. export producers and Chinese and Mexican processing exporters' respective positions at the head and tail of the global value chain. Emerging Asia Other emerging China Mexico
34. Gross exports decomposition- Electrical and optical equipment (WIOD sector14)
35. Gross exports decomposition- Transport equipment (WIOD sector15)
36. Gross exports decomposition- Textile and Products (WIOD sector 04)
37. Gross exports decomposition- Basic and Fabricated Metals (WIOD sector 12)
38. Gross exports decomposition- Basic and Fabricated Metals (WIOD sector 12)
39. Gross and Domestic Value-added-adjusted Revealed Comparative Advantage Indicators, 2004 China and India rank 1stand 4th based on gross trade data. Using domestic value-added data, China and India’s rankings drop to 7th and 15th place! For India, the sector has switched from being labeled as a comparative advantage sector to a comparative disadvantage sector (drop from 1.3 to 0.66)! The ranking of US and Japan also switches from being labeled as a comparative disadvantage sector to a comparative advantage sector! (0.93 to 1.32 for US, 0.87 to 1.05 for Japan)
40. Gross and Domestic Value-added-adjusted Revealed Comparative Advantage Indicators, 2004 - Based on gross trade data, India exhibits a strong revealed comparative advantage on the strength of its high share of business services exports in its overall exports (2.4). Once we compute RCA using domestic value-added in exports, the same sector becomes a comparative disadvantage sector for India (0.7)! - Because business services in advanced countries are often exported indirectly by being embed in their manufacturing exports, after taking into account indirect value added exports, the Indian share of the sector in its exports becomes much less impressive.
41. Gross and Domestic Value-added-adjusted Revealed Comparative Advantage Indicators: Electrical and Optical Equipment 1995-2009
42. Gross and Domestic Value-added-adjusted Revealed Comparative Advantage Indicators: Electrical and Optical Equipment 1995-2009
43. Conclusion National income accounts record domestic output in value added terms but standard trade statistics record trade in gross terms. Our method is the first in the literature to establish a formal and precise relationship between value-added measures and officially reported trade statistics. We have created a gross trade accounting framework that is fully consistent with the SNA standard. It has gone beyond the original Leontief insights and offered a way to measure double counting caused by back and forth intermediate goods trade. It has the potential to reshape our basic understanding of global trade.
44. Conclusion - continues Our accounting method identifies exports of domestic value added; domestic value-added exported first but eventually returns home; foreign value-added used in production of exports and absorbed by other countries; and double counted items arising from two way intermediate goods trade. These conceptually different components sum up to 100% of official gross exports statistics at both sector and aggregate level. By identifying which parts of the official trade statistics are double counted and the sources of the double counting, we provide a effective way to correctly interpret official trade statistics in value-added terms and provide an index to quantitatively measure what and how much is double counted for each intermediate trade flows.