\documentclass[12pt,letterpaper]{article} \begin{document} {THE INTERNATIONAL TRADE AND PRODUCTION\\ } {DATABASE FOR ESTIMATION (ITPD-E)\\ } {Ingo Borchert, Mario Larch,\\ Serge Shikher, and Yoto Yotov\\ } {ECONOMICS WORKING PAPER SERIES}\\ Working Paper 2020--05--C\\ U.S. INTERNATIONAL TRADE COMMISSION\\ 500 E Street SW\\ Washington, DC 20436\\ May 2020 Office of Economics working papers are the result of ongoing professional research of USITC Staff and are solely meant to represent the opinions and professional research of individual authors. These papers are not meant to represent in any way the views of the U.S. International Trade Commission or any of its individual Commissioners. Working papers are circulated to promote the active exchange of ideas between USITC Staff and recognized experts outside the USITC and to promote professional development of Office Staff by encouraging outside professional critique of staff research. Please address correspondence to \texttt{gravity@usitc.gov}. The International Trade and Production Database for Estimation (ITPD-E)\\ Ingo Borchert, Mario Larch, Serge Shikher, and Yoto Yotov\\ Office of Economics Working Paper 2020--05--C\\ May 2020\\ \hspace*{0.333em}\\ Ingo Borchert, University of Sussex Business School\\ Mario Larch, University of Bayreuth, CEPII, CESifo, GEP, ifo\\ Serge Shikher, United States International Trade Commission\\ Yoto Yotov, Drexel University, ifo, CESifo\\ \textbf{Disclaimer.} The ITPD-E is a public good that was created in response to market demand. Its initial development and maintenance was a substantial long-term effort by the authors. Accordingly, in return for that effort, we expect two things from all users of the ITPD-E. First, please cite this paper if you are using the database. Second, if you believe that there is a mistake in the database or that the database can be improved by incorporating additional data or more reliable data, even if only for an individual country or industry, please let us know by writing to ITC's gravity portal e-mail address (\texttt{gravity@usitc.gov}). We will try to accommodate detection of errors, inconsistencies, and suggestions as soon as possible. Please visit \href{https://gravity.usitc.gov}{USITC's gravity portal} for updates. \textbf{Acknowledgments:} Our team is grateful for the research support and encouragement that we received from the United States International Trade Commission. The USITC is not in any way responsible for any errors in the ITPD-E. \hypertarget{introduction}{% \section{Introduction}\label{introduction}} This paper describes the construction of the International Trade and Production Database for Estimation (ITPD-E).\footnote{In the future, we plan to produce two additional versions of the database. One will have missing data filled in using the latest methodological advances in structural gravity modeling. This database will constitute the second member of the ITPD family and, due to its suitability for simulations, will be called ITPD-S. A third version within the ITPD family of databases will link the trade flow information to an input-output table (ITPD-IO).} The ITPD-E contains consistent data on bilateral international trade and domestic trade, calculated using production data, for a large number of countries, industries, and years. The data are constructed at the industry level covering the broad sectors of agriculture, mining and energy, manufacturing, and services. Therefore, the ITPD-E describes nearly completely the traded sectors of each economy. The ITPD-E covers the time period starting in 2000 and going up to 2016, the most recent year for which data across all sectors are available. The ITPD-E provides information for 243 countries, 170 industries, and 17 years. This includes 26 industries in agriculture, 7 in mining and energy, 120 in manufacturing, and 17 in services. We envision statistical estimation as the main use of the ITPD-E. Hence, the database consists of the reported administrative (``raw'') data from the sources that we describe in the next section. It includes bilateral trade flows as well as domestic trade. The latter is constructed by using production data, i.e.~as the difference between production and total exports. Importantly, since exports are reported on a gross basis, production data are on a gross basis too. In order to ensure maximum coverage, we combine trade data reported by both importer and exporter to fill in as many missing values as possible in the trade and production data. It is important to note that no estimation models, such as the gravity framework, have been used to fill any missing observations in ITPD-E. This is one of the key distinguishing features of ITPD-E that makes it suitable for estimation purposes. Naturally, ITPD-E is not balanced and includes missing observations for some years and countries. The paper is structured as follows. Section \protect\hyperlink{sec_ITPD-E_creation}{{[}sec\_ITPD-E\_creation{]}} describes our approach to constructing the database and provides more detail on the dimensions of ITPD-E in terms of countries, industries, and years. In Section \protect\hyperlink{data_sources}{{[}data\_sources{]}} we describe the data sources used to construct ITPD-E. Section \protect\hyperlink{sec_comparison}{{[}sec\_comparison{]}} compares ITPD-E to existing trade and production datasets. In Section \protect\hyperlink{sec_validation}{{[}sec\_validation{]}} we show the application of ITPD-E to the standard gravity model. The results serve as a validation of this database and also demonstrate its usefulness, while being transparent about its limitations. The last section concludes and outlines future work planned on the data. \hypertarget{the-itp-database-construction-and-key-featuressec_itpd-e_creation}{% \section[The ITP Database: Construction and Key Features]{\texorpdfstring{The ITP Database: Construction and Key Features\protect\hypertarget{sec_ITPD-E_creation}{}{{[}sec\_ITPD-E\_creation{]}}}{The ITP Database: Construction and Key Features{[}sec\_ITPD-E\_creation{]}}}\label{the-itp-database-construction-and-key-featuressec_itpd-e_creation}} This section describes the principal steps involved in the construction of the ITPD-E and discusses the database's key features and dimensions. These steps are common to all four broad sectors included in the ITPD-E. More detail regarding data sources for each sector and any sector-specific steps undertaken to construct the data are described in Section \protect\hyperlink{data_sources}{{[}data\_sources{]}}. \hypertarget{construction-of-the-international-trade-datasecconstr_of_int_trade_data}{% \subsection[Construction of the International Trade Data]{\texorpdfstring{Construction of the International Trade Data\protect\hypertarget{sec:constr_of_int_trade_data}{}{{[}sec:constr\_of\_int\_trade\_data{]}}}{Construction of the International Trade Data{[}sec:constr\_of\_int\_trade\_data{]}}}\label{construction-of-the-international-trade-datasecconstr_of_int_trade_data}} Each international trade flow is recorded and reported separately by the two parties in the transaction, exporter and importer. In order to take full advantage of all reported international trade data we use a mirroring procedure described below. In goods trade, data on imports is considered more reliable than data on exports because of the oversight from governments enforcing their tariff schedule and other import regulations.\footnote{See for example \href{https://wits.worldbank.org/wits/wits/witshelp/Content/Data_Retrieval/T/Intro/B2.Imports_Exports_and_Mirror.htm}{https://wits.worldbank.org/wits/wits/witshelp/Content/Data\_Retrieval/T/Intro/B2. Imports\_Exports\_and\_Mirror.htm} and \cite{Timmeretal2012}. Specifically for developing countries, \cite{RozanskiYeats1994} show that export data are less reliable than import data.} In addition, importer-reported trade values, which are reported on the c.i.f.~basis, are consistent with gravity theoretical methodology. Therefore, we primarily use importer-reported values for goods trade, as is also done for example by \cite{Feenstraetal2005}. Since there are no tariffs levied on services trade, importing countries lack the fiscal incentive to carefully keep track of services imports. Services export data, in turn, are often collected as part of mandatory surveys run by national statistical agencies and/or central banks, and therefore are considered more accurate than imports. For these reasons, and because there is no c.i.f./f.o.b.~distinction in services trade, we primarily use exporter-reported values for services trade. In our mirroring procedure for goods, we use exports reported by partner countries to fill in missing values for the import values. For services, we use reported imports to fill in missing values of exports. To denote mirrored cases, the ITPD-E includes a flag variable named \(flag\_mirror\). In Section 5 we demonstrate that the additional observations that arise from mirroring do not affect estimated coefficients of standard gravity variables. Nevertheless, their inclusion is potentially important to ensure proper country and industry coverage. \hypertarget{construction-of-the-domestic-trade-datasecconstr_of_dom_trade_data}{% \subsection[Construction of the Domestic Trade Data]{\texorpdfstring{Construction of the Domestic Trade Data\protect\hypertarget{sec:constr_of_dom_trade_data}{}{{[}sec:constr\_of\_dom\_trade\_data{]}}}{Construction of the Domestic Trade Data{[}sec:constr\_of\_dom\_trade\_data{]}}}\label{construction-of-the-domestic-trade-datasecconstr_of_dom_trade_data}} Domestic trade is calculated as the difference between the (gross) values of total production and total exports. Total exports are constructed as the sum of bilateral trade, as reported in the ITPD-E, for each exporting country. In the relatively few instances in which our procedure results in negative domestic trade values, we delete those observations from the database. The sources of output and trade data are described in Section \protect\hyperlink{data_sources}{{[}data\_sources{]}}. \hypertarget{final-proceduressecfinal_procedures}{% \subsection[Final Procedures]{\texorpdfstring{Final Procedures\protect\hypertarget{sec:final_procedures}{}{{[}sec:final\_procedures{]}}}{Final Procedures{[}sec:final\_procedures{]}}}\label{final-proceduressecfinal_procedures}} We combine the domestic and international trade flows for each of the 170 ITPD-E industries into a single database. Then, we create a balanced database across all dimensions of the ITPD-E by filling all missing international trade observations with zeroes. In order to distinguish between the trade zeroes that exist in the original raw data and the new zeroes that are added to balance the data, we create a flag variable called \(flag\_zero\), which is equal to `r' for observations with zeroes coming from original data sources, `p' for observations with positive trade flows, and `u' for observations filled with zeroes in this step. We do this for international trade observations only, not for domestic trade observations. Since the previous procedure results in the addition of many zeroes that are irrelevant for gravity estimations, in order to eliminate outliers (e.g., countries with very few observations that would be dropped in standard gravity regressions), and to ensure that ITPD-E is suitable for even the most demanding gravity specifications, we use the Poisson pseudo-maximum likelihood (PPML) estimator with a demanding set of fixed effects (i.e., exporter-time, importer-time, and directional country-pair fixed effects) to estimate gravity for each of the 170 ITPD-E industries. Then, we retain the estimating sample for each industry as our final industry-level data. Note that this procedure will eliminate all irrelevant zeroes from our sample, e.g., if a country does not export a given industry in a given year, then the corresponding zeroes will be captured perfectly by this country's exporter-time fixed effect. Thus, even after the rectangularisation of the data that we described in the previous paragraph, ITPD-E remains an unbalanced database as some countries do not appear in some years and/or ITPD-E industries. \hypertarget{country-industry-and-year-coverage}{% \subsection{Country, Industry, and Year Coverage}\label{country-industry-and-year-coverage}} The procedures that we implement ensure that for each country in the ITPD-E there is a sufficient number of observations in at least one industry that are meaningful for estimation purposes. The final dimensions of our database are as follows. In terms of years, the ITPD-E covers the 17-year period between 2000 and 2016. After combining the raw data from the four main sectors and implementing the data cleaning and construction procedures, we end up with about two million observations in each year. About a million observation each year has positive trade flows. The number of observations varies significantly across sectors. Usually, and as expected, in each year the large bulk of the data pertains to manufacturing. However, we also have significant number of observations for agriculture (around 75,000 non-zero trade observations per year and about 160,000 observations overall), for mining and energy (around 15,000 non-zero trade observations per year and about 35,000 observations overall), and for services (around 20,000 non-zero trade observations per year and about 40,000 observations overall). The ITPD-E covers 170 industries. Of those, 26 are in agriculture, 7 are in mining and energy, 120 are in manufacturing, and 17 are in services. Table \protect\hyperlink{ITPD-E_sectors}{1} lists the 170 industries covered in ITPD (in Column 1), and their corresponding ITPD codes (in Column 2). In addition, for each industry, the table reports average positive exports (in Column 3), maximum exports (in Column 4), total number of observations (in Column 5), and total number of zeroes (in Column 6). Note that trade in this table includes both international and domestic trade, so the maximum exports are typically from the domestic trade. Industries 1-26 belong to the ``Agriculture'' broad sector. Industries 27 to 33 belong to the ``Mining and energy'' broad sector. Industries 34 to 153 belong to the broad sector ``Manufacturing''. Finally, industries 154 to 170 belong to the broad sector ``Services''. Our sectoral coverage closely follows the ISIC rev.~4 classification system. Our Agriculture broad sector corresponds to division A-01 ``Crop and animal production'' in ISIC. Mining and energy broad sector includes ISIC sections B and D. The manufacturing broad sector corresponds to section C. The services broad sector includes sections F through K, M, N, and P through S. Several ISIC divisions are not included in the initial release of ITPD-E. They include ISIC divisions A-02 ``Forestry'' and A-03 ``Fishing'', and mostly non-traded ISIC sections L ``Real Estate'', O ``Government'', T ``Household production'', and U ``Extraterritorial organizations''. We were careful to avoid double-counting of industries when assembling data from multiple sources. For example, manufactured food is included in the manufacturing broad sector and not in the agriculture. In terms of country coverage, the ITPD-E includes 243 countries.\footnote{Consistent with the practice in USITC's DGD, the number of countries in the text and in Table~\protect\hyperlink{ITPD-E_countries}{2} reflects the total number of distinct entities. In cases when countries split at certain points in history, this then includes the single code for both regions prior to the split as well as the two newly created codes post-separation. In ITPD-E this is the case for ``Serbia and Montenegro (SCG)'' prior to 2006, which appear as Serbia (SRB) and Montenegro (MNE) respectively afterwards, as well as for ``Netherlands Antilles (ANT)'', which appear as Curacao (CUW) and Sint Maarten (SXM) after 2010, even though both entities continue to be part of the Kingdom of the Netherlands.} There are only two countries (French Guiana and Monaco) in the ITPD-E that appear only as exporters but not as importers. For all other countries there is data as both exporters and importers. Table~\protect\hyperlink{ITPD-E_countries}{2} lists the 243 countries covered in ITPD-E (in Column 2), and their corresponding 3-letter ISO codes (in Column 1).\footnote{There are three countries that have multiple 3-letter alpha codes in the alternative original databases, which were used to construct ITPD-E. These countries are Chinese Taipei (Taiwan), which appears as CHT or TWN, the Democratic Republic of Congo, which appears as DRC, ZAR, or COD, and Romania, which appears as ROM or ROU. Without any strong preference and any implications, and only for consistency purposes and ease of use of ITPD-E, we selected to use TWN, COD, and ROU.} In addition, for each country, the table reports average positive exports (in Column 3), maximum exports (in Column 4), total number of observations (in Column 5), and total number of zeroes (in Column 6). Similarly to Table \protect\hyperlink{ITPD-E_sectors}{1}, trade in Table \protect\hyperlink{ITPD-E_countries}{2} includes both international and domestic trade, so the maximum exports are typically from the domestic trade. \hypertarget{file-format-and-columns}{% \subsection{File Format and Columns}\label{file-format-and-columns}} The database is distributed as a comma-separated file, ITPD\_E\_RXX.csv, where XX is the release number. Therefore, the initial release file is called ITPD\_E\_R01.csv. The columns in the database are described in Table \protect\hyperlink{tab:file_columns}{3}. The ISO 3-letter country codes and year can be used to merge the ITP database with the USITC's Dynamic Gravity Dataset as well as other datasets that use the standard ISO codes.\footnote{Note that ITPD-E uses a constant country code for Romania, ROU, while the DGD version 1.0 uses historically-accurate country codes, which means it denotes Romania by ROM in 2000-2001 and ROU in 2002 and later. Thus, the only adjustment that needs to be made when merging ITPD-E with DGD is to change all instances of ROM to ROU in DGD.} Trade values are expressed in millions of current United States dollars. Trade is international when exporter and import countries are different. Trade is domestic when the importer and exporter are the same country. As explained in Section \protect\hyperlink{sec:constr_of_int_trade_data}{{[}sec:constr\_of\_int\_trade\_data{]}}, \emph{flag\_mirror} is equal to 1 for observations that are obtained from the mirror trade data. The variable \emph{flag\_zero} shows whether the current observation contains zero or positive trade and, in case of zero, the origin of the zero. If the current observation contains positive trade, then \emph{flag\_zero} is equal to `p' (``positive''). If the current observation had a zero in the original raw data then \emph{flag\_zero} is equal to `r' (``reported''). If the current observation is filled in with a zero when the dataset is balanced (see Section \protect\hyperlink{sec:final_procedures}{{[}sec:final\_procedures{]}}), then \emph{flag\_zero} is equal to `u' (``unknown''). \hypertarget{data-sourcesdata_sources}{% \section[Data Sources]{\texorpdfstring{Data Sources\protect\hypertarget{data_sources}{}{{[}data\_sources{]}}}{Data Sources{[}data\_sources{]}}}\label{data-sourcesdata_sources}} This section discusses the original data sources that are employed in the construction of the data, as well as their strengths and limitations. We split this discussion into four subsections: (1) Agricultural data, (2) Mining and energy data, (3) Manufacturing data, and (4) Services data. For each broad sector, we discuss sources of international trade and production data. The common principles and steps in the construction were described in Section \protect\hyperlink{sec_ITPD-E_creation}{{[}sec\_ITPD-E\_creation{]}}. The steps and procedures that are specific to each broad sector are described in this section. In selecting constituent raw data sources for the ITPD-E, we considered that data sources should provide clear documentation, contain data that was not estimated by statistical procedures, and are regularly updated. \hypertarget{agriculture}{% \subsection{Agriculture}\label{agriculture}} The agriculture broad sector is divided into 26 industries that cover products contained in ISIC rev.~4 division A-01, which includes production of crop and of animal products.\footnote{The ITPD-E industrial classification for agriculture was developed with assistance from USITC staff.} The initial release of ITPD-E does not include forestry and logging (contained in ISIC division A-02) and fishing and aquaculture (contained in ISIC division A-03) because the data for those industries are not included in the FAOSTAT dataset, which is our main data source in agriculture. The agricultural industries in the ITPD-E are listed in Table \protect\hyperlink{tab:ITPD-E_class_ag}{4}. \hypertarget{trade-data}{% \subsubsection{Trade Data}\label{trade-data}} The Food and Agriculture Organization of the United Nations Statistics Division (FAOSTAT)\footnote{\url{http://faostat3.fao.org/home/E}.} collects information on an annual basis for more than 245 countries. The source data come from UNSD, Eurostat, and other national authorities as needed. This source data is checked for outliers and data on food aid is added to take into account total cross-border trade flows.\footnote{\url{http://www.fao.org/faostat/en/\#data/TM}.} FAOSTAT's Detailed Trade Matrix reports information on agricultural bilateral trade quantities (in tons) and values (in thousands of US dollars). Bilateral trade data are available from 1986 to 2017 for many countries. The original FAO data are classified according to the FAOSTAT Commodity List (FCL) which includes more than 600 items.\footnote{\url{http://www.fao.org/economic/ess/ess-standards/commodity/en/}.} As we provide a database containing agriculture, mining, energy, manufacturing, and services, we have to carefully avoid double counting. Specifically, some of the FAO FCL items contain mining and manufactured industries, so we do not include these FCL items in ITPD's agriculture industries. Specifically, we classify all industries between 1500 and 1601 of ISIC rev.~3 as manufacturing industries. Using the FCL to HS and HS to ISIC rev.~3 correspondence tables, we identify the FCL items that are part of the manufacturing data.\footnote{These are FCL items 16, 18, 19, 20, 21, 22, 23, 24, 26, 28, 29, 31, 32, 34, 36, 37, 38, 39, 41, 45, 46, 48, 49, 50, 51, 57, 58, 60, 61, 64, 66, 72, 76, 80, 82, 84, 86, 90, 95, 98, 104, 109, 110, 111, 113, 114, 115, 117, 118, 119, 121, 126, 127, 129, 150, 154, 155, 158, 159, 160, 162, 163, 164, 165, 166, 167, 168, 172, 173, 175, 212, 235, 237, 238, 239, 240, 241, 244, 245, 246, 247, 252, 253, 257, 258, 259, 261, 262, 264, 266, 268, 269, 271, 272, 273, 274, 276, 278, 281, 282, 290, 291, 293, 294, 295, 297, 298, 306, 307, 313, 314, 331, 332, 334, 335, 337, 338, 340, 341, 343, 390, 391, 392, 447, 448, 450, 451, 466, 469, 471, 472, 473, 474, 475, 476, 491, 492, 496, 498, 499, 509, 510, 513, 514, 517, 518, 519, 538, 539, 562, 563, 564, 565, 575, 576, 580, 583, 584, 622, 623, 624, 625, 626, 631, 632, 633, 634, 657, 658, 659, 660, 662, 664, 665, 666, 672, 737, 753, 768, 770, 773, 774, 828, 829, 831, 840, 841, 842, 843, 845, 849, 850, 851, 852, 853, 854, 855, 867, 869, 870, 871, 872, 873, 874, 875, 877, 878, 882, 883, 885, 886, 887, 888, 889, 890, 891, 892, 893, 894, 895, 896, 897, 898, 899, 900, 901, 903, 904, 905, 907, 908, 909, 910, 916, 917, 919, 920, 921, 922, 927, 928, 929, 930, 947, 949, 951, 952, 953, 954, 955, 957, 958, 959, 977, 979, 982, 983, 984, 985, 988, 994, 995, 996, 997, 998, 999, 1008, 1010, 1017, 1019, 1020, 1021, 1022, 1023, 1035, 1037, 1038, 1039, 1040, 1041, 1042, 1043, 1058, 1059, 1060, 1061, 1063, 1064, 1065, 1066, 1069, 1073, 1074, 1075, 1080, 1081, 1089, 1097, 1098, 1102, 1103, 1104, 1105, 1108, 1109, 1111, 1112, 1127, 1128, 1129, 1130, 1141, 1151, 1158, 1160, 1163, 1164, 1166, 1167, 1168, 1172, 1173, 1174, 1175, 1186, 1187, 1221, 1222, 1223, 1225, 1241, 1242, 1243, 1273, 1274, 1275, 1276, 1277, and 1296.} Note that these FCL items typically do not have matching production data in the FAO's database. Some FCL items could not be uniquely matched to broad sectors. In these cases, we allocated FCL items according the number of constituent HS lines.\footnote{This leads in dropping of industries with FCL items 653, 868, 948, 978, 1018, 1036, 1232, and 1259. We kept the following FCL items in agriculture: 667, 777, 780, 782, 826, 987, 1009, and 1293.} We also dropped industries we could not match to any ISIC or HS code\footnote{Specifically, we could not match FCL items 10, 30, 464, 944, 972, 1012, 1032, 1055, 1070, 1077, 1084, 1087, 1094, 1120, 1122, 1124, 1137, 1144, 1154, 1159, and 1161.} and industries with FCL item codes above 1296, which are aggregates and industries such as fertilizers, pesticides, and machinery, belonging to one of the other broad sectors. Table \protect\hyperlink{tab:ITPD-E_class_ag}{4} includes the correspondence between ITPD-E agricultural industries and FCL items. FAOSTAT does not include international trade data for FCL item 27 ``rice (paddy)''. Instead, we use data from the UN Commodity Trade Statistics Database (COMTRADE), HS sector 100610 ``Cereals; rice in the husk (paddy or rough)''. \hypertarget{production-data}{% \subsubsection{Production Data}\label{production-data}} The FAOSTAT also provides disaggregated agricultural production data.\footnote{Available at \url{http://www.fao.org/faostat/en/\#data}.} We use the data on the ``Value of Agricultural Production'' because we are interested in values in order to be able to combine and compare data across sectors. The Value of Agricultural Production data from the FAO contains information for at least some years for 261 geographical ``areas'' (which include country groups) and 220 different FCL items. We use gross production values in current United States dollars. The industry classification of the FAO production data is the same as its trade data: FAOSTAT Commodity List (FCL) (see \href{http://www.fao.org/economic/ess/ess-standards/commodity/en/}{http://www.fao.org/economic/ess/ess-stan- dards/commodity/en/}). There are 220 different items in the raw ``Value of Agricultural Production'' data. As explained in the previous section, our classification of agricultural industries contains 26 industries, described in Table \protect\hyperlink{tab:ITPD-E_class_ag}{4}. A correspondence between the country codes used by the FAO and the ISO3 country codes used by the ITPD-E can be found at \url{http://www.fao.org/faostat/en/\#data/QV}. The production data is then used to calculate domestic trade following the procedure described in Section \protect\hyperlink{sec:constr_of_dom_trade_data}{{[}sec:constr\_of\_dom\_trade\_data{]}}. There are two approaches to calculating domestic trade at a more aggregate level from disaggregated data: (a) calculate domestic trade at the disaggregated level and then aggregate and (b) aggregate output and total exports separately and then calculate domestic trade at an aggregate level. The two approaches may produce different results when there is missing data. For agriculture, there are many missing values at the disaggregated level. However, this is natural, as some countries will not produce some agricultural goods. We therefore use approach (a) and calculate domestic sales at the disaggregated level for agriculture. Note that constructing domestic sales as difference between output data and total exports leads to negative values for some observations, which we drop. \hypertarget{mining-and-energy}{% \subsection{Mining and Energy}\label{mining-and-energy}} The mining and energy broad sector in ITPD-E consists of 7 industries that cover products that are part of ISIC rev.~4 Sections B and D. Section B includes the extraction of minerals occurring naturally as solids, liquids, or gases. This section also includes supplementary activities aimed at preparing the crude materials for marketing. Section D covers operation of electric and gas utilities that provide electric power, natural gas, steam, and hot water through a permanent infrastructure. The ITPD-E mining and energy industries are listed in Table \protect\hyperlink{tab:ITPD-E_class_mining}{{[}tab:ITPD-E\_class\_mining{]}}. \hypertarget{trade-data-1}{% \subsubsection{Trade Data}\label{trade-data-1}} The international trade data for the mining and energy sector come from the UN Commodity Trade Statistics Database (COMTRADE), which is the most comprehensive source for bilateral data on merchandise trade flows. The data are available at a very disaggregated level and for a long period of time. COMTRADE reports annual bilateral trade flows (in current US dollars converted from national currencies) expressed in gross value and volume from 1962. The data can be accessed online through the UN website or through the World Bank's WITS portal.\footnote{\url{http://comtrade.un.org} or \url{http://wits.worldbank.org/Default.aspx?lang=en}.} The COMTRADE data are accessible in different nomenclatures and in different levels of disaggregation. Trade data classified according to the Harmonised System (HS) are available up to the 6-digit level (that is, at a level of detail that distinguish about 5,000 separate goods items), which is the most disaggregated classification that is consistent across countries at the international level,\footnote{When using the HS 2002 classification, mineral products are captured in Section V (25-27), and metals in Sections XIV (71) and XV (72-83). See \url{https://unstats.un.org/unsd/tradekb/Knowledgebase/50043/HS-2002-Classification-by-Section}.} thus enabling us to aggregate the mining trade data to our desired level of ITPD-E aggregation, which is determined by the availability of the production data as discussed next. \hypertarget{production-data-1}{% \subsubsection{Production Data}\label{production-data-1}} To obtain production data for the mining and energy sector, we rely on the dataset on mining and quarrying and utilities sectors from the United Nations Industrial Development Organization (UNIDO) at the 2- and 3-digit levels of ISIC Code (revisions 3 \& 4) called ``MINSTAT - Mining and Utilities Statistics Database''. MINSTAT rev.~3 covers years from 1990 and the country coverage ranges from 1 to 67, while rev.~4 data starts in 2005 with more stable country coverage, which varies from 17 to 50. In combination, revisions 3 and 4 of the MINSTAT database combine to cover a total of 110 countries. In order to take full advantage of the MINSTAT database and to ensure maximum coverage in terms of years, countries, and industries, we create a concordance between ISIC rev.~3 and ISIC rev.~4, which appears in Table \protect\hyperlink{tab:ITPD-E_class_mining}{{[}tab:ITPD-E\_class\_mining{]}}, together with the ITPD-E mining classification. As can be seen from this table, we have to aggregate some industries up in each of the two revisions of the ISIC classifications and also to construct some residual values as the difference between aggregate values and some subcategories. In order to obtain maximum number of non-missing observations, our approach is to combine sub-categories by replacing missing values with zeroes for a given sub-category which we use in the aggregation, if there are non-missing corresponding values in any of the other sub-categories that were used in the same aggregation. One motivation for this approach is that often countries report values either in one sub-category or in the other.\footnote{We also experimented with a conservative approach, where missing values in any subcategory are treated as missing values and, as a result, the corresponding aggregate category is also missing. In other words, a non-missing aggregate value is obtained only for the cases where all corresponding sub-categories were not missing. The difference between the number of observations that we obtain with the two alternative procedures was relatively small in favor of the more liberal aggregation approach.} Once we aggregate revisions 3 and 4 to the new ITPD-E concordance, we combine the data from the two ISIC revisions using the data in revision 3 as the baseline due to the larger original number of observations in this revision. This results in 1,822 observations and completes the construction of the production data for mining. We then use the procedure described in Section \protect\hyperlink{sec:constr_of_dom_trade_data}{{[}sec:constr\_of\_dom\_trade\_data{]}} to construct domestic trade flows using the national gross production data and the data on international trade described in the previous section. \hypertarget{manufacturing}{% \subsection{Manufacturing}\label{manufacturing}} The manufacturing broad sector in ITPD-E consists of 120 industries that cover products that are part of ISIC rev.~4 section C, which includes ``the physical or chemical transformation of materials, substances, or components into new products''. Note that manufacturing includes the processing of the products of agriculture into food for humans or animals. It also includes repair and installation of machinery and equipment. The list of manufacturing industries in ITPD-E can be found in Table \protect\hyperlink{tab:ITPD-E_class_manuf}{5}. \hypertarget{trade-data-2}{% \subsubsection{Trade Data}\label{trade-data-2}} Manufacturing international trade data come from the United Nations International Trade Statistics Database (COMTRADE), which is the most comprehensive source for bilateral data on merchandise trade flows.\footnote{\url{https://comtrade.un.org}.} International trade data was obtained in ISIC rev.~4 to match the classification of the production data described below. \hypertarget{production-data-2}{% \subsubsection{Production Data}\label{production-data-2}} The primary source for industry-level manufacturing production data is the United Nations Industrial Statistics (INDSTAT) database, which exists in two versions, one in ISIC rev.~3 and one in ISIC rev.~4.\footnote{\url{http://stat.unido.org}.} The latest edition of the INDSTAT database in rev.~3 covers the period from 1985 onwards for 138 countries at the 3- and 4-digit levels, which includes 151 manufacturing industries. INDSTAT in rev.~4 covers the period from 2005 onwards for 89 countries at the 3- and 4-digit levels, which includes 161 industries. It should be noted that even though ISIC rev.~4 replaced ISIC rev.~3 since 2005, the ISIC rev.~3 version of the INDSTAT data includes observations for the years 2005-2014 and we took advantage of these data to ensure maximum coverage. As with other original data sources, INDSTAT coverage in terms of years and industries varies from country to country depending on data availability. In order to ensure maximum coverage, we construct the raw manufacturing production data in two steps. First, for each revision of ISIC we combine the output data at the 3-digit level with the data at the 4-digit level to minimize the number of missing observations. Then, we combine the data from the two ISIC revisions by aggregating the two 4-digit ISIC classifications up in order to ensure matching. To describe our decisions on how to perform the first step (i.e., where we use the 3-digit data to fill in missing 4-digit values within each revision), we use the case of Colombia. Colombia is chosen because it offers good examples of possible cases that should be considered. For example, we have all 4-digit data that corresponds to the 3-digit Sector 281. When summed, the 4-digit values add to the corresponding 3-digit value, as expected. Next, consider Sector 154 for which we have data on all but one of the 4-digit categories. The missing category is 1544. However, when we add the non-missing values, they sum exactly to the corresponding 3-digit total. This may be for three reasons: (i) because the missing value for 1544 is zero, (ii) because 1544 was aggregated with another 4-digit sector in the data (this is often indicated by a flag in the INDSTAT database), or (iii) because to get the total, the reporters or the creators of the data have simply added the non-missing 4-digit. We believe that we can exclude option (iii). To see this consider the next industry in the example. Sector 191 includes two 4-digit industries, one is missing and one is not. However, the total at the 3-digit level is different from the total in the non-missing 4-digit industry. This is evidence that the reporters/developers of the data did not simply add up the non-missing 4-digit values in order to obtain the corresponding 3-digit data. This also means that, in this case, we can recover the single missing 4-digit category as the difference from the 3-digit value and the non-missing 4-digit value. Note that the same procedure can be applied to recover missing 4-digit values in any industry for which we have the 3-digit value and only one missing 4-digit value. Next, consider Sector 155, where we have two missing values at the 4-digit level and a non-missing 3-digit value. In this case we cannot recover the missing 4-digit values. We apply the procedure that we just described separately to the ISIC rev.~3 data and to the ISIC rev.~4 data. In the case of the ISIC rev.~3 data we recover 1301 values, of which 175 are negative and set to missing. In the case of the ISIC rev.~4 data we recover 948 missing 4-digit values, of which 59 are negative and set to missing. Once we fill in the missing 4-digit values in each ISIC revision as described above, we concord them to the ITDP-E classification. The concordance appears in Table \protect\hyperlink{tab:ITPD-E_class_manuf}{5}. As can be seen from this table, the correspondence between the two ISIC revisions is overall quite good, and only a few aggregations up are needed. Specifically, we only need to aggregate two 4-digit industries in ISIC rev.~3, 3312+3313, and there were only eleven cases in which we have to aggregate in ISIC rev.~4. These include, 1075+1079, 1311+1312, 1430+1391, 2029+2680, 2812+2813, 2818+2822, 2520+3040, 2620+2817, 2731+2732, 2660+3250, and 3211+3212. Consistent with our approach for agriculture and mining, and in order to obtain maximum number of non-missing observations, we replace missing values with zeroes for a given sub-category, which we use in the aggregation, if there are non-missing corresponding values in any of the other sub-categories that were used in the same aggregation.\footnote{We also experimented with a conservative approach, where missing values in any sub-category are treated as missings and, as a result, the corresponding aggregate category is also missing. The difference between the number of observations that we obtain with the two alternative procedures was relatively small.} After the above procedures, we have manufacturing production data for 146 countries. We combine the production data with trade trade to calculate domestic trade, as explained in Section \protect\hyperlink{sec:constr_of_dom_trade_data}{{[}sec:constr\_of\_dom\_trade\_data{]}}. In a small number of cases we had output data, but not total exports. \hypertarget{services}{% \subsection{Services}\label{services}} The Services broad sector includes 17 industries that perform the activities classified as services. It includes the activities in ISIC rev.~4 sections F through K, M, N, and P through S. The mostly non-traded activities contained in sections L ``Real Estate'', O ``Government'', T ``Household production'', and U ``Extraterritorial organizations'' are not included in this broad sector. \hypertarget{trade-data-3}{% \subsubsection{Trade Data}\label{trade-data-3}} Bilateral trade data on services are still relatively limited, and its quality is widely viewed as not on par with trade statistics for merchandise goods. The reasons can be seen in the inherent difficulty of reliably recording intangible and non-storable service trade flows. Nevertheless, over recent years significant progress has been made to offer better services trade statistics. Drawing on the best and most comprehensive statistics available, the services trade data in the ITPD-E come from the `WTO-UNCTAD-ITC Annual Trade in Services Database' and the UN `Trade in Services Database' (UN TSD).\footnote{The UN TSD and the WTO-UNCTAD-ITC datasets are both Balance of Payments statistics and as such capture Modes of Supply 1, 2, and 4 and exclude Mode 3 (sales of foreign affiliates).} We take the UN Trade in Services database in the form currently incorporated into Comtrade.\footnote{\url{https://comtrade.un.org}.} We use this dataset to obtain services trade statistics for the early years 2000-2004, since the WTO-UNCTAD-ITC dataset only commences in 2005. However, we also retain data for subsequent years (post-2005) from the UN TSD because there is considerable scope for blending both datasets to achieve maximum coverage. This is partly because the WTO-UNCTAD-ITC dataset formally commences in 2005 but during the early years many countries only report \emph{World} exports, rather than bilateral flows, and broad-based reporting at the bilateral level only commences after 2010. We note that the raw data contain a small number of negative export flows, which tend to be concentrated in the insurance (reinsurance) industry as well as in merchanting, and thus may reflect the particular accounting principles in these industries. Following the basic principles of ITPD-E construction outlined in Sections \protect\hyperlink{sec:constr_of_int_trade_data}{{[}sec:constr\_of\_int\_trade\_data{]}} and \protect\hyperlink{sec:constr_of_dom_trade_data}{{[}sec:constr\_of\_dom\_trade\_data{]}}, we set these negative international trade values to missing. Statistics on cross-border services trade are reported using the Extended Balance of Payments Services classification (EBOPS). The fifth edition of the Balance of Payments and International Investment Position Manual (BPM5), released in 1993, employs the EBOPS 2002 classification, whereas the latest sixth edition of the Manual (BPM6) employs the EBOPS 2010 classification, which provides the most recent and detailed services trade statistics. The trade data in UN TSD are organized according to the EBOPS 2002 classification, which we concord to EBOPS 2010 as this newer classification ensures the best level of industry-level granularity in services trade. This is also consistent with the fact that more and more countries switch their reporting from BPM5 (EBOPS 2002) to BPM6 (EBOPS 2010). Most items are in principle straightforward to map across the two classifications, and the concordance used for ITPD-E, which follows \cite{Wettsteinetal2019}, is provided in Table~\protect\hyperlink{tab:ITPD-E_class_services}{{[}tab:ITPD-E\_class\_services{]}}.\footnote{One major difference between BPM5 and BPM6 is the treatment of manufacturing services on physical inputs owned by others and repairs of goods. As a result, the EBOPS 2002 code for Merchanting (``270'') as part of `Other business services' does not have a correspondence in BPM6; conversely, the EBOPS 2010 codes for manufacturing services and maintenance (``SA'' and ``SB'') cannot be concorded backwards to BPM5.} After various data procedures including mirroring, we obtain approximately 280,000 observations from the UN TSD. The WTO-UNCTAD-ITC dataset\footnote{\href{https://www.wto.org/english/res_e/statis_e/trade_datasets_e.htm}{https://www.wto.org/english/res\_e/sta tis\_e/trade\_datasets\_e.htm}.} includes data on bilateral trade in services classified according to EBOPS 2010. Countries are identified by UN 2-digit alphanumeric codes, and we apply a mapping to supplement ISO 3-digit alphanumeric country codes. Nearly three-quarters of the observations (73.5\%) in the WTO-UNCTAD-ITC dataset are sourced from Eurostat. This is in contrast to the UN TSD dataset, which draws on different sources. After various procedures including mirroring, we obtain approximately 300,000 observations from the WTO-UNCTAD-ITC dataset, of which 130,000 are reported zeroes. Yet before trade statistics from the WTO-UNCTAD-ITC and the UN TSD databases are combined, we perform hierarchical consistency checks and mirroring within each constitutent dataset. Regarding the former, EBOPS is a hierarchical classification and so there is scope for replenishing or updating higher-level entries with sums of reported bilateral trade flows from lower-level items. Specifically, we carry out vertical consistency check across the two EBOPS categories `SI' (telecommunications, computer and information services) and `SJ' (Other business services). Regarding the mirroring protocol, as explained in Section \protect\hyperlink{sec:constr_of_int_trade_data}{{[}sec:constr\_of\_int\_trade\_data{]}}, we use reported exports as the main source of bilateral trade in services. To maximize coverage, we use mirrored import flows from reporting countries to fill in missing export values. A flag is retained in the database to identify values that derive from mirroring. Both services trade flow datasets---UN TSD and WTO-UNCTAD-ITC---are then merged to obtain the best possible coverage of bilateral services trade. The contribution from each principal source in terms of observation counts is documented in Table \protect\hyperlink{tab:serv_joint_cov}{6}, which suggests substantial gains from blending these two mirrored datasets. We emphasize that the WTO-UNCTAD-ITC dataset remains the primary source of service trade data, meaning that raw data from this dataset are never overwritten or replaced. This is because the WTO data of later years appear to be more comprehensive, and possibly of higher quality, than the data of earlier years.\footnote{For instance, the value share of reported trade in Travel, Transport and Other Commercial Services (as opposed to non-tradable, unallocated, and Government services) rises continuously throughout the years covered by ITPD-E and reaches 85\% in 2016.} At the same time, trade values replenished from the UN TSD account for 38\% of the combined services dataset. Of these additional non-missing observations, 31\% pertain to the initial years 2000-04 that are not covered by the WTO-UNCTAD-ITC dataset at all. Overall, while the nature of services trade raw data remains qualitatively unchanged, the data in ITPD-E are appreciably enhanced by blending two principal sources of statistics, a conservative use of mirroring techniques, and casting all services trade data in the latest EBOPS 2010 format, suitably grouped for a correspondence to services sections in the ISIC rev.~4 industry classification. \hypertarget{production-data-3}{% \subsubsection{Production Data}\label{production-data-3}} We obtain information on gross output at basic prices from the UN National Accounts database using the ISIC rev.~4 classification.\footnote{\url{http://data.un.org/Explorer.aspx?d=SNA}.} Production data from the UN National Accounts requires two principal modifications before it can be merged with trade flow data: first, gross output data as obtained from the UN Statistical Division are denominated in local currency units. Since the conversion of services production statistics should ideally use the same USD exchange rates that are used for services trade flow data, we apply the average annual exchanges rates from the IMF's International Financial Statistics series. Second, in terms of statistical concepts for output data, information is reported in two different frameworks within the ISIC rev.~4 industrial classification, namely SNA 1993 and SNA 2008, respectively.\footnote{It should be noted that a small number of mainly emerging economies are not included in the UN SNA data. Most of these economies do not report gross output statistics to the UN at all, neither in ISIC rev.~3 nor in rev.~4, and so there is no obvious solution within the realm of UN statistics. Examples include Australia, China, Indonesia, Malaysia, Russia, Thailand or Turkey.} The UN data show that some countries report only in the former, some only in the latter, and some countries report in both frameworks, at least for some intermittent years as their reporting transitions from SNA 1993 to SNA 2008. We blend data from both SNA frameworks so as to maximize coverage in terms of countries and industries.\footnote{We start with all gross output data reported in SNA 2008 but retain additional information by mapping data reported in SNA 1993 into SNA 2008. On the basis of a comparison of differences in values for industries in which gross output is reported simultaneously in SNA 1993 and SNA 2008, it was decided to keep SNA 1993 output information in industries in which the observed difference is less than 10 percent on average.} In preparation for merging trade and gross output information, the last step involves the construction of a mapping from ISIC rev.~4 to EBOPS 2010, shown in Table \protect\hyperlink{tab:ITPD-E_class_services}{{[}tab:ITPD-E\_class\_services{]}}. This trade to output concordance follows---with very minor exceptions---the EBOPS-ISIC bridge table used in the methodological paper for the WTO's ``Trade in Services data by mode of supply (TISMOS)'' dataset \citep[Table 14]{Wettsteinetal2019}.\footnote{For instance, the regrouping involves the combination of trade in audiovisual services with trade in telecom and information services, or likewise the combination of trade in educational services with that part of travel services that is undertaken for educational purposes.} As mentioned in the previous subsection, not all product codes are amenable to concording; in particular, five EBOPS 2010 codes---SA (manufacturing services on physical inputs), SB (maintenance and repair services), SH (charges for intellectual property rights), SL (government goods and services) and SN (services not allocated)---have no meaningful correspondence in an ISIC industry classification. Yet in order to maximize information available to users, trade values recorded under these five items are retained in ITPD-E; however, no internal trade can be constructed for these categories because output is undefined on conceptual grounds. We combine the production data with trade trade to calculate domestic trade in services as explained in Section \protect\hyperlink{sec:constr_of_dom_trade_data}{{[}sec:constr\_of\_dom\_trade\_data{]}}. To summarize the industry coverage for domestic trade in services, out of 17 service industries in ITPD-E, five industries do not have any domestic trade data because they have no equivalent production data, and two industries have domestic trade starting in 2005 because these categories (arts, entertainment, recreational and other personal services) were newly introduced in EBOPS 2010 and do not have corresponding items in EBOPS 2002. Hence, total exports cannot be calculated prior to 2005 for these industries. Overall, after various cleaning and concording procedures, we obtain 12,260 observations for services gross output from UN SNA. \hypertarget{sec_comparisoncomparison-with-other-trade-and-production-data--sets}{% \section[Comparison with Other Trade and Production Data- sets]{\texorpdfstring{\protect\hypertarget{sec_comparison}{}{{[}sec\_comparison{]}}Comparison with Other Trade and Production Data- sets}{{[}sec\_comparison{]}Comparison with Other Trade and Production Data- sets}}\label{sec_comparisoncomparison-with-other-trade-and-production-data--sets}} In this section we compare ITPD-E to other trade and production datasets. Just as ITPD-E, each of the databases referenced here are products of great efforts to construct consistent data on international and domestic trade flows, and each of them has been used in a number of influential academic and policy papers. A common feature across all databases is that each of them includes international and domestic trade flows. Below, we compare ITPD-E with each alternative dataset across the following criteria: (i) industry coverage; (ii) country coverage; (iii) time coverage; and (iv) suitability for estimations. \begin{itemize} \item \textbf{World Bank's TPP.} The World Bank's Trade, Production and Protection (TPP) database\footnote{\url{https://datacatalog.worldbank.org/dataset/trade-production-and-protection-database}. The database has been developed by Nicita, A.~and Olarreaga, M., 2007, ``Trade, Production, and Protection Database, 1976--2004'', \emph{World Bank Economic Review} \textbf{21}(1), pp.~165--171.} covers approximately 100 countries, for the period 1976-2004 where information is available in ISIC rev.~3 at the 3-digit level. Similarly to the ITPD-E, the World Bank's TPP is constructed from reported administrative data and, therefore, is suitable for estimations. In addition, the TPP covers a longer time span. However, it has been discontinued and data for the years post 2004 are not included in it. In terms of country coverage, the ITPD-E includes significantly more countries. Finally another important advantage of the ITPD-E is that it is designed to cover the complete economy, i.e., all industries, while the TPP is limited to manufacturing industries only. \item \textbf{CEPII's TradeProd.} CEPII's Trade, Production and Bilateral Protection (TradeProd) database\footnote{\url{http://www.cepii.fr/CEPII/fr/bdd_modele/presentation.asp?id=5}.} covers manufacturing (at ISIC rev.~2 at the 3-digit level) for over 150 countries during the period 1980-2006. Similarly to the ITPD-E, the CEPII's TradeProd is constructed from reported administrative data and, therefore, is suitable for estimations. In addition, the TradeProd covers a longer time span than ITPD-E. However, it has been discontinued and data for the years post 2006 are not included in it. In terms of country coverage, the ITPD-E includes significantly more countries than TradeProd. Finally, another important advantage of the ITPD-E is that it is designed to cover the complete economy, i.e., all industries, while the TradeProd is limited to manufacturing industries only. \item \textbf{GTAP Dataset.} The GTAP Dataset 10 covers the full economies of 121 countries and 20 country aggregates.\footnote{\url{https://www.gtap.agecon.purdue.edu/databases/v10/}.} It includes data for the years 2004, 2007, 2011, and 2014. It covers 65 industries that are classified according to the GTAP sector classification. Thus, the GTAP dataset and the ITPD-E both cover all industries within an economy. However, the ITPD-E has a wider country, sector, and time coverage.\footnote{GTAP also provides a Bilateral Time Series Trade Data for all regions in GTAP 10 with a wider time coverage (1995 to 2016), but only for merchandise trade.} In addition, the ITPD-E offers consistent panel data. The key difference between the GTAP dataset and the ITPD-E is that the former is suitable for simulations while the latter is suitable for statistical estimations. Specifically, the GTAP dataset relies on economic models to estimate missing data so that it can offer a fully balanced dataset that can be used for simulation analysis but not for estimations. On the other hand, the ITPD-E is not fully balanced and cannot be used for simulations. The key advantage of ITPD-E is that it is constructed from reported administrative data, which makes it suitable for estimation purposes. \item \textbf{WIOD.} The latest edition (Release 2016) of the World Input-Output Database (WIOD) covers the period 2000-2014 for 56 industries.\footnote{\url{http://www.wiod.org/home}.} Similarly to the GTAP dataset and the ITPD-E, the WIOD dataset covers the entire economies. In terms of coverage, the ITPD-E has similar time coverage but a significantly wider industry, and especially country coverage compared to the WIOD. Another key difference between the WIOD dataset and the ITPD-E is that the former is suitable for simulations while the latter is suitable for estimations. Specifically, the WIOD dataset relies on economic models to estimate missing data so that it can offer a fully balanced dataset that can be used for simulation analysis but not for estimations. On the other hand, the ITPD-E is not fully balanced and cannot be used for simulations. However, the key advantage of ITPD-E is that it is constructed from reported administrative data, which makes it suitable for estimation purposes. \end{itemize} To summarize, the ITPD-E offers superior country and industry coverage compared to other existing trade and production databases. It covers non-manufacturing industries, includes recent years, and has data for nearly all countries in the world. Importantly, unlike several of the datasets reviewed above, ITPD-E is suitable for estimation of economic models because it is constructed from reported administrative data and does not include ``data'' estimated using statistical techniques and economic models. \hypertarget{sectoral-gravity-estimates-itpd-e-validationsec_validation}{% \section[Sectoral Gravity Estimates: ITPD-E Validation]{\texorpdfstring{Sectoral Gravity Estimates: ITPD-E Validation\protect\hypertarget{sec_validation}{}{{[}sec\_validation{]}}}{Sectoral Gravity Estimates: ITPD-E Validation{[}sec\_validation{]}}}\label{sectoral-gravity-estimates-itpd-e-validationsec_validation}} This section demonstrates that the ITPD-E is suitable for disaggregated gravity estimations. Due to space constraints, our objective is merely to offer a proof of concept by using ITPD-E to obtain a set of basic estimates for each ITPD-E industry.\footnote{For a thorough gravity analysis with the ITPD-E, we refer the reader to \cite{Borchertetal2020}, who validate the use of the ITPD-E by implementing the latest developments in the empirical gravity literature, as summarized by \cite{Yotovetal2016}, and who provide a discussion of a series of stylized facts and best practice recommendations for gravity estimations in the light of the detailed industry-level estimates.} To this end, we employ the traditional (and still very widely used) log-linear gravity specification, which we estimate with the OLS estimator. We also take advantage of the separability of the structural gravity model to obtain estimates for each of the 170 ITPD-E industries. Given our purpose to put ITPD-E to the test, we employ a set of standard gravity covariates in our estimations including the logarithm of bilateral distance between two countries, and indicator variables for contiguity, common official language, colonial relationships, and the presence of free trade agreements. Data on all gravity variables come from the Dynamic Gravity Dataset (DGD), which is constructed and maintained by the U.S. International Trade Commission (see \cite{USITCGravity}).\footnote{It is encouraging to note that the country coverage of ITPD-E is very close to the comprehensive coverage of the USITC's Dynamic Gravity Dataset. Specifically, DGD covers all 243 countries that appear in ITPD-E. More importantly, there are only 10 very small regions/territories that appear in DGD but do not appear in the ITPD-E. These include ALA (Aland Islands), GAZ (Gaza Strip), GGY (Guernsey), GLP (Guadeloupe), JEY (Jersey), KSV (Kosovo), MAF (Saint-Martin), MTQ (Martinique), REU (Reunion), and SJM (Svalbard and Jan Mayen).} An important advantage of the ITPD-E is that it includes both international and domestic trade flows. This offers an opportunity to estimate and compare border effects in a comprehensive way across all industries, i.e.~the trade-reducing impact of there being an international border, conditional on all other gravity forces including distance. We therefore also include a set of country-specific dummy variables (\(SMCTRY\)) that take a value of one for internal trade and zero for international trade. Finally, following \cite{AndersonvanWincoop2003}, all models include full sets of exporter-time and importer-time fixed effects to control for unobservable multilateral resistance terms. Due to the large number of gravity estimates (170 estimates for each of the 6 gravity variables in our specifications), we present our regression results graphically, using six panels/plots (one for each gravity covariate) in Figure \protect\hyperlink{fig1}{6}. Each dot within a panel represents the point estimate for a particular industry, and the estimates are ordered from the smallest to the largest estimate. Panel A of the figure shows that distance coefficient estimates are all negative and many of them sizable. Hence, our estimates confirm that distance is a strong impediment to trade. The estimates in Panel B reveal that sharing a common border promotes trade in all but one industry. Similarly, Panel C shows that in all but 4 industries the estimates of the impact of common official language on trade are positive. The estimates in Panel D capture the empirical regularity that countries with a colonial relationship trade more (only 8 of the estimates of colonial ties are negative). Further, in Panel E, we confirm that \(FTA\)'s promote trade between members in all but 7 industries. Finally, we find that international borders reduce international trade considerably, even after controlling for the impact of geography and other significant determinants of trade, as evidenced by the very large positive coefficients in Panel F of Figure \protect\hyperlink{fig1}{6}. These estimates confirm the finding of very substantial border barriers \citep{Andersonetal2018} for a much broader set of industries. We conclude this section with a robustness exercise that evaluates the impact of the mirror data used in ITPD-E on gravity estimations. To perform this exercise, we drop all observations for which the flag variable for mirroring is equal to one. Our findings are reported in Figure \protect\hyperlink{fig_mirr}{12}. Comparison between the estimates in Figures \protect\hyperlink{fig1}{6} and \protect\hyperlink{fig_mirr}{12} suggests that the mirrored trade flows do not play a very important role for common estimates of the standard gravity variables. Thus, the benefits of adding more data through mirroring do not seem to come at the cost of biasing the standard gravity results. Nevertheless, we recommend caution with the use of mirrored trade flows, especially if the focus is on heterogeneous gravity estimates. In sum, the industry-level gravity estimates that we obtain in this section are as expected and are readily comparable to existing estimates from the related literature. For example, in terms of magnitude and direction, our estimates are very close to the meta-analysis results from \cite{HeadMayer2014}. Thus, our results offer preliminary evidence that validates the use of the ITPD-E for gravity estimations. For a more thorough analysis of the suitability and implications of using the ITPD-E for gravity estimations, we refer the interested reader to \cite{Borchertetal2020}. \hypertarget{summarysec_conclusions}{% \section[Summary]{\texorpdfstring{Summary\protect\hypertarget{sec_conclusions}{}{{[}sec\_conclusions{]}}}{Summary{[}sec\_conclusions{]}}}\label{summarysec_conclusions}} Analyzing and quantifying trade policies requires consistent information on both international and domestic trade flows. While a few databases containing this information exist, they have significant limitations, such as sectoral coverage limited to manufacturing, limited country coverage, absence of recent updates, and use of estimated data in place of reported administrative data. The International Trade and Production Database for Estimation (ITPD-E) contains consistent data on domestic and international trade for 243 countries, 170 industries, and 17 years commencing from the year 2000. It provides data for four broad sectors of the economy: agriculture, mining, manufacturing and services. Due to this coverage, ITPD-E describes each economy nearly completely. As we only use reported administrative data and not statistical estimates to fill in database entries, the ITPD-E is well-suited for statistical inference such as the estimation of gravity models. We demonstrate this usefulness with standard gravity estimates. Further insights into estimating gravity from ITPD-E are offered in \cite{Borchertetal2020}. The ITPD-E is a public good for the benefit of researchers, civil servants and many others. In return for creating and maintaining this public good, we kindly ask that all users of the ITPD-E please cite this paper. Moreover, in an effort to continuously improve the coverage and quality of ITPD-E data, we welcome feedback including error reports to the ITC's gravity portal e-mail address (\texttt{gravity@usitc.gov}). Please visit \href{https://gravity.usitc.gov}{USITC's gravity portal} for updates. In the future, we plan to use the latest methodological advances in structural gravity modeling to project bilateral trade when such information is missing. This augmented version will constitute the second member within the family of ITPD databases, which should be useful for simulations and counterfactual analysis. Hence, we may call this version of the database the International Trade and Production Database for Simulation (ITPD-S). We also plan to offer a third database version within the ITPD family by recasting the trade flow information according to the industry structure of the well-known World Input-Output Database (WIOD). This will enable users of the data to employ the latest input-output tables in their analysis in order to allow for inter-industry links on the supply side. This third version may be called the International Trade and Production Database with Input-Output information (ITPD-IO). \hypertarget{ITPD-E_sectors}{} \begin{longtable}[]{@{}rlrrrrr@{}} \caption{ \protect\hypertarget{ITPD-E_sectors}{}{{[}ITPD-E\_sectors{]}} ITPD-E: Industry Coverage and Summary Statistics\\ for International and Domestic Trade Flows}\tabularnewline \toprule ID & Industry Description & Mean Exports & Max Exports & \#Observations & \#Zeroes &\tabularnewline \midrule \endfirsthead \toprule ID & Industry Description & Mean Exports & Max Exports & \#Observations & \#Zeroes &\tabularnewline \midrule \endhead ID & Industry Description & Mean Exports & Max Exports & \#Observations & \#Zeroes &\tabularnewline & & & & & &\tabularnewline & Wheat & 84.22 & 50955 & 77845 & 50305 &\tabularnewline 2 & Rice (raw) & 204.44 & 117214 & 63997 & 45094 &\tabularnewline 3 & Corn & 78.60 & 106817 & 87885 & 53488 &\tabularnewline 4 & Other cereals & 28.68 & 11918 & 81678 & 49366 &\tabularnewline 5 & Cereal products & 1.13 & 193 & 39654 & 26564 &\tabularnewline 6 & Soybeans & 68.96 & 21637 & 55957 & 36540 &\tabularnewline 7 & Other oilseeds (excluding peanuts) & 38.57 & 95962 & 160283 & 89937 &\tabularnewline 8 & Animal feed ingredients and pet foods & 2.76 & 883 & 71448 & 43299 &\tabularnewline 9 & Raw and refined sugar and sugar crops & 340.37 & 45732 & 10651 & 7503 &\tabularnewline 10 & Other sweeteners & 2.80 & 1108 & 82972 & 50552 &\tabularnewline 11 & Pulses and legumes, dried, preserved & 9.26 & 13545 & 134352 & 76837 &\tabularnewline 12 & Fresh fruit & 60.53 & 189855 & 211150 & 114823 &\tabularnewline 13 & Fresh vegetables & 114.45 & 323976 & 175575 & 99368 &\tabularnewline 14 & Prepared fruits and fruit juices & 0.71 & 235 & 106861 & 62526 &\tabularnewline 15 & Prepared vegetables & 3.21 & 1717 & 16594 & 10079 &\tabularnewline 16 & Nuts & 11.95 & 15018 & 127163 & 72949 &\tabularnewline 17 & Live Cattle & 9.16 & 1796 & 31588 & 19813 &\tabularnewline 18 & Live Swine & 7.88 & 1207 & 16511 & 9728 &\tabularnewline 19 & Eggs & 66.85 & 60757 & 65981 & 42101 &\tabularnewline 20 & Other meats, livestock products, and live animals & 3.62 & 6513 & 173954 & 100797 &\tabularnewline 21 & Cocoa and cocoa products & 9.91 & 1121 & 38609 & 24716 &\tabularnewline 22 & Beverages, nec & 5.51 & 10386 & 192290 & 105467 &\tabularnewline 23 & Cotton & 20.17 & 27391 & 99474 & 62499 &\tabularnewline 24 & Tobacco leaves and cigarettes & 7.04 & 9484 & 104397 & 62538 &\tabularnewline 25 & Spices & 2.44 & 3958 & 186919 & 99077 &\tabularnewline 26 & Other agricultural products, nec & 7.52 & 11296 & 271908 & 132358 &\tabularnewline 27 & Mining of hard coal & 126.19 & 445807 & 75090 & 48800 &\tabularnewline 28 & Mining of lignite & 31.80 & 36894 & 19707 & 13874 &\tabularnewline 29 & Extraction crude petroleum and natural gas & 779.50 & 166676 & 92441 & 61933 &\tabularnewline 30 & Mining of iron ores & 142.48 & 119661 & 45156 & 30533 &\tabularnewline 31 & Other mining and quarring & 38.54 & 173194 & 306736 & 154467 &\tabularnewline 32 & Electricity production, collection, and distribution & 1657.40 & 711982 & 21438 & 13061 &\tabularnewline 33 & Gas production and distribution & 442.79 & 82946 & 17634 & 12996 &\tabularnewline 34 & Processing/preserving of meat & 78.49 & 310942 & 264080 & 146761 &\tabularnewline 35 & Processing/preserving of fish & 15.58 & 38660 & 304756 & 159213 &\tabularnewline 36 & Processing/preserving of fruit \& vegetables & 14.73 & 52296 & 333462 & 166554 &\tabularnewline 37 & Vegetable and animal oils and fats & 28.28 & 116795 & 266236 & 138783 &\tabularnewline 38 & Dairy products & 42.39 & 94969 & 253217 & 136888 &\tabularnewline 39 & Grain mill products & 18.02 & 125311 & 270470 & 146590 &\tabularnewline 40 & Starches and starch products & 6.89 & 36249 & 183214 & 98206 &\tabularnewline 41 & Prepared animal feeds & 30.04 & 111443 & 169836 & 90475 &\tabularnewline 42 & Bakery products & 25.36 & 57461 & 252274 & 131157 &\tabularnewline 43 & Sugar & 14.56 & 16695 & 188988 & 110896 &\tabularnewline 44 & Cocoa chocolate and sugar confectionery & 11.65 & 20310 & 280421 & 141973 &\tabularnewline 45 & Macaroni noodles \& similar products & 6.72 & 38863 & 184870 & 103418 &\tabularnewline 46 & Other food products n.e.c. & 21.51 & 112924 & 374789 & 183580 &\tabularnewline 47 & Distilling rectifying \& blending of spirits & 14.33 & 69585 & 246847 & 133915 &\tabularnewline 48 & Wines & 8.81 & 17195 & 213716 & 117754 &\tabularnewline 49 & Malt liquors and malt & 18.45 & 23818 & 179923 & 101884 &\tabularnewline 50 & Soft drinks; mineral waters & 17.70 & 58983 & 257976 & 140498 &\tabularnewline 51 & Tobacco products & 26.13 & 102640 & 235124 & 134336 &\tabularnewline 52 & Textile fibre preparation; textile weaving & 21.34 & 246517 & 344835 & 171213 &\tabularnewline 53 & Made-up textile articles except apparel & 6.36 & 17651 & 391348 & 192908 &\tabularnewline 54 & Carpets and rugs & 3.46 & 16354 & 265513 & 143581 &\tabularnewline 55 & Cordage rope twine and netting & 0.96 & 1331 & 234543 & 125027 &\tabularnewline 56 & Other textiles n.e.c. & 6.50 & 20764 & 307153 & 153166 &\tabularnewline 57 & Knitted and crocheted fabrics and articles & 10.42 & 42050 & 397698 & 196339 &\tabularnewline 58 & Wearing apparel except fur apparel & 23.01 & 91896 & 506144 & 238280 &\tabularnewline 59 & Dressing \& dyeing of fur; processing of fur & 1.63 & 6379 & 130653 & 74210 &\tabularnewline 60 & Tanning and dressing of leather & 6.78 & 22684 & 184452 & 99312 &\tabularnewline 61 & Luggage handbags etc.; saddlery \& harness & 4.66 & 8285 & 366600 & 186902 &\tabularnewline 62 & Footwear & 12.06 & 43163 & 372876 & 196295 &\tabularnewline 63 & Sawmilling and planing of wood & 17.96 & 31578 & 247071 & 134990 &\tabularnewline 64 & Veneer sheets plywood particle board etc. & 12.86 & 78298 & 224925 & 121628 &\tabularnewline 65 & Builders' carpentry and joinery & 14.97 & 46014 & 209825 & 116001 &\tabularnewline 66 & Wooden containers & 3.42 & 7112 & 195088 & 107693 &\tabularnewline 67 & Other wood products; articles of cork/straw & 3.06 & 16621 & 331583 & 170806 &\tabularnewline 68 & Pulp paper and paperboard & 30.99 & 106969 & 301621 & 150070 &\tabularnewline 69 & Corrugated paper and paperboard & 16.37 & 66759 & 283916 & 142817 &\tabularnewline 70 & Other articles of paper and paperboard & 8.35 & 20824 & 374885 & 184797 &\tabularnewline 71 & Publishing of books and other publications & 3.20 & 8351 & 378505 & 195730 &\tabularnewline 72 & Publishing of newspapers journals etc. & 13.34 & 26845 & 175255 & 99674 &\tabularnewline 73 & Publishing of recorded media & 2.82 & 1806 & 169299 & 102308 &\tabularnewline 74 & Other publishing & 1.22 & 2403 & 330683 & 168921 &\tabularnewline 75 & Printing & 18.32 & 97399 & 373622 & 181414 &\tabularnewline 76 & Service activities related to printing & 9.89 & 9353 & 104597 & 58065 &\tabularnewline 77 & Reproduction of recorded media & 22.95 & 20472 & 33613 & 21363 &\tabularnewline 78 & Coke oven products & 314.59 & 210261 & 53512 & 34302 &\tabularnewline 79 & Refined petroleum products & 158.84 & 689720 & 319655 & 169260 &\tabularnewline 80 & Processing of nuclear fuel & 12.04 & 11806 & 81830 & 48410 &\tabularnewline 81 & Basic chemicals except fertilizers & 60.20 & 225749 & 367995 & 173250 &\tabularnewline 82 & Fertilizers and nitrogen compounds & 22.32 & 99179 & 197467 & 105856 &\tabularnewline 83 & Plastics in primary forms; synthetic rubber & 41.85 & 158036 & 296165 & 146461 &\tabularnewline 84 & Pesticides and other agro-chemical products & 7.17 & 27703 & 214444 & 108127 &\tabularnewline 85 & Paints varnishes printing ink and mastics & 12.84 & 55007 & 295304 & 150218 &\tabularnewline 86 & Pharmaceuticals medicinal chemicals etc. & 57.96 & 207183 & 380241 & 180919 &\tabularnewline 87 & Soap cleaning \& cosmetic preparations & 17.09 & 87468 & 388759 & 193025 &\tabularnewline 88 & Other chemical products n.e.c. & 21.14 & 227393 & 435301 & 210509 &\tabularnewline 89 & Man-made fibres & 12.97 & 92990 & 187332 & 96573 &\tabularnewline 90 & Rubber tyres and tubes & 12.85 & 48966 & 275980 & 138581 &\tabularnewline 91 & Other rubber products & 8.35 & 34360 & 389019 & 192866 &\tabularnewline 92 & Plastic products & 40.14 & 197074 & 482105 & 225038 &\tabularnewline 93 & Glass and glass products & 15.62 & 74879 & 371857 & 182357 &\tabularnewline 94 & Pottery china and earthenware & 3.70 & 26673 & 323151 & 165148 &\tabularnewline 95 & Refractory ceramic products & 5.62 & 46413 & 163078 & 86853 &\tabularnewline 96 & Struct.non-refractory clay; ceramic products & 8.12 & 73291 & 234859 & 125520 &\tabularnewline 97 & Cement lime and plaster & 27.82 & 147585 & 167892 & 100058 &\tabularnewline 98 & Articles of concrete cement and plaster & 32.21 & 101392 & 193650 & 107092 &\tabularnewline 99 & Cutting shaping \& finishing of stone & 6.94 & 22175 & 198141 & 109049 &\tabularnewline 100 & Other non-metallic mineral products n.e.c. & 9.22 & 69710 & 275469 & 137026 &\tabularnewline 101 & Basic iron and steel & 98.27 & 964575 & 375386 & 185293 &\tabularnewline 102 & Basic precious and non-ferrous metals & 68.60 & 517119 & 331416 & 161768 &\tabularnewline 103 & Casting of iron and steel & 10.03 & 83981 & 264452 & 138221 &\tabularnewline 104 & Structural metal products & 32.57 & 109832 & 288356 & 149921 &\tabularnewline 105 & Tanks reservoirs and containers of metal & 7.73 & 13955 & 223679 & 122020 &\tabularnewline 106 & Steam generators & 11.69 & 18316 & 146238 & 83301 &\tabularnewline 107 & Cutlery hand tools and general hardware & 9.16 & 33652 & 431422 & 212001 &\tabularnewline 108 & Other fabricated metal products n.e.c. & 19.75 & 130876 & 483221 & 227877 &\tabularnewline 109 & Engines \& turbines (not for transport equipment) & 19.06 & 33046 & 225913 & 122809 &\tabularnewline 110 & Pumps compressors taps and valves & 18.31 & 73509 & 409150 & 197870 &\tabularnewline 111 & Bearings gears gearing \& driving elements & 9.91 & 37749 & 337549 & 171450 &\tabularnewline 112 & Ovens furnaces and furnace burners & 3.08 & 2852 & 197186 & 104739 &\tabularnewline 113 & Lifting and handling equipment & 15.24 & 63518 & 304791 & 156407 &\tabularnewline 114 & Other general purpose machinery & 23.37 & 152751 & 436553 & 210243 &\tabularnewline 115 & Agricultural and forestry machinery & 12.47 & 25953 & 248332 & 128833 &\tabularnewline 116 & Machine tools & 13.96 & 50621 & 349097 & 178263 &\tabularnewline 117 & Machinery for metallurgy & 6.47 & 24558 & 120723 & 66468 &\tabularnewline 118 & Machinery for mining \& construction & 16.89 & 147674 & 334163 & 173726 &\tabularnewline 119 & Food/beverage/tobacco processing machinery & 5.17 & 12330 & 255581 & 131232 &\tabularnewline 120 & Machinery for textile apparel and leather & 4.87 & 17778 & 268892 & 141765 &\tabularnewline 121 & Weapons and ammunition & 8.21 & 17551 & 148226 & 85899 &\tabularnewline 122 & Other special purpose machinery & 23.00 & 124321 & 371734 & 185338 &\tabularnewline 123 & Domestic appliances n.e.c. & 16.87 & 90216 & 367690 & 187869 &\tabularnewline 124 & Office accounting and computing machinery & 42.14 & 117097 & 448807 & 221711 &\tabularnewline 125 & Electric motors generators and transformers & 20.24 & 154165 & 457444 & 216303 &\tabularnewline 126 & Electricity distribution \& control apparatus & 17.31 & 80183 & 407076 & 198665 &\tabularnewline 127 & Insulated wire and cable & 12.39 & 166781 & 356089 & 179145 &\tabularnewline 128 & Accumulators primary cells and batteries & 6.74 & 71609 & 293490 & 149727 &\tabularnewline 129 & Lighting equipment and electric lamps & 7.21 & 25120 & 373200 & 187185 &\tabularnewline 130 & Other electrical equipment n.e.c. & 15.30 & 37101 & 416978 & 202832 &\tabularnewline 131 & Electronic valves tubes etc. & 84.59 & 295224 & 350806 & 177572 &\tabularnewline 132 & TV/radio transmitters; line comm. apparatus & 37.07 & 92857 & 401113 & 200155 &\tabularnewline 133 & TV and radio receivers and associated goods & 26.20 & 97966 & 417532 & 211872 &\tabularnewline 134 & Medical surgical and orthopaedic equipment & 17.71 & 83503 & 379228 & 184182 &\tabularnewline 135 & Measuring/testing/navigating appliances etc. & 18.55 & 82216 & 420499 & 199833 &\tabularnewline 136 & Optical instruments \& photographic equipment & 11.73 & 23638 & 326220 & 169262 &\tabularnewline 137 & Watches and clocks & 4.94 & 5179 & 267915 & 143415 &\tabularnewline 138 & Motor vehicles & 147.76 & 440148 & 362326 & 195277 &\tabularnewline 139 & Automobile bodies trailers \& semi-trailers & 11.16 & 25485 & 248525 & 137063 &\tabularnewline 140 & Parts/accessories for automobiles & 52.42 & 285369 & 427422 & 210741 &\tabularnewline 141 & Building and repairing of ships & 34.48 & 53927 & 181233 & 113507 &\tabularnewline 142 & Building/repairing of pleasure/sport. boats & 5.07 & 9713 & 155453 & 93268 &\tabularnewline 143 & Railway/tramway locomotives \& rolling stock & 17.86 & 42110 & 128073 & 72058 &\tabularnewline 144 & Aircraft and spacecraft & 44.77 & 84027 & 258794 & 144385 &\tabularnewline 145 & Motorcycles & 9.63 & 42095 & 223329 & 128282 &\tabularnewline 146 & Bicycles and invalid carriages & 3.09 & 3203 & 245497 & 137234 &\tabularnewline 147 & Other transport equipment n.e.c. & 0.93 & 2078 & 168167 & 95088 &\tabularnewline 148 & Furniture & 24.35 & 97079 & 412160 & 206453 &\tabularnewline 149 & Jewellery and related articles & 14.82 & 19996 & 302899 & 157922 &\tabularnewline 150 & Musical instruments & 1.50 & 2087 & 227752 & 128035 &\tabularnewline 151 & Sports goods & 3.81 & 11085 & 306837 & 162732 &\tabularnewline 152 & Games and toys & 7.70 & 21405 & 317275 & 168806 &\tabularnewline 153 & Other manufacturing n.e.c. & 7.92 & 43869 & 427790 & 208205 &\tabularnewline 154 & Manufacturing services on physical inputs owned by others & 117.08 & 26426 & 11693 & 5833 &\tabularnewline 155 & Maintenance and repair services n.i.e. & 46.97 & 2577 & 18995 & 9403 &\tabularnewline 156 & Transport & 1442.14 & 1047168 & 64771 & 21460 &\tabularnewline 157 & Travel & 1311.46 & 946006 & 43927 & 17748 &\tabularnewline 158 & Construction & 4201.07 & 1438899 & 41874 & 22337 &\tabularnewline 159 & Insurance and pension services & 1871.66 & 1412604 & 46687 & 22605 &\tabularnewline 160 & Financial services & 1120.47 & 886362 & 49593 & 22776 &\tabularnewline 161 & Charges for the use of intellectual property n.i.e. & 139.57 & 18713 & 44255 & 21228 &\tabularnewline 162 & Telecommunications, computer, and information services & 1890.46 & 1922669 & 56484 & 22439 &\tabularnewline 163 & Other business services & 2795.92 & 3266031 & 61736 & 23986 &\tabularnewline 164 & Heritage and recreational services & 6904.15 & 505072 & 8636 & 6166 &\tabularnewline 165 & Health services & 6883.00 & 2299522 & 24006 & 14300 &\tabularnewline 166 & Education services & 2781.07 & 1314752 & 34327 & 18931 &\tabularnewline 167 & Government goods and services n.i.e. & 19.81 & 4702 & 40507 & 19334 &\tabularnewline 168 & Services not allocated & 775.16 & 67672 & 38824 & 20635 &\tabularnewline 169 & Trade-related services & 6911.71 & 3036797 & 39111 & 20646 &\tabularnewline 170 & Other personal services & 6048.57 & 529953 & 9542 & 7064 &\tabularnewline & & & & & &\tabularnewline \bottomrule \end{longtable} \hypertarget{ITPD-E_countries}{} \begin{longtable}[]{@{}llrrrr@{}} \caption{ \protect\hypertarget{ITPD-E_countries}{}{{[}ITPD-E\_countries{]}} ITPD-E: Country Coverage and Summary Export Statistics\\ for International and Domestic Trade Flows}\tabularnewline \toprule ISO3 & Country Name & Mean Exports & Max Exports & \#Observations & \#Zeroes\tabularnewline \midrule \endfirsthead \toprule ISO3 & Country Name & Mean Exports & Max Exports & \#Observations & \#Zeroes\tabularnewline \midrule \endhead ISO3 & Country Name & Mean Exports & Max Exports & \#Observations & \#Zeroes\tabularnewline & & & & &\tabularnewline ABW & Aruba & 2.58 & 3294 & 67627 & 52887\tabularnewline AFG & Afghanistan & 0.38 & 229 & 112296 & 89008\tabularnewline AGO & Angola & 38.51 & 33373 & 80145 & 61822\tabularnewline AIA & Anguilla & 0.04 & 6 & 44262 & 37691\tabularnewline ALB & Albania & 5.68 & 6472 & 134929 & 98000\tabularnewline AND & Andorra & 0.12 & 288 & 124044 & 98825\tabularnewline ANT & Netherlands Antilles & 1.09 & 926 & 71673 & 47394\tabularnewline ARE & United Arab Emirates & 10.85 & 42341 & 373205 & 155148\tabularnewline ARG & Argentina & 8.15 & 12186 & 298090 & 139428\tabularnewline ARM & Armenia & 2.59 & 2258 & 109879 & 80495\tabularnewline ASM & American Samoa & 0.08 & 83 & 60231 & 50398\tabularnewline ATA & Antarctica & 0.16 & 42 & 13631 & 11807\tabularnewline ATF & French Southern Territories & 0.08 & 15 & 11443 & 9895\tabularnewline ATG & Antigua and Barbuda & 0.57 & 3729 & 107178 & 83528\tabularnewline AUS & Australia & 28.45 & 99275 & 411231 & 151195\tabularnewline AUT & Austria & 35.69 & 84077 & 402064 & 137296\tabularnewline AZE & Azerbaijan & 20.41 & 20094 & 129709 & 94340\tabularnewline BDI & Burundi & 3.55 & 1097 & 52196 & 42880\tabularnewline BEL & Belgium & 45.21 & 121503 & 455415 & 125430\tabularnewline BEN & Benin & 2.35 & 3794 & 84039 & 64793\tabularnewline BES & Bonaire, Sint Eustatius and Saba & 0.65 & 87 & 675 & 391\tabularnewline BFA & Burkina Faso & 2.54 & 1547 & 83832 & 63588\tabularnewline BGD & Bangladesh & 8.38 & 12596 & 197858 & 126497\tabularnewline BGR & Bulgaria & 7.14 & 17112 & 333410 & 161157\tabularnewline BHR & Bahrain & 5.28 & 10341 & 161213 & 109426\tabularnewline BHS & Bahamas, The & 4.91 & 2143 & 112764 & 87074\tabularnewline BIH & Bosnia and Herzegovina & 4.03 & 4354 & 166510 & 104976\tabularnewline BLM & Saint Barthelemy & 0.07 & 12 & 1151 & 682\tabularnewline BLR & Belarus & 6.12 & 15189 & 182347 & 101477\tabularnewline BLZ & Belize & 0.58 & 377 & 102160 & 81760\tabularnewline BMU & Bermuda & 35.77 & 32101 & 52032 & 42293\tabularnewline BOL & Bolivia & 3.57 & 3931 & 114868 & 78370\tabularnewline BRA & Brazil & 101.38 & 450192 & 390469 & 145324\tabularnewline BRB & Barbados & 0.25 & 380 & 134317 & 88128\tabularnewline BRN & Brunei & 7.91 & 5971 & 80447 & 60768\tabularnewline BTN & Bhutan & 1.07 & 130 & 38215 & 32162\tabularnewline BVT & Bouvet Island & 0.50 & 74 & 6357 & 5308\tabularnewline BWA & Botswana & 2.59 & 3872 & 103952 & 75654\tabularnewline CAF & Central African Republic & 0.64 & 741 & 59182 & 50498\tabularnewline CAN & Canada & 72.40 & 314038 & 447913 & 141673\tabularnewline CCK & Cocos (Keeling) Islands & 0.03 & 5 & 28942 & 24952\tabularnewline CHE & Switzerland & 48.73 & 192302 & 428383 & 136145\tabularnewline CHL & Chile & 19.06 & 51055 & 247812 & 131685\tabularnewline CHN & China & 204.82 & 964575 & 484848 & 90170\tabularnewline CIV & Cote d'Ivoire & 3.54 & 2668 & 173161 & 114579\tabularnewline CMR & Cameroon & 3.21 & 3927 & 162520 & 114978\tabularnewline COD & Congo, Democratic Republic of the & 3.37 & 2321 & 85860 & 67199\tabularnewline COG & Congo, Republic of the & 7.10 & 5455 & 99841 & 77618\tabularnewline COK & Cook Islands & 0.28 & 126 & 28778 & 24406\tabularnewline COL & Colombia & 12.65 & 20505 & 248553 & 134369\tabularnewline COM & Comoros & 0.16 & 23 & 24837 & 20419\tabularnewline CPV & Cape Verde & 0.54 & 164 & 66543 & 54235\tabularnewline CRI & Costa Rica & 8.71 & 8950 & 204397 & 116330\tabularnewline CUB & Cuba & 0.97 & 887 & 116830 & 86176\tabularnewline CUW & Curacao & 1.74 & 814 & 25205 & 14859\tabularnewline CXR & Christmas Island & 0.15 & 25 & 20039 & 17067\tabularnewline CYM & Cayman Islands & 1.67 & 1013 & 52393 & 43372\tabularnewline CYP & Cyprus & 6.33 & 7568 & 238844 & 144778\tabularnewline CZE & Czech Republic & 25.25 & 53165 & 361673 & 134507\tabularnewline DEU & Germany & 207.76 & 604221 & 505048 & 106212\tabularnewline DJI & Djibouti & 0.28 & 108 & 50921 & 43495\tabularnewline DMA & Dominica & 0.08 & 43 & 96735 & 76762\tabularnewline DNK & Denmark & 27.57 & 72376 & 417392 & 144334\tabularnewline DOM & Dominican Republic & 4.76 & 12838 & 188147 & 115199\tabularnewline DZA & Algeria & 32.92 & 71378 & 131418 & 97307\tabularnewline ECU & Ecuador & 12.76 & 20201 & 210291 & 129022\tabularnewline EGY & Egypt, Arab Rep. & 8.11 & 29061 & 299337 & 145862\tabularnewline ERI & Eritrea & 1.49 & 373 & 46703 & 39924\tabularnewline ESH & Western Sahara & 0.06 & 8 & 6386 & 5390\tabularnewline ESP & Spain & 99.64 & 507348 & 444857 & 119530\tabularnewline EST & Estonia & 4.96 & 5249 & 250130 & 137521\tabularnewline ETH & Ethiopia (excludes Eritrea) & 3.50 & 4111 & 169668 & 121932\tabularnewline FIN & Finland & 28.49 & 46100 & 345940 & 141456\tabularnewline FJI & Fiji & 0.68 & 502 & 112434 & 74405\tabularnewline FLK & Falkland Islands & 1.08 & 185 & 17524 & 15012\tabularnewline FRA & France & 158.17 & 597713 & 488779 & 105854\tabularnewline FRO & Faeroe Islands & 1.10 & 167 & 41075 & 31701\tabularnewline FSM & Micronesia, Federated States of & 0.28 & 57 & 16044 & 13156\tabularnewline GAB & Gabon & 4.42 & 4523 & 105156 & 81146\tabularnewline GBR & United Kingdom & 155.67 & 546990 & 493871 & 119178\tabularnewline GEO & Georgia & 1.04 & 472 & 190823 & 134294\tabularnewline GHA & Ghana & 3.73 & 6488 & 180850 & 123076\tabularnewline GIB & Gibraltar & 0.59 & 671 & 57828 & 47632\tabularnewline GIN & Guinea & 3.19 & 1074 & 97025 & 78332\tabularnewline GMB & Gambia, The & 0.25 & 50 & 73257 & 61106\tabularnewline GNB & Guinea-Bissau & 5.16 & 990 & 18911 & 16019\tabularnewline GNQ & Equatorial Guinea & 21.91 & 3188 & 34080 & 27772\tabularnewline GRC & Greece & 27.68 & 72284 & 340433 & 151943\tabularnewline GRD & Grenada & 0.09 & 20 & 58708 & 48105\tabularnewline GRL & Greenland & 1.96 & 491 & 29919 & 24355\tabularnewline GTM & Guatemala & 2.06 & 1300 & 174822 & 107039\tabularnewline GUF & French Guiana & 0.06 & 0 & 12 & 5\tabularnewline GUM & Guam & 0.12 & 41 & 36830 & 28633\tabularnewline GUY & Guyana & 0.71 & 490 & 98856 & 71973\tabularnewline HKG & Hong Kong & 29.52 & 191535 & 387871 & 135613\tabularnewline HMD & Heard Island and McDonald Islands & 0.01 & 1 & 4358 & 3638\tabularnewline HND & Honduras & 2.48 & 1805 & 145168 & 97178\tabularnewline HRV & Croatia & 9.41 & 15188 & 251509 & 135813\tabularnewline HTI & Haiti & 0.85 & 624 & 66869 & 53011\tabularnewline HUN & Hungary & 19.47 & 32754 & 322083 & 135165\tabularnewline IDN & Indonesia & 20.86 & 60480 & 398769 & 156056\tabularnewline IMN & Isle of Man & 15.28 & 123 & 54 & 23\tabularnewline IND & India & 69.54 & 458348 & 458718 & 134699\tabularnewline IOT & British Indian Ocean Ter. & 0.11 & 55 & 27357 & 24094\tabularnewline IRL & Ireland & 40.96 & 67868 & 332240 & 147634\tabularnewline IRN & Iran & 28.84 & 106805 & 250447 & 142323\tabularnewline IRQ & Iraq & 61.05 & 22701 & 71569 & 58109\tabularnewline ISL & Iceland & 5.04 & 5459 & 190033 & 123061\tabularnewline ISR & Israel & 23.88 & 44598 & 302920 & 132810\tabularnewline ITA & Italy & 141.65 & 480001 & 476930 & 110258\tabularnewline JAM & Jamaica & 1.35 & 1283 & 157336 & 107050\tabularnewline JOR & Jordan & 3.00 & 6445 & 195097 & 116938\tabularnewline JPN & Japan & 419.33 & 1415613 & 418911 & 126628\tabularnewline KAZ & Kazakhstan & 23.12 & 31599 & 161252 & 100996\tabularnewline KEN & Kenya & 5.75 & 10147 & 224274 & 143489\tabularnewline KGZ & Kyrgyzstan & 2.66 & 1859 & 104934 & 76304\tabularnewline KHM & Cambodia & 4.42 & 2662 & 131392 & 94496\tabularnewline KIR & Kiribati & 0.23 & 85 & 22107 & 19075\tabularnewline KNA & Saint Kitts and Nevis & 0.16 & 57 & 54605 & 44155\tabularnewline KOR & Korea, South & 105.16 & 203221 & 404257 & 131690\tabularnewline KWT & Kuwait & 24.96 & 37888 & 180226 & 117567\tabularnewline LAO & Laos & 3.84 & 1803 & 76981 & 59784\tabularnewline LBN & Lebanon & 0.52 & 861 & 269503 & 145285\tabularnewline LBR & Liberia & 1.86 & 595 & 66270 & 56152\tabularnewline LBY & Libya & 35.22 & 18754 & 73995 & 59691\tabularnewline LCA & Saint Lucia & 0.17 & 144 & 67450 & 48700\tabularnewline LIE & Liechtenstein & 8.46 & 295 & 3034 & 1789\tabularnewline LKA & Sri Lanka & 4.10 & 17761 & 261440 & 143403\tabularnewline LSO & Lesotho & 2.07 & 334 & 35750 & 29037\tabularnewline LTU & Lithuania & 6.43 & 11109 & 268554 & 140093\tabularnewline LUX & Luxembourg & 21.58 & 62883 & 255449 & 130761\tabularnewline LVA & Latvia & 5.85 & 11463 & 241274 & 133402\tabularnewline MAC & Macao & 1.10 & 1093 & 107253 & 71113\tabularnewline MAR & Morocco & 6.79 & 5875 & 254319 & 141308\tabularnewline MCO & Monaco & 0.02 & 0 & 4 & 2\tabularnewline MDA & Moldova & 2.49 & 2010 & 134406 & 89826\tabularnewline MDG & Madagascar & 1.47 & 1427 & 143136 & 100064\tabularnewline MDV & Maldives & 0.30 & 115 & 48908 & 38625\tabularnewline MEX & Mexico & 102.76 & 279748 & 332546 & 137594\tabularnewline MHL & Marshall Islands & 2.33 & 781 & 30833 & 26292\tabularnewline MKD & Macedonia & 3.53 & 2541 & 143330 & 88137\tabularnewline MLI & Mali & 3.06 & 2494 & 121058 & 92946\tabularnewline MLT & Malta & 5.00 & 4324 & 196532 & 124661\tabularnewline MMR & Myanmar & 8.70 & 11599 & 101074 & 72647\tabularnewline MNE & Montenegro & 1.84 & 1076 & 49748 & 31417\tabularnewline MNG & Mongolia & 6.67 & 2780 & 81905 & 61938\tabularnewline MNP & Northern Marianas & 0.07 & 42 & 20697 & 16468\tabularnewline MOZ & Mozambique & 4.43 & 3123 & 120864 & 92506\tabularnewline MRT & Mauritania & 2.60 & 1163 & 69112 & 56528\tabularnewline MSR & Montserrat & 0.02 & 3 & 33768 & 28449\tabularnewline MUS & Mauritius & 2.09 & 2159 & 178090 & 113730\tabularnewline MWI & Malawi & 3.29 & 3831 & 110255 & 81967\tabularnewline MYS & Malaysia & 21.91 & 37621 & 382376 & 148319\tabularnewline MYT & Mayotte & 0.05 & 26 & 12459 & 9373\tabularnewline NAM & Namibia & 1.37 & 1027 & 194306 & 140444\tabularnewline NCL & New Caledonia & 1.09 & 386 & 96687 & 74843\tabularnewline NER & Niger & 2.96 & 2715 & 112934 & 88409\tabularnewline NFK & Norfolk Island & 0.03 & 2 & 10002 & 8567\tabularnewline NGA & Nigeria & 86.33 & 169010 & 196703 & 133460\tabularnewline NIC & Nicaragua & 1.73 & 897 & 124986 & 88073\tabularnewline NIU & Niue & 0.13 & 95 & 23575 & 20091\tabularnewline NLD & Netherlands & 62.32 & 193707 & 484893 & 129138\tabularnewline NOR & Norway & 47.80 & 173194 & 324450 & 137472\tabularnewline NPL & Nepal & 2.38 & 2464 & 112751 & 75941\tabularnewline NRU & Nauru & 0.14 & 87 & 37406 & 32335\tabularnewline NZL & New Zealand & 15.39 & 39099 & 326632 & 148136\tabularnewline OMN & Oman & 9.26 & 22785 & 171143 & 110668\tabularnewline PAK & Pakistan & 11.10 & 47625 & 302648 & 156043\tabularnewline PAN & Panama & 1.41 & 1525 & 204284 & 117720\tabularnewline PCN & Pitcairn & 0.02 & 9 & 18244 & 15850\tabularnewline PER & Peru & 24.36 & 33889 & 224824 & 123449\tabularnewline PHL & Philippines & 14.33 & 65100 & 313383 & 151355\tabularnewline PLW & Palau & 0.04 & 10 & 15097 & 12616\tabularnewline PNG & Papua New Guinea & 5.25 & 2471 & 70730 & 54207\tabularnewline POL & Poland & 46.22 & 151853 & 389642 & 158089\tabularnewline PRI & Puerto Rico & 0.01 & 0 & 25 & 13\tabularnewline PRK & Korea, North & 1.61 & 5944 & 180083 & 122711\tabularnewline PRT & Portugal & 24.69 & 55490 & 366106 & 150324\tabularnewline PRY & Paraguay & 3.80 & 2006 & 108819 & 76205\tabularnewline PSE & Palestine & 3.95 & 1030 & 41745 & 32853\tabularnewline PYF & French Polynesia & 0.13 & 54 & 63579 & 48282\tabularnewline QAT & Qatar & 26.06 & 32236 & 166066 & 112911\tabularnewline ROU & Romania & 23.58 & 49756 & 308195 & 149590\tabularnewline RUS & Russia & 53.54 & 125608 & 350310 & 145379\tabularnewline RWA & Rwanda & 3.60 & 2403 & 71646 & 55993\tabularnewline SAU & Saudi Arabia & 34.63 & 55590 & 271794 & 150881\tabularnewline SCG & Serbia and Montenegro & 4.99 & 5331 & 27390 & 11722\tabularnewline SDN & Sudan & 10.15 & 9418 & 113170 & 91237\tabularnewline SEN & Senegal & 1.22 & 579 & 176385 & 118751\tabularnewline SGP & Singapore & 17.77 & 20860 & 387573 & 145020\tabularnewline SGS & South Georgia and South Sandwich Islands & 0.05 & 9 & 4831 & 4092\tabularnewline SHN & Saint Helena, Ascension, and Tristan da Cunha & 0.07 & 21 & 38225 & 32612\tabularnewline SLB & Solomon Islands & 0.33 & 112 & 36306 & 30422\tabularnewline SLE & Sierra Leone & 1.22 & 3321 & 109954 & 85968\tabularnewline SLV & El Salvador & 3.49 & 2310 & 135014 & 86773\tabularnewline SMR & San Marino & 0.22 & 598 & 62214 & 47068\tabularnewline SOM & Somalia & 0.71 & 313 & 43068 & 36594\tabularnewline SPM & Saint Pierre and Miquelon & 0.10 & 19 & 9439 & 7917\tabularnewline SRB & Serbia & 8.86 & 11375 & 155497 & 73732\tabularnewline SSD & South Sudan & 23.26 & 4329 & 1431 & 944\tabularnewline STP & Sao Tome and Principe & 0.04 & 8 & 43583 & 36656\tabularnewline SUR & Suriname & 1.40 & 1025 & 105557 & 82825\tabularnewline SVK & Slovakia & 15.32 & 25551 & 292232 & 139053\tabularnewline SVN & Slovenia & 7.96 & 14024 & 285543 & 132209\tabularnewline SWE & Sweden & 41.65 & 87211 & 395584 & 132315\tabularnewline SWZ & Swaziland & 1.57 & 927 & 147003 & 109008\tabularnewline SXM & Sint Maarten & 0.07 & 22 & 4914 & 3415\tabularnewline SYC & Seychelles & 0.70 & 287 & 86160 & 68440\tabularnewline SYR & Syria & 5.18 & 45330 & 192354 & 124263\tabularnewline TCA & Turks and Caicos Islands & 0.07 & 55 & 56611 & 48357\tabularnewline TCD & Chad & 9.92 & 3310 & 43268 & 36985\tabularnewline TGO & Togo & 1.46 & 843 & 99571 & 73004\tabularnewline THA & Thailand & 13.41 & 21976 & 425045 & 146228\tabularnewline TJK & Tajikistan & 4.09 & 1001 & 56713 & 44993\tabularnewline TKL & Tokelau & 0.04 & 22 & 50304 & 42177\tabularnewline TKM & Turkmenistan & 15.22 & 9441 & 53615 & 42702\tabularnewline TLS & East Timor & 0.48 & 431 & 39710 & 33827\tabularnewline TON & Tonga & 0.12 & 46 & 28536 & 23409\tabularnewline TTO & Trinidad and Tobago & 4.64 & 4764 & 161523 & 102957\tabularnewline TUN & Tunisia & 3.22 & 4383 & 228179 & 128868\tabularnewline TUR & Turkey & 22.34 & 60632 & 408159 & 134593\tabularnewline TUV & Tuvalu & 0.31 & 71 & 18968 & 16416\tabularnewline TWN & Taiwan & 33.33 & 82328 & 396874 & 127856\tabularnewline TZA & Tanzania & 2.31 & 2309 & 193263 & 129629\tabularnewline UGA & Uganda & 2.98 & 5424 & 145506 & 103693\tabularnewline UKR & Ukraine & 20.97 & 48904 & 335438 & 172535\tabularnewline UMI & U.S. Minor Outlying Islands & 0.07 & 71 & 42973 & 36875\tabularnewline URY & Uruguay & 3.07 & 5460 & 173672 & 109832\tabularnewline USA & United States & 720.86 & 3266031 & 513572 & 99200\tabularnewline UZB & Uzbekistan & 3.82 & 2861 & 83616 & 61015\tabularnewline VAT & Holy See & 0.03 & 6 & 21378 & 18508\tabularnewline VCT & Saint Vincent and the Grenadines & 0.24 & 102 & 63868 & 49271\tabularnewline VEN & Venezuela & 16.95 & 44638 & 182439 & 114485\tabularnewline VGB & British Virgin Islands & 0.77 & 988 & 94766 & 76493\tabularnewline VIR & U.S. Virgin Islands & 0.47 & 16 & 71 & 26\tabularnewline VNM & Vietnam & 14.12 & 37670 & 317891 & 156808\tabularnewline VUT & Vanuatu & 1.02 & 314 & 36771 & 30097\tabularnewline WLF & Wallis and Futuna Islands & 0.01 & 1 & 10395 & 8922\tabularnewline WSM & Samoa & 0.31 & 94 & 47287 & 37063\tabularnewline YEM & Yemen & 6.79 & 4205 & 108046 & 82108\tabularnewline ZAF & South Africa & 8.97 & 27623 & 415897 & 157415\tabularnewline ZMB & Zambia & 3.27 & 4297 & 114920 & 82221\tabularnewline ZWE & Zimbabwe & 1.17 & 680 & 136006 & 94472\tabularnewline & & & & &\tabularnewline \bottomrule \end{longtable} \hypertarget{tab:file_columns}{} \begin{longtable}[]{@{}ll@{}} \caption{\protect\hypertarget{tab:file_columns}{}{{[}tab:file\_columns{]}}Data File Columns}\tabularnewline \toprule Column name & Column description\tabularnewline \midrule \endfirsthead \toprule Column name & Column description\tabularnewline \midrule \endhead exporter\_iso3 & ISO 3-letter alpha code of the exporter\tabularnewline exporter\_name & Name of the exporter\tabularnewline importer\_iso3 & ISO 3-letter alpha code of the importer\tabularnewline importer\_name & Name of the importer\tabularnewline year & Year\tabularnewline industry\_id & ITPD industry code\tabularnewline industry\_descr & ITPD industry description\tabularnewline broad\_sector & Broad sector\tabularnewline trade & Trade flows in million of current US dollars\tabularnewline flag\_mirror & Flag indicator, 1 if trade mirror value is used\tabularnewline flag\_zero & Flag indicator:\tabularnewline & `p' if positive trade\tabularnewline & `r' if the raw data contained zero\tabularnewline & `u' missing (unknown, assigned zero)\tabularnewline \bottomrule \end{longtable} \hypertarget{tab:ITPD-E_class_ag}{} \begin{longtable}[]{@{}clcl@{}} \caption{\protect\hypertarget{tab:ITPD-E_class_ag}{}{{[}tab:ITPD-E\_class\_ag{]}}USITC Agricultural Classification}\tabularnewline \toprule ITPD-E Code & ITPD-E Description & FCL Item Code & FCL Title\tabularnewline \midrule \endfirsthead \toprule ITPD-E Code & ITPD-E Description & FCL Item Code & FCL Title\tabularnewline \midrule \endhead ITPD-E Code & ITPD-E Description & FCL Item Code & FCL Title\tabularnewline & & &\tabularnewline & Wheat & 15 & Wheat\tabularnewline 2 & Rice (raw) & 27 & Rice, paddy\tabularnewline 3 & Corn & 56 & Maize\tabularnewline 4 & Other cereals & 44 & Barley\tabularnewline 4 & Other cereals & 71 & Rye\tabularnewline 4 & Other cereals & 75 & Oats\tabularnewline 4 & Other cereals & 79 & Millet\tabularnewline 4 & Other cereals & 83 & Sorghum\tabularnewline 4 & Other cereals & 89 & Buckwheat\tabularnewline 4 & Other cereals & 92 & Quinoa\tabularnewline 4 & Other cereals & 94 & Fonio\tabularnewline 4 & Other cereals & 97 & Triticale\tabularnewline 4 & Other cereals & 101 & Canary seed\tabularnewline 4 & Other cereals & 103 & Mixed grain\tabularnewline 4 & Other cereals & 108 & Cereals, nes\tabularnewline 5 & Cereal products & 17 & Bran of Wheat\tabularnewline 5 & Cereal products & 59 & Bran of Maize\tabularnewline 5 & Cereal products & 81 & Bran of Millet\tabularnewline 5 & Cereal products & 85 & Bran of Sorghum\tabularnewline 5 & Cereal products & 91 & Bran of Buckwheat\tabularnewline 5 & Cereal products & 96 & Bran of Fonio\tabularnewline 6 & Soybeans & 236 & Soybeans\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 242 & Groundnuts, in shell\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 243 & Groundnuts, Shelled\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 249 & Coconuts\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 250 & Coconuts, Desiccated\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 251 & Copra\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 254 & Oil palm fruit\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 256 & Palm kernels\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 263 & Karite Nuts (Sheanuts)\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 265 & Castor Beans\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 267 & Sunflower seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 270 & Rapeseed or colza seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 275 & Tung Nuts\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 280 & Safflower seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 289 & Sesame seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 292 & Mustard seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 296 & Poppy seed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 299 & Melonseed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 310 & Kapok fruit\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 311 & Kapokseed in shell\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 312 & Kapokseed, shelled\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 328 & Seed Cotton\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 329 & Cottonseed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 333 & Linseed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 336 & Hempseed\tabularnewline 7 & Other oilseeds (exc.~peanuts) & 339 & Oilseeds nes\tabularnewline 8 & Animal feed ingredients \& pet foods & 169 & Beet Pulp\tabularnewline 8 & Animal feed ingredients \& pet foods & 628 & Pulp, Waste of Fruit for Feed\tabularnewline 8 & Animal feed ingredients \& pet foods & 630 & Cane Tops\tabularnewline 8 & Animal feed ingredients \& pet foods & 635 & Straw \& Husks\tabularnewline 8 & Animal feed ingredients \& pet foods & 639 & Grasses nes for forage\tabularnewline 8 & Animal feed ingredients \& pet foods & 640 & Clover for forage\tabularnewline 8 & Animal feed ingredients \& pet foods & 643 & Legumes for silage\tabularnewline 8 & Animal feed ingredients \& pet foods & 646 & Turnips for fodder\tabularnewline 8 & Animal feed ingredients \& pet foods & 651 & Forage Products nes\tabularnewline 8 & Animal feed ingredients \& pet foods & 652 & Vegetable Products for Feed nes\tabularnewline 8 & Animal feed ingredients \& pet foods & 846 & Gluten Feed \& Meal\tabularnewline 8 & Animal feed ingredients \& pet foods & 858 & Hay (Clover, Lucerne, etc.)\tabularnewline 8 & Animal feed ingredients \& pet foods & 859 & Hay nes\tabularnewline 8 & Animal feed ingredients \& pet foods & 862 & Alfalfa Meal \& Pellets\tabularnewline 9 & Raw \& refined sugar \& sugar crops & 156 & Sugar cane\tabularnewline 9 & Raw \& refined sugar \& sugar crops & 157 & Sugar beet\tabularnewline 10 & Other sweeteners & 161 & Sugar crops nes\tabularnewline 10 & Other sweeteners & 1182 & Honey\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 176 & Beans, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 181 & Broad beans, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 187 & Peas, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 191 & Chick-peas, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 195 & Cow peas, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 197 & Pigeon peas\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 201 & Lentils, dry\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 203 & Bambara beans\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 205 & Vetches\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 210 & Lupins\tabularnewline 11 & Pulses \& legumes(dried, preserved) & 211 & Pulses nes\tabularnewline 12 & Fresh fruit & 486 & Bananas\tabularnewline 12 & Fresh fruit & 489 & Plantains\tabularnewline 12 & Fresh fruit & 490 & Oranges\tabularnewline 12 & Fresh fruit & 495 & Tangerines, mandarins, clementines, satsumas\tabularnewline 12 & Fresh fruit & 497 & Lemons \& limes\tabularnewline 12 & Fresh fruit & 507 & Grapefruit \& pomelo\tabularnewline 12 & Fresh fruit & 512 & Citrus fruit nes\tabularnewline 12 & Fresh fruit & 515 & Apples\tabularnewline 12 & Fresh fruit & 521 & Pears\tabularnewline 12 & Fresh fruit & 523 & Quinces\tabularnewline 12 & Fresh fruit & 526 & Apricots\tabularnewline 12 & Fresh fruit & 530 & Sour cherries\tabularnewline 12 & Fresh fruit & 531 & Cherries\tabularnewline 12 & Fresh fruit & 534 & Peaches \& nectarines\tabularnewline 12 & Fresh fruit & 536 & Plums\tabularnewline 12 & Fresh fruit & 541 & Stone fruit, fresh nes\tabularnewline 12 & Fresh fruit & 542 & Pome fruit nes\tabularnewline 12 & Fresh fruit & 544 & Strawberries\tabularnewline 12 & Fresh fruit & 547 & Raspberries\tabularnewline 12 & Fresh fruit & 549 & Gooseberries\tabularnewline 12 & Fresh fruit & 550 & Currants\tabularnewline 12 & Fresh fruit & 552 & Blueberries\tabularnewline 12 & Fresh fruit & 554 & Cranberries\tabularnewline 12 & Fresh fruit & 558 & Berries nes\tabularnewline 12 & Fresh fruit & 560 & Grapes\tabularnewline 12 & Fresh fruit & 561 & Raisins\tabularnewline 12 & Fresh fruit & 567 & Watermelons\tabularnewline 12 & Fresh fruit & 568 & Melons, Cantaloupes\tabularnewline 12 & Fresh fruit & 569 & Figs\tabularnewline 12 & Fresh fruit & 571 & Mangoes\tabularnewline 12 & Fresh fruit & 572 & Avocados\tabularnewline 12 & Fresh fruit & 574 & Pineapples\tabularnewline 12 & Fresh fruit & 577 & Dates\tabularnewline 12 & Fresh fruit & 587 & Persimmons\tabularnewline 12 & Fresh fruit & 591 & Cashewapple\tabularnewline 12 & Fresh fruit & 592 & Kiwi fruit\tabularnewline 12 & Fresh fruit & 600 & Papayas\tabularnewline 12 & Fresh fruit & 603 & Fruit, tropical (fresh) nes\tabularnewline 12 & Fresh fruit & 619 & Fruit, fresh nes\tabularnewline 13 & Fresh vegetables & 116 & Potatoes\tabularnewline 13 & Fresh vegetables & 122 & Sweet potatoes\tabularnewline 13 & Fresh vegetables & 125 & Cassava\tabularnewline 13 & Fresh vegetables & 135 & Yautia (Cocoyam)\tabularnewline 13 & Fresh vegetables & 136 & Taro (Cocoyam)\tabularnewline 13 & Fresh vegetables & 137 & Yams\tabularnewline 13 & Fresh vegetables & 149 & Roots \& tubers nes\tabularnewline 13 & Fresh vegetables & 260 & Olives\tabularnewline 13 & Fresh vegetables & 358 & Cabbages\tabularnewline 13 & Fresh vegetables & 366 & Artichokes\tabularnewline 13 & Fresh vegetables & 367 & Asparagus\tabularnewline 13 & Fresh vegetables & 372 & Lettuce \& chicory\tabularnewline 13 & Fresh vegetables & 373 & Spinach\tabularnewline 13 & Fresh vegetables & 388 & Tomatoes, fresh\tabularnewline 13 & Fresh vegetables & 393 & Cauliflowers \& broccoli\tabularnewline 13 & Fresh vegetables & 394 & Pumpkins, squash \& gourds\tabularnewline 13 & Fresh vegetables & 397 & Cucumbers \& gherkins\tabularnewline 13 & Fresh vegetables & 399 & Eggplants\tabularnewline 13 & Fresh vegetables & 401 & Chillies \& peppers (green)\tabularnewline 13 & Fresh vegetables & 402 & Onions, shallots (green)\tabularnewline 13 & Fresh vegetables & 403 & Onions, dry\tabularnewline 13 & Fresh vegetables & 406 & Garlic\tabularnewline 13 & Fresh vegetables & 407 & Leeks \& other alliaceous vegetables\tabularnewline 13 & Fresh vegetables & 414 & Beans, green\tabularnewline 13 & Fresh vegetables & 417 & Peas, green\tabularnewline 13 & Fresh vegetables & 420 & Broad Beans, Green\tabularnewline 13 & Fresh vegetables & 423 & String Beans\tabularnewline 13 & Fresh vegetables & 426 & Carrot\tabularnewline 13 & Fresh vegetables & 430 & Okra\tabularnewline 13 & Fresh vegetables & 446 & Green Corn (Maize)\tabularnewline 13 & Fresh vegetables & 449 & Mushrooms\tabularnewline 13 & Fresh vegetables & 459 & Chicory roots\tabularnewline 13 & Fresh vegetables & 463 & Vegetables, Fresh n.e.s.\tabularnewline 14 & Prepared fruits, fruit juices & 527 & Apricots, Dried\tabularnewline 14 & Prepared fruits, fruit juices & 537 & Plums, dried\tabularnewline 14 & Prepared fruits, fruit juices & 570 & Figs, Dried\tabularnewline 14 & Prepared fruits, fruit juices & 620 & Fruit, dried nes\tabularnewline 15 & Prepared vegetables & 120 & Potato Offals\tabularnewline 15 & Prepared vegetables & 128 & Cassava, Dried\tabularnewline 16 & Nuts & 216 & Brazil nuts\tabularnewline 16 & Nuts & 217 & Cashew nuts\tabularnewline 16 & Nuts & 220 & Chestnuts\tabularnewline 16 & Nuts & 221 & Almonds\tabularnewline 16 & Nuts & 222 & Walnuts\tabularnewline 16 & Nuts & 223 & Pistachios\tabularnewline 16 & Nuts & 224 & Kolanuts\tabularnewline 16 & Nuts & 225 & Hazelnuts (Filberts)\tabularnewline 16 & Nuts & 226 & Areca nuts\tabularnewline 16 & Nuts & 229 & Brazil Nuts, Shelled\tabularnewline 16 & Nuts & 230 & Cashew Nuts, Shelled\tabularnewline 16 & Nuts & 231 & Almonds, Shelled\tabularnewline 16 & Nuts & 232 & Walnuts, Shelled\tabularnewline 16 & Nuts & 233 & Hazelnuts, Shelled\tabularnewline 16 & Nuts & 234 & Nuts nes\tabularnewline 17 & Live Cattle & 866 & Cattle\tabularnewline 17 & Live Cattle & 946 & Buffaloes\tabularnewline 18 & Live Swine & 1034 & Pigs\tabularnewline 19 & Eggs & 1062 & Hen eggs\tabularnewline 19 & Eggs & 1091 & Eggs, exc.~hen eggs\tabularnewline 20 & Other meats, livest.~pr.  live animals & 976 & Sheep\tabularnewline 20 & Other meats, livest.~pr.  live animals & 987 & Wool, Greasy\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1009 & Wool, Hair Waste\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1016 & Goats\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1026 & Skins, Wet-Salted (Goats)\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1031 & Coarse goat hair\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1057 & Chickens\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1068 & Ducks\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1079 & Turkeys\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1083 & Pigeons and other birds\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1096 & Horses\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1107 & Asses\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1110 & Mules\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1126 & Camels\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1134 & Hides, Wet-Salted (Camels)\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1136 & Hides nes, Camels\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1140 & Rabbits\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1150 & Other rodents\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1157 & Other camelids\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1169 & Live animals, non food nes\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1171 & Live animals nes\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1181 & Bees\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1183 & Beeswax\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1185 & Cocoons, reelable\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1216 & Hides nes\tabularnewline 20 & Other meats, livest.~pr.  live animals & 1218 & Hair, fine\tabularnewline 21 & Cocoa and cocoa products & 661 & Cocoa beans\tabularnewline 22 & Beverages, nec & 656 & Coffee green\tabularnewline 22 & Beverages, nec & 667 & Tea\tabularnewline 22 & Beverages, nec & 671 & Mate\tabularnewline 23 & Cotton & 767 & Cotton Lint\tabularnewline 23 & Cotton & 769 & Cotton Waste\tabularnewline 24 & Tobacco leaves \& cigarettes & 826 & Tobacco leaves\tabularnewline 25 & Spices & 687 & Pepper\tabularnewline 25 & Spices & 689 & Pimento\tabularnewline 25 & Spices & 692 & Vanilla\tabularnewline 25 & Spices & 693 & Cinnamon (canella)\tabularnewline 25 & Spices & 698 & Cloves\tabularnewline 25 & Spices & 702 & Nutmeg, mace, cardamoms\tabularnewline 25 & Spices & 711 & Anise, badian, fennel\tabularnewline 25 & Spices & 720 & Ginger\tabularnewline 25 & Spices & 723 & Spices nes\tabularnewline 26 & Other ag.~products, nec & 460 & Vegetable products, fresh or dry nes\tabularnewline 26 & Other ag.~products, nec & 461 & Carobs\tabularnewline 26 & Other ag.~products, nec & 654 & Dregs from brewing, distillation\tabularnewline 26 & Other ag.~products, nec & 677 & Hops\tabularnewline 26 & Other ag.~products, nec & 748 & Peppermint, Spearmint\tabularnewline 26 & Other ag.~products, nec & 754 & Pyrethrum, dried flowers\tabularnewline 26 & Other ag.~products, nec & 755 & Pyrethrum Extract\tabularnewline 26 & Other ag.~products, nec & 771 & Flax, raw or retted\tabularnewline 26 & Other ag.~products, nec & 777 & Hemp fibre and tow\tabularnewline 26 & Other ag.~products, nec & 778 & Kapok fibre\tabularnewline 26 & Other ag.~products, nec & 780 & Jute\tabularnewline 26 & Other ag.~products, nec & 782 & Jute-like fibres\tabularnewline 26 & Other ag.~products, nec & 788 & Ramie\tabularnewline 26 & Other ag.~products, nec & 789 & Sisal\tabularnewline 26 & Other ag.~products, nec & 800 & Agave fibres nes\tabularnewline 26 & Other ag.~products, nec & 809 & Abaca manila hemp\tabularnewline 26 & Other ag.~products, nec & 813 & Coir\tabularnewline 26 & Other ag.~products, nec & 821 & Fibre crops nes\tabularnewline 26 & Other ag.~products, nec & 836 & Natural rubber\tabularnewline 26 & Other ag.~products, nec & 837 & Rubber, Natural (Dry)\tabularnewline 26 & Other ag.~products, nec & 839 & Natural gums\tabularnewline 26 & Other ag.~products, nec & 1293 & Crude Organic Materials nes\tabularnewline & & &\tabularnewline \bottomrule \end{longtable} \begin{longtable}[]{@{}clclcl@{}} \toprule ITPD-E Code & ITPD-E Description & ISIC3 & ISIC rev.~3 Description & ISIC4 & ISIC rev.~4 Description\tabularnewline \midrule \endhead 27 & Mining of hard coal & 101 & Mining and agglomeration of hard coal & 51 & Mining of hard coal\tabularnewline 28 & Mining of lignite & 102 & Mining and agglomeration of lignite & 52 & Mining of lignite\tabularnewline 29 & Extraction crude oil and gas & 111 & Extraction crude oil and gas & 6 & Extraction of crude oil and gas\tabularnewline 30 & Mining of iron ores & 131 & Iron ores & 71 & Mining of iron ores\tabularnewline 31 & Other mining and quarring & 190 & C-(101+102+111+131) & 90 & B-(051+052+61+62+071)\tabularnewline 32 & Electricity prodcn, collcn, and distr. & 401 & Electricity prodcn, collcn, and distr. & 351 & Electric power generation, transmission\tabularnewline 33 & Gas production and distribution & 402 & Gas production and distribution & 352 & Manufacture of gas\tabularnewline & & & & &\tabularnewline \bottomrule \end{longtable} \hypertarget{tab:ITPD-E_class_manuf}{} \begin{longtable}[]{@{}clcc@{}} \caption{ \protect\hypertarget{tab:ITPD-E_class_manuf}{}{{[}tab:ITPD-E\_class\_manuf{]}} Manufacturing: ITPD-E Classification and Concordances}\tabularnewline \toprule ITPD-E Code & ITPD-E Description & ISIC3 & ISIC4\tabularnewline \midrule \endfirsthead \toprule ITPD-E Code & ITPD-E Description & ISIC3 & ISIC4\tabularnewline \midrule \endhead ITPD-E Code & ITPD-E Description & ISIC3 & ISIC4\tabularnewline & & &\tabularnewline & Processing/preserving of meat & 1511 & 1010\tabularnewline 35 & Processing/preserving of fish & 1512 & 1020\tabularnewline 36 & Processing/preserving of fruit \& vegetables & 1513 & 1030\tabularnewline 37 & Vegetable and animal oils and fats & 1514 & 1040\tabularnewline 38 & Dairy products & 1520 & 1050\tabularnewline 39 & Grain mill products & 1531 & 1061\tabularnewline 40 & Starches and starch products & 1532 & 1062\tabularnewline 41 & Prepared animal feeds & 1533 & 1080\tabularnewline 42 & Bakery products & 1541 & 1071\tabularnewline 43 & Sugar & 1542 & 1072\tabularnewline 44 & Cocoa chocolate and sugar confectionery & 1543 & 1073\tabularnewline 45 & Macaroni noodles \& similar products & 1544 & 1074\tabularnewline 46 & Other food products n.e.c. & 1549 & 1075+1079\tabularnewline 47 & Distilling rectifying \& blending of spirits & 1551 & 1101\tabularnewline 48 & Wines & 1552 & 1102\tabularnewline 49 & Malt liquors and malt & 1553 & 1103\tabularnewline 50 & Soft drinks; mineral waters & 1554 & 1104\tabularnewline 51 & Tobacco products & 1600 & 1200\tabularnewline 52 & Textile fibre preparation; textile weaving & 1711 & 1311+1312\tabularnewline 53 & Made-up textile articles except apparel & 1721 & 1392\tabularnewline 54 & Carpets and rugs & 1722 & 1393\tabularnewline 55 & Cordage rope twine and netting & 1723 & 1394\tabularnewline 56 & Other textiles n.e.c. & 1729 & 1399\tabularnewline 57 & Knitted and crocheted fabrics and articles & 1730 & 1430+1391\tabularnewline 58 & Wearing apparel except fur apparel & 1810 & 1410\tabularnewline 59 & Dressing \& dyeing of fur; processing of fur & 1820 & 1420\tabularnewline 60 & Tanning and dressing of leather & 1911 & 1511\tabularnewline 61 & Luggage handbags etc.; saddlery \& harness & 1912 & 1512\tabularnewline 62 & Footwear & 1920 & 1520\tabularnewline 63 & Sawmilling and planing of wood & 2010 & 1610\tabularnewline 64 & Veneer sheets plywood particle board etc. & 2021 & 1621\tabularnewline 65 & Builders' carpentry and joinery & 2022 & 1622\tabularnewline 66 & Wooden containers & 2023 & 1623\tabularnewline 67 & Other wood products; articles of cork/straw & 2029 & 1629\tabularnewline 68 & Pulp paper and paperboard & 2101 & 1701\tabularnewline 69 & Corrugated paper and paperboard & 2102 & 1702\tabularnewline 70 & Other articles of paper and paperboard & 2109 & 1709\tabularnewline 71 & Publishing of books and other publications & 2211 &\tabularnewline 72 & Publishing of newspapers journals etc. & 2212 &\tabularnewline 73 & Publishing of recorded media & 2213 &\tabularnewline 74 & Other publishing & 2219 &\tabularnewline 75 & Printing & 2221 & 1811\tabularnewline 76 & Service activities related to printing & 2222 & 1812\tabularnewline 77 & Reproduction of recorded media & 2230 & 1820\tabularnewline 78 & Coke oven products & 2310 & 1910\tabularnewline 79 & Refined petroleum products & 2320 & 1920\tabularnewline 80 & Processing of nuclear fuel & 2330 &\tabularnewline 81 & Basic chemicals except fertilizers & 2411 & 2011\tabularnewline 82 & Fertilizers and nitrogen compounds & 2412 & 2012\tabularnewline 83 & Plastics in primary forms; synthetic rubber & 2413 & 2013\tabularnewline 84 & Pesticides and other agro-chemical products & 2421 & 2021\tabularnewline 85 & Paints varnishes printing ink and mastics & 2422 & 2022\tabularnewline 86 & Pharmaceuticals medicinal chemicals etc. & 2423 & 2100\tabularnewline 87 & Soap cleaning \& cosmetic preparations & 2424 & 2023\tabularnewline 88 & Other chemical products n.e.c. & 2429 & 2029+2680\tabularnewline 89 & Man-made fibers & 2430 & 2030\tabularnewline 90 & Rubber tires and tubes & 2511 & 2211\tabularnewline 91 & Other rubber products & 2519 & 2219\tabularnewline 92 & Plastic products & 2520 & 2220\tabularnewline 93 & Glass and glass products & 2610 & 2310\tabularnewline 94 & Pottery china and earthenware & 2691 & 2393\tabularnewline 95 & Refractory ceramic products & 2692 & 2391\tabularnewline 96 & Struct.non-refractory clay; ceramic products & 2693 & 2392\tabularnewline 97 & Cement lime and plaster & 2694 & 2394\tabularnewline 98 & Articles of concrete cement and plaster & 2695 & 2395\tabularnewline 99 & Cutting shaping \& finishing of stone & 2696 & 2396\tabularnewline 100 & Other non-metallic mineral products n.e.c. & 2699 & 2399\tabularnewline 101 & Basic iron and steel & 2710 & 2410\tabularnewline 102 & Basic precious and non-ferrous metals & 2720 & 2420\tabularnewline 103 & Casting of iron and steel & 2731 & 2431\tabularnewline 104 & Structural metal products & 2811 & 2511\tabularnewline 105 & Tanks reservoirs and containers of metal & 2812 & 2512\tabularnewline 106 & Steam generators & 2813 & 2513\tabularnewline 107 & Cutlery hand tools and general hardware & 2893 & 2593\tabularnewline 108 & Other fabricated metal products n.e.c. & 2899 & 2599\tabularnewline 109 & Engines \& turbines (not for transport equipment) & 2911 & 2811\tabularnewline 110 & Pumps compressors taps and valves & 2912 & 2812+2813\tabularnewline 111 & Bearings gears gearing \& driving elements & 2913 & 2814\tabularnewline 112 & Ovens furnaces and furnace burners & 2914 & 2815\tabularnewline 113 & Lifting and handling equipment & 2915 & 2816\tabularnewline 114 & Other general purpose machinery & 2919 & 2819\tabularnewline 115 & Agricultural and forestry machinery & 2921 & 2821\tabularnewline 116 & Machine tools & 2922 & 2818+2822\tabularnewline 117 & Machinery for metallurgy & 2923 & 2823\tabularnewline 118 & Machinery for mining \& construction & 2924 & 2824\tabularnewline 119 & Food/beverage/tobacco processing machinery & 2925 & 2825\tabularnewline 120 & Machinery for textile apparel and leather & 2926 & 2826\tabularnewline 121 & Weapons and ammunition & 2927 & 2520+3040\tabularnewline 122 & Other special purpose machinery & 2929 & 2829\tabularnewline 123 & Domestic appliances n.e.c. & 2930 & 2750\tabularnewline 124 & Office accounting and computing machinery & 3000 & 2620+2817\tabularnewline 125 & Electric motors generators and transformers & 3110 & 2710\tabularnewline 126 & Electricity distribution \& control apparatus & 3120 & 2733\tabularnewline 127 & Insulated wire and cable & 3130 & 2731+2732\tabularnewline 128 & Accumulators primary cells and batteries & 3140 & 2720\tabularnewline 129 & Lighting equipment and electric lamps & 3150 & 2740\tabularnewline 130 & Other electrical equipment n.e.c. & 3190 & 2790\tabularnewline 131 & Electronic valves tubes etc. & 3210 & 2610\tabularnewline 132 & TV/radio transmitters; line comm. apparatus & 3220 & 2630\tabularnewline 133 & TV and radio receivers and associated goods & 3230 & 2640\tabularnewline 134 & Medical surgical and orthopedic equipment & 3311 & 2660+3250\tabularnewline 135 & Measuring/testing/navigating appliances and equipment & 3312+3313 & 2651\tabularnewline 136 & Optical instruments \& photographic equipment & 3320 & 2670\tabularnewline 137 & Watches and clocks & 3330 & 2652\tabularnewline 138 & Motor vehicles & 3410 & 2910\tabularnewline 139 & Automobile bodies trailers \& semi-trailers & 3420 & 2920\tabularnewline 140 & Parts/accessories for automobiles & 3430 & 2930\tabularnewline 141 & Building and repairing of ships & 3511 & 3011\tabularnewline 142 & Building/repairing of pleasure/sport. boats & 3512 & 3012\tabularnewline 143 & Railway/tramway locomotives \& rolling stock & 3520 & 3020\tabularnewline 144 & Aircraft and spacecraft & 3530 & 3030\tabularnewline 145 & Motorcycles & 3591 & 3091\tabularnewline 146 & Bicycles and invalid carriages & 3592 & 3092\tabularnewline 147 & Other transport equipment n.e.c. & 3599 & 3099\tabularnewline 148 & Furniture & 3610 & 3100\tabularnewline 149 & Jewelery and related articles & 3691 & 3211+3212\tabularnewline 150 & Musical instruments & 3692 & 3220\tabularnewline 151 & Sports goods & 3693 & 3230\tabularnewline 152 & Games and toys & 3694 & 3240\tabularnewline 153 & Other manufacturing n.e.c. & 3699 & 3290\tabularnewline & & &\tabularnewline \bottomrule \end{longtable} \begin{longtable}[]{@{}lllll@{}} \toprule ITPD-E Code & ITPD-E Description & EBOPS 2002 & EBOPS 2010 & ISIC Rev. 4\tabularnewline \midrule \endhead 154 & Manufacturing services on physical inputs owned by others & & SA & --\tabularnewline 155 & Maintenance and repair services n.i.e. & & SB & --\tabularnewline 156 & Transport & 205, 246 & SC & H\tabularnewline 157 & Travel & 237, 243 & SDA + SDB3 & I\tabularnewline 158 & Construction & 249 & SE & F\tabularnewline 159 & Insurance and pension services & 253 & SF & K (60\%)\tabularnewline 160 & Financial services & 260 & SG & K (40\%)\tabularnewline 161 & Charges for the use of intellectual property n.i.e. & 266 & SH & --\tabularnewline 162 & Telecommunications, computer, and information services & 247, 262, 288 & SI + SK1 & J\tabularnewline 163 & Other business services & 272, 273 & SJ excl SJ34 & M + N\tabularnewline 164 & Heritage and recreational services & -- & SK23 & R\tabularnewline 165 & Health services & 241, 896 & SDB1 + SK21 & Q\tabularnewline 166 & Education services & 242, 895 & SDB2 + SK22 & P\tabularnewline 167 & Government goods and services n.i.e. & 291 & SL & --\tabularnewline 168 & Services not allocated & 982 & SN & --\tabularnewline 169 & Trade-related services & 271 & SJ34 & G\tabularnewline 170 & Other personal services & -- & SK24 & S\tabularnewline \bottomrule \end{longtable} EBOPS codes SK23 and SK24 are new to BPM6 and have no backwards correspondence in BPM5.\\ The concordance in this table differs from \cite[Table 14]{Wettsteinetal2019} in three respects: ISIC industry L is not concorded to `other business services' (EBOPS SJ); Rather than aggregating financial services and insurance services, respectively, we keep these two services products separate and split ISIC category K according to a fixed fraction; We do not include EBOPS SW (trade margins of wholesalers and retailers) as part of SJ34 because this code is not included in the WTO-UNCTAD-ITC annual trade in services dataset. \hypertarget{tab:serv_joint_cov}{} \begin{longtable}[]{@{}lrrrr@{}} \caption{\protect\hypertarget{tab:serv_joint_cov}{}{{[}tab:serv\_joint\_cov{]}} Coverage of joint WTO-UN services trade flow dataset}\tabularnewline \toprule \endhead Year & UN TSD & UN TSD & WTO & Total\tabularnewline & only & updated & only &\tabularnewline 2000 & 7,448 & 0 & 0 & 7,448\tabularnewline 2001 & 7,945 & 0 & 0 & 7,945\tabularnewline 2002 & 10,164 & 0 & 0 & 10,164\tabularnewline 2003 & 11,045 & 0 & 0 & 11,045\tabularnewline 2004 & 21,075 & 0 & 0 & 21,075\tabularnewline 2005 & 21,297 & 10 & 3,970 & 25,277\tabularnewline 2006 & 22,387 & 33 & 5,235 & 27,655\tabularnewline 2007 & 22,403 & 39 & 6,254 & 28,696\tabularnewline 2008 & 19,671 & 82 & 11,271 & 31,024\tabularnewline 2009 & 18,009 & 87 & 15,255 & 33,351\tabularnewline 2010 & 7,316 & 46 & 30,019 & 37,381\tabularnewline 2011 & 6,741 & 39 & 32,616 & 39,396\tabularnewline 2012 & 3,542 & 35 & 34,831 & 38,408\tabularnewline 2013 & 1,580 & 3 & 37,891 & 39,474\tabularnewline 2014 & 2,602 & 3 & 39,073 & 41,678\tabularnewline 2015 & 2,140 & 5 & 40,691 & 42,836\tabularnewline 2016 & 0 & 0 & 39,481 & 39,481\tabularnewline Total & 185,365 & 382 & 296,587 & 482,334\tabularnewline \bottomrule \end{longtable} \includegraphics{ols_dist.pdf}\includegraphics{ols_cntg.pdf}\\ \includegraphics{ols_lang.pdf}\includegraphics{ols_clny.pdf}\\ \includegraphics{ols_fta.pdf}\includegraphics{ols_smctry.pdf}\\ {\textbf{Notes:} Each panel of this figure reports estimates of the effects of a standard gravity covariate across the 170 industries of the ITPD-E. The estimates are obtained with the OLS estimator and the dependent variable is the logarithm of bilateral trade. All estimates are obtained with exporter-time and importer-time fixed effects. Each dot in each panel represents an estimate for a particular industry and the estimates are ordered from the smallest to the largest. See text for further details.} \includegraphics{ols_no_mirr_dist.pdf}\includegraphics{ols_no_mirr_cntg.pdf}\\ \includegraphics{ols_no_mirr_lang.pdf}\includegraphics{ols_no_mirr_clny.pdf}\\ \includegraphics{ols_no_mirr_fta.pdf}\includegraphics{ols_no_mirr_smctry.pdf}\\ {\textbf{Notes:} Each panel of this figure reports estimates of the effects of a standard gravity covariate across the 170 industries of the ITPD-E. The estimating sample does not include observations that were obtained after mirroring the underlying data, as described in the main text. The estimates are obtained with the OLS estimator and the dependent variable is the logarithm of bilateral trade. All estimates are obtained with exporter-time and importer-time fixed effects. Each dot in each panel represents an estimate for a particular industry and the estimates are ordered from the smallest to the largest. See text for further details.} \end{document}