This repository only holds the list of bugs that have been reported for IFs. Anyone may add a bug report, but please look to see if your issue has already been added!
I haven't heard any feedback on this, I added my original email at the end just as a note.
Currently, I'm exploring the TiVA dataset from OECD, I think we can create some tables from it for value add across sectors. The coverage is not good enough but can be a good supplement. Link is here- https://www.oecd.org/en/topics/sub-issues/trade-in-value-added.html
"Across all six sectors in IFs, these are current tables used to initialize base values in IFs- VaddAg% for agriculture, VaddMan% for manufacturing, VaddSer% for service, and ICTVAdd%TotBusAdd for ICTech. ICTVAdd%TotBusAdd is a table not marked as a preprocessor and has not been updated for a while. Instead, we have a table called VaddICT%GDP, which is pretty recent. Out of these four data tables, we can get agriculture, manufacturing and service data directly from WDI. The VaddICT%GDP table, however, was manually calculated from OECD’s data.
We can do the same calculation for other two sectors as well (materials & ICTech). The raw data represent value add by industries from national accounts, categorized by International Standard Industrial Classification (ISIC), either under revision 4 or revision 3. These are reported by United Nation Statistics Division and OECD. IEA also has a consolidated version here.
Hence, to get an accurate representation of sectoral value add data, I suggest we take a closer look at the ISIC and map those 2-digit sections to our sectors. The mapping between GTAP sectors to ISIC is also helpful. The Excel file attached has the complete mapping. One potential issue is the overlapping divisions. For example, the table VaddMan% from WDI covers ISIC divisions 10-33. But according to our current GTAP sectoral mapping (also attached in this email thread), divisions 23-25 are categorized as materials and the division 26 is categorized as ICT. Another problem we will see is undercounting. For instance, both water and construction sections are categorized as manufacturing using the current GTAP mapping. But the table VaddMan% does not include these sections.
I can think of two solutions now regarding the data initialization to make our data values more harmonized with GTAP:
Maintain the 3 WDI tables, add new tables based on ISIC 2-digit level data for the remaining sectors, and do further calculations to adjust for overlapping or undercounting.
We gather and aggregate ISIC 2-digit level data for all sectors. 3 WDI tables are only used as a back-up for countries not covered by UNSD and OECD.
BTW, I think the mapping between GTAP and IFs was manually created by us. For example, we decided that Chemical, rubber, plastic products should be put under manufacturing. Can someone confirm this? The other two mapping tables (ISIC to WDI & GTAP to ISIC) are published by WDI and GTAP."
I haven't heard any feedback on this, I added my original email at the end just as a note.
Currently, I'm exploring the TiVA dataset from OECD, I think we can create some tables from it for value add across sectors. The coverage is not good enough but can be a good supplement. Link is here- https://www.oecd.org/en/topics/sub-issues/trade-in-value-added.html
"Across all six sectors in IFs, these are current tables used to initialize base values in IFs- VaddAg% for agriculture, VaddMan% for manufacturing, VaddSer% for service, and ICTVAdd%TotBusAdd for ICTech. ICTVAdd%TotBusAdd is a table not marked as a preprocessor and has not been updated for a while. Instead, we have a table called VaddICT%GDP, which is pretty recent. Out of these four data tables, we can get agriculture, manufacturing and service data directly from WDI. The VaddICT%GDP table, however, was manually calculated from OECD’s data.
We can do the same calculation for other two sectors as well (materials & ICTech). The raw data represent value add by industries from national accounts, categorized by International Standard Industrial Classification (ISIC), either under revision 4 or revision 3. These are reported by United Nation Statistics Division and OECD. IEA also has a consolidated version here.
Hence, to get an accurate representation of sectoral value add data, I suggest we take a closer look at the ISIC and map those 2-digit sections to our sectors. The mapping between GTAP sectors to ISIC is also helpful. The Excel file attached has the complete mapping. One potential issue is the overlapping divisions. For example, the table VaddMan% from WDI covers ISIC divisions 10-33. But according to our current GTAP sectoral mapping (also attached in this email thread), divisions 23-25 are categorized as materials and the division 26 is categorized as ICT. Another problem we will see is undercounting. For instance, both water and construction sections are categorized as manufacturing using the current GTAP mapping. But the table VaddMan% does not include these sections.
I can think of two solutions now regarding the data initialization to make our data values more harmonized with GTAP:
BTW, I think the mapping between GTAP and IFs was manually created by us. For example, we decided that Chemical, rubber, plastic products should be put under manufacturing. Can someone confirm this? The other two mapping tables (ISIC to WDI & GTAP to ISIC) are published by WDI and GTAP."