Introduction

 

            The Directorate General of Valuation has undertaken a project to develop an export valuation data base on the lines similar to the National import Data base (NIDB).  The project work started in the second half of 2004 as per the directions from the Board (CBEC).  The mandate is to develop a comprehensive electronic data base on export goods with a view to checking overvaluation and abuse of export incentive schemes. 

 

2.    The work on the Export commodity Database (ECDB) involves capture of export data in a pre-determined format from the Customs Stations, its consolidation and analysis with the help of a specially designed software for providing certain results which would help in detecting potential cases of valuation fraud.  The commodity analysis includes calculation of unit values, weighted averages of identical/similar goods, standard deviations and marking potential cases of overvalued consignments as outliers. The analysed data is then disseminated to field formations for on-line use, to check export valuation of live consignments, especially those, which are sensitive to overvaluation.

 

Data Capture

 

3.       Export data is captured from Customs Stations in a predetermined format.  The format has been finalized in consultation with Customs Stations and NIC. 

 

4.         The data will be extracted as End of the Day Operations (EOD) from the EDI Stations.  The software for that purpose has been developed by the NIC and the same has been distributed to all the EDI Stations.  The Directorate General of Valuation (DOV) have already started receiving data in the format as text files.

 

5.         In respect of Non-EDI stations the software for data generation is being developed and will be distributed to all Customs stations shortly.

 

6.         The scrutiny of data being received from the EDI stations indicates that the data quality needs considerable improvement.  It is noticed that the description field is not being fed with due care in some cases, especially, the specific description and specification (brand, model, grade etc) of the products are missing.  Instead, broad description of the relevant DEPB or DBK Schedule is reproduced.  Further HS/ITC classification of the goods is hardly checked by the assessing officers.  Unit Quantity Code (UQC) used is also at variance even for same category of goods.  These deficiencies in the data quality are causing problems in the data analysis, which require clustering identical or similar goods based on description and specification.  Vigorous efforts are needed at Customs Stations to improve the data quantity.            

 

Data Analysis     

 

7.         The software for analyzing the export data for the purposes of ECDB has been developed by the Center of Development of Advance Computing (CDAC, formerly knows as NCST), which is a Government of India Organisation under the Ministry of Information and Technology. 

 

8.         The software has been designed for data analysis by clustering the raw data under predetermined fields.  The clustering mechanism makes use of the description of goods (key words) HS/ITC Code, Unit Quantity Code(UQC), country of destination, DBK Serial Number, DEPB Schedule number, DFRC Schedule number, etc.  The software based on their repeated occurrence in identified HS/ITC Codes selects the key words for the purpose of data analysis.

           

9.         The software provides adequate flexibility to choose the commodities, which should be subjected to analysis, based on one or more of the parameters described above.  On the basis of identified key words, the software runs an analysis to calculate the weighted average value per unit and standard deviation from weighted average for each consignment.  The software is also capable of marking those cases, which fall outside the sum of weighted average and standard deviation as outliers of that cluster.  The software also allows one to choose the frequency of the data analysis based on the date of the shipping bills or the date of let export order.

           

11.        The raw data are presently received from 21 EDI Stations and consists of 1.5 lakhs records approximately per week (about 60 MB).  While it will be possible for the software to analyse the entire data, it is proposed to limit the data analysis to those commodities which are sensitive to overvaluation so that analysed data results would be more meaningful and specifically targeted towards overvaluation of sensitive goods.

 

Identification of Sensitive Commodities

 

12.        The Directorate General of Valuation has attempted several approaches for arriving at a list of commodities which could be considered sensitive from the export valuation angle and vulnerable to the abuse of export incentive schemes.  The best suited approach was to restrict the analysis to goods covered by the export incentive schemes such as DBK and  DEPB.  The items in the above schedules were categorized into corresponding ITC/HSN codes and then the data analysis was carried out. Further, the list was pruned by taking away items on which value caps have been fixed for the DEPB/DBK rates and these items specific rates of DBK were also eliminated. The above exercise of identifying the sensitive commodities has brought down to data size by about 12% for the purpose of analysis.  The analysed results scrutinized for different period show outliers to the extent of 1.5% of total records.  The commodities, which are generally, marked as outliers were also scrutinized and found to fall in the category of those in which overvaluation was generally noticed.  An illustrative list of such commodities is given in Annexure II.

 

13.        The outliers are only indications of possible overvaluation and they need to be  scrutinized individually by gathering additional details to determine whether they are truly overvalued cases.  However, this exercise will not hold up the exports as the goods figuring in the ECDB were already exported.  The analysed results would help to compare values declared for identical or similar goods under processing for export and identify potentially overvalued cases.  Further, the outliers marked by ECDB, which are goods already exported, could be subjected to post export verifications / investigations for deliberate overvaluation.

 

Output Data Format

 

14.        The analysed data incorporates most of the input data fields except the details like name of the exporter and the consignee, which are considered commercially sensitive.  In addition, the data will incorporate the analysed results such as unit value, weighted averages, standard deviation and outlier.  The data can be queried by a variety of ways to get the desired information.  The data can be selected for the desired period based on different query options.  The more the options used, the finer the data (more specific). 

 

15.        The software also has provision for converting the output data selected into excel format and carrying out different trend analysis study by the user (data filtering, sorting, searching for lay words, specifications etc).  There is also provision for extracting various kinds of reports from the database, such as total exports of any commodity, port-wise exports, destination-wise exports etc.     

  

Implementation Schedule   

 

16.        The ECDB project (Phase I) was started in November 2004, by combining the export data received from the EDI stations, and converting them into a simple display format (access) without analysis, and placing the data on the website of the Directorate General of Valuation (www.dov.gov.in).  This simple database was available for direct access by all Customs fields in the formations on a password-protected basis.

 

17.        The field formations were requested to give their comments and suggestions for improving the format and presentation of ECDB. The feedback received was encouraging.  The field formations were also consulted on the ways to identify sensitive commodities from export angle for subjecting to data analysis.  The data analysis was validated through trial runs and it was further demonstrated to field officers at Mumbai Customs Zones (Mumbai Custom House, Nhave Sheva and to ACC Sahar).  The data analysis software was refined taking into account the feed back received. 

 

18.        The second Phase of the ECDB (ECDB Phase II) has now been launched, after the finalization of the data analysis software developed by CDAC.  This software has been developed for use in both the LAN environment and Stand alone environment. For the LAN environment the data has been provided in both the Access format and in the Oracle format. This will be followed by weekly data analysis and transmission to the filed formations for building up the local data base by updating the initial database.  The Customs stations can choose the format depending on the type of intranet server and software used.  The customs stations which are on EDI will be provided with the transmission through ICENET.  For other stations, arrangements are being made to send CDs containing the three months data.  The weekly anlaysed data will be placed on the DOV Website (www.dov.gov.in) in a downloadable format.  The local stations can upload the data on a PC followed by updating with weekly transmissions to be downloaded from the DOV Website.  The entire ECDB (Phase II) will also be made available on the DOV Website for access by officers on a Password protected basis.

 

19.        The next phase of the ECDB will incorporate data from Non-EDI stations for which the data capture module (software) is being developed.  The data transmission will be arranged via e-mail from the respective stations as is presently done in the case of NIDB.  It is expected that this final phase of the ECDB would be started within three months after the launch of Phase II.

 

 

Expected benefits   

 

20.        Export commodity data base (ECDB), will enable the Customs officers all over India to have access to the latest information on export consignments and the valuation trends.  This would be of great assistance to them: in checking of export valuation of live consignments and preventing export fraud or abuse of the export incentive schemes; in providing a host of information on commodities under export for economic analysis and monitoring the working of export incentive schemes; in preventing shifting of exporters from port to port and overvaluation of the goods at remote Customs stations; in facilitating risk analysis and targeting of suspect commodities tendered for export; and in providing valuable assistance to the investigating agencies to check commercial fraud.