Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

CCSA - Best Practicies For Ensuring Quality in Int - Stat.ver2.2

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 14

Quality Enhancement in International Merchandise Trade Statistics

Henri Laurencin Head Development Statistics and Information Branch UNCTAD

CCSA: Special Session on Best Practice for Ensuring Quality in International Statistics Athens, Greece, 29 May 2012

CCSA: Best Practice for Ensuring Quality in International Statistics

When statistics raise questions


Mail received at UNCTAD in April 2011 from UN Headquarters: Need urgent, quick assistance on question: there continues to be reluctance by the

xx delegate to resolve a discrepancy between OHRLLS/UNCTAD on one side and WTO on the other concerning LDCs exports as a share of world trade.

OHRLLS: UN Office of the High Representative for the Least Developed Countries, Landlocked Developing Countries and Small Island Developing States

CCSA: Best Practice for Ensuring Quality in International Statistics

Discrepancies?
Gambia: total merchandise exports ($ millions)
Sources IMF - Balance Of Payments IMF - Direction Of Trade IMF - International Finance Statistics World Bank - WDI UNSD - Comtrade UNCTAD WTO Economic Intelligence Unit Common Data Set April 2011 2000 46 15 15 16 15 15 15 15 2001 27 10 10 6 10 10 4 10 2002 31 13 12 4 13 12 5 13 2003 68 18 8 8 5 8 8 8 8 2004 107 38 10 10 18 10 10 10 10 2005 104 28 8 8 5 8 8 8 8 2006 109 35 11 21 11 11 11 12 11 2007 134 43 13 13 13 13 13 13 13 2008 206 51 14 14 14 14 14 14 14 2009 174 55 15 15 66 15 15 21 15 2010 167 61 15 15 35 15 15 17 15

CCSA: Best Practice for Ensuring Quality in International Statistics

Prerequisite for improvement


1. Better quality of merchandise trade statistics received from national statistical systems, especially from developing countries
2. Better access of International Organizations (IOs) to national statistics by establishing effective cooperation

CCSA: Best Practice for Ensuring Quality in International Statistics

Ways to be followed to improve quality


Agree on a definition of international merchandise trade statistics
Act by: Providing guidance to national data compilers on the collection and compilation of international merchandise trade statistics in line with agreed definitions Building statistical capacity in countries Sharing knowledge and statistics at the international level

CCSA: Best Practice for Ensuring Quality in International Statistics

TFIMTS
Creation in 1992 of the Inter-Agency Task Force on International Merchandise Trade Statistics (TFIMTS) to improve international cooperation in the collection, processing and dissemination of international merchandise trade statistics. The Inter-Agency Task Force on International Merchandise Trade Statistics is convened by the WTO and gathers specialists from Eurostat, FAO, IMF, OECD, UN Statistics Division, UNCTAD and UN Regional Commissions.

CCSA: Best Practice for Ensuring Quality in International Statistics

Achievements of TFIMTS
Establish international standards governing the measurement of trade statistics. Publication of the International Merchandise Trade Statistics: concepts and definitions (IMTS) Manual (last version endorsed by the UN Statistical Commission in 2010); Prepare data compilation recommendations set forth in the International Merchandise Trade Statistics: Compilers Manual + National compilation and reporting practices in IMTS; Support the organization of workshops, seminars and trainings at the country and regional levels.

CCSA: Best Practice for Ensuring Quality in International Statistics

Discrepancies persist 20 years later. Why?


IMF, Eurostat, OECD and UN Statistics Division are primary data compilers, thus collect data directly from different national sources (Central Bank, National Statistical Office, Ministry of Trade or Customs Department); IMF BOP data (based on Balance of Payment Manual) may have different figures as the concepts are not 100% in-line with other sources data (based on IMTS Concepts and Definitions). EU Community concept is also slightly different; IMTS Concepts and Definitions recommends to include re-exports and to exclude goods in transit from official statistics. There is possible mix-up between re-exports and transit. If goods in transit are included, it inflates trade statistics. If re-exports are excluded, it minimizes trade statistics; Methodology for estimating missing data is not fully harmonized.

CCSA: Best Practice for Ensuring Quality in International Statistics

Towards a common data set


Eurostat, OECD, UNSD, UNCTAD and WTO decided to joint their efforts and to reconcile their time series on merchandise trade statistics using international standards as a benchmark; They created and maintain the Common Data Set (CDS); Each agency is in charge of supplying figures for a predefined set of reporting economies. Reconciliations consist of scrutinizing significant differences between the agencies' data to determine the best value. In many instances, this results in finding an agreement on substitution of values and estimates; The CDS database gives access to statistics and documentation (metadata) of both officially reported and reconciled series. It covers annual total merchandise export and import values of over 200 economies for the period from 1995 to 2010.

CCSA: Best Practice for Ensuring Quality in International Statistics

Still discrepancies exist


Gambia: total merchandise exports ($ millions)
Source IMF Balance Of Payments IMF Direction Of Trade IMF International Financial Statistics World Bank WDI UNSD-Comtrade UNCTAD WTO EIU Common Data Set CDS (Dec.2011) Common Data Set CDS (Dec.2012) 2
1 2

2000

2001

2002

2003 68 18 8 8 5 8 8 8 8 8

2004 107 38 10 10 18 10 10 10 10 10

2005 104 28 8 8 5 8 8 8 8 8

2006 109 35 11 21 11 11 11 11 11 11

2007 134 43 13 13 13 13 13 13 13 13

2008 206 51 14 14 14 14 14 14 14 14

2009 174 56 15 15 66 66 1 66 15 15 66

2010 167 62 15 15 35 35 35 15 15 35

2011

46 15 15 16 15 15 15 15 15

27 10 10 6 10 10 12 10 10

31 13 12 4 13 12 15 13 13

118

95 33 40 16 33

Break in series starting 2009. Data includes partially re-exports. Forthcoming

But they are better identified!

CCSA: Best Practice for Ensuring Quality in International Statistics

Summing up (1/3)
No official statistics should be released without metadata (cf. Principles governing international statistical activities);

National agencies involved in data collection on a same subject matter (ex. Central Bank, National Statistical Office and Customs on IMTS) need to synchronize their activities at the country level;
International organizations involved in the collection and dissemination of statistics on a same subject matter should better coordinate (for instance to minimize reporting burden and avoid the release of contradictory statistics);

CCSA: Best Practice for Ensuring Quality in International Statistics

Summing up (2/3)
Initiatives from IOs to create common data sets by reconciling their time series (ex. CDS of IMTS) using international standards as benchmarks should be encouraged; A Common Data Set should not be seen by IOs as a restriction to the release of their own sets:

IOs can refer to a different standard for any reason (for ex. in case of the calculation of a derived indicator)
Coordination needs time and IOs should not be prevented to release their advanced statistics

CCSA: Best Practice for Ensuring Quality in International Statistics

Summing up (3/3)
Improved quality of statistics released by IOs and better use of resources can be assured by sharing data, information, methodology and more generally knowledge and expertise. UNCTAD and WTO are joining their efforts to fully harmonize their statistics on trade in merchandise and services; Last but not least, never forget the users of international statistics and properly assess their needs and expectations.

CCSA: Best Practice for Ensuring Quality in International Statistics

Thank you for attention

You might also like