1. Viet Vu
Consultant to United Nations’ Statistics
Division
15-17 March 2010

2. Institutional arrangement: should
GRPs compiled by regional
statistical offices or NSO?
Why GRP compilation by regional offices?
 Strong demand of regional government to plan and monitor
the development of regional economy;
 Fiscal constraint of national statistical offices.
Disadvantages:
 Do not have full information on national activities at
regional levels;
 Corporations registered/reported to local levels may have
operations elsewhere;
 Problems in measuring services, particularly in constant
prices;
 Discrepancies between sum of GRPs and GDP;
 Total staff in GRP compilation in the whole country is high
and may not be adequately trained.
Possible improvement
 Economic data are collected within an integrated sampling
scheme to satisfy both national and regional needs;
 Benchmark value added coefficients and price deflators
prepared by NSO and provided to regional offices.

3. institutional arrangement: should
GRPs compiled by regional
statistical offices or NSO? (2)
Why should GRP be compiled by NSO
 All the disadvantages created by GRP compilation
at regional levels are eliminated;
 All information available at NSO can be utilized,
allowing for cross checking of one region against
other regions and national averages;
 Equality between the sum of GRPs and GDP is
guaranteed at both current and constant prices;
 Statistical methods can be appropriately utilized
and well coordinated by better trained
statisticians;
 Regional political interference is avoided.

4. Issues for discussion
Institutional arrangement: Should
GRPs be compiled centrally by NSO?
GRP at constant prices: Should
national price or regional price indexes
be used for deflation?
Indicators for estimation and
extrapolation:

5. Issues for discussion
(cont.)
Informal activities at regional levels:
 Should benchmark household income and expenditure be
implemented at both national and regional levels every
five years and on the basis of this benchmark information,
estimate quarterly and annual information using
employment data?
 Is it possible to have annual household income and
expenditure survey for both national and regional
purposes or is it too expensive?
 Should employment by both Establishment Survey (ES)
and by Labor Force Survey (LFS) be recommended for
developing countries to use for not only estimating
informal activities but also monitoring employment in the
country?
 Other alternatives?