Transcript A presentation on the status of economic statistics

A PRESENTATION ON THE STATUS OF ECONOMIC
STATISTICS (MANUFACTURING) IN UGANDA
By
John Bonaventure
Musoke
Senior Statistician
1
www.ubos.org
“A CANDLE LOOSES NOTHING BY
LIGHTING ANOTHER CANDLE”
Prof. Ssenteza Kajubi
2
www.ubos.org
Content





3
The Business Register 2001/2002 and
2006/7
Uganda Business Inquiry 2000/2001
Index of Industrial Production (IoP – M)
Producer Price Index (PPI)
Challenges
www.ubos.org
The Business Register 2001/2002
Defn: A business register is a record of all
active business establishments. Essentially it
contains structural information about each
business.
 Objectives
- Create a comprehensive list of businesses
- Provide a frame for future business surveys
4
www.ubos.org
The Business Register …..

-
5
Prior to the start of the Registration exercise
an attempt was made to compile the register
from administrative sources;
The Uganda Revenue Authority
Uganda manufacturers Association (UMA)
Uganda Investment Authority etc.
www.ubos.org
The Business Register …..

-
6
Variables collected
Name of Establishment, Activity ( ISIC),
Location, GPS Coordinates, Tel:, Postal
Address, Employment by sex, Type of
Ownership, Start date, if enterprise – name.
www.ubos.org
The Business Register …..

7
Coverage
We covered all businesses operating
from a fixed location, regardless of
employment size band
A total of 160,883 businesses were
registered out of which 11,968 were for
manufacturing
www.ubos.org
THE UGANDA BUSINESS INQUIRY (UBI)
2000/2001
Defn: This is an economic survey covering all
sectors in the economy for the financial year
2000/2001
8
www.ubos.org
UBI Cont………
 Objectives
-
9
Compute VA for estimating GDP
Estimate capital Formation
Update input output table
Compile SAM
Update weights for IIP
Establish wts for PPI
www.ubos.org
GO and VA for manufacturing 2000/1
Sub sector
GO (mil. shs)
VA
(mil. shs)
VA/GO
Food
processing
912,140
325,609
35.7
Other
812,823
manufacturin
g
284,023
34.9
Total
609,932
35.3
10
1,724,963
www.ubos.org
Annual Business Inquiry 2006/7

11
In order to produce high frequency data that
can enable us update these important
indicators we are now carrying out an ABI
with ref. yr 2006/7. We finished data
collection and have just finished data entry. A
draft report is expected around July 2009.
www.ubos.org
Index of Production
(Manufacturing)
By John B. Musoke
12
www.ubos.org
Historical Background of IIP


13
Work started in 1990 after the Census of
Business Establishments (COBE)
The IIP was based mostly on measuring
quantities of production for products
available in the base period.
www.ubos.org
Errors within the IIP System

-
-
14
Challenges encountered in using Quantity to
measure volume change….
Establishments that close
New Establishments and products
Non Homogeneous products
Conversion factor related problems or
problems related to human error
www.ubos.org
Index of Production (IoP)

-
-
15
Index rebranded IoP to expand and Cover
Manufacturing
Services
Utilities
Mining
www.ubos.org
IoP – Manufacturing…….

The index computation mostly utilizes
deflated Sales values to measure change. It
still remains a volume/quantity index.
Q*P=V
Q = V/P
Qindex = Vindex/Pindex
16
www.ubos.org
IoP – Manufacturing…….

17
Just as PPI, data for IoP is collected from
formal manufacturing across the country
majority of it coming from the industrial belt of
Uganda C:\Documents and
Settings\john.musoke.UBOS.000\My
Documents\Maps\ugandan map1.doc
www.ubos.org
IoP – Manufacturing…….
Monthly data is collected on a quarterly basis
but entered by month in an Excel data
file.(IoP DATA SALES)
There are three sources of data:- Survey data
- VAT data
- Quantities data from mandated Authorities

18
www.ubos.org
IoP – Manufacturing…….




19
The index is derived in two steps:
First, for each “basic heading” (product or
activity group) a value index is compiled by
adding total turnover for each business (no
weighting – full coverage assumed)
The value index is then divided by a
corresponding price index (PPI)
This operation delivers volume indices at
“basic heading” level
www.ubos.org
IoP – Manufacturing…….


20
Second, these basic level indices are
combined using the Laspeyres formula into
the major Divisions/Groups and all
Manufacturing.
The weights used to aggregate are based on
gross value added taken from the UBI
www.ubos.org
IoP – Manufacturing…….




21
The index is based on 2002
Until June 2004, the basic heading indices
were based on the old system (quantities)
From July 2004 onwards, deflated turnover
has been adopted in most cases
The new indices are “linked” to the old before
being aggregated.
www.ubos.org
IoP – Manufacturing…….
Quality control (Sales are much more variable
than prices and more subject to error)
 We calculate the response rates based on
known (or assumed) average values of
turnover in the previous 12 months and
impute for non-respondents.
 We check the month-to-month change in
sales values and query large changes
22
www.ubos.org
IoP – Manufacturing…….

-
-
23
Challenges
Correcting outliers with VAT data.
Reliability of monthly series
Using survey data – small economy
Using survey data – few estabs.
www.ubos.org
UGANDAS EXPERIENCE
WITH THE PRODUCER PRICE INDEX
(LUSAKA 4th – 7th May 2009)
24
www.ubos.org
What is a Producer Price Index (PPI)?
25

PPI is a PRICE index (number) used to measure
the rate of change in the prices of goods and
services bought and sold by producers.

PPIs can differently be calculated to suit specific
goals thus we have a PPI for Manufacturing

PPIs for Mining and Quarrying, Agriculture,
Hotels and Utilities are in the initial stages of
development.
www.ubos.org
Output PPI..


26
UBOS started compiling PPIs (Manufacturing) in
2004.What is generated is an output PPI which
measures the rate of change in the prices of products
sold as they leave the producer.
The PPI generated is categorized btn.
- the Local Market (PPI for Local Market)
- for goods and services produced for export, (PPI for
Exports)
- And PPI combined which combines the two above
www.ubos.org
Which PPI are we measuring?



27
PPI-M measures both goods produced for the local
market and those produced for export.
This PPI covers the ‘formal’ Manufacturing sector
It measures the change in the effective unit selling
prices received by manufacturers for products sold
for both the home market and for exports.
www.ubos.org
PPI - M cont…
28

Products are priced at Basic Price

Basic Price is the price per unit of goods/
services receivable by the producer.
Because the producer does not receive taxes
on products but does receive subsidies.
Taxes on products (excise, non-refundable
VAT, taxes on exports, etc) are excluded,
while subsidies on products are included.
www.ubos.org
Methodology & Sample Size
determination
WHY PPI was generated?

29
Out of the UBI which was conducted in 2000/01, the
Bureau then had comprehensive data on Value
Added (VA) and Gross Output (GO) for
establishments. The data on GO was then used for
generating weights for PPI-M.
www.ubos.org



30
Methodology
Ctd is based on a sample of
The PPI Manufacturing
establishments with 10 or more employees as
reported during the UBI.
A wide range of representative products were
selected in July 2004
PPI-M covers all ISIC Manufacturing Divisions
from 15 to 37 (Table of ISIC)
www.ubos.org
Methodology Ctd



31
A sample of 200 establishments as of now are
covered in the PPI with close to 600 products.
The sampling method used was Probability
Proportional to size combined with Concentration of
businesses and personal judgment based on
previous data/information.
There are 36 main headings based on 4-digit ISIC.
www.ubos.org
Challenges before
-
-
32
Establishments that close
Inclusion of new establishments and
products
Non Homogeneous products
Conversion factor related problems or
problems related to human error
www.ubos.org
Methodology Contd…
33

These are collapsed into 21 groups. Closely related
activity codes producing almost homogenous
commodities are grouped together to form the 21 as
shown in the table.

Which are later put into 8 major headings which are
re-grouped to suit the requirements in National
Accounts.
www.ubos.org
Challenges after
-
-
34
Correcting outliers within VAT data files.
Variability of monthly sales/ turnover series
Using survey data – respondent fatigue
Using survey data – few establishments
Need for specialized respondents
Harmonizing data collection within UBOS
and without
www.ubos.org
Challenges cont…

35
Getting transaction prices (actual) other than
price-list prices.
www.ubos.org
36
www.ubos.org
PLOT 9
COLVILLE STREET
P.O BOX 7186
KAMPALA
TEL.
041-706000
FAX
041-237553
E-MAIL
[email protected]
Website
www.ubos.org
37
WELCOME TO UGANDA WELCOME TO STATISTICS HOUSE
www.ubos.org