Transcript and the first published estimates (P)

MEETING OF THE WORKING PARTY ON
NATIONAL ACCOUNTS
Paris, 10-12 October 2006
REVISIONS IN QUARTERLY GDP OF
OECD COUNTRIES
1
Aims of the paper

Building, maintaining and using revisions databases:
the QNA-MEI experience

Some remarks on strategy and tools for revisions
analysis

Size of revisions to the first published estimates (P)
on MEI (May 1995 – Aug 2005)
2
Main Results
Revisions between three years later estimates (Y3) and the
first published estimates (P) of q-o-q GDP growth rates (%) on
MEI. Mean absolute revision (1995.1-2002.4)*
1
0.8
0.6
0.4
USA
PRT
NZL
NOR
NLD
KOR
JPN
ITA
GBR
FRA
FIN
ESP
DNK
DEU
CHE
CAN
BEL
0
AUS
0.2
Y3_P
* BEL: 96.4;2002.4, DEU: 95.2.2002.4, PRT: 1996.1.2002.4, KOR 1999.4 2002.4.
3
Main Results
Revisions between three years later estimates (Y3) and the
first published estimates (P) of q-o-q GDP growth rates (%) on
MEI. Relative mean absolute revision (1995.1-2002.4)*
1
0.8
0.6
0.4
USA
PRT
NZL
NOR
NLD
KOR
JPN
ITA
GBR
FRA
FIN
ESP
DNK
DEU
CHE
CAN
BEL
0
AUS
0.2
Y3_P
* BEL: 96.4;2002.4, DEU: 95.2.2002.4, PRT: 1997.1-2002.4, KOR 1999.4 2002.4.
4
Main Results
Comparison between previous year’s and this year’s
updated results of revisions of the 3 years later
published estimates to the first published estimates
(Y3_P) of q-o-q GDP growth rates (%) on MEI.
Mean absolute revision.
1
0.8
0.6
0.4
This year
USA
PRT
NZL
NOR
NLD
KOR
JPN
ITA
GBR
FRA
FIN
ESP
DNK
DEU
CHE
CAN
BEL
0
AUS
0.2
Last year
KOR: cannot be compared because last year data for Korea was not relevant.
5
The QNA-MEI revisions dataset
OECD-ONS workshop on “Assessing and improving statistical
quality. Revision analysis for the national accounts”, 7-8 October
2004
“The OECD also agreed to explore the possibility to expand and
maintain the revisions database that was set up for purposes
discussed in the workshop.
An international dataset and the resulting summary measures of
revisions should help to put countries’ individual revision analyses
into an international context.”
First elements presented also at the STESEG meeting, 27-28 June
2005
6
MEI as a revisions database source

MEI monthly snapshots in principle provide historical
data on subsequent releases on a wide range of key
variables
– Provides a coherent set of variables representing the
‘information set’ available at successive monthly intervals

MEI has been published as a CD Rom for many
years
7
OECD interest in a revisions database

Analysis – Are revisions random (i.e. centred around 0)? Are
they in general becoming smaller?

Support internationally coordinated research work to
improve the quality of early estimation methods

Encourage transparency in the statistical process
8
The QNA-MEI revisions dataset

Initial OECD work (Ahmad, Bournot & Koechlin,
2004) was for G7 countries, period 1996 – 2000

Current project (Di Fonzo, STESEG 2005)
– Expand coverage to all OECD countries
– Expand time period covered
– Devise a scheme which can support ongoing production of
real time database & summary revision analysis indicators
– Quarterly GDP and its components
9
Data availability for GDP (constant prices, sa) in the
QNA-MEI revisions database
MEI editions
Vintages
Countries
First
Last
First
Last
#
May-95
May-06
95q1
06q1
133
1: USA
May-95
May-06
94q4
05q4
133
13: AUS, CAN, CHE, DNK, ESP, FIN, FRA,
GBR, ITA, JPN, NLD, NOR, NZL
Nov-95
May-06
95q2
05q4
127
1: DEU
Nov-96
May-06
95q4
05q4
115
1: PRT
Mar-97
May-06
96q3
06q1
111
1: BEL
Mar-00
May-06
99q3
05q4
75
1: KOR
Oct-03
May-06
03q2
05q4
32
1: CZE*
Nov-03
May-06
03q2
05q4
31
1: AUT*
Apr-04
May-06
03q4
05q4
26
1: GRC
Nov-04
May-06
04q2
05q4
19
1: IRL
Jul-05
May-06
05q1
05q4
11
1: HUN
May-95
Sep-00
94q4
00q2
65
2: SWE, TUR
May-95
Sep-00
94q4
00q1
65
1: MEX
Oct-05
May-06
00q1
05q4
8
1: SVK
---
---
---
---
0
3: ISL, LUX, POL
10
QNA-MEI revisions dataset
Caveats for international comparisons

P are not necessarily flash estimates

Timeliness of P differs between countries

No attempt is made to isolate the different types of revisions,
(including errors)

Differences may exist between estimates and national
estimates

No investigation into influence of economic cycle

Period covers SNA93 introduction and other changes, fixed
base, chain-linking
11
Release time of GDP (P) on MEI
Further cautionary notes





MEI is commonly published during the first week of
the month
Different release timing across countries
There is no guarantee that the first number published
in MEI is really the first number ever released
‘QNA-MEI view’ on timeliness in the release of GDP
The true timeliness is ‘filtered’ through the publication
practices adopted by OECD in the last decade
12
Revisions analysis: Summary statistics
1 n
1 n
R    Lt  Pt    Rt
n t 1
n t 1

Mean revision

1 n
1 n
Mean absolute revision MAR   Lt  Pt   Rt
n t 1
n t 1

Mean squared revision

%L>P

% sign(L) = sign(P)
1 n
1 n 2
2
MSR    Lt  Pt    Rt
n t 1
n t 1
13
Elements for discussion

Presentation of revisions, evaluating the first revision
registered on MEI, effect of different time periods on the
summary statistics, stability of the results, use of other
(simple) descriptive measures?

Possible use of a QNA-MEI revisions database by member
countries?

Priority for analyzing other QNA series and how to use the
results?
- e.g. as an input to internationally coordinated
methodological work?
- To reconsider policies for communicating information on
revisions to users?
14
THANK YOU!
15