Transcript S3.1b

Principal European Economic Indicators
Flash Estimates of European Aggregates
Eurostat
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Flash Estimates of European Aggregates
 Increasing interest in timely short-term indicators for
economic and monetary policy purposes
 Principal European Economic Indicators (PEEIs)
initiative
 In this framework: flash estimates of European key
indicators:
– Monetary Union Index of Consumer Prices (MUICP) [T+0]
– Gross Domestic Product (GDP) [T+45]
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Principal European Economic Indicators
PEEIs:
 Five sets of key indicators: consumer prices, national
accounts, short-term business, labour market and external
trade indicators + housing indicators (review 2008)
 Targets: availability, timeliness, quality
 Moving focus from Member States figures to the indicators for
the euro area (and for the European Union)
 Improvements and actions: flash estimates, harmonisation
and coordination (methodology, seasonal adjustment, release
calendars, revision policy, data exchange, communication)
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
PEEIs approach
 Identification through a wide-ranging consultation of users
 High attention from political bodies to monitor the progress
 Direct involvement of high level management
 Work at technical level to sustain the approach and promote
the actions
 Supporting legislation
 Methodological developments (chain linking in quarterly
accounts, quality aspects in consumer prices, etc.)
 Benchmark with main economic partners, in particular US
 Enhanced communication strategy
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Flash estimates (1)
A flash estimate is an early estimate produced and
published as soon as possible after the end of the
reference period, using a more incomplete set of
information than the set used for traditional estimates
 With respect to a regular estimate:
–
–
–
–
–
27-29 May 2009
better timeliness
less accuracy, but acceptable
less coverage
based on a more limited set of information
estimation method, more econometrics
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Flash estimates (2)
 Flash estimates are not forecasts or leading indicators:
–
–
–
flash estimates refer to a past reference period (i.e.,
already finished);
flash estimates use the information related to the
reference period;
flash estimates rely, as much as possible, on the
compilation techniques used for the successive
traditional first estimates
Approach followed in Eurostat
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Quarterly GDP and national accounts (1)
 Method: sophisticated aggregation (temporal
disaggregation techniques) – breakdown of annual
aggregates according to the quarterly pattern
 Input: annual totals and quarterly available information
from Member States (for flash estimates, countries’ flash
estimates
 Dimension: univariate (temporal coherence) and
multivariate (accounting consistency)
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Quarterly GDP and national accounts (2)
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Quarterly GDP and national accounts (3)
 Target variable for the flash estimate:
–
–
–
–
–
–
27-29 May 2009
GDP
quarterly growth rate on the previous quarter
euro area and the European Union
seasonally adjusted and working day corrected
volume measure
as published by Eurostat
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Quarterly GDP and national accounts (4)
Specific issues
 Methodology: chain-linking (introduced in 2005)
 Seasonal adjustment: indirect (SA data from Member States)
 Revision policy: the entire time series is revised once a new
quarter is released except for flash estimates
 Enlargements: most recent composition and evolving
composition
 Common currency: conversion with variable exchange rate to
euro
 Lack of basic information for flash estimates (use of available
indicators – mainly the Industrial Production Index)
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Harmonised Index of Consumer Prices (1)
 Method: weighted average of national HICPs
 Input: HICPs and households final consumption
expenditure (weights) from Member States
 Dimension: univariate
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Harmonised Index of Consumer Prices (2)
 Target variable for the flash estimate:
–
–
–
–
27-29 May 2009
MUICP
index
euro area
as published by Eurostat
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Harmonised Index of Consumer Prices (3)
Specific issues
 Methodology: weighted average of national HICPs
 Seasonal adjustment: not applied
 Revision policy: fixed by legal act
 Enlargements: evolving composition
 Common currency: affects only weights
 Lack of basic information for flash estimates (use of
price information, energy prices, regression and time
series modelling)
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Performance (1)
 Evaluation criteria:
– ability of the flash estimates to be a good predictor of the
first regular release figures;
– accuracy and reliability
– timeliness (respect of calendar release)
 Instruments:
– revision analysis
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Performance (2)
GDP
 Compilation method: the same as 1st regular estimate
 Coverage: between 80%-95% of the euro area
 Performance:
– Average revision with respect to regular: 0.01
– Correct anticipation movements: 28 over last 32 quarters
– Growth rate:
27 correct estimation
5 times error of +/- 0.1
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Performance (2)
MUICP
 Compilation method: the same as 1st regular estimate
 Coverage: 95% total consumption expenditure weight
 Performance:
Over the last two years
– 17 correct estimation of inflation rate
– 7 times error of +/- 0.1
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
GDP flash estimates 2001 vs. 2007
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
MUICP flash estimates 2001 vs. 2007
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Future developments
 Promote flash estimates and improve their quality and
availability
 Coordination between Eurostat and Member States
 Coordination between different areas of national
accounts and with basic statistics
 30-60-90 days cycle for National Accounts
 GDP flash estimates at T+30 days
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa
Thank you for your attention!
27-29 May 2009
International Seminar on Timeliness, Methodology and Comparability of
Rapid Estimates of Economic Trends - Ottawa