Transcript Methods used by OECD countries to estimate quarterly changes in

Methods used by OECD countries to
estimate quarterly changes in
inventories from survey data
Working Party on National Accounts
3-5 October 2007
Charles Aspden, OECD
Contribution of changes in inventories to quarterly seasonally adjusted GDP
volume growth1
(%)
Country
2005Q3 2005Q4 2006Q1 2006Q2 2006Q3 2006Q4 2007Q1 2007Q2
Australia
-0.2
0.1
-0.2
-0.7
0.1
0.5
0.6
-0.4
Canada
0.1
0
-0.2
0.5
-0.3
-0.9
0.2
0.1
France
-0.3
0.4
-0.5
0.7
-0.2
-0.1
-0.2
0.1
Germany
-0.2
0.6
0
-0.4
0.1
-1.7
1.9
-0.6
Italy
-0.5
0.5
-0.1
-0.1
1.1
-0.1
-0.6
-0.1
Japan
-0.2
0
0.1
0
0.1
-0.1
0
-0.1
United States2
-0.1
1.7
-0.5
0.5
0.1
-1.3
-0.7
0.2
Why the survey was undertaken
• Follow-up to last year’s survey, which was
primarily concerned with volume estimation.
• Changes in inventories are important:
– GDP(E)
– Output for GDP(P)
– IVA for GDP(I)
Structure of report
• 4 parts plus appendix containing country
responses
– Overview of methods
– Survey methods
– Alternative methods
– Combined approach
Six different ways of estimating
changes in inventories
a)
b)
c)
d)
e)
f)
Enterprise surveys to obtain levels
Data from other sources, e.g. marketing boards
Commodity flow at product level
Residual at aggregate level (GDP(P or I)-GDP(E))
Model using some kind of indicators
Qualitative data (used in conjunction with (d)
and (e))
Overview of methods
• 20 countries responded
• 13 use enterprise surveys to obtain inventory
data for at least some industries, usually
manufacturing, wholesale and retail
• Some of remaining 7 use commodity flow
exclusively, or almost exclusively
• Some use d, e and f
Table 2 Summary of methods used by the 13 countries using surveys for some industries
ES – enterprise survey CF – commodity flow M – model OS – other source
Country
Ag &
forest
Australia
CF, M
& OS
Canada
CF
Czech Rep. ES
Denmark
Finland
Hungary
Iceland
Ireland
Japan
Korea
New
Zealand
Slovak
Rep.
US
OS
CF
ES
ES1
ES
OS2 &
CF
CF
M&
OS
ES
CF, M
& OS
Min
Manf
EG
Cons
ES
ES
ES
ES, OS
ES
ES
ES
M&
OS
CF
ES
ES
CF
ES
ES
ES
ES1
ES
ES
ES3
ES
ES3,4
M
ES
ES
OS
ES
ES
ES
OS
OS
ES
Whol
trade
ES
Retail
trade
ES
ES
ES
ES
ES
Trans. Acco & Othe
& stge food
M & OS ES
M&
OS
ES
CF
ES
ES
ES
ES
ES
ES
M
ES
ES
ES
ES
ES
ES
ES
ES
ES
OS
M
OS4
ES
CF
ES
M
ES
ES
ES
ES
ES
ES
ES5 &
M
ES
M
ES
ES
M
M
M
M
ES
ES
M
ES &
OS
OS
Accuracy of survey estimates
• Some countries do not survey small
businesses. Biased estimates, unless
adjustments made.
• 5 countries reported sampling errors
• Australia and NZ reported SE on movement
• Cannot forget non-sampling errors
Table 3a Standard errors of inventory estimates, Australia
Industry\
statistic
Manufacturing
Retail trade
Wholesal trade
SE of level
$411m
$605m
$776m
Relative SE of SE of
level
movement
% of GDP1 in
1Q07
1.0%
2.2%
2.2%
$264,589m
0.10
0.09
0.23
$257m
$241m
$601m
Table 3b Standard errors of inventory estimates, New Zealand
Industry\
[Absolute]
Standard
error of
level
Inventory
level
Manuf. raw
materials
Manuf.
finished
goods and
work in
progress
Retail trade
$123.5m
Wholesale
trade
statistic
Relative
standard
error of
level
Standard
error of
movement
Percentage
of GDP1 in
1Q07
$2,872m
4.3%
$202.0m
$7,773m
2.6%
1.5%
($43m)
1.0%($78m) 0.19
$143.9m
$5,328m
2.7%
$237.4m
$9,497m
2.5%
$41,856m
0.10
1.5%
($80m)
1.3%
($123m)
0.19
0.29
Deflation and revaluation
• General approach:
– Deflate opening and closing inventories book
values
– Closing level minus opening level at constant
prices
– Revalue to average prices of the period
• Deflators should match the book values –
scope, valuation basis, timing
• Laspeyres indexes – deflate at detailed level
Deflation and revaluation cont.
• Most countries conduct their inventory
estimation by activity – mainly 2-digit, but
some 1-digit, and others more detailed
• Japan and Korea do so by commodity
• Re-weighting annually or five-yearly
Construction of deflators
• Manufacturing
– Output – output PPIs
– Work in progress and materials – PPIs, Wis, IPIs
• Wholesale – PPIs, CPIs, IPIs
• Retail – PPIs, CPIs
Timing of deflators
• Determined by business accounting methods
and turnover periods
• US has provided a detailed description of how
it lags its deflators.
• Evidence that stocks-to-sales ratios are low
and getting lower, implying deflators should
be timed to be close to the end of the quarter.
• Care needs to be taken with high value
products subject to rapid price change.
Other methods
• Germany
– Qualitative surveys
– Production and sales data
– Assumption that enterprises desire to maintain
stocks of finished goods in proportion to output
– Consider other data
Other methods
• Belgium (as reported to WPNA in 1999)
– Qualitative surveys
– Econometric model
A combined approach
• Combined approach
– Survey estimates
– Commodity flow balancing
– Any other pertinent information (such as from a
model)
• If adjustments made to changes in inventory
estimates from a survey. Use estimates of
inventory levels as guideposts.
Summary
• Changes in inventories are major contributors
to quarterly GDP growth
• Warrant a good deal of attention and effort to
produce best possible estimates with the data
available
• Best to use survey estimates in conjunction
with commodity balancing at product level
(together with any other pertinent
information)
• Survey levels as guideposts