Download presentation
Presentation is loading. Please wait.
Published byNicolas Hase Modified over 6 years ago
1
Task Force on Seasonal Adjustment of Quarterly National Accounts
Second Meeting Item 4 Case Study Seasonal & Calendar Patterns – First Results Henning Ahnert Frankfurt, April 2007
2
2 Background TF mandate: “examine further, for selected components of ESA Table 1, the underlying causes of the significant differences in the seasonal and working day correction factors currently used”. Questionnaire sent to TF: 13 replies received (thank you!), five replies pending (BE, GR, MT, SE, UK) Summary will be part of final Task Force Report
3
Structure of presentation
3 Structure of presentation I: Value added NACE C-E, G-I, J-K II: Household and Government consumption III: Value added NACE F and GFCF IV: Summary
4
Task Force discussion What are common features?
4 Task Force discussion What are common features? What causes differences? What are economic reasons? What are statistical reasons? What might affect comparability of QNA raw data, and in particular adjusted QNA ?
5
I. Value added – Seasonal factors
5 I. Value added – Seasonal factors Value added NACEC-E Value added NACE G-I NACE J-K Q1 Q2 Q3 Q4 EU25 100.7 100.1 96.3 103.1 96.8 100.4 101.2 101.6 98.9 99.7 99.0 102.4 CZ 101.7 103.4 93.8 92.8 107.6 103.0 98.7 99.2 100.6 94.3 105.3 EE 99.3 100.2 96.2 104.2 94.4 103.8 108.4 101.5 98.3 HU 95.6 99.4 98.6 106.2 87.3 101.4 102.6 93.7 100.5 106.9 SK 103.9 102.2 97.2 97.0 92.4 112.6 108.8 85.9 106.5 98.5 95.2 FI 95.9 99.8 101.8 94.9 100.9 97.4 97.3 104.7 SE 88.5 105.1 95.1 102.5 96.7 105.7 95.3 102.7
6
I. Value added - Sources 6 CZ EE HU SE SK FI NACE C-E
Quarterly enterprise surveys for output and intermediate consumption Monthly Production indices; Industry turnover Production at current prices deflated Surveys (q) and administrative data (m/q) Main source is deflated turnover (production index for NACE E) NACE G-I See before Retail index (m); car registrations (m);nights spent * average costs; Turnover and transport statistics Deflated turnover NACE J-K Profit and loss account for NACE J (q) ; Production; dwelling new orders Profit and loss account for NACE J (q) ; Statistics on credit institutions for NACE J Turnover statistics and rents for NACE K
7
I. Value added – Seasonal effects
7 I. Value added – Seasonal effects CZ EE HU SE SK FI NACE C-E Peak in Q2 due to lower intermediate consumption to output ratio; Trough in Q3 due to NACE E Main impact from strong peaks of NACE E in Q4/Q1 Trough in Q3 due to summer vacation Q4 peak due to Christmas NACE E in Q1 (heating season) reparations (= intermediate consumption) in Q3) NACE C very low in Q1 and not compensated by peak of NACE E o NACE G-I Peak in Q 2 due to lower intermediate consumption to output ratio, and vice versa in Q4 Q4 Christmas consumption; Q3 tourism expenditure Retail sales before Christmas, counter reaction in Q1; Lower income implies higher Christmas peak; Lower income implies only one holiday season per year in summer Special seasonal factors due to NACE G due to [fiscal changes?] Retail sales high before Christmas and low in Q1; Q3 (Q1) tourism peak (trough) NACE J-K Trough in Q3 due to high ratio of intermediate consumption Trough in J in Q4 due to high costs Similar pattern as for G-I due to similar factors, mainly in NACE 74 (other business activities) NACE J peak in Q1 (compensation); trough in Q4 (insurance payments)
8
I. Value added – Seasonal effects
8 I. Value added – Seasonal effects Industrial Production NACE E (energy) - seasonal factors (period averages, )* CZ EE HU SK SE FI Quarter 1 113.3 121.8 122.3 114.1 116.3 Quarter 2 93.2 83.0 86.3 90.0 92.2 92.0 Quarter 3 89.3 81.6 79.3 85.6 83.9 84.1 Quarter 4 103.9 113.8 112.2 111.1 109.2 109.9 Value added by NACE sections, (current prices, percentages of whole economy) C D E F G H I J K CZ 1.5 25.9 4.2 6.6 12.4 2.0 10.3 3.3 13.8 EE 1.0 16.8 3.4 7.3 15.2 1.6 12.1 3.8 18.8 HU 0.2 22.2 2.9 4.8 10.9 7.6 4.6 17.5 SK 0.6 23.4 4.9 6.8 14.4 1.3 10.4 14.3 SE 0.3 19.7 3.0 4.5 10.7 8.0 19.8 FI 23.1 2.1 5.9 2.3 18.5
9
I. Value added – Seasonal effects
9 I. Value added – Seasonal effects Czech Republic - Ratios intermediate consumption to gross output Averages NACE C,D,E NACE G, H, I NACE J,K Q1 0.723 0.570 0.644 Q2 0.728 0.577 0.651 Q3 0.734 0.579 0.655 Q4 0.754 0.608 0.675
10
I. Value added – Seasonal effects
10 I. Value added – Seasonal effects Slovakia and Estonia - Average seasonal factors of value added (NACE sections) C D E G H I J K Slovakia (average ) Q1 98.15 97.95 142.72 85.70 85.64 100.01 130.82 99.89 Q2 101.02 103.79 90.82 124.06 111.63 99.99 94.01 100.02 Q3 100.10 78.19 116.29 108.97 97.79 Q4 109.28 98.10 90.63 74.08 93.31 99.91 73.61 100.09 Estonia (average ) 101.2 94.0 124.7 91.7 76.7 97.1 99.6 99.8 96.0 105.5 84.3 103.7 111.2 105.6 102.8 102.2 106.2 99.4 75.0 101.7 119.7 103.3 103.4 96.5 96.4 101.0 116.0 102.5 92.1 94.1 94.5 101.4
11
I. Value added – Calendar regressors
11 I. Value added – Calendar regressors CZ EE HU SE SK FI NACE C-E 1 regressor, no leap year correction, CZ calendar No calendar adjustment yet 1 working day regressor, HU calendar Between 0-7 regressors in DEMETRA 2 regressors (weekdays, weekends) NACE G-I Only for retail trade: 1 working day regressor Leap year for retail trade NACE J-K no adjustment for NACE J, No significant calendar effects No calendar adjustment
12
I. Value added – Calendar effects
12 I. Value added – Calendar effects CZ EE HU SE SK FI NACE C-E C: 0.03% D: 0.22% E: -0.50% Country-specific public cause differences to other countries No calendar adjustment yet n.a. C-E: 1.57% D: 1.22% G-I: 1.13% G: 1.56% NACE G-I G: -0.12% H: -0.54% I: 0.04% Effects only in retail trade n.a NACE J-K K: -0.22% No significant effects Example: Calendar factors, averages NACE C,D,E NACE G, H, I NACE J,K Q1 100.42 99.64 99.29 Q2 100.08 99.94 99.97 Q3 100.38 99.76 99.61 Q4 99.63 100.30 100.66
13
Household consumption Government consumption
13 II. Consumption – Seasonal factors Household consumption Government consumption Q1 Q2 Q3 Q4 EU25 97.2 99.4 100.3 103.1 99.2 100.2 98.1 102.6 BE 98.8 101.6 98.9 100.0 DE 96.7 99.5 103.8 98.3 98.2 104.6 ES 87.8 101.3 99.8 110.8 98.5 104.5 FR 99.1 98.4 102.7 100.1 100.9 97.8 101.2 IT 103.5 98.6 101.9 NL 100.6 100.5 99.7 96.1 107.7 94.3 101.8 GB 96.6 103.6 100.7
14
Government Consumption
14 II. Consumption – Sources BE DE ES FR IT NL GB Private Consumption Monthly retail, wholesale, hotels and restaurants turnover Q-crafts reports Car registration Energy reports Tobacco tax reports Enterprise tax data Monthly food consumption Retail sales Passenger transport Other services indicators Q-HBS main source. For food, tobacco and alcohol also retail trade; Turnover statistics for transport, communication Car registrations Overnight stays in hotels M retail trade and commercial/personal services. Energy; travel expenditure. Other services, e.g. medical, transport,communicat. Statistics on housing and population. Turnover of supermarkets and sales of vehicles. Government Consumption Q reports of government and social security statistics National Audit Office provides all data Q data from government, refers to compensation, social transfers and intermediate consumption Q-labour input, compensation and social benefits in kind
15
Government Consumption
15 II. Consumption – Seasonal effects BE DE ES FR IT NL GB Private Consumption Retail trade Christmas and post-Christmas effect Two extra wages in June and Dec, e.g. on vehicle purchase; Christmas and post Christmas effect; High restaurant consumption in Q3, but compensated by low food consumption; Spaniards spend holidays in Spain. Q1: trough due to low tourism and other Q3: trough due to low energy and durable consumption, despite peak in tourism Q4: peak due to Christmas and energy consumption Peak/trough in Q3/Q4 due to hotels/tourism; Clothing peaks in Q3/Q1 due to seasonal sales Higher share for hotels and restaurants may explain differences to other countries At aggregate level the seasonal pattern of the components largely cancels out Government Consumption Bonus Q4&Q3; Funds spent at end-of-year; Anticipated purchases; Effects decrease Two extra wages in June and Dec Due to half-annual allocation of budget Q2 and Q4 are strong 13th month salary in December; Other components less affected; No extra salary in June Holiday allowance in May, end-of year bonus; End-of year expenditures
16
II. Consumption – Seasonal effects
16 II. Consumption – Seasonal effects Retail Turnover - seasonal factors (current prices, period averages )* BE DE ES FR IT NL GB Q1 96.8 94.1 95.7 94.4 90.6 91.7 92.8 Q2 101.4 98.1 98.2 98.6 99.7 102.7 97.0 Q3 98.9 98.4 96.6 Q4 106.7 111.1 107.2 108.5 118.0 107.0 113.7
17
II. Consumption – Seasonal effects
17 II. Consumption – Seasonal effects Italy:
18
II. Consumption – Seasonal effects
18 II. Consumption – Seasonal effects Domestic consumption expenditure by COICOP (current prices, percentages of total) 01 02 03 04 05 06 07 08 09 10 11 12 BE 13.4 3.8 5.4 23.1 4.3 14.9 2.3 9.3 0.5 5.1 12.4 DE 11.4 3.6 24.3 7.0 4.6 13.6 2.8 9.5 0.7 5.3 12.0 ES 14.1 2.9 5.5 16.2 5.2 3.5 11.6 2.6 9.1 1.5 18.9 8.9 FR 13.8 3.1 4.8 24.5 5.8 14.7 9.2 6.1 11.1 IT 14.8 2.7 7.8 20.7 7.7 3.2 7.1 0.8 9.7 NL 10.6 3.0 22.2 6.2 4.5 10.1 15.5 GB 3.9 18.7 6.4 1.8 15.0 12.7 1.4 Source: Eurostat, ECB calculations. Data refer to 2005 except UK (2004)
19
II. Consumption – Seasonal effects
19 II. Consumption – Seasonal effects
20
II. Consumption – Seasonal effects
20 II. Consumption – Seasonal effects
21
Government Consumption
21 II. Consumption – Calendar regressors BE DE ES FR IT NL GB Private Consumption Detailed by sector and for m data; DE working days except in retail trade (Sundays, late night shopping, Easter, public holidays, winter/summer sales) and accommodation (public holidays) Adjusted are m/q source data Trading days, Easter and leap year for m series Separate regressors for seven days, FR calendar Use of groups, leads, lags as appropriate Adj. whenever not statistically rejected and sensible Jul, Aug & Dec tested separately Each indicator adjusted for working days, Easter effect and leap year No calendar adjustment due to short series; Research 07H2 Government Consumption No calendar effect/adjustment Trading days, Easter, Leap year ? No calendar adjustment
22
Government Consumption
22 II. Consumption – Calendar effects BE DE ES FR IT NL GB Private Consumption Coefficients provided; National public holidays and regional holidays may explain differences to other countries 8 fixed public holidays compensated with leave on following Monday if falling on Sunday. Remaining 4 holidays have seasonal character. Proxy: One additional working day: growth One additional working day: -0.17 growth . Indicators q, therefore less siginificant than for production side. Changes in shop opening hours - six regressor model and time-variant regression Government Consumption No calendar adjustment No calendar effect despite adjustment
23
II. Consumption – Calendar effects
23 II. Consumption – Calendar effects Calendar factors for private consumption in Germany, Italy and Spain 04Q1 04Q2 04Q3 04Q4 05Q1 05Q2 05Q3 05Q4 06Q1 06Q2 06Q3 06Q4 DE 100.4 100.2 100.1 100.3 99.6 100.6 100.0 99.9 99.7 IT 99.8 ES 99.1 98.8 98.6 101.0 98.9 97.0
24
Gross Fixed Captial Formation Value added Construction
24 III. Construction & GFCF – Seasonal factors Gross Fixed Captial Formation Value added Construction Q1 Q2 Q3 Q4 EU25 93.2 101.0 98.1 107.5 88.8 101.9 103.2 105.6 LT 75.9 100.3 108.5 113.5 64.2 100.6 123.6 110.5 MT 93.5 104.7 97.9 104.3 n.a. n.a AT 80.1 102.1 106.8 109.2 65.6 104.4 115.5 114.1 PT
25
Value added in construction (NACE F) Gross fixed capital formation
25 III. Construction & GFCF – Sources GR LT MT AT PT Value added in construction (NACE F) Construction work; lagged building permits, employment and investment; Q business survey Industrial production Cement and steel turnover Gross fixed capital formation Production of capital goods, export/imports Construction investment Production of capital goods, export/import Cement turnover Productin and import of machinery and metal products
26
Value added in construction (NACE F) Gross fixed capital formation
26 III. Construction & GFCF – Seasonal effects GR LT MT AT PT Value added in construction (NACE F) Climatic conditions (very low factor in Q1) Very low seasonal factor in Q1 due to climatic conditions; counter effect in other quarters Trough in Q4/winter (Oct-Jan) and Q3/Summer holidays; No unadjusted QNA Gross fixed capital formation High share of construction output and its climatic dependency; weight of machinery and equipment increasing Climatic conditions e.g. for construction See above Shares of construction across countries (%) GR LT MT AT PT Value added NACE F in total value added 7.6 4.8 6.3 Construction investment in total investment 53.5 60.7 55.3 56.4 50.8
27
Value added in construction (NACE F) Gross fixed capital formation
27 III. Construction & GFCF – Calendar regressors GR LT MT AT PT Value added in construction (NACE F) QNA directly adjusted with two regressors: Working day and leap year QNA directly adjusted; regression on Mo-Fri & Sa-Su as implemented in DEMETRA No calendar adjusted QNA; Compilation to start in late 07/08 Gross fixed capital formation See before
28
Value added in construction (NACE F) Gross fixed capital formation
28 III. Construction & GFCF – Calendar effects GR LT MT AT PT Value added in construction (NACE F) No significant calendar effects; Public holidays falling on weekend are compensated by extra leave on following Monday For GDP as a whole: 2 extra working days add 0.1% to GDP; Tourism sector reduces effect on GDP. Some underestimation due to NA construction possible (when activity estimates derived from employee data) Not applicable due to the unavailability of calendar adjusted QNA data Gross fixed capital formation No significant calendar effects Public holidays falling on weekend are compensated See before
29
29 Summary (I) Common determinants of seasonal components across countries, e.g.: Weather/climatic conditions (for NACE E, construction and tourism) Bonus payments/extra salaries (for HH and Gov consumption) (and some industries?) End-of year expenditures (for HH and Gov consumption, GFCF) But of course different magnitudes across countries
30
Summary (II) Compositional effects, e.g.:
30 Summary (II) Compositional effects, e.g.: Share of tourism sector varies Share of construction in whole economy and in GFCF rather similar Income difference effect, e.g. lower income imply higher seasonal effects for consumption Aggregation effect, e.g. differences in components may or may not cancel out
31
31 Summary (III) Common determinants for calendar component as well, but magnitude and sign differ a lot Working days, weekend days, public holidays, leap year Compositional differences and Leave compensation for weekend holidays
32
Summary (IV) Statistical effects:
32 Summary (IV) Statistical effects: M/Q source data determine seasonal and calendar component Some extreme differences, e.g. “true” value added indicators vs. material purchase proxies used to compile QNA value added But even two equally suitable sources might show different S and C components (e.g. retail trade vs. HBS?)
33
Summary (V) Statistical effects:
33 Summary (V) Statistical effects: M vs. Q frequeny matters for calendar adjustment. Many sources still quarterly Different definitions of calendar component National calendars not always used No calendar adjustment in some countries Resource and software constraints Further work on recommendations for calendar adjustment warranted (new ESA TP requires calendar adjustment !)
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.