Aurlie Lemmens

1
The European Consumer United in Diversity?

• Aurélie Lemmens
• K.U.LEUVEN, Belgium

2
De Standaard, 21 April 2005
- About the Consumer Confidence Indicator
- Often cited in the popular press
- Good indicator of the economic climate
and of the households financial situation
- Pan-European predictor ?
3
(No Transcript)
4
The European Paradox
United in Diversity

• Economic divergence
• Legal and regulatory inequalities
• Cultural disparities

• European unification process
• Increasing mobility
• Distance-spanning technologies
• Media globalisation

Multi-Domestic Marketing Strategies
Pan-Regional Marketing Strategy
Pan-European Marketing Strategy
5
Literature Review Mixed Evidence
National or regional consumer segments
Pan-European consumer segments

• Bijmolt, Paas Vermunt (2004)
• Askegaard Madsen (1998)
• Tellis, Stremersch Yin (2003)
• Stremersch Tellis (2004)

• Ter Hofstede, Steenkamp Wedel (1999)
• Gielens Dekimpe (2001)
• Gielens Steenkamp (2004)
• Kamakura et al. (1993)

6
Literature Review Conclusion

• Domain-specific segmentation bases
• Static analyses

7
4 weeks
25 weeks
Bronnenberg, Mela and Boulding (2005)
8
Literature Review Conclusion

• General segmentation basis
• The European Consumer Confidence Indicators
  (CCI)
• Dynamic analysis
• - Spectral measures of dynamic correlation and
  cohesion
• - Allows for different segmentations for
  different planning horizons

9
A General Segmentation Basis
The Consumer Confidence Indicator

• Good predictor of future expenditures, i.e.
• Willingness to buy (Katona, 1951 Eppright et al,
  1998)
• Consumption growth (Nahuis and Jansen, 2004)
• Car sales, retail sales of durables, consumer
  credit ( mood bias) (Bolger, Franses and
  Antonides, 1999)
• Durables (Gelper, Lemmens and Croux, 2005)
• Fashion merchandise (Allenby, Jen and Leone, 1996)

10
A General Segmentation Basis
The Consumer Confidence Indicator

• Good predictor of future expenditures
• Collected by the European Commission
• Publicly available
• Consistently collected over (i) 14 EU countries
  and (ii) a long time span
• US counterpart Index of Consumer Sentiment (ICS)

11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
The Research Questions

• Over various planning horizons
• Compared to the United States

• The homogeneity of the European consumers

• The homogeneity of European segments

• Scandinavian, Midwest and Mediterranean segments

• Does distance still matter ?

• Cultural, economic and geographic distance

15
Methodology Spectral Analysis

• The basic concept of spectral analysis
• Any time series yt can be decomposed into an
  infinite sum
• of components, each having a different frequency ?

16
Spectral Analysis of 2 Time Series
Some Intuition
time
17
Spectral Analysis of 2 Time Series
High Frequency - Short-run variations
Correlation
. . .
. . .
Medium Frequency - Middle-run variations
Correlation
. . .
. . .
Low Frequency - Long-run variations
Correlation
Time
18
Dynamic Correlation (example)
LR
MR
SR
19
Things to Remember...

• The planning horizon is inversely related to the
  frequency

• Time domain
• Short run
• Long run

• Spectral domain
• High frequency
• Low frequency

• A dynamic correlation can be computed for each
  frequency

• Considering n times series, dynamic correlations
  
• can be aggregated as a cohesion measure for each
  
• frequency

20
A Measure of Comovement

• We have 2 stationary series, xi and xj , each of
  length T
• The dynamic correlation between xi and xj at
  frequency ?

21

• The cross-spectrum between and at
  frequency ?
• with

• The spectrum of xi at frequency
• , with

22
Cohesion and Cross-Cohesion

• Cohesion between n1 time series
• Cross - Cohesion between two distinct subsets of
  size n1 and n2

23
Data

• The Consumer Confidence Indicator (CCI) of the
  various European countries
• The Index of Consumer Sentiment (ICS) of the
  various US regions
• The distance measures between the European
  countries

24
The European CCI the American ICS

• Monthly from November 1995 until July 2005 (117
  obs.)
• Fourteen EU countries (excluding Luxembourg)
• Over 30,000 consumers surveyed by the EU
  Commission
• Answers in balance pertaining to the financial
  situation of the households, the general economic
  situation, savings, and employment expectations

25
Extract from the questionnaire

• How has the financial situation of your household
  changed over the last 12 months? It has ...
• got a lot better
• got a little better
• stayed the same
• got a little worse
• got a lot worse
• don't know
• Compared to the past 12 months, do you expect to
  spend more or less money on major purchases
  (furniture, electrical/electronic devices, etc.)
  over the next 12 months? I will spend
• much more
• a little more
• about the same
• a little less
• much less
• don't know.

26
The European CCI the American ICS

• Monthly, same time span
• Four US regions (North-East, Midwest, South and
  West)
• Over 500 households surveyed by the University of
  Michigan
• Similar questions

• Monthly from November 1995 until July 2005 (117
  obs.)
• Fourteen EU countries (excluding Luxembourg)
• Over 30,000 consumers surveyed by the EU
  Commission
• Answers in balance pertaining to the financial
  situation of the households, the general economic
  situation, savings, and employment expectations

27
Data 3 Distance Measures

• Geographic distance Contiguity (Gielens and
  Dekimpe, 2001)
• 4 economic variables (Mitra Golder, 2002)
• Economic size (GDP), economic prosperity (GDPC),
  economic infrastructure (RAIL), economic
  accessibility (DENS).
• 7 cultural variables (Schwartz and Ros, 1995)
• Conservatism (CONS), intellectual autonomy
  (INTEL), affective autonomy (AFFECT), hierarchy
  (HIER), egalitarianism (EGAL), harmony (HARM),
  mastery (MAST).

28
Results

• The homogeneity of the European consumers
• The homogeneity of European segments
• Does distance still matter?

29

• The Homogeneity of the European Consumers
  Confidence

30
Pair-Wise Dynamic Correlations

• In total, 91 (14x13)/2 dynamic correlations
• Illustration with 3 countries France, Germany,
  UK
• Static correlations
• France-Germany 0.135
• France-UK 0.031
• Germany-UK 0.014

31
SR
MR
LR
32
MR
LR
SR
33
(No Transcript)
34

• The Homogeneity in Consumer Confidence in
  a-Priori Segments

35
A-Priori Segmentation Scheme

• The Scandinavian segment
• Denmark, Finland, and Sweden
• The Mediterranean segment
• Greece, Italy, France, Portugal, and Spain
• The Midwest segment
• Austria, Belgium, Germany, and The Netherlands
• Based on Tellis et al. (2003)

36
(No Transcript)
37

• Does Distance Still Matter?

38
Regression Analysis

• We regressed
• Correlation f (geographic, economic, cultural
  distances)

• The static correlation
• The dynamic correlation
• On the short run
• On the medium run
• On the long run

39
Does Distance Still Matter?
40
Does Distance Still Matter?
41
Does Distance Still Matter?
42
Does Distance Still Matter?
43
Conclusions

• Spectral analysis is a useful tool to explore the
  dynamics of the relationship between variables.
• The relationship between variables depends on the
  planning horizon.
• Regarding the European market,
• Short-run variations are purely idiosyncratic.
• Longer-run commonalities emerge in some European
  regions.
• These commonalities are inversely related to
  the (economic, cultural, and to a lesser extent
  geographic) distance.

44
Thanks!