Two indicators for Dutch consumer confidence
Each month Statistics Netherlands (CBS) published figures on Dutch consumer confidence. Likewise the European Commission (EC) publishes monthly consumer confidence figures for the all countries of the European Union (EU) including the Netherlands. The EC calculates consumer confidence for all EU countries in a uniform manner on the basis of data gathered by the countries themselves.
Different outcomes
The outcomes of the CBS and the EC indicators on consumer confidence differ. In the last four years especially, there was a clear difference in levels. EC confidence was considerably higher than CBS confidence. The cause lies in different definitions, calculations and source data. The developments shown by the two indicators are roughly the same. The differences are explained in this article, using the data as originally observed and not the seasonally corrected data.
Consumer confidence CBS and EC,
January 1994 - February 2006
Definition
The first difference comes from the questions on which the indicators are based. A selection of the sixteen questions in the Consumer Survey is used to calculate consumer confidence. For the EC indicator these are only questions referring to expectations for the future. The CBS indicator also includes opinions consumers have about the current situation and the situation in the previous twelve months. The EC indicator has questions about various areas; expectations about unemployment and savings are also included.
The questions
Consumer confidence according to the CBS definition is based on the following five questions:
- Opinion about the economic situation in the previous 12 months;
- Opinion about the economic situation in the next 12 months;
- Opinion about the financial situation in the previous 12 months;
- Opinion about the financial situation in the next 12 months;
- Opinion if the time is favourable for major purchases.
The first two questions together are the opinion about the economic climate. The last three questions together deal with willingness to buy.
EC consumer confidence is based on the following four questions:
- Opinion about the economic situation in the next 12 months;
- Opinion about the financial situation in the next 12 months;
- Opinion about the unemployment in the next 12 months;
- Opinion about the possibility to save in the next 12 months.
The observation for both indicators is the same. All questions of the consumer survey are asked to a thousand households each month. Each month new households are selected. The survey is conducted by telephone with in the first ten working days of each month. The questions refer to the economy in general and to the household’s own financial situation. Consumers indicate whether they feel something goes better, worse, that the situation is unchanged or that they don’t know.
Calculation
Not only the questions are different. The calculations are different as well. CBS calculates the difference between the percentage of optimists (improved) and the percentage of pessimists (worsened). In the European calculation a further distinction is made between slightly improved and improved, or slightly worsened and worsened. Consumers who feel that something is only slightly better or worse are counted as half. Consumers who feel the situation is unchanged or have no opinion are not included in the calculations.
| Economic situation next 12 months, | % of the | |
|---|---|---|
| February 2006 | consumers | |
| Clearly improved | 7 | |
| Slightly improved | 27 | |
| Unchanged | 38 | |
| Slightly worsened | 12 | |
| Clearly worsened | 9 | |
| Don't know | 7 | |
| CBS-calculation (7+27)-(12+9) | 12 | |
| EC-calculation (7+0,5*27)-(0,5*12+9) | 5 | |
| 1) | The calculations are based on non-rounded figures. These may cause slight differences. |
The other indicators that make up consumer confidence are calculated in the same way. These are averaged in order to get to a total indicator.
Source data
The third difference between the two indicators concerns the data used. Between October 1995 and the end of April 2000 the data for the EC indicator came from Bureau Intomart. Before and after the EC used CBS data. Statistics Netherlands always used its own CBS data.
Clearly an effect of the difference in definition
The differences in the questions on which the indicator is based, clearly affects the level of consumer confidence. The question about savings is only used in the EC indicator. The indicator always has a high value. Therefore the European harmonised confidence is higher than Dutch confidence. The difference increases as confidence decreases. The difference is evident in the last four years.
The figure below shows the indicators that are not included in both. They are calculated according to the CBS method. The indicators included in both, are not represented in this figure.
Consumer confidence indicators according to the CBS calculation method, May 2000 – February 2005
More extreme results due to calculation differences
Because Statistics Netherlands also count consumers who feel that something is slightly better or worse for the full 100 percent, the CBS indicator usually has more extreme results. This is shown by the indicators included in both, such as the opinion about the household’s financial situation in the next twelve months.
Indicator financial situation in the next twelve months,
January 2001 – February 2006
The effect of the different source data
The difference in source data has consequences for the level of consumer confidence. The CBS figures for the period October 1995- April 2000 were structurally lower than the Intomart figures used by the EC. The figure below shows the results as published by the EC, and the results as they would have been if the EC had used CBS data for the period October 1995-April 2000. The level of the European indicator is higher in this period, while the developments are similar.
Consumer confidence according to the European definition and calculation method, October 1995 – April 2000
Finally
In the course of 2001 the European Commission changed to the current definition and calculation of consumer confidence and revised its entire series. Before then the CBS and EC indicators were the same in these areas. The reason for the change, according to the EC, was that the new indicator has a slightly better indicative value for European consumption. CBS did not follow for continuity reasons and because the CBS indicator, particularly willingness to buy, has a high indicative value for Dutch consumption. Moreover the structure of consumer confidence would be lost since it would no longer be made up of willingness to purchase and the opinion on the economic climate.
The two indicators for consumer confidence each have their advantages and disadvantages. The advantage of the EC indicator is its international comparability. The advantage of the CBS indicator is that it is not just based on expectations for the future, but also on the consumer sentiment about the present and the past. Therefore the CBS indicator provides a wider picture and expresses the effects of unexpected events. Also, there are no breaks in the CBS series caused by switching to different source data.