The Consumer Confidence Survey is an obligatory survey for member countries of the OECD (Organization for Economic Co-operation and Development). Since Israel joined the OECD, this survey has been conducted by the Central Bureau of Statistics on a monthly basis.

The survey aims to provide monthly information for a system of indicators that can be used to assess the economy, and is conducted among a representative sample of persons aged 21 and over.

The methodology used to conduct the Consumer Confidence Survey in Israel is delineated in the OECD's guidelines. In the survey, information is gathered from individuals on their general assessments of the economic situation, expectations for change in the situation, and plans for savings and large purchases in the near future. The survey findings can be used by decision makers in the business and public sectors to evaluate and formulate business and economic policy.

The survey questionnaire is short, and relates to three time periods: Appraisals of changes that occurred during the past year, appraisals regarding the current situation, and appraisals of the situation in the coming year.

Participants in the survey are not required to provide quantitative data. Rather, they are asked to provide general evaluations of their expectations for change (increase, stability, or decrease). 



The Survey Population and Sampling Frame


The population of the Consumer Confidence Survey consists of persons aged 21 and over in Israel's permanent population, including residents of non-therapeutic institutions (such as student dormitories, absorption centers, and sheltered housing for the elderly population). In addition, new immigrants are included in the survey population only if they have been living in Israel for at least 18 months.

The Sampling Frame for the Sampling Year 2020

The sample consists of all of those sampled in the Social Survey of the previous sampling year (2018), excluding those who were not in the Social Survey population (mainly for reasons of death, staying abroad for more than one year, or living in a non-hospitalization institution) or those with a permanent disability. 

The Consumer Confidence Survey is a monthly survey, whereas the Social Survey is annual. Therefore, the Social Survey sample had to be divided into 12 months of interviewing. After eliminating those who did not belong to the survey population, the total population of participants in the 2019 survey included 8,906 individuals. This population was divided into 12 approximately equal samples (743-744 each), for each month of the year from March 2020 to February 2021.

The Social Survey sample was divided into 12 monthly samples which were balanced samples in terms of their size and in terms of their main characteristics. The balanced samples were constructed in relation to several background variables.  Four groups by population groups (Ultra-Orthodox Jews, including immigrants who arrived in Israel until 1989; non-ultra-Orthodox Jews, including immigrants who arrived in Israel until 1989; Jewish immigrants who arrived in 1990 and after, including ultra-Orthodox Jews; and non-Jews); two groups by sex (males and females); five groups by age (21-34, 35-44, 45-54, 55-64, and 65+);  six groups by level of education, according to the diploma received (no diploma; secondary school diploma with/without matriculation certificate; tertiary education; first degree; second or third degree; unknown); five groups by net total income per month (up to NIS 4,000; NIS 4,001 – NIS 6,000; NIS 6,001 – NIS 10,000; above NIS 10,000; unknown).

For those who did not respond to the Social Survey but fit the definition of the survey population, only some of the variables were known. Therefore the population was balanced in relation to the following variables: Two groups by sex (males and females), five groups by age (21-34, 35-44, 45-54, 55-64, and 65+); three groups by population groups (Israeli-born Jews, new immigrants, and non-Jews), and four non-response groups (refusal, not located, absent, not able to be interviewed or unknown reason). 


 The Survey Population and Sampling Frame for 2011-2019



Estimation Method

The estimation method is designed to reduce sampling errors as well as biases that emerge because non-respondents might have different characteristics from respondents.

In order to obtain estimates relating to the entire survey population, a "weight" was determined for each individual who was interviewed. The weighting coefficient of a respondent reflects the number of persons in the population who are represented by this individual.

The system of weighting coefficients was determined in a multi-stage process using the "raking" method. In that process, the weighted sample distribution is adjusted to a number of external distributions, according to selected distribution variables.

The adjustment was conducted separately for two population groups – persons living in Jewish and mixed localities, and persons living in non-Jewish localities – by the following distribution variables:

  • Age and sex
  • Labour force characteristics (worked/did not work in the past week)
  • Education (academic/non-academic)
  • Homogeneous groups by level of net household income[1]


During the Coronavirus crisis between March and June 2020, substantial changes were observed in the respondents' assessments over the calendar month, together with non-uniform enumeration over the days of the month. These two phenomena create biases in the survey estimates. In order to reduce them, adjustment was conducted to balance the number of respondents between the beginning and later part of the month.

In addition to the monthly estimate described above, the calculation of biweekly estimates began in July 2020. The method used for calculating the biweekly estimates is similar to that use to calculate the monthly estimates, with the required matching. Since there are fewer respondents, the weighted sample distribution cannot be matched to the number of external distributions, as is done in the monthly estimates. The adjustment is therefore done separately for two population groups – persons living in Jewish and mixed locality types, and persons living in non-Jewish locality types – according to the following distribution variables:

  • Age and sex
  • Education (academic/non-academic)


Balance Estimates

The participants' responses to the survey questionnaire were translated into values ranging from -1 to 1, in which a positive trend is represented by positive values, a negative trend is represented by negative values, and zero represents a lack of change. For example, there were five possible responses to the question dealing with Israel's future economic situation: The response "it will get a lot better" was assigned a score of 1, the response "it will get a little better" was assigned a score of 0.5, the response "it will stay the same" was scored 0, the response "it will get a little worse" was scored -0.5, and the response "it will get a lot worse" was scored -1.

The balance estimate represents the difference between the positive responses and the negative responses, as a percentage of the total respondents to the survey.

The balance estimate is calculated by weighting each participant's responses according to a weighting coefficient. The estimate ranges from -1 to 1, and when it is expressed as a percentage it ranges from -100 to 100.

The formula for the balance estimate is:


formula.png

Whereas:

i – symbolizes an individual respondent.

wi – symbolizes the weight of the individual i.

xi – symbolizes the response to a particular question of individual I.


The New Consumer Confidence Indicator (Weighted Balance)

In January 2019, a new official EU consumer confidence indicator was chosen to replace the indicator that was previously in use.

The new Consumer Confidence Indicator is the arithmetic mean (in equal weights) of four balances , which were calculated for four questions selected from the questionnaire (questions B and C remain the same as for the CCI published in the media releases up to the December 2019 data):

A. Changes in the household's financial situation over the last 12 months: 

B. Expected changes in the household’s financial situation over the next 12 months.

C. Expected changes in the general economic situation in Israel over the next 12 months.

D. Intent to make major purchases over the next 12 months, compared to the last 12 months:

The previous CCI published in CBS media releases up to the December 2019 data was the arithmetic mean (in equal weights) of the four balances , which were calculated for four questions selected from the questionnaire: 

A. Expected changes in the household's financial situation over the next 12 months.

B. Expected changes in the general economic situation in Israel over the next 12 months.

C. Expected changes in unemployment over the next 12 months.

D. Likelihood of the household putting money into savings over the next 12 months. 

The New Consumer Confidence Index

The new Consumer Confidence Index, which is derived from the new Consumer Confidence Indicator and is similar to it (a weighted balance), is based on four balance estimates that relate to the following: changes in the household's financial situation over the last 12 months, in the household's financial situation over the next 12 months, in the general economic situation in Israel over the next 12 months, and intent to make major purchases over the next 12 months compared to the last 12 months. Every month, each one of the four balance estimates is standardized to a uniform scale by means of a calculation based on a base period (March 2011 to February 2012). The index is obtained by averaging the standardized balance estimates, then multiplying by 10 and adding 100.

A value of 100 means that the population is neither optimistic nor pessimistic relative to the base period.

Values less than 100 indicate that the population is pessimistic; the lower the index's value, the more pessimistic the population, compared to the base period.

Values greater than 100 indicate that the population is optimistic; the higher the index's value, the more optimistic the population, compared to the base period.

The formula for calculating the Consumer Confidence Index is as follows: 


formula2.png


The previous Consumer Confidence Index, published in media releases up to the December 2019 data was derived from the previous Consumer Confidence Indicator.

The data collection method: The Consumer Confidence Survey is conducted by the Central Bureau of Statistics (CBS) by means of telephone interviews.

Frequency of data collection: This is an ongoing monthly survey, which is conducted from the first of each month until the end of the month. 




[1]  The sampling distribution by this variable has been conducted since June 2011.