Impact Of Sources Of Information On Consumer Confidence
Purpose: The purpose of this paper is to investigate the potential impact the consumer’s trust source for information has on factors of consumer confidence. The study here uses the consumer’s perceptions of existing business conditions, buying conditions for consumer durables and their personal finances as independent variables and consumer confidence as the dependent variable. ‘Sources of Information’ is used as the moderator and the moderating influence of ‘Sources of Information’ on the relation between each independent variable and the dependent variable is then investigated.
Design/methodology/approach: A questionnaire was administered on the target population of working professionals of age group 20-40. The results of this survey are analyzed using moderated regression analysis.
Research limitations/implications: A self-reported, cross-sectional questionnaire was used to collect all measures. Additionally, this study used a single sample. Future research needs to obtain more diversified samples and continue to expand current research by examining additional areas of finance and economics.
Practical Implications: This study can be used as a reference by the media and press to find out the areas in which they need to focus and improve so as to gain more trust from the public. It can also be used to enhance the calculation of actual consumer confidence scale.
Originality/value: The findings of this paper will help further research on the often overlooked effect that media has on consumer confidence in a country. By looking at consumer confidence from an Indian perspective, the findings of the study will help marketers, advertisers, government agencies and financiers find the best channel for focusing their efforts in reaching out to consumers and how to develop the best strategy to influence consumer sentiments.
Keywords: Consumer Confidence, Buying Conditions, Business Conditions, Personal Finance, Consumer Sentiments
Type of Research: Descriptive Research by Questionnaires on a conveniently selected sample
Consumer confidence has always been considered a key indicator of economic growth. It is often used as a predictor for household consumption and expenditure. It is understood that consumers who are confident about the current and future state of the country and their own ability to generate finances are likely to increase consumption. In a country like India, it is very important to understand the metrics driving consumer confidence because consumption expenditure accounts for over 60% of India’s GDP.
The Consumer Confidence Index gives us a great perspective on the state of the economy and its prospects for growth. The index often forms the cornerstone of financial planning for many commercial, economic and budgeting activities. Therefore is onus is not just to look at the index as a number but, should try to reason out the basis of the metrics used to calculate it.
While the forces that drive consumer confidence were never completely explained, its role in business cycle analysis has increased recently. There is a large consensus today over the power of this index in modeling real consumption in the economy. In fact, many authors have showed that the inclusion of the index have improved forecasts of real consumption. (Carroll C. F., 1994)
Since most consumers are aware of their own economic state and the recent changes to it, it is not unexpected that confidence surveys reflect the current state of the economy. Recent research by (Bram, 1998) has suggested that there is a link between consumer confidence and future saving patterns of the consumers. So the importance of consumer confidence in economic planning cannot be ruled out.
Forecasting of all macroeconomic variables depend on how accurately household consumption can be modeled and forecasted. Years of research have shown that income, wealth and interest rate movement help to account for the changes in consumption. As a complement to these macroeconomic variables, University of Michigan’s Survey Research Center collects data from households their views on economy’s recent, current and future economic and financial conditions. These surveys are intended to provide a measure of how willing the consumer is to spend. The best known measure for consumer confidence in the United States is the Index of Consumer Sentiment (ICS) (Wilcox, 2007).ICS is still the most popularly used index for economic forecast in the United States due to the large sample size and the long time series data that is available to researchers. (Carroll C. D., 1994)
The most widely used consumer confidence index in India is maintained by CNBC TV 18 Boston Analytics. Data for this index is collected from 10,000 respondents from 15 cities in India through interviews. The cities are chosen to represent India’s ethnic, regional and other heterogeneities. This is the largest confidence index in the word in regards to its sample size. The sample aims to capture information about the major contributors of personal consumption in Indian GDP. The index questionnaire contains twenty five questions covering many variables that are known to affect consumer confidence. As a measure for weeding out the biases to responses, diffusion values for each question are calculated. Then the diffusion value is weighted on the importance of the variable in the model. After all this is done, the final index is calculated as the weighted sum of the diffused values of the current and the base year survey. The Index is adapted for Indian conditions to include factors such as heterogeneity in the Indian population, regional dispersion as well as factors peculiar to India like high saving rate and social linkages.
While random sampling might weed out biases, we believe that source of information a consumer bases his judgment on will shift his/her opinion to that professed by the source (Kurpis, 2004) or reaffirm his existing bias (Burke, 2008) –be it – print media, Television Media or even FRAs (Friends, Relatives and Associates).We wish to simulate a scale of consumer confidence and test the effect of source of information on consumer confidence
Given below is a diagram depicting the components of the CNBC TV-18 Boston Analytics Consumer Confidence Index.
Consumer Confidence Index
Consumer Confidence Level The most important feature of the confidence index is probably the leading information it contains with respect to the main determinants of real consumption: it is the first published information on the consumer behavior. As such, most economic agents integrate it in their forecasting (for instance, stock markets react very quickly to the release of the confidence index, as it gives an indication of future consumption and thus future growth; in the same way, central bankers decisions to modify the monetary policy is partly based on the evolution of this index), The knowledge of the parameters affecting this index might allow to infer better its importance in terms of consumption. (Bram, 1998)
Consumer Confidence of any consumer as adapted from University of Michigan’s Index of Consumer Sentiments is a construct of:
Business Conditions and
Buying Behavior of the consumer
Personal Finance captures the financing options available to the consumer, the expected future changes in real family income of the consumer and the expected trend in the financial condition of the consumer. (Carroll C. D., 1994). Personal Finance indices will have high degree of correlation with Consumer Confidence Index as demonstrated using University of Michigan’s Index of Consumer Sentiments. When opportunities for personal financing options are higher to a consumer, more likely he is to make purchases. Also easy availability of credit will increase credit off-take by consumers as well as reduce saving rates as the precautionary use of money is reduced. Personal Finance Index therefore forms an important part in a consumers’ confidence. Consumer spending is explained partly by consumer attitude. The higher the income, higher will be the consumer confidence (McIntyre, 2002) Spending by consumers depends on availability of credit and income prospects. The perception of the consumers of the same is likely to have a significant impact on his or her consumption of goods and services. (Cashell, 2009). The idea that relative income is of more importance compared to the real income might not hold true in certain cases as is empirical from the study (Kenneth V, 2007) whose findings were opposite to their hypothesis and it revealed that people are actually more interested in real rather than relative income. Therefore in our research we have taken perceived current real family income as one of the measure of personal finance index. The perceived availability of credit or cheap finance is also one of the indicators of consumer confidence. If an individual thinks that easy finance is available then his perceived confidence goes up (Woo, 2008)Thus, personal finance index is often considered a good measure of consumer spending intentions.
Business conditions are the factors which have an impact on the operation of businesses. Business conditions are also called business fundamentals. Consumers are also directly affected by business conditions because the business climate can determine things like pricing for goods and services. An increase in the unemployment is likely to generate an increase in uncertainty among consumers, even among those who are themselves employed. This is likely to increase precautionary savings and decrease confidence and consumption As such, one can expect a negative relationship between unemployment and consumer confidence—i.e., the higher the unemployment rate, the lower sentiment and consumption are likely to be.(DION, 2006) It is often an aggregate of the consumers’ perception of macroeconomic indicators of the economy like growth rates, inflation and governments monetary and fiscal policies. Business Conditions index reflects the wariness or the confidence the consumer has in the fundamentals of the economy and how the economy is performing now as well as in the future. Consumer confidence Index also incorporates expectations regarding the future of the economy; such information might be particularly relevant for the intervening period between the announcement of a policy shift and the time it is implemented. For example, suppose that following an election, a change in administration leads households to expect an improvement in the economy. We would expect this (positive) sentiment to have an impact on consumers’ current and projected future spending patterns. Moreover the consumer confidence index also takes into account the occurrence and impact on business of unusual shock e.g., an event such as Hurricane Katrina—that cannot be systematically built into models but affect the economy in significant ways. (Maude Toussaint-Comeau) Keeping these factors in mind we have tried to capture consumer sentiment the perceived employment conditions and perceived past, present and future business conditions
Buying Behavior captures consumers perception about government policies related to prices of goods, products and services , availability of choices of goods, products and services for purchase and well as consumers expectations about future prices of goods, products and services vis-à-vis their current prices. (Mishkin, 1978). We intend to find out the buying behavior of the consumers through their perception of prevailing prices, availability of choices and the perceived government policies. Consumers rely on various information cues or product attributes in their decision-making. Price represents extrinsic cues, and is one of the most important kinds of information consumers’ use when they make a purchase decision, accounting for 40 percent of their information search. (Byoungho Jin, 2003) Consumers change their consumption habits with change in the perceived inflation rates. They tend to consume more if the perceived inflation rates are high. (Gunter Poser, 1979) For example, if an individual perceive the price of a particular product to go high in future, he would rather consume the product now at the prevailing price than wait for the price to go up. Buying behavior is also affected by the perceived availability of choices. When the consumer perceives the economic conditions to be adverse he engages in minimal search of alternatives and hence perceives limited availability of choices. (Sharon E Beatty, 1987) An individual’s buying decision is also influenced by the perceived government policies. If a person feels that the current policies are enough to fight against price rise and unemployment then he will be encouraged to buy. Not much literature review is available to demonstrate any direct relation between perceived government policies and buying behavior but indirect studies have been done on this example buying decision and inflation. (Eugene J. Kelley) Low scores on a Buying Behavior index suggests that consumers are vary of spending their income on illiquid items. This has great repercussions to market economies that derive their productivity and growth from the commerce that takes place within their domestic market. Also it leads to interesting questions for countries that export goods and services, the producer sentiment of these economies may be driven by the Buying Behavior index of foreign countries. While capturing such complex macroeconomic interplays is beyond the scope of this project, it is still an interesting point to note.
Once each of them is individually measured, the composite measure captures all dimensions of consumer confidence. The methodology used for measuring each of these constructs as well as consumer confidence will be discussed later in the proposal. (Curtin, 2002)
Mullainathan in 2002 wrote a paper on biases that exist in media. According to the paper, there are two kinds of biases. One is the ideology bias, which shows a news outlet’s attempt to change readers’ opinions into some specific direction. The other type of bias is the spin bias, which shows the outlet’s attempt to write a memorable report. While competition between media outlets tend to cancel out ideological bias, chances of spin bias increases as a result. (Mullainathan, 2002)
According to a study conducted by CES on ‘The Impact of Newspapers on Consumer Confidence’ (CES, Nov, 2006), it is argued that newspapers can make stories that are memorable, and then they should be able to affect the consumer behavior into some particular direction. According to the study, the consumers’ opinion about the state of the economy can be magnified or reduced by using spin in media reporting. Thus the consumer confidence is not just dependent on macroeconomic variables affecting the economy but also dependent on the way news about the macroeconomic variable are reported in the media. They go on to construct a model that shows how news is reported by the media. Newspapers were chosen as media because they have higher reach compared to other media types and it’s easier to present before a panel of experts. To avoid time biases, sampling was done on a fixed day of every month. However, this also leads to the situation, where the experts’ reaction to the media is influenced by only what is being reported on that particular day. Also, since newspapers differ in the way they report news; two newspapers were chosen as the basis for the study. One was a large, popular morning newspaper and a somewhat smaller evening newspaper, more directed at higher social classes. They expected that the two papers together will provide a reasonable picture of economic news being presented by the media. The paper goes on to explain the drawbacks of their sampling method, where papers from 1998 to 2002 were used for data collection, while the actual collection took place in 2004 resulting in a hindsight bias in the target group. Also they felt that the current mood of the target group will affect their interpretation of historical data presented to them. The experts taking part in the study had their observations correlated and were found to have a alpha value of 0.77, which is sufficient measure of reliability and consistency of the questionnaire used to administer the study. The paper concludes that the MEDIA variable can have a sizable impact on consumer confidence, though the duration of this impact needn’t be very long. The paper also states that there is scope for further research as finding the effect of spin from other types of media like radio and Televisions as well as effect of spin from stories that are orally distributed. This lays the foundation for our research.
Consumer confidence is not only affected by the actual state of the economy, but also by spin. This shows that while consumer confidence index claim to be unbiased to media and other variables, these extraneous variables seem to influence consumer confidence. By examining the different sources of information and their respective impacts on the consumers’ confidence, we wish to further the findings of this paper. After all, in a country like India, people exercise much more prudence in their choice of information due to greater social linkages. (Country-data)
Also while research of media bias affecting consumer confidence has been researched, the effect of media bias in each component of Consumer confidence like Business Condition, Personal Finances and Buying Behavior has not been measured separately. We believe that the extent of the moderation effect of source of information on each component in their relation to consumer confidence will be different. In most extant literature, equal weights are given to Personal Finance Index, Buying Behavior Index and Business Condition Index while calculating Consumer Confidence. However, if the effect of moderation of source of information is different for each of these indices, then they should not affect consumer confidence equally.
We wish to measure and capture the extent of this difference in our research and hence, we state our research hypotheses in the next section.
H1: Influence of Business Conditions Index on Consumer Confidence is contingent upon the source of information used by the consumer
Business Condition Index captures the consumers perceptions of macroeconomic variables like past, current and future performance of the economy and the country, employment levels and business opportunities. (Fuhrer, 1993). The spin bias in the media as explained in the previous section will alter the consumers’ perception of business condition. However, the moderating effect of the source of the information depends on the effectiveness of the source. This is what will be analyzed. We expect there to be significant change in the impact of Business Condition Index on Consumer Confidence due to the source of information with consumers more likely to trust print media than TV media. The reach and depth of reporting plays an important role here thereby, in a country like India with wide variance in income levels, print media scores over Television as the primary source of first hand information to many citizens. This however cannot discount the effect oral retelling of information will have on consumer perceptions. We feel however, that when it come to getting an idea of business conditions which have a large number of variables, consumers are more likely to trust print media rather than FRAs.
H2: Influence of Personal Finances Index on Consumer Confidence contingent upon the source of information used by the consumer
Personal Finances Index captures the financing options available to the consumer, the expected future changes in real family income of the consumer and the expected trend in the financial condition of the consumer. (Carroll C. D., 1994). Since each consumers Personal Finances scenario is likely to be unique to him, they are more likely to approach to FRAs for source of information. As a result, we expect the impact of Personal Finance Index on Consumer Confidence to be greatest for consumers using FRAs for sources of information as opposed to those to use print media and television media.
H3: Influence of Buying Behaviour Index on Consumer Confidence contingent upon the source of information used by the consumer
Buying Behavior Index captures consumers perception about government policies related to prices of goods, products and services , availability of choices of goods, products and services for purchase and well as consumers expectations about future prices of goods, products and services vis-à-vis their current prices. Since there exists a multitude of goods and services that interests consumers, FRAs and print media are unlikely to be an exhaustive source of information for information about them. As a result, we expect impact of Buying Behaviour Index on Consumer Confidence to be greatest for consumers using TV Medium as source of information as opposed to consumers who approach FRAs or use newspapers for information.
Effectiveness of Source of Information
Perceived Business Conditions
Consumer Confidence Level
Personal Finance Conditions
Sample and Data Collection
A sample of working professionals of age group of 20 to 40 years old and with the income level ranging from 2 lakh to 7 lakh is considered as the target sample for data collection. The reason behind choosing this group as our target group is that this group represents the characteristics typical to those of middle income groups in India. Increasingly middle class India has been a prime driver for growth of consumerism in India. As, the wealth trickles down and more and more individuals spend more of their income, they are driving both production and consumption of goods and services in a burgeoning Indian economy (Sharma, 2009). Also, these are the people who are bread winners for their families and ultimate decision makers in buying process. Change in their confidence level affects the overall consumer confidence index to a significant extent. A total of 150 respondents belonging to this target group was contacted and considered for the study. The respondents chosen have access to internet. Their responses are collected through an online survey. All the three indices and the consumer confidence level are measured through a single consolidated questionnaire.
Unless noted differently, the three predictors – Perceived Business Conditions, Buying Behavior, and Personal Finance Conditions as well as the outcome variable ‘consumer confidence level’ are measured using a Likert scale. We developed phrases which reflected the qualities about the all the three predictors that might influence a respondent’s attitude toward it. This attitude is grasped through the response of the respondent to that particular situation. The levels of their responses are assigned scale values 1,2,3,4 and 5. The total score of each individual predictor and the outcome variable is calculated for each respondent. (Churchil, 2009)
The moderating variable –Source of information is measured by using a simple nominal scale. Each of the respondents is asked to give their most used and trustworthy source of information for the particular predictor.
There is high correlation between the each of the three predictors and the outcome variable. (Index of Consumer Sentiments)The effect of moderation on this correlation is analyzed.
Nature of the moderator:
The importance of moderators arises from their ability to enhance understanding of the relationship between relevant independent variables and dependent variables, as well as seemingly established relationships. (Gianfranco Walsh, 2007)
Source of information which is acting as a moderator is a quantitative variable which affects the strength of correlation between the independent or the predictor variable and the dependent or the outcome variable. Specifically within this correlation analysis framework, a moderator is a third variable that affects the zero-order correlation between two other variables. A moderator-interaction effect also would be said to occur if a relation is substantially reduced instead of being reversed. (Kenny, 1986)
For example, consider that the strong correlation between buying behavior and consumer confidences scale is reduced due to the moderation effect of source of information say, print media. Thus it can be said that there is moderator effect of print media as a source of information on the correlation between buying behavior (independent variable) and consumer confidence (dependent variable).
Establishing an analytical framework for testing moderator effects:
A common framework for capturing both the correlational and the experimental views of a moderator variable is possible by using a path diagram as both a descriptive and an analytic procedure. The model diagrammed in the following figure has two causal paths that feed into the outcome variable of task performance:
Predictor and moderator
In addition to these basic considerations, it is desirable that the moderator variable be uncorrelated with both the predictor and the criterion (the dependent variable) to provide a clearly interpretable interaction term. (Kenny, 1986)
Testing the moderation by analytical procedure:
Moderation implies that the causal relation between two variables changes as a function of the moderator variable. The statistical analysis is measuring and testing the differential effect of the independent variables (perceived business conditions, buying behaviour and personal finance conditions) on the dependent variable (consumer confidence scale) as a function of the moderator (Source of information) The way to measure and test the differential effects depends in part on the level of measurement of the independent variable and the moderator variable. In our case the moderator is a categorical variable with 3 categories namely - Print media, TV and FRAs (Friends relatives and associates) and the independent variables (perceived business conditions, buying behavior and personal finance conditions) as continuous variables.
The typical way to measure this type of moderator effect is to correlate independent variables with dependent variable separately for each source of information and then test the difference. (Kenny, 1986)
We tested the correlation between independent variable and dependent variables using the regression analysis and tested its significance, (following the path a in the fig above). Then applied the effect of moderator variable and found the correlation between independent and dependent variable at each level of moderation. If the significance of the correlation for any moderator level is higher than the other moderator level for the same set of independent and dependent variable, then the effect of moderation is said to be significant for that particular level only. If it is found that the moderation effect for any level of moderator is insignificant for a set of independent and dependent variables, then it can be concluded that there is no moderation effect due to this particular moderator on the relationship between those particular independent and dependent variable.
Implications of the study:
This study is an extension in understanding consumer confidence of the particular economy and how it reacts to the information effects. It gives new dimension to the current method of determination of consumer confidence based upon the buying behavior, personal finance and perceived business conditions. Determination of the impact of source of information on the relationship between dependent and independent variables would lead to micro level comments on the macro study. For example, if a larger segment of the population refer to print media as a trustworthy source of information related to buying behavior and this source of information has reduced the extent of correlation between buying behavior and consumer confidence, then the construct used for measuring consumer confidence has to take into consideration this reduction in influence of buying behavior in determining the level of consumer’s confidence.
To look from the point of view of the people who are indulged in providing information to public through different media, this study can be used as a tool to strategize the role of media in stemming or stimulating consumer confidence. For example, TV media, from our survey and research findings, we conclude that the general population refers to TV for information related to business conditions. This can help the government to concentrate and improve on business related content being broadcasted on TV media; this might have greater influence on consumer confidence. Apart from that if TV media comes to know that there is negative impact on the correlation between the perceived business conditions and the consumer confidence due to their medium of information sharing; they can revisit the way of sharing the information and also can further verify the fact that people are actually interpreting the information in the same way as they are intended to do.
The study can be useful to the common man who is the consumer. The common man can identify the most trustworthy source of information for any of the three independent variables. Expansive research on furthering the primary findings of the survey is sufficient to come up to the conclusion.
Sample size is not large enough Consumer confidence is a macro economic entity and encompasses the whole economy. The sample size of 150 is not suitable to actually make any concrete comment. But since we are again reducing our population by selecting the target group of people with annual income level ranging from Rs 2 Lakh to Rs 7 Lakh and age group of 20-40 yrs, the random sample can be said as the true representative of the target group. Also as far as the moderation analysis is considered; if any of the dependent and independent variable is continuous in nature, then the sample size below 200 would reduce the power and effect of the test can be low. (Aguinis, 2004)
Homogeneous group Most of the respondents are belonging to same race and have similar cultural background. This leads to formation of homogeneous group of respondents whose responses may be biased. To have more accurate results, we need to take the responses from people belonging to various cultural, social and economical backgrounds.
Impact of recession Due to recession in 2008, the overall perceptions of the people about the economical factors have changed drastically. People are quite suspicious about the good news in the economy and have become little bit pessimistic about the future of the financial conditions. This effect is not taken into consideration while doing analysis.
Mutual exclusivity of categorical moderator We have assumed the categories of the source of information as mutually exclusive. There are chances that some people might be referring to one or more number of sources of information for any particular independent variable. Report does not include effect of overlapping of moderating variable.
Consumer confidence scale The research does not include all the parameters which are considered while calculating the consumer confidence scale. Hence based upon the findings of the research we cannot comment anything on the consumer confidence scale of a particular economy. The research is limited to find out the effect of source of information on the correlation between three major factors (which are affecting the consumer confidence scale) and consumer confidence level.
A time series analysis would have helped us determine temporal determinants of consumer confidence, but due to limitations, we have not been able to include such constructs into our model.
Analysis method We can do the moderation analysis using the method of checking the interaction effect of independent variable and moderator on the variance of dependent variable. This would check the moderation effect in more detailed manner. (Kenny, 1986)