CONSUMER CONFIDENCE AND ECONOMIC GROWTH

1
CONSUMER CONFIDENCE AND ECONOMIC GROWTH A CASE
STUDYBYKELVIN A SERGEANT

• Paper presented at the 4th Biennial International
  Conference on Business, Banking and Finance
  Theme
• Restoring Business Confidence Investments in
  the Caribbean
• June 22 24, 2011
• Hilton Trinidad and Conference Centre

2
Objectives of the Study

• To identify the functional identity of consumer
  confidence.
• To determine the relationship between CCI and
  economic growth( as well as interest rates and
  the exchange rate
• To determine if a long-run relationship exists
  (co-integration)
• To see if we can make conclusions on based on the
  psychological framework of consumers

3
What is a Consumer Confidence Index?

• A consumer confidence index measures how
  consumers feel about several economic factors.
  The measure is based on several questions put by
  an interviewer to the consumer. The result which
  is represented by a numerical value, speaks to
  consumers evaluation of their own financial
  situation, employment chances, expenditure
  intentions and their opinion of general economic
  conditions.

4
What is the Rationale for such an Index?

• Consumption is an important determinant in any
  economy
• Personal consumption expenditure usually
  represents the single largest sector or driver of
  GDP (income side)
• The CCI seeks to gauge consumer spending through
  consumer sentiment
• Consumer confidence is measured or observed
  through consumer surveys
• The most famous and oldest measure of CC is the
  US survey conducted by the Conference Board and
  the University of Michigan Consumer Sentiment
  index

5
Features of the CCI

• The CCI may be related to macroeconomic variables
  such as inflation
• Relating the index to macroeconomic variables
  suggest that confidence is boosted by lower
  inflation, lower interest rates and higher real
  income
• Consumer confidence is called a lagging indicator
  in that current behaviour is based on the past
  and it can take some time to turn around
• Enterprises make use of consumer indices to
  enhance the sophistication of their planning,
  advertising and marketing strategies.

6
Limitations of the CCI as an indicator of
economic well being

• Economic crisis may affect confidence and not
  spending, e.g., stock market crash of 1987
• The lag effect consumer confidence either
  persists after the end of a growth period, or
  remains depressed after a recession has ended
• The Republic Bank/MFO index has the same flaws as
  most indices
• Public sentiment about an occurrence may run
  counter to spending pattern
• Index is still in its embryonic stage

7
CCI in the Caribbean

• In the Caribbean only two countries, Trinidad and
  Tobago and Jamaica continuously measure and track
  business and consumer sentiment. In Trinidad and
  Tobago, Republic Bank Limited and Market Facts
  and Opinions produce the Consumer Confidence
  Index, and the Arthur Lok Jack Graduate School of
  Business, University of the West Indies, produces
  the Corporate Confidence Index. In Jamaica, the
  Jamaica Conference Board is responsible for both
  indices.

8
Previous Research

• One can separate the literature on consumer
  confidence into three distinct approaches.
• The first argues that there is a significant and
  strong relationship between consumer sentiment
  and consumption expenditures (carroll et al.
  1994).
• The second fails to find any supportive evidence
  of empirical significance, rejecting the validity
  of consumer confidence as a leading indicator
  (Garner, 1991).
• The third uses some form of unconventional
  methodology to bridge the gap between qualitative
  survey data and quantitative analysis, resulting
  in favourable (Jansen and Nahius, 2003) and
  non-favourable evidence (Dominitz and Manski,
  2004).
• However, the common point of all these studies is
  to focus on the explanatory power of consumer
  confidence thus restricting it to the role of an
  exogenous variable.

9
Our Approach

• This paper also employs some unconventional
  methodology by assessing the determinants of
  consumer confidence in Trinidad and Tobago so
  that the functional identity of consumer
  confidence is revealed. This is important for
  three reasons.
• a favourable finding will mean that in Trinidad
  and Tobago, consumers behave different due to the
  dynamic economic structure
• second, consumer confidence measures in Trinidad
  and Tobago should be considered with care (rather
  than just some data) and
• last, we will have insight into the psychological
  frame work of consumers in emerging markets where
  there is certainly a difference between ability
  to buy and willingness to buy.

10
Methodology of the Republic Bank/MFO Consumer
Confidence Index

• The Survey is conducted door-to-door, with a
  randomly selected nationally representative
  sample of 500 households
• The neighbourhoods used in the sample were
  randomly selected over the broad areas of North
  East, North West, Central and South Trinidad, and
  Tobago.
• The MFO/Republic Bank Consumer Confidence Index
  is based on consumers opinions of current and
  expected personal finances, business conditions
  and purchasing conditions. The Index of Current
  Consumption (ICC) and the Index of Consumption
  Expectation (ICE) are incorporated in the Index
  of Consumer Confidence or Index of Consumer
  Sentiment (ICS).

11
Behaviour of variables in the study
12
Methodology of study

• Basic Regression model
• CCI C GDP ER IR u t
• Determine whether a long run equilibrium
  relationship exist between consumer confidence
  and GDP (cointegration).
• A necessary condition for cointegration is that
  these series must be non stationary and
  integrated of the same order.
• Apply the Perron Unit Root Test to establish
  whether variables are non-stationary
• Establish the direction of causality using the
  Granger- Causality test
• Apply the Impulse Response function to observe
  the behaviour of the endogenous variable to
  external shocks to any of the exogenous variables

13
Results of the unit root tests

Variable Level First Difference
CCI -6.098 .
?GDP -6.530 .
IR -5.352 -6.389
ER -10.939 .
5 critical value -5.590 -5.590
The results indicate that the CCI, ?GDP and ER
are stationary in levels, while IR is
non-stationary in levels. Hence, a long run
equilibrium relationship among the variables
cannot be established
14
Results of the granger-causality test

• Bivariate Model 1 The null hypothesis that CCI
  does not Grangercause GDP is rejected at 5 and
  10 level of significance. Also the null
  hypothesis that GDP does not Granger-cause CCI is
  also rejected at the 5 and 10 level of
  significance. This means that GDP helps to
  predict CCI and CCI helps to predict GDP.
• Bivariate Model 2 the null hypothesis that CCI
  does not Granger-cause IR is rejected at the 10
  level. However, the null hypothesis that IR does
  not Granger- cause CCI is accepted at levels of
  significance. Therefore it can be concluded that,
  that while CCI helps to predict interest rates,
  interest rates does not help predict consumer
  confidence.
• Multivariate Model. The results indicate that CCI
  helps predict GDP and vice versa. It also
  indicates that GDP to a small extent helps
  predict the exchange rate. However, the exchange
  rate and the interest rate do not help to predict
  any variable in the model

15
Results of the impulse response function

• The response of consumer confidence to a shock in
  GDP will initially lead to a drop in consumer
  confidence in the first and second quarters,
  however, the consumer confidence will trend
  towards positive as the lag increases.
• This justifies the CCI variable as a lagging
  indicator and supports our belief that initially,
  CCI will not respond to increases in GDP, but
  after a period of time, there will be a positive
  response in accordance with a priori expectations
  depending on economic theory. We note also the
  negative response of the interest rate variable
  again in line with our initial expectations.

16
Response to cholesky one S.D. Innovations
17
Conclusion

• We found that consumer confidence can have an
  endogenous nature and has a strong relationship
  to GDP.
• However, in our model, we found only a short-run
  relationship as the results of the tests led us
  to conclude that CCI, GDP and ER are stationary
  in levels while IR is non-stationary in levels.
• the series had different order of integration and
  as such, a long-run relationship among the
  variables could not be established.
• We therefore support the use of the consumer
  confidence index as a short-run policy variable

18
Thank you