1. Introduction

Consumer confidence is a measure of sentiment about the state of the economy as reflected in the perceptions of a key economic agent – the consumer. By polling consumers’ opinions about macroeconomic conditions – as for example, employment; income; the price situation; and the outlook- it is possible to obtain a reasonable gauge of consumption spending, which is a major component of aggregate demand in the economy. The literature provides robust evidence of indicators of consumer confidence being good predictors of consumer spending with a lead content of four to five months (Gelper et. al. 2007). Consequently, economic agents that influence aggregate supply – manufacturers; retailers; investors; builders; government agencies - rely on measures of consumer confidence to make an assessment of consumption expenditure which, in turn, forms their production decisions. These indicators also provide useful insights to wielders of monetary and fiscal policies who have to make decisions on modulating aggregate demand in the economy around the path of productive capacity in the economy in order to ensure macroeconomic stability. Consequently, several central banks conduct their own consumer confidence surveys¹, while others depend on surveys carried out by independent organisations (the Consumer Confidence Index of the Conference Board and Consumer Sentiment Index of the University of Michigan for the US; Consumer Confidence Index of the Conference Board of Canada).

The Reserve Bank has been a pioneer in propagating the use of consumer confidence surveys in India on a large scale - its own quarterly survey was introduced in June 2010². At present, the survey is conducted in six major cities – Bengaluru; Chennai; Hyderabad; Kolkata; Mumbai; and New Delhi - covering 5,400 respondents in each round (approximately 900 per city). A hybrid two-stage sampling design is adopted. In the first stage, polling booths in a city are selected by systematic random sampling after arranging all polling booths according to their constituencies. About 45 polling booths spread over the entire city are selected in order to ensure the widest geographical coverage. In the second stage, 20 respondents are selected from each polling booth area by following the right hand rule^#.

In the survey, responses are canvassed around two reference points in time, viz., the current situation as compared with a year ago and expectations for a year ahead, with reference to five parameters - economic conditions; income; spending; the price level; and employment. Net responses i.e., the difference between positive and negative perceptions on the five parameters are combined to create a current situation index (CSI), which seeks to capture consumer sentiments about the prevailing state of affairs, and the future expectations index (FEI), which gauges the near-term outlook of consumers on these parameters (Annex). The CSI and the FEI can take values between 0 and 200.

The rest of this article is organised into four sections. The next section i.e., Section II presents a snapshot of the four rounds of the surveys carried out between September 2016 and June 2017. Section III evaluates these results with a view to gauging coherence in responses and in the CSI/FEI. Section IV investigates the predictive power of the FEI vis-à-vis private consumption and, thereby, its usefulness as an input for policy decision-making. Section V sets out some concluding observations.

2. Survey Results

As mentioned earlier, the survey solicits consumer sentiment around five key parameters which are discussed below.

2.1. General Economic Situation

Respondents were optimistic about the general economic situation during the major part of 2016, with net responses gradually picking up between December 2015 and September 2016. However, net responses slipped into the pessimistic zone since March 2017. Consumers’ expectations about the future economic situation also reached a turning point in December 2016 and worsened thereafter, but they seem to be stabilising in the current survey round (Chart 1; Table 1).

2.2. Employment Scenario

Consumer sentiment on the current employment situation stayed subdued, regressing further in the December 2016 round. In the first half of 2017, net responses have remained in the pessimistic zone, with 39 per cent of respondents perceiving a worsening of the employment scenario in relation to a year ago. In contrast, expectations regarding the future employment situation remained buoyant, with half of the respondents expressing optimism on future employment which largely co-moved with the income outlook³ (Chart 2; Table 2).

2.3. Price Situation

The improvement in households’ perceptions about the current price level during 2015-16 was interrupted in the March 2016 round of the survey and since then, there has been a steady deterioration extending up to the current round, barring in December 2016. Although expectations on the future price level have remained largely unchanged, there was a dramatic improvement in December 2016 - when prices of perishables crashed following demonetization - and again in the current round, which was largely driven by continuing optimism about spending (Chart 3; Table 3). As regards inflation, consumers’ perceptions and outlook worsened in the first half of 2017, reversing the improvement polled in the second half of 2016 (Table 4).

2.4. Household Income

Households’ assessments about current income levels have continued to remain sluggish, with no improvement perceived since December 2016. This has closely co-moved with their perceptions on the employment situation - in various rounds of the survey between September 2016 and June 2017, 34 to 39 percent of respondents felt that the employment situation has worsened as compared with the previous year and that their income levels had, in fact, fallen. Respondents’ level of optimism on future income has continued to dip in the first half of 2017, although it remains in the expansion zone (Chart 4; Table 5).

2.5. Spending

Consumer sentiment on spending turned down marginally in the current round of the survey, although it has generally remained buoyant, with 70- 80 per cent of the respondents perceiving an increase in current spending over the previous year (Chart 5; Table 6; Table 7). Notably, the outlook on non-essential (i.e., discretionary) spending has steadily picked up (Table 8), but 90 per cent of the respondents attributed it to the increase in prices of goods and services.

3. Coherence between the General Economic Situation and Other Parameters

With the steady worsening of respondents’ perceptions about the general economic situation, it is insightful to delve into which parameter(s) influenced consumer confidence the most. Their optimism on the general economic situation last year was mainly driven by perceptions on inflation as it plunged in the post-demonetisation period. In 2017, however, this sense of well-being appears to have been eroded by the worsening employment scenario. Of the respondents who believed that the employment situation has worsened, more than half also reported that the economic situation has deteriorated.

These diverse sentiments on the current state of the Indian economy and its near-term outlook are encapsulated in the CSI and the FEI. The CSI has declined for three consecutive quarters between December 2016 and June 2017; it entered the pessimistic zone in March 2017 as sentiments on the general economic situation, the price level and income worsened. The Future Expectations Index (FEI) had been expanding steadily between September 2015 and September 2016 to reach an all-time high in the history of the survey in December 2016 on post-demonetisation effects. Since then, however, it has been ebbing (up to June), although it remains in the expansion zone (Chart 6).

Based on 10,000 re-samples selected through simple random sampling with replacement (SRSWR), the 99 per cent bootstrap confidence intervals for CSI and FEI were quite low (Table 9), attesting to the robustness of the findings of the survey round.

4. Predictive Power of Future Expectations Index

With a reasonably long time series on consumer sentiment indices now available, an examination of their predictive power with respect to the goal variable – consumer spending – becomes feasible and worthwhile. In particular, it is of interest to evaluate whether or not the FEI offers relevant and timely insights into future consumption levels in India. If consumers feel confident about the actual and future economic and financial situation, they would be willing to increase their consumption. In contrast, if consumers are pessimistic, they would be inclined to delay their consumption and engage in precautionary saving. The empirical literature is inconclusive, however, with some studies indicating little information for forecasting consumption (Gosselin and Desroches, 2002; Roberts and Simon, 2001) and others detecting leading indicator properties (Gelper, et. al. 2007; Batchelor and Dua 1998). In the context of the Reserve Bank’s survey and its FEI, augmented Dickey-Fuller tests on the FEI and private final consumption expenditure to test the presence of a unit root suggest that both the series are I(1) and their first differences are stationary.

The ability of FEI to provide insights into future consumption levels has been examined by using Granger causality test between FEI and nominal private final consumption expenditure (PFCE) (Table 10). Pair-wise tests indicate that the FEI Granger causes PFCE with three to five quarters lag. Thus, the FEI can be usefully employed to forecast future values of private final consumption. This highlights the importance of FEI as an input for policy formulation.

5. Summary

Indicators of consumer confidence provide key insights into the state of the economy by profiling the perceptions of the consumer and thereby consumption spending. This article finds that the future expectations index predicts private final consumption expenditure with three to five quarters lag, offering exploitable potential as an input for forward looking policy formulation. In this context, it is useful to note that consumers’ reading of the general economic situation has turned pessimistic in 2017, mainly reflecting a worsening of perceptions on employment and inflation. Although consumers remain somewhat more optimistic on spending, this may reflect their expectations of higher prices. These perceptions may be transient, as expectations on all these parameters captured in the future expectations index remain in the expansion zone.

Annex A: Data and Results

Annex B

1. Sample Coverage and Survey Questionnaire

The survey is conducted in six metropolitan cities viz. Bengaluru, Chennai, Hyderabad, Kolkata, Mumbai and New Delhi. In each round of the survey, 5,400 respondents are selected (900 respondents from each city), using a hybrid two stage sampling design. At the first stage in a city, the polling booths are selected by systematic random sampling after arranging all polling booths according to their constituencies. In order to ensure wide geographical coverage, 45 polling booths spread over the entire city are selected. In second stage, from each selected polling booth area, 20 respondents are selected following the right hand rule. From Q4:2014-15 onwards, information relating to expenditure on essential and non-essential items, and households’ current financial situation are also being collected.

2. Current Situation Index & Future Expectations Index

In standard opinion surveys, respondents generally have three reply options such as up/same/down; or above-normal/normal/ below-normal; or increase/ remain-same/decrease. Because of the difficulty of interpreting all three percentages, the survey results are normally converted into a single quantitative number. One of the most common way of doing this is to use ‘Net-Responses’ (also called ‘Balances’ or ‘Net Balances’). It is defined as the percentage of the respondents reporting a decrease (negative), subtracted from the percentage reporting an increase (positive). Net Responses can take values from –100 to +100. In this survey, Net Response is used to analyse the Consumer Confidence Survey results. To combine the consumer confidence perceptions on various factors, two indices are worked out. These are Current Situation Index for reflecting current situation as compared to one year ago and Future Expectations Index to reflect the expectations one year ahead. For calculating the index, the following formula has been used.

Overall Index = 100 + Average (Net Response of selected factors)

Where Net Response = Positive perceptions (%) – Negative perception (%)

The average net responses on the current perception on various factors, viz. economic conditions, income, spending, price level and employment are used for the calculation of the Current Situation Index.

The average net responses on the future perceptions on various factors, viz. economic conditions, income, spending, price level and employment are used for the calculation of the Future Expectations Index.

* This article is prepared by Smt. Sangeetha Mathews and Shri D P Singh under the guidance of Dr. Goutam Chatterjee in the Division of Household Surveys of the Department of Statistics and Information Management. The valuable inputs and suggestions provided by Dr. Michael D Patra, Executive Director are gratefully acknowledged. The authors would also like to thank Dr. Snehal Herwadkar, Monetary Policy Department for her remarkable contribution to this article. The views expressed in the article are those of the authors and do not represent the views of the Reserve Bank of India.

¹ An illustrative list of central banks that conduct their own consumer confidence surveys includes Bank of Japan, Bank of Korea, Bank Indonesia and the Central Bank of the Republic of Turkey

² Two other agencies in India conduct consumer confidence surveys. While Thomson-Reuters initiated a survey in March 2010, it was an online poll restricted to 500 adults. The survey conducted by the Centre for Monitoring the Indian Economy (CMIE) through its consumer pyramids panel was introduced in 2015.

^# A method in which the first household from each polling booth area is selected randomly and the subsequent respondent is selected from the 11th household by moving in the right hand direction and skipping 10 houses. The process is continued till the sample size for each polling booth is achieved.

³ The coherence coefficient between the employment outlook and the income outlook varied between 52-57 per cent during the last four quarters. Coherence coefficient is the sum of similar responses on two variables as a proportion to total responses.