Can consumer attitudes forecast household spending in the United States? Further evidence from the Michigan survey of consumers.

1. Introduction

For many years, indices of consumer sentiment have been used to provide government policy makers, economic forecasters, and business managers with timely and important information on consumer attitudes. This interest in consumer attitudes reflects a widespread belief that the sentiments and expectations of individual consumers directly affect the direction of the U.S. economy. Reinforcing this belief is the fact that consumer spending accounts for about two thirds of the nation's Gross Domestic Product (GDP).

Thus far, most analyses of consumer attitudes as a leading indicator of household spending have focused primarily on the predictive power of the Michigan Index of Consumer Sentiment (ICS). The results of these studies have, however, been mixed. For example, an early study by Lovell (1975) finds that measures of consumer attitudes based on the Michigan Survey of Consumers are unreliable predictors of future consumption. (1) Mishkin (1978), using a stock adjustment model, shows that the ICS provides good explanatory power for changes in consumer durables. Carroll, Fuhrer, and Wilcox (1994, henceforth CFW) find that the Michigan Index has some incremental predictive power as regards forecasting household spending. Souleles (2001), using the microdata of the Michigan Survey, reports that consumer sentiment is useful in forecasting future consumption, even when controlling for a number of macroeconomic variables. On the other hand, Howrey (2001) finds that both lagged and current-quarter monthly values of the ICS are generally insignificant when control variables are present in the equations of total personal consumption expenditures (PCE), consumer spending on durable goods as well as on services.

Lovell (2001) recently suggests that the Index of Consumer Expectations (ICE) developed by the University of Michigan may be a better proxy for consumer confidence than the ICS. This is because the ICE is derived solely from a subset of forward-looking questions, in contrast to the ICS, which is based on both forward-looking questions and current-conditions questions. (2) In view of Lovell's (2001) insightful suggestion, the main objective of this article is to empirically examine the predictive power of the ICE in forecasting U.S. consumption growth. Moreover, it would he useful to compare the informational content of the ICE and the ICS to determine whether indices of consumer confidence reflect consumers' perception of future economic conditions.

In this article, we use the reduced-form equation given in CFW (1994) to examine the forecasting ability of the ICE and the ICS. Our empirical results indicate that the lagged values of the ICE predict changes in total PCE much better than those of the ICS. Furthermore, when tested separately in the reduced-form equation, the forward-looking questions are generally significant, suggesting that they contain valuable information about consumers' expectations of future economic outlook. We also extend our analysis to the study by CFW (1994). The results of this analysis confirm the view that the ICE has greater incremental predictive power than the ICS.

The remainder of this article is structured as follows: Section 2 describes the five core questions used in the Michigan Surveys of Consumers. Section 3 discusses the econometric methodology and data. Section 4 reports our main empirical results. Section 5 is a case study based on CFW's (1994) data set. Section 6 presents some conclusions.

2. The Michigan Surveys of Consumers

The ICS, produced by the Michigan Surveys of Consumers, is derived from the following five core questions: (3)

QF[P.sup.r] (Financial Position realization). We are interested in how people are getting along financially these days. Would you say that you (and your family living there) are better off or worse off financially than you were a year ago?

QDurs (Durables purchases). About the big things people buy for their homes such as furniture & refrigerator, stove, television, and things like that. Generally speaking, do you think now is a good or bad time for people to buy major household items?

QF[P.sup.e] (Financial Position expectation). Now looking ahead--do you think that a year from now you (and your family living there) will be better off financially or worse off, or just about the same as now?

QBCm12 (Business conditions, 12 months). Now turning to business conditions in the country as a whole--do you think that the next twelve months we'll have good times financially, or bad times, or what?

QBCy5 (Business conditions, 5 years). Looking ahead, which would you say is more likely--that in the country as a whole we'll have continuous good times during the next 5 years or so, or that we will have periods of widespread unemployment or depression, or what?

The ICS is computed using the relative scores (the percentage giving favorable replies minus the percentage giving unfavorable replies, plus 100) for each of the five core survey questions.

It is important to note that the five core questions vary in nature. Two of the five questions, QF[P.sup.r] and QDurs, ask respondents how they view current economic conditions, whereas QBCm12, QBCy5, and QF[P.sup.e] ask them how they view future business conditions--both during a 1-year and a 5-year period--as well as changes in their own financial situation during the following year.

In order to further gauge consumer confidence, the University of Michigan also designed the ICE using only the relative scores of the three forward-looking questions QBCm12, QBCy5, and QF[P.sup.e]. Because of its usefulness in predicting the future course of the U.S. economy, the ICE is included in the Leading Indicator Composite Index developed by the U.S. Department of Commerce.

3. Econometric Methodology and Data

We examine the predictive ability of various measures of consumer confidence by using the reduced-form equation given in CFW (1994, p. 1400):

(1) [DELTA] log([C.sub.t]) = [[alpha].sub.0] + [N.summation over i=1] [[beta].sub.i][S.sub.t-i] + [gamma][Z.sub.t-1] + [[epsilon].sub.t],

where [C.sub.t] is consumer spending, [S.sub.t] is consumer confidence, [Z.sub.t] is a vector of control variables, and [[epsilon].sub.t] is a non-autocorrelated error term. Following CFW (1994), we include the following control variables in Z: four lags of the dependent variable (past consumption growth) and four lags of the growth in real labor income. In this article, we use three proxies for [S.sub.t]: the ICS, the ICE, and the five individual survey questions (QF[P.sup.r], QDurs, QBCm12, QBCy5, and QF[P.sup.e]). We also consider four categories of consumption [C.sub.t]: total personal consumption expenditures (PCE), durable goods, nondurable goods, and services. Data for labor income and the four categories of consumption are obtained from the Web site of the Bureau of Economic Analysis (BEA). The other data series that include the ICE, the ICS, and the relative scores for the five survey questions are provided by the Survey Research Center of the University of Michigan.

Our empirical analysis has been carried out using quarterly data for the sample period 1960Q1-2002Q2. (4) The choice of the starting date of our sample period is constrained by data availability. This is because, prior to 1960, the Michigan Surveys of Consumers was conducted only 2 or 3 times a year. (5) More importantly, some of the original data on the pre-1960 ICE are no longer kept by the University of Michigan. (6) Consequently, it is very difficult to compare data between the ICS and the ICE using the pre-1960 survey results.

4. Empirical Results

The empirical results of the reduced-form equation are presented in Tables 1 and 2. To conserve space, we report only the increment to the [R.sup.2] (or the adjusted [R.sup.2]) provided by the lagged values of the various measures of [S.sub.t] and the p-value of the joint significance of the lags of consumer confidence. We also estimate the reduced-form equation without the control variables in order to examine whether the confidence indicator, by itself, has good forecasting ability. As CFW (1994) point out, this part of our investigation allows us to examine the validity of Hall's (1978) random walk hypothesis. (7) Finally, due to the lag structure of the model, the effective sample size for Equation 1 is from 1961Q1 to 2002Q2.

Table 1 displays the predictive power of the ICE, the ICS, and the five individual survey questions when the control variables are excluded from the reduced-form equation. From this table, we find that the ICS, by itself, is a leading indicator in all four consumption categories. For example, we observe that in the case of total PCE, the sentiment index is statistically significant at the 1% level. Moreover, adding the last 4 quarters of data from the ICS to the prediction equation explains 9.5% of the variation in the next period's growth in total PCE (row 1). A similar finding is also detected for the three subcategories of total PCE: durable goods (row 2), nondurable goods (row 3), and services (row 4), where the 4 lags of the index in these cases are statistically significant at the 5% level or better. A careful inspection of the results of Table 1 further indicates that the incremental [R.sup.2]s for nondurable goods as well as for services are quite high, ranging from 8.9% to 9.9%. As regards the durable-goods case, lagged consumer sentiment contributes only 2.8% of the l-quarter-ahead variation in the growth of this consumption category.

Similar to the results based on the ICS, the coefficients on the four lags of the ICE, as reported in Table 1, are highly significant. Moreover, the self-forecasting capability of the ICE is better than that of the ICS in all consumption categories considered. For instance, we notice that in the case of total PCE, the incremental [R.sup.2] from the prediction equation using the ICE as a predictor is larger than that based on the ICS by 6.3% (15.8% vs. 9.5%). As for the three subcategories of total PCE, our regression results reveal that predictive power of the ICE is still higher than that of the ICS. This is particularly evident in the services category where the increase in the [R.sup.2] can be as high as 4% (13.9% vs. 9.9%).

The self-predictive ability of the individual survey questions is generally impressive. Focusing on the three forward-looking questions, QF[P.sup.e], QBCm12, and QBCy5, we find that they provide good explanatory power for movements in future consumer spending. This observation holds regardless of the choice of the consumption category. Furthermore, it is important to point out that the question concerning short-term business conditions, QBCm12, performs best in terms of tracking future consumption growth. The [R.sup.2]s generated from the reduced-form equations utilizing QBCm12 as a sole explanatory variable are always higher than those generated by QF[P.sup.r] and QDurs in all four consumption categories examined. This finding is especially obvious with respect to total PCE and spending on services. (8)

Table 2 presents our regression results when the control variables are included in the prediction equation. From Table 2, we see that, in the case of total PCE as well as spending on services, the predictive ability of the ICS is substantially reduced in the presence of the control variables. Furthermore, we find that the coefficients of the four lags of the sentiment index in the prediction equation of total PCE are only marginally significant at the 10% level. As regards nondurable goods, the ICS in this consumption category is statistically insignificant at any chosen level of confidence, indicating that the Index is not incrementally informative about movements in spending for nondurable goods. However, in the durable-goods category, the inclusion of lagged consumer sentiment raises [R.sup.2] by 3.3%, with a significant p-value.

As Table 2 indicates, while the predictive power of the ICE is lowered when the control variables are present, it continues to be a leading indicator in three of the four consumption categories considered. In particular, we find that adding the ICE to Equation 1 explains an additional 5.7% of the variation in the future PCE. Moreover, the four lags of the Index are jointly significant at any level of confidence. (9) This conclusion can "also be generalized to two of the three subcategories of total PCE, namely, durable goods and services where the incremental [R.sup.2]s are higher than those based on the ICS by 3.4% and 2.1%, respectively. As regards the case of nondurable goods, the incremental [R.sup.2] is only 0.1%, with a significant p-value. This result suggests that in the prediction equation for nondurable goods, almost all the information in the ICE is held in common with the control variables.

From Table 2, we find that the incremental forecasting power of the forward-looking questions is again higher than that of the current-conditions questions. Also, we can see that the question concerning short-term business conditions, QBCm12, performs best in terms of predicting three consumption categories, namely, total PCE, durable goods, and services, with the incremental [R.sup.2] varying from 3.6% to 7.7%. Moreover, our estimation results indicate that QBCm12 possesses an even greater explanatory power than the ICE. For example, the incremental [R.sup.2]s from regressions using QBCm12 as a predictor for the above three consumption categories are all higher than those of the ICE. Thus, much of the ICE's predictive power appears to stem from its ability to convey information on consumers' assessment of the short-term business and economic environment. As for spending on nondurable goods, we find that QF[P.sup.e] is the only question with forecasting capability. However, its incremental [R.sup.2] is quite small (1.6%), suggesting that the additional information gained from using this forward-looking question is limited. (10)

5. A Case Study Based on CFW's (1994) Data

In this section, we extend our analysis to examine the robustness of the results of CFW (1994) when alternative measures of consumer confidence, namely, the ICE and the five individual survey questions, are used in Equation 1. In their study, CFW (1994) partition total PCE into three subcategories (motor vehicles, goods excluding motor vehicles, and services) rather than the three traditional categories (durable goods, nondurable goods, and services) used in this article. Here, we follow CFW's (1994) definitions of consumption and obtain the required data from Carroll's Web site (www. econ.jhu.edu/people/ccarroll). Our regression results shown below are based on the sample period from 1961Q1 to 1992Q3. (11)

The forecasting ability of the ICE, the ICS, and the five individual questions are summarized in Tables 3 and 4. We again report only the increment to the [R.sup.2] provided by the lagged values of various measures of consumer confidence and the p-value of their joint significance. (12) On the basis of these results, several points deserve special attention.

First, similar to the findings reported in Table 1, the empirical results of Table 3 indicate that the ICS is able to self-forecast future household spending. This is true regardless of the types of consumption.

Second, the self-predictive power of the ICE as well as the question regarding short-term business conditions, QBCm12, is again higher than that of the ICS. This is particularly apparent in the case of total PCE and spending on services.

Third, the self-predictive power of the ICE and ICS, as shown in Tables 1 and 3, clearly indicates that lagged consumer confidence is able to predict current consumption growth. The robustness of this evidence across differing sample periods as well as various consumption categories can be construed as a clear rejection of Hall's (1978) random walk hypothesis that no lagged variable can be used to forecast changes in current consumption purchases.

Fourth, as Table 4 indicates, although the coefficients on the 4 lagged values of the ICS in the case of total PCE are statistically significant at the 5% level, the incremental [R.sup.2] is relatively small (4.1%). The ICS does, however, provide good explanatory power in forecasting growth in two of the three subcategories of PCE examined, namely, motor vehicles and goods excluding motor vehicles. The incremental [R.sup.2]s for these two consumption categories range from 6.3% to 9.9%. In the case of services, the lags of the ICS are not jointly significant at the 10% level, indicating that the sentiment index is not incrementally informative about the future growth of this consumption category.

Fifth, the incremental predictive power of the ICE is consistently higher than that of the ICS. This observation stands, regardless of the choice of the consumption category.

Finally, the two business-conditions questions, QBCm12 and QBCy5, once more dominate the two current-conditions questions, QF[P.sup.r] and QDurs, in terms of predicting movements in future consumer spending. Also, QBCm12 tends to have a higher incremental forecasting power than the ICE itself in every single consumption category examined. (13)

The results reported in this section indicate that use of the ICE improves CFW's (1994) estimation results. They also confirm Lovell's (2001) suggestion that the ICE is a much better proxy for consumer confidence than the ICS because it is derived solely from forward-looking survey questions. This indicates that the incremental predictive power of both the ICE and the ICS is directly related to their ability to capture consumers' perception of the general business and economic climate. It is this perception that affects the timing of consumer purchases of major household items. This in turn implies that consumer sentiment can act as a channel for aggregate economic fluctuations.

6. Conclusions

In this article, we have examined the usefulness of various measures of consumer confidence in forecasting household spending in the United States. Using the reduced-form equation provided in CFW (1994), we find that for the sample period 1961Q1-2002Q2, the ICE is incrementally more informative about household spending than the ICS for all categories of consumption examined. A similar conclusion emerges when CFW's (1994) data set is used. The clear message of these findings is that the econometric models currently used to forecast U.S. consumption growth could be made more accurate by relying on the ICE rather than the ICS, as is the current practice.

Overall, our results reconfirm the popular view that indices of consumer confidence reflect consumers' perception of future economic trends. As such, confidence indices constitute a convenient and useful tool for predicting future household spending. Forecasters and government policy makers would do well to pay attention to them, especially during periods of economic fluctuation.

Appendix: Description of the Data Used in Section 4

Consumption

Four categories of real personal consumption expenditures are used in Section 4: Total personal consumption expenditures, durable goods, nondurable goods, and services. The data are from the U.S. Department of Commerce, BEA.

Labor Income

Labor income is defined as wages and salaries plus transfer minus personal contributions for social insurance. These data are from the U.S. Department of Commerce, BEA.

Price Deflator

Data on nominal labor income are deflated by the personal consumption expenditure implicit price deflator (1982 = 100). These data are from the U.S. Department of Commerce, BEA.

Michigan Surveys Data

The survey data can be obtained from www.sca.isr.umich.edu.

Table 1. Self-Predictive Power of Various Measures of Consumer
Confidence (Incremental [R.sup.2]s from Simple Prediction
Equations) (a)

                                     1961Q1-2002Q2

Row   Category of Real PCE     ICS       ICE     QP[F.sup.r]

1       Total                 0.095     0.158       0.104
                             (0.000)   (0.000)     (0.000)
2       Durables              0.028     0.059       0.026
                             (0.022)   (0.000)     (0.027)
3       Nondurables           0.089     0.117       0.125
                             (0.004)   (0.003)     (0.001)
4       Services              0.099     0.139       0.071
                             (0.000)   (0.001)     (0.005)

                          1961Q1-2002Q2

Row      QDurs       QP[F.sup.e]      QBCm12        QBCy5

1        0.067         0.107           0.172        0.130
        (0.000)       (0.000)         (0.000)      (0.000)
2        0.016         0.048           0.061        0.039
        (0.060)       (0.001)         (0.000)      (0.005)
3        0.064         0.090           0.131        0.093
        (0.004)       (0.011)         (0.001)      (0.016)
4        0.111         0.061           0.162        0.117
        (0.000)       (0.051)         (0.000)      (0.001)

PCE, personal consumption expenditures: ICS, Michigan Index of Consumer
Sentiment; ICE, Index of Consumer Expectations; QF[P.sup.r], Financial
Position realization; QDurs, Durables purchases; QF[P.sup.e], Financial
Position expectation; QBCm12, Business conditions, 12 months; QBCy5,
Business conditions. 5 years.

(a) The numbers in parentheses are p-values of the joint significance
of the lags of consumer confidence. Hypothesis tests are conducted
using a heteroskedasticky and serial correlation robust covariance
matrix (allowing serial correlation at lags up to 4).

Table 2. Incremental Predictive Power of Various Measures of Consumer
Confidence (Incremental [R.sup.2]s from Simple Prediction
Equations) (a)

                                        1961Q1-2002Q2

Row   Category of Real PCE      ICS        ICE      QP[F.sup.r]

1       Total                  0.011      0.057        0.018
                              (0.102)    (0.000)      (0.107)
2       Durables               0.033      0.067        0.014
                              (0.042)    (0.000)      (0.100)
3       Non-Durables           0.005      0.001        0.024
                              (0.248)    (0.098)      (0.180)
4       Services               0.006      0.027       -0.003
                              (0.069)    (0.041)      (0.327)

                          1961Q1-2002Q2

Row      QDurs      QP[F.sup.e]     QBCm12          QBCy5

1        0.001         0.040           0.066        0.040
        (0.160)       (0.000)         (0.000)      (0.026)
2        0.022         0.042           0.077        0.038
        (0.058)       (0.000)         (0.000)      (0.004)
3       -0.007         0.016          -0.002       -0.008
        (0.449)       (0.036)         (0.253)      (0.449)
4        0.037        -0.006           0.036        0.028
        (0.005)       (0.630)         (0.025)      (0.018)

PCE, personal consumption expenditures; ICS, Michigan Index of Consumer
Sentiment; ICE, Index of Consumer Expectations; QF[P.sup.r], Financial
Position realization; QDurs, Durables purchases; QF[P.sup.e], Financial
Position expectation; QBCm12, Business conditions, 12 months; QBCy5,
Business conditions, 5 years.

(a) See notes at bottom of Table 1.

Table 3. Self-Predictive Power of Various Measures of Consumer
Confidence
(Incremental [R.sup.2]s from Simple Prediction Equations) (a)

                                       1961Q1-1992Q3

Row   Category of Real PCE     ICS        ICE       QP[F.sup.r]

1     Total PCE               0.156      0.189         0.118
                             (0.000)    (0.000)       (0.000)
2     Motor vehicles          0.030      0.049        -0.026
                             (0.000)    (0.000)       (0.002)
3     Goods excluding         0.196      0.209         0.152
        motor vehicles       (0.000)    (0.000)       (0.000)
4     Services                0.123      0.175         0.086
                             (0.000)    (0.000)       (0.003)

                           1961Q1-1992Q3

Row      QDurs      QP[F.sup.e]        QBCm12          QBCy5

1        0.060         0.107           0.200           0.162
        (0.001)       (0.000)         (0.000)         (0.000)
2        0.009         0.009           0.051           0.036
        (0.055)       (0.028)         (0.000)         (0.000)
3        0.089         0.132           0.210           0.164
        (0.000)       (0.000)         (0.000)         (0.007)
4        0.050         0.078           0.203           0.163
        (0.008)       (0.000)         (0.000)         (0.000)

PCE, personal consumption expenditures; ICS, Michigan Index of
Consumer Sentiment; ICE, Index of Consumer Expectations; QF[P.sup.r],
Financial Position realization; QDurs, Durables purchases; QF[P.sup.e],
Financial Position expectation; QBCm12, Business conditions, 12 months;
QBCy5, Business conditions, 5 years.

(a) See notes at bottom of Table 1.

Table 4. Incremental Predictive Power of Various Measures of Consumer
Confidence (Incremental [R.sup.2]s from Simple Prediction
Equations) (a)

                                    1961Q1-1992Q3

Row   Category of Real PCE     ICS        ICE      QP[F.sup.r]

1     Total PCE               0.041      0.069        0.027
                             (0.001)    (0.000)      (0.004)
2     Motor vehicles          0.099      0.116        0.041
                             (0.000)    (0.000)      (0.001)
3     Goods excluding         0.063      0.081        0.048
        motor vehicles       (0.000)    (0.000)      (0.001)
4     Services                0.013      0.037        0.005
                             (0.177)    (0.027)      (0.198)

                             1961Q1-1992Q3

Row      QDurs      QB[F.sup.e]    QB[Cm.sup.12]    QB[Cy.sup.5]

1        0.002         0.020           0.075           0.052
        (0.105)       (0.004)         (0.000)         (0.006)
2        0.050         0.020           0.126           0.076
        (0.001)       (0.011)         (0.000)         (0.000)
3        0.009         0.038           0.084           0.055
        (0.013)       (0.001)         (0.000)         (0.007)
4        0.021         0.006           0.049           0.033
        (0.045)       (0.107)         (0.008)         (0.042)

PCE, personal consumption expenditures; ICS, Michigan Index of Consumer
Sentiment; ICE, Index of Consumer Expectations; QF[P.sup.r], Financial
Position realization; QDurs, Durables purchases; QF[P.sup.e], Financial
Position expectation; QB[Cm.sup.12], Business conditions, 12 months;
QBCy5, Business conditions, 5 years.

(a) See notes at bottom of Table 1.

We would like to thank Michael Lovell and two referees for useful comments. We are also grateful to Christopher Carroll and Richard Curtin for directing our attention to some useful data series. This research was supported by a research grant (no. 2020607) from the Chinese University of Hong Kong for the first author.

(1) For earlier studies on the predictive ability of the ICS, see Curtin (2000).

(2) Lovell (2001) makes this suggestion in a comment on an article by Howrey (2001).

(3) The Surveys of Consumers are conducted by the Survey Research Center at the University of Michigan. Each month, the Survey Research Center carries out a minimum of 500 telephone interviews. The respondents are asked to answer 50 questions, each of which represents various aspects of consumer attitudes and expectations.

(4) All estimation procedures are carried out using the Gauss programming language.

(5) The ICS is available quarterly from the first quarter of 1960 and monthly from January 1978.

(6) We would like to thank Richard Curtin, the Director of the Surveys of Consumers, for directing our attention to this problem.

(7) Hall (1978) argues that consumption evolves according to a random walk with drift under the Life Cycle and Permanent Income hypotheses. His seminal work has generated a significant amount of empirical investigations for the United States. However, at this stage, the empirical evidence on the validity of Hall's (1978) random walk hypothesis is mixed.

(8) The [R.sup.2]s for the equations of total PCE and spending on services are, respectively, 17.2% and 16.2%.

(9) We also note that the ICE is incrementally more informative than the ICS in this consumption category.

(10) Following CFW (1994), we examine the robustness of the results presented in Table 2 when influential observations such as those in 1975Q2, 1980Q2, 1992Q4, and 1993Q1 are excluded from the prediction equation. Our empirical results indicate that, in general, the incremental predictive power of the ICE and the question concerning short-term business conditions is still higher than that of the ICS. This set of regression results is available from the authors upon request.

(11) CFW (1994) ends their sample period in 1992Q3.

(12) The detailed regression results of Tables 1 through 4 are available from the authors upon request.

(13) CFW (1994) estimate the reduced-form equation based on two sample periods: 1955Q1-1992Q3 and 1978Q1-1992Q3. In order to provide a fair comparison with their results, we also estimate Equation 1 using the data from the shorter sample period (1978Q1-1992Q3). Our results indicate that the conclusions drawn from this shorter sample period (1978Q1-1992Q3) are qualitatively the same as those of our longer sample period (1955Q1-1992Q3). This set of regression results is available from the authors upon request.

References

Carroll, Christopher D., Jeffrey C. Fuhrer, and David W. Wilcox. 1994. Does consumer sentiment forecast household spending? If so, why? American Ecvnomic Review 84:1397-1408.

Curtin, Richard T. 2000. Psychology and macroeconomics: Fifty years of the surveys of consumers. Special Report, Surveys of Consumers, University of Michigan.

Hall, Robert E. 1978. Stochastic implications of the life cycle-permanent income hypothesis: Theory and evidence. Journal of Political Economy 86:971-87.

Howrey, Philip E. 2001. The predictive power of the index of consumer sentiment. Brookings Papers on Economic Activity 1:175-216.

Lovell, Michael C. 1975. Why was the consumer feeling so sad? Brookings Papers on Economic Activity 2:473-9.

Lovell, Michael C. 2001. The predictive power of the index of consumer sentiment: Comment and discussion. Brookings Papers on Economic Activity 1:208-13.

Mishkin, Frederic. 1978. Consumer sentiment and spending on durable goods. Brooking Papers on Economic Activity 1: 38-48.

Souleles, Nicholas S. 2001. Consumer sentiment: Its rationality and usefulness in forecasting expenditure--Evidence from the Michigan micro data. Journal of Money, Credit, and Banking. In press.

Andy C. C. Kwan * and John A. Cotsomitis ([dagger])

* Department of Economics, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong; E-mail [e-mail]; corresponding author.

([dagger]) Department of Economies, Concordia University, Montreal, Quebec, Canada, H3G 1M8; E-mail John.Cotsomitis@ eni.on.ca.

Received July 2003; accepted October 2003.