¹ Kramer, Gerald H., ‘Short-Term Fluctuations in U.S. Voting Behavior, 1896–1964’, American Political Science Review, LXV (1971), 131–43CrossRefGoogle Scholar; Kramer, Gerald H. and Lepper, Susan, ‘Congressional Elections’, in Aydelotte, William O., ed., Dimensions of Quantitative Research in History (Princeton, N.J.: Princeton University Press, 1972)Google Scholar; Lepper, Susan, ‘Voting Behavior and Aggregate Policy Targets’, Public Choice, XVIII (1974). 67–81CrossRefGoogle Scholar; Tufte, Edward R., ‘Determinants of the Outcomes of Midterm Congressional Elections’, American Political Science Review, LXIX (1975), 812–26CrossRefGoogle Scholar, and Political Control of the Economy (Princeton, N.J.: Princeton University Press, 1978)Google Scholar; Bloom, Howard S. and Price, H. Douglas, ‘Voter Response to Short-Run Economic Conditions: The Asymmetric Effect of Prosperity and Recession’, American Poliiical Science Review, LXIX (1975), 1240–54CrossRefGoogle Scholar; Li, R. P. Y., ‘Public Policy and Short-Term Fluctuations in U.S. Voting Behavior: A Reformulation and Expansion’, Political Methodology, III (1976), 49–70.Google Scholar

² Meltzer, Alan H. and Vellrath, Mark, ‘The Effects of Economic Policies on Votes for the Presidency: Some Evidence from Recent Elections’, Journal of Law and Economics, XVIII (1975), 781–98CrossRefGoogle Scholar; Fair, Ray C., ‘The Effects of Economic Events on Votes for President’, Review of Economics and Statistics, LX (1978), 159–73CrossRefGoogle Scholar; Tufte, , Political Control of the Economy.Google Scholar

³ Frey, Bruno and Schneider, Friedrich, ‘An Empirical Study of Politico-Economic Interactions in the United States’, Review of Economics and Statistics, LX (1978), 174–83CrossRefGoogle Scholar; Kernell, Samuel, ‘Explaining Presidential Popularity’, American Political Science Review, LXXII (1978), 506–22CrossRefGoogle Scholar; Monroe, Kristen, ‘Economic Influences on Presidential Popularity’, Public Opinion Quarterly, XLII (1978), 360–9CrossRefGoogle Scholar. For comparative results for the French chief executive see Lewis-Beck, Michael, ‘Economic Conditions and Executive Popularity: The French Experience’, American Journal of Political Science, XXIV (1980), 306–23.CrossRefGoogle Scholar

⁴ See, for example, Bloom, and Price, , ‘Voter Response to Short-Run Economic Conditions’Google Scholar, and Tufte, , Political Control of the Economy.Google Scholar

⁵ Popkin, Samuel, Gorman, John W., Phillips, Charles and Smith, Jeffrey A., ‘Comment: What Have You Done for Me Lately? Toward an Investment Theory of Voting’, American Political Science Review, LXX (1976), 779–805, p. 788.CrossRefGoogle Scholar

⁶ Kramer, , ‘Short-Term Fluctuations’Google Scholar; Downs, Anthony, An Economic Theory of Democracy (New York: Harper, 1957), p. 36.Google Scholar

⁷ Downs, , An Economic Theory of Democracy, p. 42Google Scholar. To be sure, Downs's formulation does permit voters to take into account considerations other than their own personal economic circumstances: ‘It is possible for a citizen to receive utility from events that are only remotely connected to his own material income… There can be no simple identification of acting for one's greatest benefit’ with selfishness in the narrow sense because self-denying charity is often a great source of benefits to oneself. Thus our model leaves no room for altruism in spite of its basic reliance upon the self-interest axiom' (p. 37). Thus our empirical probings of the pocketbook predictions should not be equated in any simple way with testing Downs's analysis of voter rationality. The pocketbook hypothesis should be construed rather as an expression of a central theme of Downs's analysis.

⁸ Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., The American Voter (New York: Wiley, 1960), p. 205.Google Scholar

⁹ Converse, Philip E., ‘Public Opinion and Voting Behavior’, in Greenstein, Fred I. and Polsby, Nelson W., eds., Handbook of Political Science, Vol. VI (Reading, Mass: Addison-Wesley, 1975), p. 98.Google Scholar

¹⁰ Nie, Norman and Andersen, Kristi, ‘Mass Belief Systems Revisited: Political Change and Attitude Structure’, Journal of Politics, XXXVI (1974), 540–91, p. 571.CrossRefGoogle Scholar

¹¹ LaPalombara, Joseph, ‘Decline of Ideology: A Dissent and Interpretation’, American Political Science Review, LX (1966), 5–16, p. 9.CrossRefGoogle Scholar

¹² This term was used originally in Meehl, Paul E.'s ‘The Selfish Voter Paradox and the Thrown-Away Vote Argument’, American Political Science Review, LXXI (1977), 11–30.CrossRefGoogle Scholar

¹³ Both the pocketbook and sociotropic predictions – at least as we have characterized them – describe voters who react retrospectively. The concept of retrospective voting has a distinguished heritage – see Downs, , An Economic TheoryGoogle Scholar, and Key, V. O., The Responsible Electorate (Cambridge, Mass: Belknap, 1966)CrossRefGoogle Scholar. For a recent and extensive treatment of retrospective voting see Fiorina, Morris P., Retrospective Voting in American National Elections (New Haven, Conn.; Yale University Press, forthcoming).Google Scholar

¹⁴ Logan, Mikal Ben-Gera, ‘Short-Term Economic Changes and Individual Voting Behavior’ (manuscript, Yale University, 1977)Google Scholar; Fiorina, Morris P., ‘Economic Retrospective Voting in American Elections: A Micro-Analysis’, American Journal of Political Science, XXII (1978), 426–43CrossRefGoogle Scholar; Kinder, Donald R. and Kiewiet, D. Roderick, ‘Economic Discontent and Political Behavior: The Role of Personal Grievances and Collective Economic Judgments in Congressional Voting’, American Journal of Political Science, XXIII (1979), 495–527CrossRefGoogle Scholar; Klorman, Ricardo, ‘Trends in Personal Finances and the Vote’, Public Opinion Quarterly, XLIII (1978), 31–48.CrossRefGoogle Scholar

¹⁵ Campbell, et al. , The American Voter, pp. 416–21Google Scholar; Fiorina, , ‘Economic Retrospective Voting’Google Scholar; Klorman, , ‘Trends in Personal Finances’Google Scholar; Tufte, , Political Control of the EconomyGoogle Scholar; Wides, Jeffrey W., ‘Self-Perceived Economic Change and Political Orientations: A Preliminary Exploration’, American Politics Quarterly, IV (1976), 395–412.CrossRefGoogle Scholar

¹⁶ The argument that voting in response to economic issues might be policy-oriented was made by Stigler, George J. in his ‘General Economic Conditions and National Elections’, American Economic Review, LXIII (1973), 160–7Google Scholar. Furthermore, several studies of Western democracies suggest that macroeconomic policy is a function of the extent to which the executive branch has been controlled by parties of the Left as against parties of the Right. In particular, according to Douglas Hibbs, long periods of control by right-wing parties are associated with higher rates of unemployment – see his ‘Political Parties and Macro-Economic Policy’, American Political Science Review, LXXI (1977), 1467–87Google Scholar. Control by left-wing parties, in contrast, is associated with larger and more rapidly growing public sectors, and with downwardly redistributive tax systems – see Cameron, David's ‘Politics, Public Policy, and Economic Inequality: A Comparative Analysis’, paper delivered at the annual meeting of the American Political Science Association, Chicago 1976Google Scholar, and ‘The Expansion of the Public Economy: A Comparative Analysis’, American Political Science Review, LXXII (1978), 1243–61Google Scholar. To act in a policy-oriented fashion voters would need to have registered these differences. It should be noted, however, that as regards the issues of inflation and unemployment, policy-oriented voting does not require voters to believe in the existence of a trade-off between these two maladies. It only requires that they perceive differences between the amount of effort and/or skill the two parties are able to apply against whichever of the two problems they find especially troublesome.

¹⁷ An exception is Weatherford, M. Stephen's ‘Economic Conditions and Electoral Outcomes: Class Differences in the Political Response to Recession’, American Journal of Political Science, XXII (1978), 917–38CrossRefGoogle Scholar, which utilizes the Center for Political Studies' 1956–60 panel study.

¹⁸ This is due to the fact that previous studies have all, in one way or another, controlled for party identification. It could be that considerations which prompt individuals to change their votes might also prompt them to change their avowed partisanship; by controlling for partisanship the impact of such considerations, including unfavourable economic conditions, would fail to be detected. There is, after all, considerable evidence indicating that party identification is susceptible to just this sort of change – see, for example, Brody, Richard A., ‘Stability and Change in Party Identification: Presidential to Off-Year’, paper delivered at the annual meeting of the American Political Science Association, Washington, 1977Google Scholar; Shively, W. Phillips, ‘Information Costs and the Partisan Life Cycle’, paper delivered at the annual meeting of the American Political Science Association, Washington, 1977Google Scholar; Page, Benjamin I. and Jones, Calvin, ‘Reciprocal Effects of Policy Preferences, Party Loyalties, and the Vote’, American Political Science Review, LXXIII (1979), 1071–89CrossRefGoogle Scholar. Most analyses to date have been quite sensitive to this issue, but it is certainly prudent to probe more deeply for potential problems. For this reason, the analyses in this paper will examine both changes in voting and changes in partisanship, as later discussion will reveal.

¹⁹ It should be noted, however, that several recent investigations into a number of different domains have also found personal concerns and discontents of various kinds have negligible political consequences. See, among others, Gatlin, Douglas S., Gile, Michael W., and Cataldo, Everett F., ‘Policy Support Within a Target Group: The Case of School Desegregation’, American Political Science Review, LXXII (1978), 985–95CrossRefGoogle Scholar; Kinder, Donald R. and Sears, David O., ‘Prejudice and Politics: Symbolic Racism versus Racial Threats to the Good Life’, Journal of Personality and Social Psychology, forthcomingGoogle Scholar; Lau, Richard, Brown, Thad, and Sears, David O., ‘Self-Interest and Civilians' Attitudes Toward Vietnam’, Public Opinion Quarterly, XLII (1978), 464–83CrossRefGoogle Scholar; Sears, David O., Hensler, Carl P., and Speer, Leslie K., ‘“Whites” Opposition to “Busing”: Self-interest or Symbolic Racism?’, American Political Science Review, LXXIII (1979), 369–84CrossRefGoogle Scholar; Sears, David O. and Kinder, Donald R., ‘Racial Tension and Voting in Los Angeles’, in Hirsch, W. Z., ed., Los Angeles: Viability and Prospects for Metropolitan Leadership (New York: Praeger, 1971)Google Scholar; Sears, David O., Tyler, Tom R., Citrin, Jack C., and Kinder, Donald R., ‘Political System Support and Public Responses to the Energy Crisis’, American Journal of Political Science, XXII (1978), 56–82CrossRefGoogle Scholar; Sears, David O., Lau, Richard, Tyler, Tom R. and Allen, Harris M. Jr., ‘Self-Interest Versus Symbolic Politics in Policy Attitudes and Presidential Voting’, American Political Science Review, LXXIV (1980), 139–51.Google Scholar

²⁰ Kinder, and Kiewiet, , ‘Economic Discontent and Political Behavior’.Google Scholar

²¹ Although our attention will be confined to pocketbook and sociotropic conceptions of politics, these two certainly do not exhaust the possibilities. For example, one alternative perspective emphasizes social group identification, essentially arguing that much of American politics is carried on at a level intermediate between pocketbook and sociotropic interests. This is a venerable idea; we are happy to notice indications of its revival; see, for example, Miller, Arthur, Gurin, Patricia, and Gurin, Gerald, ‘Electoral Implications of Group Identification and Consciousness: The Reintroduction of a Concept’, paper delivered at the annual meeting of the American Political Science Association, New York, 1978Google Scholar; Rhodebeck, Laurie, ‘Group-Based Politics’ (Ph. D. dissertation, Department of Political Science, Yale University, 1980).Google Scholar

²² One of the major problems was that in 1976 complaints about unemployment were lumped together with all other job-related references, which ranged from despising the public (mentioned quite often by sales clerks) to having to work too many hours. Only about half of the references in the employment related category were about unemployment. It also turned out that the many complaints in 1976 about taxes were coded as ‘other or less specific economic woes’. Furthermore, the 1972–74 codes often contained a hodge-podge of various and sundry responses. One category, for example, contained references to inflation, taxes, governmental waste, and the cost of college. We obviously needed to disaggregate these responses.

The staff at CPS was extremely accommodating and helped make the task of receding the interviews much less arduous than it might have been. We would like to thank Warren Miller for his approval of the project, and Ann Robinson, Alice Hayes and Maria Sanchez for their valuable assistance. A more detailed analysis of the political consequences of personal economic problems and perceptions of national economic problems appears in Kiewiet, D. Roderick's ‘Policy-Oriented Voting in Response to Economic Issues’, American Political Science Review, forthcoming, 1981.Google Scholar

²³ Despite superficial appearances these measures are not at all mutually redundant. The family's recent experience with unemployment was indeed correlated with the respondent's subjective assessment of personal economic problems but quite imperfectly (this evidence is described in detail in Kinder, and Kiewiet, , ‘Economic Discontent and Voting Behaviour’Google Scholar). In any case, it makes absolutely no difference whether the set of pocketbook interests are included together in the regression analysis or whether their effects are estimated separately.

²⁴ Alt, James E., The Politics of Economic Decline (Cambridge: Cambridge University Press, 1978).Google Scholar

²⁵ McCracken, Paul W., ‘The Practice of Political Economy’, American Economic Review, LXIII (1973), 168–71Google Scholar; Yankelovitch, , Skelly, and White, , Inc., ‘The General Mills American Family Report: A Study of the American Family and Money’ (Minneapolis, Minn.: General Mills, Inc., 1975).Google Scholar

²⁶ The major determinant of voting in congressional elections is the citizen's partisan attachment to a political party – see Campbell, et al. , The American VoterGoogle Scholar, and Converse, Philip E., ‘The Concept of a Normal Vote’, in Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., eds., Elections and the Political Order (New York: Wiley, 1966)Google Scholar. Thus it is essential to remove the possible confounding effects of partisan identification. At the same time we must avoid over-controlling for partisanship. Two recent analyses have implied that ‘nominal party identification’ – i.e. identification as a Republican, an Independent, or a Democrat – is a stable, enduring characteristic for most citizens. Strength or intensity of partisanship – e.g. the difference between strong Republicans and weak Republicans – is much more responsive to a variety of short-term forces (see Brody, , ‘Stability and Change’Google Scholar, and Converse, , ‘Public Opinion and Voting Behavior’Google Scholar). We have therefore operationalized partisanship in its nominal form for the purposes of these vote analyses. The third section of our analysis takes up more directly the causal interplay between partisanship and economic discontents, both pocketbook and sociotropic.

We also ran the regressions with a set of social background measures included – age, sex, race, and a composite measure of socioeconomic status. Unless otherwise noted in the text, adding these variables as additional controls had no effect on the pattern of results.

²⁷ The 1972 interview schedule was split into two overlapping but non-identical forms, with each administered to half of the full sample. This was done in order to increase Ihe total number of questions that could be asked, and was accomplished in such a way as to ensure that the resulting sub-samples each constituted a representative cross-section of the electorate. As a consequence of this procedure, questions representing some elements of the basic regression model were asked of the entire sample – vote choice and partisanship, in particular. Other questions were asked only of one half-sample or the other. The major complication here is that the personal and national problem questions were posed to one half-sample, the remainder of the personal economic grievances and national economic assessments to the other. This rules out the possibility of any genuine multivariate analysis. Rather than putting the 1972 study aside altogether, we generated estimates for the personal and national problems separately from the rest. We would be much more reluctant about presenting these findings if the personal and national problems people name were highly correlated with other predictor variables. From the 1974 and 1976 studies, we know that they are correlated, but quite modestly so – see Kiewiet, D. R. and Kinder, D. R., ‘Political Consequences of Economic Concerns – Personal and Collective’ (paper delivered at the annual meeting of the American Political Science Association, New York, 1978).Google Scholar

²⁸ Mayhew, David R., Congress: The Electoral Connection (New Haven, Conn.: Yale University Press, 1974)Google Scholar; Ferejohn, John, ‘On the Decline of Competition in Congressional Elections’, American Political Science Review, LXXI (1977), 166–76CrossRefGoogle Scholar; Cover, Albert D., ‘One Good Term Deserves Another: The Advantages of Incumbency in Congressional Elections’, American Journal of Political Science, XXI (1977), 532–42.Google Scholar

²⁹ The probit analysis program used was written by McKelvey, Richard and Zavoina, W., which they describe in their ‘A Statistical Model for the Analysis of Ordinal Level Dependent Variables’, Journal of Mathematical Sociology, IV (1975), 103–20CrossRefGoogle Scholar. For a very readable, less technical description of probit analysis see Wolfinger, Raymond E. and Rosenstone, Steven J., Who Votes (New Haven, Conn.: Yale University Press, 1980), Appendix C.Google Scholar

³⁰ For a good introduction on the specification and interpretation of interaction terms in multivariate analysis see Cohen, J. and Cohen, P., Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences (Hillsdale, New Jersey: Erlbaum, 1975).Google Scholar

³¹ The latter term is actually an estimate of the respondent's vote in the previous election. Based on an OLS regression, it is the first stage of the two-stage procedure. Substituting this estimate for the actual reported vote solves the estimation problems resulting from autocorrelated error. Any good econometric text discusses this procedure. For a very readable introduction to these problems and procedures see Marcus, Greg B., Analyzing Panel Data (Beverly Hills Calif.: Sage Publications, 1979).CrossRefGoogle Scholar

³² Essentially the same pattern held for sporadic voters as well. Among citizens who failed to vote for Congress in 1972 but did so in 1974, their choice in that year had nothing to do with economic misfortunes suffered in private life; the same is true for voting in 1976 among 1974 non-voters.

Nor did the evidence improve when we examined a second change hypothesis: that voters whose pocketbook experiences worsened from one election to the next would be inclined to switch their votes from the in-party to the opposition. We could find only traces of support for this hypothesis. Voters whose unemployment experiences worsened between 1972 and 1974 were somewhat more inclined to switch their support to Democratic candidates during the same period. But in general the collective eflects due to changes in pocketbook experiences were at least as frail as those reported in Table 3.

³³ So far we have neglected the question of pocketbook discontents and turnout. It has been argued that the economically aggrieved will turn to the political system for reparations and, consequently, will participate more, other things equal, than economically secure citizens; see, for example, Lipset, Seymour Martin, Political Man (Garden City, New York: Doubleday, 1960)Google Scholar. Others contend that pocketbook discontents are largely irrelevant to conventional political activity, e.g. Meltzer, and Vellrath, , ‘The Effects of Economic Policies’Google Scholar, and Schlozman, Kay L. and Verba, Sidney, Injury to Insult (Cambridge, Mass.: Harvard University Press, 1979)Google Scholar. The most persuasive evidence, however, is found in Rosenstone, Steven J.'s ‘Economic Adversity and Voter Turnout’ (manuscript, Department of Political Science, Yale University, 1980)Google Scholar. Analysing data from a Census Bureau Current Population Survey, Rosenstone finds that personal economic adversity suppressed congressional turnout in 1974. Preoccupation with personal economic problems appears to diminish somewhat the resources that can be allocated to politics. Rosenstone's results corroborate our own analysis of congressional turnout. We find, too, that personal economic discontents have slight (and erratic) dampening effects on participation. The economically aggrieved evidently do not look to the national political system for solutions – another blow to the pocketbook prediction.

³⁴ Fenno, Richard J., Home Style (Boston, Mass.: Little, Brown, 1978).Google Scholar

³⁵ This effect of unemployment experiences on presidential voting may be less robust than it appears. When a set of social background variables (age, race, sex, socio-economic status) is added to the basic regression model, the effect vanished in 1976 (although not in 1972).

³⁶ Kiewiet, 's ‘Policy-Oriented Voting in Response to Economic Issues’Google Scholar extends this analysis of personal and national level concern about inflation and unemployment back to the 1956 election. Evidence of voting in response to unemployment specifically was uncovered early in the series, e.g. 1958. This does not imply, however, that a political trade-off between unemployment and inflation used to exist, because during these years very few voters perceived inflation to be a serious problem.

³⁷ Klorman, , ‘Trends in Personal Finances’Google Scholar; Fiorina, , ‘Economic Retrospective Voting’Google Scholar; Tufte, , Political Control of the EconomyGoogle Scholar; Wides, , ‘Self-Perceived Economic Change’.Google Scholar

³⁸ Although the evidence on unemployment and presidential voting shown in Table 5 constitutes little in the way of support for the Downsian ‘throw the rascals out’ hypothesis, it is more compatible with the alternative formulation that also emphasizes voters' pocketbook interests: namely, that the unemployed look to the Democratic party for solutions to their problem, and therefore support the Democratic presidential candidate, regardless of which party controls the White House.

³⁹ A good exposition of this technique can be found in Kenny, D. A., ‘Cross-Lagged Panel Correlation: A Test for Spuriousness’, Psychological Bulletin, LXXXII (1975), 345–62.CrossRefGoogle Scholar

⁴⁰ Converse, Philip E., The Dynamics of Party Support: Cohort-Analyzing Party Identification (Beverly Hills, Calif.: Sage, 1976)Google Scholar; Brody, , ‘Stability and Change in Party Identification’Google Scholar; Converse, Philip E. and Markus, Greg B., ‘Plus Ça Change… The New CPS Election Study Panel’, American Political Science Review, LXXIII (1979), 32–49.CrossRefGoogle Scholar

⁴¹ For details of this procedure see Kenny, , ‘Cross-Lagged Panel Correlation’.Google Scholar

⁴² The t-statistics reported in this section are associated with differences between cross-lagged correlations. Again, see Kenny, , ‘Cross-Lagged Panel Correlation’ for an explanation of how they are calculated.Google Scholar

⁴³ There is a voluminous literature on issue-voting, but most of the major points of contention are addressed in Brody, Richard A. and Page, Benjamin, ‘Comment: The Assessment of Issue Voting’, American Political Science Review, LXVI (1972), 450–8CrossRefGoogle Scholar, and in Niemi, Richard G. and Weisberg, Herbert F., eds., Controversies in American Voting Behavior (San Francisco: Freeman, 1976).Google Scholar

⁴⁴ Brody, , ‘Stability and Change in Party Identification’.Google Scholar

⁴⁵ Sociotropic economic judgements may not always play such an active role, of course. When times are unequivocally good or (perhaps especially) unequivocally bad, then economic conditions are less subject to misperception. When economic signals are less clear, rationalization is encouraged. A similar argument presumably applies to issue voting. When candidates take ambiguous positions on issues, as they often do, the public's tendency to ascribe opinions to them so as to reinforce or rationalize its own view is enhanced, thereby subverting the possibility of genuine issue-based voting. See Brody, and Page, , ‘Comment: The Assessment of Issue Voting’Google Scholar; Kinder, Donald R., ‘Presidents, Prosperity, and Public Opinion’, Public Opinion Quarterly, forthcomingGoogle Scholar; and especially Page, Benjamin I., Choices and Echoes in Presidential Elections (Chicago: University of Chicago Press, 1978).Google Scholar

⁴⁶ One way to account for the pocketbook prediction's dismal record here is to explain the findings away. Perhaps we have measured economic grievances inadequately; if so, the frail relationships between political preferences and pocketbook interests would not reflect the true state of affairs at all, but rather deficiencies in our assessment of personal economic grievances.

This is a legitimate criticism, but not a telling one: substantial evidence supports the construct validity of our measures. Part of this evidence is drawn from the CPS surveys themselves, in that replies to the questions on personal economic grievances did show the expected degree of consistency. People who complained about unemployment or declining real income, for example, were also likely to express dissatisfaction with their family's financial condition. (This evidence is presented in detail in Kiewiet, D. Roderick and Kinder, Donald R., ‘Political Consequences of Economic Concerns – Personal and Collective’, paper delivered at the annual meeting of the American Political Science Association, New York, 1978Google Scholar.) More persuasive, perhaps, is evidence drawn from external sources, such as Klorman's ‘Trend in Personal Finances’. Analysing data from an Institute for Social Research (University of Michigan) income dynamics panel survey, he found strong relationships between citizens' retrospective judgements of their financial progress, on the one hand, and changes in family income, on the other. For more corroborative evidence based on still another data source see Rosenstone, Stephen J., Wolfinger, Raymond E., and McIntosh, Richard A., ‘Voter Turnout in Midterm Elections’, paper delivered at the annual meeting of the American Political Science Association, New York, 1978Google Scholar. And at the aggregate level such retrospective judgements fluctuate (if sluggishly) with changes in national economic conditions. From 1956 to 1976, for instance, the proportion of respondents who reported improvements in their family's financial condition was greatest in 1964 (41 per cent), also the year registering the sharpest increase in real per capita disposable income; the survey question recorded the least satisfaction in 1974 (25 per cent), when decline in real income was the most pronounced. Over the eleven election surveys spanning this twenty-year period, the Pearson correlation between change in real income and retrospective judgements of family financial progress was 0·68. Finally and most tellingly, improving the measurement of personal economic adversity does nothing to alter our verdict on the pocketbook's minimal political effect. We have recently developed an extensive battery of questions to tap personal economic experiences and worries. Aided by a more comprehensive, sensitive, and reliable gauge of pocketbook grievances, we nevertheless find virtually no support for the pocketbook prediction – see Kinder, Donald R., Denney, W. Michael, and Hendricks, J. Stephen, ‘Political Consequences of Inflation’ (manuscript, Department of Political Science, Yale University, 1980)Google Scholar. The measurement of personal economic grievance does not seem to be the problem. The measures are up to customary methodological standards. But the discontent they tap seldom gets expressed in political judgements.

⁴⁷ For further corroboration see Denney, W. Michael, Hendricks, J. Stephen, and Kinder, Donald R., ‘Personal Stakes and Local Polities’, paper presented at the thirty-fifth annual meeting of the American Association for Public Opinion Research, Cincinnati, Ohio, 1980Google Scholar; Kiewiet, , ‘Policy-Oriented Voting in Response to Economic Issues’Google Scholar; Kinder, , ‘Presidents, Prosperity and Public Opinion’Google Scholar; Kinder, , Denney, , and Hendricks, , ‘Political Consequences of Inflation’Google Scholar; and Sears, , Lau, , Tyler, , and Allen, , ‘Self-Interest versus Symbolic Politics’.Google Scholar

⁴⁸ Downs, , An Economic Theory of Democracy.Google Scholar

⁴⁹ Page, , Choice and Echoes in Presidential Elections.Google Scholar

⁵⁰ The same incumbent-challenger asymmetry characterizes the evidence on personal economic grievances. In 1972, for example, evaluations of Nixon were correlated with voters' employment experiences (partial r = −0·047, P < 0·05, witn simultaneous controls on party identification, age, race, sex, and socioeconomic status) and their retrospective judgements regarding their own financial conditions (partial r = −0·115, p < 0·01). Both indicators were essentially unrelated to evaluations of McGovern (partial r = 0·021, n.s.; 0·059, p < 0·05 respectively). Virtually identical results were obtained in 1976.

⁵¹ Popkin, et al. , ‘What Have You Done for Me Lately?’ (italics added).Google Scholar

⁵² This, of course, is a central question in both normative and empirical democratic theory, addressed, among other places, in Downs, , An Economic Theory of DemocracyGoogle Scholar; Converse, , ‘Public Opinion and Voting Behavior’Google Scholar; and Lippman, Walter, Public Opinion (New York, Macmillan, 1922).Google Scholar

⁵³ Converse, , ‘Public Opinion and Voting Behavior’Google Scholar; Sears, David O., ‘Political Behavior’, in Lindzey, Gardner and Aronson, Elliot, eds., The Handbook of Social Psychology, Vol.5, 2nd edn. (Reading, Mass.: Addison-Wesley, 1969).Google Scholar

⁵⁴ Hibbs, Douglas A., ‘The Mass Public and Macroeconomic Performance: The Dynamics of Public Opinion Toward Unemployment and Inflation’, American Journal of Political Science, XXIII (1979), 705–31CrossRefGoogle Scholar; Kernell, , ‘Explaining Presidential Popularity’Google Scholar; Kinder, and Kiewiet, , ‘Economic Discontent and Political Behavior’, Figure 3.Google Scholar

⁵⁵ Sniderman, Paul M. and Brody, Richard A., ‘Coping: The Ethic of Self-Reliance’, American Journal of Political Science, XXI (1977), 501–22CrossRefGoogle Scholar; Schlozman, and Verba, , Injury to Insult.Google Scholar

⁵⁶ This speculation implies both that pocketbook voting will be more likely among those citizens who see their own problems as having social or collective causes, and that sociotropic voting will be enhanced among those who define national economic problems as having political causes and political solutions. Both implications are borne out in our preliminary analysis of new national survey data from the United States, reported in Kinder, , Denney, , and Hendricks, , ‘Political Consequences of Inflation’.Google Scholar

⁵⁷ Lane, Robert E., Political Ideology (New York: Free Press, 1962).Google Scholar

⁵⁸ Some provocative evidence on this point is reported in Nisbett, Richard E., Borgida, Eugene, Crandall, Rick and Reed, Harvey, ‘Popular Induction: Information Is Not Necessarily Informative’, in Carroll, J. and Payne, J., eds., Cognition and Social Behavior (Hillsdale, N.J.: Erlbaum, 1976), pp. 113–34.Google Scholar