¹ For reviews of the literature, see Clarke, Harold D. et al. , Controversies in Political Economy: Canada, Great Britain, the United States (Boulder, Colo.: Westview Press, 1992)Google Scholar; Lewis-Beck, Michael S., Economics and Elections: The Major Western Democracies (Ann Arbor: University of Michigan Press. 1988)Google Scholar: Miller, William L., ‘Studying How the Economy Affects Public Attitudes and Behavior: Problems and Prospects’, in Clarke, Harold D., Stewart, Marianne C. and Zuk, Gary, eds, Economic Decline and Political Change: Canada, Great Britain, the United States (Pittsburgh: University of Pittsburgh Press, 1989).Google Scholar

² Examples include Butler, David and Stokes, Donald E., Political Change in Britain, 2nd college edn (New York: St. Martin's Press, 1976)Google Scholar, chap. 17; Clarke, Harold D., Stewart, Marianne C. and Zuk, Gary, ‘Politics, Economics and Party Popularity in Britain, 1979–1983’, Electoral Studies, 5 (1986), 123–41CrossRefGoogle Scholar; Goodhart, Charles A. E. and Bhansali, R. J., ‘Political Economy’, Political Studies, 18 (1970), 43–106CrossRefGoogle Scholar; Hibbs, Douglas A., ‘Economic Outcomes and Political Support for British Governments Among Occupational Classes: A Dynamic Analysis’, American Political Science Review, 76 (1982), 259–79CrossRefGoogle Scholar; Hudson, John, ‘Prime Ministerial Popularity in the UK: 1960–81’, Political Studies, 32 (1984), 86–97CrossRefGoogle Scholar; Miller, William L. and Mackie, William M., ‘The Electoral Cycle and the Asymmetry of Government and Opposition Popularity’, Political Studies, 21 (1973), 263–79CrossRefGoogle Scholar; Mishler, William, Hoskin, Marilyn and Fitzgerald, Roy, ‘British Parties in the Balance: A Time Series Analysis of Long-Term Trends in Public Support for Major Parties’, British Journal of Political Science, 19 (1989), 211–36CrossRefGoogle Scholar; Norpoth, Helmut, ‘Guns and Butter and Government Popularity in Britain’, American Political Science Review, 81 (1987), 949–59CrossRefGoogle Scholar; Norpoth, Helmut, Confidence Regained: Economics, Mrs Thatcher and the British Voter (Ann Arbor: University of Michigan Press, 1992)CrossRefGoogle Scholar; Price, Simon and Sanders, David, ‘Modeling Government Popularity in Postwar Britain: A Methodological Example’, American Journal of Political Science, 37 (1993), 317–34CrossRefGoogle Scholar; Whiteley, Paul, ‘Macroeconomic Performance and Government Popularity–The Short Run Dynamics’, European Journal of Political Research, 14 (1986), 45–61CrossRefGoogle Scholar; Yantek, Thom, ‘Government Popularity in Great Britain under Conditions of Economic Decline’, Political Studies, 33 (1985), 467–83.CrossRefGoogle Scholar

³ The forecasting model is described in Sanders, David, ‘Government Popularity and the Next General Election’, Political Quarterly, 62 (1991), 235–61CrossRefGoogle Scholar; Sanders, David, ‘Forecasting the 1992 British General Election Outcome: The Performance of an “Economic Model”’, in Denver, David et al. , eds, British Elections and Parties Yearbook 1993 (London: Harvester Wheatsheaf, 1993).Google Scholar See also Sanders, David, ‘Why the Conservatives Won - Again’, in King, Anthony et al. , eds, Britain at the Polls 1992 (Chatham: Chatham House Publishers, 1993)Google Scholar; Sanders, David, Marsh, David and Ward, Hugh, ‘A Reply to Clarke, Mishler and Whiteley’, British Journal of Political Science, 20 (1990), 83–90CrossRefGoogle Scholar; Sanders, David, Marsh, David and Ward, Hugh, ‘The Electoral Impact of Press Coverage of the British Economy, 1979–87’, British Journal of Political Science, 23 (1993), 175–210CrossRefGoogle Scholar; Sanders, David, Ward, Hugh and Marsh, David (with Tony Fletcher), ‘Governmental Popularity and the Falklands War: A Reassessment’, British Journal of Political Science, 17 (1987), 281–313.CrossRefGoogle Scholar The pioneering work in the field is Alt, James E., The Politics of Economic Decline: Economic Management and Political Behaviour in Britain Since 1964 (Cambridge: Cambridge University Press, 1979).Google Scholar

⁴ See, for example, Charemza, Wojciech W. and Deadman, Derek F., New Directions in Econometric Practice (Aldershot, Hants: Edward Elgar, 1992), pp. 83–4.Google Scholar

⁵ Sanders, , Ward, and Marsh, , ‘Government Popularity and the Falklands War’.Google Scholar

⁶ See, for example, Sanders, , Marsh, and Ward, , ‘The Electoral Impact of Press Coverage’, p. 191.Google Scholar

⁷ Sanders, , ‘Why the Conservatives Won’, p. 177.Google Scholar

⁸ Key, V. O. Jr, The Responsible Electorate: Rationality in Presidential Voting, 1936–1960 (New York: Vintage Books, 1968).Google Scholar

⁹ See, for example, Kinder, Donald R. and Kiewiet, D. Roderick, ‘Sociotropic Politics: The American Case’, British Journal of Political Science, 11 (1981), 129–61.CrossRefGoogle Scholar On the theory of retrospective voting, see Fiorina, Morris, Retrospective Voting in American National Elections (New Haven, Conn.: Yale University Press, 1981)Google Scholar; Kiewiet, D. Roderick and Rivers, Douglas, ‘A Retrospective on Retrospective Voting’, in Eulau, Heinz and Lewis-Beck, Michael S., eds, Economic Conditions and Electoral Outcomes (New York: Agathon Press, 1985).Google Scholar

¹⁰ MacKuen, Michael B., Erikson, Robert S. and Stimson, James A., ‘Peasants or Bankers? The American Electorate and the US Economy’, American Political Science Review, 86 (1992), 597–611.CrossRefGoogle Scholar

¹¹ Sanders, , ‘Why the Conservatives Won’, p. 179.Google Scholar

¹² The Gallup economic evaluation questions are: (a) personal prospections–‘How do you think the financial situation of your household will change over the next 12 months?’ (b) personal retrospections–‘How does the financial situation of your household now compare with what it was 12 months ago?’ (c) national expectations–‘How do you think the general economic situation in this country will develop over the next 12 months?’ (d) national retrospections–‘How do you think the general economic situation in this country has changed over the last 12 months?’ The response categories for these questions are: ‘get a lot better’, ‘get a little better’, ‘stay the same’, ‘get a little worse’, ‘get a lot worse’. The economic evaluation variables are constructed by subtracting the percentage offering negative responses from the percentage offering positive ones.

¹³ The Gallup party support questions are: (a) ‘If there were a General Election tomorrow, which party would you support?’ (b) [If ‘don't know’] ‘Which party would you be most inclined to vote for?’ Conservative support is calculated as the sum of the percentages of persons responding ‘Conservative’ to (a) or (b).

¹⁴ The poll tax variable is scored + 1 for April 1990 and 0 for other months.

¹⁵ Although most analysts now agree that a wide variety of political events can have important, if frequently transient, impacts on party support, there is no ‘canonical list’ of such events and judgement calls are unavoidable. Here, months in which an event judged to be beneficial to Conservative support occurred are scored + 1, months in which an event judged to be harmful occurred are scored − 1, and all other months are scored 0. A list of the events used is available from the authors upon request.

¹⁶ See Norpoth, , Confidence RegainedGoogle Scholar, chap. 8–10. The 1983, 1987 and 1992 election variables are scored + 1 for the month immediately preceding the election and + 1 for the month in which the election occurred. Other months are scored 0.

¹⁷ Sanders, et al. , ‘The Electoral Impact of Press Coverage’, p. 207.Google Scholar

¹⁸ See Granato, Jim, ‘An Agenda for Econometric Model Building’, in Stimson, James A., ed., Political Analysis, Vol. 3 (Ann Arbor: University of Michigan Press, 1992).Google Scholar

¹⁹ The tests are Davidson and MacKinnon's J-test and Mizon and Richard's encompassing test. See Charemza, and Deadman, , New Directions in Econometric Practice, chap. 8.Google Scholar

²⁰ Sanders, , ‘Government Popularity and the Next General Election’Google Scholar; Sanders, , ‘Why the Conservatives Won’.Google Scholar Unlike his analyses, which use macroeconomic forecasts in an auxiliary model predicting expectations, we utilize the Gallup expectations data.

²¹ The 1992 election variable is necessarily omitted from these forecasting models. In all cases, model estimates and diagnostics are similar to those presented in Table 1.

²² This is the Conservative vote share in Great Britain; the party's vote in the United Kingdom as a whole was 41.9 per cent. See Butler, David and Kavanagh, Dennis, The British General Election of 1992 (London: Macmillan, 1992), p. 286, Table A 1.2.CrossRefGoogle Scholar

²³ Sanders' variant of a personal expectations forecasting model does somewhat better, missing the 1992 Conservative share by approximately 1 percent. See Sanders, , ‘Government Popularity and the Next General Election’, p. 248Google Scholar; Sanders, , ‘Forecasting the 1992 British General Election Outcome’, p. 104–5.Google Scholar Since both his analysis and ours show that personal expectations models forecast well over short and longer time horizons, there is no need to entertain alternative specifications.

²⁴ Mills, Terence C., Time Series Techniques For Economists (Cambridge: Cambridge University Press, 1991), pp. 64–5.Google Scholar

²⁵ Granger, Clive W. J. and Newbold, Paul, ‘Spurious Regressions in Econometrics’, Journal of Econometrics, 2 (1974), 111–20CrossRefGoogle Scholar; Hendry, David, ‘Econometrics - Alchemy or Science?’, Economica, 47 (1980), 387–406.CrossRefGoogle Scholar

²⁶ Beck, Nathaniel, ‘Comparing Dynamic Specifications: The Case of Presidential Approval’Google Scholar, in Stimson, , ed., Political Analysis, Vol. 3, pp. 67–9.Google Scholar

²⁷ See Appendix.

²⁸ See, for example, Clarke, Harold D., Mishler, William and Whiteley, Paul, ‘Recapturing the Falklands: Models of Conservative Popularity, 1979–1983’, British Journal of Political Science, 20 (1990), 63–81CrossRefGoogle Scholar; Clarke, Harold D. and Whiteley, Paul, ‘Perceptions of Macroeconomic Performance, Government Support and Conservative Party Strategy, 1983–1987’, European Journal of Political Research, 18 (1990), 97–120CrossRefGoogle Scholar; Norpoth, , ‘Guns and Butter’Google Scholar; Norpoth, , Confidence Regained, chaps 8–10.Google Scholar On Box-Jenkins-Tiao models, see Beck, , ‘Comparing Dynamic Specifications’, pp. 73–81.Google Scholar

²⁹ See, for example, Clarke, and Whiteley, , ‘Perceptions of Macroeconomic Performance’Google Scholar; Hudson, , ‘Prime Ministerial Popularity’.Google Scholar

³⁰ Crewe, Ivor and King, Anthony, ‘Did Major Win? Did Kinnock Lose? Leadership Effects in the 1992 Election’, in Heath, Anthony et al. , (eds), Labour's Last Chance? The 1992 Election and Beyond (Aldershot, Hants: Dartmouth, 1994), pp. 125–48.Google Scholar

³¹ See, for example, Butler, and Stokes, , Political Change in BritainGoogle Scholar, chap. 16; Crewe, and King, , ‘Did Major Win? Did Kinnock Lose?’Google Scholar; Stewart, Marianne C. and Clarke, Harold D., ‘The (Un)Importance of Party Leaders: Leader Images and Party Choice in the 1987 British Election’, Journal of Politics, 54 (1992), 447–70.CrossRefGoogle Scholar Although some analysts have demonstrated significant leader effects, they have been reluctant to accord them much importance. Butler and Stokes, for example, conclude that ‘the pull of the leaders remains but one among the factors that determine transient shifts of party strength; it is easily outweighed by other issues and events of concern to the public, including the movements of the economy, which do so much to set the climate of the party battle’. See Butler, and Stokes, , Political Change in Britain, p. 244.Google Scholar See also Crewe, Ivor, ‘How to Win a Landslide Without Really Trying: Why the Conservatives Won in 1983’, in Ranney, Austin, ed., Britain at the Polls, 1983: A Study of the General Election (Durham, NC: Duke University Press, 1985)Google Scholar; Dunleavy, Patrick and Husbands, Christopher T., British Democracy at the Crossroads: Voting and Party Competition in the 1980s (London: Allen & Unwin, 1985)Google Scholar, chap. 5; Rose, Richard and McAllister, Ian, The Loyalties of Voters: A Lifetime Learning Model (London: Sage Publications, 1990), pp. 132–42.Google Scholar

³² The Gallup prime ministerial satisfaction question is: ‘Are you satisfied or dissatisfied with—as Prime Minister?’

³³ Consistent with Sanders' argument, the data in Figure 3 suggest that the initial impact of the party's leadership change on Conservative support gradually eroded during 1991. See Sanders, , ‘Government Popularity and the Next General Election’, pp. 253–8.Google Scholar However, analyses presented below indicate that prime ministerial approval has both short-term and long-term effects on governing party support. Given that Major's approval ratings were much higher than Thatcher's had been prior to her ouster, it may be argued that both types of effects were beneficial for the Tories in the run-up to their 1992 re-election bid.

³⁴ Charemza, and Deadman, , New Directions in Econometric PracticeGoogle Scholar, chap. 5. The critical values for these tests and the cointegration tests reported below are nonstandard. See Dickey, David A. and Fuller, Wayne A., ‘Distribution of the Estimators for Autoregressive Series With a Unit Root’, Journal of the American Statistical Association, 74 (1979), 427–31Google Scholar; MacKinnon, James, ‘Critical Values for Cointegration Tests’, in Engle, Robert F. and Granger, Clive W. J., eds, Long-Run Economic Relationships (Oxford: Oxford University Press, 1991).Google Scholar

³⁵ Engle, Robert F. and Yoo, Sam, ‘Cointegrated Economic Time Series: An Overview with New Results’Google Scholar, in Engle, Robert F. and Granger, Clive W. J., eds, Long-Run Economic Relationships.Google Scholar

³⁶ The error correction mechanism terms used in these analyses are, then, the residuals from the cointegrating regressions summarized in Table 5.

³⁷ This is not to say that the analyses of the ECMs indicate that national economic evaluations are wholly irrelevant for Conservative support. Recall that the cointegrating regressions (Table 5) reveal that national retrospections, but not national prospections, are influential net of prime ministerial approval. This, together with the significant error correction mechanism in the NR model, suggests that national retrospections may exert long-term effects on Tory popularity. Encompassing tests pitting the PE and PR ECMs against their NR rival do not indicate a preferable specification. However, the fact that personal expectations and personal retrospections, but not national retrospections, have significant short-term effects suggests that the PE and PR models are preferable.

³⁸ Engle, Robert F., Hendry, David and Richard, Jean-François, ‘Exogeneity’, Econometrica, 51 (1983), 277–304.CrossRefGoogle Scholar

³⁹ Stewart, and Clarke, , ‘The (Un)Importance of Party Leaders’Google Scholar; see also Miller, William L. et al. , How Voters Change: The 1987 British Election Campaign in Perspective (Oxford: Oxford University Press, 1990), chap. 7.Google Scholar

⁴⁰ Crewe, and King, , ‘Did Major Win? Did Kinnock Lose?’Google Scholar

⁴¹ Charemza, and Deadman, , New Directions in Econometric Practice, pp. 265–6.Google Scholar

⁴² The variable measuring the replacement of Thatcher by Major is scored + 1 for December 1990 (Major became Conservative leader on 28 November 1990), and 0 for other months.

⁴³ Granato, , ‘An Agenda for Economic Model Building’, pp. 128–30.Google Scholar

⁴⁴ These models do not include error correction mechanisms to capture possible long-term relationships between prime ministerial approval and subjective economic evaluations. Analyses indicate that national and personal economic retrospections are not cointegrated with prime ministerial approval–the R²s are very small (0.12 and 0.08, respectively) and the residuals are nonstationary. Cointegration analyses using national and personal economic prospections yield stationary residuals, but again the R²s are small (0.28 and 0.25, respectively). PE and NE models that include error correction mechanisms produce very similar estimates to those in Table 8; in particular, the coefficients for personal and national expectations are 0.12 and 0.05, respectively, (p ≤ 0.05). The coefficients for the error correction mechanisms are significant (p ≤ 0.05), but quite small ( − 0.10 and − 0.11), thereby suggesting that short-term shocks to prime ministerial approval would be eroded only very slowly by any long-term relationships involving approval and economic expectations.

⁴⁵ Norpoth, , ‘Guns and Butter’Google Scholar; Norpoth, , Confidence Regained, pp. 146–51.Google Scholar

⁴⁶ The Falklands impacts are permanent step-level effects in these models. An alternative gradual-temporary specification, with impacts in May and June 1982 that subsequently decay, i.e., (β_t−1 + β_t−2)/(1 − δB), yields very similar estimates to those in Table 8 (the estimates for other variables are very similar as well). The adjustment (δ) parameter is very large ( + 0.97), thereby indicating that the war's effect on Mrs Thatcher's approval ratings eroded very slowly. This large (δ approximates that ( + 0.94) reported by Norpoth, , Confidence Regained, p. 147.Google Scholar

⁴⁷ Stigler, George, ‘General Economic Conditions and National Elections’, American Economic Review, 63 (1973), 160–7, p. 160.Google Scholar

⁴⁸ See especially MacKuen, , Erikson, and Stimson, , ‘Peasants or Bankers?’Google Scholar Note, however, that their claim that national economic prospections are all that matters in presidential approval models is not universally accepted. For example, Lewis-Beck, Economics and Elections, chap. 8, argues that both national prospections and national retrospections are influential. For a similar argument, see Clarke, Harold D. and Stewart, Marianne C., ‘Prospections, Retrospections and Rationality: The “Bankers' Model” of Presidential Approval Reconsidered’, American Journal of Political Science, 38 (1994), 1104–23.CrossRefGoogle Scholar

⁴⁹ The major controversies are reviewed by contributors to a recent symposium on party identification in Political Behavior, 14 (1992), 193–360.Google Scholar See also Budge, Ian, Crewe, Ivor and Farlie, Dennis, Party Identification and Beyond (New York: Wiley, 1976).Google Scholar

⁵⁰ See Conover, Pamela J. and Feldman, Stanley, ‘Emotional Reactions to the Economy: I'm Mad as Hell and I'm Not Going to Take It Anymore’, American Journal of Political Science, 30 (1986), 50–78CrossRefGoogle Scholar; Conover, Pamela J., Feldman, Stanley and Knight, Kathleen, ‘The Personal and Political Underpinnings of Economic Forecasts’, American Journal of Political Science, 31 (1987), 559–83.CrossRefGoogle Scholar

⁵¹ The research is being conducted in collaboration with Paul Whiteley, College of William and Mary, with funds provided by the National Science Foundation (U.S.), grant #SES-9309018.

⁵² See, for example, Gilbert, Christopher L., ‘Professor Hendry's Econometric Methodology’, in Granger, Clive W. J., ed., Modelling Economic Series (Oxford: Oxford University Press, 1990).Google Scholar

⁵³ See Engle, Robert F. and Granger, Clive W. J., ‘Co-integration and Error Correction: Representation, Estimation and Testing’, Econometrica, 55 (1987), 251–76.CrossRefGoogle Scholar Beck recently has argued that the ECM specification also is applicable in situations when one variable is stationary, but very ‘long-memoried’, whereas another is nonstationary (Beck, , ‘Comparing Dynamic Specifications’, pp. 69–73Google Scholar; Beck, Nathaniel, ‘The Methodology of Cointegration’, in Freeman, John R., ed., Political Analysis, Vol. 4 (Ann Arbor: University of Michigan Press, 1993)).Google Scholar

⁵⁴ See Engle, and Granger, , Long-Run Economic Relationships, chap. 1.Google Scholar

⁵⁵ Engle, and Granger, , ‘Co-integration and Error Correction’Google Scholar; Engle, and Yoo, , ‘Cointegrated Economic Time Series’.Google Scholar

⁵⁶ The literature on ECMs and the related topic of cointegration has expanded rapidly in recent years. The works by Beck cited in fn. 53 provide a useful introduction. Charemza, and Deadman, , New Directions in Econometric PracticeGoogle Scholar, chap. 5, discuss ECMs in terms of the broader context of the LSE approach.