It is widely assumed that citizens are myopic, weighing policies’ short-term consequences more heavily than long-term outcomes. Yet no study of public opinion has directly examined whether or why the timing of future policy consequences shapes citizens’ policy attitudes. This article reports the results of an experiment designed to test for the presence and mechanisms of time-discounting in the mass public. The analysis yields evidence of significant discounting of delayed policy benefits and indicates that citizens’ policy bias towards the present derives in large part from uncertainty about the long term: uncertainty about both long-run processes of policy causation and long-term political commitments. There is, in contrast, little evidence that positive time-preferences (impatience) or consumption-smoothing are significant sources of myopic policy attitudes.
Department of Political Science, University of British Columbia (email:
1 Weaver R. Kent, ‘The Politics of Blame Avoidance’, Journal of Public Policy 6 (1986), 371–398
Nordhaus William D., ‘The Political Business Cycle’, Review of Economic Studies, 42 (1975), 169–190
Przeworski Adam, Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America (New York: Cambridge University Press, 1991)
Pierson Paul, Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment (New York: Cambridge University Press, 1994)
Alesina Alberto and Roubini Nouriel, ‘Political Cycles in OECD Economies’, Review of Economic Studies, 59 (1992), 663–688
2 Lewis-Beck Michael, ‘Comparative Economic Voting: Britain, France, Germany, Italy’, American Journal of Political Science, 30 (1986), 315–346
Erikson Robert S., MacKuen Michael B. and Stimson James A., ‘Bankers or Peasants Revisited: Economic Expectations and Presidential Approval’, Electoral Studies, 19 (2000), 295–312
3 Adam S. Booij and Bernard M. S. van Praag, ‘A Simultaneous Approach to the Estimation of Risk Aversion and the Subjective Time Discount Rate’, Journal of Economic Behaviour & Organization, 70 (2009), 374–388
4 Japelli Tullio and Pistaferri Luigi, ‘Intertemporal Choice and Consumption Mobility’, Journal of the European Economic Association, 4 (2006), 75–115
5 North Douglass C. and Weingast Barry R., ‘Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England’, Journal of Economic History, 49 (1989), 803–832
Weingast Barry R. and Marshall William J., ‘The Industrial Organization of Congress; or, Why Legislatures, Like Firms, Are Not Organized as Markets’, Journal of Political Economy, 96 (1988), 132–163
6 Huber Evelyne and Stephens John D., Development and Crisis of the Welfare State: Parties and Policies in Global Markets (Chicago: University of Chicago Press, 2001)
Stokes Susan C., ‘Public Opinion and Market Reforms: The Limits of Economic Voting’, Comparative Political Studies, 29 (1996), 499–519
7 See, for example, Lewis-Beck, ‘Comparative Economic Voting’; Erikson, MacKuen and Stimson, ‘Bankers or Peasants Revisited’.
8 Rhodebeck Laurie A., ‘The Politics of Greed? Political Preferences among the Elderly’, Journal of Politics 55 (1993), 342–364
Chong Dennis, Citrin Jack and Conley Patricia, ‘When Self-Interest Matters’, Political Psychology, 22 (2001), 541–570
Jon Krosnick et al., ‘The Origins and Consequences of Democratic Citizens’ Policy Agendas: A Study of Popular Concern About Global Warming’, Climatic Change, 77 (2006), 7–43
Walstad William B., ‘The Effect of Economic Knowledge on Public Opinion of Economic Issues’, Journal of Economic Education, 28 (1997), 195–205
9 Healy Andrew and Malhotra Neil, ‘Myopic Voters and Natural Disaster Policy’, American Political Science Review, 103 (2009), 387–406
10 Healy and Malhotra find no temporal effect. This might derive from the relatively small (two-year) variance in timing examined.
11 Samuelson Paul A., ‘A Note on Measurement of Utility’, Review of Economic Studies, 4 (1937), 155–161
12 Friedman Milton, A Theory of the Consumption Function (Princeton, N.J.: Princeton University Press, 1957)
Kenneth K. Kurihara, ed., Post-Keynesian Economics (New Brunswick, N.J.: Rutgers University Press, 1954)
Fisher Irving, The Theory of Interest as Determined by Impatience to Spend Income and Opportunity to Invest It (New York: Macmillan, 1930)
13 Keren Gideon and Roelofsma Peter, ‘Immediacy and Certainty in Intertemporal Choice’, Organizational Behaviour and Human Decision Processes, 63 (1995), 287–297
Halevy Yoram, ‘Strotz Meets Allais: Diminishing Impatience and the Certainty Effect’, American Economic Review, 98 (2008), 1145–1162
14 Vonk Roos, ‘The Negativity Effect in Trait Ratings and in Open-Ended Descriptions of Persons’, Personality and Social Psychology Bulletin, 19 (1993), 269–278
Lau Richard R., ‘Two Explanations for Negativity Effects in Political Behaviour’, American Journal of Political Science, 29 (1985), 119–138
Klein Jill G., ‘Negativity Effects in Impression Formation: A Test in the Political Arena’, Personality and Social Psychology Bulletin, 17 (1991), 412–418
15 Vaughn Leigh Ann and Weary Gifford, ‘Roles of the Availability of Explanations, Feelings of Ease, and Dysphoria in Judgments About the Future’, Journal of Social and Clinical Psychology, 21 (2002), 686–704
Tversky Amos and Kahneman Daniel, ‘Availability: A Heuristic for Judging Frequency’, Cognitive Psychology 5 (1973), 207–232
16 Gneezy Uri, List John A. and Wu George, ‘The Uncertainty Effect: When a Risky Prospect Is Valued Less Than Its Worst Possible Outcome’, Quarterly Journal of Economics, 121 (2006), 1283–1309
17 Some investments may convert fungible tax dollars into forms of physical or human capital that are difficult for politicians to ‘raid’ later. Until the non-fungible capital has been fully created, however, the extracted tax dollars are still at risk of diversion, and that risk should be higher the longer it will take to produce the capital.
18 Alesina Alberto and Tabellini Guido, ‘Credibility and Politics’, European Economic Review, 32 (1988), 542–550
Persson Torsten and Tabellini Guido, ‘Representative Democracy and Capital Taxation’, Journal of Public Economics, 55 (1994), 53–70
19 Horn Murray J. and Shepsle Kenneth A., ‘Commentary on “Administrative Arrangements and the Political Control of Agencies”: Administrative Process and Organizational Form as Legislative Responses to Agency Costs’, Virginia Law Review, 75 (1995), 499–508
20 Claude Ménard and Mary M. Shirley, eds, Handbook of New Institutional Economics (Berlin: Springer, 2008)
21 Chanley Virginia A., Rudolph Thomas J. and Rahn Wendy M., ‘The Origins and Consequences of Public Trust in Government: A Time Series Analysis’, Public Opinion Quarterly, 64 (2000), 239–256
Hetherington Marc J., Why Trust Matters: Declining Political Trust and the Demise of American Liberalism (Princeton, N.J.: Princeton University Press, 2005)
Scholz John T. and Lubell Mark, ‘Adaptive Political Attitudes: Duty, Trust, and Fear as Monitors of Tax Policy’, American Journal of Political Science, 42 (1998), 903–920
22 Shanto Iyengar and William J. McGuire, eds, Explorations in Political Psychology (Durham, N.C.: Duke University Press, 1993)
Sniderman Paul M., Brody Richard A. and Tetlock Philip, Reasoning and Choice: Explorations in Political Psychology (New York: Cambridge University Press, 1991)
Feldman Stanley and Zaller John, ‘The Political Culture of Ambivalence: Ideological Responses to the Welfare State’, American Journal of Political Science, 36 (1992), 268–307
23 Carpini Michael X. Delli and Keeter Scott, What Americans Know About Politics and Why It Matters (New Haven, Conn.: Yale University Press, 1997)
Lupia Arthur and McCubbins Mathew D., The Democratic Dilemma: Can Citizens Learn What They Need to Know? (New York: Cambridge University Press, 1998)
24 The survey experiment was fielded by Knowledge Networks, Inc. (KN) in two waves in 2008: 7 to 18 February and 23 July to 15 August. A comparison of the sample distributions of key demographic variables in our sample with the American National Election Studies 2008 Time Series Study reveals close comparability.
25 Board of Trustees, ‘Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds’ (Washington, D.C.: Government Printing Office, 2007).
26 This solution resembles the Social Security reforms of 1977 and 1983.
27 Schieber Sylvester J. and Shoven John B., The Real Deal: The History and Future of Social Security (New Haven: Yale University Press, 1999)
28 For instance, at the annual i of 2.9 per cent assumed by the Social Security Administration, B would have to be roughly 17 times higher in the 40-year than in the 5-year condition. See Board of Trustees, ‘Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds’.
29 Lusardi Annamaria and Mitchell Olivia S., ‘How Ordinary Consumers Make Complex Financial Decisions: Financial Literacy and Retirement Readiness’ (Cambridge, Mass.: National Bureau of Economic Research, 2009)
30 We might attempt to avoid the confound with periodic costs and benefits by informing subjects of the policy's total costs and benefits and holding these quantities (and thus the policy's NPV) constant across timing conditions. Subjects could be told, for instance, that they will pay total costs with ‘a present value of $14,095’ (the sum of the discounted costs described in the present study, in the forty-year condition, assuming i = 2.9 per cent) and, after five (or forty) years, will receive benefits with ‘a present value of $28,190’ (twice the discounted costs). This strategy would, however, leave subjects to guess about a critical temporal feature of the policy – the distribution of those costs over time. Would, for instance, the $14,095 be paid all at once in Year 1 or in equal instalments over five (or forty) years? It is unclear how subjects could engage in meaningful intertemporal reasoning in the absence of such information. Moreover, if a non-trivial proportion of subjects assumed a smooth distribution of costs over time (the likeliest profile), then the average perceived periodic costs of the policy tradeoff would still vary with the timing manipulation.
31 The three items were: ‘Now we would like to know what you think about the proposed reform to the Social Security system described in the presentation. Do you support or oppose the reform plan?’ ‘Now, please indicate how strongly you agree or disagree with the following statement. “I would be willing to pay the costs of the Social Security reform plan described in the presentation.” ’; ‘Now, please indicate how strongly you agree or disagree with the following statement. “It would be a good idea for the government to adopt the Social Security reform plan described in the presentation.” ’ Each item offered subjects four response levels. We code these items with equal intervals between responses.
32 All pairwise correlations (r) between index components exceed 0.80. The mean of the index is 0.40 and the standard deviation is 0.29.
33 In identifying these controls, we began by estimating a regression model of reform support that included all variables in our dataset potentially relevant to subjects’ orientation towards the reform. This included variables affecting individuals’ distributive stake in the reform (age, income, employment status, retired status, number of children); cognitive engagement with politics (political and economic knowledge, need for cognition, education); political orientations (egalitarianism, congressional and presidential approval, support for ‘less government’ and for Social Security spending, gender, race); economic perceptions (personal current, personal prospective, national current, national prospective); and interview start and completion dates. We employ as our standard controls those variables that were statistically significant in this larger model. All demographic variables were collected separately by KN.
34 Complete question wordings for all measures are available in the on-line appendix.
35 Tomz Michael, Wittenberg Jason and King Gary, ‘Clarify: Software for Interpreting and Presenting Statistical Results’, Journal of Statistical Software, 8 (2003)
Brambor Thomas, Clark William Roberts and Golder Matt, ‘Understanding Interaction Models: Improving Empirical Analyses’, Political Analysis, 14 (2006), 63–82
36 Frederick Shane, Loewenstein George and O’Donoghue Ted, ‘Time Discounting and Time Preference: A Critical Review’, Journal of Economic Literature, 40 (2002), 351–401
37 Citations to studies correlating these variables to time preferences can be found in the on-line appendix.
38 In the curvilinear model, we centre age to reduce multicollinearity between age and its square.
39 In the context of this experiment, we note that older respondents should have more reason to care about the survival of the Social Security programme and may, thus, be more supportive of reform that would save it. This fact, however, should not affect age's interaction with the timing of policy effects, as older respondents (assuming they are less patient than younger respondents) should still have a stronger preference for the quicker delivery of benefits.
40 We report results for just one model specification for each impatience-proxy. The results are substantively unchanged across all alternative specifications described above; full results for all models are available in the on-line appendix.
41 The questions are based on the standard ANES economic perceptions battery. However, whereas in the ANES respondents are asked about ‘the next 12 months’, our item asks about the period ‘5 to 10 years from now’ in order to elicit judgments over a time horizon relevant to the reform issue.
42 The difference of means is 0.04 (p = 0.017).
43 Predicted values in this article are based on simulations produced using the clarify add-on to stata with variables other than those of interest set at their sample means. See Tomz, Wittenberg and King, ‘Clarify: Software for Interpreting and Presenting Statistical Results’.
44 See, e.g., Lupia and McCubbins, The Democratic Dilemma.
45 Results are given in the on-line appendix.
46 As only twenty-five respondents gave the most trusting response on the ‘waste’ measure, we group together responses at the two most trusting levels of the measure.
47 Separate models were estimated for each variable and included both ‘main’ and interactive effects (Timing×control) for all controls. Note that age is included as both a linear and a quadratic term. For variables not described elsewhere in this article, descriptions are available in the on-line appendix.
48 See on-line appendix for these results.
49 We also tested for another potential explanation: that the moderation of timing effects by fiscal trust might have reflected a tendency by high-trusters not to scrutinize – and, thus, not to absorb – the temporal features of the reform described in the policy brief. We collected data on two indicators of attention to and reception of the information in the policy brief: length of time subjects spent reading the policy brief and, as discussed earlier, subjects’ timing perceptions. We find that, in the presence of appropriate controls, fiscal trust has no significant effect on either of these variables (analyses not reported), indicating that the trust effect does not operate through inattention to policy detail.
50 Bullock John G., Green Donald P. and Ha Shang E., ‘Yes, but What’s the Mechanism? (Don’t Expect an Easy Answer)’, Journal of Personality and Social Psychology, 98 (2010), 550–558
51 Our measure of time preference mentions includes all considerations suggestive of pure impatience, including references to the immediacy of the reforms costs; to the lower intrinsic value of future costs or benefits relative to present costs or benefits; and to concerns that benefits will arrive after the respondent has died. Consumption smoothing mentions consist of references to current and future expected economic circumstances, whether personal, group-specific (e.g., seniors), or national in scope; and to the greater ease of paying costs later than sooner. Uncertainty mentions include references to any form of uncertainty about whether the reform's benefits will emerge or be enjoyed by the respondent, including the possibility that the funds accumulated will be diverted or misspent by politicians; the past failure of politicians to responsibly manage the Social Security program; or uncertainty about Social Security's future financial difficulties.
52 The results are reported in the on-line appendix.
53 In all conditions, the reform's costs were described numerically.
54 This result also disposes of a further potential influence on intertemporal attitudes: inflation. With monetary payoffs, subjects might hypothetically perceive longer-delayed benefits to be smaller because of expected price increases over time. However, inflationary reasoning is inconsistent with the presence of a significantly stronger timing effect in the verbal condition, where respondents were given no monetary benefit value that could be adjusted for inflation. In addition, respondents in the numerical condition were told that all figures were expressed in current dollars.
55 Given a current US life expectancy of 78 years, we would not expect any age effect for subjects over about 40 years in the forty-year condition because age has virtually no effect on expected benefit-stream length within this subgroup: all subjects above 40 could expect personally to receive 0 years of benefits in the forty-year condition.
56 We add to our standard controls variables correlated with age that could influence reform support: party identification, political and economic knowledge, and support for Social Security spending.
57 Full results are provided in the on-line appendix.
58 Lawrence C. Dodd and Calvin Jillson, eds, The Dynamics of American Politics: Approaches and Interpretations (Boulder. Colo.: Westview Press, 1994)
Peters B. Guy and Savoie Donald, eds, Taking Stock: Assessing Public Sector Reforms (Montreal: McGill-Queen's University Press, 1998)
59 Scholz and Lubell, ‘Adaptive Political Attitudes: Duty, Trust, and Fear as Monitors of Tax Policy’; Chanley, Rudolph and Rahn, ‘The Origins and Consequences of Public Trust in Government’; Hetherington, Why Trust Matters.
60 Putnam Robert D., Leonardi Robert and Nanetti Raffaella, Making Democracy Work: Civic Traditions in Modern Italy (Princeton, N.J.: Princeton University Press, 1993)
61 Christiaan Grootaert and Thierry Van Bastelaer, eds, The Role of Social Capital in Development: An Empirical Assessment (Cambridge: Cambridge University Press 2002)
Knack Stephen and Keefer Philip, ‘Does Social Capital Have an Economic Payoff? A Cross-Country Investigation’, Quarterly Journal of Economics, 112 (2006), 1251–1288
62 See, e.g., Weingast and Marshall, ‘The Industrial Organization of Congress’; North and Weingast, ‘Constitutions and Commitment’.
* Department of Political Science, University of British Columbia (email: email@example.com); and Department of Political Studies, Queen's University at Kingston, Ontario, respectively. The authors acknowledge the generous support of the Social Sciences and Humanities Research Council of Canada (#410-2006-1174) and the UBC Hampton Fund. For detailed comments on the paper or study protocol, we thank John Bullock, Fred Cutler, Elizabeth Goodyear-Grant, Benjamin Nyblade, Andrew Owen, Dimitrios Panagos, Paul Quirk, and the journal's three anonymous reviewers. We also received helpful advice from Jason Barabas, Mick Couper, Brian Gaines, Donald Green, Sunshine Hillygus, Donald P. Haider-Markel, Simon Jackman, Richard Johnston, Jeffrey Mondak, Angela O'Mahony, Diana Mutz and Nicholas Winter. For superb research assistance, we thank Nico Dragojlovic, Anna Drake and Steven Klein. A Supplementary Appendix containing additional information is available online at: http://dx.doi.org/10.1017/S0007123412000117
Email your librarian or administrator to recommend adding this journal to your organisation's collection.
Full text views reflects the number of PDF downloads, PDFs sent to Google Drive, Dropbox and Kindle and HTML full text views.
* Views captured on Cambridge Core between September 2016 - 21st October 2017. This data will be updated every 24 hours.