Retrospective Economic Voting and the Intertemporal Dynamics of Electoral Accountability in the American States George A. Krause † University of Pittsburgh and Benjamin F. Melusky ‡ University of Pittsburgh Draft as of Monday, August 27, 2012 * An earlier version of this paper was presented at the 2012 annual meetings of the Midwest Political Science Association. Chicago, IL. April. We thank Dave Peterson for helpful comments on a preliminary draft of this essay. † Professor, Department of Political Science, University of Pittsburgh, 4442 Wesley W. Posvar Hall. Pittsburgh, PA 15260. [email protected](e-mail address). Corresponding Author. ‡ Ph.D. student, Department of Political Science, University of Pittsburgh, 4600 Wesley W. Posvar Hall. Pittsburgh, PA 15260. [email protected](e-mail address). Keywords: Dynamic Electoral Accountability, Intertemporal Attribution Discounting, Retrospective Economic Voting, Elections in the American States
38
Embed
Retrospective Economic Voting and the Intertemporal ...gkrause/Intertemporal Attribution...into an attribution (Ericsson and Simon 1981: 18-19). Hence, it becomes difficult for individuals
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Retrospective Economic Voting and the Intertemporal Dynamics of Electoral Accountability in the American States
George A. Krause† University of Pittsburgh
and
Benjamin F. Melusky‡
University of Pittsburgh
Draft as of Monday, August 27, 2012
* An earlier version of this paper was presented at the 2012 annual meetings of the Midwest Political Science Association. Chicago, IL. April. We thank Dave Peterson for helpful comments on a preliminary draft of this essay.
† Professor, Department of Political Science, University of Pittsburgh, 4442 Wesley W. Posvar Hall. Pittsburgh, PA 15260. [email protected] (e-mail address). Corresponding Author.
‡ Ph.D. student, Department of Political Science, University of Pittsburgh, 4600 Wesley W. Posvar Hall. Pittsburgh, PA 15260. [email protected] (e-mail address).
Keywords: Dynamic Electoral Accountability, Intertemporal Attribution Discounting, Retrospective Economic Voting, Elections in the American States
Abstract A theory of intertemporal attribution is advanced to explain how voters discount
sanctions and rewards for politicians’ past performance in office when the latter runs for elective
office as a non-incumbent candidate. Analysis of aggregate electoral data on incumbent and
non−incumbent (i.e., outgoing and former) governors in the American states generally offers
strong support for the testable implications of this theory. For both recent term and tenure
evaluative time frames, voters possess the strongest economic attributions for incumbent
(current) governors, while dynamically discounting the economic stewardship of outgoing or
former governors. Intertemporal attribution discounting is increasing in the width of the
evaluative time used by voters to assess economic stewardship. These findings underscore the
dynamic slippage present in electoral accountability by demonstrating how time elapsed from
prior service in elective office not only makes it increasingly difficult for voters to effectively
reward competent politicians, but to also sanction incompetent counterparts.
1
INTRODUCTION
Holding politicians electorally accountable for their performance in office is a central
tenet in representative democracies (Key 1966). Voters’ capacity to hold politicians accountable
for their performance in office is directly linked to not only their attributions regarding
policymaking competence via macroeconomic conditions (e.g., Ansoloabehere, Meredith, and
Snowberg 2011; Gomez and Wilson 2003; Healey and Lenz 2012; Powell and Whitten 1993;
Markus 1988; Lewis-Beck and Stegmaier 2007), but also linking this competence to a particular
politician (or party) in a consistent manner. With few exceptions (e.g., Schwabe 2011), canonical
theories of retrospective economic voting pertain to a canonical two-period logic, whereby
re−election to the same office in period t is based on economic performance in period t −1
when a narrow evaluative time frame is employed. In other words, the weight attached to a
governor’s economic stewardship during a former governor’s last year in office increases as time
elapses between their exit from office and their re-entry into the electoral arena.
25
DISCUSSION
Electoral accountability is a fundamental criterion by which all representative
democracies are evaluated. If citizens cannot reward competent politicians and sanction
incompetent ones, then it calls into question the viability of democratic institutions to serve the
interests of the broader polity. At its core, electoral accountability most often translates into
understanding the voters’ ability to effectively reward and sanction politicians for their
policymaking performance at the polls. Retrospective economic voting is the most common, and
perhaps the most effective, channel by which voters make assessments regarding the
performance of their elected representatives (e.g., Barro 1973; Downs 1957; Ferejohn 1986;
Fiorina 1981). Current theories and empirical evidence of retrospective voting are generally
rooted in a two-period view of the world in which the incumbent party is evaluated for a given
political office at time t based on their stewardship of the economy at time t-1 (but see Schwabe
2011). The extant research does not explicitly consider the intertemporal dynamics of
retrospective economic voting behavior. Extant research that focuses on party responsibility
undervalues the importance of the actual stewards (persons) that actually appear on the ballot
seeking elective office, and their corresponding individual reputation for policymaking
competence. That is, whether voters deem a politician as being a ‘good type’ or ‘bad type’ in
principal-agent terms, depends upon a given individual’s record in elective office.
To address both issues, a theory of intertemporal attribution discounting (IAD) has been
advanced to facilitate our understanding as to how voters (in aggregate) retrospectively discount
past economic stewardship of elected officials seeking other elective offices or the same office at
some point in the future. This puzzle deals with the tangible reality that elected officials often
have more than one act to their political careers. Attention is restricted to aggregate electoral
voting outcomes for incumbent governors seeking re-election and non-incumbent (i.e., outgoing
26
or former) governors running in either gubernatorial or U.S. Senate election contests. Further,
attention is restricted to a common political jurisdiction (i.e., major statewide elective office) to
ensure comparability in analyzing electoral accountability. The statistical evidence offers strong
support for several of the predictions generated by IAD theory. For instance, voters discount a
former politician’s past economic stewardship through time when determining whether to offer
electoral support to these particular individuals in subsequent election contests when evaluating
them on their tenure or most recent term in elective office. Voters’ attributions of economic
stewardship are not only weaker when evaluated over the politician’s tenure in office, but also
discount at a faster rate than compared to when the voters’ reference point for assessing
economic stewardship is anchored to the more recent past (i.e., last term in office). In the
exceptional case of most recent (final) year retrospective economic attributions, voters exhibit an
intertemporal anchoring bias, whereby they compound their assessments of the governor’s
performance in office when evaluating these individuals for elective office at a later date.
Although dynamic electoral accountability appears to be more effective in this latter instance, it
is a mirage since it is premised on a highly biased evaluation of governors’ economic
stewardship in office because it only comprises a single year that represents anywhere from
12.5% to 25% of their time in this elective office for almost 90% of the effective sample cases.
These findings have critical implications for understanding electoral accountability in a
dynamic setting. Because voters increasingly discount past economic performance associated
with particular politicians through time with respect aggregate electoral choice, politicians
bearing a strong record of economic stewardship, based on a sufficiently informative record in
office, should seek higher office not long after stepping out of their previous office. Conversely,
politicians with a poor record of economic stewardship under such circumstances can benefit
27
from a prolonged hiatus seeking elective office in order to ensure that ‘old wounds’ can heal with
the passage of time. That is, more distant retrospective economic voting assessments are
discounted at a greater rate than less distant ones. This claim is the mirror image of voters
possessing information and cognitive limitations, and thus employ simplifying heuristics in order
to effectively hold elected officials accountable (Lupia 1994), including the weighting of only
the most recent past information in electoral choice (Healy and Lenz 2012).
On a normative level, this study suggests that electoral rules such as term−limits, that
often produce former governors seeking statewide office at some point in the future, may
actually make it more difficult for voters to collectively weed out competent politicians from
incompetent ones. Both the logic and evidence presented here thus complements scholarship
which claim that reducing institutional restrictions associated with elective office actually
increases the quality of electoral accountability in a democracy (Besley and Case 1995; Besley
2006). Moreover, the application of overly-simplified heuristics, such as an end-heuristic
identified by Healy and Lenz (2012), will only exacerbate the difficulty that voters confront
when discriminating between competent and incompetent politicians in dynamic settings. From
the politician’s perspective, differential strategies should be pursued by competent and
incompetent agents. When politicians are competent types (i.e., excellent economic stewards),
they should seek elective office either concurrently or in the not too distant future by running on
their positive economic record based on the most recent term of service. Conversely,
incompetent politicians (i.e., poor economic stewards) should defer to seek elective office for a
lengthy period of time into the future until voters’ collective attributions begin to reflect
intertemporal regressive bias, and when they do re−enter the electoral arena they should
emphasize their overall economic record from their tenure in office. Because voters’ ability to
28
distinguish between competent and incompetent politicians based on prior service becomes
increasingly blurred through time, politicians’ individual-level reputations can be thought of as a
rapidly depreciating asset that loses value when a politician takes a hiatus from elective office.
Future research should pursue developing unified theories of attribution that integrate insights
from the IAD theory, with insights generated spatial theories of heterogeneous attributions
rooted in political sophistication (Gomez and Wilson 2001, 2003), partisan identification
(Rudolph 2003), or geographical considerations (Ansolabehere, Meredith, and Snowberg 2011).
Acquiring a better understanding of the context in which the reputational durability of politicians
systematically varies can serve to improve electoral accountability in dynamic settings.
29
REFERENCES
Angrist, Joshua D., Victor Chernozhukov, and Ivan Fernández-Val. 2006. “Quantile Regression
Under Misspecification, with an Application to the U.S. Wage Structure.” Econometrica
74(March): 539-563.
Ansolabehere, Stephen and James M. Snyder Jr. 2002. “The Incumbency Advantage in U.S.
Elections: An Analysis of State and Federal Offices, 1942–2000.” Election Law Journal:
Rules, Politics, and Policy 1(June): 315-338.
Ansolabehere, Stephen, Marc Meredith, and Erik Snowberg. 2011. “Mecro−Economic Voting:
Local Information and the Micro−Perceptions of the Macro−Economy.” Harvard
University. Typescript.
Barro, Robert J. 1973. “The Control of Politicians: An Economic Model.” Public Choice
14(Spring): 19-42.
Berry, William D., Evan J. Ringquist, Richard C. Fording, and Russell L. Hanson. 1998. “Measuring Citizen and Government Ideology in the American States, 1960-93.” American Journal of Political Science 42(January): 327-348.
Besley, Timothy and Anne Case. 1995. “Does Electoral Accountability Affect Economic Policy
Choices? Evidence from Gubernatorial Term Limits.” The Quarterly Journal of
Economics 110(August): 769-798.
Besley, Timothy. 2006. Principled Agents: The Political Economy of Good Government.
New York: Oxford University Press.
Bloom, Howard S. and Douglas H. Price. 1975. “Voter Response to Short-Run Economic
Conditions: The Asymmetric Effect of Prosperity and Recession.” American Political
Science Review 69 (March): 124-154.
Brady, David W., Hahrie Han, and Jeremy C. Pope. 2007. “Primary Elections and Candidate
30
Ideology: Out of Step with the Primary Electorate?” Legislative Studies Quarterly
32(February): 79-105.
Chubb, John E. 1988. “Institutions, The Economy, and The Dynamics of State Elections.” The
American Political Science Review 82(March): 133-154.
Cohen, Jeffrey E. and James D. King. 2004. “Relative Unemployment and Gubernatorial
Popularity.” The Journal of Politics 66(November): 1267-1282.
Dometrius, Nelson. 1999. “Governors: Their Heritage and Future.” In American State and Local
Politics: Directions for the 21st Century, edited by Ronald Weber and Paul Brace. New
York: Chatham House Publishers, Seven Bridges Press.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper Row.
Ericsson, K. Anders, and Herbert A. Simon. 1981. “Sources of Evidence on Cognition: A
Historical Overview.” In Thomas V. Merluzzi, Carol R. Glass, Myles Genest, eds.
Cognitive Assessment. New York: Guilford Press. Chapter 2 (pp. 16-51).
Ferejohn, John. 1986. “Incumbent Performance and Electoral Control.” Public Choice 50 (March): 5-25.
Fiorina, Morris P. 1981. Retrospective Voting in American National Elections. New Haven: Yale
University Press.
Gomez, Brad T. and J. Matthew Wilson. 2001. “Political Sophistication and Economic Voting in
the Electorate in the American Electorate: A Theory of Heterogeneous Attribution.”
American Journal of Political Science 45(October): 899-914.
Gomez, Brad T. and J. Matthew Wilson. 2003. “Causal Attribution and Economic Voting in
American Congressional Elections.” Political Research Quarterly 56(September): 271-
282.
Hanks, Christopher and Bernard Grofman. 1998. “Turnout in Gubernatorial and Senatorial
31
Primary and General Elections in the South, 1922-90: A Rational Choice Model of the
Effects of Short-Run and Long-Run Electoral Competition on Relative Turnout.” Public
Choice 94(3/4): 407-421.
Hansen, Susan B. 1999. “’Life Is Not Fair’: Governor’s Job Performance Ratings and State
Economies.” Political Research Quarterly 52(March): 167-188.
Healy, Andrew, and Gabriel S. Lenz. 2012. “Substituting the End for the Whole: Why Voters
Respond Primarily to the Election−Year Economy.” University of California-Berkeley.
Typescript.
Key, V. O. 1966. The Responsible Electorate: Rationality in Presidential Voting 1936-1960.
Cambridge, MA: Harvard University Press.
Kiewiet, D. Roderick, and Michael Udell. 1998. “Twenty-Five Years after Kramer: An
Assessment of Economic Retrospective Voting Based Upon Improved Estimates of
Income and Employment.” Economics and Politics 19(November): 219-248.
Kramer, Gerald H. 1971. “Short Term Fluctuations in U.S. Voting Behavior, 1896–1964.”
American Political Science Review 65(March): 131-43.
Lewis-Beck, Michael and Mary Stegmaier. 2007. “Economic Models of Voting.” In The Oxford
Handbook of Political Methodology, eds., Russell Dalton and Hans-Dieter Klingemann.
Lupia, Arthur. 1994. “Encyclopedias versus Shortcuts: Information and Voting in California
Insurance Reform Elections.” American Political Science Review 88(March): 63-76.
Machados, J.A.F., and J.M.C. Santos Silva. 2000. “Glejser’s Test Revisited.” Journal of
Econometrics 97(March): 189-202
Markus, Gregory B. 1988. “The Impact of Personal and National Economic Conditions on the
Presidential Vote: A Pooled Cross-Sectional Analysis.” American Journal of Political
Science 32(February): 137-154.
32
Niemi, Richard G., Harold W. Stanley, and Ronald J. Vogel. 1995. “State Economies and States
Taxes: Do Voters Hold Governors Accountable?” American Journal of Political Science
39(November): 936-957.
Partin, Randall W. 1995. “Economics Conditions And Gubernatorial Elections: Is the State
Executive Held Accountable?” American Politics Quarterly 23(January): 81-95.
Peltzman, Sam. 1987. “Economic Conditions and Gubernatorial Elections.” American Economic
Review 77(May): 293-297.
Powell, G. Bingham Jr. and Guy D. Whitten. 1993. “A Cross-National Analysis of Economic
Voting: Taking Account of the Political Context.” American Journal of Political Science
37(May): 391-414.
Rudolph, Thomas J. 2003. “Who's Responsible for the Economy? The Formation and
Consequences of Responsibility Attributions.” American Journal of Political Science
47(October): 698-713.
Schwabe, Rainer. 2011. “Reputation and Accountability in Repeated Elections.” Typescript.
Bank de Mexico.
Taber, Charles S., and Everett Young. 2011. “Information Processing, Public Opinion, and
Accountability.” In Accountability Through Public Opinion: From Inertia to Action.
Sina Odugbemi and Taeku Lee, editors. Washington, D.C.: The World Bank
Woehr, David J., and Jack. Feldman. 1993. “Processing Objective and Question Order Effects on
the Causal Relation Between Memory and Judgment in Performance Appraisal: The Tip
of the Iceberg.” Journal of Applied Psychology 78(2): 232-241.
Wolfers, Justin. 2007. “Are Voters Rational? Evidence from Gubernatorial Elections.” National
Bureau of Economic Research. Typescript.
33
Figure 1: Distribution of Major Statewide Elections For the Intertemporal Range: Sample of Current, Outgoing, and Former Governors
0.0
1.0
2.0
3.0
4D
ensi
ty
0 100 200 300 400
Election Delay (in Months)
34
TABLE 1: Regression Analysis Predicting Election Vote Share for Incumbent, Outgoing, and Former Governors in the American States Tenure Net Δ in UE Rate Recent Term Net Δ in UE Rate Recent Year Net Δ in UE Rate
OLS Beta Median OLS Beta Median OLS Beta Median Economic Stewardship & Attribution Discounting
∆ State Quarterly UE Rate × Election Delay
0.004 [0.007]
0.0001 [0.0003]
−0.015+ [0.010]
−0.004 [0.005]
−0.0002 [0.0002]
−0.012* [0.006]
0.011 [0.019]
0.0008 [0.0008]
0.065** [0.028]
∆ State Quarterly UE Rate 0.309
[0.393] 0.002
[0.0162] 0.641
[0.573] 1.085***
[0.388] 0.032**
[0.016] 1.190***
[0.435] −0.278
[0.927] −0.004
[0.038] −0.823
[1.016] Election Delay
(months) −0.258***
[0.061] −0.010***
[0.002] 0.130+ [0.096]
−0.099*** [0.031]
−0.005*** [0.001]
−0.065+ [0.042]
−0.097*** [0.031]
−0.005*** [0.001]
−0.075* [0.040]
Ancillary Controls Previous Gubernatorial
General Election Results 0.196* [0.100]
0.008* [0.004]
−0.023 [0.126]
0.212** [0.101]
0.008** [0.004]
0.165+ [0.125]
0.180* [0.101]
0.007* [0.004]
0.080 [0.113]
Non-Incumbent Governor −10.137***
[3.854] −0.351** [0.163]
−6.306 [6.348]
−15.299*** [3.776]
−0.538*** [0.159]
−12.138* [6.417]
−15.347*** [3.828]
−0.533*** [0.161]
−10.347+ [6.708]
General Election Contest
−17.363*** [2.494]
−0.722*** [0.109]
−19.363*** [5.582]
−17.004*** [2.431]
−0.696*** [0.105]
−18.164*** [5.810]
−16.744*** [2.459]
−0.685*** [0.105]
−17.841*** [3.914]
Incumbent Party −10.021+ [6.285]
−0.495* [0.270]
−14.484+
[9.751] −11.186* [5.899]
−0.553** [0.250]
−9.834 [9.289]
−11.235* [6.065]
−0.543** [0.256]
−16.986** [8.125]
Incumbency Deficit 1.647
[2.749] −0.165+ [0.119]
−3.831 [6.701]
−0.095 [2.612]
−0.209* [0.112]
−0.675 [5.039]
−0.476 [2.636]
−0.216* [0.112]
−0.998 [4.688]
State UE Rate (Election Quarter)
−0.726 [0.869]
−0.034 [0.037]
−0.747 [1.297]
0.361 [0.745]
0.0001 [0.032]
−0.228 [1.016]
−0.284 [0.732]
−0.018 [0.031]
−1.949** [0.796]
State UE Rate (EQ) × Incumbent Party
0.218 [0.945]
0.004 [0.040]
0.806 [1.279]
−0.259 [0.831]
−0.006 [0.035]
0.405 [1.087]
−0.289 [0.851]
−0.008 [0.036]
1.592* [0.852]
Absolute ∆ in VAP Since Governor Left Office
1.646*** [0.600]
0.047* [0.026]
0.223 [1.164]
0.364** [0.157]
0.016** [0.007]
0.248 [0.283]
0.378** [0.158]
0.016** [0.007]
0.236 [0.217]
Negative Exit Reason −16.691***
[3.955] −0.823***
[0.163] −15.833 [12.417]
−18.733*** [3.734]
−0.884*** [0.155]
−11.264 [14.587]
−18.659*** [3.774]
−0.889*** [0.155]
−11.051 [9.194]
∆ State Citizen Ideology (Party−Adjusted)
0.030 [0.139]
−0.002 [0.006]
0.058 [0.321]
0.007 [0.131]
−0.003 [0.005]
0.202 [0.260]
−0.011 [0.133]
−0.003 [0.005]
−0.012 [0.259]
Constant 77.251***
[8.499] 1.257***
[0.362] 91.755***
[15.754] 73.812***
[8.374] 1.133***
[0.352] 75.318***
[17.150] 79.129***
[8.312] 1.272***
[0.349] 89.732***
[11.778] Number of Observations 296 290 296 309 303 309 309 303 309
[0.000] Notes: Dependent variable is defined as the Former Governor’s Vote Share in the given election.. Standard errors are inside parentheses (Median regression SEs are heteroskedastic-consistent and robust to misspecification of the variance-covariance matrix). *** p ≤ 0.01; ** p ≤ 0.05; * p ≤ 0.10; +significant at the 0.10 level (one-tailed test).
35
Figure 2: The Impact of Intertemporal Attribution Discounting on Retrospective Economic Voting Behavior
‐8
‐7
‐6
‐5
‐4
‐3
‐2
‐1
0
1
2
3
0 1 20 40 60 80 100120140160180200213
Cond
ition
al M
argina
l Effects of
Past Econo
mic Stewardship on
Expe
cted
Electoral Vote Share
Election Delay (Months)
Panel 2A: Conditional Marginal Effects: Tenure Model
ConditionalMarginal Effects
95% C.I.: Upper
95% C.I.: Lower
‐5
‐4
‐3
‐2
‐1
0
1
2
3
0 1 20 40 60 80 100120140160180200213
Cond
ition
al M
argina
l Effects of
Past Econo
mic Stewardship on
Expe
cted
Electoral Vote Share
Election Delay (Months)
Panel 2B: Conditional Marginal Effects: Recent Term Model
ConditionalMarginal Effects
95% C.I.: Upper
95% C.I.: Lower
‐5
0
5
10
15
20
25
0 1
20
40
60
80
100
120
140
160
180
200
213
Cond
ition
al M
argina
l Effects of
Past Econo
mic Stewardship on
Expe
cted
Electoral Vote Share
Election Delay (Months)
Panel 2C: Conditional Marginal Effects: Recent Year
ConditionalMarginal Effects
95% C.I.: Upper
95% C.I.: Lower
36
Figure 3: The Consequences of Intertemporal Attribution Discounting on Expected Electoral Vote Share