Gubernatorial Stability in Iowa: A Stranglehold on Power

A Stranglehold on Power

Christopher W. Larimer

• Publisher: Palgrave Pivot
• Year: 2015
• Print ISBN:
  9781137528124
• eBook ISBN:
  9781137528131
• Topic: Politics & International Relations
• Subtopic: European Politics
• Index: Politics & International Relations

This book uses a multi-method approach to explain why recent Iowa governors have been able to stay in office significantly longer than their peers. Voters in Iowa value a personal connection with their governor and those governors who ignore that expectation are held accountable at the polls.

1

The “Personal” Power of Iowa Governors Since 1969

Abstract: This chapter provides an overview of existing research on gubernatorial popularity and how this framework applies to recent governors in Iowa. Iowa governors faced declining economic conditions, shared party affiliation with unpopular presidents and, at times, had unified control of state government. Yet, when applied to the case of Iowa, these factors, while significant, do not fully explain the strong incumbency bias observed in the state. Considerable attention is given to expanding Beyle’s notion of the “personal” power of governors as well as consideration of the strength of the challenger when examining gubernatorial popularity. This chapter concludes by asking what explains the discrepancy between observed electoral outcomes in Iowa and what existing research says should happen.

Keywords: challenger quality; gubernatorial popularity; institutional power; personal power

Larimer, Christopher W. Gubernatorial Stability in Iowa: A Stranglehold on; Power. New York: Palgrave Macmillan, 2015. DOI: 10.1057/9781137528131.0006.

This book is about the unusual staying power of Iowa governors in recent history going back to the election of Robert Ray in 1968 through the election and reelection of Terry Branstad in 2010 and 2014, a former governor of Iowa who also served in office from 1983 to 1999. As with any research question, the first step to finding an answer is theoretical. This chapter proceeds by reviewing the literature on gubernatorial popularity, with speculation as to how existing research applies to the study of Iowa governors. Along the way, I present descriptive data that addresses how various political and economic factors may have shaped the way Iowans look at their governor. By grounding this analysis in the existing literature on American governors, this chapter (and book) not only speaks to the attitudes Iowans hold toward their chief executive but also to a more general audience interested in state politics.¹

Gubernatorial popularity and reelection in Iowa

Perhaps the first question to ask is: Why should we care about gubernatorial popularity? Common sense suggests that more popular governors are likely to do better at the polls than less popular ones, and indeed this relationship has been demonstrated empirically. Patrick Kenney and Tom Rice (1983) were the first to assess whether the popularity of governors affects the proportion of votes received in the most immediate election. Despite a small number of observations (N = 19), the authors found a significant and positive relationship; when approval increases, vote share increases. The breakeven point in their analysis was 48 percent, indicating that incumbent governors with a 48 percent approval rating can expect, on average, to receive just over 50 percent of the vote. But does this relationship hold in the case of Iowa? This is an important question to consider as this book attempts to discern the importance of gubernatorial popularity in Iowa, and there is no more direct measure of popularity than approval ratings.² If approval ratings are not predictive of election outcomes, there is little point in spending time analyzing the nuances of such data.

In Figure 1.1, I plot approval ratings and percent vote share received by the incumbent for the ten elections from 1970 to 2014 in which the incumbent candidate for governor in Iowa sought reelection (open seat elections in 1982, 1998, and 2006 were not included). All ten ratings were based on the same poll, the Des Moines Register Iowa Poll, to eliminate the possibility of polling effects. Two effects are worth noting. First, there is a strong positive correlation between the last approval rating taken prior to the election and the vote share received by the incumbent in that election (r = .077).³ For these ten observations, percent approval explains nearly 60 percent of the variation in vote share received. Second, the minimum approval required for electoral victory is considerably less than 50 percent. Using the regression line equation in Figure 1.1, an approval rating of 44 percent prior to the election is the predicted minimum rating required to generate a large enough vote share (50 percent plus 1) to ensure victory (50.02 = 35.24 + 0.34 (44)). Even a governor polling at 40 percent can expect to receive just over 48 percent of the vote in the general election.

The minimum approval rating predicted to generate 50 plus percent of the vote is quite modest. So while approval ratings certainly do matter, and are statistically significant in predicting vote share, Iowa voters do not readily abandon a governor for whom they harbor less than enthusiastic feelings. Having answered the why of why do we care about gubernatorial popularity (because it predicts election outcomes),⁴ the next question to consider is: What factors affect gubernatorial popularity?

FIGURE 1.1 Gubernatorial popularity and vote share, 1970–2014

Note: Results are for gubernatorial elections held from 1970 to 2014 involving an incumbent. Open seat elections in 1982, 1998, and 2006 were not included. The source for all ten poll numbers is the Des Moines Register Iowa Poll. The x-axis represents the percent approval for the ten elections for the most recent poll prior to the upcoming gubernatorial election; the y-axis represents the percent of the vote received by the incumbent in the election.

Research on gubernatorial popularity

Presidents, like governors, are chief executives, and so naturally a question to consider is: Are governors held to the same standard as their federal counterpart, the president?

Beginning with Robert Stein’s (1990) work, scholars began collecting and analyzing the popularity of governors as measured by approval ratings. Stein was the first to systematically test whether voters were capable of distinguishing between the “functional responsibilities” of governors and presidents. Stein’s work showed that while governors were held accountable for macroeconomic conditions such as perceptions of the state’s economy, such perceptions were tied to evaluations of the sitting president. Governors of the same party as the president were evaluated more negatively when national economic conditions were perceived to be going in the wrong direction. Lonna Atkeson and Randall Partin (1995) similarly report that economic conditions do matter but find little support for the notion that evaluations of the president affect gubernatorial approval ratings; that is, voters assess the state’s economic condition independent of national politics (see also Niemi, Stanley, and Vogel 1995; Partin 1995). Tom Carsey and Gerald Wright (1998) challenged Atkeson and Partin on this latter point. Reevaluating the data used by Atkeson and Partin, Carsey and Wright demonstrated that presidential approval does matter for governors. Voters have difficulty disentangling national factors from vote choice decisions in subnational elections.

Stein’s research (and that of others) was based on exit surveys and not actual approval measures of governors. The debate among scholars has since shifted to the extent to which macro-level economic indicators, such as a state’s unemployment rate, affect the job approval ratings of governors. Robert Crew and Gregory Weihler (1996) were among the first to test this proposition using job approval ratings of governors in three states: California, Minnesota, and Iowa. The authors found no support for state unemployment rates as significant predictors of gubernatorial popularity. Rather, in line with previous work by Stein (1990) and Carsey and Wright (1998), national politics took precedence. Governors who share party affiliation with the president are rewarded and punished when the national economy sails and sinks (see also Peltzman 1987). For governors, the popularity of the president matters (see also Cohen 1983), and that popularity is intertwined with national economic indicators such as the national unemployment rate. Similar to later work by Adam Brown (2010), perceptions of governors are filtered through a partisan lens, and that lens is shaped by individual partisan identification as well as approval or disapproval of the sitting president.

While Crew and Weihler (1996) found state unemployment rates to be insignificant predictors of gubernatorial popularity, more recent work demonstrates the opposite. Drawing on job approval measures of California governors from 1967 to 1997, and governors from seven other states from 1980 to 1997, Susan Hansen (1999a) showed that state unemployment rates do in fact matter for gubernatorial popularity, independent of national political factors. As state unemployment increases, so too does the percentage “fair” or “poor” ratings of governors (see also Hansen 1999b). Importantly, Hansen found a positive relationship between year of the rating and state unemployment rate, indicating the importance of the state’s unemployment rate in job approval ratings has been growing over time.

These studies certainly demonstrate that political, and to some extent, economic factors are important, but comparing findings across studies is difficult. While some studies were based on exit poll data, others used job approval ratings from a small number of states or for a limited time period. The Job Approval Ratings (JAR) database created by Thad Beyle, Richard Niemi, and Lee Sigelman (2002), and later updated through March of 2010, includes measures of gubernatorial approval, US Senatorial approval, and state-level presidential approval across all 50 states taken from over 2,000 approval ratings from 1947 to 2010.⁵ The data are publicly available and when combined with political and economic factors commonly theorized to affect the popularity of elected officials, allow for a systematic test of the extent to which state economic conditions matter when it comes to assessing the job governors are doing. Scholars of the American governor have since used this data set quite regularly for just such purposes and established a consistent set of findings pertaining to the popularity of governors.

Using the JAR data set for the years 1980–2001, Jeffrey Cohen and James King (2004) find that both national and state unemployment rates matter, but that the latter is best considered in the context of the former. Rather than just including a monthly measure of state unemployment, Cohen and King define what they call “relative unemployment,” a measure calculated as the state unemployment rate minus the federal unemployment rate (both lagged one month). When the state’s economy improves relative to the federal economy, the approval rating of governors should increase, when it doesn’t, approval should decrease. Indeed, Cohen and King (see also King and Cohen 2005) find that both national and “relative unemployment” significantly and negatively affect the approval ratings of governors, showing a 3–4 percentage point negative effect on approval for every 1 percentage point change in national unemployment and state unemployment.

Cohen and King’s work not only clearly demonstrates that state economic conditions do matter when it comes to gubernatorial approval, but also that such effects hold independently of national political factors. King and Cohen (2005) also found significant effects for presidential approval, shared party affiliation, state political control, the tenure of governors, and even population size. As with previous work, a governor’s fate is tied to the party and popularity of the current president. Governors who share party affiliation with an unpopular president (which of course may be linked to declining economic conditions) can expect to see a drop in their approval ratings. Similarly, as discussed by other scholars, governors who have unified control of state government can actually expect a drop in approval rating (see also Nicholson, Segura, and Woods 2002; Leyden and Borelli 1995). Unified control provides the public an easy target to blame when things go bad, and, in fact, governors seem to recognize that unified control does not automatically translate into easy governing. In Alan Rosenthal’s (2013: 207–208) survey of governors, chief executives were equally positive about their experiences with divided government as with unified government. But Iowa, where variation in who holds the governor’s office is considerably less than in other states, appears to buck these trends.

“It’s the economy, stupid” . . . maybe

The aforementioned research suggests the simple answer to Chet Culver’s failed reelection bid is to blame the struggling economy, and there is considerable empirical support for such an answer. Figure 1.2 shows the approval ratings of Governors Branstad, Vilsack, and Culver alongside the state and federal unemployment rates (right y-axis) as measured by the US Bureau of Labor Statistics. The approval ratings are taken from the JAR database mentioned earlier as well as from the Des Moines Register Iowa Poll.⁶ The figures generally show the expected relationship: as unemployment rates increase, popularity decreases. The state’s unemployment rate during Culver’s tenure went from 3.7 percent in January of 2007 to 6.3 percent in October of 2010, the month before the election (it reached 6.4 percent during January, July, and August of his reelection year). The 2.6 percentage point swing contrasts sharply with Govern...

Table of contents

• Cover
• Title
• Introduction: The Year 2010 as a Watershed Moment in Iowa Politics
• 1  The Personal Power of Iowa Governors Since 1969
• 2  Gubernatorial Popularity and Power in Iowa Since 1969
• 3  Iowa Comfort and Doing the Full Grassley
• 4  Iowa Voters on Iowa Governors
• 5  The Big Wide-Mouthed Frog Theory of Iowa Politics
• Appendix
• Bibliography
• Index