Political Mandate and Clarity of Responsibility: Economic Policies under Rightist Governments in Latin America

Since the mid-1990s, some rightist governments in Latin America have adhered to a strict market orientation while others have shown less attachment to doctrinaire neoliberal policies, a puzzle as rightists are expected to favor minimal government intervention in the economy. In an environment over the past two decades in which market-oriented policies, in general, have grown increasingly unpopular with many Latin Americans, we contend that rightists have less political cover to endorse neoliberal policies. Using panel data for eighteen Latin American countries from 1995 to 2015, we find that, because of the clarity of responsibility that occurs under political mandates and the unpopularity of market reforms, mandate-holding rightist governments will tend to go against their ideological preferences and decrease neoliberal policies. Our findings indicate that as presidential vote margins increase and responsibility for unpopular economic policies becomes clearer, rightist executives will be less willing to support such policies, but only to a point. The results suggest that clarity of responsibility can influence presidential decision-making concerning unpopular policies, especially microeconomic policies, but this influence diminishes as presidents become more electorally secure.

like the left, not all rightists have followed the same policies. Although some rightist governments have adhered to a strict neoliberal model, others have shown less doctrinaire attachment to market reforms, which is puzzling since rightists have traditionally favored minimal government intervention in the economy.
Political economists have developed a litany of theories to explain economic policies under leftist governments that also appear applicable to rightist regimes. Among the theories, some studies stress party institutionalization, contending that high party institutionalization promotes interparty and interbranch cooperation, encouraging presidents to maintain status-quo, market-oriented policies (Flores-Macías 2010Levitsky and Roberts 2011). Others suggest that the boom in industries and commodities tied to production for the global market has generated greater export-oriented influence in domestic labor markets, increasing the probability for market reforms (Frieden 1991;Remmer 1998). Additionally, factor endowment research indicates that countries abundant in natural resources support interventionist policies because the state typically controls resource sectors, providing politicians with the financial means to placate popular interests (Weyland 2009;Hidalgo 2009). This article will focus on the importance of political mandates. Although previous research has shown that leaders holding mandates have greater discretion to implement their preferred policies (Biglaiser 2016;Johnson and Crisp 2003), 3 we have reason to believe that this discretion is limited, especially when many of the economic policies are not particularly popular with the general populace.
Using panel data for up to eighteen Latin American countries from 1995 until 2015, we find support for the mandate theory but in a way largely unpredicted for rightist rule. The scholarly literature suggests that rightist presidents favor neoliberalism, including a limited role for the state to address economic inequality, because their richer core constituencies benefit from market policies (Gibson 1996;Luna and Rovira Kaltwasser 2014;Middlebrook 2000, 3-4;Wiesehomeier and Benoit 2009). 4 Given their preference for market reforms and the enhanced policy discretion that comes with a mandate, practitioners of mandate theory would expect rightist leaders to back neoliberal policies. However, we argue that because of the clarity of responsibility that arises between mandate holders and the electorate, rightist presidents experience cross pressures that often make them less likely to uphold the most strict market policies as compared to governments that lack mandates. 5 With regard to cross pressures, on the one hand, the advance of leftist governments in power since the late 1990s has forced the right, for electability reasons, to moderate the most orthodox neoliberal policies introduced in the late 1980s and early 1990s. The right realizes that, with the exception of trade liberalization (Baker 2003;Baker and Greene 2011), orthodox market reforms have grown less popular with many Latin Americans over the past two decades, calling into question the "political viability of the neoliberal model" (Roberts 2014, 25). Indeed, one reason for the electoral success of left parties since the late 1990s has been their willingness to moderate or even abandon the most extreme neoliberal reforms. 6 On the other hand, rightist governments recognize the challenges of breaking with market-oriented reforms, particularly macro-level policies, that places them at odds with global policy-making circles (Luna and Rovira Kaltwasser 2014, 16;Roberts 2014, 26). Rightist leaders understand that moving completely away from neoliberalism could shut them out of foreign capital markets and increase foreign capital flight. Moreover, an influential minority of rightist constituents strongly favor market reforms. Thus, rightist governments are crosspressured, attempting to maintain market reforms to appease the global economy and a minority of its constituents while retaining enough domestic political support to win, or remain in, office.
We argue that the pressures to serve the broader electorate, and the electoral consequences of supporting less popular policies, are greater for mandate-holding rightists-who have greater clarity of responsibility, defined by unified government and presidential margin of victory-making them less prone to sustain the strictest market reforms. Unlike under divided government, where the different parties often blame each other for maintaining an orthodox neoliberal model, the parties still support the reforms because 3 Johnson and Crisp (2003) discuss mandate in the context of winning an election but without consideration of victory size or if the president and legislature come from the same party. 4 Rightist ideology differs from parties on the left, as leftists usually prefer increased government involvement in the economy (Baker and Greene 2011; Magaloni and Romero 2008;Stevenson 2001). Similarly, Johnson and Crisp (2003, 134) find that right-of-center legislatures are correlated with more market-oriented policies. 5 As presidents not winning by a landslide, and those who govern under divided government, hold less control over policy outcomes, they are less clearly responsible for said policies. 6 As Baker and Greene (2011, 44) report, "voters' declining enthusiasm for market reforms" is an important reason for leftist victories in presidential elections. they satisfy the investment community, while clarity of responsibility fostered by a mandate informs voters as to who is responsible for the unpopular neoliberal policies (Powell and Whitten 1993, 398). The large number of voters opposing strict market reforms results in mandate holders moderating doctrinaire neoliberal reforms, especially micro-level economic policies, because of global financial markets tendency to exert macroeconomic discipline in developing countries (Kaplan 2013;Mahon 1996;Mosley 2003;Wibbels 2006). Although we expect there are limits on clarity of responsibility, we contend that rightist leaders holding mandates will call greater attention to more popular measures that reflect the right party's strengths including managing public order and security, pledging support for conservative social policies, and promoting government programs serving the very poor, all as means to win elections and stay in office (Meléndez 2014;Koivumaeki 2014;Wiesehomeier and Doyle 2014). This project contributes to the literature in several ways. First, this study provides new insights for understanding economic policies under rightist administrations in Latin America, an area understudied in the literature. 7 Second, the findings complement a range of political economy research showing a schism between adherence to macro-and micro-level economic policies (Boix 1998;Murillo 2009;Kaplan 2013). Third, and most importantly, the research explains why mandate-holding executives from the right appear to go against their presumed ideological preferences, helping to solve a puzzle in the literature.

Theories on Economic Policies under Rightist Governments in Latin America
Since the late 1980s and early 1990s, nearly all Latin American countries abandoned import-substitution industrialization (ISI) and adopted orthodox neoliberal reforms. The swing to a market orientation seemed consistent with the ideology historically held by many rightist parties in Latin America. The Conservative Party in Chile under President Jorge Alessandri in 1958-1964 typified the view held by much of the Latin American right during the post-World War II era, preferring the market to state intervention. Moreover, the fact that many rightist parties took office in the late 1980s and early 1990s and played central roles in initiating neoliberal reforms, further bolstered connections between market reforms and the right (Middlebrook 2000, 46-47). Lastly, survey work conducted by Wiesehomeier and Benoit (2009) also appears to show the strong market orientation of the right. Wiesehomeier and Benoit (2009) asked experts to assess the policy positions of different political parties and party leaders (including current presidents) in eighteen Latin American countries. Experts placed parties and presidents on a 1-20 scale, with higher scores on economic issues reflecting greater backing for the market. The survey results indicated that rightist parties and their leaders in all eighteen countries scored above 10 in support of globalization. Similarly, rightist parties and their leaders favored cutting public services to reduce taxes rather than raising taxes to increase public services in sixteen of the eighteen countries. 8 Despite the many factors suggesting that rightist governments should favor a market strategy, economic policies have varied under rightist administrations. Some parties have tended to back the status quo, maintaining or even expanding on market-oriented reforms, while others have been less doctrinaire in their support for the neoliberal model. Differences among rightist governments present a puzzle in the political economy literature that has received limited scholarly attention. We propose a theory that seeks to explain the conditions in which rightest governments in Latin America would move away from their ideological preference on neoliberal economic reforms and implement less market-oriented policies. Our theory indicates that political mandates for rightest governments produce a cross-pressured environment where mandates allow them to enact their preferred policies more freely while making the responsibilities for those policies more clear. We argue that when neoliberal policies appear less popular, political mandates can actually make it harder to enact such policies due to potential blame attribution. Even in situations of improved economic performance, the fact that neoliberal reforms have tended to exacerbate economic inequality may contribute to more moderating policies that seeks greater economic redistribution. 9 7 See works by Gibson (1996), Luna andRovira Kaltwasser (2014), andMiddlebrook (2000), who investigate the politics of the right in Latin America. 8 See also Luna and Rovira Kaltwasser (2014, 4) and Eaton (2014, 78), who discuss differences between the right and left on economic policy, with rightist parties preferring a free and unfettered market and minimal state intervention. In the context of the OECD, Boix (1998) similarly argues social democrats use the public sector to support economic growth and promote equality while conservatives employ the private provision of investment to address development. 9 See also Remmer (2012, 953), who contends that a growing economy "creates opportunities for political leaders to offer voters programs of poverty reduction, improved social equity, economic nationalism, and increased government spending." The basis for our theory evolves from the literature, showing how political mandates can affect the policies that governments implement (Grossback, Peterson, and Stimson 2007;Johnson and Crisp 2003;Light 1989, 26;Mainwaring and Shugart 1997). A political mandate, as indicated by a landslide presidential victory (Conley 2001;Jones 1999) and the executive's party holding a majority of seats in the legislature (Weinbaum and Judd 1970), creates the conditions necessary for presidents to install their preferred policies (Biglaiser 2016;Mainwaring and Shugart 1997). Political mandate theory would anticipate that, because of their ideological preferences, rightist executives favor neoliberal reforms (Gibson 1996;Middlebrook 2000).
Building on research on clarity of responsibility, we present a nuanced version of what political mandate theory would predict for rightist governments and market policies. Before developing the argument, we first develop the concept of clarity of responsibility. Clarity of responsibility dates back to the work of Powell and Whitten (1993), who argued that a voter's assignment of responsibility depends largely on the coherence and control government exerts over policy. Unlike in political systems with divided government, where the executive and majority in the legislature come from different parties-or in cases where the president won office in a close race and voters may not know who to credit (or blame) for economic policies, the legislative or executive branch-a mandate suggests unified control of policymaking as the government comes from the same party (Powell and Whitten 1993, 398).
Beyond the importance of mandates for assigning responsibility, clarity of responsibility matters most during periods of economic crises, as voters can observe shocks to the macroeconomy and assign blame (Duch and Stevenson 2008). Although there has been some variation in economic performance among countries in the region (see Kaplan 2013), Latin America has experienced numerous economic crises in most countries over the past two decades. The high frequency of economic crises across much of the region increases the chance that clarity of responsibility resonates with both Latin American leaders and voters.
Complementing the work of Duch and Stevenson (2008), and Powell and Whitten (1993), we contend that clarity of responsibility affects the policies of mandate-holding rightists. The logic of the argument is that in attempting to serve constituents as well as protect the interests of the global (and domestic) financial community, leaders on the right frequently face cross pressures and will give more support to their broader constituent base by moderating neoliberal policies at the minor expense of the financial community and a small number of influential rightist backers. On the one hand, right parties need to support the interests of constituents, many of whom question the benefits of most orthodox neoliberal policies. Survey data conducted by the polling service Latinobarómetro between 2002 and 2009 indicate fairly strong public disapproval for market policies in all Latin American countries. In a question about satisfaction with the functioning of the market economy, on average, less than 27 percent of responses expressed satisfaction with the operation of the market while more than 73 percent responded not very satisfied, not at all satisfied, don't know, or no answer to the question (see Table 1). 10 Many Latin Americans also have been deeply critical of privatization, a fundamental component of neoliberal models. In a Latinobarómetro survey conducted between 1998 and 2009 asking whether privatization of state companies has been beneficial, less than 35 percent, on average, agreed or strongly agreed about the merits of privatization (see Table 2). As much as any economic reform, privatization draws intense domestic protest tied to concerns over sovereignty, unemployment, and regulation (Roberts and Arce 1998). But privatization is not the only policy that unleashes public harangues. Indeed, with the exception of trade openness, support for the neoliberal model has fallen throughout Latin America (Baker 2003; Baker and Greene 2011). 11 Ironically, previous research suggested that Latin American politicians initiated orthodox market reforms in the 1990s, often breaking promises in political campaigns to support greater state intervention, because elected officials calculated that such reforms would benefit their constituents and ultimately would serve their own political ambitions (Stokes 2001). 12 As many Latin American politicians have since discovered and polling information suggests, following a rigid neoliberal course may backfire for survival-minded leaders (Madrid 2010;Roberts 2012). Rightist politicians cannot rely on a minority of rightist constituents 10 Mahon (2003, 58) similarly reports increased hostility to extreme market solutions. He attributes much of Latin America's turn away from the "Washington Consensus" to adverse external trends (e.g., growing economic uncertainty, slowing credit flows, and the rise in protectionism by the North) and worsening social and economic problems (e.g., increases in crime, unemployment, and inequality). 11 As Baker (2003, 428; emphasis added) notes, in comparing trade with other economic reforms, "many citizens favorably single out free trade from the list of Latin America's recent market reforms." 12 See Campello (2014), who convincingly argues that left-leaning candidates have a high probability of betraying campaign promises and switching to neoliberal policies when elected during a currency crisis.
who fully embrace the most orthodox neoliberal model if they hope to succeed electorally. On the other hand, most rightist executives and their parties recognize that in an integrated global economy, as Latin American countries have operated under for more than two decades, politicians need to maintain some measure of market reforms or they will confront economic challenges from the international (and domestic) financial community (Luna and Rovira Kaltwasser 2014, 16;Roberts 2014, 26). There is, of course, liable to be variation in the degree of market openness countries will uphold, which we will discuss later, but the point is that the global financial community imposes constraints on economic policies. Most leaders realize that disregarding the interests of the global financial community and overturning neoliberal policies risks their countries losing access to desperately needed foreign capital markets. We contend that clarity of responsibility that comes with a mandate and general dissatisfaction with most market reforms increases the probability that mandate-holding rightists will moderate from their presumed preferences and the interests of the investment community and support a less doctrinaire neoliberal model. Because mandate holders from the right likely would be held accountable if they maintained or amplified orthodox market policies, such governments have tended to downplay neoliberal policies and have focused on their perceived strengths tied to public order and security, conservative social issues, and pro-poor policies, measures favorably received by most of their constituents. Although mandates provide presidents with greater discretion to institute their preferred policies, mandates also enhance clarity of responsibility between voters and elected officials that compel presidents to follow voter demands, in this case softening neoliberal policies.
However, we expect the more moderate neoliberal stance will have greater impact on micro-level economic policies as compared to macro-level policies. Based on literature for the developing world, the financial community particularly opposes changes at the macro level. Foreign investors rail against macro-level reforms that restrict trade openness, increase government deficit spending, and raise levels of inflation and price controls, as such policies have the most effect on their economic bottom line (Kaplan  Mahon 1996;Mosley 2003;Wibbels 2006). Because foreign investors likely will respond with large capital flight against governments that institute macro-level policy changes, mandate holders will tend to uphold macro-level reforms. Instead, rightist governments have leaned toward more moderate micro-level economic policies, including labor and business reforms. By instituting micro-level changes such as wage increases and higher regulations on starting new businesses, the governments can gain support from labor interests and already established businesses, which may favor restrictions that reduce competition from potential newcomers while imposing minimal damage on global finance. In situations where rightist executives do not hold a mandate, such as in minority governments, coalitions, or a bicameral opposition, we maintain that blame (or responsibility) for policies is harder to assign to presidents or the party in power. The fact that voters cannot clearly apportion responsibility for policies to non-mandate-holding governments provides politicians with increased political cover to maintain stricter market reforms. Politicians from both the left and right often have incentives to support such reforms because they placate the investment community, an important group for creating jobs and whose supporters frequently contribute to political campaigns on both sides of the aisle. 13 However, we do not believe that the moderating effect of clarity of responsibility on the enactment of neoliberal policies is unlimited. Research on clarity of responsibility asserts that as assignment of responsibility for policies becomes easier, political actors face greater electoral consequences for enacting unpopular policies (Powell and Whitten 1993). This assertion is at odds with the mandate literature, which indicates that greater electoral margins insulate political actors from electoral vulnerability and allow politicians to use more discretion (Biglaiser 2016). Our theory seeks to unify these seemly disparate theories by showing that clarity of responsibility can mitigate the discretionary power of political mandates, but only to a point. When mandates reach a size necessary to reduce electoral vulnerability, the mitigating effects of clarity of responsibility will cease. Therefore, we expect that Latin American presidents with political mandates will support more moderate neoliberal policies as clarity of responsibility increases, but as their mandates increase, and their positions become less vulnerable, they will revert to their support of more orthodox neoliberal economic policies. Rightest governments thus enact greater neoliberal market reforms, when such reforms are fairly unpopular, under two conditions: when the election is close enough to cloud the clarity of responsibility, and when the margin of victory is so great that the executives are seemingly impervious to political challenges. Based on work related to political mandates and clarity of responsibility, we propose the following hypotheses: H 1 : Increases in presidential margins of victory for rightest executives will decrease the likelihood of implementation of strict market-oriented policies. H 2 : The effect of margin of victory on the likelihood of implementation of strict market-oriented policies will be curvilinear and eventually become positive. H 3 : Unified government for rightist executives will decrease the likelihood of implementation of strict market-oriented policies. H 4 : Presidential mandates and clarity of responsibility are more likely to affect micro-level economic policies rather than macro-level policies.

Research Design and Methods
We collected annual data for up to eighteen presidential democracies in Latin America from 1995-2015 (see the online supplementary Appendix for the list of countries and years of the administrations). We code the country's governing parties as either right/right-center, center, left-center/left, using a consensus method that compares ideological placements from multiple sources. We combine ideological placements from World Bank's Database of Political Institutions (Beck et al. 2001) 14 as well as placements from Doyle (2012), Pop-Eleches (2009), and Baker and Greene (2011). We use the four independent measures developed by the authors to arrive at a consensus on the ideology of the government. For each observation there was a placement reached by a majority of the sources, which provides us confidence that the regimes are appropriately classified. For the purposes of establishing the effects of mandates on rightists' economic policies, we run separate sets of models for only rightist executives, rightist and centrist executives, and finally a model which includes the entire ideological spectrum.
The primary dependent variable for this analysis is the Heritage Foundation's Index of Economic Freedom. The index includes ten economic freedoms most associated with market reforms: property rights protection, freedom from corruption, fiscal freedom, labor freedom, government spending, business freedom, monetary freedom, trade freedom, investment freedom, and financial freedom. The index uses sources from multiple organizations including the Economist Intelligence Unit, the International Monetary Fund (IMF), the World Bank, and the US Department of State to determine the value for each freedom. 15 The index has been called on by many previous studies (Barro 2013;Beck, Demirgüç-Kunt, and Levine 2006;Lee, Biglaiser, and Staats 2014;Roberts 2014, 41). 16 We also test our hypotheses on each of the ten component measures of the index to determine if the effects of mandates are uniform across the measures. Each economic freedom is graded on a scale from 0 to 100, with higher numbers indicating a greater market orientation. A country's score is a weighted average of the ten economic freedom indexes.
The primary independent variables, comprising previous operationalizations of political mandates, are unified government, and presidential margin of victory. Executives possess unified governments when their parties hold absolute majorities in their houses that have lawmaking powers. In the case of a bicameral 14 Party orientation with respect to economic policy, coded based on the description of the party in the sources, using the following criteria: Right: for parties that are defined as conservative, Christian democratic, or right-wing; Left: for parties that are defined as communist, socialist, social democratic, or left-wing; Center: for parties that are defined as centrist or when party position can best be described as centrist. 15 See the Index of Economic Freedom website for more information about the methodology used for each freedom measure: http://www.heritage.org/index/book/methodology. 16 As Flores-Macías (2012, 26) notes, there are limitations with the index, as some indicators are fully or partially qualitative and depend on the judgment calls of experts. Despite the limitations, the index provides the most comprehensive measure for market openness.
legislature, executive parties must have a majority in both houses. 17 We measure presidential margin of victory as the difference in the vote percentage received between the current president and the closest challenger in the most recent election. We coded presidential margin from the ultimate round, as some elections required a second round; in these cases, a dummy variable for second round is included. Presidential and legislative electoral data come from Adam-Carr.net and the Parline database on national parliaments (http://www .ipu.org/parline-e/parlinesearch.asp). Since a presidential mandate comprises both of these variables, we employ an interaction term between presidential margin of victory and unified government. Additionally, since we contend that the effects of a mandate are curvilinear, we include a squared presidential margin of victory term. Latin America has seen left parties become increasingly popular, with almost all countries having elected a leftist president in the last decade (Biglaiser 2016). With left parties becoming more popular, there is some reason to believe that parties on the far right may be receiving smaller electoral margins, and therefore our mandate measure could be picking up unobserved ideology effects, which explain policy outcomes. We thus include a measure of ideology, used in previous research on mandates (Baker and Greene 2011), which is a score on a continuous 0-20 scale with 0 coded as extreme left and 20 being extreme right. 18 We also control for alternative explanations that could account for strict market-oriented policies. Many studies have investigated the effect of party institutionalization on differences in policies, maintaining that countries with institutionalized party systems are more likely to retain market status-quo policies (Flores-Macías 2010Mainwaring 2006;Kaufman 2011). We control for party institutionalization by including electoral volatility, a common proxy for party institutionalization, because low electoral volatility and high party institutionalization tend to go together (Flores-Macías 2010, 2012). We utilize electoral volatility developed by Pedersen (1979), which is exactly one-half of the net change in the percentage of votes won by parties from the previous legislative election. We also include a measure of presidential duration to test whether the length of tenure has an effect on enacting market-oriented policies.
The boom in industries and commodities tied to production for the global market, along with the collapse of most firms geared toward ISI since the late 1980s, has generated greater export-oriented influence in domestic labor markets (Murillo 2000;Wibbels and Arce 2003). Because of their increasing influence, export-oriented interests are successful in lobbying politicians, who, in turn, support more extensive, market-oriented reforms (Frieden 1991;Remmer 1998). To address the growth of trade-oriented industries in influencing economic policies, we control for the export sector using a measure of exports as a percentage of GDP.
Other researchers contend that country natural resource endowments influence policies. Specifically, countries that derive much of their revenues from natural resources favor statist policies, as the abundant proceeds earned from resource exports, often under the control of the public sector, give political leaders added financial flexibility to benefit larger segments of the populace (Weyland 2009;Hidalgo 2009). For testing the natural resource endowment argument, we add annual ores and metals exports (as a percent of merchandise exports) to fuel exports (as a percent of merchandise exports), a measure used in the political economy literature (Jensen 2003). We also include economic controls for annual GDP growth and logged GDP per capita (2005 U.S. dollars), which address possible economic situations and crises experienced by individual countries. 19 All economic data come from the World Bank (2015).
International factors including international organizations and trends also potentially affect economic policy. Researchers have found that when countries enter into agreements with the IMF, as many Latin American governments have, they are frequently required to enact policies that are contrary to their interests and are often detrimental to economic growth (Dreher 2006;Stiglitz 2002;Vreeland 2003). Indeed, as we note in our theoretical discussion of political mandates, financial markets tend to impose pressures on developing countries to support more open markets. We control for IMF conditionality with a dummy variable for countries currently under an IMF extended fund facility or stand-by agreement, as these arrangements come with strict loan conditions (Biglaiser and DeRouen 2010). Additionally, global capital mobility can affect neoliberal economic policies. Mosley (2003) finds that, in the developing world, financial markets exert a considerable influence on government policy. To control for global capital mobility, we include the Chinn-Ito Index (2008), which is a measure of financial capital openness. This measure has been used in previous research on Latin America as a proxy for global capital mobility (De la Torre, Ize, and Schmukler 2011; Huber and Stephens 2012). We also include decade dummy variables to ensure that our findings are not a result of decade differences. All independent variables are lagged a year.
For the analysis, we employ a panel data approach that accounts for the unbalanced nature of our time series. Specifically, we use a fixed effects generalized least squares (GLS) procedure employing weighted least squares that can handle a wide range of unequally spaced panel data patterns (Baltagi and Wu 1999). This approach allows for the estimation of unequally spaced panel data and provides natural estimates of the serial correlation and variance components parameter, while still accounting for the AR(1) disturbances. The Wooldridge test for unit-specific first order autoregressive AR(1) process in the panel data calculated a probability of F > 0.000, which rejects the null hypothesis of no first-order autocorrelation. Additionally, since we are examining the effects of within-country change in neoliberal economic policies, we include country-and year-specific fixed effects. The Hausman specification test indicated that the model was best fit by a fixed effects structure with a probability over the chi 2 distribution > 0.000.

Results
We begin to present the results of our analysis in Table 3. Models 1-3 examine the effects of mandates and clarity of responsibility on right and center-right parties, Models 4-6 include center parties, and Model 7 examines the entire ideological spectrum. Looking at Models 1 and 2, we find that our mandate variables have a statistically significant effect at the p < 0.01 level, with unified government having a positive relationship, and presidential margins having a negative relationship. This indicates that countries with unified government are more likely, and countries with greater presidential margins less likely, to promote the most strict market-oriented reforms. This finding offers support for Hypothesis 1, but not Hypothesis 2, while the statistically significant relationship found for the squared term indicates support for Hypothesis 2. Model 2, which tests the interactive relationship between our mandate variables, provides clearer evidence of our expected relationship. Model 2 supports our expectations that unified government and presidential margins are conditional on each other, with the effect of each being affected by the value of the other.
Due to the difficulty of interpreting interactive relationships from regression coefficients, Figure 1 presents the findings from Model 2 graphically. Figure 1 indicates, consistent with Hypotheses 1 and 2, that the effects of presidential margins of victory are negative and curvilinear. Initial increases in the presidential margin of victory, which produces greater clarity of responsibility, results in an expected decrease in the likelihood that right presidents will promote the strictest market-oriented reforms. However, this effect is not continuous across the entire spectrum of margin of victory. Figure 1 indicates that once the presidential margin of victory surpasses about 25 percent, increasing presidential margins produce greater likelihood of the implementation of the strictest market-oriented reforms. This finding is consistent with our theory since clarity of responsibility will reduce the likelihood of implementation of unpopular policies insofar as presidents face electoral vulnerability. Once a president's margin of victory becomes large enough to produce a safe electoral position, the negative effects of clarity of responsibility on the implementation of preferred policy should dissipate, which is what we find. Inconsistent with Hypothesis 3, Model 3 indicates that presidents with unified government are actually more likely to promote the strictest market-oriented reforms. However, the initial negative effects of presidential margin of victory is stronger for presidents with unified government. This finding indicates that right presidents with unified governments are actually more vulnerable to clarity of responsibility, though this effect also dissipates once these presidents are less vulnerable to electoral challenge. Together this indicates that right presidents with unified government are more vulnerable to clarity of responsibility, however, they are also more likely to implement strict market-oriented policies.
Model 3 presents the results from our analysis where presidential mandate is interacted with the president's ideology score on a left-to-right-spectrum. Figure 2 presents the results graphically and indicates that extreme-right presidents, with a predicted ideological score of 18, are less likely to implement marketoriented reforms as presidential margin of victory increases, indicating that they are more vulnerable to clarity of responsibility than center-right presidents are. Center-right presidents, with a predicted ideological score of 12, increase the likelihood of implementing market-oriented reforms, which is more consistent with the previous research on mandates (Johnson and Crisp 2003;Biglaiser 2016). Additionally, we find that the presidential margin necessary for increasing the likelihood of producing greater market-oriented reforms Dependent variable: Heritage Foundation's Index of Economic Freedom (Overall Score). All independent variables lagged 1 year. All regressions include country and year fixed effects.
is greater for extreme-right presidents (31 percent), than it is for center-right presidents (9 percent). The results bolster our argument that clarity of responsibility carries the most weight with politicians farthest to the right, who ideologically should be most committed to neoliberal policies, but who are more vulnerable electorally to upholding market reforms. Looking now at the controls, we find that electoral volatility has a positive and significant expected effect on the implementation of market-oriented reforms, which may indicate that political volatility could be making clarity of responsibility harder. When support for political parties change quickly, and increasing segments of the population move support from one party to another, correctly identifying which party  is responsible for the enactment of unpopular policies becomes more difficult. Voters find it increasingly difficult to indicate which party is responsible for unpopular policies after only a few years (Johnson and Ryu 2010); this makes clarity of responsibility more challenging, and perceptive politicians know this. We also find support for the natural resource endowment argument. The negative and statistically significant relationship for natural resource endowment indicates that countries with greater natural resource wealth are less likely to implement market-oriented reforms.
Models 4-6 indicate that the effects found for right presidents also hold once center party governments are included. Presidential margin has a negative and significant relationship with the likelihood that marketoriented policies are enacted, though that relationship is again curvilinear and becomes positive after a significant interval, which provides further support to Hypotheses 1 and 2. Unified government produces a greater likelihood of enacting market-oriented policies, but the negative effect of increasing presidential margins is greater for unified government, which indicates split support for Hypothesis 3. Additionally, we test our hypotheses across the full ideological spectrum and find similar results for right parties. The effect of presidential margin of victory is initially negative but curvilinear, and unified governments are slightly more likely to implement market-oriented policies, though the results are less robust than the model for only right parties. A graphical representation of the findings, similar to Figure 1, is presented in Appendix Figure A2. Table 4 presents the results for the models, which examine the effects of mandates and clarity of responsibility, on the component parts of the Heritage Foundation's Index of Economic Freedom. We generally find support for our theory and Hypothesis 4 concerning presidential mandates and clarity of responsibility on micro-level economic policies rather than macro-level policies. We find that rightist leaders holding mandates tend to moderate micro-level policies including business freedom (e.g., starting and closing a business), labor freedom (e.g., wages and labor rigidities), property rights (e.g., laws enforced by the state that guarantee the ability of individuals to acquire private property), freedom from corruption, financial freedom (e.g., government regulation of financial services), fiscal freedom (e.g., individual and corporate tax policies), and investment freedoms, while most macro-level policies do not have a statistically significant relationship. Moreover, as shown in Figure 3, macro-level policies including government spending and monetary freedom (e.g., inflation and price controls) have nearly straight lines, and for trade freedom (e.g., tariff rates and non-tariff barriers) the slopes of the curves are fairly shallow. In contrast, with the exception of financial freedom, the lines are highly curvilinear for micro-level policies. The statistically significant relationships found from the models of the constituent indices and the graphs are largely consistent with our theory that presidential mandates and clarity of responsibility have important effects on moderating neoliberal policies, and particularly for micro-level economic policies.
A brief review of anecdotal cases from Colombia and El Salvador also suggest that rightist leaders holding mandates tend to respond to voter preferences, downplaying doctrinaire neoliberal policies and promoting initiatives that have broader constituent support, including maintaining public order and security, and instituting targeted spending programs to help the poor (Wiesehomeier and Doyle 2014;Koivumaeki 2014;Meléndez 2014). 20 The experience of Colombian president Álvaro Uribe shows how a mandate affects policies under rightist rule. For more than forty years, Colombia has faced security issues and armed conflict with guerrillas, particularly with the Marxist-Leninist Revolutionary Armed Forces of Colombia (FARC). In 2002, Uribe presented himself as a political outsider who would address problems that the two long-standing political parties in Colombia, the Liberal Party and Conservative Party, could not seem to solve, namely, to root out and end armed conflict with the FARC (Wills-Otero 2014, 199). Security concerns took precedence in both of Uribe's terms (2002)(2003)(2004)(2005)(2006)(2006)(2007)(2008)(2009)(2010). Indeed, the budget for the military and police force increased from 2 percent of GDP in 1990to 6.35 percent in 2008(Wills-Otero 2014. A milder form of neoliberalism also emerged under Uribe. Although Uribe encouraged foreign direct investment and private ownership and negotiated free trade agreements, part and parcel of the neoliberal model, he recognized that not all Colombians favored market policies. 21 Uribe responded by backing government-financed social development policies in education, health, and public services as part of a strategy to encourage more 20 Pledging conservative stances on social issues also is a common course followed by rightist leaders (e.g., Argentina, Chile, Ecuador, and Mexico; see Eaton 2014, 83;Morresi and Vommaro 2014, 328) -Otero 2014, 206). Thus, Uribe enhanced security, promoted more equitable income distribution, and implemented more targeted spending programs for the poor, all apparently as measures for helping to achieve his electoral goals. A similar situation occurred in El Salvador, where rightist leaders promoted security issues and implemented social policies to galvanize support from the masses. Prior to 2009, the rightwing Nationalist Republican Alliance (ARENA) Party had won every postwar presidential election and "dominated the legislative assembly, either itself or with its rightist allies" (Koivumaeki 2014, 268). Part of ARENA's success stemmed from selling itself as a capable defender of public security, an important concern as El Salvador had suffered a long and bloody civil war (1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992). Related to the wartime cleavage, ARENA attempted to link its main party rival, the Farabundo Martí National Liberation Front, with its violent past. ARENA's success also arose from the government's softening of its neoliberal image. Although over the decades, ARENA governments had implemented market-oriented policies, a backlash formed against neoliberalism, particularly in the early 2000s. 22 President Antonio Saca (2004)(2005)(2006)(2007)(2008)(2009) responded to extensive criticism, much of it coming from poor voters, a key constituency of ARENA, by initiating Plan Oportunidades. Oportunidades offered conditional cash transfer and microcredits to the poorest Salvadorans. As of 2008, Oportunidades served seventy-seven municipalities, with the aim of lowering extreme poverty (Koivumaeki 2014, 276, 287). ARENA cast its electoral fortunes not on adhering to the 22 Table 1   most orthodox form of neoliberalism but rather with highlighting its security competence and offering government programs for the poor. 23 A caveat is needed for the clarity of responsibility and political mandate theory developed here. The implication of the argument is not that rightist governments without mandates draw attention to neoliberal policies at the expense of security and social issues. Rightist governments of all stripes will support policies that give themselves the best chance of winning elections. The point here is that rightist governments holding mandates are more likely to downplay orthodox market reforms and highlight popular issues consistent with core conservative themes because voters can clearly assign responsibility under such institutional arrangements.

Conclusion
Although many works have investigated the variation in economic policies under leftist governments, policy differences among rightist governments in Latin America have received minimal attention. The different economic policies of rightist governments, which arguably should prefer the same policies based on their presumed ideological preferences, present a puzzle in the political economy literature.
Building on the political mandate and clarity of responsibility argument, this research helps to understand why rightist executives holding political mandates appear to go against their presumed preferences for the neoliberal model. We argued that because of the growing opposition to most orthodox market reforms, rightists holding mandates often downplay neoliberalism, particularly with regard to micro-level economic policies, and instead rely on public stances to maintain order and security, uphold conservative positions on social issues, and increase government programs for the poor to win elections and stay in office.
There are limitations to this study. First, the years available for the dependent variable hamper efforts to extend the analysis prior to 1995. Similarly, the relatively small sample of countries and years available for rightist governments in Latin America also limit empirical investigation. Despite the limitations, the work holds important implications on policy. First, the fact that relatively few empirical works investigate policies under rightist governments in Latin America provides an opportunity to make an initial step into understanding economic policies beyond leftist governments. Second, the research also builds on earlier studies in American and Latin American politics primarily outside of international political economy, showing the benefits of borrowing theories developed in other research areas to broaden our ideas about policies. Third, the research explains why some rightist governments support policies that seem to challenge their expected ideological preferences, helping to solve a puzzle in the literature.
The work presented here offers opportunities for further assessment of the mandate and clarity of responsibility theory. Previous work has investigated the effect of mandates on leftist governments in Latin America (Biglaiser 2016). Future studies could investigate the effects of mandate and clarity of responsibility on left and right governments throughout the developing world. Is Latin America an exception? Are the factors that affect policies under left and right governments in Latin America relevant elsewhere? Additionally, it might be useful to test the theories not only in presidential systems but also in parliamentary governments. Do parliamentary governments with dominance by one party affect economic policies relative to narrowly won elections or where the winning parties need to form a coalition? Future empirical studies also could explore the trade-off between policies that support security and social conservative values, and maintaining hands-off government policies such as neoliberalism. The takeaway from the results is that clarity of responsibility produced by political mandates potentially gives voters more voice in economic policies. 23 The decision by Mexican President Ernesto Zedillo, a US-trained economist, to expropriate a few foreign firms between 1995 and 1998 (Hajzler 2012), at the same time as growing unpopularity with neoliberal policies (e.g., a 1995 Latinobarómetro poll showed that more than 77 percent of the respondents believed that the country was in a bad or very bad economic situation and only 13 percent responded that they expected the situation to get better), and during mid-term elections, also is an example of clarity of responsibility and political mandate at work.