Filling the Demand Gap: The Success of Centrist Entrepreneurial Populism in the Czech Republic

Abstract

This article argues that a new type of populism is emerging that combines entrepreneurial populism with centrist populism and takes as an example the success of the Czech party ANO and its leader Andrej Babiš. Entrepreneurial populist leaders are businesspeople who claim that because they have been successful in running a business, they can run a government efficiently, like a business. Entrepreneurial populists lack a coherent ideology and instead take advantage of where the political opening is greatest in the political spectrum. In the Czech Republic in the 2010s the opening was greatest in the centre, as much of the population had social liberal attitudes, although there were no social liberal parties to represent them. To test these hypotheses, the article presents a series of regression models explaining party vote choice in the 2013 and 2017 Czech elections. The findings confirm a demand for a centrist social liberal party that did not exist at that time.

Emerging political grievances and desire for political change can also be channelled by credible change agents such as wealthy businesspeople who are able to convert their economic capital into political capital (Heinisch & Saxonberg 2021). They appeal to voters by claiming that they can do for the country what they have done for their business. A related claim is that, as businesspeople and thus non-politicians, they stand apart from the political corruption that has allegedly infected the political system. These economic entrepreneurs do not differ from other populists in that they present society as divided between a nefarious or compromised elite and ‘the people’ whose interests have been betrayed. As we will show in our theory section, the role of the successful entrepreneur is intrinsic to the political capital of these populists. This role provides them with the resources to run for offices, the status that elevates them above other new entrants into politics, and also a compelling narrative of having built a business in the ‘real world’ while typical politicians merely talk.

Their considerable resources allow these entrepreneur populists to build their own parties and concentrate power in their own hands. As a result, they are not beholden to a particular position on the political spectrum but can choose to connect their populism with different host ideologies. Some entrepreneur change agents, such as Donald Trump, propagate radical ideas and focus on nativism in order to appeal to voters under the pressure of economic, social and political change (Minkenberg 2000; Betz 2002; McGowan 2002; Piero 2003; Demertzis 2006). Here the drivers are often resentment, fear of decline and a sense of marginalisation that lead to alienation from the political mainstream. However, entrepreneurial populists can find other ways to appeal to voters, if voters are not marginalised losers of globalisation but rather, for the most part, centrist in their socioeconomic and sociocultural orientations and have democratic values. These voters are still open to the populist core message that corrupt elites ignore ordinary people but may otherwise not subscribe to radical right or left positions.

From this point of departure, we attempt to explain the electoral success of the Czech billionaire Andrej Babiš and his party ANO 2011 (Action of Dissatisfied Citizens, Akce nespokojených občanů 2011—ANO) in terms of his appeal as an entrepreneurial populist who filled the demand for a change agent, one who appeared to be an apolitical outsider and offered a centrist policy agenda. We argue that voters with centrist, culturally and socially liberal orientations who support generous social benefits but distrust the state to carry out social services will indeed support a centrist entrepreneurial populist leader if they favour change from an undesirable status quo and if political elites have not supplied a viable social liberal alternative.

This article helps fill several gaps in the literature. First, it shows that entrepreneurial populist parties are likely to arise when the supply of political parties does not meet voters’ demand in terms of value orientations, and that voters of populist parties do not always hold populist or radical attitudes. This challenges research on voting in general and populism in particular. Second, it shows why entrepreneurial populist leaders might choose different positions on the left–right spectrum, depending on the opportunity structure (Kitschelt 1986), which helps us better understand why we see such diversity among entrepreneurial populist parties. Third, we present the case of a successful populist party in power, ANO, that had to adapt from radical change agency to accepting stewardship of national politics, which populists often struggle to do (Heinisch 2003). Last, this argument places the Czech case in a larger international context in which entrepreneurial populists have succeeded in elections (for an overview see Heinisch & Saxonberg 2021).

Our article proceeds as follows. We first present our theoretical argument in detail, showing where our approach departs from the existing literature. We then explain our methodology and operationalisation. Following a brief description of our case, we present our models and describe our findings.

Theoretical discussion

Conceptualising entrepreneurial populism

We follow the ideational approach (Hawkins 2009; Mudde 2017; Hawkins et al. 2018), which defines populism as a ‘thin-centred ideology’. However, the ideationalist approach generally assumes that populists consider ‘society to be ultimately separated into two homogeneous and antagonistic groups, “the pure people” and “the corrupt elite”, and argues that politics should be an expression of the volonté générale (general will) of the people’ (Mudde 2004, p. 543).

Despite these core beliefs, entrepreneurial populists are less likely to embrace the Manichean and conspiratorial aspects of populism and do not necessarily mobilise along the strongly morally connoted dimensions juxtaposing sinister elites and exclusivist conceptions of ‘the people’. Although they also engage in ‘de-differentiation’ (Schedler 1996, p. 2) by claiming that all the other parties are the same and that ordinary people constitute a homogenous group whose interests are poorly served, entrepreneurial populists emphasise that since they were successful in business, then they can run the country like a firm. Entrepreneurial populist parties (EPPs) are centred around a charismatic leader who portrays themselves as a successful businessperson and non-politician. Given this hierarchical orientation, the leader emphasises efficiency rather than direct democracy and popular sovereignty: they argue that the will of the people is that the country needs to be run more efficiently, like a business.

Regardless of what additional political orientation they choose, as we argue, entrepreneurial populists have all been business leaders, economic entrepreneurs, turned populists (Heinisch & Saxonberg 2021). They are united in presenting themselves as outsiders ‘taking on the political establishment, making government more efficient, less regulated, and more business-like’ (Buštíková & Guasti 2019, p. 304). They have been able to convert their economic capital into political capital and are often helped by their ability to buy much of the national media in order to get positive press coverage; for example, Silvio Berlusconi in Italy and Babiš in the Czech Republic. They present themselves as ‘self-made’ (Dahl 1961, p. 25), having the social status and financial resources to command special political attention.

Centrist populism and the entrepreneurial populist

Given the conceptually thin centre of populist core beliefs, populism tends to connect itself with ‘thick’ host ideologies that supply diagnostic and prognostic narratives as well as political antagonists against whom to mobilise. Typically, these ideologies can be radically right-wing, emphasising nativism, xenophobia and right-wing authoritarianism (Mudde 2007; Rydgren 2007, 2018), or radically left-wing, focusing on capitalist exploitation (March & Mudde 2005; March 2007, 2017; Stavrakakis & Katsambekis 2014; Bonikowski et al. 2019). However, EPPs often eschew such ‘thick’ ideological linkages in order to retain their thin core and be able to position themselves flexibly along the political spectrum. Above all, populism seeks to maximise its appeal to desirable voter groups while being unencumbered by ideological principles. Taggart (2004, p. 4) refers to this as the chameleonic nature of populism: if a window of opportunity emerges in the centre, populists would arguably position themselves there. This is where entrepreneurial populists have a specific advantage over other forms of populism wedded to thick ideologies because they control their parties and are not beholden to intraparty impediments such as a contravening host ideology, grassroots activists or intraparty factionalism. We are not suggesting that entrepreneurial populism is by definition centrist, but that entrepreneurial populists flexibly position their party where it is electorally most beneficial. In the US, Trump found an opening in the radical right after conservative Republicans had lost the previous two elections and the popular vote in three of the four last presidential elections. Meanwhile, in Italy, Silvio Berlusconi successfully filled the vacuum that emerged when the once-dominant centre-right Christian Democrats imploded, while in Thailand, Thaksin Shinawatra found an opening in the left and became prime minister in 2001 before he was eventually overthrown in a military coup and his Thai Rak Thai Party was outlawed (Mizuno & Phongpaichit 2009; Sakwa 2016).

Centrist forms of populism have received less attention because the literature focuses on the radical right and, to a lesser extent, on radical left populism. Učeň provides a useful description:

Centrist populism defines itself as an alternative to traditional mainstream parties (and their blocs) without inclining towards extremist policies. … It is not an anti-system extreme, as it does not position itself on the periphery of the system. (Učeň 2004, p. 47)

Stanley discusses the emergence of centrist populism in Central and Eastern Europe, where ‘attacking the elites from a distinct thin-ideological position restricted the electoral potential of new parties’ (Stanley 2017, p. 149). Engler et al. (2019) locate the emergence of centrist populism as a thin ideology in Central Europe, specifically mentioning ANO and explain how it differs from radical populist discourses. According to Pop-Eleches (2010), such parties are non-ideological and anti-political and emphasise the need to increase living standards. As Havlík and Voda note, ‘instead of a coherent ideology, the appeal of CPPs (centrist populist parties) is often based on the promised competence of the leader, or on the claim to increase the participation of the people in the policy-making process’ (Havlík & Voda 2018, p. 163).

In our conceptualisation, we incorporate Mudde’s (2007) view of populist formations as ‘catch-all’ parties lacking a strong or ‘thick’ political ideology. This makes them flexible on economic issues. In contrast to radical right populists, centrist populist do not tend to target cultural ‘outsiders’ (Taguieff 1995; Albertazzi & McDonnell 2008) and thus are less likely to employ ‘welfare chauvinism’ in which ‘deserving’ insiders are entitled to benefits as opposed to ‘non-deserving’ outsiders (van Oorschot 2006; van der Waal et al. 2010; De Koster et al. 2013). Instead, centrist populists advocate maintaining the existing welfare state but differ from their leftist counterparts in that they reject the latter’s emphasis on national protectionism against European integration and globalisation (Kriesi 2014, p. 370).

Entrepreneur-led centrist populist parties are thus characterised by successful businesspeople who draw on their personal and business wealth to advance their political ambitions. They run their parties like businesses and claim to be agents of change. Like other populists, entrepreneurial populists take aim at unaccountable elites and pursue a catch-all strategy, making ambivalent claims and offering popular but vague proposals on a number of salient policy issues. They blame corruption and the incompetence of the political elites for the nation’s problems and adjust to the political marketplace. They are also flexible and adopt more extreme positions if it is advantageous. However, if the political window of opportunity is in the political centre, they are better equipped to occupy a centrist position than are other populists wedded to radical host ideologies. These entrepreneurs all use populism to delegitimise opponents, but they may differ in their intentions and specific messages.

Conceptualising ANO as centrist entrepreneurial populist

There is little disagreement that ANO is a populist party that does not fit a radical right or radical left characterisation (Havlík & Voda 2018). The PopuList 2.0 classifies it as populist and non-radical (Rooduijn et al. 2019). We define it as centrist because this party places itself in the centre of the party system. Its essence is distancing the party from (the sins of) mainstream politics; the central position is, therefore, very suited to this type of distancing. It is not an anti-system extremist party, as it does not position itself on the periphery of the system.

Some scholars (Buštíková & Guasti 2019) have labelled ANO ‘technocratic populist’ because of its apolitical nature. However, we disagree with this label because the core of entrepreneurial populism is not the technocratic notion that the party will form a government of technical experts but, rather, it will form a government of pragmatic businesspeople who know how to get things done. Thus, although Engler et al. label Babiš ‘technocratic’, they describe him in terms of entrepreneurship: ‘In the 2013 campaign Babiš argued that he is not a politician, but a manager, and that in order to fix the functioning of the state, the Czech Republic needs someone who could run it just like a company’ (Engler et al. 2019, p. 1325). As summarised by Maškarinec and Bláha (2014, p. 708), ANO legitimised its leadership by choosing candidates from the private sector.

In summary, what sets entrepreneurial populist leaders apart from other types of populist leaders are four important characteristics: their emphasis on their ability to rule the country efficiently, like a business, which prevents them from supporting notions of direct democracy; their ability to turn economic capital into political capital (including often the ability to buy much of the media); their centralised rule and nearly complete control over their party as the boss of the firm; and their flexibility in not being bound by any host ideology, which allows them to look for political openings in the political space where supply does not equal demand. The centrist self-placement of the ideology differs from the rightist forms of entrepreneurial populism by a commitment to political centrism, preservation of a generous welfare state and a tendency to show relative moderation in its discourse on race, culture and ethnicity. In contrast to left-wing entrepreneurial populists, centrist populists emphasise the alleged benefits of private enterprise and the need to ensure the smooth running of the market and oppose tax increases. We would therefore expect this to be reflected in the preferences of their voters.

The voters’ perspective

Entrepreneurial populists are well-placed to attract centrist voters in that they claim to be pragmatic problem-solvers who know how to get things done (Hopkin & Paolucci 1999; Krouwel 2006; Brunnerová 2019). First and foremost, their basic appeal is their seemingly non-ideological notion that they know how to run a business and thus will know how to run the country. By presenting themselves as non-politicians, their approach promises new, ‘no-nonsense’ solutions beyond the traditional ideological divide. If our claim that supply does not always meet demand is correct, then entrepreneurial populists will position themselves depending on where they see the window of opportunity, and if the opening is in the centre, they will move to fill that space.

The fact that entrepreneurial populists are established business leaders makes them appear less dangerous and unpredictable than other types of populist leaders. At the same time, their high profile and stature as national figures make them credible change agents. The resources that a well-connected and wealthy entrepreneur can bring to bear to achieve political goals are considerable and thus potentially larger than those of other change agents. Viewed from a centrist perspective, an entrepreneurial populist is the least risky choice of radical change agent while presenting a clear departure from the political status quo. Thus, voters see them as responsible radical change agents. At the very least, business tycoons will not pursue economically risky changes because they must protect their own business.¹

The third significant appeal of entrepreneurial populists for centrist voters is that of policy. In contrast to market liberals, they can offer increased support for social welfare programmes by claiming they can afford to do without increasing taxes because they will make the government run more efficiently. Thus, their stance on such issues is flexible and depends on where they find a political opening.

Below we argue that, in the Czech Republic, a large portion of the population hold centrist, social liberal socioeconomic views. In contrast to market liberals, they want generous social benefits, but in contrast to social democrats, they do not trust the state to provide welfare services. In a situation where the supply of political parties did not meet the demand for a centrist, social liberal party, an opening occurred in the early–mid 2010s for a centrist entrepreneurial populist party to fill the gap.

The Czech case: when supply does not meet demand

A central argument is that entrepreneurial populist parties are not bound by host ideologies, so they are free to look for political openings where supply does not meet demand. In the Czech case, this is in the social liberal centre. Survey data show that a large portion of the Czech population have constantly supported generous welfare benefits and held social liberal values, yet no social liberal parties have represented them in parliament. After the fall of the communist regime until 1998, the dominant Czech party was the Civic Democracy Party (Občanská demokratická strana—ODS), which presents itself as a liberal conservative party; however, some have argued that one of the reasons for the ODS’s early success was that despite the Thatcherite rhetoric of its leader, Václav Klaus, the government actually followed social liberal welfare policies (Orennstein 1995). Klaus’s decision later to follow more radical market-oriented policies, combined with the emergence of corruption scandals, led to his downfall (Saxonberg 1999).

Until ANO was elected to parliament in 2013, a pattern emerged in which voters elected parties that they perceived as being centrist, but which were actually rather right-wing. As such, these parties quickly faded when voters realised their false interpretation of the parties. The Civic Democracy Alliance (Občanská demokratická alianca—ODA) and the Freedom Union (Unie svobody) both dropped out of parliament after two terms, while TOP09 suffered a radical decrease in support when voters started to perceive them as being right-wing rather than centrist (Saxonberg & Sirovátka 2014; Heinisch & Saxonberg 2021). Since the population is mostly centrist, with a social liberal orientation, the average centrist voter saw these parties as being more centrist and social liberal than the ODS because they wanted to have such a party—it was a case of wishful thinking. Some scholars have referred to these parties as the ‘liberal centre’ and pointed to the significant electoral alternation among these parties in entering parliament (Hanley 2010; Haughton et al. 2011; Haughton & Deegan-Krause 2015; Maškarinec 2019). However, it turned out that these parties, in reality, were rather anti-ODS versions of the ODS. We call them ‘anti-ODS’ rather than ODS rivals because although these parties differed from the ODS in some areas, such as claiming to support civil society (Hanley 2012), they were similar to the ODS in their substantive economic and social policies, generally advocating market-liberal measures and implementing austerity measures after 2008 when holding the finance ministry as part of a coalition government (Saxonberg & Sirovátka 2014). Thus, rather than competing with ODS on the basis of opposing political-economic platforms, they basically wanted to enact ODS policies but with different personnel, as they considered ODS leaders to be corrupt. In other words, they wanted to be like ODS but without its political baggage. When these parties pursued free-market policies and welfare cutbacks while in government, many voters abandoned them and turned to new anti-ODS ODS parties in the hope that these parties would be more socially liberal. Thus, as Table 1 shows, the anti-ODS ODS parties either imploded and were eliminated from parliament after two elections (ODA and Unie svobody), or their support radically decreased by the third election, and they were barely able to keep their parliamentary seats (TOP09 in 2017).

TABLE 1 Czech Election Results

Year	Social Democratic Party (ČSSD)	Civic Democratic Party (ODS)	Communist Party (KSČM)	Christian & Democratic Union–Czechoslovak People’s Party (KDU–ČSL)	Green Party (SZ)	Anti-ODS ODS parties	Public Affairs (VV)	ANO	Right-wing populists: Republicans
						ODA
1992	6.53	29.73	14.05	6.28		5.93			5.98
1996	26.44	29.62	10.33	8.08		6.36			8.01
						US
1998	32.31	27.74	11.03	8.99		8.6			–
						Coalition of four (KDU–ČSL, US–DEU)
2002	30.2	24.47	18.51		–	14.27	–	–	–
2006	32.32	35.382	12.81	7.23	6.29	–	–	–	–
						TOP 09			Dawn of Direct Democracy (Usvit)/SP
2010	22.08	20.22	11.27	–	–	16.7	10.88	–	–
2013	20.45	7.72	14.91	6.78	–	11.99	–	18.65	6.88
2017	7.27	11.32	7.76	5.8	1.46	5.31	–	29.64	10.64

Note: Only parties which won seats in parliament are included in the table.

Source: Czech Statistical Office, available at: http://www.volby.cz/index_en.htm.

It is difficult to confirm this argument directly since the term ‘social liberal’ rarely emerges in Czech discourse, and voters would not likely use it to identify themselves. However, the indirect evidence for this interpretation is reasonably strong. First, survey data consistently show that people who vote for anti-ODS ODS parties place themselves to the left of ODS voters, which indicates that they consider themselves to be more centrist, even though in reality these parties have pursued socioeconomic policies that were often more market-oriented than those of the ODS (Saxonberg 2003, pp. 50, 215). People who voted for these parties also were clearly much more centrist on socioeconomic issues than the leaders of the right-wing parties (Kitschelt et al. 1999, pp. 312–17).

Second, surveys show very high support for welfare benefits, despite the centrist tendency of the Czech electorate. Crucially, this indicates that most Czechs share a social liberal vision of the welfare state rather than a statist, social democratic version. Support for a generous welfare state is so strong that, for example, on average, those who voted for ODS in 1996 had a more favourable attitude toward welfare policies than the average Swede living in a model social democratic country (Saxonberg 2003, p. 50). We can see how by going against the social liberal values of their voters, the anti-ODS ODS parties lose support. To give an example, at the same time that the TOP09 finance minister was carrying out austerity measures in the early part of the 2010s, 76.8% of those voting for his party thought the government should do more to fight poverty (Saxonberg & Sirovatka 2014, p. 455). Meanwhile, a survey taken in 2013 showed that 67% of Czechs believed that the state should provide social security for its citizens, while only 11% agreed that individuals were responsible for their own social situation (Saxonberg & Sirovatka 2014, p. 455).

Third, according to the latest Eurobarometer survey, trust in government and the parliament is the lowest among all EU countries.² This combination of centrist leanings and support for generous social benefits while distrusting the state is typical of social liberalism, which seeks market solutions to welfare issues.

So far, we only know of one survey that asks precise questions about what types of welfare Czechs prefer so that one can clearly differentiate between social democratic and social liberal attitudes. This survey clearly shows that Czechs are very supportive of benefits, but they prefer cash over state provision of services, which is the social liberal viewpoint (Saxonberg & Sirovatka 2009).

If we are correct in our assertion that most people in the Czech Republic have social liberal values although there is no social liberal party to represent them, it becomes clear that other anti-ODS ODS parties declined because voters had originally supported them based on a flawed assumption. Namely, voters initially considered these parties socioeconomically centrist, only to then discover that their goal was in fact to oppose ODS’s leaders rather than its policies. It also becomes clear why centrist populist parties were able to enter parliament by playing on the liberal notions of not trusting the government and believing in markets while at the same time having a ‘thin ideology’ that included support for social benefits. The combination of centrist self-placement, anti-corruption appeals and support for social benefits allowed centrist populist parties to fill the gap on the supply side.

This raises the question of why ANO’s populist centrist competitor VV (Věci veřejné) quickly declined after being elected to parliament in 2010. By comparison, ANO gained more seats than VV in its first elections, in 2013, and then grew rather than declining. While we will not go into detail about VV here, entrepreneurial populists have an advantage over more decentralised populist parties in that since they run their parties very hierarchically, like a company, they have complete control, which prevents the formation of factions. In contrast, the more decentralised VV, whose leader at the time, John Radek, was not its founder-financier, eventually collapsed when some MPs left the party in protest over a bribery scandal involving its founder. Many centrist VV voters switched to the ANO in 2013 as the clearest ‘centrist’ alternative (Pink & Voda 2014; Haughton & Deegan-Krause 2015; Havlík & Voda 2016).

After the 2013 elections, another non-entrepreneurial populist party in the Czech parliament also collapsed from infighting, the right-wing populist Dawn of Direct Democracy (Úsvit přímé demokracie, henceforth shortened to ‘Dawn’). Even though its leader, Tomio Okamura, was a businessman, he did not run his party like a business, nor did he want to run the country like one. Because of his lack of control over the party, infighting arose, causing Okamura to leave in 2015 and found a new one, Freedom and Direct Democracy (Svoboda a přímá demokracie—SPD) (Engler et al. 2019).

By contrast, as a wealthy entrepreneur, Babiš was able to follow in Berlusconi’s footsteps and gain control over much of the mass media (Hanley & Vachudova 2018, p. 286). This ensured that he gained positive media coverage when scandals broke out, such as his alleged misuse of EU funds.

In this article, we argue that in the Czech case, we can best explain the election of Andrej Babiš in terms of the appeal of centrist entrepreneurial populism. We will show that Babiš adopted socioeconomic positions that appealed to the majority of Czech voters who hold centrist rather than social liberal values. These tend to be more highly educated, middle-class professionals than the marginalised, unemployed male voters, who are more typically supporters of right-wing populist parties (Betz & Meret 2013; Bornschier & Kriesi 2013; Hochschild 2016).

ANO and the Czech case

ANO entered the Czech parliament in 2013 and formed a coalition government with the Czech Social Democratic Party (ČSSD) and the Christian and Democratic Union–Czechoslovak People’s Party (KDU–ČSL), with ANO leader Babiš becoming finance minister. In this role, Babiš continued to stress that he was running the economy more efficiently (ANO 2017). As Naxera notes, despite being forced to resign as finance minister a few months before the 2017 elections due to corruption scandals, Babiš presented himself in typical populist fashion to be ‘a dissident fighting against the political system’ (Naxera 2018, p. 45). Despite Babiš’s resignation, in 2017 ANO increased its vote, becoming the largest party and coalition leader in a government that included social democrats and relied on the communists for support.

We may, therefore, summarise our argument as follows. First, Babiš and his party, ANO, are an example of a distinct form of populism: entrepreneurial populism. Second, such parties tend to be non-dogmatic and position themselves on the political spectrum based on where the political opening lies. In the Czech Republic, the opening consists of centrist voters with generally social liberal attitudes. The voters of these parties do not vote for entrepreneurial populist parties primarily because they have populist values; rather, many choose such parties because the supply of political parties does not meet the criteria that voters demand. Given the political elites’ inability to supply these voters with the type of party with which they identify, these voters are willing to accept the argument that pragmatic business entrepreneurs can also succeed in running the government efficiently.

Based on our discussion of entrepreneurial populism and its centrist variant, we may form the following hypotheses as to who votes for centrist entrepreneurial populist parties (EPPs).

First (H1), voters who support centrist EPPs are more likely to perceive little distinction between the existing major parties of the right and left. This is because populist parties often succeed in portraying the rest of the political spectrum as all the same and equally culpable in the undesirable status quo. As populists, the EPPs take no exception.

Second (H2): voters supporting centrist EPPs are more likely to have a middle-class background in terms of education and income. This is because centrist EPPs appeal to discontented middle-class voters. This contrasts to right-wing populist parties, where the voters are often marginalised people who feel like ‘losers’ because of globalisation (namely, less educated men, who are often unemployed).

Third (H3): voters supporting centrist EPPs are more likely to be accepting in their sociocultural views. Discontented middle-class voters who desire sweeping change are not necessarily intolerant and thus reject radical right-wing parties in favour of centrist EPPs.

Fourth (H4): voters supporting centrist EPPs are likely to regard leadership as central because people who vote for centrist EPPs are persuaded that the leader or top candidate of a centrist EPP is likely to be a better steward of national politics on the basis that the ability to run a successful enterprise applies equally to government.

Fifth (H5a): voters supporting centrist EPPs are likely to display social liberal attitudes by supporting existing generous welfare states. Relatedly (H5b): voters supporting centrist EPPs are likely to display social liberal attitudes by opposing high taxes. These are the voters least likely to find alternative options in the party system that appeal to them.

Data sources and method

We used the post-election polls conducted by the Centre for Research on Public Opinion for the periods 28 October–11 November 2013 and 23 October–6 November 2017.³ Our tests confirmed that the data are ‘missing completely at random’, that is, there is no relationship between the missingness of the data and any values, observed or missing.

Unfortunately, the data are limited in that there were no socioeconomic questions about income or profession. In addition, there were some questions about welfare attitudes in 2013 that were not repeated in 2017, which makes it more difficult to obtain a clear picture of welfare attitudes. Nevertheless, both surveys did repeat some questions about welfare and economic issues, making it possible to have some basic idea of voters’ attitudes towards such matters. Despite some limitations, this survey is the best publicly available post-election survey.

We should note that to forestall the issue of data overload by having too many questions, we tried making a scale for populism so we could group some questions together; however, the significance levels decreased further, giving us worse results. We could not make a scale of welfare attitudes because there are several dimensions to social liberal welfare attitudes, and we did not have enough questions to measure each dimension. While, for example, support for high taxes, health and education benefits, and unemployment benefits could make a good scale for measuring social democratic values, they would not measure social liberal values, because we would expect social liberals to agree with social democrats that the state should support health and education, while also agreeing with free-market liberals that taxes should not be high and that generous benefits to the unemployed would give people disincentives to work. Finally, to test if the overload issue presented problems, we ran regressions with only one independent variable, such as change, but it did not give significantly different results. For example, change was not significant in 2017 for voting for ANO even when it was the only independent variable.

Analysis

The parties in the analysis and their social liberal voters

Our starting point is the assumption that a large portion of Czech voters have centrist, social liberal values, although there are no parties that occupy the social liberal space.

As Table 2 shows, both the general population and ANO voters were basically in the centre. On the issue of left–right placement, ANO voters were always in the centre, but in 2013 they were more to the right than non-ANO voters, although the gap greatly decreased in 2017 (in the scale 0 = extreme left and 10 = extreme right). Even those who voted for the right-wing populist parties (Dawn in 2013 and the SPD in 2017) and the Christian Democrats placed themselves in the centre. ODS voters were more to the right, and ČSSD voters were more to the left, but the real outliers were the communists, whose voters were the only ones who positioned themselves in an extreme position. (We left out TOP09 because they had too few supporters in 2017, but their self-placement was between ANO and ODS.)

TABLE 2 Attitudes Among Voters of the Different Parties (10-point scale)

	ANO	ČSSD	ODS	KSČM	KDU–ČSL	Dawn/SPD
Self-placement on a left–right scale (0 = extreme left, 10 = extreme right) (2013)	6.4	3.37	7.98	1.79	5.46	5.23
(2017)	6	3.62	8.02	1.81	5.4	5.83
The state should support healthcare and education (2013)	6.58	7.7	5.05	8.49	7.2	6.92
(2017)	7.46	8.07	6.2	8.96	7.43	8.43
The state should continue owning public enterprises (2013)	6.29	7.22	5.05	8.49	6	6.65
(2017)	6.96	7.79	5.58	8.35	7.15	7.23
The state should support the unemployed (2013)	5.15	5.65	3.41	6.78	5.91	5.35
(2017)	4.01	4.90	3.47	4.7	4.53	4.75
The state should tax the wealthy more (2013)	5.07	6.85	3.02	8.19	6.37	6.62
(2017)	5.15	6.28	3.56	7.15	5.56	5.59
The state should have stricter immigration laws (2013)	7.19	7.51	7.14	8.1	6.34	7.42
(2017)	7.77	7.84	7.72	7.93	6.48	8.83
The state should intervene in the economy (2013)	4.66	5.65	3.95	6.83	5.54	5.38
(2017)	4.89	5.28	3.82	6.12	4.85	4.59
The state should restrict rights to fight crime (2013)	4.16	4.99	4.97	5.14	4.83	4.65
(2017)	4.36	4.78	4.21	5.46	5.18	4.67
The state should restrict foreign capital (2013)	4.19	4.78	3.77	5.71	4.91	5
(2017)	5.57	6.19	4.26	6.93	5.58	6.69
The EU has gone too far (2013)	6.27	6.59	6.79	7.99	5.46	7.31
(2017)	7	7.39	6.66	7.36	5.45	8.34

Source: own calculations using the following databases: Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

Both ANO and non-ANO voters were also basically centrists on welfare issues, such as support for the unemployed, taxing the wealthy, intervention in the economy and supporting human rights, as well as the need to restrict foreign capital investment. Once again, communist voters were the outliers being much more extreme in their support for welfare policies, but among the other parties, what was surprising was the general lack of variance and generally moderately high support for such parties. Both ANO and non-ANO voters thought the state should support health and education, which was also typical of social liberals, who valued universal education and healthcare. Thus, excluding KSČM voters, in 2013, the range was quite narrow on the 10-point scale, between 5.05 for ODS voters and 7.22 for ČSSD voters, with ANO voters in the middle at 6.58. (Support for welfare increased somewhat among all voters in 2017.) Social liberals were less supportive of welfare policies than those with social democratic values, because social liberals tended to blame the unemployed for not doing more to find new jobs, while those with social democratic values supported full employment measures and universalist policies that did not single out certain groups (such as the unemployed) for being less deserving. Furthermore, social liberals were more likely than social democrats to fear that generous unemployment benefits would discourage the unemployed from seeking work. Thus, we see in Table 2 that support for the unemployed was lower than support for health and education among voters of all parties. Still, once again, the voters tended to be centrist in their views, with the mean score for non-communist voters ranging from 3.47 for ODS voters to 5.91 for Christian Democratic voters in 2013, with a slight decline of support in 2017 for voters of all parties except ODS.

General support for continued state ownership was also somewhat above the halfway point; however, one must remember this question was asked at a time and in a society in which most state enterprises had already been privatised. Thus, when respondents read ‘continued state ownership’, they were likely to think of services, such as the postal service, which are publicly owned in most Western countries. Again, excluding communist voters, mean score between voters of the parties is quite small and near the centre, ranging from 5.05 for ODS voters to 7.22 for ČSSD in 2013, with ANO supporters again being between them. Support for state ownership increased among voters of all parties except KSČM.

In two areas that we normally associate with right-wing populism, the mean scores were actually somewhat high: the belief that immigration should be limited and that the EU integration had gone too far. Nevertheless, there were no major differences between ANO voters and voters of other parties. For example, in 2013, the range for believing the EU had gone too far went from 5.46 amongst KDU–ČSL supporters to 6.79 for ODS supporters with ANO in the middle. In this case, Dawn joined communist supporters in being slight outliers, but even their scores of 7.31 (Dawn) and 7.99 (KSČM) were only moderately higher than the rest. In addition, as Table 3 shows, immigration and the EU were not important issues for ANO voters and, according to the respondents, these issues were not the main reason for voting for ANO. In other words, it seems that, with the exception of communist voters, supporters of all the main parties had somewhat centrist, social liberal values on welfare and socioeconomic issues. However, although voters were in the centre, with the exception of ANO, the parties were not.

TABLE 3 Reasons for Voting for ANO

	Switched from ČSSD (%)	All ANO voters (%)
Quality of its programme	28.8	30.9
Trust in and sympathy for its leader*	22	29.6
Good results	15.3	10.3
Quality of the party	11.9	6
Desire for change	5.1	6

Source: own calculations using Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

Note: *‘důvěra a sympatie’.

Table 4 shows questions that are usually associated with populist attitudes, which are scaled from 1 to 5. With the exception of communist voters, the only issues where ANO voters and non-ANO voters had a mean score above the middle point of 3 concerned dissatisfaction with the political situation. Even here, the mean score decreased for voters of all parties in 2017, giving further evidence that populist voters do not necessarily have populist attitudes. In addition, ANO voters actually had lower averages than non-ANO voters in both years. The survey shows that there was substantial support for the claim that democracy is the best system, with ANO voters averaging nearly 4 on a 5-point scale, which indicated that ANO voters did not hold authoritarian views. The mean scores were below the midpoint of 3 for both ANO and non-ANO voters, both for the questions of dissatisfaction with the way the democratic system was working in the Czech Republic and the claim that it did not matter who was in power.

TABLE 4 Attitudes Among Voters of the Different Parties (5-point scale)

	ANO	ČSSD	ODS	KSČM	KDU–ČSL	Dawn/SPD
Dissatisfaction with the political situation (2013)	3.4	3.65	3.88	4.03	3.43	3.77
(2017)	2.82	3.31	3.1	3.67	3.16	3.52
Democracy is the best system (2013)	3.96	3.77	4.33	3.03	3.89	3.93
(2017)	4.08	3.94	4.43	3.41	4.18	3.81
Dissatisfaction with how democracy is functioning (2013)	2.58	2.75	2.68	3.33	2.63	3
(2017)	2.81	2.52	2.39	2.79	2.36	2.9
It does not matter who is in power (2013)	1.95	1.88	1.81	1.82	2.03	2.23
(2017)	1.8	1.93	1.7	1.96	1.94	1.86

Source: own calculations using the following databases: Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

Consequently, there is no evidence that ANO voters espoused populist views. They were generally centrists who wanted a somewhat generous welfare state. As social liberals, they supported healthcare and education more than aid to the unemployed (whom they saw as less deserving). They also supported democracy and human rights and placed themselves in the centre (for reasons of space we do not include a disaggregated table for these questions).

In a situation in which the population is generally social liberal but there are no social liberal parties, it seems that a window of opportunity opened for a centrist-populist party such as ANO. People did not vote for ANO because they had populist beliefs but because they were centrists. However, as we will show below with the logit and multinomial logit regressions, there is evidence that these centrist, social liberal voters also sympathised with the entrepreneurial argument about the ability of a successful entrepreneur to run the country if they could run a business. There was also support for the general populist argument that a charismatic leader could bring change.

Empirical findings: comparing ANO voters in 2013 and 2017

When testing our first hypothesis, that voters perceived little difference between parties, we saw that in 2013 the only significant variable was wanting change (see Table 5). In 2017, wanting change was no longer significant. This makes sense, because in 2013, the ANO ran as a party that would inject fresh blood into the system, but since the party spent four years as part of a coalition government, it could no longer claim to represent newness. Instead, the party emphasised that it should be supported on the grounds of its performance in power. There were no major differences between the 2013 (ANO 2013) and 2017 party programmes (ANO 2017). However, in the 2017 programme, Babiš emphasised his alleged successes as finance minister. Both programmes claimed the party wanted to fight corruption, although Babiš himself had been involved in corruption scandals. Unfortunately, the two polls that comprise our data did not include the exact same questions concerning corruption. In 2013, the question was whether fighting corruption was a reason for voting for the party, while in 2017, it was whether fighting corruption was the most important issue. In both cases, the correlations were actually negative but not significant.

TABLE 5 Logistic Regression Models Explaining Vote Choice for ANO (2013 and 2017)

	2013 n = 657	2017 n = 712
Hypothesis 1: Voters perceive little difference between parties
Dissatisfied with how the democratic system is functioning	−0.29 (0.23)	−0.45 (0.3)
Dissatisfied with political situation	−0.25ˆ (0.14)	0.05 (0.17)
Does not matter who is in power	0.21 (0.16)	0.12 (0.2)
Want change	2.71*** (0.41)	0.5 (0.74)
Fighting corruption is reason for voting	−0.16 (0.22)
Fighting corruption is the most important issue		−0.28 (1.12)
Hypotheses 2: Supporters are middle class
Has ‘maturita’^†	0.06 (0.29)	0.26 (0.35)
Self-employed	−0.08 (0.84)	−0.15 (0.93)
Student	0.22 (0.93)	−0.79 (1.12)
Pensioner	−1.03 (0.85)	−0.66 (0.85)
Unemployed	−1.94ˆ (1.02)	3.1 (1.46)*
Parental leave	0.19 (1.02)	−0.85 (1.11)
Fulltime employee	−0.31 (72)	−0.27 (0.79)
Hypothesis 3: More likely to be tolerant
EU integration gone too far (0–10)	0.07 (0.06)	0.02 (0.08)
Fight crime	−0.02 (0.05)	−0.05 (0.06)
Stricter immigration policies	0.02 (0.06)	0.22 (0.1)*
Democracy best system	−0.01 (0.17)	0.24 (0.20)
Hypothesis 4: Leadership important
Sympathise with Babiš	1.01*** (0.09)	1.27 (0.12)***
Hypothesis 5: Support for welfare
Centrist (3–7 on scale of 0–10)	1.01** (0.32)	0.49 (0.39)
State should intervene in the economy (0–10)	0.12ˆ (0.06)	0.27** (0.09)
Restrictions on foreign capital (0–10)	−0.03 (0.06)	−0.07 (0.07)
Higher taxes for the rich (0–10)	−0.13* (0.05)	−0.1 (0.06)
State should be responsible for unemployment (0–10)	0.07 (0.05)	−0.07 (0.07)
State should continue owning public enterprises (0–10)	−0.1 (0.06)	−0.08 (0.08)
State should support education and healthcare	−0.07 (0.06)	−0.25 (0.09)**
Control variables
Male	−0.44 (0.29)	0.75* (0.36)
Age	0.01 (0.01)	0.02 (0.02)
Constant	−6.38*** (1.91)	−10.43*** (2.12)
Model statistics
Log Likelihood	−173.997	−130.284
Prob > chi^2	0.000	0.000
Pseudo R^2	0.54	0.69

Note: Standard error in parenthesis, ˆ = significant at 0.1 level, * at 0.05 level, ** at 0.01 level, *** at 0.001 level. ^†‘Maturita’ is a school leaving examination. This examination is taken by the students at the end of their secondary education, and must be passed in order to gain admittance into the higher education institutions in the Czech Republic.

Source: own calculations using the following databases: Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

When testing hypothesis 2, that voters likely to support centrist EPPs are middle class, no socioeconomic variables were significant for 2013. In 2017 there seemed to be a slight shift, as being unemployed became positive and significant at the 0.05 level. Thus, we had to reject this hypothesis, as the socioeconomic background of the voters did not seem to matter.

In hypothesis 3, we tested whether voters of centrist EPPs were more likely to be tolerant towards immigrants and positive about the EU. Since our hypothesis was that ANO voters were social liberals without a social liberal party rather than traditional populist voters, we did not expect them to be xenophobic or Eurosceptic. It turned out that for 2013, none of the variables measuring intolerance were significant. This supported our hypothesis by suggesting that even though the ANO is a populist party, its voters do not display the intolerance that is common for voters of right-wing populist parties. However, as with hypothesis 2, we saw a shift in 2017 in that wanting stricter immigration policies became significant and positively correlated with voting for the ANO. Was the ANO’s base shifting farther to the right and embracing more traditional right-wing populist views? Below we argue against this after first testing all our hypotheses.

In hypothesis 4, we tested whether leadership was central to centrist EEP voters. ANO voters were much more likely to sympathise with Babiš than are other voters. This was highly significant in both years, which we would expect from an entrepreneurial populist party that is centred on a charismatic leader from industry who claims to be able to run the country efficiently like a business. If a party has a thicker ideology, then voters might likely vote for the party for ideological reasons even if they do not feel great sympathy for its leader.

In hypothesis 5a, we examined the connection between voting for EPPs and voters’ social liberal welfare attitudes. Our calculations for 2013 showed that ANO supporters in the Czech Republic appeared to be centrist, as self-placement between 3 and 7 on a scale of 0–10 was strongly correlated with voting for ANO. They also believed that the state should intervene in the economy, although this was only significant at the 0.1 level. As predicted in hypothesis 5b, although ANO voters would like generous welfare benefits, they were against higher taxes for the rich. The other welfare variables were not significant.

In 2017, we saw some changes. Being centrist was no longer significant, while believing that the state should intervene in the economy became significant at the 0.01 level. Opposition to higher taxes for the rich was no longer significant, while opposing state support for education and healthcare became significant. In other words, voters’ welfare attitudes were somewhat contradictory in both samples. They thought the state should intervene in the economy, a ‘leftist’ view, but in 2013 they were against higher taxes for the rich, which is what we would expect for voters with social liberal values, namely support for state intervention and generous social benefits (except for unemployment benefits) and opposition to higher taxes.

In 2017 voters shifted to the left. Believing that the state should intervene in the economy became significant at the 0.01 level rather than the 0.1 level, and they no longer opposed taxes for the rich. On the other hand, they shifted to the right in opposing state support for healthcare. However, we would expect that centrist voters would hold socioeconomic views that are neither clearly leftist nor rightist.

Causes of the shift

At first glance, it might seem as if ANO voters had shifted from being centrist, somewhat socially liberal in 2013, to becoming more like traditional radical right voters in 2017. Their social base was more in line with radical right parties as being unemployed was now positively correlated with voting for the party; their view of immigration was more negative, which we would expect from a right-wing populist party; and even though they thought the state should intervene in the economy and even though they no longer opposed higher taxes on the wealthy, in 2017 they were against state support for healthcare and education.

A more careful analysis offers a different explanation. The change in the relationship between ANO and non-ANO voters was not due to a change in attitudes among its base but rather to the implosion of the ČSSD and the shift of many of its voters to ANO. Both ANO and the social democrats were in a coalition government in the period 2013–2017, but ANO was able to take credit for the economic successes as Babiš was finance minister. Meanwhile, the social democrats were split and hurt by constant infighting, losing almost two-thirds of their votes in the 2017 elections (Maňák 2018). The focus of our article is the rise of ANO rather than the collapse of the ČSSD, but we should note here that infighting within the social democratic party began after the 2013 elections, when President Miloš Zeman—who previously had been the party leader but had gone on to found his own left-wing populist party, Party of Civic Rights—supported a failed putsch against party leader Bohuslav Sobotka. When Sobotka fired Babiš as finance minister after the corruption scandal, this led to further infighting within the party, which led to Sobotka’s resignation as party leader, although he continued as prime minister.

What is important for our study is that after the ČSSD imploded, many former social democratic voters went over to the ANO in the 2017 elections. Our survey data show that fully 69% of those who switched to the ANO were former ČSSD voters. While there are inherent reliability problems in asking respondents for whom they voted in previous elections, the fact that the ČSSD lost a lot of votes while the ANO gained a lot of support makes this assumption appear reasonable. Our assumptions are backed by the findings of other authors using different data, who also concluded that ANO took votes from the social democrats in the 2017 elections.⁴

The mean score on ideological self-placement moved slightly leftward from 6.5 to 6.0 on a scale of 0–10 (see Figure 1), which reflects the support from former social democratic voters. Thus, even though being more centrist than other voters was no longer statistically significant in 2017, most ANO voters still placed themselves in the centre. Since many social democratic voters went over to ANO, the self-placement curve shifted slightly to the left and became somewhat flatter. This result is not surprising: in a country with no social liberal parties and where most voters are social liberals, we would expect the social liberal vote to be split. Many social liberals would have refrained from supporting Babiš because of his populism and alleged corruption; therefore, a portion of social liberals could have been expected to vote for the social democrats in 2013, since they shared the social democratic belief in a generous welfare state, even if they disagreed on how it should be implemented. As the social democratic party imploded, it is not surprising that some of its former social liberal voters turned to the most centrist party as an alternative.

FIGURE 1. Self-placement Scores of ANO Voters

Notes: 0=extreme left, 10=extreme right.

Source: own calculations using the following databases: Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

If most of the new ANO voters were former social democratic supporters, how can we explain the rightward shift on such issues as immigration? It turned out that these former social democratic voters were strongly anti-immigrant. These crossover voters scored 8.42 on a scale of 0–10 (compared to 7.77 for all ANO voters and 7.84 for ČSSD voters), where 10 means support for stricter immigration laws. Meanwhile, our multinomial regression using ANO as a baseline for 2017 shows that support for stricter immigration laws was insignificant among social democratic voters (see Table 6). This indicates that when the ČSSD imploded, those who were more strongly anti-immigrant tended to go over the ANO, while those who were not as strongly anti-immigrant remained with the social democrats.

TABLE 6 Party Vote Choice in Czech National Elections 2017, Multinomial Logit Regression with ANO as the Baseline, N = 678

	ODS	ČSSD (Soc Dem)	KSCM (communist)	Pirates	SPD (right Populist)	Other
Hypothesis 1: Voters perceive little difference between parties
Dissatisfied with how the democratic system is functioning	0.16 (0.38)	0.11 (0.39)	0.52 (0.41)	0.23 (0.37)	1.02 (0.37)**	0.34 (0.34)
Dissatisfied with political situation	0.17 (0.22)	−0.17 (0.22)	−0.04 (0.25)	−0.31 (0.21)	0.92 (0.22)	0.06 (0.19)
Does not matter who is in power	−0.37 (0.28)	0.11 (0.25)	0.01 (0.28)	−0.2 (0.24)	−0.13 (0.24)	−0.11 (0.22)
Want change	−1.04 (1.31)	−15.82 (1,243.12)	−0.38 (1.17)	1.2 (0.85)	−0.34 (0.92)	−1.17 (1.07)
Fighting corruption most important issue	−15.38 (1,659.63)	1.12 (1.13)	0.17 (1.7)	−0.99 (1.55)	0.02 (1.44)	−1.10 (1.4)
Hypotheses 2: Supporters are middle class
Has ‘maturita’†	0.26 (0.47)	−0.62 (0.45)	−1.18 (0.52)*	−0.56 (0.43)	0.27 (0.46)	0.07 (0.4)
Self-employed	−0.19 (1.07)	0.31 (1.51)	1.08 (1,231)	−0.46 (1.1)	0.27 (1.09)	−1.17 (1.03)
Student	−0.38 (1.61)	2.58 (1.71)	1.3 (1,905.92)	0.54 (1.25)	0.32 (1.47)	0.27 (1.22)
Pensioner	−0.80 (1.00)	0.78 (1.32)	15.85 (889.64)	−0.16 (1.07)	0.44 (1.04)	−0.36 (0.89)
Unemployed	−18.12 (2,151.27)	−16.55 (2,629.52)	−3.63 (2,280.4)	−3.68 (1.57)*	−1.65 (1.7)	−4.91 (2.14)
Parental leave	1.355 (1.31)	−12.53 (1,701.89)	1.92 (1,487.29)	0.19 (1.3)	−14.69 (1,660.12)	1.15 (1.18)
Full-time employee	−0.08 (0.89)	1.36 (1.26)	14.79 (889. 64)	0.12 (0.89)	0.52 (0.92)	−0.3 (0.81)
Hypothesis 3: More likely to be tolerant
EU integration gone too far (0–10)	0.02 (0.1)	0.02 (0.11)	−0.03 (0.11)	−0.01 (0.1)	0.18 (0.11)´	−0.12 (0.09)
Fight crime	0.02 (0.08)	0.13 (0.08)´	0.1 (0.09)	0.01 (0.08)	0.01 (0.08)	0.05 (0.07)
Stricter immigration policies	−0.15 (0.12)	−0.21 (0.13)	−0.46 (0.14)***	−0.29 (0.11)**	−0.01 (0.13)	−0.33 (11)**
Democracy best system	−0.12 (0.28)	−0.01 (0.27)	−0.52 (0.28)´	−0.07 (0.27)	−0.18 (0.25)	−0.29 (0.23)
Hypothesis 4: Leadership important
Sympathise with Babis	−1.3 (0.13)***	−1.29 (0.13)***	−1.22 (0.14)***	−1.36 (0.13)***	−1.22 (0.13)***	−1.33 (0.12)***
Hypothesis 5: Support for welfare
Centrist (3–7 on scale of 0–10)	−1.07 (0.49)*	0.36 (0.54)	−1.42 (0.57)*	0.53 (0.5)	−0.13 (0.51)	−0.02 (0.45)
State should intervene in the economy (0–10)	−0.28 (0.11)*	−0.3 (0.11)*	−0.13 (0.12)	−0.4 (0.11)***	−0.34 (0.11)**	−0.32 (0.1)**
Restrictions on foreign capital (0–10)	−0.06 (0.09)	0.06 (0.09)	0.19 (0.11)´	−0.01 (0.09)	0.08 (0.09)	0.06 (0.08)
Higher taxes for the rich (0–10)	0.03 (0.08)	0.16 (0.08)*	0.22 (0.09)*	0.06 (0.08)	0.11 (0.08)	0.07 (0.07)
State should be responsible for unemployment (0–10)	−0.01 (0.09)	0.18 (0.09)*	−0.04 (0.1)	−0.01 (0.09)	0.13 (0.09)	0.03 (0.08)
State should continue owning public enterprises (0–10)	−0.03 (0.1)	0.25 (0.11)*	0.23 (0.12)`	0.01 (0.1)	0.05 (0.10)	0.06 (0.09)
State should support education and healthcare (0–10)	0.1 (0.11)	0.33 (0.11)**	0.53 (0.12)***	0.39 (0.11)***	0.21 (0.11)`	0.2 (0.1)*
Control variable
Male	−0.89 (0.46)´	−0.71 (0.45)	−0.58 (0.51)	−0.93 (0.43)*	−0.42 (0.45)	0.9 (0.41)*
Age	0.01 (0.02)	0.03 (0.22)	0.02 (0.03)	−0.03 (0.02)	−0.03 (0.02)	−0.01 (0.02)
Constant	10.63	1.40	−11.25	3.52	3.52	11.65
Model statistics
Log Likelihood −682.2204
Prob > chi2 0.0
Pseudo R2 .445

Note: Standard error in parenthesis, ˆ = significant at 0.1 level, * at 0.05 level, ** at 0.01 level, *** at 0.001 level. ^†‘Maturita’ is a school leaving examination. This examination is taken by the students at the end of their secondary education, and must be passed in order to gain admittance into the higher education institutions in the Czech Republic.

Source: own calculations using the following databases: Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences.

Nonetheless, anti-immigrant feelings seemingly did not provide an impetus for these voters to switch over to the ANO. Respondents were asked an open-ended question about their choice of party. Amongst those who had switched from the social democrats to ANO, 28.8% stated it was because they thought the party had the best programme; another 22% gave trust in the party and sympathy for its leader; 15.3% said the party had good results and 11.9% claimed the quality of the party (‘kvalita strany’) was their main reason for selecting ANO. In addition, ANO did not run an anti-immigrant campaign.

Table 3 compares ANO voters who switched from the social democrats with ANO voters in total. If any of them have given immigration or anti-immigration views as a reason for choosing the ANO, their reply would have fallen under the category of ‘other’, which was 1.69% for crossover voters and 1.99% for all ANO voters. The major difference between the two groups is that those who switched from the social democrats were much more likely to give ‘good results’ and ‘quality of the party’ as the main reasons for their choice than ANO voters in general. This shows, once again, the effect of the ANO having been in government and convincing voters that its entrepreneurial populist style worked. Thus, a plausible interpretation is that since Babiš was able to run a business, he was seemingly also able to run the country well. Being disappointed in the social democrats, who were in turmoil, the voters gave the ANO rather than the ČSSD credit for the economic successes, especially since as finance minister, Babiš was the cabinet member with the greatest ministerial responsibility for economic policy.

Thus, entrepreneurial populist claims rather than anti-immigrant views were arguably the main reason for social democratic voters to switch to ANO in 2017. In addition, the ANO’s electoral programme did not give much reason for xenophobes to choose the party (ANO 2017). Furthermore, since neither ANO voters as a whole nor the newly won-over social democratic voters considered immigration an issue, we would not claim that the party was becoming an anti-immigrant, right-wing populist party or that those voting for it held such views. Moreover, if a voter had been strongly motivated by anti-immigrant views, then the logical party to support would have been the SPD, which ran a strongly anti-immigrant and Eurosceptic campaign supporting leaving the EU. Yet, among SPD voters, support for stricter immigration policies was statistically insignificant, the same as for votes for the social democrats or the ODS (see Table 5). Thus, xenophobia was not an important issue for voters, as relative support for or opposition to stricter immigration rules was not an issue that differentiated the voters of these parties.

Although the variables ‘male’ and ‘unemployment’ were significant in 2017 in the full model with all the control variables, this cannot be taken as evidence of a radical right profile among ANO voters. First, the majority of ANO voters were women, as were the majority of former social democratic voters who shifted to ANO. Second, in the case of unemployment, only six of 306 ANO voters in this survey claimed to have been unemployed. Thus, the socioeconomic background of ANO voters does not match what we would expect from populist parties. They did not have xenophobic or Eurosceptic views and they were not disillusioned with mainstream politics.

The results are apparent on the most important issue for entrepreneurial populism; namely, sympathy for the leader, since its main idea is that if someone can run a business, they can run the country. Not only was this variable highly significant in both surveys, the multinomial regression shows that this was the only variable significant for voters of all parties. For voters of other parties, sympathy for Babiš was significant at the 0.001 level, but in this case, it was negatively correlated. It seems that either a voter accepted the entrepreneurial populist argument and sympathised with Babiš and, therefore, voted for ANO, or that they rejected this argument and, therefore, voted for another party. Since many of those who voted for the other parties—whether the more left-leaning social democrat ČSSD or right-leaning ODS—supported generous welfare policies, they might have been open to a truly social liberal party, but for various reasons, they were not willing to vote for a populist leader. This can explain why attitudes towards Babiš turned out to be more important than welfare attitudes. In other words, it is likely that many ODS and ČSSD voters had social liberal values but did not want to vote for a populist party.

Conclusion

In this article, we analysed the electoral success of Babiš and his party ANO as a case of entrepreneurial populism, a distinct form of populism. In that it deviates from the more common radical variants of populism, entrepreneurial populism appeals to voters in different ways and under different circumstances. This conceptualisation is not only intended to explain the emergence of an ‘instant catch-all party’ in Czech Republic (Innes 2002, p. 88) but to place this phenomenon in a broader international context. Proceeding from the argument that the supply of political parties does not always equal the demands of the electorate, we show that this opened a window of opportunity for entrepreneurial populist parties to fill the emerging gap. In entrepreneurial populism, the appeal of change agents promising to run the country as successfully as their businesses is matched with a ‘thin’, catch-all ideology that can be adapted according to political expediency. Thus, the placement of entrepreneurial parties in the political spectrum depends on the national context. In the Czech Republic, the window of opportunity was squarely in the centre, given that the majority of the population held centrist, social liberal views.

The Czech case also indicates that when supply does not meet demand, voters might be willing to vote for populist parties even if they do not have populist attitudes. Babiš’s appeal was the only statistically significant variable for voters of all parties: those who sympathised with him were more likely to vote for ANO, and those who did not sympathise with him were more likely to vote for other parties. Thus, his personality seemed to matter. Predictably, the variable change was no longer significant in 2017 when ANO was an incumbent party. This indicates that ANO voters did accept the entrepreneurial populist argument that since Babiš knows how to run a business, he knows how to run the country, a claim validated by his performance as finance minister in 2013–2017. Otherwise, ANO voters did not show the traditional populist disdain for democracy, its institutions or its functioning. Thus, if the Czech case shows that a country can have successful populist parties without populist voters, we may need to reconsider our understanding of the relationship between voters and parties.

Finally, our concept of non-ideological entrepreneurial populists searching for openings can explain why we find such a diversity of left–right placement among such movements. Further studies applying our theory could investigate whether Berlusconi positioned his party at the centre-right because he saw an opening there, since the traditionally dominant party, the Christian Democrats, had recently imploded due to scandals, thus leaving an opening in the centre-right. Meanwhile, further studies could also explore the development of Trumpism in the United States. Trump was known to have socially liberal views on such issues as abortion and gay rights, but when he decided to run for president, he likely saw the political opening at the more extreme right given the recent failures of more traditional conservatives in the previous two elections. In order to run on the far right, he needed the support of evangelical voters, so being a basically non-ideological entrepreneurial populist, he had no trouble switching his views so he could place himself where he found a political opening.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Grant Agency of the Czech Republic, Grant 19-12289S and the Slovak Grant Agency: APVV-17-0596. It was also co-funded by the Erasmus+ Programme of the European Union, project number 611572-EPP-1-201 9-1-SK-EPPJ MO-CHAIR while Saxonberg was at the Comenius University, Faculty of Social and Economic Sciences, Institute of European Studies and International Relations.

Notes on contributors

Steven Saxonberg

Steven Saxonberg, Södertörn University, Department of Sociology, 141 89 Huddinge, Sweden; Research associate at the Department of Social Policy and Social Work, Masaryk University, Joštova 218/10, 602 00 Brno, Czech Republic. Email: [email protected]

Reinhard Heinisch

Reinhard Heinisch, Department of Political Science, University of Salzburg, Rudolfskai 42, Salzburg, Austria. Email: [email protected]

Notes

1 See Hloušek et al. (2020) for a more detailed background on Babiš and his business career.

2 Standard Eurobarometer 94, Winter 2020–2021, available at: https://europa.eu/eurobarometer/surveys/detail/2355, accessed 12 October 2022.

3 Povolební studie, 28.10–11.11.2013, Listopad and Povolební studie 2017, Centrum pro výzkum verejného mínení (Centre for Research on Public Opinion) at the Institute of Sociology of the Czech Academy of Sciences. To get access to these datasets, contact the centre at: https://cvvm.soc.cas.cz/cz/.

4 For example, Maškarinec (2019).