Do Parties Adjust Their Policy Proposals to the Political Context? A Study of Election Promises During the 2022 Italian General Election

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.


Introduction
The strategic emphasis placed by political parties on specific policy domains constitutes a pivotal aspect of party competition. Political actors can direct their attention to political issues over which they are deemed competent (Petrocik, 1996), or they can intercept citizens' demands on issues perceived salient by the electorate (Ansolabehere & Iyengar, 1994;Klüver &Spoon, 2014). Whether party competition is interpreted as a top-down process or conceived according to a bottom-up perspective, the political science literature has extensively explored the importance attributed by political parties to key policy issues (Green-Pedersen, 2007) alongside their attempt to prioritize the implementation of those policies on which they have most engaged during an electoral campaign (Bevan et al, 2011;Hobolt & Klemmemsen, 2005).
In this article, we explore the emphasis attributed by Italian political parties to specific policy issues through the analysis of the parties' election commitments contained in their programmatic manifestos before the 2022 general election. Differently from issue salience studies which usually employ Manifesto Project data -and therefore look at the whole text of party programs (Wagner & Meyer, 2014;Abou-Chadi et al, 2020;Spoon & Klüver, 2015;Green-Pedersen, 2007) -or parliamentary speeches (Hobolt & Klemmemsen, 2005), the focus of this paper is on election promises. Talking about something and making concrete promises on a specific issue are two different things and, as such, require a different commitment and effort from parties. When a party, for instance, discuss an issue at length in its campaign manifesto, parliament, or the media, this does not automatically mean that the party takes some sort of commitment to act or even just prioritise that specific policy domain in its governing agenda. It might just address the existence of a problem and explain the state of the art of the legislation as well as passing judgements on previous actions or competitors' positions.
Campaign programs, for instance, do not merely include policy proposals, but also 'lots of rhetoric' (Harmel, 2018, p.235). Whereas, when a party pledges to adopt a specific measure or attain a certain goal, it takes a concrete commitment that it is then expected to keep if elected. While it is harder to hold partisan actors accountable for what they talk about during the electoral campaign, there is a direct and immediate link between promising and acting in the chain of representation. As the Downsian model of representative democracies and Mansbridge's (2003) promissory representation theory suggest, parties are elected on the basis of their policy proposals, among other things (Artés, 2011;Naurin et al, 2019). Voters, in turn, will assess the work of the incumbents at the polls according to their ability to fulfil their mandate (Matthieß, 2020). As a result, formulating a new promise does not come cheap: Differently from merely addressing one new topic, making an additional pledge means that the party should truly commit to enacting it (or at least trying to). In this light, this paper captures the actual commitments that Italian parties make in the run-up to the 2022 election and the policy issues in which they engage themselves. Accordingly, we manually selected all manifesto statements that correspond to a campaign promise. The resulting sample includes 3,992 pledges within the campaign programs of the main six political parties and two pre-electoral coalitions competing in the 2022 Italian general election.
This study contributes to the literature on party competition and party responsiveness. By examining parties' electoral commitments, we show that political conflict and parties' mutual interactions are structured around the same policy domains, with parties' engagements devoted mainly to economicrelated issues. Furthermore, we provide empirical evidence of the limited role played by the political context that formed the backdrop to the 2022 elections in influencing the distribution of parties' electoral pledges on the most salient issues. The Covid-19 pandemic, the implementation of EUfunded investments in the ecological transition, the war in Ukraine and the resulting energy crisis, did not lead to an increase in parties' commitment towards environmental issues and to a more marked engagement in the energy and defence sectors. At the same time, examining the congruence between voters' priorities and the share of promises made in different policy sectors, we show that the election promises included in the 2022 parties' manifestos only partially respond to the electorate's concerns.
The analysis of electoral promises contributes to better clarifying parties' electoral dynamics in several aspects. First, it allows us to grasp the differences between political actors with regard to their ideological positioning on specific issues. By proposing divergent solutions to the same problem, political actors signal not only the degree of importance attributed to that specific issue but also their ideological placement (Green-Pedersen, 2007). Second, the analysis of election promises allows for predictions about what a political party will do once in power. The selective salience attributed by political actors to certain issues affects the content of governing agreements and the decision-making process of coalition partners (Harmel, 2018). Finally, the analysis of electoral promises can serve as a proxy for assessing the degree of accountability of political parties. A political party that does not keep its electoral promises can be punished by voters who can use this information to evaluate its performance (Matthieß, 2020).

Pledge making in the literature
To tackle the question of which election pledges parties make, it is essential to first understand why political parties decide to make these promises. Aside from electoral purposes, campaign promises fulfil various functions in representative democracies (Håkansson & Naurin, 2016;Naurin et al, 2019). First of all, they are instruments through which parties can convey their position on different issues to the electorate. A party is expected to promise to achieve a certain goal or adopt a specific measure in line with its ideological stance (Naurin et al, 2019). As the promises contained in election manifestos mirror parties' policy preferences, citizens are presented with a clearer choice on election day. They are made aware of what different candidates commit to doing if elected, their priorities, and aims. At the same time, looking at pledge fulfilment rates, the electorate can assess and, hence, evaluate governments' past performances and their ability to deliver on their electorate mandate (Matthieß, 2020). Election promises are also important for intra-party dynamics, as manifestos are often the outcome of compromises and concessions between party factions (Harmel, 2018), and for coalition building. In those countries in which governments are formed by two or more political forces, as is the case in Italy, promises can be either the result of bargaining between partners of a pre-electoral coalition sharing a common platform (Moury & Timmermans, 2008) or 'a basis for negotiation' for prospective alliances (Håkansson & Naurin, 2016). Political parties are, thus, driven not only by policy but also vote and office-seeking motivations when formulating their policy proposals (Naurin et al, 2019).
Existing studies on pledges are mainly concerned with political parties' capacities to carry out their electoral commitments (Artés, 2011;Naurin et al, 2019). Very few studies examine which promises parties actually make. In the volume edited by Naurin, Royed, and Thomson (2019) about pledge fulfilment in twelve countries, a whole chapter is devoted to the topic. They suggest that parties make different types of promises depending on their position in the previous legislature (in government or opposition) and their party family: Incumbents, for instance, are found to take more commitments to preserve the existing situation ('status quo' promises) compared to opposition parties (Naurin et al, 2019). Similarly, in his analysis of pledge fulfilment in Spain, Artès (2011) shows that the socialist party (PSOE)'s manifestos included a higher percentage of expansionary promises while the Popular Party prioritised more restrictive measures. While these studies provide evidence that parties 'make substantively important pledges' (Naurin et al, 2019, p.294), they fail to systematically explore the issue areas of the promised policies. They do sometimes indicate the policy domains in which the highest amount of election promises are made 1 , without, however, providing explanations or investigating this aspect more in-depth. As the aim of this paper is to explore the issues of the election promises in the 2022 Italian general election and, more specifically, whether parties adapt their policy proposals to the context in which the electoral campaign takes place, we complement pledge literature with insights provided by party and issue competition studies.

Which policy issues do parties prioritise?
Existing literature has long recognised that party competition results from a combination of two main strategies (Green-Pedersen, 2007; see also Abou-Chadi et al, 2020). On the one hand, parties compete by taking different stances on the same issues (positional competition). According to this approach, parties promote the adoption of policy measures that are somehow different to the ones advertised in their competitors' election programs in the same policy domain (Abou-Chadi et al, 2020). On the other hand, the so-called issue competition literature underlines that parties' agendas also differ in the emphasis given to different policy domains (Dolezal et al, 2014;Green-Pedersen, 2007).
Regardless of their policy preferences, some politicians might discuss more questions related to social benefits or social and civil rights, while others might centre their interest around immigration and security matters. A third approach to party and electoral competition prefigures an alternative mechanism for conceiving the interactions between parties and voters. According to the theory of valence politics, candidates compete by claiming possession of reputational resources that allow the attainment of those policies that are largely advocated by the electorate, i.e. valence issues (Stokes, 1992).
In this study, we build on issue competition literature for one main reason: It is easier to observe a change in the emphasis attributed to a variety of policy domains in manifestos presented in two subsequent elections rather than a shift in party position. It is true that even avoiding a salient issue could harm the party's image (Sides, 2007). Nevertheless, shifts in policy positions are electorally riskier as they are usually frowned upon by the party base (Tavits, 2007).
There are several reasons behind parties' choice to focus on one issue instead of another and, hence, to make more promises in a specific policy domain. First of all, political parties are expected to pay more attention to currently salient problems. Clearly, party manifestos have to refer to the issues, concerns, and challenges that are salient at the time of the election for citizens (Klüver & Sagarzazu, 2016) and receive quite intense media coverage (Green-Pedersen & Stubager, 2010). In other words, parties should ride the wave of public interest. As the theories of party responsiveness have long underlined, parties' programmatic agenda should reflect social problems and their constituencies' concerns (see the theoretical discussion in Spoon & Klüver, 2015). This argument is based on the assumption that voters would sanction those politicians that they perceive as distant and indifferent (Ansolabehere & Iyengar, 1994;Sides, 2007). As a result, parties' campaign proposals should provide policy responses to the issues deemed as the most relevant by the electorate and the new challenges that arise.
Parties' choices do not, however, occur in a vacuum. One party alone cannot establish around which issues the electoral competition is played. Instead, partisan actors have to also respond to the matters raised by their competitors. As the concept of party-system agenda developed by Green-Pedersen and Mortensen (2010) well summarises, when a party increasingly discusses a certain topic, it forces all other parties to take a stand on it as well. 2 An issue can therefore become salient when an external event or even another party in the system bring it to the forefront of the governing agenda. Following this reasoning, the election promises made by different parties for the same electoral campaign are expected to deal with a similar range of issues.
Furthermore, all parties should make more commitments in the policy domains on which they enjoy ownership and are capable of credibly claiming competence (Petrocik, 1996). Parties compete by strategically manoeuvring the emphasis attributed to specific policy domains for which they ought reputational advantages in the eyes of the electorate. In this perspective, partisan actors are likely to dwell on the policy issues they own in their speeches, their agenda, and their manifestos (Petrocik, 1996). Concerning this last point, drawing on Manifesto Project data on 17 countries, Wagner and Meyer (2014) show that election programs are largely focused on the issues the party owns. Dolezal et al. (2014), employing another dataset (the AUTNES), come to the same conclusion in the case of Austria. As issue ownership is obtained, among other things, by past government performance -what Petrocik calls 'the record of the incumbent' (Petrocik, 1996, p.827) -making promises in those sectors can make parties gain more credibility and, hence, support.
Nevertheless, this approach is not necessarily a mutually exclusive framework for the analysis of party competition. With its reference to the reputational component of the party/candidate, the idea of issue ownership indirectly relates to valence politics. A recently developed theoretical framework that incorporates position and valence within the same paradigm is the issue yield theory (De Sio & Weber, 2014). According to this approach, political parties should emphasise those policy domains with a high yield. In other words, they should prioritise the issues that offer the opportunity to expand electoral consensus beyond the party's constituency whilst minimising the risk of eroding existing support. For this to happen, these policy domains must be shared by both the party's voter base and the electorate at large and must be those on which the party is perceived as most credible. Several studies have investigated parties' strategies and issue opportunities in recent European countryspecific elections relying on the issue yield theory (see inter alia Franzmann et al, 2020;Plescia et al, 2020). These scholarly contributions highlight that in recent national elections, political partiesrather than prioritizing ideological coherence -have been tailoring their campaign agendas and, thus, their electoral commitments to focus on those issues that have proven to be electorally rewarding, based on the aforementioned risk opportunity assessment (De Sio & Weber, 2020).
In addition to riding-the-wave and issue ownership arguments, parties are more inclined to highlight a specific issue if the issue is divisive (positional issues; Spoon & Klüver, 2015;Green-Pedersen, 2007). As the positional competition argument maintains, one of the easiest ways in which parties differentiate themselves in the electoral campaign is by taking different stances on the same issues.
This strategy is of course only successful when the electorate is polarised around the issue under consideration or, at least, when there is not a broad consensus among voters. In a comparative longterm analysis of nine European countries, Spoon and Klüver (2015) distinctly show that parties' manifestos are increasingly focused on immigration and European integration in those periods when voters' preferences on those issues are more conflictual.

The XVIII Legislature: from multiple-systemic crises to election
The XVIII legislature in Italy (2018-2022) was marked by an unstable and complex political climate that saw three highly heterogeneous coalition governments alternate in office (see Chiaramonte, 2023;Garzia, 2022) In the aftermath of the early dissolution of the XVIII legislature, the electoral context assumed a fourpole configuration (Garzia, 2022). Brothers of Italy (FdI), emerging as the favoured party in the preelectoral polls, formed a coalition with two of the main partners of the Draghi government, the League and Go Italy. On the left camp, the three key players who challenged the centre-right coalition were the Democratic Party, united with More Europe (+E), the Greens and Left Alliance (AVS), the Five Star Movement, and the coalition formed by Action and Italy Alive (A-IV).

Expectations
As discussed in the theoretical section, existing literature suggests that political parties should orient their emphasis towards those issues that hold greater salience. The premature end of the Draghi government and the resulting electoral campaign took place in a political context in which energy, defence, and environmental issues dominated the public debate. In the analysis conducted in this study, we formulate our expectations by focusing primarily on these policy domains, to further explore the potential link between the political context and electoral promises. Rising gas prices resulting from the energy crisis, the allocation of resources designated by the PNRR for ecological transition investments, and the war in Ukraine represented an opportunity for these dimensions to acquire renewed attention. These issues also became a top priority for voters, who expect parties to position themselves and propose solutions to current problems. In their study of the evolution of pledge-making, Håkansson and Naurin (2016) show that parties are increasingly making a large number of promises on the issues that voters deem as the most important. Pre-electoral individual data from the 2022 CISE/ICPP survey (data presented in Improta et al, 2022) portray that the key priorities for the Italian electorate in 2022 concern valence issues. For example, 92% of respondents support the need to guarantee adequate energy costs, 90% support the fight against unemployment, whilst 82% consider the need to reduce global warming a top priority. At the same time, some questions about the environment, the management of the energy crisis, and foreign policy are highly divisive, providing the opportunity for political parties to differentiate their political offering from their competitors. In fact, on these issues, voters' concerns are expressed through support for alternative policy goals. On the environment, voters are divided on the trade-off between environmental protection and economic growth, whilst, in foreign policy, the electorate is divided on the economic sanctions against Russia and the provision of arms. Based on these premises, our main expectation is: Proposition 1: In 2022, Italian political parties devoted a high share of promises towards issues related to the environment, defence, and energy sectors.
Nevertheless, other factors can help us explain why parties decide to prioritize one issue instead of others. Firstly, being in government or opposition might affect political parties' electoral engagement (Schumacher et al, 2015). We, therefore, posit that the inter-party dynamics that characterized the XVIII legislature will also be reflected in the policy areas towards which Italian political parties oriented their commitments.
We first look at the main opposition party during the Draghi government, Brothers of Italy.
Throughout the entire mandate, FdI sharply criticized the government's action, especially regarding the economic measures adopted to deal with rising gas prices, and has repeatedly called for a redefinition of EU-funded PNRR. FdI's members were indeed no strangers to harsh statements directed at the previous government both in Parliament and in the media and at its 'failures' and 'inability' to solve the current main problems for citizens, as the ones struggling with paying utility bills. It is therefore reasonable to expect FdI to continue to highlight alternative policy solutions and focus its campaign promises on the energy sector. At the same time, during the electoral campaign, FdI's leader Giorgia Meloni, favoured at the polls, tried to depict herself as a viable governing candidate by, among other things, underlying her Atlanticism and support for the Ukrainian resistance. It follows that: Proposition 2: We expect Brothers of Italy to emphasise energy and defence-related promises in its electoral program.
Turning now to the parties that supported the Draghi government, we expect different trends depending on the party.
Until the end of the legislature, the Democratic Party has consistently showcased unwavering backing for the cabinet's priorities and agenda. The party solidly stood in support of Ukraine and was in favour of the measures taken by the Draghi government to deal with the adverse effects of the energy crisis.
It can be therefore anticipated that the Democratic Party would opt for a programmatic agenda in line with these measures. However, policy proposals in continuity with Draghi's energetic policies could be electorally risky. As the Draghi government's measures did not effectively solve all existing problems, resulting in citizens still grappling with the negative repercussions of the energy crisis during the electoral campaign, we might expect the PD to downplay these issues in its program to avoid being held accountable for this negative performance (see Schumacher et al, 2015). At the same time, from its privileged position as an opposition party, FdI managed to successfully make energyrelated issues its own, making it harder for the PD to defend its stances. Our third expectation is: Proposition 3: We expect the Democratic Party to emphasise energy promises less while committing to a high extent to the defence sector.
All other three main parties supporting the Draghi government (M5S, L, FI) have adopted a critical attitude towards some of the policies implemented during his mandate. The Draghi cabinet faced significant criticism from the Five Star Movement primarily due to its handling of the energy crisis, its decision to send weapons to Ukraine, and the (perceived) lack of environmental policies, which was the first step that led to the government's resignation and early election (Garzia, 2022). The M5S should consequently emphasise its alternative policy proposals during the electoral campaign. At the same time, as Conte tried to position his party within the new multipolar configuration (Garzia, 2022), M5S is expected to highlight its pacifist and environmental stances in order to differentiate itself from the PD-led centre-left bloc. It follows that: Proposition 4: We expect the Five Star Movement to devote a large number of promises to the environment, energy, and defence.
The League and Go Italy faced electoral setbacks due to their involvement in the grand-coalition government, with their voter base shifting towards FdI (Chiaramonte, 2023), eventually leading them to exit the governing coalition. In light of the various criticisms directed at Draghi's agenda and at its powerlessness to prevent energy costs from escalating, they should not propose policy measures in line with the previous cabinet. At the same time, they are expected to not pay too much attention to alternative energy policy proposals for two reasons. They may, firstly, decide to minimize their commitment towards energy-related matters to avoid appearing less credible as being former members of the government. Secondly, being part of the centre-right electoral coalition, they need to differentiate themselves from the partners in order to increase their vote share and, thus, their relevance in the alliance. While FdI is expected to focus more on energy issues, L and FI should emphasise other policy domains, such as reforms to the tax system or control over immigration.
Concerning defence-related matters, both L and FI had a pro-Russia and, more importantly, a pro-Putin history, hard to forget and even harder to reconcile with the need to take a clear stand on the Russian-Ukrainian war. We propose that: Proposition 5: We expect the League and Go Italy to emphasise less defence and energy promises.
Finally, considerations on issue ownership lead us to expect single-issue parties to campaign mostly on their key policy area. In particular, we expect: Proposition 6: The Green and Left Alliance to have the highest share of electoral pledges related to environmental issues.

Data
For this paper, we collected data on campaign promises included by Italian parties and coalitions in their election programs and, more specifically, the topics touched on by these promises. Differently from other studies that look at speeches (Hobolt & Klemmemsen, 2005), press releases (Klüver & Sagarzazu, 2016;Meyer & Wagner, 2016) or party manifestos (Wagner & Meyer, 2014;Abou-Chadi et al, 2020;Spoon & Klüver, 2015;Green-Pedersen, 2007), we aim to explore the issues (and their shifts) in which parties concretely engaged themselves in front of their rank-and-file voters and the general public during the last election. Accordingly, we argue that, in addition to exploring what partisan actors talk about and discuss in any election campaign, it is also important to consider the actual policies they promote and the policy domains to which they commit themselves.
Politicians can convey their promises to the electorate through different platforms, from public rallies and interviews to campaign speeches and political debates. In this analysis, we focus on the promises written down in the policy programs presented during the electoral campaign. While the policy proposals put forward in oral declarations could also be the result of a spur-of-the-moment reply to a comment or an attack, or the expression of a personal preference of a single politician, election programs are programmatic documents whose drafting process usually involves a discussion between the different party officials, parliamentary actors, core supporters, experts, and interest groups (Dolezal et al, 2012;Harmel, 2018). Manifestos and the election pledges contained in them can be therefore argued to be the expression of the official party line and its policy preferences.
To identify all the election pledges contained in the campaign manifestos, we manually select all the statements that correspond to a pledge. To this end, we employed Royed's (1996) definition, which is the one commonly used by the pledge literature: A promise is thus 'a commitment to carry out some action or produce some outcome, where an objective estimation can be made as to whether or not the action was indeed taken or the outcome produced' (Royed, 1996, 79). To be considered as a 'pledge' a sentence (or a paragraph) should (1) refer to a policy measure the party wants to adopt or a goal to attain, and (2) its fulfilment should be testable (see Naurin et al, 2019). The majority of the promises selected are introduced by expressions such as 'we want to', 'we plan to', 'we need to', and 'we will adopt', or contain terms like 'we will achieve' or 'our goal is'. The final version of the dataset leaves out also several international policy proposals. Indeed, the fulfilment of promises such as '[we will promote] a European common army' 3 is not due to governments' willingness, the presence of financial and political resources, nor a favourable institutional context. Several additional factors play a major role in the international and European context. The success (or the failure) of these proposals cannot be directly ascribed (or blamed) to the Italian government. Additional information on the data collection process can be found in the Appendix.
Overall, for the 2022 election, in this study, we analyse 3,992 pledges included in the campaign programs of the eight main competing parties and/or coalitions ( Table 1). As Action and Italy Alive decided to run in the election together and to promote a joint agenda, the coalition manifesto is the only document analysed for both parties. Similarly, the Greens and the Italian Left both rallied around the program presented by their pre-electoral coalition. Since the aim of this analysis is to also observe differences in issue engagements between parties, for all other political forces, we look at the election programs of individual parties even when an electoral alliance was formed. Brothers of Italy, Go Italy, and the League, for instance, deposited a shared program for their centre-right coalition 4 , but here we only focus on the manifestos written by each coalition partner. Election promises were then grouped depending on the type of policy issue they refer to. Categories were created on the basis of the Comparative Agendas Project (CAP) codebook for Italy. Differently from the Manifesto Project, whose primary goal is to assess parties' ideological preferences, the coding rationale behind the CAP is to identify the actual content of the sentences. Each promise was classified according to the policy dimension addressed, not the instrument or its goal. In this way, these categories capture the issues of the proposals promised by the different parties/coalitions during the election campaigns. Examples are provided in the appendix. The original 21 categories of the CAP have been recoded into 17 policy issues (Appendix 1).

Results
Policy issues of the promises in the 2022 election campaign   The second policy domain in which Italian parties engaged themselves during the electoral campaign of August-September 2022 is related to social, health, and welfare issues. Overall, 14.0% of the total promises analysed were included in this category. In this case, excluding the Five Star Movement, the percentage of social promises differs less according to the party or the coalition than economic promises: Italian parties seem to all pay a similar level of attention to social issues. It is particularly noteworthy that the lowest percentage of social promises (9.8%) is found for the M5S since Conte had run his electoral campaign focusing on fighting inequalities, preserving the Reddito di Cittadinanza (citizenship income), and implementing other social measures. However, while promises related to these questions were included in the M5S's electoral program, the party devoted a lot of space also to other issues, such as security, agriculture, transportation, and education matters.
Issues related to the justice system, security, and civil and penal code (see the category Law and Order) had an average percentage of 10.5 points, which makes them the third most-discussed topic within the 2022 campaign. Looking at Figure 1, there is a marked difference between parties and within parties on the same side of the political spectrum. Specifically, being usually classified as a more right-wing issue, we would have expected right-leaning parties to commit to law and security policies measures and aims more than left-leaning ones. However, this only applies to the League, and, to a lesser extent, Go Italy. While the League had the highest percentage of law-and-order promises made (14.1%), the second largest percentage corresponded to +E (13.0%). The level of pledges in the FdI's manifesto fell considerably short of the average (8.5%), a rate similar to the M5S's one (8.6%). It is particularly noteworthy that the main right-wing candidate in the 2022 election (Meloni's FdI) did not prioritise these types of issues in its program, but rather favour other matters such as the economy, social issues, education, government and administrative concerns.
Following the riding-the-wave approach, our expectation is that, since the election was held right in the middle of an energy crisis and that the political debate was dominated by rising energy costs, proposals for dealing with the crisis and becoming independent from the Russian gas, and the inevitable criticisms towards past government actions Improta et al, 2022), parties should have increased the emphasis on those promises as well. Instead, only 6.5% of the total number of pledges pertained to energetic issues. The topic was particularly salient for the League and AVS, whose percentages of 9.6% and 8.7%, respectively, exceed by far the average rate. The percentage of energy promises by FI was at the average level (6.3%), but for FdI and, especially, M5S and PD it remained far below (4.9%, 3.7%, and 2.5% respectively). Differently from our expectations, FdI had a very low share of energy promises (Proposition 2) and FI and L made significantly more promises in this sector than we anticipated (Proposition 5). The low attention to these promises by PD is, on the contrary, consistent with our assumption (Proposition 3).
A similar discussion can be made for environmental promises. The need to fight global warming was listed among the top priorities for Italian voters in the 2022 general election (Improta et al, 2022).
However, once again, except for FI (6.3%) and -in accordance with our expectation (Proposition 6) -the Greens-Left alliance (19.7%), all remaining parties did not devote too much attention in their manifestos to this issue: On average, only 5.4% of the policy proposals were related to the environment. In particular, the Five Star Movement does not seem to prioritise its environmental stances to differentiate its position from the centre-left coalition (see Proposition 4), while Go Italy appears to have emphasised this topic more than its coalition partners.
Defence is another policy domain to which we would have expected Italian party manifestos to pay more attention given that the Russian invasion of Ukraine and discussions on providing weapons to Ukraine were a central part of the 2022 political campaign Improta et al, 2022). Instead, Figure 1 shows very low percentages of defence promises in each party manifesto analysed (0.9% on average). Only one defence promise ('No to the arms race') was included in the M5S's program, while PD and +E had not even made a single promise related to this policy dimension. Taken together, these data do not corroborate our expectations regarding defence promises: FdI, PD, and M5S did not make more promises than FI or L in the defence sector (see Propositions 2, 3, 4, and 5).
All main Italian parties seem to place particular emphasis on education and government issues in their 2022 programmatic agenda. Immigration, civil and political rights, transportation, housing, technology, international affairs, agriculture, public resources, and culture are among the policy domains in which Italian parties committed less in the last general election. 5 In short, Figure 1 shows that Italian parties emphasised economic, social, and security issues in their election programs. While differences exist among parties, environmental, energy, and defence promises did not receive the share of attention we expected. If we look at the average data, pledges related to energy issues came only sixth on the list of priorities, the environment eighth, and the defence second to last. Conclusions do not change when single parties are considered. Except for the Green and Left Alliance, the vast majority of campaign proposals made by the main competitors in the 2022 Italian general election are not related to the energy crisis, the environment, or the conflict in Ukraine.
Socio-economic questions are the key aspects around which the electoral competition is played. One might assume that the negative repercussions on the state of the economy and labour market generated by the war in Ukraine, the European sanctions against Russia, and the consequent inflationary pressures -which come on top of ongoing socio-economic distress triggered by the Covid-19 pandemic -have spurred parties to increasingly commit in these sectors. It is striking, however, that, in addition to economic and social issues, Italian parties emphasise other policy domains, such as security, government, education, and infrastructures, which appear to have nothing to do with the current situation of the country. In contrast to our main expectation, the vast majority of commitments parties take do not seem to be related to the issues that were widely discussed during the electoral campaign or that were considered most salient by citizens.
There might be two possible explanations for this finding. The first one could be related to the different functions of party programs. As discussed in the theoretical section, parties write their electoral manifestos to fulfil different purposes, from attracting votes to providing a unitary platform for the party and a coherent governing agenda for elected officials (Harmel, 2018;Dolezal et al, 2012). Election promises are not only made with the main goal of reflecting voters' concerns and preferences and providing answers to the most salient problems. To cite an example, the issue of the introduction of a presidential system in Italy was the only government-related question that received some attention from citizens during the 2022 election, but it was by far not among their top priorities (see data in Improta et al, 2022). Nevertheless, many promises pertaining to the political system, public administration, bureaucracy, and federalism were included in parties' manifestos. Parties are used to include these types of policy proposals in their programs, even though they are not the issues voters are most concerned about. A similar discussion can be made for matters related to security and judiciary powers. Also in this case, other goals -e.g. parties' desire to implement some reforms that are deemed important (Naurin et al, 2019) -might push partisan actors to take on many commitments in not so (at least, apparently) salient policy areas.
A second potential reason explaining the low emphasis paid to key proposals could be attributed to substantial differences among the policy domains under consideration. Education-related promises, for instance, deal with primary and secondary education, universities, academic research, teachers and professors; In other words, sectors in which parties can put forward alternative policy measures and goals, and in which office-holders have more power to act. In other policy domains, on the other hand, the concrete policy options and/or the possible goals might be limited, because solutions have a more technical rather than political nature, or because decisions have to be made at a non-domestic level. As a result, parties might always devote less space to these types of promises in their manifestos.
We should not indeed forget that, whatever reason pushes parties to include a certain policy within their program, at the end of their mandate they are evaluated for their capacity to fulfil their promises, among other things (Matthieß, 2020).
It is important to acknowledge that a specific issue (i.e. defence) might only apparently seem less important when we restrict our assessment to comparing the number of promises in that particular sector with the electoral commitments made in high-priority policy areas (i.e. economy). To rule out this possibility, in the next section, we compare the share of promises made in one sector with the promises made by the same political parties in the same sector during the electoral campaign of 2018.
In this way, we can investigate whether, even though the environmental, energy, and defence pledges did not receive the emphasis we expected, their share increased from the previous campaign.

Electoral promises in 2018 and 2022: a comparison
In this section, we compare the percentage of promises made in 2022 with data from 2018 (Appendix 2). We consider the main five parties that contested both elections (Figure 2). Differently from the previous analysis, we merely focus on the issues that received more attention in 2022 (namely economic and social issues, law and order, and government), plus the three issues we expected to be salient (environment, energy, defence). Unlike in 2022, environment, energy, and defence-related issues did not dominate the 2018 electoral competition, which, in turn, was primarily centred around immigration and EU matters (Giannetti et al, 2018). The M5S, which since its early formation presented itself as an environmentalist party (Pirro, 2018), drastically reduced the emphasis paid to this issue: Its percentage of environment promises dropped from 21.4% in 2018 to 2.5% in 2022. Similarly, the League which devoted 9.4% of its electoral promises to environmental issues in 2018 reduced its commitment to just 2.6% in the 2022 election.
Brothers of Italy presented a slightly higher percentage in 2022, but the only party to substantially increase its share of promises dedicated to the environment compared to 2018 is FI: Its share of promises rose from 0.8% to 6.3%.
Similarly, in light of the war in Ukraine, we would have expected Italian political parties to pay more attention to defence promises in the 2022 election. However, their average percentage in 2022 (0.9%) is even lower than in 2018, when the share of defence promises was 1.0%. It is even more surprising that the parties most in favour of providing military support to Ukraine through the provision of arms dedicated little space to defence promises in their electoral programs. Whilst FdI slightly reduced its commitment from 1.0% to 0.9%, the PD dedicated no promises to the defence issue in its 2022 electoral program (while in 2018 it devoted 0.6% of its total number of electoral promises to this topic). The reduction in the M5S's commitment, which dropped from 2.4% to 0.6%, can be interpreted in light of the growing criticism expressed by the party regarding the deployment of arms to Ukraine.
Only the League slightly increase the percentage of promises related to defence issues.
While the space reserved for both environmental and defence promises in 2022 parties' electoral programs is reduced compared to 2018, the same is not true for energy-related promises. In this case, the average percentage went from 5.0% in 2018 to 6.5% in 2022. This increase is, however, mainly driven by two right-wing parties, FdI and L, which significantly made a higher share of energy promises compared to 2018. The percentage of FdI's energy promises raised from 1.0% to 4.9%, while the League's passed from 2.7% to 9.6%. As expected, being critical of the Draghi government's energy policy and measures taken to deal with rising energy costs, the party led by Giorgia Meloni increasingly emphasised its energy-related proposals. PD, M5S, and FI all reduced the percentage of promises devoted to energy issues compared to 2018. Once again, the salience of the topic due to the energy crisis and price increase triggered by the invasion of Ukraine and the sanctions against Russia (Improta et al, 2022;Chiaramonte et al, 2023) does not seem to boost the emphasis attributed to the topic by Italian parties and, in particular, the share of energy-related policy measures promoted.

Discussion and conclusions
The aim of this study was to examine the policy issues in which Italian parties concretely engage during the 2022 general election. To this end, we collected data on the pledges contained in those programs published by parties and pre-electoral coalitions. Overall, by looking at the issues of the policy proposals made during the electoral campaign, we find that Italian parties seem to compete around the same issues. Only in very few cases do they show sharp differences in the emphasis attributed to different policy dimensions (see for instance the categories 'law and order' and 'government' in Figure 1). The economy is the issue in which all parties -with the exception of the  (Petrocik, 1996;Dolezal et al, 2014;Wagner & Meyer, 2014), political parties do not seem to increase the share of electoral engagements related to the issues they enjoy ownership in, at least as long as the topics of campaign promises are considered. Additionally, no policy domain stands out as a more left or right-wing issue or as a specific hallmark for one pre-electoral coalition. While FI and M5S both made a high share of promises covering the same policy areas, security measures and education matters are central in several manifestos across the political spectrum. Clearly, these findings do not suggest that ideologically-different politicians promote the same measures or that their policy positions are now closer, but rather that Italian parties do not seem to compete by taking commitments on a different range of topics. The case of AVS deserves a separate discussion. The programmatic agenda supported by the Greens, Italian Left, and other environmentalist political actors, in line with their issue ownership and priorities, mainly promoted policies dealing with environmental issues, renewable energies, sustainability, and zero-waste strategy.
In the theoretical section, we argued that, as parties compete in the election campaign by proposing policy solutions to the country's main problems and by responding to their electorate's concerns (Klüver & Sagarzazu, 2016), their promises should primarily address salient issues. Since existing studies have already established that parties do indeed focus more on the issues voters deem as most relevant (Klüver & Sagarzazu, 2016;Ansolabehere &Iyengar, 1994), we explored whether parties adapt their election promises to the socio-economic context as well. Data from Italian parties do not support this expectation. Even though the 2022 electoral race was marked by rising energy costs and the war in Ukraine, the percentage of promises related to the energy industry, the environment, and the defence sector was not very high and it even decreased compared to the 2018 election. Moreover, in contrast with our expectation -with only a few exceptions (i.e. energy promises made by FdI)being in government or in opposition as well as strongly criticising past government's actions do not seem to affect election promises.
These findings seem to suggest that party competition dynamics in multi-party settings cannot be adequately framed only relying on the salience of the issues or considerations on the credibility political actors enjoy in the eye of the electorate. A potential explanation could come from the issue yield theory debate (De Sio & Weber, 2020). Political parties strategically avoid engaging in those issues that are not profitable from an electoral perspective. This could be attributed to the lack of a cohesive electoral base on specific issues, making it risky for political parties to address those concerns as it may result in internal divisions and a loss of existing electoral support. An alternative explanation could come from the nature of the sectors in which political parties engaged less than expected. The potential remedies to address the rising gas prices or the conflict in Ukraine, which were among the main concerns of the electorate, are surely influenced by decisions made at the supranational level. Thus, political parties find themselves with constrained opportunities to take action within their own country in these policy domains. Therefore, to avoid being punished by the electorate for failing to fulfil their election promises, parties might avoid committing themselves to these issues.
Clearly, in this study, we merely focus on the Italian case and especially on the last general election in Italy. However, we would expect to find similar dynamics in other European countries as well, given the centrality of matters arising from the war in Ukraine, the energy crisis, and the EU-funded recovery package across Europe. Probably, if the prominent issues had been related to political domains where parties can implement tangible solutions at the domestic level, we might have achieved outcomes that align more closely with our expectations. By extending the analysis to encompass other multi-party systems, we can determine whether the observed pattern is unique to Italy and these specific policy areas or if it applies to other countries as well.
Results of this study contribute to both literatures on issue salience and party responsiveness. While previous research has provided evidence that parties devote a lot of attention in their annual Parliamentary speeches (Hobolt & Klemmemsen, 2005), press releases (Klüver & Sagarzazu, 2016 problems of utmost importance and, more specifically, to the issues they enjoy ownership in, the same is not true for election pledges. All in all, it is clear that talking about a relevant issue and making concrete commitments to adopt some policies or to achieve a specific goal are not the same thing and require different kinds of effort from political actors. Further research would therefore benefit from accounting for this difference and taking into consideration parties' policy proposals as well when investigating the issues political actors emphasise.