Microsoft Word - PDF_Issue_16_2_Ceron_Zarra.docx Italian Political Science, VOLUME 16 ISSUE 2 Published in 2021 under Creative Commons BY-NC-ND license by: Italian Political Science. ISSN 2420-8434. All copyrights of article contents are retained by the authors. Volume 16, Issue 2, 74–99. Contact Author: Matilde Ceron, University of Pavia. E-mail address: matilde.cer@gmail.com Women’s substantive representation in the pandemic: assessing the parliamentary debates on the Italian Recovery Plan Matilde Ceron UNIVERSITY OF PAVIA Antonella Zarra INSTITUTE OF LAW AND ECONOMICS, HAMBURG UNIVERSITY Abstract During the Covid-19 emergency, women have been largely under-represented in decision-making concerning the health crisis and the recovery effort. This dynamic complements growing pressure from women’s interest groups against the scarce attention to issues relevant to the achievement of gender parity in the pandemic. As a result, the outbreak and its response raise the question of the importance of women’s representation for the saliency of policies directly supporting their empowerment. The parliamentary debate on the Italian National Recovery and Resilience Plan (NRRP) offers a well-defined case for assessing the relevance of substantive representation in Italy within the Covid-19 recovery. Position papers and proposals from women’s interest groups vocally campaigning for parity offer the benchmark against which we code over 200 parliamentary speeches for their gender parity-relevant content. The analysis compares the different prevalence of parity-relevant issues across women and men Members of Parliament (MPs), accounting for ideological differences. We argue that representation matters for women’s empowerment policies, as women policymakers within the Parliament, are those primarily raising such concerns in parliamentary debates over the Italian NRRP. The paper contributes to the extant literature on women’s representation by evidencing a stark gender gap in the saliency of parity-related issues. 1. Introduction he Covid-19 pandemic has posed an abrupt halt to, if not reversal of, the improve- ment of gender parity in many domains of our society. Extensive evidence has emerged of the detrimental effect of the outbreak and prolonged lockdown on women, who have endured increased episodes of violence and additional overwhelming care responsibilities deriving from school closures and work-from-home arrangements (Adams-Prassl et al., 2020; Boca et al., 2020; Mongey & Weinberg, 2020; Alon et al., 2020). From this perspective, Covid-19 has been claimed to represent a tale of two pan- demics from a gendered perspective (Ceron and Zarra, 2021). By the same token, women have been scarcely represented in the management of the health crisis and the subse- quent reconstruction phase. Henceforth, the pandemic provides a critical case to assess the contribution of women in putting forward parity-relevant issues in the policy debate. T CERON and ZARRA 75 The topic of substantive representation of women, through analysing to what extent the involvement of women leads to more equal policies, is particularly salient because of the unique opportunity offered by the National Recovery and Reconstruction Plans (NRRPs). On the one hand, gender parity and its mainstreaming are among the mandated guide- lines Member States are committed to in the drafting of national plans within the scope of Next Generation EU (NGEU),1 with the explicit acknowledgement of the additional toll the pandemic has imposed on women. On the other, the national implementation of Next Generation EU through the NRRPs, mobilizes an unprecedented magnitude of invest- ments, especially in a country like Italy. As a result, the NRRP is well-suited to deliver substantial progress toward gender parity, raising the question both of whether such an opportunity was fully capitalised on in national choices over the plan, and the role of women in pushing towards such an objective. Additionally, the discussion on the NRRP is a well-delimited policy-debate in comparison, for example, to the Covid-19 response measures. Nevertheless, the policy process in Italy saw substantial changes in key actors with the end of the Conte government and the inception of the Draghi premiership, under which auspices the final version of the plan was submitted to the European Commission. The parliamentary debate over the plan, while not per se fully reflecting the policy out- comes, presents a stable and distinct arena in which to assess the saliency of equality and its gender divide. Against this background, our work considers the prevalence of parity-relevant issues among parliamentary debates on the key document preceding the official presentation of the Italian NRRP. By manually coding 226 parliamentary speeches from four plenary de- bates between October 2020 and April 2021, we conduct a text analysis to study women’s substantive representation. The work scrutinises the main parliamentary debates relat- ing to the three key documents reflecting the evolution of the NRRP, from the initial guidelines submitted to Parliament by the Government to the Conte and Draghi iteration of the NRRP. We derive the key parity-relevant issues from the position papers and hear- ings of women’s interest groups, proceeding then to code accordingly over 200 parliamentary speeches. The focus on parliamentary debates, in line with a stream of the literature on substantive representation (e.g. Wängnerud, 2006), does not contemplate the impact of the gender of Members of Parliament (MPs) on policy outcomes, but rather it measures gender gaps in issues relevant to parity raised in MPs’ speeches. We then compare the saliency of equality measured by the prevalence of parity-relevant issues across the gender divide overall and within parties. On such premises, we provide descrip- tive statistics and run a logistic regression to assess whether the gender of MPs and their political group membership play a role on the extent to which their speeches include any parity-relevant issue. Specifically, we test whether gender differences arise in the preva- lence of equality concerns and whether they remain robust at the party level. Our hypothesis is that parity-related issues are more prevalent among the speeches of women MPs, in line with the expectation that descriptive representation matters substantively. We choose not to refer to our dependent variable as ‘women’s issues’ but rather ‘parity- relevant issues’, as it includes all policies with implications for gender equality (e.g. school 1 Next Generation EU, the common post-pandemic reconstruction initiative, is an umbrella term for measures and funding sources, whose main component is the Recovery and Resilience Facility (RRF) which finances nationally drafted NRRPs subject to their approval at the EU level. Women’s substantive representation in the pandemic 76 and early childcare intervention). In the same vein, in our analysis we refer to ‘women’s empowerment’ as ‘the process by which women gain power and control over their own lives and acquire the ability to make strategic choices’ and to ‘gender mainstreaming’ as ‘a strategic approach to policy-making that aims to achieve equality of opportunity be- tween women and men in all spheres of society and to integrate a gender perspective into policy-making activities’.2 The analysis shows that women play a central role in advocating for gender main- streaming within the Italian parliamentary debate in the NRRP, corroborating the argument that their under-representation negatively affects progress on gender parity within the policy agenda. Results confirm that the limited overall saliency of gender par- ity is driven predominantly by women MPs, with a gender divide that extends to most political parties regardless of their different overall sensitivity to equality. The contribu- tion of the analysis to the extant literature on women’s representation is twofold. First, it pinpoints the relevance of substantive representation in the parliamentary debate over the Italian NRRP. Second, it indicates whether and to what extent women’s representa- tion (under-representation) may foster (hinder) the inclusion of parity-relevant issues within the parliamentary debate on the NRRP, supporting further analysis of the broader implications for the prospects of prioritisation of gender equality in the overall policy- making process outside Parliament, as well as the post-pandemic reconstruction at large. Our in-depth assessment of the Italian case stresses how women’s politicians and de- scriptive representation within Parliament may be crucial for advocating more equal policies. The derived implication is that women’s empowerment in the aftermath of the health emergency heavily hinges on women's voices and their representation. The work is in line with cross-country studies of gender equality in the NRRPs indicating that against the objective of gender mainstreaming mandated by NGEU, parity plays a mar- ginal role (Zarra and Ceron, 2021). Such dynamics raise particular concerns for countries relatively lagging behind in terms of gender parity, such as Italy within the EU27. 2. Substantive representation: expectations for parity-relevant issues The representation of women in parliaments and other decision-making bodies is increas- ing worldwide (Wängnerud, 2009). In the last elections, for a few seats, Iceland missed the chance to become the first European country with a majority of women in the parliamen- tary seats (‘Iceland Misses out on Europe’s First Female-Majority Parliament after Recount’, 2021). On average, as of October 2021, the proportion of women in national par- liaments is 25.8% globally, with an increase of more than 6 percentage points in 10 years (IPU, 2021). This upward trend can be attributed to several factors, including (i) the transi- tion to proportional electoral systems (Rule, 1994), (ii) the inclusion of gender quotas in constituencies, and (iii) the early empowerment of women (McAllister & Studlar, 2002). More broadly, the upward trend in welfare state spending as a percentage of GDP recorded in the past half-century also plays a role in achieving a more balanced political 2 Both definitions are retrieved from the taxonomy provided by the European Institute for Gender Equality (EIGE). See: https://eige.europa.eu/taxonomy/term/1185 and https://eige.europa.eu/taxon- omy/term/1102 . CERON and ZARRA 77 representation: more welfare spending leads to a better socio-economic status for women, which enhances their chances of getting elected (Krook, 2010; McDonagh, 2010; Rosen- bluth et al., 2006; Siaroff, 2000; Thames & Williams, 2010). A last relevant driver can be traced to changing social and gender norms which, along with increasing pressure from women's interest groups, have been contributing to the more meaningful participation of women in politics (Wängnerud, 2009). When it comes to the theorisation of women’s representation, the literature tradition- ally distinguishes between descriptive and substantive representation, where the former concentrates on the proportion of women elected and the latter on the effects of more bal- anced parliaments, investigating to what extent the involvement of women in the war room leads to more equal policies and better democracies. Wängnerud (2009) argues that the re- nowned theory of the politics of presence by Phillips (1995), which suggests that women politicians are better equipped to represent women’s interests as they – at least to some ex- tent – share the experiences of other women, constitutes the link between the two types of representation. To provide an example of such a connection, countries’ and parties’ imple- mentation of quotas contributes to the enhancement of ‘the quality of decision making on a substantive level’ (Celis, 2006). While the research on descriptive representation enjoys a longer and richer tradition, it is only recently, thanks to a greater number of seats held by women MPs, that the scholarship on substantive representation has flourished, producing extensive empirical evidence. However, it is worth emphasising that the study of substantive representation is strictly intertwined with key concepts such as ‘gender equality’ and ‘women’s interests’ and their definition and meaning; hence any related study should also be evaluated in light of the connotation of such terms given by the authors (Wängnerud, 2009). For instance, the very concept of women’s interests is questioned by some scholars, who argue that their def- inition should not be provided in a top-down fashion, but rather subjectively determined by women themselves (Celis, 2006). All in all, however, any attempt to catalogue what women’s interests are and how gender equality should be understood results in overlapping lists of topics, ranging from the division of paid and unpaid labour, women’s exclusion from political and economic leadership, sexual harassment and gender-based violence, and fam- ily and social policies (Phillips, 1995). Because of such a context, the approach we follow in our analysis consists in inferring key themes of relevance to gender parity from women’s interest groups which, as further elaborated in the section to follow, contributed to the de- bate over post-pandemic management. When it comes to the evidence on the extent to which the number of women elected affects their interests, existing empirical works suggest that, despite a wide variation across regions of the world, there is a positive correlation between countries’ overall performance in gender equality and their increased proportion of women in decision-making bodies. A stream of the literature suggests that more elected women lead to significant advancements in policies that specifically concern them (Schwindt-Bayer, 2006; Thomas, 1991), such as family policies (Wilensky, 1990) or abortion legislation (Berkman & O’Connor, 1993). Also, higher proportions of women in parliaments translate into more laws that benefit children (Besley & Case, 2003; Lijphart, 1991). Wängnerud (2006) investigated MPs’ priorities in Sweden across a 20-year timeframe and found that women politicians were more likely to prioritise issues belonging to domains of social policy, family policy and care for the elderly Women’s substantive representation in the pandemic 78 compared to men politicians. Observational studies from several EU countries show that the introduction of gender quotas was associated with measures supporting maternal em- ployment and work-family policies (Weeks, 2019). In the same vein, elected women are more likely to interact and have contact with women’s organisations outside Parliament. Looking at quasi-experimental settings that allow for the identification of causal rela- tionships between women elected and policy outcomes, the evidence is still limited in terms of geographical coverage and does not follow clear patterns (Hessami & da Fonseca, 2020). The randomized assignment of gender quotas in India allowed scholars to identify the effect of women’s leadership in shaping policies in areas such as health and education (Beaman et al., 2006; Chattopadhyay & Duflo, 2004). At the local level, research on municipalities in Italy (Baltrunaite et al., 2019), Spain (Bagues & Campa, 2021) and Norway (Geys & Søren- sen, 2019) found no significant effect of women’s representation on public spending. Two recent contributions, however, showed that instead of focusing on public expenditure, the impact of substantive representation can be demonstrated by uncovering dynamics within the policy process and looking at women politicians’ behaviour. In France, women politi- cians were more likely to draft amendments on parity-relevant topics than men (Lippmann, 2020). At the regional level, in Germany, a study analysing council meetings found that the presence of women led to increased discussions on childcare and the in- creased participation of other elected women (Hessami & Baskaran, 2019). Conversely, however, opposite results from another stream of scholarship have stressed that more women in parliaments can lead to detrimental effects on their substan- tive representation. Hawkesworth (2003) found that the increased number of women MPs negatively affected the behaviour of men MPs, who act as bottlenecks for the access of women to positions of power within the assembly as well as for policies favourable to women. In the same vein, Carroll (2001) argues that more women enhance heterogeneity in the parliamentary forum, thus pushing other women MPs to focus on other (not gen- dered) policy subjects, thinking that others will take care of women’s issues. Kathlene (1994) found that as the number of women increased, men MPs became more aggressive and obstructive towards legislative proposals by women. Crowley (2004) pointed out that fewer women in parliaments can be more effective in representing women’s issues since they look less threatening. Finally, other research highlights that women politicians may be more prone to work on women’s issues because men MPs tend to dominate the other topics (Heath et al., 2005; Schwindt-Bayer, 2006). Hence, particularly in a delicate moment such as the Covid-19 crisis, it is of sheer in- terest to assess to what extent women’s participation in the policy process is key to advocating for parity-relevant issues and gender mainstreaming. 3. Covid-19 and its decision-making through gendered lenses The Covid-19 crisis has negatively impacted progress toward gender equality by dispro- portionately affecting women, who have suffered the harshest consequences of the recession in areas that go far beyond the labour market. The pandemic has caused a re- duction in women’s employment rates in many advanced economies, including Italy, where a real ‘she-cession’ is taking place, with 72.9% of the jobs lost in the country in 2020 being women’s (Lippmann, 2020). Moreover, being the main caregiver of the household in a country where stereotypical gender norms are still predominant, women CERON and ZARRA 79 experienced increased care duties due to the closure of schools and childcare institu- tions. A survey carried out during the first wave of the outbreak confirmed that in Italy work from home arrangements and housework responsibilities fell mainly to women, although men enjoyed more time with children to the benefit of more gratifying family work (Del Boca et al, 2020). In addition, by forcing millions of individuals to stay at home, the Covid-19 emergency contributed to more domestic abuse, with a drastic in- crease in the number of reports to the police authorities and helplines: calls reporting abuses increased by 72% in March and April 2020 compared to 2019 according to the Ital- ian National Statistics Institute (ISTAT). Against the context of such detrimental dynamics, the scarce representation of women in the decision-making bodies ruling on the measures to contain the contagion and provide economic relief to the population, particularly in the early months of the pandemic, revamped the debate on women’s leadership and its substantive effects on policies. Though compared to men they tend to perceive the pandemic as a serious prob- lem (Galasso et al., 2020), and although they have been lauded for their collaborative and collective managerial style (Hong Fincher, 2020; Zednik, 2020), their inclusion in the strategic management of the pandemic was relatively limited. Thus, similarly to previ- ous crises such as the Great Recession (Kantola & Lombardo, 2017), the scarce representation of women at leadership levels may have impinged on the presence of re- covery policies that could directly cater to their needs. In the Italian case, women have been excluded from decision-making bodies estab- lished specifically for Covid-19. During the spring of 2020, the former government chaired by Conte established two main strategic bodies in charge of containing the con- tagion while designing post-pandemic measures. On the one hand, the Scientific Technical Committee (Comitato Tecnico Scientifico) advises the head of the Civil Pro- tection Department on the adoption of prevention measures necessary to cope with the spread of the virus. On the other hand, the Committee of Experts in Economic and Social Issues (Comitato di esperti in materia economica e sociale, hereinafter Colao Task Force) was established in the spring of 2020 to lead the post-Covid reconstruction. The former body was made up exclusively of men while the latter included 17 members, of which only four were women. The striking gender imbalance within the task forces led to protests from women’s interest groups and politicians and the creation of petitions and the organisation of flash mobs such as ‘Dateci Voce’, with which women from civil society asked the government to restore gender parity in the composition of advisory bodies. After the movement's ap- peal, the former head of government intervened by supplementing the groups with 11 women. The initial severe imbalance may suggest that gender parity and women’s rep- resentation in crisis management were not a priority for the Italian leaders given the highly men-dominated 16 task forces convened. The only exception was the task force ‘Women for the Renaissance’ chaired by the Minister for Gender Parity Elena Bonetti, launched by the Italian Department for Equal Opportunities, composed by a team of women entrusted with proposals for enhancing gender equality. These proposals put at the forefront women’s representation across all domains within the workforce – includ- ing in STEM – with a pillar devoted to women’s leadership and its monitoring through an Observatory on Gender Equality tasked with gender impact assessments and the Women’s substantive representation in the pandemic 80 extension of gender quotas in a broad array of bodies (Dipartimento per le Pari Oppor- tunità, 2020). Specifically, the task force report proposed five areas of intervention to promote women’s leadership, enhance women’s participation in the workforce, em- power women’s skills in STEM, eradicate gender stereotypes and promote financial independence. While these actions cover a wide range of areas where the gender gap is significant, the related areas reflected in the NRRP formulation predominantly concern women’s participation in the workforce and the promotion of female entrepreneurship. The entry of more women in the task forces contributed to flagging gender equality as one of the three axes of the Colao Task Force Plan ‘Initiatives for the relaunch of Italy 2020-2022’. In parallel, at the European level, while the MEP Alexandra Geese launched ‘Half of It’, a petition calling for half of the funds of NGEU to be devoted to women, the European Commission announced that national plans would be assessed through gen- dered lenses: ‘[...] Member States should demonstrate that the objectives of gender equality and equal opportunities for all are mainstreamed into the plan’ (European Com- mission, 2021). These factors contributed to the saliency of gender equality within the public debate, suggesting there may still be a fundamental role played by women in voic- ing parity-relevant issues, which may extend to parliamentary policy-makers. 4. Gender parity in NGEU and the Italian NRRP Against the backdrop of the gendered implications of the pandemic, the EU response of- fers a well-defined case for the evaluation of equality concerns in the recovery phase. Following an initial period in which the Member States were left entirely on their own to sustain their economy through the crisis and fuel reconstruction, and after months of divisive negotiation, a common recovery instrument (i.e. NGEU) found political agree- ment in July 2020. The programme, and more specifically its financing instrument (i.e. the Recovery and Resilience Facility (RFF)), underwent a lengthy legislative process re- quiring not only the green light of the EU institutions but also the ratification of the national parliaments. Its complex form, however, is of value given that priorities – in- cluding the transversal one of gender equality – should generally direct those of all EU Member States. Gender mainstreaming has long been within EU policy priorities. Currently, the Gender Equality Strategy 2020-2025 drives EU policy on the matter, complementing mainstreaming with intersectionality as horizontal principles guiding its implementa- tion (European Commission, 2020). Its objectives are five-fold. The first three – (1) being free from violence and stereotypes; (2) thriving in a gender-equal economy; (3) leading equally throughout society – reflect specific areas of policy intervention which can be summarized respectively as dedicated to the area of gender-based violence, eco- nomic empowerment and political representation. Two further points – (4) gender mainstreaming and an intersectional perspective in EU policies and (5) funding actions to make progress in gender equality in the EU – reflect its implementation stage which, beyond the financing of the strategy, stresses indeed the two above-mentioned horizon- tal principles. Such an approach has been included within NGEU. In this context, the commonly defined priorities within the NGEU regulatory framework foresee the main- streaming of gender equality. The RRF Regulation (Regulation (EU) 2021/241) refers to gender in Recital 28 which, acknowledging the uneven burden of the pandemic on CERON and ZARRA 81 women, calls for gender-mainstreaming within the plans. Recital 39 likewise calls for the detailing of expected contributions to gender equality within the NRRP, which is also recalled in article 4 among the elements for the Member States to include for the plan to be ‘duly reasoned and substantiated’. The Italian NRRP, which was approved on 13 July 2021, requires the country to im- plement reforms and investments in response to the pandemic crisis. The plan is envisioned for the 2021-2026 timeframe and it is part of a broader set of measures, in- cluding inter alia NGEU and the European Cohesion Policy funding for 2021-2027. The Plan revolves around three intervention areas, namely digitalization and innovation, ecological transition and social inclusion. According to the Italian government’s fore- casts, the NRRP will lead to a 3.6% increase in GDP and a 3.2% increase in employment. More in detail, the NRRP develops along 16 Components. These components are grouped into six core missions. Each mission indicates the reforms necessary to more effectively implement the measures. The plan includes 63 reforms in total, which can be divided into: • Horizontal reforms, namely structural innovations of the system aimed at im- proving equity, efficiency, competitiveness and the country's economic framework; • Enabling reforms, functional to ensure the implementation of the Plan and to remove administrative, regulatory and procedural obstacles that affect eco- nomic activities and the quality of services provided; • Sectoral reforms (included in the individual Missions), namely regulatory in- novations in specific areas of intervention or economic activities, designed to introduce more efficient regulatory and procedural regimes in the respective sectoral areas; • Competing reforms, i.e. measures not directly included in the Plan, but neces- sary for the achievement of its general objectives. Against this backdrop, the Italian NRRP considers gender as a cross-cutting priority within the broad category of ‘social inclusion’. Such a role for gender equality was intro- duced in the last iteration of the plan, under the auspices of the Draghi government in April 2021. In the final stage, the saliency of parity increased in terms of financial com- mitments. The first version of the plan presented by the Conte government devoted only 4.52 billion euros to gender equality out of the 209 billion, which were deemed insuffi- cient to address the gap. The draft was received with sharp criticism from women’s interest groups, denouncing insufficient funding and prioritisation of parity-relevant measures, especially in light of the detrimental effect and additional burden of the pan- demic. The final version of the plan increased parity-relevant funding to 7 billion. Specifically, it contains two mechanisms to reduce gender gaps. On the one hand, it en- visages direct investments to stimulate women’s employment – for instance through a fund for women’s entrepreneurship. On the other, it allocates funds to sectors that are particularly relevant for women’s empowerment, such as 4.6 billion for childcare facili- ties. Nevertheless, the evolution of the plan, which especially in the early stages was negatively evaluated by women’s interest groups, begs the question of the centrality of women actors in raising concerns for parity-relevant issues. Indeed, such activism was widely present within civil society. While assessment of the impact of women’s interest Women’s substantive representation in the pandemic 82 groups on the overall policy process is beyond the scope of the analysis, gender-based dy- namics within the parliamentary arena offer a first well-defined step in assessing the gender divide over parity-relevant issues. As such, it offers the opportunity to evidence the relevance of substantive representation in the specific case of the Italian NRRP par- liamentary debate, with broader implications for the policy process and pandemic reconstruction at large. 5. Research design: delimiting the corpus and the codebook The research question of the substantive representation of women in the parliamentary debates on the Italian NRRP is assessed through a manual content analysis. The research design hinges on three key aspects in relation to the data and methods: delimiting the corpus of speeches on the NRRP, selecting the codebook of parity-relevant issues and identifying how to assess the gender gap in the prevalence of the latter. In the first ac- count, the debate over NGEU and the Italian plan within Parliament is composite, spanning across the different phases in the process of delineating the NRRP and arenas, ranging from floor debates, a multitude of committee meetings, formal and informal hearings, to Q&A with Ministers and communications of the Prime Minister. For the re- search question at hand, it is of value to be widely inclusive in terms of the timeframe of the debate across the various iterations of the plan, presented in Figure 1, especially given the changes in governing coalition. Conversely, minimising heterogeneity of the fora and procedures yields a more consistent corpus in which to minimise potential bias linked, for example, to the specific policy domain and (gender) composition of a com- mittee or actor in the case of hearings and Q&A. Figure 1. Timeline of the phases in the delineation of the Italian NRRP Accordingly, debates are selected – with the reference document as an object – to cover all three phases across the Senate and Chamber of Deputies considering: ● Debate on the Guidelines in the Senate: Relazione delle Commissioni riunite 5ª e 14ª sulla proposta di «Linee guida per la definizione del Piano nazionale di ri- presa e resilienza» (Doc. XVI n. 3) ● Debate on the Conti proposal in the Chamber of Deputies: Discussione della Re- lazione della V Commissione sulla proposta di Piano nazionale di ripresa e resilienza (Doc. XXVII n. 18-A) Debate on the Guidelines in the Senate 13 Oct. 2020 Debate on the Conte Proposal 31 Mar. 2021 Debate on the Draghi Proposal 27 Apr. 2021 EU greenlight on the Italian plan 13 July 2021 CERON and ZARRA 83 ● Debate on the Draghi proposal in the Chamber of Deputies: Comunicazioni del Presidente del Consiglio dei Ministri in vista della trasmissione alla Commis- sione europea del Piano nazionale di ripresa e resilienza ai sensi dell'articolo 18 del regolamento RRF (UE) 2021/241. ● Debate on the Draghi proposal in the Senate: Comunicazioni del Presidente del Consiglio dei ministri in vista della trasmissione alla Commissione europea del Piano nazionale di ripresa e resilienza e conseguente discussione The resulting sample of 226 speeches does not allow us to overtly assess differences across the drafting phases or Prime Ministerships and related majorities. However, the potentially evolving cleavages imply that it is of great importance to capture within the analysis both the early and the late phases of drafting. The first allows for a broader emer- gence of parity-relevant themes that may have initially been given low priority in the recovery. The latter captures MPs’ prioritising within their speech, either welcoming the gender parity-relevant content of the final plan or signalling insufficient progress. Con- comitantly, the life-cycle of the NRRP approval provides for comprehensive coverage of gender and party differences concerning the saliency of equality. Against this back- ground, the selection of the documents within the corpus includes the three phases in the delineation of the NRRP, namely the early guidelines communicated by the govern- ment to Parliament, the first plan presented under the auspices of the Conte government and its final iteration presented by Draghi. The second step in the analysis is the selection of the codebook through which the dependent variable of the analysis is constructed, capturing whether each speech refers to parity-relevant issues. The codebook is generated on the basis of two levels of sources. Firstly, the parliamentary dossier on the final plan provides an overview of how the transversal priority of equality is translated into practice within the NRRP (Documenta- tion Services of the Chamber of Deputies, 2021). The document offers the classification of which measures are considered parity-relevant within the plan, as summarised in the previous section. The policy content of the plan itself offers guidance on the overarching classification of parity-relevant measures to guide the measurement strategy in coding speeches. We derive a complementary source for parity-relevant concerns by consider- ing the priorities of the main advocates for equality: women’s interest groups. Arguably, their position offers a hard test of parity-relevant priorities regarding the NRRP overall, as well as the least consensual components of a recovery catering equality, hence in- cluded in their platform for gender mainstreaming within the plan. Additionally, going beyond the final policy outcome in the NRRP allows us to include within the coding strat- egy also parity-relevant issues which were not successfully included in the plan. The complementary approach is facilitated by the saliency of parity-relevant con- cerns within the policy and public debate surrounding the Italian NRRP. Indeed, women’s interest groups active in the arena of gender parity have participated in parlia- mentary hearings on the NRRP. Besides, ad-hoc associations and petitions emerged. Specifically, two key associations founded during the Covid-19 crisis – Il Giusto Mezzo and Half of It – put forward a joint petition #UnaVoceNonBasta (Il Giusto Mezzo, 2021a) endorsed by a multitude of women’s interest groups. Beyond such text, Half of It (2021) put forward a Manifesto. Il Giusto Mezzo (2021b) developed proposals on the NRRP to- gether with a technical (gendered) analysis of how to mainstream parity across all pillars Women’s substantive representation in the pandemic 84 within NGEU. On this account, Il Giusto Mezzo mobilised other women’s interest groups through a jointly undersigned letter to the government voicing those concerns and proposals, along with the parallel request of equal representation in committees ded- icated to the pandemic and recovery effort. As mentioned, the lack of women representation in task forces for crisis management led to a petition endorsed by 86 or- ganisations (Dateci voce, 2020). In addition to the petitions and policy proposals, three women’s interest groups took part in the parliamentary hearings, also supplying written documentation to the committees, namely Il Giusto Mezzo (2021c), Ladynomics (2021) and Ingenere (2021). Because of such widespread endorsement of the key petitions emerging from women’s interest groups, their policy documents are representative not only of the position of the specific organisation but rather of a broader constellation of organisations promoting gender parity. The documents mentioned above and presented systematically in Table A1 in the appendix offer the reference for outlining parity-relevant concerns against which par- liamentary speeches can be classified. The mapping exercise delineates the primary parity-relevant elements within the plan along with concerns and policy proposals ad- vanced by key civil society organisations. Themes include: ● employment, firms with references to sectors and the crisis; ● services and infrastructures; ● social policies, assistance and instruments; ● investments and evaluation; ● youth and education. More specifically, certain terms related to the categories above are highly recurrent within the policy documents referring to the above-mentioned categories. The most rel- evant examples within the employment arena are women’s entrepreneurship/ employment and maternity, especially in connection to leave. Another cluster refers to children, childcare and early education services, referring both to kindergartens and fa- cilities for the 0-3 age group. A further arena is that of stereotypes, completed by terms relating to gender-based violence. On such a basis, we selected a coding strategy which delineates the near-automatic positive coding as parity-relevant of speeches containing the following terms, presented in alphabetical order in the original language: asilo/i (kin- dergarten/s); bambino/i (child/ren); cura (care); donna/e (woman/en); famiglia/e (family/ies); femminile/i (feminine); femminicidi (feminicide); figlie/i/o/a (offspring); genere (gender); lavoratrice/i (female worker/s); maternità (maternity); nido/i (nursery/ies); parità (parity); scuola/e (school/s); stereotipo/i (stereotype/s); violenza/e (violence). A broad classification which may be derived can distinguish across: (i) interven- tions aimed at supporting employment, both directed at workers and employers (ii) policies aimed at supporting care responsibilities, lessening a burden predominantly falling on women’s shoulders and (iii) gender-based violence, which remains, however, only a secondary focus within the position papers of the key interest groups and the plan. These themes and terms constitute the benchmark against which we code the parlia- mentary speeches as raising parity-relevant issues. All speeches are coded manually and independently by the two authors over their parity-relevant content to account for inter- CERON and ZARRA 85 coder reliability, leading to an agreement of over 95%.3 It should be noted that the ap- proach followed by the authors gives priority to the parity-relevant implications of the speeches rather than the mere presence of the terms potentially associated with equality. A glaring example is the use of a generally parity-relevant term without any specific con- cerns for equality. School, a particularly divisive element in pandemic management, is at times not mentioned in connection with the highly parity-relevant policy choices of closures. Rather, for example, schools may be mentioned in contesting infrastructural and procurement choices such as the provision of ad-hoc desks to facilitate distancing. The choice of the authors – which always aligned empirically in the coding – was that of giving primacy to the substantive content of the speech, hence not classifying as parity- relevant those speeches merely using some of the above-highlighted terms in ways unre- lated to equality. The final step concerns the operationalisation of the independent variable and con- trols. For each speech we account for speaker gender and party. Excluding the two speeches by members of the government, Table A2 shows the gender distribution of the MP who took the floor in those debates. Overall, women MPs within the sample amount to 37.5%, spanning at the party level from the lowest proportion of Lega of 25% to the high- est of 58% for IV. Accordingly, the proportion of women MPs within Parliament overall and in each party is not necessarily reflected in the corpus considered for the analysis. Nevertheless, the analysis does not consider the absolute number of parity-relevant speeches among women and men MPs or in each party. Rather, it considers its relative proportion comparing the percentage of women and men overall or in a specific party mentioning parity-relevant issues. Accordingly, we analysis the data through descrip- tive statistics of the prevalence of parity-relevant issues across women and men MPs overall and within each party. Additionally, we employ a logistic regression assessing whether the emerging gender gap remains when controlling for the party of the MPs. 6. Results: the gender gap in parity-relevant issues The analysis reveals substantial differences across gender and party of the actors voicing parity relevant concerns. As shown by Figure 2, the overall gender divide is stark and sig- nificant when considering a t-test comparing, overall, the proportion of parity-relevant speeches across men and women. Indeed, nearly half of speeches by women mention parity-relevant concerns, while the proportion falls under 20% for men. 3 In one instance, a discrepancy between the two independent coding exercises was found. The term ‘school’ was classified as gendered by only one author, but then it was decided to remove it from the gen- dered speeches as the term was used in the context of architectural barriers. Women’s substantive representation in the pandemic 86 Figure 2. Proportion of parity-relevant MPs’ speeches by gender Note: On the left the proportion of men (m) mentioning themes related to gender equality, on the right the proportion across women (f). The overall gender divide over equality speeches may, however, derive from cross- party differences in the makeup of women and men MPs as Table A2 shows heterogene- ous gender balances within the corpus across political groups. Indeed, the most balanced representation of speakers in the corpus clusters at the centre, with Italia Viva (IV) reaching nearly 60%, followed by Forza Italia (FI) with nearly 45%. Conversely, the right has among the lowest scores of women MPs within the speakers considered, notably, only 25% for Lega and 32% for Fratelli d’Italia (FdI), respectively holding the two bottom rankings. As a result, the pool of women and men MPs is ideologically heterogeneous, potentially contributing to the overall divide shown in Figure 2. Indeed, more women be- long to the centre to the centre-left side of the spectrum, while less represented among speeches by Lega and FdI, hence implying that the ideological heterogeneity in the sali- ency of equality may be at play rather than substantive representation per se. Figure 3 reinforces such a concern, highlighting substantial differences in parity- relevant speeches across party lines. The highest proportion (above 60 per cent) of speeches calling for gender equality is associated with the MPs of the Democratic Party (PD), closely followed by the left (LEU) and centre. On the opposite end of the spectrum, the far (FdI) and centre-right (FI) obtain the lowest score for mentioning parity-relevant issues. Against this backdrop, robust evidence pointing towards the importance of sub- stantial representation in the making of the NRRP would be represented by a gender divide at the party level in the saliency of equality within the parliamentary debate. 0 .1 .2 .3 .4 .5 m ea n of c od ed g en de re d_ sp ee ch m f CERON and ZARRA 87 Figure 3. Proportion of gendered speeches by the party of the MPs Note: proportion of MPs mentioning themes related to gender equality across political party. Figure 4 below shows there is indeed such a gender divide also across parties. With the sole exception of LEU and the centre, consisting, in both instances, of a very limited number of speeches (9 and 4 respectively), all parties show a gap in the saliency of equal- ity across women and men MPs in favour of the former. Moreover, the effect is robust to differences in overall sensitivity across parties. On one side, we have the party shown in Figure 2 as the one with the lowest proportion of equality speeches overall (FdI) for which any parity-relevant issue is driven entirely by women. On the other, we have the party marked by Figure 2 as the most sensitive to equality (PD) where all women MPs mention parity-relevant issues in their speeches on the NRRP while that is the case for only half of the men. Similarly, only 7% of speeches by men MPs from FI concern parity-relevant issues compared to 25% of those by women within the same party. Lega displays a divide that is somewhat narrower, amounting to nearly 24% for men which almost doubles for women MPs at nearly 43%. Moving towards the centre, the IV gap is more marked, with men scoring substantially below 14% while women reach 40%. Conversely, the M5S is among those with the smallest gender gap with a proportion of 22% for men and only 36% for women. The emerging picture is that while cross-party differences are substantial, they do not alone drive the gender divide in the saliency of equality as gaps in parity-rel- evant speeches remain across women and men MPs of the same party. 0 .2 .4 .6 .8 m ea n of c od ed g en de re d_ sp ee ch centro fdi fi iv lega leu m5s misto pd Women’s substantive representation in the pandemic 88 Figure 4. Proportion of gender parity-relevant speeches across party and gender of MPs Source: own elaboration. Results show that the substantive representation of gender equality in the context of the Italian NRRP is overwhelmingly driven by women. At the same time, this is not to discount the remaining relevance of heterogeneous saliency across parties: men from the party with the highest prevalence overall of parity-relevant concerns (PD) – which, overall, mentions equality 67% of the times – display sensitivity for such themes more often than women in all other parties other than LEU and the centre. Nevertheless, the more worrying dynamic is that even within the context of a party that does display a sen- sitivity to equality, the gender divide remains well-marked and even larger than in some instances in which saliency is lower overall. In this context, it may be likewise of value to briefly go beyond the numerical com- parison of the frequency with which women and men mention parity-relevant issues across party lines. That is, the themes themselves that are most common may likewise translate into substantive differences in which types of concerns are raised and the ex- tent to which those are translated into support for policies fostering gender equality. Specifically, certain themes, such as families, a generic reference to children and the call for support for increased natality is prominent within right-wing parties (FdI, LEGA), yet also in this instance carried predominantly if not exclusively (e.g. FdI) by women. In this context, references to parity-relevant terms – which do often result in coding within the equality camp – may at times align with policy concerns voiced by women’s interest groups (e.g. supporting families with children) while displaying rhetoric that may not be fully consistent with the promotion of women’s empowerment (e.g. focusing on improv- ing birth rates rather than, for example, the expansion of care services or women’s 0 .2 .4 .6 .8 1 m ea n of c od ed g en de re d_ sp ee ch centro fdi f iv lega leu m5s misto pd m f m f m f m f m f m f m f m f m f i CERON and ZARRA 89 employment opportunities). As a result, assessed through the lenses of concerns of women’s interest groups, the contribution of some parity-relevant speeches to advocat- ing women’s empowerment within the NRRP may be minimal. Nevertheless, against the research question at hand, the gender divide between MPs in voicing such concerns, even in a weak form, remains. Findings, hence, evidence substantive representation even in the context of low overall saliency of equality. Moving to the opposite end of the spectrum, in the centre-left party (PD) more frequent reference to equality is paired with greater alignment with the policy proposals and concerns raised by women’s interest groups overall. An overt reference to gender parity, and its mainstreaming within the NRRP and women’s empowerment, is more prominent within the discourse of PD MPs. Once again, however, not only is the proportion of women MPs raising such concerns greater, but also the scope of their support for empowerment policies is more extensive. Women MPs within such parties are more prone to referring to specific policies in sup- port of equality, especially as regards the strengthening of care services and parental leave, as well as highlighting the impact of the pandemic on equality and the related need for intervention in the prevention of gender-based violence. As a result, men and women MPs not only display a different saliency overall for equality, persistent when consider- ing ideology, but the associated narrative and level of alignment with women’s interest groups also differ, providing robust support for substantive representation within the parliamentary debates relating to the Italian NRRP. The characteristics and structure of the datasets offer a warning against deriving conclusions on the size of the gender gap. In line with the hypothesis under considera- tion, the key element of interest is, however, directional: namely, whether women MPs display higher saliency of parity-relevant themes. In such a context, going beyond the significant differences in the proportion of parity-relevant speeches overall and at the party level reported above, controlling for gender and party concomitantly allows for ro- bust conclusions of whether indeed such a trend is pervasive across party lines. Figure 5 reports the marginal effect by gender and party of a logit model with the dummy reflect- ing whether each speech is parity-relevant as a dependent variable, and as independent variables a dummy distinguishing between women and men MPs and a categorical vari- able capturing political group membership. Marginal effects only by gender overall and by parties without distinguishing between men and women MPs are reported in the ap- pendix in Figure A1 and Figure A2 respectively. A significant gender difference is confirmed, as shown by Figure A1 and Table A3 in the appendix. The Democratic Party is confirmed as the party with the most sensitivity to the themes, with all other parties displaying significantly lower saliencies. When pitted against the PD all parties indeed perform worse. Figure 5 highlights that the left-right gradient is not fully reflected in terms of sensitivity to parity. Indeed, considering FI as a reference, the League displays a significantly higher saliency of gender. Additionally, a gender gap emerges across party lines, albeit more markedly in some instances rather than others: at a 10 percent level differences are not significant only for IV, FI and FdI. Women’s substantive representation in the pandemic 90 Figure 5. Proportion of parity-relevant speeches by gender and party of the MPs Note: marginal effects for women (f) and men (m) MPs of the parties for which a sufficient number of speeches across gender is available. Additionally, a brief comparison of the nature of parity-relevant themes which are more pervasive across different political parties shows that saliency of equality may not reflect the same priorities across political groups. An in-depth analysis of what we in- clude within parity-relevant discourse may further discount the scoring of some parties overall and the contribution of men in more equality-oriented parties. As a result, further research is warranted comparing at a more granular level the specific keywords emerg- ing across speeches. Nevertheless, the hypothesis of a gender gap within the parliamentary debate on the NRRP is indeed supported: across party lines a higher pro- portion of women MPs included within the corpus speak in support of gender parity compared to the proportion of men. 7. Final remarks: the equality cost of women’s under-representation In Italy, the pandemic has profoundly impacted women, who bore the cost of forced lock- downs and the closure of schools and childcare facilities with an increased level of domestic violence and unemployment. At the same time, while the race for gender parity has been halted by the outbreak, it regained centre stage in the public debate, with women’s interest groups fiercely contesting the lack of women’s representation in deci- sion-making. Initially, the country scored the highest with regard to the restrictiveness of its Covid-19 management, putting forward measures that heavily penalise women, such as school closures (Hale et al., 2020). Down the line of crisis management and in the recovery effort, the prioritisation of gender equality in the Italian NRRP has similarly been deemed insufficient by women’s interest groups, especially in its early iterations. 0 .2 .4 .6 .8 1 P r( ge nd er ed . s pe ec h) PD M5S IV FI Lega FdI Political Group m f Adjusted predictions with 95% CIs CERON and ZARRA 91 Against such a benchmark, our text analysis shows that women played an important role in advocating for parity-relevant issues in the parliamentary debate on the formula- tion of the NRRP. Parity-relevant issues are mentioned in nearly half of the speeches by women MPs while only less than 20% in those by men parliamentarians, proving that parity-relevant concerns are highly women-driven. Although men MPs are nearly twice as numerous as women within the corpus of NRRP speeches, they account for only 28 parity-relevant speeches out of 226. Conversely, the sparser women MPs account for 39 out of the 67 parity-relevant speeches. Such an overall gender difference has been as- sessed in the analysis across party lines, indicating that it is not entirely driven by the heterogeneous ideological composition of the MPs pool. At the institutional level, women MPs were, overall, those contributing the most to the advancement of themes relating to women’s empowerment within parliamentary debates. Albeit sensitivity to equality varies substantially across parties, the gender gap in concerns over parity and women’s empowerment remains and is robust to accounting concomitantly for gender and party of MPs. Our analysis based on manual coding of the corpus of over 200 speeches allowed us to account for irrelevant or biased uses of terms associated with potentially parity-rele- vant issues. Nevertheless, our findings provide only a limited overview of differences in the content and priorities within gender parity-relevant speeches across men and women MPs as well as party lines. Further exploration of the corpus of NRRP speeches may be warranted to better delineate such differences. At the same time, in broad terms, the analysis already highlighted the predominance of women MPs, especially within the centre-left, in raising concerns which more closely align with the petitions and hearing contributions of women’s interest groups, in particular concerning childcare and em- ployment. The implication is that within the parliamentary debate on the NRRP, women played a key role in raising parity-relevant concerns, pointing especially in the direction of mitigating the most detrimental legacies of the pandemic for equality. The broader policymaking process – albeit beyond the focus of the research question at hand – further reinforces the centrality of women in promoting gender mainstream- ing in the NRRP. Campaigns of women’s interest groups over gender equality in the NRRP may indeed further compound the contribution of women policy actors to advo- cating for parity within parliamentary debates. A broader analysis of the whole policy cycle may indeed uncover further arenas in which gender gaps emerge over the saliency of equality. The parliamentary scope of our preliminary analysis is a limitation which, however, reinforces the need for further research to verify whether throughout the pol- icy network women remain central in flooring concerns over equality. Our findings of gender differences in advocating women’s empowerment should be read against a broader policy outcome within the NRRP. The overview of the evolution of the plan through gendered lenses suggests a substantial progress from the early stages of the Italian NRRP. The improvement from the Conte to the Draghi plan is undeniable in terms of funding, growing from 4.5 to 7 billion, an amount that from the perspective of the interest groups nevertheless pales against the scale of the overall plan worth 204.5 billion. The resulting allocations for parity remain indeed far below the ‘half of it’ de- manded by women’s interest groups. Given the gender divide in advocating for gender mainstreaming in the parliamentary debate, the improvement of funding throughout Women’s substantive representation in the pandemic 92 the policy cycle raises two further key questions. On one hand is the extent to which such progress saw the contribution of women policy actors and women’s interest groups; on the other, the extent to which the under-representation of women in the policy network – evident, for example, in the context of task forces – contributed to a prioritisation of parity-relevant issues deemed unsatisfactory by women’s interest groups. The contribution of this work is twofold. Firstly, it confirms the peril of women un- der-representation for gender parity (Hessami & da Fonseca, 2020), with implications that may well expand beyond the NRRP and post-Covid reconstruction. Additionally, the analysis paves the way for further in-depth assessment not only of the saliency of con- cerns raised by women’s interest groups in the parliamentary chamber but also, more broadly, of actors within the policy network, of particular relevance given the govern- mental nature of the NRRP. In this respect, future research on the social media debate around the NRRP may help better identify policy-makers and interest groups raising concerns on parity-relevant issues, outlining their contribution to shaping the plan. Ad- ditionally, further research may likewise expand the institutional analysis in scale – considering pandemic decision-making beyond the NRRP – and scope – providing a richer understanding of gender and party differences in the discourse surrounding equality. As a result, the analysis not only opens further avenues of research in gender equality in the context of the Covid-19 crisis but also points towards the substantive im- plications of women’s under-representation in decision-making as pushing for equality largely remains a women’s task. Acknowledgements We are grateful for the valuable comments of the editor and anonymous referees. We are thankful for the precious feedback on earlier drafts of this work of Pamela Pansardi, Li- cia Papavero and the attendees of the SISP 2021 Gender and Representation panel. References Adams-Prassl, A., Boneva, T., Golin, M., & Rauh, C. (2020). Inequality in the impact of the coronavirus shock: Evidence from real time surveys. Journal of Public Economics, 189, 104245. https://doi.org/10.1016/j.jpubeco.2020.104245 Alon, T., Doepke, M., Olmstead-Rumsey, J., & Tertilt, M. (2020). The Impact of COVID-19 on Gender Equality (No. w26947; p. w26947). National Bureau of Economic Research. https://doi.org/10.3386/w26947 Bagues, M., & Campa, P. (2021). Can gender quotas in candidate lists empower women? Evi- dence from a regression discontinuity design. Journal of Public Economics, 194, 104315. https://doi.org/10.1016/j.jpubeco.2020.104315 Baltrunaite, A., Casarico, A., Profeta, P., & Savio, G. (2019). Let the voters choose women. Jour- nal of Public Economics, 180, 104085. https://doi.org/10.1016/j.jpubeco.2019.104085 Beaman, L., Duflo, E., Pande, R., & Topalova, P. (2006). Women politicians, gender bias, and policy-making in rural India. Berkman, M. B., & O’Connor, R. E. (1993). Do Women Legislators Matter?: Female Legisla- tors and State Abortion Policy. American Politics Quarterly, 21(1), 102–124. https://doi.org/10.1177/1532673X9302100107 CERON and ZARRA 93 Besley, T., & Case, A. (2003). Political Institutions and Policy Choices: Evidence from the United States. Journal of Economic Literature, 41(1), 7–73. https://doi.org/10.1257/jel.41.1.7 Boca, D. D., Oggero, N., Profeta, P., & Rossi, M. (2020). Women’s work, housework and child- care, before and during COVID-19. Calò, X., Occhiuzzi, R., & Profeta, P. (2021). COVID-19 and its economic impact on women and women’s poverty. FEMM European Parliament. Carroll, S.J. (2001). The Impact of Women in Public Office. Bloomington: Indiana University Press. Celis, K. (2006). Substantive Representation of Women: The Representation of Women’s In- terests and the Impact of Descriptive Representation in the Belgian Parliament (1900– 1979). Journal of Women, Politics & Policy, 28(2), 85–114. https://doi.org/10.1300/J501v28n02_04 Celis, K., Childs, S., Kantola, J., & Krook, M. L. (2008). Rethinking women’s substantive rep- resentation. Representation, 44(2), 99–110. Ceron, M., & Zarra, A. (2021). A Tale of Two Pandemics? COVID-19 and Gendered Policy Re- sponses in the EU. Cambio. Rivista Sulle Trasformazioni Sociali, OpenLab on Covid-19. https://doi.org/10.13128/cambio-10280 Chattopadhyay, R., & Duflo, E. (2004). Women as policy makers: Evidence from a randomized policy experiment in India. Econometrica, 72(5), 1409–1443. Comunicazioni del Presidente del Consiglio dei Ministri in vista della trasmissione alla Com- missione europea del Piano nazionale di ripresa e resilienza ai sensi dell'articolo 18 del regolamento RRF (UE) 2021/241. Comunicazioni del Presidente del Consiglio dei ministri in vista della trasmissione alla Com- missione europea del Piano nazionale di ripresa e resilienza e conseguente discussione. Crowley, J. E. (2004). When Tokens Matter. Legislative Studies Quarterly, 29(1), 109–136. https://doi.org/10.3162/036298004X201113 Dateci voce. (2020). #datecivoce https://www.datecivoce.it/firma-la-petizione/ Del Boca, D., Oggero, N., Profeta, P., & Rossi, M. (2020). Women’s and men’s work, housework and childcare, before and during COVID-19. Review of Economics of the Household, 18(4), 1001–1017. Dipartimento per le Pari Opportunità. (2020). Women for a New Renaissance. http://www.pariopportunita.gov.it/wp-content/uploads/2020/06/Women-for-a-New- Renaissance.pdf.pdf Discussione della Relazione della V Commissione sulla proposta di Piano nazionale di ripresa e resilienza (Doc. XXVII n. 18-A). Documentation Services of the Chamber of Deputies. (2021). Dossier PNRR http://docu- menti.camera.it/leg18/dossier/pdf/DFP28a.pdf EIGE. What is gender mainstreaming. (n.d.). European Institute for Gender Equality. Re- trieved 29 January 2022, from https://eige.europa.eu/gender-mainstreaming/what-is- gender-mainstreaming European Commission. (2020). Gender equality strategy. https://ec.europa.eu/info/poli- cies/justice-and-fundamental-rights/gender-equality/gender-equality-strategy_en European Commission. (2021). Guidance to Member States Recovery and Resilience Plans. Women’s substantive representation in the pandemic 94 Galasso, V., Pons, V., Profeta, P., Becher, M., Brouard, S., & Foucault, M. (2020). Gender dif- ferences in COVID-19 attitudes and behavior: Panel evidence from eight countries. Proceedings of the National Academy of Sciences, 117(44), 27285–27291. Geys, B., & Sørensen, R. J. (2019). The impact of women above the political glass ceiling: Evi- dence from a Norwegian executive gender quota reform. Electoral Studies, 60, 102050. https://doi.org/10.1016/j.electstud.2019.102050 Hale, T., S. Webster, A. Petherick, T. Phillips, & B. Kira. (2020). Oxford Covid-19 Government Response Tracker. Oxford: Blavatnik School of Government, n. 25. Half of it (2021). Manifesto https://www.halfofit.it/wp-content/uploads/2021/02/mani- festo_completo.pdf Hawkesworth, M. (2003). Congressional Enactments of Race-Gender: Toward a Theory of Raced-Gendered Institutions. The American Political Science Review, 97(4), 529–550. Heath, R. M., Schwindt-Bayer, L. A., & Taylor-Robinson, M. M. (2005). Women on the Side- lines: Women’s Representation on Committees in Latin American Legislatures. American Journal of Political Science, 49(2), 420. https://doi.org/10.2307/3647686 Hessami, Z., & Baskaran, T. (2019). Competitively elected women as policy makers. Hessami, Z., & da Fonseca, M. L. (2020). Female political representation and substantive ef- fects on policies: A literature review. European Journal of Political Economy, 63, 101896. Hong Fincher, L. (2020, April 16). Women leaders are doing a great job at handling the pan- demic. So why aren’t there more of them? - CNN. https://edition.cnn.com/2020/04/14/asia/women-government-leaders-coronavirus- hnk-intl/index.html Iceland misses out on Europe’s first female-majority parliament after recount. (2021, Septem- ber 26). BBC News. https://www.bbc.com/news/world-europe-58698490 Il Giusto Mezzo. (2021a). #UnaVoceNonBasta https://ilgiustomezzo.it/wp-con- tent/uploads/2021/04/gm_unb_manifesto.pdf Il Giusto Mezzo. (2021b). Le proposte del Giusto mezz.o https://ilgiustomezzo.it/wp-con- tent/uploads/2021/03/le-proposte-del-giusto-mezzo.pdf Il Giusto Mezzo. (2021c). Hearing document il giusto mezzo https://www.camera.it/applica- tion/xmanager/projects/leg18/attachments/upload_file_doc_acquisiti/pdfs/000/004/ 730/Giusto_mezzo_02.02.21_.pdf Ingenere. (2021). Hearing document Ingenere https://www.camera.it/application/xmana- ger/projects/leg18/attachments/upload_file_doc_acquisiti/pdfs/000/004/728/Roberta _Carlini_02.02.21_.docx IPU. (2021). Monthly ranking of women in national parliaments. Parline: The IPU’s Open Data Platform. https://data.ipu.org/women-ranking Kantola, J., & Lombardo, E. (2017). Conclusions: Understanding Gender and the Politics of the Crisis in Europe. In J. Kantola & E. Lombardo (Eds.), Gender and the Economic Cri- sis in Europe (pp. 257–270). Springer International Publishing. https://doi.org/10.1007/978-3-319-50778-1_12 Kathlene, L. (1994). Power and Influence in State Legislative Policymaking: The Interaction of Gender and Position in Committee Hearing Debates. The American Political Science Review, 88(3), 560–576. JSTOR. https://doi.org/10.2307/2944795 Krook, M. L. (2010). Women’s Representation in Parliament: A Qualitative Comparative Analysis. Political Studies, 58(5), 886–908. https://doi.org/10.1111/j.1467- 9248.2010.00833.x CERON and ZARRA 95 Ladynomics. (2021). Hearing document Ladynomics https://www.camera.it/applica- tion/xmanager/projects/leg18/attachments/upload_file_doc_acquisiti/pdfs/000/004/ 729/Ladynomics_02.02.21_.pdf Lijphart, A. (1991). Double-Checking the Evidence. Journal of Democracy, 2(3), 42–48. https://doi.org/10.1353/jod.1991.0041 Lippmann, Q. (2020). Gender and Lawmaking in Times of Quotas. Available at SSRN 3330617. McAllister, I., & Studlar, D. T. (2002). Electoral systems and women’s representation: A long-term perspective. Representation, 39(1), 3–14. https://doi.org/10.1080/00344890208523209 McDonagh, E. (2010). It Takes a State: A Policy Feedback Model of Women’s Political Repre- sentation. Perspectives on Politics, 8(1), 69–91. https://doi.org/10.1017/S1537592709992672 Mongey, S., & Weinberg, A. (2020). Characteristics of workers in low work-from-home and high personal-proximity occupations. Becker Friedman Institute for Economic White Pa- per. Phillips, A. (1995). The politics of presence. Clarendon Press. Regulation (EU) 2021/241 of the European Parliament and of the Council of 12 February 2021 establishing the Recovery and Resilience Facility. (2021) OJ L 57, 18.2.2021, p. 17–75. Relazione delle Commissioni riunite 5ª e 14ª sulla proposta di «Linee guida per la definizione del Piano nazionale di ripresa e resilienza» (Doc. XVI n. 3). Rosenbluth, F., Salmond, R., & Thies, M. F. (2006). Welfare Works: Explaining Female Leg- islative Representation. Politics & Gender, 2(02). https://doi.org/10.1017/S1743923X06060065 Rule, W. (1994). Women’s Underrepresentation and Electoral Systems. PS: Political Science and Politics, 27(4), 689. https://doi.org/10.2307/420369 Schwindt-Bayer, L. A. (2006). Still Supermadres? Gender and the Policy Priorities of Latin American Legislators. American Journal of Political Science, 50(3), 570–585. https://doi.org/10.1111/j.1540-5907.2006.00202.x Siaroff, A. (2000). Women’s Representation in Legislatures and Cabinets in Industrial De- mocracies. International Political Science Review, 21(2), 197–215. https://doi.org/10.1177/0192512100212005 Thames, F. C., & Williams, M. S. (2010). Incentives for Personal Votes and Women’s Repre- sentation in Legislatures. Comparative Political Studies, 43(12), 1575–1600. https://doi.org/10.1177/0010414010374017 Thomas, S. (1994). How women legislate. Oxford University Press. Thomas, S. (1991). The Impact of Women on State Legislative Policies. The Journal of Politics, 53(4), 958–976. https://doi.org/10.2307/2131862 Wängnerud, L. (2009). Women in Parliaments: Descriptive and Substantive Representation. Annual Review of Political Science, 12(1), 51–69. https://doi.org/10.1146/an- nurev.polisci.11.053106.123839 Wängnerud, L. (2006). A step-wise development: Women in parliament in Sweden. Women in Parliament: Beyond Numbers, 238–248. Weeks, A. C. (2019). Quotas Matter: The Impact of Gender Quota Laws on Work-Family Poli- cies. Mimeo. Women’s substantive representation in the pandemic 96 Wilensky, H. L. (1990). Common problems, divergent policies: An 18-nation study of family policy. Public Affairs Report, 31(3), 1–3. Zarra, A. and Ceron, M. (2021). Covid-19: the place for women in pandemic response in the EU. APSA Preprints. https://doi.org/10.33774/apsa-2021-r44lc Zedník, R. (2020, April 15). A shaken world demands balanced leadership. Medium. https://medium.com/@rickzednk/a-shaken-world-demands-balanced-leadership- ef140e658579. CERON and ZARRA 97 Appendix Table A1. Corpus of official and interest groups’ documents used for the generation of parity-relevant issues Organisation Document type datecivoce letter camera dossier pnrr gender parity in the NRRP giustomezzo analysis giustomezzo hearing giustomezzo letter giustomezzo manifesto giustomezzo proposals halfofit manifesto Ingenere hearing ladynomics hearing Table A2. Gender of MPs by group and overall number of speeches in the corpus Group Proportion of women MPs Total speeches centro 0.5 4 fdi 0.3214286 28 fi 0.4615385 26 iv 0.5882353 17 lega 0.25 28 leu 0.4444444 9 m5s 0.3333333 33 misto 0.3673469 49 pd 0.3666667 30 Total 0.375 224 Women’s substantive representation in the pandemic 98 Figure A1. Marginal effects on whether a speech mentions equality across gender of the MPs within the corpus Figure A2. Marginal effects on whether a speech mentions equality across the political group of the MPs within the corpus .1 .2 .3 .4 .5 .6 P r( ge nd er ed s pe ec h) m f Gender of MPs Predictive margins with 95% CIs 0 .2 .4 .6 .8 P r( ge nd er ed s pe ec h) PD M5S IV FI Lega FdI Political Groups Predictive margins with 95% CIs CERON and ZARRA 99 Table A3. Gender of MPs by group and overall number of speeches in the corpus (1) (2) (3) reference: FdI reference: FI reference: PD Parity-relevant speech Parity-relevant speech Parity-relevant speech f 1.689*** f 1.689*** f 1.689*** (0.429) (0.429) (0.429) PD 2.766*** PD 2.926*** M5S -1.876*** (0.718) (0.735) (0.598) M5S 0.889 M5S 1.050 IV -2.219*** (0.706) (0.713) (0.736) IV 0.546 IV 0.706 FI -2.926*** (0.801) (0.798) (0.735) FI -0.160 Lega 1.295* Lega -1.631*** (0.802) (0.741) (0.609) Lega 1.135 FdI 0.160 FdI -2.766*** (0.729) (0.802) (0.718) Constant -2.566*** Constant -2.726*** Constant 0.200 (0.621) (0.649) (0.420) Observations 162 Observations 162 Observations 162 Notes: standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1. Logistic regression including parties with non-negligible number of speeches.