Labour market policy in Italy’s recovery and resilience plan. Same old or a new departure?

ABSTRACT Liberalizing labour market reforms have topped the agenda of structural reforms implemented in Italy over the last two decades, with detrimental effects on employment quality, wage dynamics and productivity. In 2021, Italy’s then Prime Minister, Mario Draghi, promised that the investments outlined in Italy’s National Recovery and Resilience Plan (NRRP) would ‘transform Italy’s labour market’. How and to what extent does the labour market policy agenda enshrined in Italy’s NRRP deviate from the prior trajectory of policy change? What balance of economic, political and class interests does it reflect? And to what extent does it adequately tackle the long-standing challenges of Italy’s labour market? This article addresses these questions combining in-depth analysis of the labour market policy measures in Italy’s 2021 NRRP and interviews with experts and elites involved in the policy process. Contrary to claims of discontinuity, the findings highlight substantive continuity of the NRRP labour market policy agenda with the prior trajectory of liberalization. The Plan maintains a narrow focus on supply-side labour market interventions – primarily the strengthening of active labour market policies (ALMPs) – without re-regulatory interventions to tackle labour market insecurity or wage stagnation. Exogenous conditionality and domestic political dynamics that systematically advanced the preferences of employer organizations in the design of the NRRP account for the limited extent of policy change. Due to the neglect of demand-side labour market interventions and the uncertainties surrounding the implementation of the ALMP reforms, the transformatory potential of the NRRP’s labour market agenda is likely to remain limited.


Introduction 1
Liberalizing labour-market reforms have been at the centre of the agenda of structural reforms that Italian policymakers have pursued for the last three decades to stimulate employment and boost competitiveness (Afonso et al. 2021;Sacchi and Vesan 2015). In the aftermath of the Great Recession, the pressures arising from external creditors and from the Eurozone's new economic governance mechanisms combined with the reformist ambitions of domestic policymakers to mark a critical juncture in the trajectory of Italian labour-market liberalization, amid high political contestation (Ferragina and Arrigoni 2021;Picot and Tassinari 2017). The effects of the labour-market liberalization agenda on employment creation, employment quality and productivity have been limited if not detrimental. Critical issues persist in the Italian labour market with regard to segmentation, youth unemployment, low female labour-force participation, stagnant wages and the high incidence of informal and precarious employment -all issues made more acute by the COVID-19 crisis. In early 2021, Italy's newly appointed Prime Minister, Mario Draghi, promised that with the investments outlined in Italy's National Recovery and Resilience Plan (NRRP), his government would 'transform Italy's labour market '. 2 This article contributes to the debate around the policy and politics of Italy's NRRP by focusing on its labour market and employment-policy component, concentrated in the fifth 'Mission' of the Plan. It addresses the following questions. What is the overall thrust and trajectory of intervention embodied in Italy's NRRP in the field of labour-market policy? Do the measures deviate from or reinforce the prior trajectory of policy change in this field -and if so, in what direction? What is the configuration of class and sectoral interests that the Plan's agenda of labour-market reforms entails? And what are the potential impacts and shortcomings of the envisaged measures? These questions are tackled through an in-depth case-study approach that combines analysis of the relevant policy measures included in Italy's NRRP and of the related implementation guidelines and monitoring reports, with original qualitative interview data. These were gathered through 14 elite interviews conducted by the author with key policymakers, peak-level representatives of social-partner organizations and experts/policy advisers during 2021 and 2022. 3 The article consists of four sections. The first summarizes the trajectory of labourmarket policy (LMP) change in Italy up until the COVID-19 crisis, providing the background to the interventions outlined in the NRRP. The second analyses the content of the NRRP's main measures in the field of labour and employment policy. The third discusses the politics of the NRRP formulation process in this policy area. The fourth concludes by assessing the trajectory of policy change and its main criticalities, and by discussing future prospects.

Labour-market policy in Italy from the great financial crisis to COVID-19
The liberalization of labour market and employment policy has headed the agenda of structural reforms of successive Italian governments for the last three decades. The first two reform cycles, in the 1990s and early 2000s, followed a trajectory of labour-market deregulation 'at the margins', liberalizing the use of atypical contracts and temporary agency work. This was not accompanied by a restructuring of the unemployment benefits system (Picot 2013), thus leaving large segments of the non-standard labour force in a situation of 'flex-insecurity' (Berton, Richiardi, and Sacchi 2012). Active labour-market policies (ALMPs) also remained underdeveloped. A nominal strengthening of the jobsearch conditions imposed on benefit recipients was introduced -without however matching it with adequate investment in public employment services (PES), which remained under-staffed and with significant regional disparities in performance (Sacchi and Vesan 2015).
The third cycle of labour-market liberalization took place in the aftermath of the Eurozone crisis of 2011-2012 -first with the Fornero labour-market reforms of 2012, and then with the Jobs Act of 2015. This latter wave of reforms -implemented partly as a result of conditionality arising from the EU (Sacchi 2015) -went straight to the 'core' of the labour market. It entailed an unprecedented limitation of the reach of employment protection legislation (EPL) for workers on open-ended contracts in large firms, who previously enjoyed considerable protection against unfair dismissal guaranteed by 'Article 18' of the 1970 Workers' Statute (Ferragina and Arrigoni 2021;Picot and Tassinari 2017). This did not amount to a substantive de-segmentation of the Italian labour market, as the use of fixed-term contracts was also further deregulated, contributing to the persistently high share of atypical employment and to increasing wage inequality in the post-crisis period (Giangregorio and Fana 2021).
In contrast with the measures of the two previous two reform cycles, however, the 2015 Jobs Act -at least in some aspects of policy design and in the narrative surrounding its implementation (Galanti and Sacchi 2019) -embodied a vision of LMP inspired by the rhetorical principles of 'flexicurity' (Wilthagen and Tros 2004) and of recalibration towards 'social investment' (Morel, Palier, and Palme 2012). This amounted, in the vision of the reform's supporters, to a model of liberalized labour markets where the focus of safety nets is not placed upon the preservation of the job, but rather centred on the provision of security in the labour market -to be provided both through incomeprotection measures during spells of unemployment, and through ALMPs that should facilitate individual transitions towards new employment through re-skilling and orientation services.
Accordingly, EPL liberalization was accompanied by some 'expansionary' reforms (Vesan and Ronchi 2019). The unemployment benefits system was restructured to expand coverage for labour-market 'outsiders' -a reform trajectory which has been described as 'embedding flexibilisation' (Picot and Tassinari 2017). A far-reaching reform of the ALMP system was also envisaged, through the introduction of a National Employment Agency (ANPAL), supposed to combat the regional fragmentation of PES; and the integration of unemployment benefit and ALMP management through a onestop shop approach (Sacchi and Vesan 2015;Giuliani 2022).
The ALMP component of the reform was nonetheless poorly implemented. First, ALMPs remained inadequately financed, and initiatives to improve the performance of Job Centres (Centri per l'Impiego) scant. Under the Renzi government, the main instrument used to favour the creation of permanent employment was rather the provision of incentives to employers lowering the cost of social security contributions (Giuliani 2022, 93). Second, the failure of the 2016 constitutional referendum proposals impeded returning to the central government ALMPs under the control of the regions. Hence, ANPAL did not have adequate powers to implement a centralized strategy, and significant differences between northern and southern regions persisted (Giuliani 2022, 93).
Italy's ALMP system continued therefore to suffer from several weaknesses: understaffing, inadequate coordination with related services and across administrative levels, underdeveloped capacity to support workers' placement in new jobs. Since 2016, the weakness of Italy's ALMPs has been the object of repeated criticism by the European Commission in the Country Specific Recommendations (CSRs) issued in the context of the European Semester.
In 2019, the 'yellow-green' government staffed by the Movimento Cinque Stelle (Fivestar Movement, M5s) and the Lega (League) and led by Giseppe Conte (the 'Conte I' cabinet) introduced a novel universal social assistance scheme not tied to previous employment history. This was the so-called Citizenship Income ('Reddito di Cittadinanza'), a flagship proposal of the M5s. The Government also introduced an EPL reform, the so-called 'dignity decree', also sponsored by the M5s, reducing the maximum number of renewals allowed for temporary contracts (Bulfone and Tassinari 2021, 531). Although the re-regulatory reach of the 'dignity decree' was limited, in order not to alienate excessively the electoral base of the League among small-and mediumsized enterprises (SMEs) (Afonso and Bulfone 2019; Bulfone and Tassinari 2021), these measures deviated partly from the previous labour-market liberalization trajectory. Hence, they attracted strong opposition from employer organizations and political parties of a centrist and centre-right orientation.
To overcome the resistance to selective universalistic anti-poverty measures, which has historically characterized the Italian welfare state, the Citizenship Income was designed -at least in principle -as a labour-market re-insertion measure with activation requirements (Busilacchi, Gallo, and Luppi 2021). However, the accumulated backwardness and regional disparities in ALMP capacity meant that implementation of its activation component remained uneven (ibid.). Accordingly, an intense public debate emerged around the supposedly problematic lack of effective job-search obligations on recipients of the Citizenship Income, making the measure hyper-politicized and controversial (Natili, Jessoula, and Caizzi 2021). Employers' organizations, centrist and centre-right parties started demanding the introduction of stronger 'workfare-type' (Dingeldey 2007) activation requirements for recipients and a shift away from excessively 'passive' income protection, condemning their absence as responsible for creating 'inactivity traps' and thus labour shortages, especially in low-paid sectors.
From 2019, the recommendations of the European Commission to strengthen the link between the receipt of income assistance and ALMPs became more specific and extensive. This emphasis on activation and skills development reflected a broader post-crisis shift in the EU's discourse and policy recommendations. These moved away from an explicit emphasis on liberalization-qua-deregulation (which had arguably reached its limits, at least in Southern Europe), towards promoting more 'enabling' LMP centred around ALMPs and social investment policies (Zeitlin and Vanhercke 2015; Vesan and Pansardi 2021) -which represent the 'soft', compensatory side of liberalization and that contribute to its sustainability in the long run.
Against this background, the outbreak of the COVID-19 pandemic in March 2020 severely aggravated extant labour-market challenges. As a consequence of the severe downturn in 2020 (−8.9% of GDP), in the first three quarters of 2020 the equivalent of 2.4 million full-time jobs were lost (ISTAT 2021). Overseeing the policy response to the pandemic was the Conte II cabinet, a coalition between the M5s, the centre-left Partito Democratico (Democratic Party, PD), and two smaller parties: the centre-left Liberi e Uguali (Free and Equal, LeU) and Matteo Renzi's centrist liberal party, Italia Viva (IV). In line with its broadly centre-left orientation, and often against the resistance of IV Ministers, 4 the Conte II government implemented several protective employment policy measures to deal with the consequences of the pandemic -among which a massive expansion of short-time work schemes (CIG), heavily subsidized by the state, and a temporary ban on dismissals.
These measures successfully prevented large-scale employment destruction and an increase in unemployment, which remained stable at around 10% (Ebbinghaus and Lehner 2022, 11). Employment destruction was however particularly concentrated among the weakest segments of the labour market: temporary and solo self-employed workers (not sheltered by the ban on dismissals), women and young people (ISTAT 2021). The dynamics of labour-market recovery from 2021 then exhibited a problematic predominance of precarious contracts and stubborn wage stagnation (ISTAT 2021).
Between autumn 2020 and the initial months of 2021, as Italy's NRRP was being finalized, significant pressures on the Conte II government intensified. These came especially from employers, but also from opposition parties and from IV ministers. These aimed at overcoming the extant focus on 'passive' forms of labour-market protection (especially the ban on dismissals, but also the use of short-time working schemes to prevent restructuring) and facilitating the return to a 'normal' dynamic of labour market functioning, with freedom for employers to dismiss, whilst also liberalizing the use of temporary contracts to facilitate employment creation. 5 This went alongside growing demands for employment policy to shift to a focus on supporting displaced workers to manage their transitions in the labour market -by strengthening ALMPs -rather than preventing transitions through a logic of job preservation (Confindustria 2021). The Conte II government found it increasingly difficult to build compromises within its divided coalition, whilst facing strong pressures from employer organizations. Eventually, the impossibility of managing this tension resulted in a political crisis leading to the collapse of the Conte II cabinet, triggered by IV's withdrawal of parliamentary support, and to its replacement by Draghi's technocratic cabinet in February 2021. It is against this background that the content of the NRRP in the field of LMP, analysed in the next section, must be contextualized.

Labour market and employment policy in Italy's NRRP: contents and goals
The content of Italy's NRRP in terms of labour market and employment policy reflects a supply-side interpretation of labour-market dynamics, one that identifies the main problems of Italy's labour market and employment performance as resulting from a misallocation of resources rather than from insufficient demand or investment. Accordingly, most measures aim to intervene in the labour market by 'smoothening'more or less forcefully -labour-market transitions, accompanying rather than guiding the operation of market forces.
The main objectives of the LMP interventions outlined in the Plan's Mission Five are: 1) strengthening ALMPs and professional training, to support unemployed workers' 'employability'; 2) strengthening of the PES system; 3) favouring women's employment, in particular through the promotion of female entrepreneurship; and 4) favouring skills acquisition to reduce skills mismatch. To achieve these goals, the plan earmarks spending of €6 billion, roughly 3.4% of the Plan's total resources (Orsitto 2021). The measures include two structural reforms: a reform of ALMPs (which receives the lion's share of the funds: €4.4 billion) and a national plan (with no resource allocation) to combat informal employment. They also include five investment priorities: strengthening of PES (€0.6 billion); incentives for female entrepreneurship (€0.4 billion); creation of a system for certification of gender equality in the labour market (€0.01 billion); strengthening of the dual training system (€0.6 billion), and strengthening of the youth civil service (€0.65 billion).
The most significant intervention in terms of resource commitment and potential institutional innovation is the structural reform of ALMPs, which aims to address some of the historic weaknesses of the system (Busilacchi, Gallo, and Luppi 2021). The reform aims to support the employability of the unemployed and of individuals going through 'labour market transitions', following in the spirit of the CSRs that Italy received through the European Semester in the preceding years. The flagship measure, formalized in legislation in November 2021, 6 is a new 'National Programme for the Guarantee of Workers' Employability' (Programma Nazionale per la Garanzia Occupabilità dei Lavoratori, GOL). The GOL programme envisages introducing a unified system for onboarding, into the ALMP system, all individuals either unemployed or at risk of unemployment.
One crucial difference as compared with the status quo ante of Italy's ALMPs, is that the new programme would include, under a unified ALMP umbrella, all recipients of different forms of unemployment support. These include recipients of the Citizenship Income and of 'standard' unemployment benefits, workers on short-time work (CIG) schemes, and other vulnerable groups in the labour market (such as NEETs), previously dealt with under different channels, or not subject to activation requirements at all. The majority of the GOL measures could be considered, taken in isolation, as 'enabling' (Dingeldey 2007) ALMPs aimed at human capital acquisition. However, the guidelines of GOL also envisage strengthening conditionality for benefit recipients, in a direction of greater workfare. Indeed, for benefit recipients without an ongoing employment relationship (i.e. not on short-time work schemes) and for the recipients of the Citizenship Income, participation in the new ALMP measures becomes compulsory, and nonparticipation implies loss of eligibility to benefits. Hence, overall, the thrust of the GOL programme shifts the Italian ALMP system in a more 'workfarist', commodifying direction. Implementation of the GOL programme is delegated to the regional authorities. These retain competences for ALMPs, and have to draw up regional action and implementation plans under the supervision of the national ALMP agency, ANPAL. This is a potentially crucial weakness when it comes to implementation, as the capacities of the regional PES are very unequal as between northern and southern regions. To address the structural weaknesses of Italy's PES, Mission Five of the NRRP includes an investment priority of €0.6 billion aimed at capacity building for the Job Centres. These measures act as a complement to those outlined in the 'Extraordinary Plan for the Strengthening of PES' ('Piano straordinario di potenziamento dei centri per l'impiego e delle politiche attive del lavoro'), already approved in mid-2020, which provided for the hiring of 11,600 new Job Centre staff. 7 The other measures outlined in M5C1 are residual in terms of resource commitment, and limited with regard to the depth of policy change envisaged. Of the investments aimed specifically at favouring the labour-market insertion of young people and lowskilled adults, the main one aims to increase the reach of the 'dual training system' and of the apprenticeship system of job-based training as a channel of labour market insertion of young people. The 'dual system' indicates forms of vocational education and training (VET) that involve an alternation between work activities, on-the-job and off-the-job training activities: a type of VET that has historically been very weak in Italy (D'Agostino and Vaccaro 2021). The NRRP does not intervene in the institutional form and governance of the dual system to address some of its long-standing weaknesses: first and foremost, the lack of systematic involvement of and coordination with the social partners in planning VET provision and addressing emerging skills needs. Rather, it limits itself to allocating more resources to the regions to provide more on-the-job training activities in better coordination with firms, and to strengthening the delivery of professional training in marginalized areas, with the aim of reaching in particular young people and adults without a secondary school qualification to favour their labour-market insertion. 8 The investment measures aimed at favouring the labour-market integration of women are also very narrowly conceived. They provide for the allocation of €0.4 billion in support of female entrepreneurship, aiming to restructure the existing systems of support and incentives, providing funding to support the growth of women-led start-ups through mentoring and technical-managerial support, and setting up a 'Women's Enterprise Fund' to finance female-led entrepreneurial initiatives. 9 The only measure included in the first chapter of Mission Five explicitly focused on the issue of quality of work and working conditions is the second structural reform -i.e. the planned formulation of a 'National plan to combat informal employment' to be drawn up by October 2022. The Plan should comprise several actions aimed at combating informal employment, which is particularly widespread in the Italian economy. This reform however receives no earmarked funding, and stands out as potentially very weak in its reach.
Noteworthy for their absence are all issues relating to the quality of employment. With the exception of the programmatic measures on combatting informal employment, all other issues relating to employment quality and working conditions are excluded from the NRRP's interventions. They include the re-regulation of atypical contractual forms and of outsourcing and subcontracting rules; the reduction of labour market temporality, and measures to combat social dumping.
Unemployment benefits and the broader system of income support also remained outside the Plan's perimeter, although policy change did take place in this field. Proposals for a comprehensive reform of the income support system were drawn up by a commission of experts, the so-called Catalfo commission, convened in July 2020 by the Labour Minister, Nunzia Catalfo, during the Conte II government. The proposals of the experts' commission deviated from the supply-side logic of the NRRP reforms, as they advocated a further universalization of income protection for all categories of workers and the unemployed, including solo self-employed workers. They also advocated simplifying the system, increasing its generosity and improving its linkages with both ALMPs and industrial policies (for further details, see Barbieri and Guarascio 2021). The expansion of income support was advocated not only for reasons of social justice but also because of its potentially beneficial impact on internal demand and thus employment creation (Barbieri and Guarascio 2021). Unsurprisingly, these proposals encountered the opposition of most political parties with the exception of the M5s, due to their potential budgetary implications, and the drawing up of the reform stalled over time (Guarascio 2021). These reforms were not included in the NRRP, but only referred to in very general terms in the Plan (p.32, p.88) as 'concurrent to the realization of the general objectives of the plan' (so-called riforme di accompagnamento). A structural re-organization of the income-support system in a trajectory of greater universalization was instead eventually included in the Finance Law for 2022 (law 234/2021), but with a much less ambitious universalizing reach than the Commission proposals had envisaged (Guarascio 2021). 10

The politics of LMP in the NRRP
Analysis of the policy process, carried out by triangulating documentary and interview evidence, highlights that the politics of LMP in the NRRP have been shaped by the interplay between supra-national conditionality arising from the European Commission, and domestic political dynamics conferring great influence on centrist political forces and employers' interests.
According to interviewees, the influence of the European Commission and the external conditionality it exercised on the Plan's design played an important role in shaping policy content. Several interviewees remarked that the 'conditioned' nature of the Plan and the bilateral interactions with the Commission during its formulation impacted significantly on the policy orientation, 11 drawing the emphasis towards ALMPs and activation -on which Italy had received numerous CSRs within the European Semester in previous years -and away from other more controversial regulatory interventions, especially in EPL. The NRRP guidelines also dictated that the financial resources could not be used to fund permanent increases in social security expenditure. Accordingly, structural interventions on the income support system could not, by design, be included in the NRRP reforms. 12 One expert interviewed reported that the European Commission was also wary of the unemployment benefit reform proposal of the Catalfo Commission, as it was concerned about the possible budgetary implications. Hence, these did not feature more prominently among the 'complementary' interventions mentioned in the NRRP. 13 Domestic politics also played a role, especially in accounting for what remained excluded from the plan and for the details of policy design. With regard to intragovernmental political dynamics, despite being politically very different, both the Conte II and the technocratic Draghi cabinet were heterogeneous coalitions comprising diverse political forces with hardly compatible electoral bases and ideological orientations. For the Conte II government, the divergence between the M5s and the 'left' component of the PD on the one hand, more oriented towards a 'protective' approach to LMP, and the 'centrist' component of the PD and IV, of more liberal orientation, on the other, created difficulties in crafting policy compromises within the coalition. In particular, the centrist IV -aligned to entrepreneurial interests for ideological and electoral reasons -opposed any re-regulatory interventions in EPL. In the Draghi government, supported by an all-parliamentary grand coalition (with the sole exception of the radical right Fratelli d'Italia (Brothers of Italy, FdI)), the difficulty of finding compromises was even more acute, and centre-left oriented components were even more marginalized. Furthermore, the broad 'pro-business' policy orientation and 'supply side' understanding of the problems of Italy's labour market was also shared, within the Draghi cabinet, 'at the top', by the Prime Minister -who, according to interview sources, played a central role in setting the fundamentals of the government's policy orientationand by his team of policy advisors responsible for finalizing the plan, mostly drawn from circles of liberal neoclassical economists. 14 The combination of political divisions and the ideational predominance of a supply-side paradigm contributed to keeping re-regulatory labour market interventions firmly off the agenda. The proposals of the Catalfo Commission for reforming the income-support system in a universalizing direction, which anyway lay beyond the scope of the Plan strictly defined, also proved politically controversial within the Draghi cabinet due to their possible budgetary implications (Barbieri and Guarascio 2021).They were thus pushed to the arena of cabinet and parliamentary negotiations, and eventually diluted in their universalizing reach (Guarascio 2021). 15 In this context, ALMPs emerged as the 'minimum common denominator' around which diverse political forces in both cabinets could find a politically viable minimal consensus. As a result of the strong EU emphasis on this front and of the objective underdevelopment of this policy field in Italy, the importance of investments in ALMPs emerged as an issue where a general consensus could be found across parties and across the labour-capital divide. 16 This also reflects the nature of the policy field, which is a 'profoundly versatile set of "multi-purpose tools" that policymakers across the political spectrum can use as a means to very different distributional ends' (Cronert 2022, 210). Hence, whilst all parties could converge on the importance of general investments in ALMPs, partisan differences were more relevant in defining the precise content and orientation of the ALMP measures contained therein. In this regard, the mix between 'supportive' ALMPs (training measures) and 'demanding', workfare-type ALMPs (compulsory job search and activation as a condition for the receipt of benefits) reflect the cross-partisan composition of the two cabinets, where a balance between diverging orientations had to be struck. Interestingly, with the exception of some policy details, the broad contents of Mission 5 remained broadly the same between the first draft of the NRRP drawn up by the Conte II cabinet and the final version of the Draghi cabinet -thus underscoring how the influence of the Commission's CSRs and the political equilibria were comparable across the two cabinets in this policy area.
The second noteworthy dynamic that accounts for the 'supply side' skew of the NRRP's LMP measures is the relationship between the governments and major stakeholders in the policy area, i.e. unions and employer organizations. LMP in Italy has historically been a field of intense discussion between governments, the three main tradeunion confederations (CGIL, CISL and UIL) and the main employers' organizations (the manufacturing employers' confederation, Confindustria, the service-sector employers' confederation, Confcommercio, the craft sector confederations, Confartigianato and CNA, and other smaller organizations).
After the period of 'social pacts' in the 1990s, tripartite social dialogue in Italy stood on uncertain institutional footing (Tassinari and Sacchi 2021). During the Eurozone crisis and in its aftermath, cabinets of both technocratic (Monti) and centrist/centre-left orientation (Renzi) explicitly rejected social concertation when crafting their liberalizing reforms, condemning it as an obstacle to structural transformation which granted excessive veto powers to producer groups and led to distortive outcomes (Tassinari 2021). Social concertation practices were, however, often re-activated by the same cabinets at times of weakness in order to extract political legitimation (Tassinari and Sacchi 2021). During the COVID-19 crisis, social dialogue with unions and employer organizations had been used by the Conte II cabinet to govern specific aspects of the health emergency, in particular the re-organization of production, where the social partners' know-how was crucial; whilst 'important social and employment policy measures were left to political and technocratic players' (Meardi and Tassinari 2022, 8-10).
The process of NRRP formulation was, in this regard, in continuity with the previous phase. Social dialogue was virtually absent, despite labour policies being an area of direct competence of the social partners. For both cabinets, the policy design process was 'topdown' and centralized in the hands of the executive -in particular the Prime Minister's Office (Presidenza del Consiglio) and the Ministry of the Economy -supported by a narrow pool of technical advisors from different ministries, and from institutions such as Bankitalia and academia.
During the Conte II cabinet, interviewees reported that after an initial moment of open consultation organized by the Government with stakeholders in June 2020, there was effectively no other formal moment of joint discussion. The main channel of union influence took place through participation in hearings held by the relevant parliamentary committees. Interviewees hypothesized that the absence of structured dialogue arose from the Cabinet's internal divisions and challenges in crafting a coherent policy agenda in a phase of intense political tensions. 17 Dialogue on the contents of the Plan was described as marginally more substantive, but still very centralized in the case of the Draghi cabinet, which took over in February 2021 and brought the process to completion. Interviewees perceived that the Draghi government also had limited political room for manoeuvre to seek substantive compromises with the unions, because the main challenge was finding consensus within the heterogenous governmental coalition. Once compromise solutions had been found therein, there was no space to subject these to interlocution with the unions -despite attempts by the Minister of Labour to maintain bipartite dialogue channels with them. 18 The lack of substantive dialogue and discussion with union organizations impacted on the contents of the Plan, as the main policy issues of primary importance to the more leftleaning union confederations, CGIL and UIL, remained side-lined. First, union representatives from CGIL and UIL complained that the Plan was excessively focused on ALMPs, considered in isolation from the broader labour-market dynamics. 19 Concerns were also expressed about the unrealistic targets around increasing PES capacity (which seemed disproportionate considering the weak starting point), and about the risk that, in order to meet the (over-ambitious) targets, Job Centres and ALMP providers would offer individuals low quality training and work experience (e.g. low paid traineeships) that would have little impact on employability and contribute to creating low-quality employment.
More fundamentally, representatives of both CGIL and UIL objected to the exclusive supply-side focus of the Plan, which excluded from the scope of the NRRP issues of employment quality, working conditions and demand-side measures to stimulate the creation of quality employment. 20 Interventions to counter the spread of employment insecurity in the post-crisis recovery were seen as particularly urgent. CISL, for its part, put the emphasis, rather, on the absence of stronger social conditionality around employment creation and the hiring of women and young people by the firms that were recipients of the NRRP's structural investments (Bottos 2021).
Despite not having managed to intervene incisively on the overall design of the reform package, union representatives nonetheless extracted some concessions on issues of concern via bilateral talks with the Labour Ministry. The first concession was the maintenance, within GOL, of an important role for the PES. Whilst employer organizations would have preferred a greater delegation of ALMPs and skills training to private actors, such as large temporary work agencies, staffing agencies and private training providers, 21 the PES retains in the GOL programme an important role for coordinating training measures offered by other private actors, in a logic of cooperation between public and private, rather than of substitution of the former by the latter. 22 The second concession was the re-insertion, during the very last stages of the Plan's drafting in April 2021, of the 'National plan for combating informal employment', initially omitted. This was attributed to the influence of the unions on the minister, Andrea Orlando. 23 Nonetheless, union representatives complained that the plan had not been allocated any financial resources, and that the first concrete measures were planned for very far ahead in time, in 2025-2026. 24 The evaluation of employers' organizations was quite different. Overall, employer organizations reported that, whilst also not being formally involved in formulation of the Plan, the content of the labour-market measures was broadly aligned with their vision of what 'needed to be done'. 25 Employer organizations espoused a supply-side vision of LMP priorities very close to that of the NRRP. 26 This alignment reflects both the ideological convergence and the political closeness between employers and the Draghi cabinet, considered by all interviewees as generally attentive to the needs of entrepreneurs and firms. The lack of social dialogue did not, hence, constitute a source of major concern, as generally employers trusted that the orientation of the Draghi government would be aligned with their priorities. 27 To the extent that they expressed concerns about the NRRP, it was primarily with regard to uncertainty around its implementation. 28 Employers also reported full agreement with the omission of measures intervening in labour-market re-regulation. These were regarded as 'not suitable' for a phase of economic recovery, where excessively rigid regulation of atypical contracts and other forms of labour contracting such as staff leasing would be counter-productive. One interviewee said, 'What we need are instruments and contracts that make transitions fluid, rather than making contractual instruments more rigid' 29 -while others expressed a preference for interventions further reducing the costs of temporary contracts. 30 The main policy issues on which employer organizations wished the Plan had been more ambitious was the specific design of ALMP measures. Preferences were expressed for stronger linkages between the receipt of income support and compulsory training; whilst others expressed explicit support for more stringent workfare-type measures -i.e. compulsory job search and acceptance of job offers by benefit recipients -to deal with issues of workforce shortages and skills mismatches. 31 Employers also expressed a preference, not fully reflected in the Plan, for a greater involvement of private actorstemporary work agencies and staffing agencies -in the delivery of ALMPs, considered to be more effective than Job Centres in identifying employers' skills needs and vacancies due to their market orientation.
Third, some of the 'missing points' in the Plan can be accounted for by the complexities of the specific policy field at hand, too complicated to be tackled in the time-limited and fragmented process of formulation of the NRRP -made even more discontinuous by the change of government that occurred in the middle of the process. This was, for instance, the case of the 'non-reform' of the governance and institutional design of the VET system. A profound re-configuration would have required a complex and potentially controversial revision of competences between the central government and the regions. It would also have required coordination between numerous diverse actors with heterogeneous preferences -such as training agencies, third-sector organizations, regional authorities and the PES -that lacked appropriate institutional fora to coordinate and draw up reform proposals in the short timeframe available (Salerno 2021). Hence, this complex issue remained untouched, and the bulk of interventions on vocational education focused instead on the state school system and on strengthening tertiary vocational education via reform of the Advanced Technical Institutes (Istituti Tecnici Superiori, ITS), comprised under Mission 4.

Discussion and conclusions
This article began with three objectives: first, assessing the trajectory and depth of policy change in the LMP component of Italy's NRRP compared to the pre-pandemic period; second, investigating the politics accounting for its contents, and the balance of economic-political interests that it reflects; third, assessing the NRRP's adequacy and potential shortcomings in addressing Italy's labour-market challenges.
With regard to the overall policy trajectory, the analysis has highlighted that the depth of change in policy goals and overall orientation of LMP as compared to the pre-COVID period is limited. The NRRP focuses almost exclusively on introducing measures that intervene on the supply side of the labour market. This approach to LMP aims to 'embed' a flexibilised labour market by equipping individuals to cope better with labour-market risks and favour 'smoother' transitions through increased conditionality, support for individualized human-capital acquisition, and reduction of 'skill mismatches', without directly hindering the operation of market forces and employers' discretion to hire, fire and adjust labour utilization according to firms' operational and economic needs. The role of the state shifts therefore to accompanying workers in navigating transitions and supporting market-induced transformations in labour-market structure, rather than seeking to govern them with more explicit protective interventions. As before COVID-19, re-regulatory interventions in EPL in a de-liberalizing trajectory remain firmly off the table (Bulfone and Tassinari 2021).
The only considerable discontinuity with the past legacies of Italian LMP is the envisaged reform of ALMPs, which, if implemented, would constitute a new departure with regard to the extant design of Italy's LMP institutions, where ALMPs have historically been under-developed. The overall focus on facilitating labour-market transitions and human capital acquisition is nonetheless in continuity both with the recommendations issued to Italy by the European Commission in the context of the European Semester CSRs since 2016, and with the 'social investment' oriented vision of labourmarket governance already enshrined in the 2015 Jobs Act. In this regard, the NRRP stands in continuity with the pre-crisis trajectory of the 'neoliberalisation' of Italian LMP (Ferragina and Arrigoni 2021) -although with a different focus as compared with previous reform waves. It reflects a sort of 'neoliberalisation 2.0' where the focus shifts from EPL de-regulation, by now brought to completion, towards activation, which constitutes the next logical step. To be sure, the NRRP does not exhaust the complex of LMP interventions implemented in Italy post-COVID, which have also comprised reform of the unemployment benefits system included in the 2022 Finance Law. This intervention partly deviates from the thrust of the LMP policy agenda enshrined in the NRRP, insofar as it increases protection coverage and thus potentially contributes to boosting demand. However, concerns around budgetary sustainability within the Draghi cabinet have limited the universalizing ambition of the reform, thus curbing its pathdeparting potential and its demand-boosting capacity.
With regard to the politics of LMP in the NRRP, the analysis has highlighted how the overall supply-side orientation of the NRRP's LMP component resulted from the interplay between supra-national conditionality arising from the European Commission, and domestic political dynamics, which conferred considerable influence on centrist political forces and employers' interests in the Plan's design. Exogenous conditionality -in particular, the requirement that the Plan be aligned with the CSRs received in the previous years -put activation and human capital acquisition at the top of the agenda. This top-down element of the Plan's design was reflected in the domestic dynamics of policy-making, centralized in the hands of the executive, with no space for social dialogue with the social partners and limited influence of interest groups on the Plan's content, as also found by Bitonti et al. (2021). This represents another point of similarity with the pre-pandemic dynamics of marginalization of social concertation in policymaking (Tassinari and Sacchi 2021).
The complex balance of power within the Conte II and Draghi cabinets further contributed to taking deviations from the supply-side orientation of LMP measures off the table. During the Conte II cabinet, the weakness of an internally divided coalition, and the veto power of the centrist IV within it, made it politically difficult to put any reregulatory or demand-side interventions averse to employers' preferences on the agenda of LMP reform. During the grand coalition Draghi cabinet, centre-left forces became even more constrained in their room for manoeuvre, and the orientation towards a supply-side understanding of LMP further entrenched among key policymakers responsible for finalizing the Plan, first and foremost the Prime Minister himself. Hence, the focus settled on ALMPs as a 'minimum common denominator' on which all political forces could to some extent converge, leaving more controversial issues untouched.
The focus on supply-side policy measures reflects not only an ideational consensus among EU institutions and Italian elites, but also the broader balance of class power in post-crisis Italy. On the one hand, the more left-oriented components of the union movement, the CGIL and UIL, have been on the defensive and increasingly marginalized from the policy process, thus leaving their demands for demand-side interventions and restoration of labour rights disempowered. On the other hand, the policy preferences of employer organizations, consistent with a supply-side oriented vision of labour market regulation, have been closely reflected in the contents and omissions of the Plan. In this respect too, the politics of LMP in Italy's NRRP are in continuity with pre-pandemic dynamics of union disempowerment and overall dominance of employer interests (cf. on this also Bitonti et al. 2021).
Many uncertainties surround the implementation and outcomes of the LMP component of the NRRP. Problematic issues pertain to the limited disbursement of resources for LMP (especially the lack of resources for interventions against informal employment); the complications relating to the overlap of regional and national competences in the administration of ALMPs; the significant regional disparities in PES capacity, and the limited involvement of the social partners in the governance of the ALMP and training system. Whether the utilization of the funds will reduce or further re-entrench extant territorial disparities in public service delivery, and whether it will result in genuinely high-quality interventions, is therefore a crucial open question. Considering past policy failures (cf. Busilacchi, Gallo, and Luppi 2021), there are reasons to suspect that the pathbreaking shift in ALMP development might remain only on paper.
The broader issue hanging over the potential impacts of the NRRP on the Italian labour market, however, pertains arguably to its exclusive supply-side orientation. The quality of the employment opportunities created in the post-COVID recovery has been very low, with stagnant wage dynamics, which suppress domestic demand even further and thus slow down the recovery. Without adequate interventions on the regulatory side to increase employment quality (with regard to wages, employment security and working time), and without a more systematic vision of demand-side interventions for quality employment creation and broader economic development, serious doubts remain about the quantity and quality of jobs that the revamped ALMP system will direct transitions towards. Arguably, some of the most problematic, topical issues facing the Italian labour market in the post-COVID world have remained, for a complex of ideological and political reasons, outside of the scope of the NRRP's interventions. Without addressing them, it seems unlikely that the goal of 'transforming Italy's labour market' will ever be meaningfully achieved.