Summary of the impact

Professor Pedro Martins' research expertise is on labour economics, including dismissals, wages and social returns of education. From June 2011 until February 2013, he was seconded from Queen Mary to undertake the key role of Portuguese Secretary of State for Employment. Informed by his research, he initiated and implemented a programme of effective labour market legislative reforms over a relatively short period of time. A reformed labour code (four changes of law) and several ordnances, resolutions and other policy change were the outcome, affecting individual dismissals, working time, collective bargaining, training and active labour market policies. His aim was to reduce unemployment by partly deregulating the labour market, so reducing the disincentives for firms to employ people, and by promoting active labour market policies to raise skill levels within the economy. The reforms affected approximately five million people. Following the reforms, Portugal moved to eighth place (in 2009 it was first) out of 34 countries in terms of the strength of permanent employees' protection (OECD Employment Outlook 2012); and unemployment fell, from 17.7% to 15.6%, and GDP increased, by 1.3%, between the first and third quarters of 2013.

Underpinning research

An effective labour market reform programme must be based on a strong evidence-based understanding of how these markets function. It is imperative to achieve the right balance between protecting employment rights and avoiding over-regulating employers. Strong employment protection legislation increases job security for those already in employment. Yet such legislation, by placing constraints on the circumstances under which employees can be dismissed, tends to discourage employers from recruiting new employees. Thus too much legislated employment protection is likely to reduce the employment chances of those seeking a first or a new job. It may also constrain worker and firm performance - and therefore economic growth — as workers may exert less effort if they feel that their employment is permanent (moral hazard).

Martins (2009a), in a key contribution on this issue, found that restrictive employment protection had a detrimental impact on company performance. He demonstrated that reducing dismissal constraints on small firms, in the late 1980s in Portugal, had a positive effect on their performance, a conclusion based on detailed longitudinal data on several thousand firms and their workers over a period of over 15 years. The study applied state-of-the-art econometric techniques, namely a difference-in-differences propensity score matching estimator, based on taking as a counterfactual an appropriate subset of larger firms which were not exposed to the same reduction in dismissal costs which applied to smaller firms. The paper's findings, together with other research (Blanchard and Portugal, American Economic Review, 2001) in this area, reinforced the conclusion that labour law reforms that promote greater flexibility make a vital contribution to the promotion of economic growth. Such an effect is particularly pronounced in countries, like Portugal, where labour law had become especially constraining for employers.

The wage determination mechanism is another area of great importance in labour market policy reform. Wage determination mechanisms have a vital function in loosening up labour markets, by equating demand and supply, and thus facilitating the fight against unemployment and the achievement of a more balanced income distribution. Again a good, evidence-based understanding of the forces that shape the wage distribution over time must underpin policy reforms. This underpinning was provided in four areas: (a) schooling (Martins and Pereira, 2004, and Martins and Jin, 2010), (b) rent sharing and collective bargaining (Martins, 2009b), (c) multinational firms (Hijzen et al, 2013), and (d) the business cycle (Martins et al, 2012). Martins and Pereira (2004), using comparative data on the wage returns in relation to individuals' schooling in a large number of countries, found a wide range of average rates. Martins and Jin (2010) examined the externalities of education by presenting theoretical results and empirical evidence about the wage returns to co-workers' schooling. Martins (2009b) assessed the extent to which workers receive part of the rents collected by firms through a process of profit sharing consistent with collective bargaining. Hijzen et al (2013) contrast the wage and working conditions practices of domestic and foreign firms across different countries, of different levels of development, drawing on acquisitions and divestitures as well as spells of worker mobility to circumvent issues of spurious correlation. Finally, Martins et al (2012) points to the sensitivity of real wages with respect to the business cycle, as proxied by the unemployment rate.

These papers generated key findings that have led to the impact described below, such as:

• how individual dismissal provisions (and too stringent employment protection legislation in general) have a detrimental impact on firm performance and economic growth;
• the importance of mechanisms that ensure that firms and workers can respond to the business cycle appropriately, particularly during downturns, and can adjust through prices (wages) - or hours of work - rather than quantities (employment levels) thus minimizing the negative impact on unemployment;
• the merits of attracting foreign direct investment in terms of improved wages and work conditions, justifying the removal of several constraints present in the labour code, which other research presents as a negative force in the attraction of foreign direct investment;
• the significant role of unions and work councils in extracting rents from firms, potentially limiting the incentives towards job creation and generating greater wage inequality, in particular in periods of downturn; and
• the very large returns to schooling - and skill provision in general - in Portugal, when compared to other countries, both in private and in social terms, justifying the introduction of the several active labour market policies, which all involved training components.

References to the research

**submitted to REF2014, * submitted to RAE 2008

Details of the impact

By 2011 the Portuguese labour market was suffering from serious structural problems and reform had become a pressing issue. Employment legislation was very generous towards workers in permanent positions, but disadvantaged those entering the labour market for the first time or seeking to change jobs. The Portuguese regulatory framework reinforced segmentation and inequality in the labour market, and attenuated firm performance and the incentives for new or existing firms to expand their workforces.

A labour market reform programme emerged as part of the wider economic and financial adjustment programme agreed between Portugal and the troika of the European Commission, the International Monetary Fund, and the European Central Bank. Informed by the above research, Professor Pedro Martins, as Secretary of State (2011-13), shaped, prepared and presented these employments laws to Parliament, supported by his legal team, and signed almost all the ordnances listed below.

The keystone labour reforms were a reformed labour code embodied in a series of laws passed by the Portuguese Government and currently in force:

• Law 53/2011, of 14 October, which reduces and harmonizes severance pay;
• Law 3/2012, of 10 January, which allows an extraordinary renewal of fixed-term contracts;
• Law 23/2012, of 25 June, which introduces a large number of changes to the labour code, in areas such as individual dismissals, severance pay, working time organization, overtime pay, short-time working schemes, and collective agreements; and
• Resolution of the Council of Ministers 90/2012, of 31 October, which sets criteria for the extension of collective agreements; and a minimum wage freeze.

These reforms reduced severance pay levels towards the European mean, allowed for a longer maximum duration of fixed-term contracts, widened the range of economic circumstances under which permanent workers can be dismissed (Martins, 2009a), facilitated the adoption of short-time working schemes, allowed for the greater involvement of works councils in collective agreements and restricted the formerly widespread extension of collective agreements to firms or workers not affiliated in employer organisations or unions, respectively (Martins, 2009b).

The other vital component of this reform programme involved training and hiring incentives. Any period of structural adjustment and redirection towards tradable sectors requires significant investments in skills, particularly for the unemployed. The research outlined above provided compelling evidence that in Portugal private returns to education tend to be particularly high and that skills generate important positive spillovers (Martins and Pereira, 2004, and Martins and Jin, 2010). Accordingly several ordnances, resolutions and other policies to promote a new active labour market and complementary training policies were launched:

• Ordnance 45/2012, of 13 February, created the "Estímulo 2012" measure;
• Resolution of the Council of Ministers 51-A/2012, of 31 October, created the "Impulso Jovem" programme; Ordnance 207/2012, of 6 July, created an incentive for the acceptance of job offers; Ordnance 225-A/2012, of 31 July, which initiated the "Passaporte Emprego" traineeships; Ordnance 229/2012, of 3 August, provided a social security reimbursement for the hiring of unemployed aged 30 or below;
• Ordnance 3-A/2013, of 4 January, giving employers a social security reimbursement when hiring unemployed workers aged 45 and above; and
• the "Vida Ativa" measure, providing short-duration training modules to specific groups of unemployed.

The amendments to the labour code included provisions applicable to all private sector employees: 3.5 million individuals, from the fourth quarter of 2012. Some provisions apply only to those workers hired from the day when the provision was in force, which correspond to approximately 75,000 individuals per month. For instance, in the case of Law 53/2011, in force since November 2011, will apply to up to 1.5 million individuals (as of June 2013). The active labour market policies apply potentially to all registered unemployed: 810,000 individuals, as of June 2013. In terms of their actual impact, the "Estímulo 2012" active labour market policy (ALMP) supported the hiring of over 15,000 unemployed, while the different ALMPs included under the "Impulso jovem" ALMP programme supported the hiring of over 20,000 unemployed (numbers as of February 2013). The "Vida Ativa" measure involved training provision to over 200,000 unemployed during its first year of operation, since its inception in February 2012.

Following the reforms, Portugal moved to eighth place (in 2009 it was first) out of 34 countries in terms of the strength of permanent employees' protection (OECD Employment Outlook 2012); and unemployment fell, from 17.7% to 15.6%, and GDP increased, by 1.3%, between the first and third quarters of 2013.

Sources to corroborate the impact

Resident Representative in Portugal, IMF

Head of Labour Market Reforms Unit, DG ECFIN, European Commission

Former Minister of the Economy and Employment, Government of Portugal (2011-2013)

Secretary of State for Employment, former President of the Institute for Employment and Training, Portugal

Senior Economist, Bank of Portugal