Chapter 2. Removing demand-side barriers to youth employment in Peru

2.2.1. Youth lose most from size-contingent employers’ social security contributions

Special rules for promoting the creation of micro and small enterprises were introduced by Peru in 2003. These include a combination between tax incentives and reduced labour obligations for such firms (Act No. 28015 on the Promotion and Formalization of Micro- and Small Enterprises of 2003). There exist three categories of private enterprises in Peru today, each subject to a distinct labour cost scheme (Weller, 2009; ILO, 2014; and Ministerio de Trabajo y Promoción del Empleo, MTPE, 2016):

• Micro-enterprises: up to ten workers and maximum sales of 150 Tax Levy Units (TLUs)1;

• Small enterprises: up to 100 workers and maximum sales of 1 700 TLUs; and

• Other enterprises: i.e. medium-sized and large enterprises, which fall into the “general” labour scheme (Legislative Decree No. 1086 of 2008).

A detailed discussion of the different components of the tax wedge in Peru is provided in Box 2.3.

Figure 2.2 decomposes the tax wedge at the minimum wage for Peru’s three legal labour cost schemes. It portrays a complex picture of tax schemes which, if not properly addressed, may alter economic activity and employment in several unintended ways. Namely, from 14.7% for micro-enterprises the social insurance contributions rise sharply for the small enterprises (21.6%) and the larger firms to (23.4%). This reflects a spike in the employers’ social security contributions, from 2% for micro-enterprises to close to 11% and 14% for small enterprises and medium-sized and large companies, respectively. Another salient feature of differentiation between Peruvian firms is the comparatively high level of social security contributions paid by the employees in micro enterprises. At 12.7% this level is 2 and 3 percentage points higher than for small enterprises and medium-sized and large companies, respectively.

Figure 2.2. Decomposition of the tax wedge at the minimum wage level for different categories of enterprises

As a percentage of the total cost of labour

Note: OECD is an unweighted average of 27 countries legislating the statutory Minimum Wages.

Source: Data for the OECD come from Taxing Wages 2017, data for Peru are OECD calculations based on information provided in Policy Questionnaire by MTPE and ENAHO 2016.

Evidence analysis shows that binding labour cost regulations, contingent upon the size of the enterprises, may engender strong asymmetries in the distribution of firms (Garicano, Lelarge and Van Reenen, 2016).2 Many highly productive firms that would have been larger without the regulation will likely refrain from expanding, to avoid the higher labour costs incurred by the change of the legal threshold. When they decide to expand, they often do so by splitting into small units, at the cost of considerable efficiency losses. Recent work on Peru by Dabla-Norris, Jaramillo Mayor, Lima and Sollaci (2018) find that size-dependent policies may even led to increase labour informality, since some firms ultimately prefer to hire informal workers to avoid the regulation. The result can manifest in substantial productivity and employment gaps with firms producing too little output, using too few employees (Restuccia and Rogerson, 2008; Bartelsman, Haltiwanger and Scarpetta, 2013). In France, for example, a country where the cost of labour increases significantly when firms reach 50 workers, aggregate output losses are estimated to represent 3.4% of GDP. Hsieh and Klenow (2009) argue that such misallocations account for a significant proportion of the difference in aggregate productivity between the United States, China, and India.

Youth job seekers are likely to lose the most from these adverse dynamics. Evidence from a wide range of countries at different levels of per capita incomes (including Peru) shows that rates of job creation are higher amongst small young firms, provided that they are allowed to grow3 (Haltiwanger, Jarmin, and Miranda, 2013; Ayyagari, Demirguc-Kunt and Maksimovic, 2014). Three are the main reasons explaining why small young firms are particularly likely to hire young workers (Ouimet and Zarutskie, 2014):

• Young job seekers typically have just attained education and therefore can be expected to carry the most up-to-date technical skills at a relatively low cost. Therefore, they might be particularly attractive to young firms aiming to develop new products and/or methods of production;

• Youth are known to be less risk averse and therefore more willing to bear the risk of working for a young firm whose survival is more uncertain;

• Finally, assortative mating implies that young workers are more attractive to young firms and vice-versa.

The above segmentation appears exacerbated by the regulations concerning paid annual leave, profit sharing and family allowances, which also vary significantly across categories of enterprises in Peru. They are considerably more stringent for medium-sized and large firms in the general regime than for small firms. Paid annual leave due corresponds to 30 calendar days in medium-sized and large companies, compared to 15 calendar days in small and micro-enterprises. Furthermore, firms under the general regime that are involved in trade, mining and industry must transfer a share of their annual revenues before tax to their employees. Firms with less than 21 employers are exempt from such a profit sharing mechanism, with the overall share of profits to be distributed varying by industry (Dabla-Norris, Jaramillo Mayor, Lima and Sollaci, 2018).4 Finally, employees working in medium-sized and large companies are entitled to a monthly family allowance equal to 10 per cent of the gross minimum wage (approximately PEN 85; USD 25 or EUR 22) if they have dependent children under 18 or children below 24 receiving higher education.5

2.2.2. Policy insights to address size-contingent social security contributions

The question as to whether the current size-based threshold structure should be maintained goes beyond the scope of this report. In fact, addressing it properly would require a thorough examination of the overall system of tax policies and tax arrangements in Peru, assessing the pros and cons of various approaches. The recent OECD study on the Taxation of SMEs in OECD and G20 countries (OECD, 2015) provides insights into the influence of tax systems on SMEs. The report concludes that the heterogeneity of the SME population means that careful targeting is required to ensure that any government interventions, including tax preferences, achieve their stated policy objectives. With the exception of the disproportionately high compliance costs on SMEs, the size of a business alone may not be sufficient justification for government intervention in the form of special rules. Careful targeting of special tax rules can reduce their costs and potential distortions while ensuring that the intended goals are met.

The report acknowledges that there may be a particular case for targeting preferences and simplification measures toward younger SMEs. These could be the SMEs who are most affected by finance and cash flow difficulties, face barriers to entry and growth from incumbent firms, are more likely to grow than older SMEs, face the highest compliance cost burdens and are likely to have higher spillover effects from innovation. In effect, some of these SMEs may play a particularly strong role in supporting the expansion of youth employment. Nonetheless, even within this group, measures should be carefully targeted to address the specific problem (e.g. access to finance, compliance costs) or particular objective (e.g. innovation). Young, small firms are also the riskiest and most likely to go out of business.

Moreover, the OECD report on Taxation of SMEs in OECD and G20 countries underlines that a high degree of caution is needed to ensure that tax preferences or simplification measures do not introduce further distortions. These distortions can result in incentives to alter economic activity in unintended ways to benefit from special tax rules, horizontal inequities in the treatment of different firms or individuals depending on their characteristics. They can also stem from the creation of additional barriers to SME growth owing to the creation of sized-based thresholds, which provide incentives to remain under that threshold, whether artificially or by restraining growth.

An additional policy insight of the report is that, when introducing special tax rules for SMEs, care should be taken to ensure that these measures do not increase complexity. The costs associated with tracking eligibility, keeping specific records and interacting with the tax system for multiple different preferences or simplification measures can increase the complexity of the system. In this regard a simpler general tax system may be more advantageous to SMEs than a series of simplification measures.

Furthermore, the reports points to process simplifications, particularly through targeted use of technology, as offering many advantages in lowering compliance costs by streamlining and reducing the steps required to comply. They can, therefore, be a powerful tool to enhance compliance and reduce its costs.

2.2.3. Lessons from the past: The importance to strengthen social dialogue

Supported by rigorous evidence based analysis, social dialogue is a key pre-requisite to the development and the implementation of policy change. However, social dialogue has been missing during the recent negotiations to reform youth labour market policies in Peru. In December 2014, a Special Youth Labour Scheme (Law No. 30288) was introduced to reduce non-wage labour costs paid by small enterprises and firms under the general regime willing to hire youth workers. The new scheme, expected to stay in vigour for a period of five years, foresaw significant cuts of employers’ social security contributions as a way of supporting the hiring of youth workers aged between 18 and 24 years. Contributions to health insurance would be subsidised by the Peruvian State, while contributions to unemployment benefits, the so-called CTS, would be eliminated. At the same time, the scheme envisaged to abolish existing entitlement of young workers to bonuses and profit sharing. Furthermore, paid annual leave would drop from 30 to 15 calendar days for youth employed under the general regime.

The law was bitterly opposed by Peruvian youth. In particular, the abolishment of the CTS implied that youth workers would be excluded from unemployment savings accounts and, hence, deprived of any social protection following a dismissal (unless the latter be judged as “unfair”). The Special Youth Labour Scheme was dubbed the Ley Pulpín (Pulpín Law), after the name of a brand of fruit juice targeted at children – with the term Pulpín evocating a naïve or immature young person who can be easily mislead. Against rising social pressures, it was almost immediately repealed (as early as January 2015). The example of the Ley Pulpín points to the need for ensuring a strong social dialogue framework that supports the identification of reform needs while at the same time ensuring protection (OECD, 2017). Together with the rigor and quality of ex-ante evaluation analysis, social dialogue is essential to create a consensus for reform and to the effective implementation of policy decisions once agreed upon.

2.3.1. The framework of prudent changes of the minimum wage should be maintained

According to the 1993 Constitution the minimum wage in Peru is regulated by the state in consultation with the social partners. This principle was reaffirmed by two rulings issued in 2004 (laws 27711 and 28318), which established that the minimum wage is set by the government taking into account a proposal by the National Council for Labour and Employment Promotion (CNTPE), a tripartite body including the government, the business associations and the labour unions.

Since 2007, the CNTPE relies on the support of an expert body, in charge of providing a technical advice on how to adjust the minimum wage. This is done taking into account two criteria: i) inflation trends, which ensures maintaining the purchasing power of workers who earn the minimum wage; and ii) the pattern of average labour productivity, which allows taking into account the broader economic and competitiveness outlook of the country.

There is a single value of the minimum wage in Peru, which applies nationwide and accordingly does not vary between sectors or regions (MTPE, 2016). It is calculated on a monthly basis, assuming a maximum limit of 48 working hours per week. Although the law does not mention the frequency of adjustments, the evidence available suggests that the minimum wage has not been increased “mechanically” throughout the years but rather prudently and in a way broadly in line with labour-market conditions.

During the period between 2007 and 2017, the cumulative increase of the real net minimum wage (i.e., excluding taxes and social security contribution) has approximated 12%, which is smaller than the corresponding increase of the real net median wage (Figure 2.3). It is also significantly smaller than the corresponding increase of real GDP (56.7%). The minimum wage was last increased in March 2018 -- to PEN 930 (EUR 245, or USD 284) per month.

Figure 2.3. Annual real net minimum and median wages of full-time employees in Peru, 2007-17

In Peruvian Sol, using 2016 as base year

Note: The annual real median wage is computed for employees in the formal and informal sector who work between 40 and 48 hours a week. Indeed, the official workday in Peru is eight hours a day, with full-time employees working between five and six days a week. Self-employed individuals are excluded from the computation.

Source: Policy Questionnaire by MTPE (2016) and ENAHO 2007-17.

While the real value of the minimum wage should not be allowed to erode in years to come, the current stance that prioritises avoiding important (real) increases should be maintained. This conclusion is supported by the results of recent evidence analysis, which points to the fact that formal employment in Peru, if not strongly affected, tends to respond negatively to changes in the minimum wage (ILO , 2017; Céspedes and Sánchez, 2013; Jaramillo, 2012; Palomino Samaniego, 2011; Del Valle, 2009; Céspedes Reynaga, 2006). These works also conclude that the risk of unemployment following a rise of the minimum wage is higher among youth workers. There is also evidence that the minimum wage is used as a benchmark for the determination of wages in the Peruvian labour market. This reflects the existence of mechanisms of indexation to the minimum wage, both at the firm level and regarding the allocation of family benefits.

2.3.2. Supporting decent minimum wages for youth while keeping costs in check

The international comparison depicts a mixed picture for the level of the minimum wage in Peru. Figure 2.4 (Panel A) shows that regardless to the category of enterprise, the minimum wage in Peru is well below the average of the OECD countries. On average, labour costs at the minimum wage level are between two (general regime) and three times (micro-enterprises) lower in Peru than in the average OECD country. This result is as expected, insofar as it reflects the existence of broad differences in productivity and average wage levels between Peru and the OECD countries.6 While the minimum wage in Peru does not depart significantly from the level observed in Chile, it is much higher than in Mexico.

Figure 2.4. Annual labour costs and minimum wage relative to median wage in Peru and OECD countries

Note: The annual median wage in Peru is computed for employees in the formal and informal sector who work between 40 and 48 hours a week. Indeed, the official workday is eight hours a day, with full-time employees working between five and six days a week. Self-employed individuals are excluded from the computation. The national poverty line in Peru is computed by the National Institute of Statistics and Informatics (Instituto Nacional de Estadísticañ e Informática, INEI) for year 2016. OECD is an unweighted average of 27 OECD countries where minimum wage is enforced. Annual net minimum and median wages are expressed in USD PPP.

Source: Data for the OECD come from Taxing Wages 2017 and the OECD Employment database (estimates as of year 2015) while data for Peru are from Policy Questionnaire by MTPE (2016) as well as ENAHO 2016.

However, the net minimum relative to the net median wage is considerably higher in Peru than in the mean OECD country. It ranges from 60% in micro-enterprises to 66% in small enterprises and 72% in firms under the general regime (Figure 2.4, Panel B). By contrast, this ratio is 55% in the mean OECD economy. No OECD country shows a ratio greater than that which prevails in Peru for firms under the general regime, with the exception of Turkey. Accordingly, the ratio is high also in comparison to Mexico and Chile.

Future policies should ensure that the minimum wage remains attractive to Peruvian youth job seekers but also that the minimum wage is set in a way that does not create a disincentive for employers to hire workers formally. Mechanisms that link minimum wages with productivity or price levels can help reduce such disincentives. A related action would be to allow for a differentiated minimum wage across different regions in Peru so that they are more closely linked to actual levels of worker productivity and/or price levels in the region.

One policy approach used by many OECD countries and other Latin American countries (Table 2.2) consists in the introduction of a legal subminimum wage for young people. This scheme could allow for progressive increases of the minimum wage up to or above the standard minimum wage, as is the case for instance in Australia and the Netherlands. By avoiding large jumps in labour costs from one level to the next, the advantage of this approach is twofold: i) to limit risks of job separation around the point of eligibility to the standard minimum wage; and ii) to help youth workers remain on a path of gradual career progression.

2.4.1. Policy options to reduce labour market dualism

The high protection of Peruvian workers on permanent contracts can discourage employers from hiring youth on this type of contracts for fear of not being able to adjust to employees’ misconduct or incapacity, or changes in demand and technology. This implies that employers would rely on temporary contracts as “dead ends”, rather than stepping stones during the transition to a permanent job, which is especially desirable in the case of youth workers. The function of these contracts as on the “job training” and “screening tool” of young job seekers, would be reduced and the duality of the labour market would increase as a result. Box 2.3 provides a review of the empirical evidence about the links between employment protection legislation and the duality of the labour market.

Figure 2.6 shows the proportion of full-time workers in the formal sector who are employed with a permanent contract (rather than a temporary one) in OECD countries and Peru. On average, workers under a permanent contract constitute a majority in the OECD countries, but a minority in Peru. Moreover, with strict EPL requirements on permanent contracts being particularly harmful to youth access to stable jobs, the probability for young people (15-24) in Peru to work under a permanent contract is half than that of the working age population (21% vs 42%). By contrast, it is only one third lower in the mean OECD country (62% vs 87%).

Tackling the strictness of the EPL on permanent contracts can therefore be an important lever to reduce the large proportion of Peruvian young workers who are either trapped in precarious jobs in the formal sector or entirely excluded due to high informality. For collective dismissals grounded on economic, technological and structural reasons, today’s cumbersome compliance procedures preventing the recourse to these dismissals could leave the way to a simpler notification requirement. For individual dismissals, a number of outstanding legal barriers that prevent employers from the possibility to plea for a justified dismissal could be relaxed. The recognition of adverse economic circumstances among the reasons justifying a dismissal is a primary candidate. Moreover, the high penalty following an unfair dismissal could be reduced with a view to aligning it more closely to the standards prevailing in OECD countries. Recent reforms undertaken by a number of OECD countries go in this direction (Estonia, Spain Slovenia and Italy). Regarding temporary contracts, Peru could benefit from a reduction of the administrative barriers that discourage the creation of temporary work agencies. There also seems to remain scope for expanding the circumstances under which Peruvian firms can use temporary contracts for permanent tasks related to the core activities of the firm. At the same time, the maximum cumulated duration should be fixed by regulation as a way of preventing young workers from being trapped in precarious jobs, but also ensuring that the employers can screen youth workers for future permanent hiring.

Figure 2.6. Incidence and composition of permanent employment, Peru and OECD countries in 2015

As a percentage of dependent employment in each age group

Note: Permanent employees are wage and salary workers under contracts without a clear end date. The figures for Peru include individuals working under a verbal contract since in Peru verbal contracts are as valid as written contracts. The OECD figure is calculated as an unweighted average.

Source: OECD Employment database and ENAHO 2016.

In setting up policies to address EPL challenges, policy makers typically face some key trade-offs. In particular, the effect of relaxing employment protection on permanent contracts may not be the creation of more and better quality jobs for youth, if temporary contracts are eased at the same time. Indeed, the overall effect could be that, as a way of benefiting the smaller cost of terminating an employment relationship at the end of a fixed-term contract, firms will adjust by substituting temporary for permanent workers. To counter the risk of this offsetting effect, several OECD countries have introduced measures recently to harmonise the regulation of temporary and permanent contracts. The objective of such reforms should be to reduce incentives for employers to hire people on temporary contracts when not justified by the temporary nature of the tasks.

Several OECD countries have introduced measures recently to harmonise the regulation of temporary and permanent contracts. One policy option to achieve this would be through merging the two employment contracts into one single open-ended contract. However, firms not only make recourse to temporary contracts as an on-the-job training or screening device, possibly leading to a permanent hiring. They also rely on temporary contracts to respond to fluctuations in activity, or for filling their truly temporary job openings. Therefore, removing temporary contracts altogether may not be a realistic solution. Moreover, since not all temporary jobs can be expected to be converted into permanent ones it could result in serious employment losses (OECD, 2014).

In order to address these problems, several observers have sueggested implementing a unified contract (OECD, 2014). The idea would be to both maintain all types of contracts and have termination costs increasing with seniority, independently of the type of contract. In addition, in the case of termination, firms would pay a layoff tax to the public authorities, while dismissals would be unfair only in cases of discrimination and prohibited grounds (see e.g. Blanchard and Tirole, 2003; Cahuc, 2012). The layoff tax would yield resources to mutualise the reallocation costs of displaced workers and induce firms to internalise the social cost of dismissals, without any need of reinstating workers, if set at a sufficiently high level (Cahuc and Zylberberg, 2008). The clear advantage of this proposal is that it would leave unchanged the cost of termination of short-term contracts, thereby not making more burdensome their use for tasks that are truly temporary.

However, real situations are complex and actual reform approaches differ, reflecting the specific circumstances of each country. Box 2.4 discusses recent EPL reforms implemented by four European countries during the past decade, Estonia (July 2009), Spain (February 2012), Slovenia (April 2013) and Italy (December 2014). Italy opted for a standard contract, according to which employment protection is set to increase with tenure as a way of taking into account the experience and seniority acquired during a career path.

One important requirement to increase the acceptability of the reforms to address excessive job protection is that worker protection is concomitantly reinforced. Policy options for introducing such an element of security are discussed in Chapter 3.