Chapter 1. All together: Making growth more inclusive in Costa Rica

In the past 30 years Costa Rica has grown steadily and social indicators have improved markedly. Well-being indicators are comparable or even above the OECD average in several dimensions, such as health, environment or life-satisfaction. This chapter reviews the social progress that Costa Rica has achieved and identifies reducing inequality and poverty as the main challenges. To tackle those challenges, the chapter argues that there is a need to upgrade existing social assistance programmes to maximise their impact. Social policies should put more emphasis on getting more people into formal work, including by raising their skill levels. This is the most effective way to get people out of poverty. Education is the area where the largest gap with respect to OECD countries is observed. Policy efforts are also warranted to tackle informality, which is increasing rapidly, and to close the gender gap in the labour market. The health and pension systems play a fundamental role in maintaining social cohesion in Costa Rica and it is crucial to modernise them and to make them sustainable in the face of demographic challenges. Costa Rica exemplifies the benefits of preserving natural resources in generating growth and employment opportunities, thereby providing a way out of poverty. Building on its achievements in this area, Costa Rica should reinforce environmental protection efforts, such as reducing emissions from the transport sector and improving wastewater treatment.

Costa Rica stands out among Latin America countries for its commitment and efforts to maintain social cohesion by providing virtually universal education, health and pensions. This commitment, together with stable economic growth, has translated in non-monetary well-being dimensions being comparable to levels found in higher-income OECD countries (Figure 1.1). Absolute poverty rates are among the lowest in Latin America. Average life expectancy at birth has improved markedly from 62 years in 1960 to 80 years today, very close to the OECD average. Subjective well-being is relatively high (Gallup, 2015), suggesting high levels of perceived happiness, which is also confirmed by the results of the recently released Gross National Happiness index. Costa Rica is the oldest democracy in Latin America. It abolished its army in 1948 and led pacification processes in the Central American region. It is the only tropical country that has substantially reversed deforestation (Word Bank, 2015) by increasing the area covered by forest from 26% at the beginning of the 1980s to 52% today.

Inclusive growth in Costa Rica

Inequality in Costa Rica is relatively high by OECD standards, as measured by the GINI index (Figure 1.2, panel A). Despite an improvement in overall macroeconomic performance, inequality has risen since the mid-1990s and is currently at its maximum historical value. In the same vein, poverty has remained largely unchanged over the last two decades (Box 1.1). This is in stark contrast with other Latin American economies (Figure 1.2, panel B), such as Brazil, which has recently made significant progress in reducing inequality and poverty. Households at the top of the income distribution get a large share of national income: the top 10% gets 32 times the average income of the bottom 10%, compared to an OECD average of 10 times (Figure 1.2, panel C).

The increase in inequality has taken place despite an overall increase in household’s disposable income and reflects that the gap between the rich and the poor has increased recently (Figure 1.4). Consistently with evidence from OECD countries (Hoeller et al. 2012), earned labour income is the main determinant of income inequality. Over the 2010-14 period, public sector salaries made the largest contribution to inequality (Table 1.1 and Gabriel and González Pandiella, forthcoming), particularly the salaries of qualified workers in public agencies outside central government. Private sector wages also contributed to inequality, reflecting high skills premium and weak job creation for the low-skilled (more below).

Box 1.1. Poverty measurement in Costa Rica

Poverty is measured in Costa Rica against an estimate of the minimum income required for a person to satisfy food and non-food needs. These needs are based on a basket of goods and services deemed to be basic for subsistence. This basket is built based on information obtained from the Survey on Income and Expenditure (ENIGH). Both composition and cost are determined separately for urban and rural areas. The value of the basket, i.e. the poverty line, is updated according to monthly changes in the prices of food and non-food items, weighted by area of residence, according to the calculations underpinning the consumer price index.

A household is considered to be poor if its per capita income is equal or below to the poverty line of the respective area (urban or rural). It would be classified in extreme poverty if the per capita income is below the value of the basic food basket, which includes only items related to food needs.

At July 2015, the poverty line is estimated to be CRC 107293 (around USD 201) per capita per month for urban households and CRC 82418 (around USD 155) for rural households. Extreme poverty lines are CRC 49607 (around USD 92) and CRC 40673 (around USD 76), respectively.

The resulting poverty indicators signal that total poverty has remained broadly stable over the last 20 years (Figure 1.3). It is estimated that total poverty reaches 21.7% of households in 2015. This implies that 317 660 households are poor, concentrating 1 137 881 people (INEC, 2015a). Extreme poverty has seen a slight increase in recent years. From 5.8% in 2010, it has increased to 7.2% in 2015. As a result, 104 712 households and 374 185 people are estimated to be in extreme poverty.

Table 1.1. Relative contribution of each source of income to total inequality
	2010	2011	2012	2013	2014	Average 2010-14
Relative contribution of different sources (%)	100	100	100	100	100	100
Qualified public sector employee salaries	31.2	32.5	29.7	31.0	31.1	31.1
Central government	12.3	13.8	13.1	14.5	14.6	13.6
Other public agencies	18.9	18.7	16.6	16.5	16.5	17.4
Unqualified public sector employee salaries	1.5	1.7	1.3	1.5	1.5	1.5
Qualified private sector employee salaries	26.0	26.6	28.0	28.2	30.2	27.8
Unqualified private sector employee salaries	3.6	3.7	3.4	2.6	3.6	3.4
Other employees salaries (e.g. domestic service)	0.6	0.7	1.0	0.3	0.2	0.5
Professionals and tech. self-employed business income	5.5	4.0	3.8	6.2	3.4	4.6
Others self-employed business income	3.4	3.7	5.5	3.9	2.8	3.9
Employers business income	8.6	10.1	8.6	9.3	10.4	9.4
Capital income	11.0	9.4	10.3	9.0	9.1	9.7
Contributory pensions	8.5	7.8	8.0	8.0	8.1	8.1
State transfers (e.g. non-contributory pensions and IMAS programmes)	-0.8	-0.7	-0.7	-0.8	-0.9	-0.8
Other transfers (private and non-monetary)	1.0	0.6	1.1	0.7	0.8	0.8
Note: These calculations are based on the income categories reported in ENAHO, which includes contributory pensions, non-contributory pensions and transfers provided by IMAS. This is in line with Sauma and Trejos (2014), Lustig et al. (2013), Estado de la Nación (2014) and Trejos and Oviedo (2012). The underlying GINI coefficient is based on the per capita income concept and therefore is slightly different from those based on the equivalised income concept such as those reported in Figure 16 (panel A) and Figure 1.2 (panel A).
Source: Gabriel and González Pandiella (forthcoming).

Indicators of income concentration confirm that income inequality in Costa Rica is driven by both ends of the income distribution. Around 38% of income goes to the top 10% (Figure 1.5, panel A). As in other OECD countries, this share has been increasing over time. On the other side of the income distribution, the share of income that goes to the bottom 20% is decreasing (Figure 1.5, panel B).

The tax and transfer system does not redistribute income to offset these widening income disparities. Comparing market income inequality and disposable income inequality shows that the tax-transfer system only reduces inequality by 4% compared to 26% on average across OECD countries. Conversely, in-kind transfers (benefits in the form of healthcare and education services provided by the state) are quite effective in reducing inequality. They reduce it by 23%, which is a higher reduction than is achieved in other Latin America countries (Figure 1.6).

Increasing the progressivity of the tax system

The tax system has a very limited redistributive impact. Tax revenue to GDP is low when compared with OECD countries and direct taxes contribute to redistribution less than in other countries in the region (Sauma and Trejos, 2014b). Costa Rica raises more revenue from social security contributions and less revenue from personal and corporate income taxes than most countries in Latin America (OECD, 2015a) and the OECD. There are also numerous tax expenditures amounting to 6% of GDP (Ministerio de Hacienda, 2014). At the same time the progressivity of the personal income tax is modest, and the employment income threshold below which no personal income tax has to be paid is very high in international comparison (Figure 1.7). As a result, a large share of employees does not have to pay personal income tax, thereby reducing tax receipts and eroding the role the tax can play in reducing inequality. Taxes on capital income at the personal level are also relatively low in Costa Rica, and tax evasion of free professions is large (Brys, forthcoming).

A large tax reform proposal is currently under consideration. It would create two additional income brackets, with rates of 20% and 25% at 5 and 10 times the average income. This is a step in the right direction to increase the progressivity and the redistributive impact of the tax, although its effect is modest, as the two new brackets will affect only 2% of total wage earners given the high tax-free income threshold. To make the personal income tax system more progressive, the additional income brackets could be set at lower income levels and the threshold under which no personal income tax is paid should be gradually lowered.

The tax reform proposal includes also the introduction of a fully-fledged VAT tax. The current standard sales tax rate is 13%, considerably below the 19% average rate for OECD countries. Moreover, services are currently exempted. The VAT reform bill will increase the rate to 14% in the first year and to 15% in the second year. It will also broaden the base to include services. In addition, it will be accompanied by a refund system to compensate households in the lowest four income deciles, which is welcome from a distributional point of view. Thus, the proposed VAT reform will not have a significant impact on inequality, but the government expects it will lower the poverty rate by about 3 percentage points. These tax changes should be complemented by additional efforts to eliminate existing exemptions across the tax system that reduce its progressivity. Moreover, Costa Rica would greatly benefit from a rebalancing of the tax mix. Shifting taxation away from social security contributions towards less distortive or more progressive taxes – such VAT, or income and property taxes – would have a positive effect on stimulating growth and formal employment (more below) and reduce inequality and poverty.

Less fragmentation and better targeting would make social assistance more effective

Public sector cash transfers, such as non-contributory pensions and social programmes run by the Institute of social welfare (IMAS; Instituto Mixto de Ayuda Social) reduce inequality (Table 1.1), although their overall impact is small. This partly reflects that they only represent 1.3% of household income, although they have increased by 11% in the period 2010-14 (Gabriel and González Pandiella, forthcoming). But their impact could be larger if their targeting were improved.

The Family Allowances Fund (FODESAF; Fondo de Desarrollo y Asignaciones Familiares) is the main government mechanism to finance social assistance programmes. In 2015, FODESAF funds accounted for 2% of GDP and 76% of social spending. According to the law that established its creation and functioning, FODESAF is aimed at those living in poverty and extreme poverty. However, the beneficiaries of several programmes financed by FODESAF spread well beyond those in a situation of poverty. Only half of the beneficiaries are actually in the first income quintile (Table 1.2), which corresponds to those in poverty and extreme poverty; 25% goes to second quintile, which corresponds to those not in poverty but in a vulnerable situation; and the remaining 24% goes to middle and high income households. This suggests that there is room to improve the targeting of some of the programmes financed by FODESAF.

Table 1.2. Distribution of beneficiaries of social programmes by income quintile
% of beneficiaries in each income quintile (2013)
Programme	Income quintiles
Programme	I	II	III and IV	V
Youth centres (Centros infantiles)	57	25	18	0	100
School meals	42	27	28	4	100
Scholarships for secondary education (Avancemos)	49	29	21	1	100
Non-contributory pensions	64	15	19	2	100
Family welfare centres (Promoción y bienestar familiar)	50	28	21	1	100
Family housing fund (Bono familiar vivienda)	30	23	39	8	100
Insured on the account of the state (Asegurados por cuenta del estado)	70	16	14	1	100
Total Fodesaf	51	25	22	2	100
Source: Estado de la Nación (2014).

Bono familiar vivienda, a subsidy to allow poor households acquiring a house, had a positive effect in facilitating access to owned housing. Currently about 67% of poor households own their house. However, only 30% of the beneficiaries of the subsidy are in poverty, and, remarkably, around 40% are in the third and fourth income quintiles (Table 1.2). Moreover, this subsidy excessively promotes homeownership. Around 20% of households in Costa Rica access housing through renting (Blanco et al., 2014), a rate which is in line with the Latin America average but well below rates observed in most OECD countries and in some Latin America countries such as Colombia. A negative side effect of a small rental market is a low degree of residential mobility. It is more costly for homeowners to move than for renters, and a small rental market may prevent households from moving close to where jobs are and undermine their economic opportunities, especially for cash-strapped workers. This may be an important constraint for Costa Ricans, given the lack of public transportation options and the heavy traffic congestions they face (more below). Hence, Bono familiar vivienda should be better targeted and, in parallel, a more tenure-neutral scheme, such as rental cash allowances, could be introduced. Similar schemes exist in many OECD countries, such as the United Kingdom and some Nordic countries (Andrews et al., 2011), and has also been introduced recently in Chile (OECD, 2013a) and Mexico (OECD, 2015b).

Another scheme that could be better targeted is the school meals scheme, which reaches only 40% of students in the lowest income quintile. Avancemos, which offers cash transfers to poor families conditional to their children attending secondary education, seems to be successful in avoiding too many beneficiaries among medium and high income earners (Table 1.2), but has few recipients among those in extreme poverty (Estado de la Nación, 2013), as it is linked to attendance to education. Non-contributory pensions and the scheme providing health insurance coverage to poor households are the programmes that better target low-income households.

Better targeting of social programmes would allow the government to focus scarce resources on those that need them the most. At the same time, targeting should be carefully designed to avoid poverty or unemployment traps, whereby the withdrawal of social benefits or support that comes with entering formal paid work dissuades people from taking a job. Removing social benefits in a gradual way would avoid the creation of such traps while, at the same time, help to focus support on neediest households.

Reducing poverty is a priority

Analysing the characteristics of poor households in Costa Rica indicates that these tend to be larger than non-poor ones (Table 1.3). They have more children and more members that are economically dependent (i.e. below 15 and above 64 years old). They also tend to be headed by women. Poverty is also strongly connected with labour market performance. Poorest workers tend to work in the informal sector: 77% of those employed have informal jobs. There is also a strong association between poverty and education. Poor tend to have 6.4 years of education, which correspond to primary education, while the non-poor reach higher levels of education. Children in poor households are less likely to attend secondary education. No difference between the poor and the non-poor is observed concerning children’ attendance to primary education, which is nearly universal in Costa Rica. Poor children also accumulate larger educational underachievement. This analysis suggests that policies conducive to tackling unemployment, reducing informality and improving education – as discussed below – are fundamental to help Costa Ricans escape poverty. It also suggests that targeted support to some groups, such as low-skilled, youth and women, is warranted.

Table 1.3. Who are the poor in Costa Rica?
Indicator	Poverty Status
Indicator	Poor	Non-Poor
Demographic
Total persons	1 137 881	3 685 834
Total households	317 660	1 144 475
Household size	3.6	3.2
Household members of 5 years old or less	0.4	0.2
Percentage of households headed by women	43.5	34.1
Percentage of households headed by women without a partner	35.7	25.5
Demographic dependency ratio	0.7	0.4
Employment
Unemployment rate	25.4	5.3
Employment rate	34.2	60.3
Percentage of employed population with informal employment	77.3	36.6
Economic dependency ratio	2.2	0.9
Education
Average schooling of people 15 years old or more	6.4	9.3
Percentage of population of 6 to 12 years old who do not attend regular education	1.1	1.0
Percentage of population aged 13 to 17 years who do not attend regular education	15.9	10.9
Percentage of population aged 7 to 17 years with educational underachievement	32.7	20.9
Housing and access to services
Percentage of households without an owned home	33.4	28.3
Percentage of households with housing in regular or bad physical condition	62.4	34.6
Percentage of households with crowding by housing bedroom	7.5	1.5
Percentage of households without internet service	60.1	34.1
Access to subsidies and pension
Percentage of households receiving a grant or scholarship	45.5	16.9
Percentage of elderly without pension	35.6	33.7
Note: The table excludes domestic servants and pensioners living in households.
Source: INEC, Encuesta Nacional de Hogares Julio 2015 Resultados Generales.

Existing social programmes have prevented an increase in poverty rates in Costa Rica, despite the rise in unemployment. Nevertheless, with poverty rates remaining largely unchanged over the last 20 years, Costa Rica has rightly established fighting poverty as a priority. Structural reform to address institutional fragmentation and lack of coordination in social policy is needed to reduce poverty and make growth more inclusive. Several steps have already been taken but further efforts are warranted to increase efficiency and make the system more evaluation-based.

IMAS, the Institute of social welfare, is gradually executing Puente al Desarrollo, a strategy to reduce extreme poverty. Such strategy encompasses social assistance, to ensure that basic needs are met, and also a social protection component, whereby social workers establish a family plan comprising education, training, health and childcare. There is conditionality attached to the plan, so that beneficiaries need to comply with the duties established in it, such as sending children to school or completing training programs. The design of Puente al Desarrollo seems to adequately reflect that poverty is a multifaceted phenomenon that requires coordinated actions in different policy areas. Taking advantage of its gradual execution, IMAS should promptly set-up evaluation mechanisms, and adjust the design of the programme based on that evaluation. IMAS has also put in place mechanisms to assess needs and check progress, including the compilation of social maps, which help to geographically monitor and target efforts, and of a multi-dimensional poverty index (INEC, 2015b).

In general terms social programmes currently run in Costa Rica to reduce poverty show scope for achieving greater efficiency. Programmes are run independently by different institutions, although they offer similar benefits, use different eligibility criteria and different registries of beneficiaries. This suggests that there is room for gaining efficiency and economies of scale by reducing fragmentation, especially considering that Costa Rica is a relatively small country. FODESAF currently finance 26 different programmes, managed by 22 different institutions. And there are other social programmes run by other ministries or autonomous agencies. This translates into a complex institutional framework, with difficulties to coordinate efforts, delineate responsibilities and check performance, and very prone to duplication.

For example Avancemos, aimed at supporting attendance to secondary education and run by IMAS, coexist with other grant schemes that offer cash transfers to those attending primary, secondary or higher education. These programmes are run independently by other institutions, such as FONABE (Fund for national scholarships). Similar duplication is observed in areas such as nutrition programmes or housing. This implies a lack of spending efficiency and a duplication of efforts by both beneficiaries and government agencies, preventing from taking advantage of economies of scale.

A key obstacle to a more efficient delivery of social assistance is that different programmes use different eligibility criteria and keep unconnected registries of beneficiaries. Moving towards a unified framework, based on a common database, would improve the efficiency of the system, reduce duplication and better verify the eligibility of applicants. Efforts in this direction have started. IMAS has integrated their information systems with those of some other social assistance programmes such as those of the non-contributory pensions or the housing programme. In June 2015, the budget for the creation of a national registry of beneficiaries (SINIRUBE – Sistema Nacional de Información y Registro Único de Beneficiarios) was approved. This registry would be a key tool to improve the coverage and the targeting of the programmes so that more social spending goes to those more in need.

There is also a need to make the social protection system more accountable and performance-based. At the heart of the current structure is FODESAF, which acts as a financial intermediary. It receives a transfer from the central government based on a formula and channels those funds to the institutions in charge of executing social assistance programmes. Part of the funding is channelled back to other departments in the central government that are in charge of executing social assistance programmes. More than 50% of FODESAF budget is earmarked and cannot be controlled either by FODESAF or the line ministries. All this implies that is difficult to associate financial flows to individual programmes, making the system inflexible and evaluation difficult.

The national registry of beneficiaries would help to establish systematic and robust mechanism of evaluation. Establishing a culture of systematic evaluation would help to adjust programmes to make them more successful, making possible to reallocate more resources to those that actually help to escape poverty. An evaluation culture is well-established in some OECD countries such as United Kingdom, the Nordic countries and the Netherlands. Costa Rica should also take decisive steps in that direction. Evaluations should be able to isolate the effect of the programme from the outcome that would have been achieved without programme participation (OECD, 2010). This can be done either via experimental evaluations or quasi-experimental evaluations based on statistical techniques.

Rice price controls and tariffs are regressive

Lack of competition and inadequate regulations in key markets (see Chapter 2) can have a direct impact on living conditions of poor households and increase inequality. This is the case of regulations affecting rice. Rice is a basic staple food for the poor in Costa Rica, and local production is not sufficient to satisfy local demand. At the same time rice cultivation is protected by tariff and non-tariff barriers and through price controls. The average tariff for rice amounts to 28%, and prices for locally produced rice are strictly controlled in every step of the production process. As a result, rice prices paid to local producers are double international prices (Umaña-Alvarado, 2014). Rice price support mechanisms also exceed by almost five times the country’s commitment under the World Trade Agreement on Agriculture (OECD, 2015c). Monge González et al. (2010) estimate that, in the period 1995-2005, tariffs and price support mechanisms entailed a transfer from consumers to rice producers equivalent to USD 396 million. The impact of this transfer on the poorest households represents about 8% of their per capita income, which contrasts with a burden on the richest households that amounts to 0.5%. At the same time, price controls have blunted incentives among local producers and productivity has therefore been falling (Monge González et al., 2010). Conditions for small local farmers have not improved either, as large producers are the ones receiving the rents (Umaña-Alvarado, 2014). Costa Rica has announced plans to gradually reduce rice price support mechanisms, which is long overdue.

Making the labour market more inclusive

The recent labour market performance has been weak. Unemployment has been increasing steadily since 2007 and is closed to 10%. Contrary to the situation before 2007, the unemployment rate is now above OECD levels and also well above unemployment rates seen in Latin America (Figure 1.8, panel A). The gaps in labour market performance across different socioeconomic groups are large and increasing. The unemployment rate for the bottom quintile is 25% and is increasing rapidly (Figure 1.8, panel B). There is also a stark contrast in labour market participation. Only 40% of those in the bottom quintile participate in the open labour market and this percentage is falling. The differential in unemployment and participation rates between bottom and top quintiles is higher in Costa Rica than in other countries in the region and is increasing, which highlights the need to put in place targeted policies to make the labour market more inclusive. As in other emerging countries, the quality of jobs in Costa Rica is below the one observed in OECD countries, especially as concerns incidence of long work hours, informality and risk of entering extreme low-pay status (OECD, 2015d).

Women and youth also face difficulties to get jobs. Unemployment is three percentage points higher for women than for men. Poor women who are heads of a household are particularly affected by high unemployment rates. The youth unemployment rate is 24%. There are no significant differences in terms of unemployment between urban and rural areas. But Costa Rica indigenous population faces lower employment opportunities. Indigenous people represent 2% of total population (Census, 2011) – the second lowest proportion of indigenous population in the Latin America. Immigrants, particularly from Nicaragua, represent a large share of the labour force. In 2013, 10% of those employed were foreign born (World Bank, 2015). They tend to be low qualified and the unemployment rate of women born in Nicaragua and other Central American countries is double the unemployment rate for native women. Conversely, for men the rate is below the one observed for Costa Ricans.

In addition to higher levels of unemployment, an obstacle to making the labour market more inclusive is informality. The informal sector in Costa Rica is smaller than in other Latin America countries (Figure 1.9, panel A) but large for OECD standards. Contrary to many Latin America countries that have seen recently a decline, informality is increasing rapidly in Costa Rica. In the period 2011-14, informal employment has grown by 40%, while employment in the formal sector has remained flat (Figure 1.9, panel B). Informality accounts now for 45% of total employment and affects the less educated the most. Half of those with only primary education that are working are employed in the informal sector. Among those with tertiary education the rate is 8%.

The labour market is also characterised by increasing duality. On the one hand traditional sectors, such a domestic-oriented agriculture, manufacturing, construction and domestic services, employ workers that tend to be low-qualified, earn lower salaries and have stagnating or decreasing employment opportunities. On the other hand, exporting sectors, such as those operating in the free trade zones including high value manufacture and services, employ higher qualified workers and are expanding and offering new job opportunities. This shift away from labour-intensive activities into higher value added activities (more in Chapter 2) has increased demand for high-skilled labour. Nevertheless, Costa Rica has a large relative supply of low-skilled workers who are not equipped for those new job opportunities. Thus, many employers report difficulties in finding workers with the skills they need, especially medium level technicians and engineers (ManPower, 2015). This structural mismatch of skills and jobs has translated into high skills wage premia, contributing to the increasing inequality, and into a growing structural unemployment rate, which is one of the largest in the Latin America region.

Costa Rica faces three fundamental challenges to make its labour market more inclusive. First, it should put in place policies to facilitate the acquisition of skills by the unemployed. Second, it should improve incentives for employers and workers to stay or move to the formal economy (informality may be a source of jobs and livelihood in the short-term, but it hampers growth, job quality and productivity in the medium-term). Third, it should implement policies supporting women to participate in the labour market.

Helping unemployed Costa Ricans get back to work

Active labour market policies are barely developed in Costa Rica. As other Latin America economies, Costa Rica prioritises social programmes such as conditional cash transfers, and the provision of basic services to socio-economically disadvantaged groups. Nevertheless, it has been less successful in increasing employability of certain groups. Public employment offices provide basic intermediation services but need to expand its services, coverage and effectiveness (Mazza, 2013). There is evidence that stepping-up job-search support and activation programmes boost both growth and equity (Causa et al., 2015). Improving active labour market policies would be particularly beneficial for low-skilled workers, as they are likely to gain the most from active support and guidance as concerns what skills are needed to find formal employment and how to obtain those skills. Hence, policies should be targeted to those experiencing more difficulties in finding a job. In this respect Mi Primer empleo, a recently launched scheme providing financial incentives to employers that hire young and female workers, seems to be well designed. Going forward, the government should evaluate the programme to assess its effectiveness in increasing employability when taking into account the participants’ profiles. This would contribute to further adjusting eligibility criteria to limit potential deadweight costs.

Training programmes for unemployed have been found to increase employability both in the OECD and in Latin America (Ibarrarán and Rosas, 2009), where they have been associated with higher probability of getting a formal job. For many Costa Ricans, learning new skills will be the only way to take advantage of new employment opportunities. Thus, social assistance programmes should encompass training opportunities and be made conditional on attending them. Costa Rica faces significant skill shortages in technical occupations, which could be partly addressed via retraining of unemployed workers. The National Learning Institute (INA), the main provider of technical training in Costa Rica, plans to develop a dual vocational track and it is involved in parliamentary discussions to implement an initiative on the matter (OECD, 2015e). Such tracks should be also made available to the adult low-skilled unemployed. Targeted support in the acquisition of new skills is also especially warranted for those that have abandoned the education system without finalising secondary education (more below).

As in many other Latin American countries, there is no unemployment insurance scheme in Costa Rica. In case of unemployment, workers are compensated through a severance pay scheme, amounting to approximately 20 days per year of service. The scheme is financed by employers, which contribute 1.5% of the payroll. As Costa Rica becomes a more open economy, it is expected to be subject to more shocks and adjustment needs, which would imply more frequent spells of unemployment. The high level of structural unemployment, 9% according to OECD estimates, and persistent skill mismatches in Costa Rica suggest that workers face difficulties to transition between jobs. An unemployment insurance system would provide them with income support and lead them to assistance and training to get ready for a new job.

Unemployment insurance schemes are well established in OECD countries, and emerging economies have also been gradually stepping-up their unemployment schemes. These are often mandatory savings into individual accounts that may be accessed during a specified period of unemployment. Mexico recently introduced a 6-month benefit to all workers who lose their job in the formal sector, partly funded by the employer and complemented with a solidarity fund. Chile has in place an unemployment insurance system of individual unemployment savings accounts, also in combination with a solidarity fund. Colombia has recently introduced a new mechanism of protection against unemployment based on individual unemployment savings accounts complemented with small allowances for unemployed people. Brasil has in place an individual account system although in this case is complemented with a firing penalty. Taking into account Costa Rica fiscal resources and administrative capacity, running a comprehensive unemployment protection scheme may not be currently feasible. Instead, to provide income compensation in the case of job loss and promote active job search and acquisition of new skills, Costa Rica could consider setting-up a scheme based on individual unemployment savings, as recently done by other OECD Latin American countries.

Boosting employment in the formal sector

Reducing informality requires a comprehensive strategy, with actions covering multiple dimensions such as fiscal, labour market, social policies, education, or business regulations. Reducing non-wage labour costs, simplifying some labour market policies, improving training and education and adapting business registrations procedures to the needs of micro and small firms would help to boost formal employment.

High non-wage labour costs deter the creation of formal jobs

An important part of the strategy to reduce informality is to avoid excessive non-wage labour costs that could deter the creation of formal jobs. Several studies (e.g. Ramirez-Alfaro, 2010) signal high social security obligations as the main obstacle for a business to become formal. The total social security contribution rate is 36%, of which 26% are borne by the employers, a high rate by international standards (Figure 1.10). Out of the total contribution, the social security actually receives 23.5% of the salary. The rest is used to finance other institutions, such as public banks, or social programmes. Empirical evidence based on a panel of Latin America countries shows that lowering the tax wedge can be an effective tool to combat informality (Lehmann and Muravyev, 2012). Such policies tend be particularly effective if structural unemployment is high (Katz, 1998), as is currently the case in Costa Rica. Thus, in the medium-term, a gradual reduction of social security contributions would help to raise formal employment. Across-the-board cuts in social security contributions are easier to implement and administer but could lower fiscal revenues. Targeted cuts may be more difficult to administer but would contribute to broaden the tax base without engendering large revenue losses. Experiences of countries that have introduced targeted cuts in employer social security contributions suggest that they can be effective in increasing formal employment and cost-effective if they are targeted at those accounting for the bulk of the non-employed or informal employment (IMF, 2014). Costa Rica could then focus cuts in social security contributions on those sectors of the economy where informality is rampant, specifically agriculture, construction and domestic work services, which account for 60% of all informal employment. The government should also avoid using social security contributions to fund public banks or antipoverty programmes but use other more progressive taxes instead, and fight domestic and international tax evasion, as these are less distortive ways to raise revenue.

The minimum wage structure should be simplified

The level of the minimum wage is not excessively high in Costa Rica relative to other countries in Latin America (Figure 1.11). Recent changes in the minimum wage have been relatively modest and overall the real minimum wage has evolved in line with labour productivity. Nevertheless, one out of three workers is being paid below the minimum (Estado de la Nación, 2014). This affects mostly young workers, women, workers in rural areas, immigrants, people with disabilities, poor, and workers in agriculture, construction and domestic work. This suggests that the current minimum wage structure is not acting as a wage floor for more vulnerable workers.

The minimum wage structure is complex. Simple minimum wage systems are most likely to improve compliance (OECD, 2015d). There are more than 200 minimum wage categories depending on sector and occupation. This makes it very difficult for employees and employers to know which level of the minimum wage corresponds to a given occupation. And minimum wage levels are updated twice per year. This biannual update was established when inflation was very volatile in order to correct for large changes in inflation. Nowadays, with inflation better anchored, the biannual revision introduces transaction costs without clear benefits. In international perspective countries with a simple national minimum wage system tend to have higher compliance rates than those with multiple sectoral or occupational minimum wages (Rani et al., 2013). Hence, compliance with the minimum wage in Costa Rica would be facilitated by a simplification of its structure, reducing significantly the number of categories, and by moving to an annual update. The level of the minimum wage should be established so that it is not an obstacle for more vulnerable workers such as young or low skilled workers. Moreover, there is also an urgent need to improve enforcement mechanisms (more below).

Informality affects the less educated the most. Half of the workers with only primary education are employed in the informal sector. This suggests that providing Costa Ricans, especially those with less education, with more job-relevant skills will be fundamental to lower informality (more below).

Formalising a business is also more difficult in Costa Rica than in most countries in Latin America and the OECD. Administrative burdens for sole proprietor firms are relatively high according to OECD’s Product Market indicators (Figure 1.12), and this is also reflected in World Bank’s index measuring the ease of starting a business, where Costa Rica ranks relatively low (121 out of 189 countries). Major perceived obstacles are the difficulties and expenses needed for obtaining a licence. OECD best practices suggest that limiting licensing to those activities where it is justified on health, safety or environmental grounds, and reducing the associated cost are important steps to curb informality. Establishing single points for issuing or accepting all notifications and licences required to register a business can significantly reduce administrative burdens to become a formal firm. One-stop shops are increasingly used in OECD countries, and have also been set-up in Brazil and Mexico to establish more flexible and simple registration procedures and facilitate that micro and small firms move to the formal sector.

Evidence from other emerging economies suggests that stricter enforcement can also play a role in increasing formality (Almeida and Carneiro, 2012), as exemplified by Mexico and Brazil. Monitoring and enforcement can be enhanced not only by increasing the number of labour inspectors, which may be currently difficult for Costa Rica given the fiscal situation, but also by other initiatives, such as linking labour inspectors remuneration to performance targets. Such initiatives have proved to be effective in increasing formality in other Latin American emerging countries (Berg, 2010).

Boosting employment of women

Female labour force participation in Costa Rica has recently increased, but it remains low in international comparison (Figure 1.13). OECD evidence suggests that the availability of affordable, high-quality childcare is a key factor explaining cross-country differences in women’s labour market participation (OECD, 2012a). In Sweden, for example, the expansion of child care services during the 1970s is thought to have helped increase women’s employment rates from 60% to over 80% (OECD, 2012b). Positive evidence has also been found in Latin American countries (Mateo-Diaz and Rodriguez-Chamussy, 2013). Facilitating access to good quality child care is important not only to encourage greater female labour supply, but also because access to early childhood education and care can improve school outcomes later in life for disadvantaged children and mitigate social inequalities (OECD, 2011; Ruhm and Waldfogel, 2011). It is during children’s early years that it matters most, as the longer that mothers stay out of the labour force to take care of their children, the more difficult it is for them to get back to work (Thévenon, 2013).

The main gender gap in Costa Rica concerns labour market participation (Table 1.4). Contrary to other Latin American countries, women in Costa Rica do not seem to face wage discrimination at the lower end of the income distribution (Carrillo et al., 2013). Nevertheless, similarly to other countries in the OECD, glass ceilings (wage gaps at the top of the distribution) are significant. Costa Rica stands out in Latin America by a large presence of women in parliament; 33% are women, compared with an average of 20% in Latin America and 28% in the OECD

Table 1.4. Gender gaps in Costa Rica versus the OECD
How do men and women perform in Costa Rica?			And in the OECD?
How do men and women perform in Costa Rica?	Women	Men	Women	Men	Year
Health
Life expectancy at birth (years)	82	78	83	78	2013
Share of people who reported good/very good health conditions	73%	79%	66%	70%	2006 (Costa Rica) 2013 (OECD)
Education and skills
Share of 25-64 yrs. old with at least upper secondary education	42%	39%	75%	75%	2014
Share of 25-64 yrs. who have attained tertiary education	19%	17%	35%	31%	2014
Skills (PISA mathematics results)	396	420	489	499	2012
Jobs
Unemployment rate (15-64 years)	12%	8%	8%	7%	2014
Labour force participation rate (15-64 years)	55%	82%	63%	80%	2014
Share in informal employment	48%	42%	-	-	2015
Civic Engagement and governance
Share of seats in national parliament	33%	67%	28%	72%	2015
Subjective well-being
Levels of life satisfaction on a 0 to 10 scale	7.4	7.4	6.6	6.5	2014
Source: OECD, Health Status Database; OECD (2015), Education at a Glance; OECD (2012), PISA 2012 Results; OECD, Labour Force Statistics Database; OECD, Better Life Index – Edition 2015 Database; Centro Centroamericano de Población, Encuesta Nacional de Salud 2006 (ENSA); INEC, Encuesta Continua de Empleo (ECE) 2015; Inter-Parliamentary Union Database; and Gallup World Poll Database (Gallup, 2015).

Scaling-up childcare services

The constitution mandates that the state should provide education starting at 2 months of age and until six years old. The Ministry of Education offers education as of 4 years old. For children below 4 years old, public services are offered by Red Nacional de cuido, involving IMAS, PANI (Patronato Nacional de la Infancia; National Children’s Trust), CEN-CINAI (Centros de Educación y Nutrición – Centros Infantiles de Atención Integral; Education and Nutrition Centres – Children’s Comprehensive Care Centers) and the Ministry of Education. But supply is limited resulting in a low coverage. Only 15% of those below three years attend, while the OECD average is 35%. Attendance increases to 62% for children age 5 and to 85% for those aged 6. There are also important differences across income levels and geographical areas. Attendance is lowest in the bottom two quintiles, especially of children below 4 years old (Figure 1.14). These differences arise mainly because higher income households resort to private childcare services (Estado de la Educación, 2015). Coverage rates are higher in those regions more economically developed. Attendance is also positively related to the education level of parents; attendance is higher for children whose parents have higher levels of education. There is also a significant geographical mismatch between the supply of childcare education and the areas where disadvantaged households in need of those services, such as households with low education background and large children population, are concentrated. All this is a source of important inequalities, both concerning the integration of women in the labour market and concerning children education outcomes. The authorities should expand the number of places available in early childhood education with a focus on children below 4 years old living in the least developed areas of the country and from more disadvantage households.

As in OECD countries, female labour market participation is also affected by prevailing social norms as concerns the role of men and women towards work and care. Large gender gaps are observed in Costa Rica regarding the sharing of non-remunerated activities such as eldercare, childcare, and care for others. Costa Rican women tend to devote an average of 6 hours per day to non-remunerated activities, while men devote only 3.5 (Estado de la Nación, 2014). These gender gaps start at very early ages and remain large thereafter. As a consequence, women, especially those assuming care responsibilities early, face difficulties to continue in education and tend to drop out of the labour force, especially those in low income households (Figure 1.15, panel A). Eighty per cent of the poor women that are out of the labour force report care responsibilities as the cause (Figure 1.15, panel B). This further highlights the need for specific policies, such as increasing coverage of childcare or eldercare, and for targeting them at women facing difficulties to access the labour market, such as those in low-income household or those taking family care responsibilities at an early age.

At the same time the proportion of female-headed households has increased notably in the last 25 years (Estado de la Nación, 2014). Currently one third of the population live in a female-headed household. This indicates that reducing gender gaps in labour market participation would have a significant positive impact, not only for women themselves, but also for their families and for the Costa Rican economy as a whole. Some estimates suggest that gender gaps currently observed in Costa Rica reduce output per capita by 23% (Teignier and Cuberes, 2014).

Complying with labour rights

Gradual progress in increasing compliance with labour rights has been recently observed in Costa Rica. Still one out of three workers is being paid below minimum wage conditions. Incompliance is also relatively large as concern the payment of overtime and the payment of the Christmas salary bonus (aguinaldo). The number of workers not enjoying any of the labour rights recognised in the legislation, such as sick leave, holidays, overtime payment or social security insurance, has been falling, although it is still 7% of all workers (Estado de la Nación, 2014). Vigilance should continue to ensure that compliance keeps increasing.

Another area which requires monitoring concerns trade union rights. The number of collective agreements in the private sector is low and, instead, direct agreements with non-unionised workers are frequent. Existing laws prohibit anti-union discrimination but enforcement has been weak. Thus, Costa Rica has attempted to promote trade union labour rights in recent years and a new law was approved in the late 2015. The new law explicitly bans any form of discrimination in the workplace, besides shortening labour related disputes, by introducing oral arguments in courts and establishing out-of-courts resolution mechanisms. However, the law still prohibits strikes in public services considered as essential – such as police but also hospitals and ports – and allows for hiring temporary staff to replace striking workers. The government should continue legal reforms so as to comply with ILO conventions and recommendations.

Improving the quality of education

Costa Rica is a pioneer of universal access to education in Latin America and has shown a strong commitment to invest in education. Costa Rica’s public education system encompasses education from primary to university level. Education is free of charge up to secondary level while universities fees are very low. Attendance to primary and basic secondary education is compulsory. In 2013, spending on education represented 6.9% of GDP, excluding spending done by the institution devoted to vocational training, the INA (Instituto National de Aprendizaje, the National Learning Institute), which is independent from the academic education system. In the OECD, only Nordic countries and New Zealand invest a higher share of its GDP in education (Figure 1.16, panel A). There is a constitutional mandate to raise education spending to 8% of GDP by 2018.

Making the most of education spending

Costa Rica has achieved significant progress in reducing illiteracy as 21% of population faced illiteracy problems in 1953 and only 2% does now according to latest census. Moreover, the country has achieved almost full enrolment (94%) in primary education, which is comparable to OECD countries. The net enrolment rate has also risen and reached 61% for 4 year-olds in 2014, and 87% for 5 year-olds in 2014. Nevertheless, the average educational attainments remain low, especially considering Costa Rican level of GDP per capita and education spending. Only 40% of the workforce has completed secondary education (Figure 1.16, panel B). Similar results are also observed for recent cohorts: less than half of the 25-29 cohort has completed secondary education. These results are far from graduation rates of OECD countries and are significantly lower than other Latin America countries such as Colombia, Peru or Panama.

The quality of education is also comparatively low, given the level of spending. PISA scores are low in the three subjects: reading, scientific and, especially mathematics (Figure 1.17). Moreover, since tests are only taken by those in schools, and given the high dropout rates, the average skill level of Costa Rican children are likely to be lower. There are also increasing disparities depending on socioeconomic status (more below). Given this large mismatch between the level of spending and the resulting educational outcomes, Costa Rica should move away from an exclusive emphasis on increasing spending as a policy target and, instead, establish the improvement in educational outcomes as the main target. OECD evidence shows that a high level of spending is not always associated with better outcomes (OECD, 2015f); how money is spent is critical. High-performing systems tend to allocate resources more equitably across socio-economically advantaged and disadvantaged schools (OECD, 2013b). Moreover, demographic dynamics imply a gradual decrease in the number of students entering primary and secondary education. Thus, the target of increasing spending on education should be also weighed against increasing needs in health or in fighting poverty.

Besides improving outcomes, an additional fundamental challenge is to spread education benefits more fairly across the society. Gaps in the access of education depending on household’s income have widened in the last two decades (Estado de la Nación, 2014). Enrolment rates in secondary education are significantly higher for children in higher income quintiles. PISA tests indicate that Costa Rican schools, in addition to low performance, show low equity (Figure 1.17), as disparities in learning outcomes depending on socioeconomic status are large. In Costa Rica, as in other Latin America countries, family background is a bigger determinant of student learning than in OECD countries. Students from lower socio-economic backgrounds perform well below their peers and are less likely to enter and complete tertiary education. The gap in performance between public and private schools is also large, partly reflecting differences in socioeconomic background of students (Fernández and Del Valle, 2013). There are also disparities originated within the public educational system, with important differences across the system in terms of instruction time or access to key subjects such as foreign language or IT (Estado de la Educación, 2015).

Improving students retention

Inequalities in education outcomes start early. At the end of primary education, the share of students coming from low-income households lagging behind is high. These difficulties are aggravated in lower secondary, when many students repeat grade and finally drop-out. This suggests that grade repetition has high costs and doubtful benefits, as also observed in other OECD countries. Grade repetition is negatively related to equity (OECD, 2013b). The percentage of students who have repeated a grade during primary, lower secondary or upper secondary school is one of the highest among PISA-participating jurisdictions. More than 33% of 15 year-old students reported having repeated a grade at least once, compared to an OECD average of 13% (Figure 1.18, panel A). Furthermore repetition is highly concentrated in low-income households (Figure 1.18, panel B). Focusing on early and targeted support to those students with a higher risk of leaving the education system would be more efficient and render better outcomes than grade repetition. This would require providing additional support to disadvantaged students and schools, which would require prioritising further lower secondary education. Spending seems to be skewed towards primary and tertiary levels, where Costa Rica spends more than the average OECD country, to the detriment of early childhood and secondary education, where public spending per student is low by international standards.

Costa Rican authorities have put in place several initiates to foster retention in secondary education and also facilitating the return to the education system of those that dropped out. These include Yo me apunto and Proeduca, a joint project with the European Union. The design of the schemes seems appropriate as they are specifically targeted at those that have previously abandoned the education system or that are experiencing more difficulties to remain in education. These initiatives should be evaluated and, in case of a positive outcome, expanded.

Another action to curb drop-outs at secondary level is to reinforce the vocational or technical tracks. This has proved to be an effective tool across OECD countries to improve educational outcomes and employability, especially when vocational programmes are designed and implemented to be responsive to labour market needs (OECD, 2014a). Acknowledging this, Costa Rica is in the process of discussing the implementation of a dual system in vocational education. International experience suggests that a key element for a successful development of dual education is to give employers a central role in workplace learning, so that the training mix reflects labour market needs. Consultations with relevant stakeholders are maintained in Costa Rica, but these are not sufficient (OECD, 2015e), as educational institutions are the dominant players of the system, with employers having a lesser role. Moreover, many technical courses do not include workplace learning. Work-based learning clearly contributes to better labour market outcomes and therefore should become mandatory (OECD, 2015e), as successfully done in other OECD countries such as Denmark and Belgium. International experience also signals that apprenticeship systems are associated with lower youth unemployment rates and the development of job-relevant technical skills in the workforce. Hence, as part of its strategy to improve technical education, Costa Rica should develop an apprenticeship system engaging all social partners in its design and delivery.

Making tertiary education more responsive

Despite the significant educational effort made by Costa Rica and the relatively high level of unemployment, employers face difficulties to fill vacancies for certain posts (ManPower, 2015). Some skills gaps have also been identified as an important bottleneck to attract foreign direct investment. These gaps are in technical and scientific professions (Monge-González et al., 2015). The tertiary education system is heavily biased towards social science and humanities, producing few Science, Technology, Engineering and Mathematics (STEM) graduates. In the last decade, 33% of graduates were in education area, followed by Economics (23%), Social Science (15%) and Health (12%). Engineering only accounted for 6% of all graduates.

As a result of this mismatch between the demand and supply for skills, Costa Rica has one of the largest skills premiums in Latin America, with high-skilled workers earning on average more than three times as much as low-skilled workers (Tosunta and Osuke, 2012). Furthermore, contrary to other countries in Latin America, the premium has been increasing in Costa Rica, reflecting that the mismatch is largely structural.

Reducing skill mismatches require actions both on the supply and on the demand. On the supply side there is a need to improve the governance of the universities to make them more accountable, performance-based and responsive to Costa Rica’s skills needs. At the core of the reform there should be a change in the universities funding model to encourage a more rapid response to labour market needs and to avoid an excess of supply of graduates in certain fields and shortages in others. The current funding model creates incentives to increase places in fields that are less expensive to deliver, such as the humanities and social sciences. Conversely, it does not encourage the expansion of programmes that are relatively more costly to provide such as engineering, which requires more expensive equipment. Incorporating better incentives into the funding formula should help to ensure a better alignment of courses and curricula with skills demand.

On the demand side it should be also facilitated that students obtain early on in their student life comprehensive information about employment options in the country, as well as wage levels in different industries and about labour market status by degree and university. Making available this information could encourage more students to enrol in those institutions that offer the prospect of better employment and earnings. This would create greater competition among institutions, thereby helping to raise quality standards. It would also help students to better understand the relative value that qualifications have in the labour market, promoting that students follow educational paths conducive to careers within more dynamic sectors such as technical and scientific fields. While much of this information is already available in Costa Rica, it is often in the form of ad hoc reports. To be effective, such information needs to be made available more generally, kept up-to-date and tailored to the needs of different users, such as students initiating secondary education, who are making study choices that might enable or constrain their future educational options.

Finally, there is a need to ensure that the education system as a whole remains inclusive. The current situation, in which students from low-income households face high risks of being left behind in public secondary education and not reaching tertiary education, while students from high-income families complete secondary education in private schools and then move on to virtually free public tertiary education, is not optimal from an equity point of view. Ultimately, rising tuition fees in public universities, accompanied by means-tested waivers, could be an option to increase equity.

Improving governance and evaluation mechanisms

Improving efficiency and evaluation mechanisms and increasing accountability across all the education system, including universities, is also needed. Recent significant increases in education spending have not been accompanied by an improvement in outcomes. A significant share of recent increases in education spending was dedicated to raise teachers’ wages, which increased by more than 20 percent in real terms in the period 2009-13 (Estado de la Educación, 2015). This could be beneficial if it leads to hire more high-quality teachers. In general, the countries that perform well in PISA attract the best students into the teaching profession by offering them higher salaries and greater professional status (OECD, 2013b). Improving teacher’s professional development, including those in vocational tracks, and harmonising their qualifications would also help to improve the quality of teaching.

The salary structure includes numerous bonuses completely delinked from educational outcomes that are perceived as part of the fixed salary (Estado de la Educación, 2015). Annual increases are granted quasi-automatically without any relationship with performance. The school assessment system needs to be reformed and rewards should be linked to improving outcomes. Those rewards should not depend exclusively on an improvement in test results but should also reward student’s retention. Countries that have embarked recently in similar reforms have seen an improvement in educational outcomes. Brazil established recently nationwide competency tests that allow to benchmark schools and to introduce incentive mechanisms (OECD, 2013c).

Increasing efficiency in the health system

Costa Rica has a long-standing commitment to provide universal health care, which has translated into outstanding health outcomes, similar to those in OECD countries. Life expectancy is 82 for women and above 77 for men, and child mortality is low. Nevertheless challenges are also surmounting in this area as concerns the growth, sustainability, and level of equity of the health system.

Public health services in Costa Rica are provided by the Costa Rican Social Security Agency (Caja Costarricense de Seguro Social, CCSS), which is the largest decentralised autonomous entity in the Government. This institution also administers the contributory pension fund and implements the non-contributory pension scheme (more below). The health insurance provided by the CCSS is based on the solidarity principle. It is financed with contributions from workers and employers, via social security contributions. Around 86% of population is insured by the CCSS. It also receives transfers from the national budget, to cover the health needs of poor and uninsured.

Health spending has grown since 2000s and is now 10% of GDP, which is above average spending in OECD countries (Figure 1.19, panel A). The recent increase in spending has not been coupled with an increase in revenue growth, raising concerns about the sustainability of the system, especially in view of demographic trends. Recent spending increases have been nearly exclusively devoted to raising the number and salary of employees. At the same time, average daily hospital production has dropped and the number of outpatient visits per professional has declined (Word Bank, 2015). Users face increasingly long wait times, especially for surgery, diagnosis, and specialised treatments. As a consequence, out of pocket spending is increasing and now is higher than the OECD average (Figure 1.19, panel C). Private health services are increasingly used by those can afford it to get services covered by the CCSS. This is generating inequities in access and quality of treatment, and increasing patient dissatisfaction.

CCSS performance is hampered by excessive fragmentation and dispersed information systems. There are multiple non-linked IT tools, some of which take the form of spread sheets, whose update and reconciliation is done dispersedly and manually. An update of the information system underpinning the CCSS is a crucial and much needed step in order to improve the management of the system. Compiling in a more efficient, timely and accurate way key information, such as financial statements, unitary costs per intervention or waiting lists, would help to make the system more efficient, accountable and performance-based. In this direction, progress is underway to roll-out the single digital medical file, which will provide real time statistics and performance indicators. The single medical file has already been deployed to half of the medical centers providing primary care and it will be gradually deployed to other medical centres (also those providing other than primary care) in 2016 and 2017. The information gathered through this initiative will be fed into SINIRUBE, the single social assistance beneficiaries’ database, and hence could also help to improve the targeting of social assistance programmes.

Changes are also needed in the way resources are allocated across the system. So far Costa Rica allocates resources based on a historical basis, without taking into account demand originating from population needs and morbidity patterns. This leads to inefficiencies and inequities. A move towards a budget model based on epidemiological needs that respond more swiftly to demand and demographic changes is needed. This should be accompanied by introducing diagnosis-related funding schemes, i.e. hospitals should be funded according to patient characteristics. OECD evidence shows that diagnosis-based schemes are very effective at containing costs without prejudice to the quality of services, as they provide stronger incentives to control spending than fee-for-service schemes, which can result in service oversupply (Pisu, 2014). On the human resources side, wage increases should be linked to performance.

Costa Rica needs also to update its health model to respond to the aging of the population, which has already brought changes in the epidemiological profile of the country and increasing demands. Chronic diseases are becoming more prevalent. This will require changes in the health model, putting a stronger emphasis on prevention and on formal home care services. The latter are increasingly used across OECD countries as a preferable and less costly alternative to long hospital stays.

Buttressing the pension system

Pension coverage in Costa Rica, though not universal, is relatively large, especially when compared with other countries in Latin America. Seventy per cent of Costa Rican workers contributed to or were affiliated with a pension scheme in 2010, which is one of the largest coverage in Latin America and Caribbean countries (OECD/IDB/The World Bank, 2014).

The Costa Rican pension system comprises four pillars: 1) a contributory defined-benefit scheme; 2) a non-contributory regime paying a minimum pension (below the poverty line), which is financed from the government budget; 3) a compulsory contributory defined-contribution scheme; and 4) a voluntary defined contribution scheme. In addition there had been special regimes for public employees, but they were mostly closed to new beneficiaries in the 1990s.

Nearly all workers in the highest income quintile contribute to the first pillar, while around 20% of those in the lowest quintile do so. Costa Rica is also one of the few Latin America countries getting significant pension savings from the self-employed. Given the low labour participation of women, Costa Rica has the largest gender pension coverage gap in Latin America (OECD/IDB/TheWorld Bank, 2014). Coverage is 12 percentage points higher for men than for women.

The non-contributory scheme covers 44 % of the population aged 65 and over. It is managed by the CCSS, the Costa Rican Social Security Agency (Caja Costarricense de Seguro), and financed with resources from the Social Development and Family Assistance Fund (FODESAF), and certain specific taxes. It is means-tested. The scheme seems well targeted at those more in need and contributes to reduce income inequality (Gabriel and González Pandiella, forthcoming).

Two main weaknesses are affecting the contributory system. On the one hand, it does not generate enough resources to be sustainable in the medium-term, even after allowing for the 2005 reform that mandated a progressive increase in contributions from 8.5% to 10.5% by 2035. There are different estimates concerning when the system will turn into deficit. According to CCSS’s projections, it will be in 2038; according to the pension regulator it would be in 2025. Either the system will require funding in addition to contributions, or it will need reform. Pension entitlement parameters, for example pension benefits or the statutory retirement age, could be indexed to changes in life expectancy. To increase the linkage between contributions and pensions the government could also consider phasing in a pension benefit calculation formula that reflects a larger proportion of participants’ working life. Currently only the last 300 monthly salaries are used to determine pension entitlements.

The second weakness is that its assets are highly concentrated: 96% of CCSS assets are invested in Costa Rica sovereign debt. A downgrade in the credit classification of Costa Rica, for example, would have a severe impact on the system sustainability. The CCSS is guaranteed by the government, which make it an implicit government liability. A more diversified financing strategy that decreases investment in government securities would reduce risks.

There is also a need to increase the transparency of the system. The pension regulator has reported unmet requests for key information to perform its role as regulator (SUPEN, 2014). To increase transparency and accountability, a road map to adhere to the OECD Guidelines for Pension Fund Governance (OECD, 2009) should be established.

Sustaining and reinforcing environment protection

Costa Rica has built a world-distinguished green trademark, centred on conservation, reforestation and national parks. This has been an important engine of economic growth and wellbeing for Costa Ricans. Thanks to the Payments for Environmental Services program (PES), Costa Rica is the only tropical country which has substantially reversed deforestation (Word Bank, 2015). Costa Rica is a pioneer in carbon pricing. In 1997, it enacted a tax on fossil fuels, whose revenue funds the PES. Costa Rica is also a pioneer in the development of eco-tourism, with a positive impact on the income of local workers (Robalino and Villalobos, 2014). Costa Rica is one of the most visited countries in Latin America. More than 2.5 million tourists travelled to Costa Rica in 2014. Protected areas, which cover 26% of the country, have contributed to reduce poverty in local communities, mostly through new job opportunities afforded by tourism (Ferraro and Hanauer, 2014). Thus, Costa Rica exemplifies that protecting national resources can pay off and be an additional source of growth and jobs and provide a way out of poverty.

Costa Rica has announced ambitious targets to decarbonise its economy while reiterating its aspiration to align its economy towards carbon neutrality by 2021 and establish a long term goal for decarbonising its economy throughout the rest of the century, including milestones for emissions reductions by 2030, 2050 and 2100 (MINAE, 2015). These encompass a target to reduce total net emissions (i.e. taking into account CO₂ removals by forests) by 25% by 2030 compared with 2012.

In 2012, Costa Rica established the Domestic Voluntary Carbon Market, whereby carbon credits can be generated and exchanged among companies and individuals. Further efforts are ongoing to consolidate and expand this market accompanied by other recent initiatives, such as the creation of certificates of carbon neutrality for companies and several sectoral mitigation programmes, the so–called Nationally Appropriate Mitigation Actions (NAMAs). These efforts should be pursued as they are key towards Costa Rican carbon neutrality aspiration.

Costa Rica has in place a number of green fiscal policy tools. For example, a tax on fossil fuels was enacted in 1997, and Costa Rica ranks above the OECD average in terms of revenues collected from environmentally-related taxes (Figure 1.20).

Costa Rica faces a number of important environmental challenges, especially as concerns the impact that increasing urbanisation is having in the environment and quality of life of Costa Ricans. The urban population has been growing rapidly, from 50% of the total population in 1990 to 75% in 2013. More decisive policy actions will also be required to achieve progress towards the objective of decarbonising its economy. This mainly centres on the necessity to reduce emissions originated by oil energy consumption. Infrastructure gaps need also to be addressed since they hurt not only Costa Rican competitiveness (Chapter 2) but also its environment.

The transport sector is the largest source of emissions, accounting for 60% of all energy consumption and 67% of GHG emissions from energy in 2010. This is explained by the persistent increase in the number and use of motor vehicles, in particular private cars, which grew by 68% from 2003 to 2014. The number of private cars per capita in Costa Rica is above that of OECD countries in the region, like Chile, or neighbouring countries such as Panama. The extended and increasing use of private cars reflects an underdeveloped and inefficient public transportation system network (see Chapter 2), which means that in many cases a private car is the only transport option. This is particularly the case in the San José Greater Metropolitan Area, where traffic congestion is heavy and commutes are long, as a result of an obsolete public transport system. This imposes very high environmental costs, such as climate change, noise and air pollution. Air pollution, measured by the concentration of particulate matters in the air of residential areas, is above the OECD average (26 versus 20).

Costa Rica has put in place a number of initiatives to foster the use of clean cars and encourage the vehicle fleet becoming more energy efficient. In March 2015, it announced the Acquisition Programme for Efficient Vehicles (PAVE), which subsidises buyers of new cars in certain categories. Since 2013, importing hybrid and electric vehicles benefit from a tax rebate and a tax exemption respectively. However, these initiatives may take a long time to deploy their effect. A meaningful reduction in emissions from the transport sector will only come from more decisive steps in the design and implementation of a public transport network, especially in San José Greater Metropolitan Area, home to almost two-thirds of the population. Increasing and improving urban public transport should be a priority for public capital spending. Better coordination between land-use and transport planning would also facilitate a lower use of private cars.

Beyond transportation, the second set of challenges originated from the increasing urbanisation refers to sewage treatment and solid-waste management. Seventy five percent of residential wastewater is diverted to septic tanks. Sewers are used by only 20% of the population, and only 4% of that fraction receives remedial treatment. This is well below OECD standards and also below standards in other neighbouring countries (Table 1.5). Of all urban wastewater collected in Costa Rica, 96% of it is discharged into rivers and receiving water bodies without any treatment, generating water contamination. Water contamination has been in same cases severe, resulting in beaches being declared unsuitable for swimming. This has consequences for public health and for the sustainability of Costa Rica environmental protection model. To respond to these challenges, the first ever collection system in the Metropolitan Area of San José is being built. The government should aim at building additional waste-treatment plants with the aim of gradually covering 100% of water discharges.

Table 1.5. Waste water treatment is poor
Country	Access to water source (% of urban pop)	Access to sanitation (% of urban pop)	Connection to sewage treatment (%)	Sewage treatment (%)
Costa Rica	99.6	94.9	8.0	4.0
Chile	99.6	100.0	91.0	72.0
Lithuania	99.3	98.7	74.0	48.9
Brazil	99.7	87.0	26.0	20.0
Mexico	96.1	87.0	62.0	13.0
OECD average	99.8	93.4	76.0	-
Source: World Bank, World Development Indicators Database, 2012; Estado de la Nación 2014; UN stat; Eurostat; Aquastat FAO data; UNEP 2000; OECD Water Database; OECD Environment at a Glance 2013.

Costa Rica generates most of its electricity (88%) from local renewable sources, mostly hydropower and geothermal (Figure 1.21). This is well above clean electricity generation observed in OECD countries and contributes to low CO₂ emissions per capita. At the same time, the electrification rate is relatively high and reaches 99% of population. Nevertheless, during the last years, due to hydrological factors, electricity generation based on bunker oil and diesel has increased. Only in 2013 it grew by 45% (Blanco, 2014). Increased urbanisation implies that demand for electricity is expected to double over the next 15 years (MINAE, 2012). The high dependence on hydropower makes Costa Rica electricity model particularly vulnerable to climate challenges, especially droughts, as recently exemplified by El Niño phenomenon. Diversifying renewable electricity sources would make Costa Rican electricity model more robust and avoid the eventual need to resort to oil sources. This would imply boosting the use of geothermal, solar, wind and biomass energy sources also with the participation of the private sector. Positive steps have recently been taken as concerns distributed electricity generation such as solar. After a successful pilot project, a regulation is in preparation so that households and businesses will be able to generate photovoltaic electricity for their own use.

These efforts should be complemented by efforts on the demand side. The aim should be to promote a more efficient use of energy. Indeed energy efficiency is an important element in the policy agenda of the Ministry of Environment and Energy. That agenda includes plans to establish hourly tariff schemes for the residential sector, to encourage more cost-effective demand patterns, and the establishment of an energy-efficiency certification programme, among many other actions, to encourage energy efficiency across all sectors.

Policy recommendations

Making the labour market more inclusive

Foster active labour market policies by strengthening public employment services and by providing training for the unemployed, with a focus on the low-skilled.
Increase the supply of publicly-funded child care services to facilitate women participation in the labour market.
Consider establishing an unemployment insurance scheme based on individual savings accounts complemented with a solidarity fund.

Reducing informality

Adopt a comprehensive strategy to reduce informality by strengthening enforcement, reducing administrative burdens to entrepreneurship, and enabling the poor to become formal workers.
In the medium term, lower social security contributions with a focus on those segments where informality is rampant, such as workers in construction, agriculture and domestic work.
Set-up one stop shops for facilitating the formalisation of micro and small firms.
Simplify the minimum wage structure and enforce compliance with the law.

Improving education

Establish better educational outcomes, such as raising graduation rates and PISA results in secondary education, as the main policy target.
Provide additional support to disadvantaged students and schools by scaling up early detection and tutoring classes.
Develop an apprenticeship system that closely involves employers; make technical education more responsive to labour market needs by increasing workplace education.
Establish schools evaluation mechanisms and introduce performance-based teachers pay; provide teachers with support and professional development opportunities.
Make the University system more responsive to labour market needs by introducing funding mechanisms that encourage a better alignment of courses and curricula with skills demand.

Improving the tax and transfer system

Reduce fragmentation of social programmes and, based on a common registry of beneficiaries, adjust eligibility criteria to ensure that recipients are those most in need.
Set-up systematic and sound evaluations of social assistance schemes, reallocate more resources to those empirically proven to lift people out of poverty and scale down ineffective ones.
Increase the progressivity of the tax system and broaden tax bases by eliminating deductions benefiting more affluent taxpayers, gradually lowering the threshold under which no personal income tax is paid and setting the new top income brackets to lower income thresholds. Go forward with the planned VAT reform.

Buttressing the pension system

To improve the sustainability of the contributory system, index pension entitlement parameters, such as benefits or the statutory retirement age, to changes in life expectancy.
To increase transparency and accountability, adhere to the OECD Guidelines for Pension Fund Governance.

Modernising the health system

Move to a hospital budget allocation system based on needs stemming from population profiles, rather than on a historical basis.
Update the information system underpinning the health system, so as to allow for a more accurate and timely control of costs and performance.
To respond to aging challenges, put a stronger emphasis in preventive and home care.

Sustaining environment protection

Improve public urban transport and wastewater management facilities.
Continue efforts to develop the carbon market and other climate-change mitigation schemes.

References

Almeida, R. and P. Carneiro (2012), “Enforcement of Labor Regulation and Informality”, American Economic Journal: Applied Economics, American Economic Association, Vol. 4(3), pp. 64-89, July.

Andrews, D., A. Caldera Sánchez and Å. Johansson (2011), “Housing Markets and Structural Policies in OECD Countries”, OECD Economics Department Working Paper, No. 836, January.

Blanco, J.M. (2014), “Energía en Costa Rica: un desafío perenne”, Proyecto de Capacitación y Diálogo sobre Ambiente y Economía, Academia de Centroamérica, San José, 19 de setiembre 2014.

Blanco, A, V. Fretes and A. Muñoz (2014), “Rental Housing Wanted: Policy Options for Latin America and the Caribbean”, Inter-American Development Bank, Monograph 181.

Berg, J. (2010), “Laws or Luck? Understanding Rising Formality in Brazil in the 2000s”, ILO Office in Brazil. Working Papers, No. 5.

Brys, B. (forthcoming) “Tax Policy and Tax Reform in Costa Rica”, Internal OECD document.

Carrillo, P. , N. Gandelman and V. Robano (2014), “Sticky Floors and Glass Ceilings in Latin America”, Journal of Economic Inequality, Springer, Vol. 12(3), pp. 339-361, September.

Causa, O., A. de Serres and N. Ruiz (2015), “Can Pro-growth Policies Lift all Boats?: An Analysis Based on Household Disposable Income”, OECD Economics Department Working Papers, No. 1180, OECD Publishing, Paris, https://doi.org/10.1787/5jxrh8dh5wg7-en.

Census (2011), “Costa Rica – X Censo Nacional de Población y VI de Vivienda., Censo 2011”, Instituto Nacional de Estadísticas y Censos.

CONARE (2015), “Seguimiento de la Condición Laboral de las Personas Graduadas 2008-2010 de las Universidades Costarricenses”, Consejo Nacional de Rectores, OPES No. 06/2015.

Estado de la Nacion(2014), “Equidad e Integración Social”, Capítulo 2, San José, Costa Rica: Programa Estado de la Nación.

Estado de la Educación (2015), “Quinto Informe Estado de la Educación”, San José, Costa Rica: Programa Estado de la Nación.

Fernández A. and R. Del Valle (2013), “Desigualdad educativa en Costa Rica: la brecha entre estudiantes de colegios públicos y privados. Análisis con los resultados de la evaluación internacional Pisa”, Revista CEPAL, 11 December 2013.

Ferraro, P.J. and H, Merlin (2014), “Quantifying Causal Mechanisms to Determine How Protected Areas Affect Poverty Through Changes in Ecosystem Services and Infrastructure”, Proceedings of the National Academy of Sciences, Vol. 111, No. 11, pp. 4332-4337.

Gabriel, M. and A. González Pandiella (forthcoming), “Recent Evolution of Inequality in Costa Rica: A Decomposition by Income Source Analysis”.

Gallup-Healthways Global (2015), Country Well-Being Rankings.

Hoeller, P. et al. (2012), “Less Income Inequality and More Growth – Are They Compatible? Part 1. Mapping Income Inequality Across the OECD”, OECD Economics Department Working Papers, No. 924, OECD Publishing, https://doi.org/10.1787/5k9h297wxbnr-en.

Ibarraran, P. and D. Rosas (2009). “Evaluating the Impact of Job Training Programmes in Latin America: Evidence from IDB Funded Operations”, Journal of Development Effectiveness 1 (2), pp. 195-216.

IMAS (2015), “Puente al Desarrollo: Estrategia Nacional para la Reducción de la Pobreza”, Costa Rica Gobierno de la República and IMAS”, http://presidencia.go.cr/wp-content/uploads/2015/03/Documento-Estrategia-Puente-al-Desarrollo.pdf.

IMF (2014), “Back To Work: How Fiscal Policy Can Help”, Fiscal Monitor, International Monetary Fund, Washington.

INEC (2015a), “Encuesta Nacional de Hogares Julio 2015 – Resultados Generales”, San José, Costa Rica. Instituto Nacional de Estadística y Censos (INEC).

INEC (2015b), “Índice de Pobreza Multidimensional (IPM)”, October.

Katz, Lawrence F. (1998), “Wage Subsidies for the Disadvantaged”, in Richard B. Freeman and Peter Gottschalk (Eds.), Generating Jobs: How to Increase Demand for Less-Skilled Workers, Russell Sage Foundation, New York.

Lehmann, H. and A. Muravyev (2012), “Labor Market Institutions and Informality in Transition and Latin American Countries”, IZA Discussion Papers, No. 7035.

Lustig, N. , C. Pessino and J. Scott (2013), “The Impact of Taxes and Social Spending on Inequality and Poverty in Argentina, Bolivia, Brazil, Mexico, Peru and Uruguay: An Overview,” Working Papers 1313, Tulane University, Department of Economics.

Manpower Group (2015), “Talent Shortage Survey”, www.manpowergroup.com.

Mateo-Diaz, M. and L. Rodriguez-Chamussy (2013), “Childcare and Women’s Labor Participation: Evidence for Latin America and the Caribbean”, Inter-American Development Bank / IDB Technical Note; p. 586.

Mazza, J. (2013), “Connecting Workers to Jobs: Latin American Innovations in Labor Intermediation Services”, Latin American Policy, 4: 269–284, http://doi:10.1111/lamp.12020.

Ministerio de Hacienda (2014), El Gasto Tributario, Metodología y Estimación, San José, Costa Rica.

MINAE (2015), National Energy Plan 2015-2030; www.minae.go.cr/recursos/2015/pdf/VII-PNE.pdf.

MINAE (2012), “Agenda energética”, www.minae.go.cr/index.php/es/2012-06-08-20-20-39/gris.

Monge González, R., L. Rivera, and J. Rosales Tijerino (2010), “Productive Development Policies in Costa Rica: Market Failures, Government Failures, and Policy”, IDB Working Paper Series, No. IDB-WP-157, Inter-American Development Bank, Washington, DC.

Monge González, R.; Hewitt, J.; and Torres-Carballo, F. (2015), “Do Multinationals Help or Hinder Local Firms? Evidence from the Costa Rican ICT Sector”, Comisión Asesora en Alta Tecnología Working Paper, http://caatec.org/sitio1/index.php/en/publications/others-.

OECD (2015a), LAC Revenue Statistics, OECD Publishing, https://doi.org/10.1787/rev_lat-2015-en-fr.

OECD (2015b, OECD Economic Survey of Mexico, OECD Publishing, Paris, https://doi.org/10.1787/eco_surveys-mex-2015-en.

OECD (2015c), “Market Openness Review of Costa Rica”, Trade and Agriculture Directorate Trade Committee (unpublished).

OECD (2015d), OECD Employment Outlook, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2015-en.

OECD (2015e), A Skills Beyond School Review of Costa Rica, OECD Reviews of Vocational Education and Training, OECD Publishing, Paris, https://doi.org/10.1787/9789264233256-en.

OECD (2015f), OECD Education at a Glance, OECD Publishing, Paris, https://doi.org/10.1787/eag-2015-en.

OECD (2015g), OECD Economic Outlook Database, OECD, Paris, https://doi.org/10.1787/bd810434-en.

OECD (2014a), Skills beyond School – Synthesis Report, OECD Publishing, Paris, https://doi.org/10.1787/9789264214682-en.

OECD/IDB/TheWorld Bank (2014), Pensions at a Glance: Latin America and the Caribbean, OECD Publishing, Paris, https://doi.org/10.1787/pension_glance-2014-en.

OECD (2013a), OECD Economic Survey of Chile, OECD Publishing, Paris, https://doi.org/10.1787/eco_surveys-chl-2013-en.

OECD (2013b), PISA 2012 Results: What Makes Schools Successful? Resources, Policies and Practices (Volume IV), Pisa, OECD Publishing, https://doi.org/10.1787/9789264201156-en.

OECD (2013c), OECD Economic Survey of Brazil, OECD Publishing, Paris, https://doi.org/10.1787/eco_surveys-bra-2013-en.

OECD (2012a), Closing the Gender Gap, OECD Publishing, Paris, https://doi.org/10.1787/9789264179370-en.

OECD (2012b), OECD Education at a Glance, OECD Publishing, Paris, https://doi.org/10.1787/eag-2012-en.

OECD (2011), Starting Strong III: A Quality Toolbox for Early Childhood Education and Care, OECD Publishing, Paris, https://doi.org/10.1787/9789264123564-en.

OECD (2010), Off to a Good Start? Jobs for Youth, OECD Publishing, Paris, https://doi.org/10.1787/9789264096127-en.

OECD (2009), OECD Guidelines for Pension Fund Governance, OECD Publishing, Paris, www.oecd.org/dataoecd/18/52/34799965.pdf.

Pisu, M. (2014), “Overcoming Vulnerabilities of Health Care Systems”, OECD Economics Department Working Papers, No. 1132, OECD Publishing, Paris, https://doi.org/10.1787/5jz159228n6j-en.

Puente al Desarrollo (2015), Estrategia Nacional para la Reducción de la Pobreza.

Ramírez-Alfaro, L. (2010), “La informalidad y semiformalidad empresarial en Costa Rica: un acercamiento desde la mirada de personas expertas”. Universidad de Educación a Distancia. Serie observatorio de MIPYMES. San José, Costa Rica.

Rani, U., P. Belser, M. Oelz and S. Ranjbar (2013), “Minimum Wage Coverage and Compliance in Developing Countries”, International Labour Review, Vol. 152/3-4.

Robalino, J. and L. Villalobos (2015), “Protected Areas and Economic Welfare: An Impact Evaluation of National Parks on Local Workers’ Wages in Costa Rica”, Environment and Development Economics, 20(3), pp. 283-310. https://doi.org/10.1017/S1355770X14000461.

Ruhm, C. and J. Waldfogel (2011), “Long-Term Effects of Early Childhood Care and Education”, IZA Discussion Paper, No. 6149, Bonn.

Sauma, P. and J. D. Trejos (2014a), “Impacto de la política fiscal en la distribución del Ingreso y la pobreza en Costa Rica”. Ciencias Económicas 32-No. 2.

Sauma, P. and J. D. Trejos (2014b), “Social Public Spending, Taxes, Redistribution of Income and Poverty in Costa Rica”, CEQ Working Paper, No. 18.

SUPEN (2014), Informe a la Junta Directiva de la CCSS y al Comité de Vigilancia del Regimen de Invalidez, Vejez y Muerte. Superintendencia de Pensiones (SUPEN). August 2014.

Teignier, M. and D. Cuberes (2014), “Aggregate Costs of Gender Gaps in the Labor Market: A Quantitative Estimate”, UB Economics Working Papers, 2014/308, Universitat de Barcelona, Facultat d’Economia i Empresa, UB Economics.

Thévenon, O. (2013), “Drivers of Female Labour Force Participation in the OECD,” OECD Social, Employment and Migration Working Papers, No. 145, OECD Publishing.

Tsounta, E. and A. Osueke (2014), “What is Behind Latin America’s Declining Income Inequality?” IMF Working Paper, WP/14/124.

Umaña-Alvarado, E. (2014), “Welfare Effects of a Change in the Trade Policy Regime for Rice in Costa Rica”, United Nations Conference on Trade and Development Study.

Word Bank (2015), “Costa Rica Systematic Country diagnostic”, Central America Country Unit, World Bank Group.