Chapter 5. Labour market developments and pensions

This chapter analyses labour market developments in Portugal that are relevant from a pension perspective. It shows that while short career histories are uncommon among current retirees, risks of short careers have been building up for future generations of retirees, among other things as a result of the economic crisis. The chapter also describes the different pathways into retirement in the Portuguese pension system, including specific early retirement regulation for long-term unemployed people. It then shows that some forms of non-standard work, such as work on temporary contracts, are frequent in Portugal and discusses the impact of changing labour markets on old-age protection systems.


5.1. Introduction

When the global financial crisis and the European debt crisis hit Portugal in the late 2000s and early 2010s, the country’s economy was deeply affected, with severe repercussions for its labour market (OECD, 2017[1]). Unemployment surged, reaching about 17% of the labour force in 2013, and barriers to labour market entry grew larger, especially for young people. The unemployment rate among 20-24 year-olds spiked at 39% in 2013 and the rate of youth not in employment, education or training (NEET) climbed to its highest level in decades, at 18% in 2014. Worries arose that these labour market difficulties could have knock-on effects and result in disrupted careers, eventually causing low pensions. Despite remarkable improvements over the last years, the Portuguese labour market still faces major challenges.

This chapter takes stock of the employment situation from a pension perspective, acknowledging the tight link between labour market experiences and pension entitlements. In a first step, it documents the prevalence of short careers and summarises how interrupted careers impact on old-age provisions in Portugal. After that, it analyses pathways into retirement, stressing that early retirement, including after spells of unemployment, is relatively frequent. The chapter then discusses how non-standard forms of employment have been evolving over the last years in Portugal and how workers in such jobs are covered by old-age social protection. Finally, building on this analysis, the last part proposes policy options to improve pension outcomes.

5.2. Career length, career interruptions and pensions

As most other OECD countries, Portugal has a contributory pension system in which career length, the frequency of career interruptions and the age at which workers stop working are decisive factors for the level of pension entitlements. Pension schemes generally pay higher pensions to workers with extended working lives, thereby creating incentives to work longer. Conversely, workers with short contribution histories face penalties and may be at risk of low old-age income. Especially when contributory schemes are the primary source of old-age income, long career breaks and early labour market exits can be a hurdle to adequate pension levels. In the OECD, both the frequency of short careers and the impact of career breaks on pension entitlements vary considerably across countries.

5.2.1. Short careers are uncommon among older workers, but risks have grown

Employment rates at older ages fall quickly with age in Portugal as in the OECD on average. The employment rate among the 55-59 year-olds was 68% in 2017 against 71% on average across OECD countries (Figure 5.1). It reaches 44% among 60-64 year-olds and 19% among 65-69 year-olds in Portugal, compared with 48% and 22%, respectively, on average in the OECD. Some non-EU countries, including Iceland, Japan and New Zealand, are able to avoid such a sharp fall and have much higher employment among the 65-69. Moreover, there is a stark employment gender gap in Portugal, of about 10-15 percentage points in the age-groups 55-59, 60-64 and 65-69, which is also in line with the OECD average.

Figure 5.1. Employment rates at older ages are slightly lower than the OECD average
Employment rate by age, in %, 2017

Source: OECD Labour Force Statistics.


While employment among people in their late 50s and 60s is close to, but below the OECD average, a comparatively large proportion of workers in Portugal remains employed beyond age 70 (Figure 5.2). Among 70-74 year-old men, 18% still work in Portugal – very close to the level for the 65-69 age group - while they are only 15% in the OECD.1 Among women, the employment rates of 70-74 year-olds in 2017 stood at 7% both in Portugal and the OECD average. These levels are high compared to other EU countries.

Short careers are relatively rare in Portugal among people who are currently close to retirement age (Figure 5.3). Only 11% of 65-69 year-olds report a professional experience of less than 30 years, suggesting that long career breaks are not widespread. In Luxembourg, Italy and Greece, for example, short careers are by far more frequent, with shares of over 30%. Only few countries report lower levels than Portugal, reaching a low of 6% in the Czech Republic and the Slovak Republic.

As with differences in labour market exit ages across genders, short careers are more frequent among women in Portugal. While only 3% of 65-69 year-old men report careers of less than 30 years, the level is much higher among women of the same age, at close to 17%. Therefore, pension penalties for short and interrupted careers affect substantially more women than men. Accordingly, old-age poverty is more frequent among women, with the relative income poverty rates – defined as living on less than 50% of median disposable household income - reaching 12% among women over 65 against 7% among men over 65 (OECD, 2017[2]). This compares to 14% and 9% on average in the OECD, respectively.

Figure 5.2. Employment rates among 70-74 year-olds are high
Employment rates among 70-74 year-olds, 2017

Source: OECD Labour Force Statistics.


Figure 5.3. Short careers are rare among older people in Portugal today
Share of 65-69 year-olds with less than 30 years of professional experience, 2016

Source: EU-SILC 2016.


The relatively limited incidence of short careers results from long-term extensions in career length. The share of the population with very short careers at a given age fell from one birth cohort to the next in Portugal among those born between 1945 and 1960 (Figure 5.4, Panel A). However, in recent years, such improvements have come to a halt; the share of people with a career length of less than 15 years at a given age has been very similar for those born in 1960, 1965 and 1970 (Panel B). While longer periods of education partly explain why such improvements have not been extended for younger generations, other factors seem to be at play.

Figure 5.4. Short careers are now less frequent, but improvements have come to a halt
Share of Portuguese population with less than 15 years of work experience at a given age, by birth cohort

Source: EU-SILC, different waves.


The Great Recession deteriorated employment prospects markedly at least in the short-to-medium term, thus increasing the risk of short careers among future generations of retirees. Unemployment rates at all ages surged during the crisis (Figure 5.5), triggering longer and repeated career breaks. Unemployment rates had already been higher in the 2000s than in the 1990s, growing from about 4% among 35-55 year-olds in the 1990s to close to 6% in the 2000s. When the economic effects of the crisis set in, unemployment spiked, raising the average unemployment rate of 35-55 year-olds between 2010 and 2017 to 11%.

The negative labour market effects were strongest among young people. The average unemployment rate between 2010 and 2017 reached 17% among 25-29 year-olds and 29% among 20-24 year-olds, up from 9% and 14% a decade earlier, respectively. High unemployment rates among young people are worrisome by themselves, but even more so as career difficulties early in life can have knock-on effects, with repercussions on career prospects later in life and ultimately low pension entitlements. While unemployment among young people has fallen since the peak of the crisis, it is still high today (Chapter 1).

Figure 5.5. Unemployment soared due to the crisis, especially among the young
Average unemployment rate by decade and age

Source: OECD Labour Force Statistics.


5.2.2. Impact of career breaks on pensions

Career length is an important factor for pension entitlements and short and interrupted careers usually lead to lower pension levels. Pension entitlements are not equally sensitive to incomplete careers across the OECD, however. While very short career breaks tend to reduce future pension levels to a limited extent only in most OECD countries, longer breaks pose serious challenges for old-age income levels in some countries.

On average across the OECD, average-wage workers who enter the labour market at age 25 and experience a 10-year unemployment spell during their career face a pension reduction of 20% in mandatory schemes compared to workers with a full career from age 20 (Figure 5.6). In most OECD countries, the loss ranges between 15% and 30%, exceeding 30% in three OECD countries. Conversely, in Ireland, New Zealand and the United Kingdom, such career breaks are fully cushioned as the mandatory schemes only include flat-rate benefits in these countries.

If there was no offsetting mechanism to limit the impact of short careers on pension levels, such an incomplete career scenario would lead to a drop of about 35% in pension benefits (OECD, 2017[3]). This means that redistributive and stabilisation devices in the OECD on average offset more than one-third of the shortfall, bringing it down from 35% to 20%.

Portugal is among the OECD countries, along with among others France, Germany and Spain, where individuals with such a career break have to work longer than full-career workers to avoid additional penalties beyond lost accrual. While full-career workers entering the labour market in 2018 will be able to retire at age 67 and two months with a full pension (Chapter 3), workers with such a break will have to work two more years. Retiring later and the fact that accruals stop after 40 years of contributions for full career workers, limits the impact of an incomplete career on pension benefits to 10% in Portugal. Pension credits granted during periods of unemployment benefit receipt on the basis of previous earnings also help cushioning the effect. Because of these offsetting mechanisms in the pension system, almost two-thirds of the shortfall in lifetime earnings are not passed on to lower pensions. However, retiring at the same age as the full-career worker would generate a pension reduction of about 50%, more than half of which comes from the sustainability factor.

During unemployment spells, social security contributions corresponding to full-time employment are registered for pension accrual, thereby preventing an interruption of people’s contributory period and buffering the effects of career shocks. The maximum duration of unemployment benefit allowance, and hence of the period for which pension credits are granted, is not identical for all unemployed in Portugal. It depends on how long workers have contributed to the system since their last unemployment period and on their age, ranging from 150 days for people under 30 with short contributory histories to 540 days for people aged 50 or older with at least 24 months of contributions since their last unemployment spell. Beyond this maximum period, the unemployed do not accrue further pension entitlements.

As for career breaks due to childcare, periods of maternity and paternity leave (Licença parental) of usually up to 150 days (180 days if shared with the possibility of an extension of three months) are considered as periods of actual work and are taken into account to determine pension entitlements. Periods of childcare leave (Licença para assistência a filho) of up to two years (three years if three children or more) are considered as contributory periods, too. Beyond that, childcare-related crediting mechanisms do not exist, which generates fairly large cuts in pensions in case of long breaks due to childcare.

Figure 5.6. Loss in pension benefits due to incomplete careers is below the OECD average
Entry at age 25 with 10-year unemployment versus full career from age 20, average-wage workers, mandatory schemes

Note: The numbers in parenthesis in the country labels indicate the extra years individuals with incomplete careers need to work to access a full pension, i.e. without actuarial penalty although the pension might be lower than for a full-career worker. The incomplete-career case is based on entry at age 25 versus 20 in the baseline with a ten-year unemployment period between age 35 and 45.

Source: (OECD, 2017[3]).


5.3. Pathways into retirement

Given its link to life expectancy, the statutory retirement age in Portugal is expected to increase, thereby encouraging people to work until older ages. However, not all workers pursue a standard work-to-retirement transition at the statutory retirement age, some rather enter retirement early or follow alternative pathways into retirement. In Portugal, several early-retirement schemes permit retirement entry before the statutory retirement age, including an early-retirement scheme for long-term unemployed workers. While alternative pathways into retirement were widely used in many OECD countries 25 years ago, most countries have made sizable efforts to restrict their availability in an attempt to strengthen the financing of social security systems in general and pension systems in particular.

5.3.1. Standard early retirement

Workers aged between 60 and the official retirement age can claim a reduced pension if they have contributed at least 40 years. Until recently, 55-59 year-olds were also entitled to a reduced pension if they fulfilled a minimum of 30 contributory years. Between 2012 and 2015 the early retirement path for 55-59 year-olds was interrupted, and it was abolished in March 2016.

While permitting to draw pensions early, these schemes involve strong monetary penalties, leading to permanently reduced retirement benefits. For example, if legislated pension rules remain unchanged, currently young workers who will retire early at age 62, i.e. about 6 years short of the future retirement age, will face a pension reduction of about 60% at age 68. This very pronounced reduction is the result of three different types of penalties that add up. First, a sweeping 31% reduction will come into effect because the sustainability factor is applied when people retire early (Chapter 3).2 Second, benefits are reduced by 6% per year of early retirement. Finally, the indexation of benefits in payment to price inflation is less generous than based on wage growth according to long-term assumptions in the OECD pension model, resulting in an additional reduction of 1.25% per year of anticipation.

In contrast to the majority of employees, some groups of workers are exempt from such reductions and benefit from specific conditions that allow retirement at younger ages without penalties. Retirement entry is possible at earlier ages for workers in occupations that are considered as arduous. Seafarers (55 years), underground miners (50 years), classical ballet and contemporary professional dancers (45 years) and a few other precisely defined occupations fall into this category (Chapter 3). The number of workers in these types of occupations is limited, however.

5.3.2. Unemployment-related early retirement

Early retirement is also possible from age 62 if a worker has been unemployed since age 57, conditional on having made 25 years of contributions. While the sustainability factor takes effect in this case and reduces pensions, exiting employment through the unemployment route does not lead to a 6% penalty per year of early retirement, as is the case of the standard early-retirement scheme discussed above. This makes unemployment-related early retirement substantially more attractive financially.

A reduced pension is also payable to unemployed workers from age 57 if they have been unemployed since age 52 and have paid contributions for at least 22 years. In this case, in addition to the pension reduction through the sustainability factor, a penalty of 6% for each year of early retirement before age 62 also applies. This additional penalty is lower than in the case of standard early retirement, however, because it is not calculated up to the full retirement age, but only to age 62.

Such early-retirement options for unemployed workers might be intended to address the problem that older workers with unemployment histories tend to have difficulties to find new employment. However, they blur the signalling role of the normal retirement age and also bear the risk of both reducing job-search efforts among older unemployed people and limiting the willingness of employers to maintain older workers. They make dismissing older workers more socially acceptable and hardly serve as an incentive for the older unemployed to actively search for work.

Entering retirement before reaching the statutory retirement age is relatively common (Table 5.1). In 2016, over 30 000 people retired early, half of them through the unemployment-related schemes. This represents 45% of the total number of 66 700 new retirees in the general pension scheme. The number of new pensions in the unemployment-related early-retirement scheme has been high and relatively stable over time whilst the number of new pensions in other early retirement schemes has varied greatly, from very low levels in 2006 to high levels in 2011. Anticipated and actual legislative changes such as frozen access to standard (not unemployment-linked) early retirement and the introduction of the sustainability factor might explain much of this variation.

Table 5.1. New retirees in the general scheme and new early retirees (in thousands)




Annual average


Average single-year population size, 60-69 year-olds





New retirees in general

scheme, of which:





New early retirees,






New early retirees,

not unemployment-linked





Source: Portuguese administrative records.

5.3.3. Disability-to-retirement transition

Disability pensions are payable to permanently disabled workers below retirement age. The Portuguese system distinguishes between partial disability (workers unable to earn more than one-third of their normal wage, i.e. the wage that could be expected without disability in this occupation), complete disability and disability due to specific conditions with a high likelihood of impacting on work capacity, such as cancer and multiple sclerosis. Disability cases need to be confirmed by the Verification Committee for Permanent Incapacity (Commissão de verificação da incapacidade permanente, CVIP) and subsequent revisions of the disability status are possible, both at the initiative of the relevant institutions or upon request of the person concerned. In recent years, Portugal has intensified medical checks in order to supervise the use of disability benefits. For instance, the “National Plan to Combat and Prevent Welfare Benefits Fraud and Contributory Evasion” was introduced in 2010, leading to greater care in cross-checking and more frequent medical assessments. (OECD, 2018[4])

When beneficiaries reach retirement age, disability pensions are automatically transformed into old-age pensions. From the beginning of 2018, the sustainability factor no longer applies when a relative disability pension is converted into an old-age pension. As the sustainability factor applies to the early-retirement schemes discussed above and reduces retirement benefits substantially, individuals may have an incentive to attempt exiting the labour market through the disability scheme rather than early retirement.

5.3.4. Retirement entry after unemployment, disability and inactivity

Compared to other EU countries, Portugal is among those with the highest share of people transiting from unemployment to retirement, at 17% of all new retirees (Figure 5.7). Only Finland reports a higher level while the average across the 13 other countries for which data are available is 9.6%. Transitions from disability or other forms of inactivity to retirement are less widespread in Portugal. This situation contrasts with the case of a few other countries in which disability-to-retirement transitions are very common, such as Estonia and Finland. However, although only few retirees self-declare that their last labour force status prior to retirement was inactivity due to disability (Figure 5.7), according to administrative records disability pensions are paid to about 12 000 people for each birth cohort aged 55-64 on average, against an average birth-year cohort size of about 140 000 people, i.e. they are paid to about 8.6% of 55-64 year-olds. These numbers should therefore be interpreted cautiously.

Figure 5.7. Unemployment is a common pathway into retirement in Portugal
Last self-declared labour-force status (other than work) before retirement among 55-69 year-olds, 2016

Note: The remainder of pathways into retirement are direct work-to-retirement transitions (not shown).

Source: EU-LFS.


5.4. Non-standard employment and pensions

Most pension systems were designed with the case of workers with stable full-time careers in mind. Non-standard forms of employment such as self-employment, part-time work and temporary work are not marginal phenomena in OECD countries, and workers in such jobs are often less covered in terms of old-age provisions. This section focuses on recent trends in non-standard work in Portugal and describes the pension coverage of non-standard workers. It also discusses how future labour market developments may affect the frequency of non-standard work.

5.4.1. Self-employment rates have fallen in Portugal, but remain above-average

As in most other OECD countries, self-employment as a share of total employment has been decreasing over the last years, from 26% in 2000 to 17% in 2017 in Portugal (Figure 5.8). While this drop was among the largest in the OECD, Portugal’s current level of self-employment remains slightly above the OECD average of 15%. Only few OECD countries report higher self-employment rates, with the highest rates in Turkey (33%) and Greece (34%).

Self-employment as a share of total employment rises with age in Portugal. Self-employment rates are below 10% among people under 35 while they are considerably higher among older age groups, especially among workers close to the retirement age. About one-third of workers in their early-60s declare themselves as self-employed according to OECD calculations. Employment in agriculture, which is a sector many self-employed work in, has been declining in Portugal much more rapidly from a high level over the last years than in most other OECD countries, thus contributing to the decrease in self-employment.

It is uncertain whether the share of self-employment will continue to fall. Digitalisation, automation and globalisation can affect the functioning of labour markets and might trigger fundamental changes, possibly leading to higher levels of self-employment (OECD, 2018[5]). This might apply to younger workers who currently have a low rate of self-employment, for instance through platform work.

So-called “false” self-employment (falsos recibos verdes) is a matter of concern in Portugal (OECD, 2017[6]). This term refers to workers who are officially self-employed, but work de facto as dependent employees. While employers face lower risks and costs when they employ “false” self-employees, the workers face lower job security and a lower degree of social protection. There are legal mechanisms to detect and regularise such cases, but systematic monitoring is currently not guaranteed. However, ongoing modernisation processes may help facilitate the identification of such situations, namely by enhancing the interconnection between Social Security data and Tax Authority data.

While dependent employees pay a social security contribution rate of 34.75% on their earnings (including the employer and employee part) – 22.65 percentage points go to pensions and 5.14 p.p. to unemployment insurance – self-employed workers are subject to specific regulations. The rules regarding their pension insurance were reformed in the summer of 2018, with the goal of extending coverage and of increasing the overall transparency of the scheme.

According to the new regulation, most types of self-employed workers pay social security contributions amounting to 21.41% of their average reference income while the contribution rate is higher for specific types of self-employed workers with an entrepreneurial activity, at 25.17%. These contribution rates are applied to the average reference income over the last three months which is defined as 70% of the total value of the service provision and 20% of the total amount of income associated with the production and sale of products. Freelancers in the service industry, for instance, have a contribution base of 70% of their service provisions while sole traders in the production sector have a contribution base of 20% of their sales. In some occupations, this may lead to low contribution levels even after the reform. Prior to the reform, self-employed workers paid higher contribution rates (29.6% for most types of self-employed workers and 34.75% for some types of self-employed workers with an entrepreneurial activity) but could self-rank into income categories to determine the contribution base, i.e. it was very easy for them to pay very low contributions by choosing a low contribution base. Using average reference income over the past months aims to ensure that contributions are commensurate with actual earnings but do not react too sensitively to income fluctuations.

Figure 5.8. Self-employment rates have decreased in Portugal, but remain above-average
Self-employment as % of total employment, 2000 and 2017

Note: Countries marked by * refer to years shortly after 2000 or shortly before 2017. Estimates including people until 65 rather than all ages may vary.

Source: OECD (2018), Self-employment rate (indicator),


From a policy perspective, it might be desirable to ensure that social security in general and pension insurance in particular are aligned between self-employed workers and dependent workers. Contribution rates that are much lower for the self-employed provide implicit subsidies for self-employment; the policy objective pursued by this mechanism might be questionable. However, identifying the labour income part of self-employed revenues is not easy, and harmonising total contribution rates between self-employment and dependent-employment might encourage underreporting of income. Several countries use the minimum wage as the contribution base of the self-employed to cope with this problem.

When self-employed workers depend significantly on one single contracting entity – the so-called ordering customer - the latter is legally obliged in Portugal to pay social security contributions for the self-employed worker, at a rate that depends on the degree to which the worker relies on the ordering customer. Portugal is one of only few OECD countries with a legal definition of dependent self-employment (OECD, 2018[7]). When self-employed workers receive between 50% and 79% of their income from one single ordering costumer, a social security contribution rate of 7% applies. The rate increases to 10% when they receive 80% of their income or more from one ordering customer. Below 50%, customers do not pay contributions. Prior to the reform, ordering customers paid a contribution rate of 5% in case self-employed workers received at least 80% of their income from them and nothing if it was less. The reform thus extended the definition of dependency and increased the contribution rate of ordering customers.

The specific regulation for self-employed workers concerns a large number of older workers as self-employment is particularly frequent among people close to retirement age. The new measures, such as an increase in contributions by ordering customers, reduce the attractiveness of “false” self-employment, which is in part generated by lower contributions paid by the self-employed. However, the regulation is relatively complex and introduces discontinuities. Other ways to counteract self-employment that is in fact dependent employment may be more efficient, e.g. through improved processes to facilitate labour inspections.

5.4.2. Many workers are on temporary contracts

The Portuguese labour market remains highly segmented (OECD, 2017[6]). Over recent years, labour market reforms have reduced the stringency of regulations related to both permanent and temporary contracts. Given the large regulatory gap between both types of contracts, temporary work is widespread: 22% of workers are on temporary contracts, against 13% in the OECD on average (Figure 5.9). Most OECD countries report substantially lower rates of temporary employment, reaching a low of under 2% in Lithuania. While the use of temporary contracts shifts risks away from firms, potentially favouring their willingness to make new hires, workers on temporary contracts face a low degree of stability, with a higher risk of unemployment and interrupted contribution histories.

Men and women are equally likely to be on temporary contracts (22.3% and 21.7%, respectively). However, differences across age-groups are marked in Portugal. Two-thirds of workers under 25 are on temporary contracts, against one-fifth of 25-54 year-olds and one-tenth of 55-64 year-olds. These figures underpin that especially young people are exposed to a lack of stability, weakening their ability to develop professionally and build a long-term future.

Temporary workers are covered by the general Portuguese social protection scheme. Pension entitlements are identical for temporary and permanent contracts. As in the case of permanent contracts, the contribution rate for workers on temporary contracts amounts to 34.75% and is split between employers and employees. There is an exception for very short fixed-term contracts, however, with contributions reduced to 26.1%. In this case, contributions are entirely financed by employers. Apart from this exception, workers on temporary contracts and workers on permanent contracts are treated identically from a pension perspective. However, temporary work indirectly impacts on pension entitlements, as it increases the risk of career breaks and working less than 120 days, which are necessary to validate a contributory year (see above).

Figure 5.9. Temporary work is widespread in Portugal
Share of workers with temporary contracts

Source: OECD Statistics.


5.4.3. Part-time work is infrequent, but has been rising among the young

Part-time work is not a frequent phenomenon in Portugal. Only 9% of dependent employees are in part-time work in Portugal, against 16% in the OECD (Figure 5.10). Portugal belongs to the countries with the lowest shares of part-time work, contrasting with Australia (26%), Switzerland (27%) and the Netherlands (37%). Only few OECD countries, all of which are located in Central and Eastern Europe, report lower shares of part-time work than Portugal. The use of flexible working-time practices is likely to remain low in Portugal as long as there is a high share of temporary employment, which makes it easier for firms to adjust labour inputs by not renewing temporary contracts (OECD, 2017[6]).

Figure 5.10. Part-time is infrequent in Portugal
Share of part-time workers among dependent workers in OECD countries, 2017

Source: OECD statistics.


Working part-time is more widespread among women (11%) than among men (6%) in Portugal. With almost two-thirds of part-time workers being female, women are more exposed to the financial consequences in terms of lower pension entitlements. The frequency of part-time work also changes considerably with age (Figure 5.11). In the past, part-time work was primarily concentrated among older age-groups close to retirement age. However, during the crisis part-time work increased sharply among people in their 20s and has remained at high levels ever since, reflecting labour market difficulties among the youth. When the economic crisis hit the country, part-time work increased among older age-groups too, but rates have fallen significantly since 2012. Today, part-time is almost equally prevalent among 20-24 year-olds and 60-64 year-olds.

Figure 5.11. Part-time work increased among the young, but declined among older workers
Share of part-time workers by age in Portugal, 2006, 2012 and 2016, in %

Source: EU-LFS.


The financial consequences of working less than full-time are marked. Part-time workers are more likely to be at risk of poverty than full-time workers in all OECD countries for which data is available (Figure 5.12). In Portugal, the difference is particularly wide. Less than 10% of full-time workers in Portugal are at risk of poverty when poverty risk is defined as having a disposable income below 60% of the national median, against over 30% of part-time workers, suggesting that working part-time is often not a matter of choice. Only in Greece are poverty risks among part-time workers (slightly) higher than in Portugal, while they are significantly lower in most other countries.

The fact that part-time work has increased strongly among young workers is worrisome because of the high poverty risk that part-time workers face. It also raises concerns regarding young people’s ability to accumulate long-term savings early on and build pension entitlements. Long-term savings are essential to facilitate the resilience to adverse financial shocks and ultimately boost income levels at older ages. An increasing share of part-time work among young workers bears the risk of repercussions on their financial stability later in life and the pension levels they will have access to.

Part-time workers in Portugal are covered by the general social protection scheme, as are workers in standard employment. The social security contribution rate is identical at 34.75% for all types of employment contracts, be they part-time or full-time. To validate a calendar year of contributions, 120 working days are necessary. At least six hours of work are needed to validate one working day. Part-time workers working less than six hours per day are granted a number of working days that corresponds to a sixth of their total monthly working hours. For instance, a person working three hours per day and 22 days per month, i.e. 66 hours per month, will be granted 11 working days. Part-time workers working fewer than 120 days per year can aggregate the days worked in several years to validate one contributory year. However, days worked beyond the threshold of 120 days/year cannot be counted for another year (Chapter 3).3 While part-time workers are at high risk of poverty during their working lives, they are relatively well covered in terms of pensions, as the threshold of 120 days/year is low.

Figure 5.12. Part-time workers are at a much higher risk of poverty than full-time workers
Share of people at risk of poverty, by working status, 2016

Note: Risk of poverty is defined as working and having a disposable income below 60% of the national median equivalised disposable income (after social transfers).

Source: EU-SILC.


5.4.4. Informal economy workers are excluded from the social security system

While workers in non-standard employment are in general covered by the social security scheme, informal economy workers are excluded from earnings-related pension schemes, thus facing a higher risk of inadequate old-age income. The informal economy is a matter of concern in Portugal, as in many other OECD countries (Figure 5.13). Currently about 12% of workers are in the informal economy according to ILO estimates, which is lower than the EU average (14%), but considerably higher than in the best-performing countries such as Estonia (7%), Finland (6%) and Sweden (5%). Workers in the informal economy may also face very unstable professional circumstances and often experience long career breaks. Integrating informal workers in the formal sector is an important step towards attenuating old-age poverty risks and improving the financial long-run stability of pension systems.

Figure 5.13. Informal employment is below average, but is not a marginal phenomenon
Share of informal employment in total employment, 2016

Source: International Labour Office (ILO).


5.5. Automation and digitalisation will contribute to shaping future labour markets

Digitalisation and automation can change labour markets at a fast pace and will contribute to shaping future jobs. While there is a large uncertainty on how exactly the future of work will look like, it could be that these trends contribute to an increasing number of non-standard workers and that workers will be required to adapt their skills more often to remain competitive in the labour market.

5.5.1. Automation and digitalisation in Portugal

As of today, the Portuguese economy relies less on automated processes than most other OECD countries. In the manufacturing sector, firms in Portugal currently use 5.8 robots per 1 000 workers, against an OECD average of 13.8 (Figure 5.14). These figures suggest that regardless of further technological developments an increase in automation could occur if Portugal were to catch up with other OECD countries. In quickly ageing societies with a shrinking labour force, the prospects of automation seem to be particularly marked: Germany, Japan and Korea are currently the OECD countries where automation is the most pronounced (Figure 5.14). While automation provides ripe growth opportunities, it also raises concerns that finding employment might become more difficult for some groups of workers, including low-skilled and vulnerable workers.

Figure 5.14. Automation in the Portuguese industry is, for now, less pronounced than in other OECD countries
Number of robots per 1000 workers in the manufacturing sector, 2016

Source: World Robotics 2017, International Federation of Robotics.


Similarly, digitalisation is gaining pace. Currently the percentage of the population using the internet in Portugal is below the OECD average, at 70% against 83% (Figure 5.15). However, the gap was much bigger a decade ago when only 38% of people in Portugal accessed the internet while they were 59% in the OECD. This trend indicates that Portugal is catching up quickly with other countries. The strong increase in digitalisation can be expected to spill over to the business area and expand digital use at work.

Figure 5.15. Internet use is below average, but Portugal is catching up
Individuals using the internet (%), 2006 and 2016

Source: World Development Indicators, World Bank.


5.5.2. Changing labour markets raise questions for old-age protection systems

Future labour market prospects raise questions regarding the social protection of non-standard workers. Especially if digitalisation and automation lead to an increase in self-employment, temporary work, part-time work or platform work, ensuring that workers in these types of occupations benefit from a sufficient level of old-age insurance might be of paramount importance. Diverging rules regarding the pension insurance of non-standard workers exist in OECD countries (OECD, 2018[8]) and non-OECD countries (ILO, 2018[9]). In light of the fundamental labour market changes ahead of us, discussions on new social protection rules for non-standard workers have started in many countries.

While automation and digitalisation provide ripe job opportunities, including for people who struggle to find employment in the traditional labour market, they also bear risks. Companies may become reluctant to offer standard employment, aiming to reduce their social security contributions. If increasingly many firms outsource work tasks to workers in non-standard forms of employment, however, the effectiveness of pension systems might be undermined through a shrinking number of contributors. Fostering pension coverage among non-standard workers is thus crucial, not only to ensure pension adequacy among future retirees but also to guarantee the functioning of pay-as-you-go pension systems themselves.

While most forms of non-standard work such as self-employment and part-time work are not new and OECD countries have experience in dealing with these types of jobs in terms of social security, the situation of platform work is different. Platform work is a relatively new phenomenon. As of today, the number of platform workers is still very low in OECD countries, accounting for a small part of the labour force only. However, up until recently their number was growing very rapidly, at least in some countries, as the example of the United States points out (Katz and Krueger, 2016[10]).

In Portugal, non-standard jobs in general and platform work in particular have the potential of becoming widespread due to digitalisation for several reasons. First, wages in standard employment are low compared to other countries and incentives to engage in other forms of activities might be higher. Second, unemployment rates among young people are high, both as a remaining consequence of the crisis and due to structural factors, and non-standard forms of employment offer alternatives for job seekers struggling to find employment. Third, Portugal is a major tourist destination (OECD, 2018[11]). The high number of tourists may contribute to a sound demand for certain types of non-standard work, including platform work. These specific characteristics suggests that ensuring adequate pension insurance for non-standard workers could be even more important in Portugal than in other countries.

5.6. Policy options

Overall labour market conditions have improved remarkably since the Great Recession, with rapidly decreasing unemployment and higher labour market participation. According to ILO (2018[12]) job-creation and training policies have substantially contributed to the recovery after the crisis. Efforts should be continued to ensure that the Portuguese labour market offers today’s working-age population sufficient opportunities to pursue successful careers and prepare for retirement.

With fast population ageing ahead, policy action should favour longer working lives. This goal is achievable through a combination of policy measures in different areas. Standard early-retirement rules should be improved to reduce the number of early labour market exits and the minimum age of early retirement should be automatically linked to life expectancy gains, as the normal retirement age is.

Furthermore, sustainability factor rules should be reformed, as discussed in Chapter 3. The sustainability factor does not currently apply to retirement at the normal retirement age, which implies that early retirement triggers sweeping benefit reductions. Retiring early does not seem rational in most cases given these very strong penalties. This suggests that people who retire early despite these rules either do not understand the drastic consequences of their decision or have no other choice, for example due to bad health conditions. In both cases, the policy objective pursued by penalising early retirees so strongly is unclear. Furthermore, the sustainability factor does not apply to disability pensions, thereby creating financial incentives to exit employment through a disability spell. The sustainability factor should be recalibrated and apply to all pensions - including pensions granted at the official retirement age - as an ultimate instrument ensuring financial sustainability given all the other pension parameters. Then, the penalties for each month of early retirement should be increased as needed to deter early retirement in a close to actuarially neutral way.

Moreover, the labour market has to be sufficiently sound to absorb a larger number of older workers and these workers need to have the skills and ability to work longer. Appropriate health policies and lifelong learning possibilities for workers of all socio-economic groups and ages are essential. Over the past years, Portugal has made sizable efforts to provide workers with more training and upskilling possibilities. The Qualifíca adult learning programme, for instance, came into force in 2017 and aims to increase lifelong learning activities by offering specialised courses in a large number of centres throughout the country (OECD, 2018[4]). Such programmes should be continued and extended when appropriate while their efficiency should be assessed on a regular basis. This is particularly important in light of automation and digitalisation prospects that point to a future of work in which skill needs might change rapidly and workers will have to adapt their skills continuously to remain competitive in the labour market.

Policy measures should also ensure that unemployment schemes do not encourage early retirement. More than 50% of new early-retirement entries occurred through an unemployment-retirement transition in 2016. Rather than permitting older long-term unemployed to enter retirement very early with lower penalties, the unemployment benefit system should be adapted to offer good protection against unemployment while providing effective active labour market programmes to strengthen job-search efforts and employability before the retirement age. Such policies can consist of job-search assistance through intensive counselling interviews of high quality, including in the form of tailored guidance in job search and potentially mandatory interviews. All programmes should be monitored to evaluate their cost-efficiency (OECD, 2015[13]). In recent years, Portugal has taken steps to reinforce activation policies. For example, the way the eligibility criteria for unemployment benefits are applied was tightened, leading to some success in increasing unemployment exits (OECD, 2017[6])

Younger age-groups were hit hardest by the economic crisis and still struggle to recover, with persistently high rates of unemployment and a high frequency of part-time work and temporary positions. These difficulties undermine young people’s ability to accumulate pension entitlements and start savings early on, thereby increasing the risk of financial woes later in life. Improving the situation of today’s young is thus particularly important, also from a pension perspective. The high number of temporary contracts, especially among young workers, bears the risk that career breaks will become more common. Indeed, workers on temporary contracts do not benefit from the same stability as workers in standard employment and are particularly vulnerable in case of economic downturns.

The economic crisis led in particular to career interruptions and major difficulties to enter the labour market. The impact of short and interrupted careers on pension levels are fairly strong, close to the OECD average, but Portugal was hit harder than the average OECD country. In order to cushion the long-run effects of the crisis on pension entitlements of those who had to cope with such strong career shocks, policy makers should consider granting pension rights to people who were unable to find employment during the crisis, for example during the 2009-14 period. This would be an exceptional measure offsetting the negative pension consequences of exceptional employment difficulties encountered during the crisis.

Self-employment is more widespread than in other OECD countries. Self-employed workers in Portugal face specific regulation regarding their social security, including their pension insurance. The 2018 reform reduced the social security contribution rate for self-employed workers from 29.6% to 21.41%, against 34.75% for dependent employees. Except for unemployment insurance, social security including pension insurance for self-employed workers should in principle be aligned to the case of dependent employees. The current rules imply that self-employment is subsidised, which might be difficult to justify. However, the grey line separating capital from labour income for the self-employed makes this general principle difficult to implement fully in practice.

“False” self-employment (falsos recibos verdes), i.e. self-employed work that is de facto dependent work, is a matter of concern in Portugal. When self-employed workers depend significantly on one ordering customer, the latter pays parts of their social security contributions. This measure may help to discourage “false” self-employment to some extent – which low contribution rates for self-employment encourage - but is not a sufficient policy tool. In order to counteract the rise in “false” self-employment in recent years, labour inspectors have been given additional tools to detect abusive “false” self-employment cases and to regulate them. Due to limited resources, however, a systematic detection of such cases might not be guaranteed. The capacity of the labour inspectorate force should be strengthened further to improve and extend their monitoring capacities (OECD, 2017[6]).

Key recommendations

  • Modify the way the sustainability factor is applied as its current use overly penalises early retirement and provides incentives to exit the labour market through alternative routes (e.g. disability). Instead, use the sustainability factor to adjust pension benefits across the board as an ultimate instrument to ensure financial sustainability given the other pension parameters.

  • Restrict early retirement by linking the minimum retirement age to life expectancy gains and by increasing penalties per month of early retirement once the sustainability factor has been reshaped.

  • Eliminate the option for long-term unemployed people to enter retirement very early and without the full penalties applied to other early-retirement entries. Provide effective active labour market programmes to strengthen job-search efforts and employability of older workers in particular.

  • Consider granting pension rights to people who were unable to find employment during the crisis, e.g. for the period 2009-14, as an exceptional measure to cushion the long-run effects on pension entitlements of those who unluckily had to face such strong career shocks.

  • Avoid social protection rules that create distortions and encourage self-employment unless they are justified by a clear policy objective.


[12] ILO (2018), Decent Work in Portugal 2008-18, From crisis to recovery, (Geneva).

[9] ILO (2018), Innovative approaches for ensuring universal social protection for the future of work, (Geneva).

[10] Katz and Krueger (2016), “The Rise and Nature of Alternative Work Arrangements in the United States, 1995-2015”, NO 22667, NBER, Cambridge, MA, (accessed on 16 August 2018).

[7] OECD (2018), Good Jobs for All in a Changing World of Work: The OECD Jobs Strategy, OECD Publishing, Paris,

[4] OECD (2018), Key policies to promote longer working lives in Portugal, OECD 2018,

[11] OECD (2018), OECD Tourism Trends and Policies 2018, OECD Publishing, Paris,

[8] OECD (2018), The Future of Social Protection: What Works for Non-standard Workers?, OECD Publishing, Paris,

[5] OECD (2018), The Future of Social Protection: What works for non-standard workers? Policy Brief on the Future of Work, (accessed on 02 September 2018).

[6] OECD (2017), Labour Market Reforms in Portugal 2011-15: A Preliminary Assessment, OECD Publishing, Paris,

[1] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris,

[2] OECD (2017), Pensions at a Glance 2017: OECD and G20 Indicators, OECD Publishing, Paris, (accessed on 02 August 2018).

[3] OECD (2017), Preventing Ageing Unequally, OECD Publishing, Paris,

[13] OECD (2015), OECD Employment Outlook, OECD Publishing, Paris,

[14] OECD (2015), Pensions at a Glance 2015: OECD and G20 Indicators, OECD Publishing, Paris,


← 1. The proportion of older workers is particularly high in the agricultural sector.

← 2. The 2019 reform discussed in endnote 6 in Chapter 3 significantly lowers the estimates of the penalties for someone entering the labour market at age 20 and retiring early. However, the penalties presented in this section apply for anyone entering the labour market after age 20 since they do not fulfil the 40 years of contributions requirement at age 60.

← 3. Two workers working 60 days in 2017 and 180 days in 2018 (worker A) and 60 days in 2017 and 60 days in 2019 (worker B) will both be granted 1 contributory year even though worker A worked twice as much as worker B over the period 2017-18.

End of the section – Back to iLibrary publication page