Towards Better Social and Employment Security in Korea

Assessment and recommendations

A strong economy but a fragmented labour market

Korea has gone through remarkable economic development over the past 40 years. Korea’s export‑led, manufacturing‑driven growth strategy has yielded faster economic growth than virtually anywhere else in the world, rapidly diminishing the gap between Korea and the OECD average in GDP per capita terms. Parallel to its rapid economic transformation, Korea has witnessed a considerable social transformation towards becoming one of the world’s most highly educated societies alongside rapid population ageing as a result of low fertility and rising life expectancy.

The labour market, however, has not kept pace with this fast development. Many Koreans struggle with jobs of poor quality and low social protection. A strong focus on labour market flexibility has served larger companies and export-oriented industries well. However, such gains have been distributed unevenly between the individuals employed in such companies and those affiliated with them through outsourcing and subcontracting. Labour regulations and agreements are geared towards protecting permanent jobs but often fail to provide for those in less regular employment situations.

The Korean labour market has a number of characteristics that distinguish it from those in most other OECD countries: a) a high share of workers employed in small and micro-businesses, especially in low‑productivity services; b) a high share of self‑employment (more than 20% of the workforce); c) a short average survival rate of small businesses (with only around half of them surviving for more than four years); and d) short job tenure for the average worker (with one in three workers having less than one year of job tenure, and one in two among those in small firms – the highest share in the OECD).

These labour market characteristics are related to the deep segmentation in Korea between regular and non‑regular jobs and explain the high prevalence of labour market “outsiders” who have not benefited to the same degree from the country’s fast economic growth. This has led to considerable disparities among certain groups of non‑regular workers, own‑account workers and those not able to work long hours, especially sick or disabled people. In addition, female labour force participation is low and women are highly over‑represented in poorly‑paid non‑regular jobs, leaving Korea with one of the biggest gender wage and gender employment gaps in the developed world.

To fulfil Korea’s growth potential, these labour market problems will have to be addressed and both job quality and social protection improved. In particular, the situation of own‑account workers and employees of small businesses will have to be tackled as they face low wages, short job tenure and weak social protection. People who lose their regular job often become trapped in such forms of employment. Changing this situation will require concerted action by the government and social partners on various policy fronts some of which – such as the situation of redundant workers or that of older workers – have been addressed by other OECD reports.

This report looks in more depth at the policies and institutions that contribute to better and more widespread social protection, labour market inclusion and job quality in Korea. In particular it looks at the effectiveness of four government programmes:

• The Employment Insurance (EI) programme, which provides contributory unemployment benefits for eligible jobseekers who lose their job involuntarily;

• The Basic Livelihood Security Programme (BLSP), which provides means‑tested, non‑contributory social assistance for people living below the poverty line;

• The Employment Success Package Programme (ESPP), which is a form of employability support with a non‑contributory benefit component (means‑tested for some but not for others); and

• The Earned Income Tax Credit (EITC), which provides in‑work support for both salaried and non‑salaried workers who earn a low income.

In addition, the report looks at the situation of Koreans struggling with health problems and temporary work incapacity, a group that often falls between the cracks. This is an issue that receives too little attention in policy circles and in the public debate.

Social protection reforms have generated strong momentum

Headline labour market statistics in Korea highlight many positive trends. Employment rates are on an increasing path and have reached the OECD average of 66% of the working‑age population. The unemployment rate, at just over 3.5%, is among the lowest for an OECD country while long‑term unemployment – a big problem in many OECD countries – is virtually non‑existent. These positive outcomes, however, must be seen against the background of a system that provides relatively limited protection to those without a job. For many in Korea, unemployment entails no entitlement to income or employment support. Under such circumstances, jobseekers are compelled to accept any available job as quickly as possible, which contributes to the enduring existence of poor‑quality jobs and the persistent fragmentation of the labour market.

Survey data suggest that the share of workers in Korea who benefit from social support when they lose or choose to change a job is much lower than in other OECD countries – lower than 10%, compared with around 20% in Australia and the United Sates and over 30% in Canada, for example. Data also suggest that Korea’s social protection measures do much less to lift people out of relative income poverty. Moreover, the data show that people with health problems fare particularly badly. All of these findings are related directly to the limited accessibility and scope of public income support in Korea.

Korean policy makers have not been idle. Continuous reforms enacted during the past decade have sought to introduce both a broader safety net and more effective employment support for jobseekers while improving the effective coverage and enhancing the support available for both low‑income workers and jobseekers. These reforms have mostly focused on: a) offering EI to a wider group within the labour force; b) better customising BLSP entitlements to beneficiaries’ needs; c) introducing targeted ESPP for low‑income jobseekers and vulnerable groups; and d) introducing and gradually expanding in‑work support through EITC, the statutory minimum wage and other measures.

Reforms have been successful in offering better support to ever more Koreans: EI, BLSP, ESPP and EITC caseloads have increased continuously in the past decade. Nevertheless, the principal ongoing issue social and labour market programmes in Korea encounter are their limited take-up and coverage. Combined together, Korea’s four main working‑age benefits (EI, BLSP, ESPP and EITC) were received by almost 4 million individuals in 2015 – equivalent to around 10% of the working‑age population. While this represents important gains made, the overall number still appears somewhat low considering that more than 13 million Koreans of this age are inactive or unemployed while another 7 million are employed but earning less than a full-time statutory minimum wage.

Despite strong progress, therefore, each of the four main measures has further to go:

• EI eligibility criteria have expanded several times since the measure was introduced in 1995. In principle, EI now covers virtually all employees (including non‑regular and daily workers) on a compulsory basis while self‑employed persons (employers and own‑account workers) can choose to opt in. In practice, however, only 53% of the entire labour force in Korea is entitled to EI. The rest include three principal groups: self‑employed people who almost never choose to insure; a large group of workers who should be insured but go undocumented and miss out on coverage (especially among small enterprises); and those legally not entitled to support, including contributing family workers.

• BLSP today covers around 3.2% of the population, following major changes in 2015 to expand coverage and better customise payments to households’ needs. However, the family support obligation – a unique Korean feature that implies the earnings and assets of a claimant’s children and parents are also taken into account in determining entitlement – continues to exclude an estimated half of all potential beneficiaries.

• ESPP was introduced in 2009 to help jobseekers who are not entitled to EI and not receiving BLSP but facing considerable disadvantages, especially in the form of low income. The eligibility criteria have since been expanded to include not only low‑income jobseekers but other disadvantaged groups, elderly jobseekers earning below the median wage and youth. The fast-increasing number of ESPP participants currently encompasses about 300 000 per year although there is still potential to expand the programme to a multiple of this.

• EITC was introduced in 2008 to benefit low‑income workers and their families. Entitlement now covers regular and non‑regular workers as well as, since 2015, self‑employed persons. Currently about six in ten EITC recipients are non‑permanent employees while around 25% more are self‑employed. With a total number of over 900 000 recipients, EITC coverage may still be considered low relative to Korea’s 7 million workers earning below 50% of the average wage. EITC’s low income threshold and low take‑up rates, especially among the self‑employed, contribute to this shortcoming.

Improving social protection

Korea has made significant progress over the past 25 years to develop and expand its social safety net. In doing so, it took many of the experiences from other OECD countries on board and successfully avoided many of their mistakes. Most importantly, Korea managed to maintain strong work incentives while keeping dependence on social benefits low. This was achieved by introducing, as part of each new measure, a strong focus on activating jobseekers alongside relevant employment services. It was also achieved, to some extent, by setting benefit levels relatively low and keeping entitlement criteria strict. In combination with a low tax on income from labour, this means that work generally pays in Korea while waiting on a benefit is neither attractive nor easy to achieve.

Strong activation measures are in place for jobseekers entitled to EI; they have clear job‑search obligations and are closely monitored by Employment Centres. Under BLSP, the link between social transfers and activation has also been strengthened over the years and seems stronger in Korea than in many other OECD countries. BLSP recipients with work capacity (as assessed by the pension insurance authority on request of a local welfare office) receive a conditional payment and must get in touch with their local Employment Centre for case‑managed counselling including, if necessary, support to address financial and social problems and potential participation in ESPP.

Despite a strong activation framework and rising beneficiary rates, Korea’s approach can be improved. First, much of the spending on active labour market measures goes to direct job creation programmes. This leaves little funding for other measures. Direct job creation programmes were shown to do little to help jobseekers into private‑sector employment and have been downsized considerably or even abolished in most other OECD countries. Second, total spending on social benefits is low relative to other OECD countries. Despite continuous expansion of the welfare state by expanding programmes and introducing new ones, total public social spending in Korea is only half the level of the average OECD country and just one‑third of the level of most European countries including France and Italy. For those covered by the measures in place in Korea available support may be sufficient but many workers and jobseekers do not receive any support.

Korea is now at a critical crossroads. If it wants to strengthen its safety net and improve job quality for a larger part of the working‑age population, significant additional action will be needed. Critical choices will have to be made on what programmes or measures to expand without damaging the strong work incentives currently in place. Experience from many OECD countries has shown that it can be difficult to find the right balance between entitlements and obligations, especially for low‑income groups. Moreover, future projections suggest that public social spending will increase rapidly in the coming decades in line with Korea’s changing demographic realities and because the system will gradually mature. This makes it important that better outcomes are achieved for the monies currently spent and new action is funded in a sustainable way.

Continuing EI, BLSP, ESPP and EITC reforms

Korea must continue and maybe accelerate incremental reform of its labour market and social protection institutions to expand the reach of its social and employment support. Blind spots in social protection coverage exist, to some extent, in every OECD country as a result of the coverage conditions they impose. However, the impact of these blind spots on overall unemployment protection is greater in Korea than elsewhere since the excluded groups represent a much larger share of the total labour force.

Boosting EI coverage

Key issues for Korea’s EI include the low voluntary registration of self‑employed persons, low registration and effective coverage of those working in small businesses and the exclusion of contributing family workers and other groups of non‑regular workers.

Fuller EI coverage is important in Korea because: a) those who receive EI benefits are well supported in their job search by the Employment Centres; b) their efforts are closely monitored, with weekly counselling meetings; c) special support is offered for those who have low employability or lack motivation; and d) benefit deferrals are available for jobseekers in case of temporary work incapacity due to sickness. All this contributes to better job matches and better social and economic outcomes.

A number of OECD countries manage to extend employment insurance coverage to some of the workers currently legally omitted or effectively excluded in Korea. These countries’ examples light the way for broader coverage in Korea under EI. Workers who leave their jobs voluntarily, for example, gain unemployment benefit coverage in many OECD countries, though typically with a period of benefit suspension. Korea is relatively strict in this regard, disqualifying most jobseekers deemed to be voluntarily unemployed.

Self‑employed persons may opt in for EI coverage on a voluntary basis, though this approach has not resulted in significant coverage rates in Korea nor elsewhere. Several OECD countries therefore have mandatory EI registration for self‑employed workers, sometimes excluding those with very low income or employers with a certain number of employees. Greece and Slovenia are two countries that have switched from voluntary to mandatory coverage for self-employed persons. Mandatory coverage would be especially important for the many rarely insured dependent self‑employed workers in Korea, whose income is concentrated on one client and who are hardly different from salaried workers. Mandatory coverage for as many of the labour force as possible has the great advantage that more people would be brought in contact with the Employment Centre and offered employment services, thus, be helped into better jobs and better careers.

With further expansion of EI, attention should be paid to maintaining its high efficiency. Where broader EI coverage results in higher EI revenues, scaling up may be possible without a change in premium rates – provided the new groups face a similar average risk of unemployment. An expansion of the system to cover voluntary unemployment to some degree, however, might require a premium increase.

Enforcing EI regulations

A relatively high rate of undocumented work in Korea has resulted in low EI coverage among a group of employees who, by all rights, ought to be covered. Some 4 million employees were excluded from EI coverage in this way in 2016 – about 23% of all employees in the private sector aged below 65 – most of them in smaller enterprises.

The situation of workers in this “effective blind spot” can only be improved through better enforcement of EI regulations and stricter monitoring of employers’ responsibility to enrol their workers in social insurance. This may require stricter penalties for those who flout the rules. It may also require tightening the reporting requirements of employers for non‑regular workers and harmonising them across all types of workers.

Stricter enforcement of existing EI rules may also require boosting the resources of the tax authorities and the labour inspectorate to monitor enterprises to ensure all workers and labour costs are properly documented. Korea’s labour inspectorate currently has very limited resources to fulfil such a monitoring role. Employers are rarely sanctioned for not registering their workers while the penalties involved are too low to be considered a real deterrent. Recent research has shown that 80% of informal employees in Korea are accounted for by non‑compliance with existing labour standards and laws while only 20% can be attributed to actual exclusions observed within the law.

Maybe surprisingly, most workers in Korea are covered by Industrial Accident Compensation Insurance (IACI): More than 17.5 million workers are IACI‑insured while only 12.5 million are EI‑insured. The recent shift in responsibility for EI registration, as of January 2017, to the Korea Workers’ Compensation and Welfare Service – which is now responsible for both EI and IACI registrations – should help in expanding EI coverage to all those workers who should rightly be affiliated under both measures.

Stricter enforcement of EI legislation would not only secure fair coverage for those who should be entitled but also ensure fair competition among employers. Reneging on EI contributions by neglecting to document workers, after all, offers an unfair competitive advantage that can add pressure on otherwise lawful employers to do the same.

In this regard, more could be achieved through one key feature of EI (already in place since its inception in 1995) that enables undocumented workers to claim the entitlements they should have been due. Undocumented workers for whom EI contributions have not been made but who are otherwise entitled to its support (by having lost their job involuntarily; and fulfilled at least 180 days of work within an 18‑month period) can request a “confirmation of insured status” from their local Employment Centre. If their claim is assessed as valid, such jobseekers pay the EI and national pension contributions they should ordinarily have paid during their undocumented employment (for up to a maximum of three years) in exchange for entitlement to EI benefits. Under such circumstances, the employer (or employers) who hired but neglected to document such workers are mandated to pay the corresponding part of their EI and national pension contributions (also up to three years) plus a one‑off fine of KRW 30 000 per worker.

If promoted and applied more rigorously, this procedure has considerable potential to boost employers’ compliance with their formal obligations under EI. Currently, every year only up to about 1 000 workers benefit from this regulation. This number could increase very significantly if more workers knew of the procedure. The compliance of employers could also be strengthened if the cost they could encounter was higher: by, for example, mandating them to pay all of the unpaid insurance contributions (including the employees’ part) under such circumstances or imposing a more meaningful fine. Expanding EI entitlement to voluntarily jobseekers, as proposed above, would further increase the effectiveness of this procedure to punish wrongful employers.

Using EI funds more effectively

The Social Insurance Premium Subsidy Programme covers part of the EI contribution of low‑income workers in small companies and their employers. Such subsidies are funded through general government expenditure and paid on a permanent basis. They provide an added incentive for small employers to register their workers formally and comply with their social insurance obligations. The programme has contributed to some increase in EI coverage although the associated deadweight costs are large. Targeting the subsidy more closely and phasing it out after some time could improve the programme’s effectiveness. More rapid phase‑out periods could apply to higher earners or to larger enterprises.

The significant deadweight cost associated with the premium subsidy programme would become less of an issue if the confirmation‑of‑insured‑status procedure was strengthened under EI. The message to employers would then become very clear: registering workers for EI is obligatory and will be monitored closely; small companies and low‑paid workers will be supported through subsidies; but those who continue to neglect to document their workers will incur significant costs, ex post and irrespective of the reasons for job loss, through a swift, effective, well‑publicised arbitration process.

The Early Re‑employment Allowance (ERA) enables EI beneficiaries returning to work before their benefits are exhausted to receive a lump‑sum payment, calculated as a share of their remaining EI entitlement. Given the relatively low EI benefit payment duration and the near inexistence of long‑term unemployment in Korea, this measure is costly and not very effective as suggested by past evaluations. For instance, ERA is used mostly by men in their 30s and 40s who find new employment easily. It may be more effective to invest these funds into closer monitoring of small and medium‑sized enterprises to make sure they document workers and their wages and comply by EI rules. Abolition of ERA is currently being considered and a bill is pending at the national assembly.

There is also a need for the Korean government to act on the EI benefit structure and payment level. At the time of its introduction, EI payments replaced 50% of jobseekers’ previous wages, with minimum and maximum payment levels set at around 20% and 80% of the average wage, respectively. Over time, this margin has narrowed (because the floor is linked to the minimum wage and the ceiling changed discretionarily) turning EI into a virtually flat‑rate payment of KRW 46 584‑50 000 per day for all beneficiaries. This shift has changed the nature of the system. Compared with other OECD countries’ floor and ceiling amounts, the EI benefit floor in particular emerges as strikingly high.

Looking ahead, Korea has several options:

• To turn EI into a genuine flat‑rate payment: If fixed at 90% of the minimum wage, this would be the highest‑value flat‑rate unemployment benefit in the OECD. At 50% of the average wage, the benefit would be similar to the average payment in many countries with contributory unemployment benefit measures and still relatively high for low‑wage earners. If the planned minimum wage increase materialises, the payment would be higher than the benefit ceiling in most of the other OECD countries (relative to their average wage).

• To re‑establish EI’s original insurance rationale: This would require delinking the EI floor from the minimum wage and raising the floor and the ceiling in the future in line with the average wage. To re‑establish the situation back in 1995, the benefit floor would have to be lowered to around 33% of the current minimum wage and the ceiling set to roughly 150% of its current level.

• To introduce a stronger insurance element than is currently the case but keep the link between the EI benefit floor and the minimum wage: Such a solution could include one or several of the following elements: a) lowering the benefit floor back to 70% of the minimum wage, as was the case when EI was introduced, or lower than this; b) rising the formula replacement rate from its current 50% of previous gross earnings to something closer to 80%; and c) rising the benefit ceiling to a fixed multiple of the floor.

The government could also consider delivering EI benefits in five instalments per week as is done in Finland and Sweden, for example, instead of in seven as is currently the case. This would effectively spread the payments out over a longer period but offer EI more flexibility to maintain the current daily benefit minimum and maximum levels. This reduction could offset the large predicted increase in the benefit floor (in line with the forthcoming increase of the minimum wage) while increasing the potential duration EI can be claimed for – which is rather short by international standards – by an equivalent amount, thus potentially leaving premium rates unchanged.

It is a political choice which of these different routes to follow. Each way comes with different economic and social implications and might entail different degrees of public acceptance. For example, high benefit floors or high flat‑rate payments can weaken the work incentives of low‑income groups considerably. Genuine or effectively flat‑rate payments, like those made today, weaken the insurance principle and put in question the financing structure; such type of EI payment is normally better funded from taxes than insurance premiums. And increases in the benefit ceiling or the aspired replacement rate will most likely require a corresponding increase in the premium rate.

Expanding access to BLSP

Recent BLSP reform has addressed many of the measure’s previous weaknesses through smoother taper rates and the use of different income thresholds for the measure’s four main components (living benefit, medical benefit, housing benefit, education benefit). BLSP benefits are now better customised to individual needs; eligibility has been broadened; and work incentives have been improved.

Probably the biggest remaining bottleneck for broader coverage under BLSP is the family support obligation: a rule – specific to BLSP – by which not only the income of the applicant’s household is means-tested to determine eligibility for support but also that of any children and parents not living in the same household. This rule creates significant inequality and poverty as it does not depend on whether relatives actually provide any care or income support and is unfit for purpose in an increasingly individualistic society where values have changed. There is a strong case for further easing (and eventually phasing out) the family support obligation, as was done in other OECD countries that previously had comparable rules including Austria, Belgium or France; in Germany, a similar rule is still operated but only in cases where relatives actually provide regular payments. Abolishing the family support obligation could potentially double or more than double the BLSP caseload, according to some estimates. A further expansion of the caseload seems justified given the relatively low share of working‑age individuals who receive BLSP. Nevertheless, older people would probably benefit the most from a lowering and elimination of the family support obligation.

Any expansion of BLSP coverage should ensure to continue the strong activation of conditional claimants. To this effect, social, welfare, health and financial services could be better harmonised to address all labour market barriers. Ideally this should be done under the responsibility of the Employment Centres, to ensure that the necessary services reach beneficiaries with the highest needs. The current co‑location of various counselling services and dispatching of local welfare officers, who remain under local government control, is only a second‑best solution. It bears the risk for beneficiaries to be handed over from one service body to another, diluting the necessary focus on employment promotion.

Further expansion of BLSP eligibility conditions should bring more low‑income jobseekers with work capacity under the measure. Further monitoring and strengthening of job‑search requirements of these conditional BLSP recipients – by, for example, harmonising them with the behavioural requirements of EI recipients – would help make the entire process more efficient. It is important in this regard to enforce conditionality and job‑search requirements for all BLSP household members of working age, as foreseen in the legislation, not only the main applicant or household head.

Maximising the impact of activation under ESPP

ESPP is a well‑designed labour market programme targeting jobseekers not entitled to EI, especially those with low incomes, thus capturing some of those moving in and out of unemployment and low‑paid, precarious work. ESPP offers customised job‑search support and, if necessary, training. Intervention is structured in three phases lasting a total of 9‑12 months and unsuccessful jobseekers can reapply after 3‑30 months.

ESPP fills the large gap between EI and BLSP in a rather effective way, with a potential target group of, initially, close to 4 million people (all working‑age individuals with an income below 60% of the median) and possibly many more given that more recently some discretionary target groups (youth and certain vulnerable groups including lone parents and disabled people, among others) can participate without a means test. ESPP eligibility has been broadened continuously since its introduction in 2009 and the caseload has increased accordingly. Nevertheless, there is still considerable room for further expansion of, perhaps, 3‑5 times the current number of participants.

A significant further expansion of the programme will require a corresponding increase in the number of ESPP counsellors as well as continuous quality improvements. The latter will require: a) additional resources to reduce the current caseload of 100‑120 participants per counsellor nearer to the lower number in the first years of the measure; b) better training for counsellors; and c) more ambitious targets on employment outcomes for each Employment Centre, varied according to the local economic situation and the degree of participants’ disadvantage or distance from the labour market.

Strengthening private employment service provision under ESPP

The recent expansion of ESPP has encouraged many new private providers of employment services to emerge. While low‑income and disadvantaged groups are currently served by public Employment Centres, ESPP services for all other participants (including in particular youth) are subcontracted to private providers.

The measurement of provider performance could be broadened further to ensure only the most effective providers delivering a high‑quality service remain in place. Like in Australia’s Star‑Rating System, provider placements and outcomes could be made public with regression adjustments taking client characteristics and the state of the local labour market into account. Meanwhile, the short duration of private providers’ contracts should be reconsidered: at just one year in length, Korea’s short contracts are likely to hinder longer-term investment in specialised competencies among providers. Australia and the United Kingdom, for example, have shifted towards contracts for private providers lasting several years following such an experience.

Another question is whether the chosen split in participants between public Employment Centres and private service providers is efficient and can be maintained under any future expansion of ESPP. Australia and the United Kingdom, again, have chosen to outsource employment services irrespective of participants’ vulnerability or disadvantage as it is not clear public providers would necessarily serve them better nor in a more cost‑effective way. Both countries keep the monitoring of behavioural conditions and participation requirements exclusively under the control of public authorities, however. Under an expanding ESPP caseload, such an approach could also be worth considering for Korea.

Developing ESPP into an unemployment assistance measure

One option for ESPP would be to expand eligibility conditions and its payment level in such a way as to reach out in a more comprehensive way to jobseekers ineligible for both EI and BLSP benefits. One concrete proposal made along these lines by the Korea Labor Institute suggests a “Korean‑style unemployment assistance scheme” could reach twice ESPP’s current number of participants (jobseekers earning less than 60% of the median income) and provide a benefit worth 20% of the average wage for up to 12 months. This would certainly be one valuable way to expand ESPP. The current government has plans to take steps into this direction, with higher benefits in a first step provided to youth participating in ESPP and at a later stage, from 2020 onwards, also to other groups.

How widely ESPP can be expanded in terms of scope and coverage entirely depends on what additional public investments are willing to be made. Evidence suggests that ESPP significantly improves employment outcomes among all types of participants, including the most disadvantaged, and in a cost‑effective way. An expanded ESPP measure could eventually reach all jobseekers earning below 60% of median income, irrespective of their employment history, delivering to them income support, customised employment support and opportunities for training.

ESPP has a number of features in common with unemployment assistance measures found in other OECD countries, including its benefit component and means‑testing requirements. Nevertheless, unemployment assistance measures only exist in a minority of OECD countries and their characteristics vary considerably: some target new labour market entrants predominantly while others are oriented towards jobseekers that have exhausted their entitlement to contributory unemployment benefits. Similarly, benefits in some countries may be paid indefinitely but only for a limited period (typically up to one year) in others. The main aim of an expanded ESPP measure might be to bring effective employment services to more people in dire need of them, alongside a modest benefit component. The ESPP’s current design reflects this goal very well.

ESPP’s current income support is relatively low (about 10‑15% of the median wage, taking all payments into account) and its structure strongly geared towards promoting employability and lasting employment outcomes. This payment structure should be maintained if participation is further expanded in order to keep work incentives high. Raising the payment level, however, will be necessary both to increase the motivation of people to actively participate in the measure and to ensure participating households gain adequate income support. A fixed‑rate benefit worth 20% of median income for all participants could be a good starting point and would be in line with similar payments provided in other OECD countries – although the short duration of participation in ESPP could enable higher payments. Indeed, in no other OECD country is the gap between the minimum unemployment insurance benefit and the unemployment assistance benefit (where such benefit exists) as large as in Korea. This implies a strong case for Korea for lowering the EI benefit floor in exchange for higher ESPP payments.

Increasing support for in-work poverty

Better supporting low‑income jobseekers will be important for Korea but not sufficient. More can be done to boost support for Korea’s working poor, most of whom are own‑account or non‑regular workers and many of them in older age. Today, 60% of Korea’s working poor (defined as everyone earning less than 60% of the median wage) receive no public assistance at all and 80% are not covered by EI.

EITC was introduced in 2008 to provide some support for such workers in the form of a tax credit, with the credit initially varying by the number of children in a family. In 2014, a separate Child Tax Credit (CTC) was introduced and EITC modified to only vary by the number of earners in the family. In 2015, EITC was expanded to include self‑employed persons. The take‑up of EITC, however, is still low because it is closely targeted to the lowest-income families and phased out completely already at the level of 50% of average household income. Moreover, for those eligible the tax credit itself is also low, amounting to an average claim worth only about 2% of the average wage per year.

Take‑up of EITC should be promoted more actively among groups not sufficiently covered by other measures. Korea’s tax authorities already collect excellent records on this and may be well placed to more proactively enlist workers entitled to EITC but not claiming it. In addition, EITC take-up of low‑income households could also be increased through modified assets tests while smoother taper rates would maximise the work incentives of qualifying households.

EITC is not the only labour market institution geared at helping workers make ends meet. The Korean government has announced to increase the statutory minimum wage from its current KRW 51 760 per day, for 2017, to KRW 80 000 by 2020. The average daily wage for a regular salaried worker in Korea, by comparison, was KRW 92 100 in August 2016, according to the Economically Active Population Survey. Such a rapid increase in the minimum wage by 2020 would represent a compound average rise of 15.6% per year – probably fast outpacing the growth of Korea’s average wage and increasing the ratio between the two to above its current 50‑55%. Such a significant increase will bring a rapid pay rise to Korea’s formal low‑wage workers, with likely knock‑on effects for those in the informal sector, too, as other countries have experienced.

Although its benefits may be debated, a minimum wage increase of such magnitude will have important consequences on social protection measures further down the line. It is, therefore, important for policy makers to investigate and fully understand the direct interactions EI, BLSP and EITC have with the minimum wage.

If the increase in the minimum wage and further promotion and expansion of EITC fail to reduce in‑work poverty to the intended extent, BLSP could also provide a stronger top-up for low-income working households. Providing in-work benefits through a central social assistance measure of this kind is, indeed, a common practice in many OECD countries.

Helping sick workers stay in employment

Korea’s social protection system is especially weak for workers who encounter temporary work incapacity due to acute or chronic health problems. Korea is among a small minority of OECD countries without a statutory (or a coherent privately‑regulated) cash sickness benefit measure to help this group.

Limited support may reach sick workers in Korea, under two sets of circumstances. First, jobseekers who encounter sickness when they are already receiving EI benefits – and are thus prevented from looking for a new job – can continue receiving EI until the end of the regular entitlement period. Second, workers working in an EI‑affiliated job who are forced to quit because of a sickness can postpone their EI claim for up to a maximum period of four years until they are well enough to look for work. They are not entitled to any income support during this period but can delay the start of their claim period until after they have recovered. Entitlement criteria are relaxed somewhat for these workers who can qualify for EI when “it is difficult for them to perform their job due to lack of physical strength, mental and physical disability, illness or injury”.

Less than 20 000 people every year profit from these two regulations, however. The large majority of workers gain no support in Korea in case of sickness or temporary work incapacity. None will gain any income support for as long as they are in employment and those who want to retain their jobs have virtually no avenue of support open to them. Consequently, many may not regain their original job and face difficulties in entering a new one and, thus, often face long periods out of work. This is likely to have a negative impact on i) their wellbeing and recovery, further delaying their return to work and entrenching an overall loss of productivity, as well as ii) the labour market and the economy, as these workers and their capacities remain unused or underutilised.

The experience of many other OECD countries has demonstrated such workers find it much easier to return to their original positions (or take on another role with their original employer) than to gain an altogether new employment. Therefore, every effort should be made to improve job retention of workers experiencing health problems. Introducing a period of liability for sickness upon employers would go a considerable way in mitigating the chances of a return to work. Further introducing a statutory cash sickness benefit – either under an expanded mandate for EI or as a stand‑alone social protection measure – would increase the chances workers with health problems have of staying in work.

Employers in most OECD countries are liable for part of the risk associated with temporary work incapacity among their workers. In most countries, such obligations cover between a week and a month of absence or, in some cases, much longer (for example, up to 36 weeks in Italy and up to two years in the Netherlands). During this time, employers are obliged to pay all or part of their absent workers’ wages. Introducing such a mechanism in Korea could secure a minimum level of protection for all salaried workers. Regulating workers’ minimum entitlement to payments in case of sickness would end the large discretion employers currently have over this area, thus acting to reduce labour market duality. Such mechanism would also safeguard the jobs of hitherto disadvantaged workers who would otherwise have lost them under such circumstances.

Cash sickness benefits are provided in almost all OECD countries in addition to a period of employer liability. In most countries, such benefits are contributory (financed from premiums collected from both employers and employees) and form part of a common insurance fund: either a broader health insurance fund; a broader employment insurance fund; or as a stand-alone measure with its own individual fund. Though all three kinds of arrangements exist in OECD countries, most operate cash sickness benefits through their general health insurance plans.

For Korea, this approach could provide a more level playing field for all workers, given the high coverage rates already achieved under the existing health insurance measure. Nevertheless, perhaps the most straightforward option for Korea would be to link cash sickness benefits directly with EI, as Canada and Ireland do, and use the existing EI infrastructure to administer it. This would be a big advantage for salaried workers in Korea by providing them the support they need for their recovery and, thereafter, enabling them to return to their original jobs instead of having to re‑enter the labour market afresh. Such a solution, however, would make it even more important to broaden EI coverage to a larger share of the working population.

Under all scenarios it would be important that any new cash sickness benefit measure is coupled with a robust rehabilitation component, which includes tools to help recovering workers to return to work; clear protocols defining the rights of workers and employers as well as doctors and insurance authorities; and regular work capacity reassessments.

Korea’s progress on implementing smooth and effective social protection measures over the past 20 years has been exemplary. Starting from a very low base around the time of the Asian Economic Crisis, Korea’s social protection institutions now provide income and employment support to millions of households in need. What is more, Korea has successfully navigated many of the pitfalls other OECD countries have faced – not least in light of its highly robust approach to activation. Taking this powerful momentum forward, Korea is now very well placed to go further: addressing its remaining coverage gaps; strengthening the implementation of what is already legislated for; and designing effective new ways to address the many troubles faced by workers who become sick and, upon recovery, help ensure their timely return to work.

Key recommendations

The Korean government should consider taking the following policy actions:

• Expand EI coverage further by: a) seeking effective ways to ensure coverage of self‑employed workers, especially dependent self‑employed, possibly by making EI contributions mandatory for this group; and b) retaining EI entitlement for workers who leave their jobs voluntarily, including a benefit suspension for a suitable period instead of disqualifying them outright.

• Better enforce EI regulations by: a) expanding the resources and mandate of the relevant authorities to help ensure all workers are formally documented and offending employers sanctioned; b) promoting and rigorously applying the use of EI’s arbitration procedure through which initially undocumented workers can claim their EI entitlements retrospectively; and c) sharing information by the tax authority and the Korean Workers’ Compensation and Welfare Service to bring EI coverage in line with the higher coverage rates already achieved by IACI.

• In the light of a reconsideration of recent EI developments: a) decide on the future structure of payments – whether to return to earnings‑related EI benefits or instead turn to a purposefully flat‑rate measure; b) make the Social Insurance Premium Subsidy Program more targeted and temporary; and c) abolish the ineffective Early Re‑employment Allowance.

• Improve the situation of workers with health problems by: a) introducing a statutory employer liability of several weeks for workers’ sick pay; b) introducing a cash sickness benefit, perhaps integrated into a somewhat broader EI measure and implemented through its existing infrastructure; and c) matching the new benefit with a strong focus on rehabilitation and return to work.

• Maximise the impact of ESPP by: a) continuously increasing the number of people participating by promoting the programme among low‑income jobseekers and providing higher income support; b) increasing the number of ESPP counsellors and providing them with better training; and c) improving the performance measurement framework for local Employment Centres.

• Strengthen private ESPP service provision by: a) expanding services in line with increases in the ESPP caseload; b) possibly reconsidering the current split in responsibilities between private providers and public entities; and c) expanding the duration of contracts with these providers while strengthening guidance and performance assessment.

• Ease access to BLSP by: a) phasing out the family support obligation; b) strengthening the activation approach for conditional BLSP recipients; and c) better co‑ordinating and integrating social, welfare, health, financial and employment services, under the responsibility of the Employment Centres.

• Help the large group of working poor by: a) actively reaching out to people who should be eligible so as to expand EITC coverage to a wider range of poor people; b) investigating the implications of the planned increase of the minimum wage on labour market outcomes and on EI, EITC and BLSP; and c) if in‑work poverty remains high, promoting the use of BLSP as a top‑up to low income.