THE MINIMUM INCOME SCHEME REFORM IN LITHUANIA

The article reviews the current social assistance scheme in Lithuania and its recent reform. The aim of this paper is to evaluate the scheme in three dimensions: coverage of poor people, adequacy of benefit amounts, and social assistance reconciliation with work incentives. The analysis of the minimum income protection reform is based mainly on the data from Survey of Income and Living Conditions and Administrative Data from the Ministry of Social Security and Labour. The paper discovers the changing role of the minimum income scheme in the context of an economic cycle, its targeting to low-income strata of the population and an adequate amount at least for families with children. However, the growing incentives of local authorities to reallocate funds envisaged for social assistance to other needs can erode the scheme. The article ends with several recommendations to policy makers.


Introduction
Minimum income schemes provide cash benefits intended to guarantee a minimum level of support when other incomes (from the market, from other cash benefits or of other family members) are insufficient.The general term "minimum income scheme" covers a wide variety of schemes that have developed over time within the social protection systems established by countries of the economically developed word.This paper is analysing non-contributory means-tested social assistance schemes aimed at the provision of minimum income for people in poverty.
The need for an adequate minimum income has been acknowledged at the European level on numerous occasions.In 1992, the European Commission proposed a directive on a minimum income, which was then downgraded to a recommendation on common criteria concerning sufficient resources and social assistance in social protection systems (Council of the European Communities, 1992).The Commission gives the idea of a guaranteed minimum income a central place in its Recommendation of 3 October 2008 on the active inclusion of people excluded from the labour market (European Commission, 2008), in which it recommends that the Member States should draw up an integrated comprehensive strategy for the active inclusion of people excluded from the labour market, combining three elements: adequate income support, inclusive labour markets, and access to quality services.
The European Parliament went even further in its Resolution of 6 October 2010 on the role of minimum income in combating poverty and promoting an inclusive society in Europe, not only stating that "minimum income schemes should be embedded in a strategic approach towards social integration", but also adding that "adequate minimum income schemes must set minimum incomes at a level equivalent to at least 60% of median income in the Member State concerned" (European Parliament, 2010).
In Lithuania, the cash social assistance benefit (SAB) is paid to families and single residents unable to provide themselves with sufficient resources for living.SAB is means-tested and is paid if the value of property does not exceed the average property value set for the residential area and if the monthly income is below the level of the Sate-Supported Income (SSI) (i.e.LTL 350 or 101 Euro per family member per month).Readiness for work and training is taken into account.Refusal of job offer, training, or works supported by the Employment Fund may lead to the suspension of social assistance benefit.Until 2014, SAB was paid by municipalities from targeted subsidies allocated to them from the national budget (Piniginės socialinės paramos, 2003).
During the last several years, the reform of the social assistance benefit scheme set up some changes.
First, the work benefit was introduced.The benefit is granted during the six months if a person has started working after a long-term unemployment even if the income of the family exceeds the SSI.The benefit amount is 50% of the SAB paid before the employment.
Second, the waiting period was abolished for unemployed people.The requirement to be registered as an unemployed for at least six months was abolished, and the application for SAB is allowed immediately after the registration at the Labour Exchange.
Third, the introduction of the economy of scale differentiated the amount of SAB.It starts to decrease from the second member of the family.The monthly benefit level is 100% of the difference between the actual income of a family or a single resident and the State Supported Income of LTL350 or €101 per person per month for the first family member, 80% for the second member, 70% for the third and any additional family member.For case examples of benefit amounts, see Table 1.
Fourth, the SAB amount for long-term unemployed was reduced.SAB will be gradually reduced by 20% after three years of payment, by 30% after four years of payment, by 40% after six years of payment for families having children; benefit payment has to be abolished for person a without children.
Fifth, workfare was introduced.Recipients of SAB must take part in "Socially useful activities" organised by the municipal administration (40 hours per months).Avoiding participation in socially useful activities may cause cancelling of SAB payment.
Sixth, there is a decentralisation of SAB financing.Municipalities provide SAB as a municipal function from their budgets.They have received a strong incentive to spend less for SAB and reallocate money for other purposes (Piniginės socialinės paramos, 2003).
Seventh, involvement of stakeholders was foreseen.In providing cash social assistance, the municipalities have the right to engage the representatives of local communities.
To sum it up, the first two points of social assistance reform were intended to relax from entitlement to SAB.Others are mainly focused on a stronger control of applicants and restrictions of benefit payment.In the following chapters, we shall analyse the coverage and adequacy of SAB and its impact on incentives to work.
In this paper, we focus our analysis mainly on three issues of the minimum income scheme reform in Lithuania.They are: coverage of poor people, adequacy of benefit amounts, and a social assistance scheme linked to the active labour market inclusion policy.

Coverage of minimum income safety net
The coverage of cash social assistance benefit in Lithuania has been undergoing significant changes in the recent years.Before the onset of the crisis, in 2007-2008, around 37,000 inhabitants (1.2% of total population) were receiving these benefits (Fig. 1).During the crisis, in 2010-2012, the number shot up to 180,000-220,000 (6-7.3% of the total population and about one third of people at risk of poverty according to the 60% poverty threshold).These are the average monthly numbers of benefit receivers, calculated on the basis of each person receiving the benefit a full year.In real life, of course, some persons receive benefits for just several months per year; therefore, the number of those who received benefit at least once per year is at least twice as high.
No reliable information exists on what prevents some persons (those having legal grounds to reside in Lithuania and having ID documents) from accessing social assistance benefits.It should be noted that immigrants (migrants, asylum-seekers, undocumented migrants) in Lithuania are quite few.Homeless persons with IDs on their hands can apply to receive social assistance benefit.They don't need to have a residence, but have to check in as claimants at the municipality.There is no research as to what extent the access to the social assistance benefit system is limited by other factors (lack of information, complexity of system, conditionality, discretionary nature of benefits, lack of support on application, fear of stigma).However, data from the Survey of Income and Living Conditions (SILC) provide an opportunity to indirectly assess the accuracy of the social assistance benefit system.According to SILC, in 2011, 404,000 persons (13.5% of the population) received these benefits at least for one month (Fig. 2).
Around 30 per cent of them (124,000) earn a monthly annual average income (LTL 255 in 2011, Table 2) which is below the benefit qualification criteria, the State Supported Income (SSI), which has not changed since 2008 and equals LTL 350 per capita per month.
The other 70 per cent (around 280,000) have been receiving social benefit even though their income is above the SSI (a monthly annual average income is LTL 702 per capita per month).One of the reasons may be the undeclared income of claimants.The second reason is related to different periods of average income calculation for cash social assistance benefit application and in the SILC.The average income for eligibility of social assistance benefit is calculated on the basis of a three-month period.Consequently, some people do receive social assistance benefit for several months, despite their relatively high income during the next months and despite their average monthly income per year being higher as the qualification criteria -SSI.
Around 84,000 persons (2.8%) whose income is below the SSI (LTL 350) do not receive social benefits (Fig. 2), even though they meet the income criteria to qualify with an average monthly income of LTL 235.There may be several reasons for that: failing to satisfy the assets criterion, or being able to work but not satisfying employment criteria.For some of such persons, the required paperwork associated with applying for and getting social benefits may be too complex, or they may not have the required documents or information on their right to claim for social benefits, or they may simply avoid the social benefit system for fear of stigma.
The take-up of SAB depends a lot on its administration.Local authorities are responsible for SAB payments, and they can act more friendly to clients or less so.Statistical data on SAB recipients show a strong impact of local administration on the number of assistance beneficiaries.Municipalities do apply the same law on social assistance, and they use the same criteria for processing client applications.Nevertheless, disparities in the number of beneficiaries among municipalities are huge.Even if demographic, economic, and labour market differences are taken into account, the number of beneficiaries differs up to twofold (Fig. 3) -a solid proof for a very high dependence of SAB take-up on the administrative culture of municipalities.
Having in mind the high variation of coverage even in such a highly centralised scheme as social assistance, recent reforms may put the minimum income protection in the country at risk.A major factor of the social assistance scheme reform is fiscal  decentralisation.Municipalities have been granted power to reallocate financial resources of social assistance: resources that are not utilised for social assistance cash benefits can be used for any other purpose at the discretion of local authorities.This new legal basis for the utilisation of financial resources was implemented in the five so-called pilot municipalities (Akmenė, Panevėžys district municipality, Radviliškis, Raseiniai, Šilalė) on 1 January 2012.Since 1 January 2014, the reform has extended to include the remaining 55 municipalities of the country.
The reform was sold as the decentralisation of decision-making and as a fight against misuse of social assistance by dishonest receivers.The pretext for this reform was the growth of beneficiaries' number in the recent years.Indeed, as Fig. 3.1 shows, during 2011 the number of beneficiaries shot up fivefold as compared with the pre-crisis years 2007-2008.However, the increase was due to important reasons.
These were as follows: a rapid rise in unemployment (from 4 per cent in 2007 to 18 per cent in 2010), wage cuts in the private sector, as well as the government's drive towards a greater fiscal consolidation.The public sector wages were cut, pensions and unemployment benefits were slashed while the categorical family benefit was redefined as income-tested and the number of beneficiaries decreased dramatically (Fig. 4).On the other hand, during the peak of the pre-crisis economic boom, the government had significantly boosted the State-Supported Income that acts as a threshold to qualify   for social benefits.This threshold has not changed as the household income decreased during the crisis.Social assistance benefits became the basic measure of income protection during the crisis (Fig. 4).Data for the period 2008-2013 demonstrate the effectiveness of the existing social safety net.Social assistance benefits guarantee the minimum income for the majority of persons in need during difficult times, and the number of beneficiaries is decreasing in periods of economic growth.
Still, as it has been mentioned above, a social assistance scheme reform was launched in 2012 in five municipalities in order to decrease the number of assistance benefit receivers.As a consequence, the number of receivers and the amount spent on assistance benefits fell markedly.Special commissions were set up to assess the right of applicants to receive assistance benefits.Commission members were not elected.Instead, they were invited to participate by local municipality administration.This made such commissions an additional obstacle to overcome for assistance applicants, thus introducing stricter application and assessment rules in order to meet municipality administration's expectations to cut the number of benefit receivers rather than to find persons in need of such assistance.
Since the social assistance benefit reform was launched in the beginning of 2012, five pilot municipalities have started saving money.The number of recipients per 1000 municipality inhabitants and expenditure for SAB per municipality inhabitant has decreased dramatically.The same trend has been observed in the other municipalities that joined the reform since January 2014.During the first two months of 2014 alone, all municipalities reduced expenditures on social assistance by 20 per cent as compared to the first two months of 2013.The Minister of Social Security and Labour Mrs. Algimanta Pabedinskienė assessed this fact as a better targeting (Lietuvos Rytas, 2014).However, saving money is neither the main priority of the social assistance scheme nor the main function of a municipality.
To sum it up, Lithuania's social assistance system worked well during the crisis and fiscal consolidation period.The assistance coverage has increased to include a far greater number of impoverished inhabitants.Unfortunately, as the demand for funds went up, the government decided to provide strong financial incentives for local governments to reduce the assistance coverage.The reform was started in 2012-2014 when the economic growth returned and the unemployment fell.The social assistance coverage is declining rapidly thanks to both the improving economic situation and the social support system reform.It is difficult to determine what proportion of persons is losing social assistance due to their improved financial situation, the reduced misuse of social assistance, and the tightened social assistance benefit rules.The reform requires fine-tuning in order to highlight these entangled factors and to move its emphasis from deterrence from applying for support benefits to their take-up.

Adequacy of social assistance cash benefit
From the very beginning of the 1990s, social assistance benefits (SAB) in Lithuania were related to the State Supported Income (SSI) which, respectively, was based on the Reference Budget (RB).However, since 1993, the RB has reduced dramatically.The number of food items reduced from 52 to 12, while the share of food products in the basket increased from 45 per cent to 80 per cent1 and the SSI lost its real value.Updates of reduced RB and the SSI for cash social assistance were implemented yearly until 2008; after, these updates were discontinued, and the RB lost its application.Nowadays, SSI is set up on the basis of purely political grounds.
The amount of assistance benefits varies according to the income of a single resident or of each family member.The monthly benefit level is 100% of the difference between the actual income of a family or a single resident and the State Supported Income (SSI) of LTL 350 (€101) per person per month for the first family member (including the cases where SAB is granted only to children), 80% for the second member, and 70% for the third and any additional family member.
The adequacy of the SAB amount varies greatly depending on family size.The maximum SAB amount for a single person has been very low during the last ten years, and now it equals just a half of at risk of poverty rate (Table 3).To a large extent this could be explained by a negative attitude towards single supported grown-ups held by politicians and society at large.A grown-up person is expected to earn his or her living or having already earned his / her right to a certain kind of social security coverage (pension, unemployment, sickness or maternity benefit).Compared to most OECD member countries, the net social assistance benefit for a single person in Lithuania is very small and makes up just 23% of median income (Fig. 5).The support to families with children (both lone-parent and two-parent families) is considerably higher, amounting to 50-60% of median income, which in this respect places Lithuania among the most generous countries (Figs. 6 and 7) due to generous equivalence scales applied in the country (80-70%).Therefore, each additional family member receives a benefit which is just a little less than that of the first member.The State Supported Income, which is used to determine the size of the social assistance benefits, was raised substantially in 2005-2008 (from LTL 135 to LTL 350 for a single person or a family's first member) but later on, as mentioned above, it, was not raised.As we can see, the size of social assistance benefit has several shortcomings.First, it lacks a procedure to be reviewed and adapted to the changing economic conditions and is not inflation-indexed.The economic crisis, which struck Lithuania in 2009, pushed down the annual consumer price inflation from 8 per cent in 2008 to just 3 per cent in 2010-2012 (Table 4).This produced a positive impact on low-income inhabitants since they are least protected from rising prices.In Lithuania, not only the SSI and consequently social assistance benefits, but also the minimum monthly wage, pensions, and social security benefits are not indexed.Instead, they are usually hiked by one-off government decisions when the state budget and the social security fund boost their revenue, most often during an economic upturn.During a crisis, the government is not capable of increasing its social benefit spending due to a lower budget revenue.Decisions of the government depend on the political cycle as well.The last time social benefits were raised back in 2008 (while the minimal wage was hiked as of the beginning of 2013).Consequently, even a lower inflation rate after a crisis peak hurts the real income of social benefit receivers and makes paid services less accessible.
The equivalence scales are too wide, which makes the support benefit difference too big for persons of a differing family status.It creates deep poverty and inactivity traps for families with children, whereas the support for single persons is inadequate.
As it has been shown, social support benefits are more generous to families but rather small to single persons.In both cases, they do not exceed the at-risk-of-poverty thresholds and encapsulate the recipients in the zone at risk poverty.The only thing we can measure is to what extent the benefits help alleviate poverty, i.e. increase income of those in poverty.
Social assistance benefits make up 8% of total household income in Lithuania; an average monthly social assistance benefit per household is LTL 309.Because the average size of a socially supported household is 4.1 persons, an average monthly social assistance benefit per capita is LTL 75 (Table 5).
The average monthly income of a social assistance benefit receiver is LTL 565, while the average social assistance benefit is LTL 75, thus amounting to 13% of a social assistance benefit receiver's total monthly income.It may seem odd that the average monthly income of a social assistance benefit receiver is LTL 565 -much more than the threshold to qualify for social assistance benefits.A person would not qualify if his or her monthly income would be LTL 565 each month of the year; however, the income may vary a lot from month to month, and so social assistance benefits are allocated only during those months when other income is below LTL 350.
A major part (75%) of funds used for the social assistance benefit are allocated to the first three population income deciles.Their income is below the at-risk-of-poverty threshold.The social assistance benefit makes up almost half of per-capita household income in the first decile, 20% in the second decile, and around 15% in the third decile.This means that the social assistance benefit is an important source of income for 30% of the most impoverished population.
The fact that almost 75% of all social assistance benefit receivers fall into the first four deciles lends credence to the social assistance benefit scheme's accuracy.Persons in these four deciles receive up to LTL 650 in monthly per capita income, including the social assistance benefit (this makes up around 85% of the at-risk-of-poverty threshold for a single person).All this shows that the accuracy of the social assistance benefit scheme in terms of social assistance spending allocation is relatively high -most of the social assistance benefit funds actually reach impoverished households.However, the social assistance benefit also fails to help such households to break away from the poverty zone, and in this respect the social assistance benefit is not effective.There are other two shortcomings of social assistance benefit amount: • it lacks a procedure to be reviewed and adapted to changing economic conditions and is not inflation-indexed; ad hoc decisions of the government depend on the political cycle; • the equivalence scales are too wide, which makes the support benefit difference too big for persons of a differing family status; it creates deep poverty and inactivity traps for families with children, whereas the support for single persons is inadequate.

The link between minimum income and the active inclusion policy
The minimum income scheme in Lithuania does not encourage the inclusion of unemployed persons into the labour market due to low wages, especially those paid for unskilled work.The minimum net wage is usually below the at-risk-of-poverty threshold for a single person (Table 6).The low-wage trap is most likely to occur at relatively low wage levels due to the fact that the withdrawal of social transfers (mainly social assistance, in-work benefits, and housing benefits), which are usually available only to persons with a low income, adds to the marginal rate of income taxes and social security contributions. 2he low-wage trap describes a situation where employed persons refrain from working more hours or taking better-paid jobs because this extra income would be too low.This trap shows the tax and benefit losses that specific household categories (a single person, a lone parent with a child, a two-parent family of four, etc.) suffer when wages rise from 33% of the average gross wage to 67%.In other words, the indicator shows the percentage of gross income lost after paying the income tax, social security contribution, and losing social assistance benefits when his or her wage rises from 33% of the average gross wage to 67%.A high low-wage trap reading workers lose a substantial part of their income after taxes and upon losing the social assistance benefit.This undermines the motivation to work more hours or seek better-paid jobs.
Viewed in the EU context, such losses suffered by a single person in Lithuania are not very high.However, for a household of one working person and one unemployed and two dependent children, the low-wage trap is quite deep (Table 7), because a single person can only hope for a meagre social assistance benefit, while the equivalence scales produce more generous social support for bigger families.The unemployment (poverty) trap3 in Lithuania is below the EU average (Table 8).A high unemployment trap value usually indicates that upon starting on a job, the disposable income undergoes minor changes, which deters from working and encourages living on the social assistance benefit.The unemployment (poverty) trap value rose before 2009, while afterwards it has registered a visible decline that could be attributed to the social assistance benefit not being indexed and cuts in the unemployment social insurance benefit.The contributory unemployment benefit scheme is fragmented and does not provide sufficient income protection for the unemployed.Only 15-17% of the unemployed receive contributory unemployment benefits (Lazutka, R., Poviliunas, A., 2013).The low coverage of the contributory unemployment scheme and the low amount of the unemployment benefit during a period of high unemployment are the major factors that have contributed to an increase in the number of recipients of social assistance.
The benefit is paid for 6-9 months depending on the length of insurance records.The amount of benefit decreases after three months of payment and has a ceiling.Due to the public finance crisis, in 2009 the Government decided to reduce the ceiling for the unemployment benefit amount.It was capped at LTL 650 per month (which means it is below the per-capita level of at-risk-of-poverty threshold).
Low wage and unemployment traps may be reduced by increasing social assistance receivers' income during the working period and decreasing the means-testing social benefit amount.As presented in Chapter 1, two measures of this kind were included into the reform package.First, Lithuanian authorities have introduced a reduction of the social assistance benefit as a work-incentive measure for long-term receivers.Since 1 January 2012, the social assistance benefit amount does depend on benefit duration.Social benefit is reduced for those beneficiaries who are entitled to social benefit for a long time: by 20% if social benefit is paid for 36-48 months, by 30% for 48-60 months, by 40% for more than 60 months (in this case, social benefit is not paid for beneficiaries without children).Second, social assistance benefit is now paid to former long-term (12 or more months) receivers if she / he has started working and earns between 100 and 200 per cents of the minimum wage.The benefit amount is 50% of the full benefit before starting work.The duration of payment is six months.It may activate long-term unemployed beneficiaries.
Moreover, the regulatory framework for social assistance in Lithuania provides sanctions against claimants.All social assistance applicants able work have to register as unemployed with local labour exchange offices.If they fail to accept to job offers, training course or active labour market programmes, they would be removed from the list of unemployed with the right to apply again to the labour exchange agency after six months.Application for social assistance benefit is available only after the renewal of registration in the labour exchange agency.Besides this requirement to accept job offers, training course or active labour market programmes, Lithuania has a typical workfare programme (i.e.work for a benefit programme).Its official title is Socially Useful Activities.The principal objective of this programme is working in return for social assistance benefits (participants receive only social assistance benefits before Socially Useful Activities but not any remuneration for this activity (work)).The target group is social assistance benefit receivers who have been ong-term unemployed l (for more than six months).Participation in this programme is mandatory for those assigned to the programme by a municipality.The municipality has the right to set the participation duration for each participant, but usually it is 40 hours per month.
Municipalities design and govern local workfare programmes at their own discretion; their aim is twofold.First, municipalities use manpower of social assistance beneficiaries for some simple works (cleaning of squares and streets).Second, municipalities seek for self-selection of formally unemployed social assistance applicants.Unemployed persons who officially are unemployed but participate in the shadow economy are unable to combine their participation in a workfare programme and undeclared labour activities.This workfare programme is seen as a very strong factor for decreasing the number of social assistance receivers.

Conclusion and recommendations
In the recent years, municipalities have focused their efforts on targeted social assistance with the support of local communities and NGOs helping to fight abuse.It is now necessary to more actively seek out and find impoverished persons who genuinely need social assistance but who are unable to apply, put their documents in order or overcome stigma.
It seems reasonable to reinstall the centralised funding of social assistance schemes in order to weaken the incentives by local authorities to reallocate funds envisaged for social assistance to other needs that are not directly related to poverty reduction.Centralised funding would also help cut regional disparities in the provision of social assistance and ensure equal rights to social assistance across the country.
Raising minimal income requires harnessing other social policy measures.Instead of solely relying on the social assistance benefit, it needs to be used in harmony with the minimum wage, social security and categorical benefit system.Developing a procedure for providing the basis for the minimum wage size and consistent review (indexation) would be a very important step forward.Today, it is reviewed ad hoc and depends on the political cycle.
The absence of effective unemployment insurance puts too much burden on the social assistance benefit system that caters for low-income population.The application of unemployment benefits must be expanded.Currently, only a small portion of the unemployed qualify, while the benefit amount itself is inadequately small and its duration is too short.The recent findings of the Employment and Social Developments in Europe 2013 indicate the positive impact of social benefits on the likelihood of getting back into employment "contrary to commonly held beliefs that people receiving unemployment benefits are more likely to get a job than people not receiving benefits" (Employment and Social Developments in Europe, 2013).
Similarly, the absence of a universal support system for families with children leads to a situation where having children in a family becomes a strong poverty risk factor.When only one of the two parents is employed, the likelihood of the need for the social assistance benefit to support such a family increases.
Even the size of the social assistance is an issue.The current equivalence scales for the minimum income scheme penalize single beneficiaries and prioritize families.This creates low-wage and unemployment traps for parents.The equivalence scales for the minimum income scheme should be redesigned in order to reduce the low-wage and unemployment traps for individuals with children, while the scheme itself should be complemented by a universal child allowance.The issue of indexing the social assistance benefit should be addressed immediately.Otherwise, the minimal income protection measures will fail to maintain their value due to inflation.

FIG. 3 .
FIG. 3. The number of recipients of social assistance benefits per 100 of vulnerable families* (right scale) and average monthly wage (left scale) in municipalities Source: Socialinės paramos šeimai informacinė sistema.The internet link: www.socmin.lt/index.php?-2063405871 ; Author's calculations.* Note: the number of beneficiares in each municipality ir devided by the sum of three types of families: families of unemployed, single families, and families with thre and more children.** Dotted bars are municipalities of five big cities.
FIG. 4. The number of recipients of various benefits (in thousands)Source: Lithuanian statistics.The internet link: http://www.stat.gov.lt/

TABLE 2 . Low-income households covered and not covered by social assistance Low-income households (average monthly income up to 350 LTL) Number of people Disposable income per member of household (LTL)
Source: Survey of Income and Living Conditinons, 2012.The internet link: http://www.stat.gov.lt/Author's calculations.

Income levels provided by cash minimum-income benefits for a lone parent with two children. Net income value in % of median household incomes, 2011 and 2005
Source: Income adequacy.Reliant on minimum income benefits2012, 2011, 2010, 2007, 2005.OECD.The internet link: http://www.oecd.org/els/benefitsandwagesstatistics.htm
* Equivalence scales are not applied in calculations, which are based on the assumption of the equal size of all households of different decile (4.1 members of a household).

TABLE 6 . The at-risk-of-poverty threshold and the minimum monthly wage (after taxes)
Lithuanian Statistics.The internet link: http://www.stat.gov.lt/ .Author's calculations. Source: