69011
            Social Impact and Social Mitigation of Enterprise Restructuring:
                               Implications for Croatia
                                      Policy Note-
                                                               Paula F. Lytle, ECSSD

Background on Enterprise Restructuring and Privatization

There has been growing recognition within the World Bank of the need to design
interventions directed at enterprise restructuring and privatization1 with greater attention
to the social impacts of those interventions and to the design of possible social mitigation
measures. A principal instrument for enterprise restructuring2, structural adjustment
lending, has been revised under the new Operational Policy 8.60 for development policy
lending to include a determination of the likely poverty and social impact of the
operation.3 Country Assistance Evaluations have indicated the need for attention to the
distributional impact of private sector development activities and to linkages among the
various reforms;4 weak institutions and the lack of comprehensive, country-specific
private sector development strategies have hampered development effectiveness in this
area.5

Despite relative agreement on recommending privatization of state-owned enterprises
(SOEs), policy advice to transition countries has differed on the method, speed and
sequencing.6 Competition and the relative strengths of the various stakeholders
(government, ministry officials, managers, workers, and unions) have tended to influence
the design and implementation of privatization programs.7 Within ECA, the experience
with the political economy of privatization has been characterized as “reveal[ing] a


1
  This policy note is focused on how social impact and social mitigation measures have been integrated
into enterprise restructuring. The complete debate on enterprise restructuring and privatization is well
beyond the scope of this paper. See, John Nellis, “The World Bank, Privatization and Enterprise Reform in
Transition: A Retrospective Analysis� OED, 2002 and also OED, Economies in Transition.
2
  Some enterprise restructuring has occurred in the context of investment lending, but the adjustment
lending constitutes the primary use. OED’s Project Performance Assessment Report on the Poland
Privatization and Restructuring Loan, for example, notes that programmatic adjustments loans are a better
instrument to address these types of projects with the need to maintain momentum of the privatization
process.
3
  See OP 8.60, paragraph 10. During the corporate review process, a decision should be made on whether
the policies supported by the DPL are likely to have significant poverty and social impacts. The Program
Document should summarize the analytic knowledge about those impacts and the borrowers’ systems for
mitigation of negative impacts and enhancement of positive impacts. See also OPCS, Good Practice Notes
for Development Policy Lending #2: Using Poverty and Social Impact Analysis to Support Development
Policy Operations, October 2004.
4
  CAEs have also noted the necessity for institutional development through technical assistance loans
supporting adjustment. See for example, the recent Romania CAE (May 2005) observes that as the World
Bank became concerned with the mode and quality of privatization, institutional development became an
explicit objective and adjustment loans were accompanied by parallel TA focusing on institutional
development.
5
  OED, Private Sector Development: Findings and Lessons from Selected Studies: #17.
6
  OED, Assistance to Transition Economies: Were there Alternatives?
7
  From Plan to Market: World Development Report 1996, p. 53


                                                    1
severe and politically charged tension with promoting efficiency and rewarding existing
stakeholders.�8

ECA project documents and CASs often cast the rationale for privatization in terms of
potential efficiency gains to the economy and the extent of direct subsidies from
governmental budgets for state-owned enterprises. The first stage of transition focused on
the establishment of market-driven, private sector-oriented economies. In recent years,
the requirements of European integration and accession have led to an emphasis on fiscal
discipline. Closer attention has been paid to implicit subsidies such as budgetary credits,
nonpayment of taxes and accumulation of tax arrears, transfers from off-budget funds,
and government payment of guaranteed loans. EU candidate countries have faced similar
policy challenges in mitigating the impact of unemployment due to structural shifts in the
economy.9

Background and Implications for Croatia

This policy note provides an overview of social impact and social mitigation measures
have been addressed in World Bank- supported enterprise restructuring10 programs in
ECA.11 The focus is on lessons learned from the experience elsewhere in the region that
could provide guidance to and policy options for Croatia in the on-going process of
enterprise restructuring.

In the past few years, WB evaluation and policy guidance to the GoC has stressed the
need to tackle subsidies to state-owned enterprises. The 2001 Country Assistance
Evaluation noted that the targets for reduction of the role of the state were relatively
modest,12 and the 2003 Country Economic Memorandum emphasized that the
privatization strategy pursued in 1990s has not facilitated restructuring. Policies of
implicit subsidies (see above) have delayed SOE restructuring, discouraging new
enterprises (particularly small and medium), expansion of existing firms, and sustainable
job creation.13 More recently, the 2004 Country Assistance Evaluation characterized
privatization of SOEs as problematic. Under the new Programmatic Adjustment Loan

8
  From Plan to Market: World Development Report 1996, p. 143.
9
   Bernard Funck and Lodovico Pizzati, eds. Labor, Employment and Social Policies in the EU Enlargement
Process, World Bank, 2002. See Johannes Linn’s introduction on page 19 in which he notes that “But as
one permits or encourages the downsizing of old industries, the social costs of transformation must be
appropriately addressed (as, for example, in the current programs of coal sector restructuring in Central and
Eastern Europe, supported by the World Bank) and an effort must be made to explain the approach to the
affected population groups.�
10
    Restructuring refers to the cost cutting, efficiency- and competitiveness-enhancing steps required to
position a firm to survive and thrive in a market environment. Privatization is a principal means of
achieving restructuring. Privatization, however, does not guarantee that restructuring will occur, but is more
strongly associated with enterprise restructuring than other methods. Djankov and. (World Bank, 2000.)
11
   While the primary focus will be on the ECA experience, selected experiences from other regions are also
included. Existing reviews of the labor impacts of enterprise restructuring were also consulted.
12
   OED, Croatia: Country Assistance Evaluation, February 2001, p.5-6. The CAE notes that modest targets
“suggest either excessive timidity on the part of the Bank, continued lack of ownership by the government
or the difficult circumstances in the region.�
13
   Croatia, Country Economic Memorandum, p. 5. and vol. 2, p.129.


                                                      2
(PAL) program, the GoC has committed to privatization or liquidation of 1,000 SOEs by
the end of 2007. This is consistent with the diagnosis of the needs for enterprise
restructuring and privatization in the Pre-Accession Assistance Needs (PAAN) approved
in July 2004. 14 As of April 2005, there are still over 100 SOEs with majority state-
ownership. With respect to the former agricultural and food SOEs, a small number (13)
of the most difficult cases remain. Half of these are agrokombinats with a large number
of employees constituting a significant proportion of employment in the area.15

Overall, regulation of the Croatian labor market has tended to maintain existing jobs
rather than create new ones.16 High unemployment is due to strict employment protection
legislation. 17As with large SOEs elsewhere, SOEs in Croatia can be characterized as
having too many employees with rigid labor contracts and often wages and benefits
higher than those in the private sector. The restrictiveness in the labor market caused by
employment protection legislation is reinforced by rigidities due to collective bargaining
agreements in SOEs. Labor regulations require multiple steps for dismissals, such as
cancellation of the contract and a lengthy, legally defined period of notice. Both wages
and the unemployment rate have risen in Croatia, additional manifestations of rigidities in
the labor market.18 For example, Hrvatske Zeljeznice (Croatian Railways) has been
characterized by low productivity due to a staffing surplus that had been increasing
relative to the volume of freight and passenger rail traffic (which was decreasing).
Furthermore, collective wage agreements with unions had raised labor costs by 6% as of
March 2005.

Protecting the existing employees in Croatia has its costs. Only 43% of the working age
population is employed; high unemployment (14.3%), nearly double that of the EU and
there is low labor force participation (50.2%) Although poverty is relatively low, the
main economic causes of income poverty are limited employment opportunities and
inadequate targeting of the social safety net.19

Social welfare legislation has been subject to frequent amendment, adding various
categories of beneficiaries.20 This has resulted in considerable expenditures on non-
insurance social transfers such as social assistance, maternity, children, war veterans and
invalids. Although the overall level of social protection spending is high, poor targeting

14
    As of July 2004, there were142 companies with majority state ownership, 163 companies in which the
state’s share is between 25% and 50%, and 740 companies in which the state holds a stake of 25% or less.
15
   Jos Hayes, “A Strategic Framwork for Improving Competitiveness in the Agricultural and Food Sectors
in Croatia,� paper commissioned under World Bank technical assistance loan.
16
   Predrag Bejaković and Viktor Gotovac, “Employment and Unemployment in the Republic of Croatia,� in
Croatian Accession to the European Union: Economic and Legal Challenges, Institute for Public Finance,
2004. See also Jan Rutkowski, “Does Strict Employment Protection Discourage Job Creation; Evidence
from Croatia, Policy Research Working Paper 3104, August 2003.
17
   Croatia Country Economic Memorandum, July 2003, p. 13.
18
   Predrag Bejaković and Viktor Gotovac, “Employment and Unemployment in the Republic of Croatia.�
Employment protection legislation is stricter in Croatia than in any other EU accession country. Croatia
CEM, p.13.
19
   Croatia, Country Assistance Strategy, November 2004.
20
   Predrag Bejaković and Viktor Gotovac, “Employment and Unemployment in the Republic of Croatia,�
p.217


                                                   3
and administration of benefits means that the most vulnerable members of society are not
necessarily protected, including those affected by enterprise restructuring. Cash benefit
expenditures are much higher in Croatia than in other countries at the similar income
level, and benefits are also managed by different ministries with categorical benefits
handled by the Ministry of Intergenerational Solidarity, Families and Veterans, (MHSW)
and social benefits and services managed by the Ministry of Health and Social Welfare.
Under the PAL program, the GoC has agreed to convene an inter-ministerial working
group21 to prepare and oversee the implementation of a strategy for reform of social
benefits spending. The goal of the strategy, agreed in the PAL negotiations, will be to
reduce the overall level of social benefits spending while increasing the share of the
means-tested social benefits.

Taken together, these various policy challenges in Croatia indicate the need for careful
design and targeting of social mitigation measures as the enterprise restructuring
proceeds. The remaining enterprise restructuring will require retrenchment, and this
process may face more obstacles due to the delays in implementation. Although targeting
of social benefits is being addressed, more vulnerable unemployed workers may not have
not adequate support. As outlined below, experience with World Bank-supported
enterprise restructuring indicates that measurements of social impact and social impact
analysis can provide inputs to this process.

Measuring Social Impact

In ECA, recognition of the implication of social impact on reform progress occurred
when the first wave of reforms encountered obstacles in implementation of well-designed
programs. OED characterized the First and Second Industrial Restructuring Projects in
Hungary as adopting an innovative approach to systemic reform, but continued along the
following lines:

        However, some of these measures (e.g., adequate bankruptcy procedures), although in place, were
        never enforced, due to the absence of an appropriate framework to absorb the social cost of
        adjustment. At the time of the design of the ISAL, although the social costs of the program were
        anticipated, no provision was made in the loan to mitigate the adverse impacts of bankruptcy and
        closures on the population. This may have contributed to the reluctance by the authorities, at the
        time, to accelerate the bankruptcy proceedings. Issues concerning the social cost of adjustment and
        other inter-related fiscal matters were subsequently addressed under the SALs which followed.



Expected social impact has at times been cast in terms of opposition to privatization,
rather than measured concretely, and social impact analysis has been missing or
incomplete.22 Organized labor and employees of state-owned enterprises (SOEs) have
posed the main opposition to privatization due to fears of job loss and of changes in
21
   The inter-ministerial working groups is comprised of representatives from the Ministry of Health and
Social Welfare; Ministry of Families, Veterans and Intergenerational Solidarity; Ministry of Economy,
Labor and Entrepreneurship and Ministry of Finance.
22
   The ILO recommends early identification of stakeholders and establishment of institutional arrangements
for their involvement. See B. Martin, “The Social and Employment Consequences of Privatization in
Transition Economies: Evidence and Guidelines, IPPRED Working Paper #4, ILO, 2000.


                                                    4
working conditions and terms of employment.23 Employment in privatized firms
generally decreases, at least in the short-term.24 Fears are often fueled by inaccurate or
incomplete information about social impact,25 and proponents of labor often are highly
critical of privatization, focusing on the negative effects on employment, income
distribution, and worker welfare.26 Labor opposition can create high political costs,
leading to governmental reluctance or slowing the speed of privatization.27 Labor unions
tend to prefer that SOE workforce issues be addressed by governmental programs, as
unions can have greater impact on the government, including on severance terms.28
Another factor that may influence the timing of restructuring is that private buyers prefer
or require the government to address retrenchment before privatization, so as minimize
labor disputes. Government-led labor restructuring, however, tends to lead to great
attention to development of social mitigation measures.29

In certain cases, despite an initial period of confrontation, unions and workers have
served as sources of industry-specific knowledge that can provide input to the
restructuring process, identifying possible areas for internal reorganization and skill
shortages with the industry. In Russia coal restructuring, technical assistance facilitated
the role of trade unions in managing social impact, including conducting seminars,
maintaining legal services in the coal regions, and monitoring.

The form of privatization most widely used in Croatia, management/employee buyouts,30
is also the method likely to have the least impact on employment.31 In Croatia, even with
the slow pace of privatization, accelerated inflows into unemployment have occurred due
to enterprise restructuring.32 Although privatization and restructuring is generally
expected to increase employment opportunities in the longer-term, short-term job losses
are typical.

23
   OECD, Privatizing State-Owned Enterprises, p. 16. Kikeri, “Privatization and Labor.�
24
   Megginson and Netter, “ From State to Market�
25
   Sunita Kikeri, “Privatization and Labor: What Happens to Workers When Governments Divest� World
Bank Technical Paper No. 396. 1998.
26
   Birdsall and Nellis, Winners and Losers, p. 51.
27
   Sunita Kikeri, “Privatization and Labor: What Happens to Workers When Governments Divest� World
Bank Technical Paper No. 396. 1998, and Yi Chen, “The World Bank and the Provision of Assistance to
Redundant Workers,� Social Protection Discussion Paper Series #112.
28
   Birdsall and Nellis, Winners and Losers, p. 52.
29
   Sunita Kikeri, “Privatization and Labor: What Happens to Workers When Governments Divest,� p. 13.
30
   This form of privatization has been common in the Yugoslav successor states, due in part to the pre-
transition structure. The discussion in Slovenia: From Yugoslavia to EU Accession weighs the positive and
negative legacy of self-management. On the one hand, managers had some responsibility for running firms
in a quasi-market environment; on the other hand, the strong role of internal ownership (by managers,
workers and pensioners) hindered efficient restructuring. The gradualist approach adopted by Slovenia
aimed at “making the costs of transition socially bearable.� Other countries that used management-
employee buyouts are: Poland and Romania, and the voucher program in Lithuania tended to function in
that manner. This form of insider ownership tends not to be stable and evolves into another form. Insider
buy-outs pose other problems (asset stripping, illicit use of assets) as detailed in the 2004 Croatia Country
Assistance Evaluation, p.19-20.
31
   Sanjeev Gupta, Christian Schiller, Henry Ma, and Erwin Tiongson, “Privatization, Labor and Social
Safety Nets,� Journal of Economic Surveys, vol. 15, No. 5, p. 655. N.B. Authors are from the IMF.
32
   Croatia, Country Economic Memorandum, vol. 1, p. 12


                                                      5
Privatization in heavily subsidized SOEs that have been protected from competition (such
as steel, railways and energy enterprises) has been more likely to require significant
reductions in staffing levels. Of the various sectors33 in which such significant enterprise
restructuring has occurred, the mining sector has focused consistently on measuring
social impact and developing social mitigation measures. In ECA, mining was heavily
subsidized and has undertaken significant restructuring. Poland, Romania, Russia and
Ukraine have all borrowed from the World Bank to finance this process; all have
analyzed social impact in some form, using this analysis in design, implementation and/or
monitoring of mitigation measures.34 Furthermore, labor activism is a common response
to coal sector restructuring, and the isolated, mono-industrial dimensions of mining
results in significant community impact and second-order effects due to job loss.35
Although Croatia does not face mine restructuring issues, the SOEs remaining to be
restructured share some of the characteristics: a high level of subsidies, activist labor
(shipyards), and protection from competition.

Until recently, systematic research has been lacking, however, on the dimensions of job
loss, the consequences for remaining employees, and implications for dismissed
workers.36 Pay levels, terms of employment, and job security for workers who remain at
privatized enterprises have not been addressed or have been assumed to stay constant.37
In addition, many workers who are formally employed at a SOE are on reduced work
schedules, forced leave, or have accumulated pay arrears.38 Often the extent to which
such workers have found alternative employment is not fully known. Distributional
impacts of reforms have been analyzed through Poverty and Social Impact Analyses
(PSIA)39 carried out in the context of programmatic poverty assessments and through
projects.

As part of the Privatization Social Support Project in Turkey, three studies were
undertaken to monitor the social impact of privatization: 1) a privatization economic

33
   Social impact analysis of power sector reforms has focused on consumers and the impact of increased
tariffs on households and the poor. See Julian Lampietti, ed. Power’s Promise: Electricity Reforms in
Eastern Europe and Central Asia. World Bank, 2004.
34
   Haney and Shkaratan, p.17.
35
   Michael Haney and Maria Shkraratan, Mine Closure and Its Impact on the Community: Five Years after
Mine Closure in Romania, Russia and Ukraine. Social Development Paper #42, June 2003.
36
   Kikeri, p. 7. Birdsall and Nellis observe that the effects of privatization on employment levels have
received surprising little rigorous analysis. Winners and Losers: Assessing the Distributional Impact of
Privatization, Center for Global Development, Working Paper #6 (May 2002), p. 21.
37
   See the literature review in the Annex to Birdsall and Nellis, Winners and Losers.
38
   C. Evans-Clock and A. Samorodov, “The Employment Impact of Privatization and Enterprise
Restructuring in Selected Transition Economies� ILO- Working Paper IPPRED-16., 1998. The ILO study
found that these workplace practices can contribute to growing poverty among the employed.
39
   “Poverty and social impact analysis (PSIA) refers to analysis of the distributional impact of policy
reforms on the well-being or welfare of different stakeholder groups, with a particular focus on the poor
and vulnerable. PSIA also addresses sustainability and the risks to policy reform and help to monitor
poverty and social outcomes and impacts of policy changes.� User’s Guide for Poverty and Social Impact
Analysis. See for example, the BiH “Poverty and the Labor Market� 2005 as part of the Programmatic
Poverty Assessment for BiH. A number of PSIAs in ECA have focused on the impact of utility reforms.


                                                    6
impact study, 2) a privatization coping strategies study, and 3) a labor redeployment
services study. The first consisted of three rounds of surveys, comparing the economic
status of workers in firms that have been privatized and in firms to be privatized. Workers
were asked about their uses of severance payments and the impact of privatization on
their family.40 In the Macedonia Social Support project (more below), social monitoring
surveys were financed under the project. Similarly, impact evaluations of targeted social
mitigation programs were conducted to assess the effects of measures adopted in the
Romania Mine Closure and Social Mitigation project and to contribute to the design of
the follow-on project. Ex-post social impact monitoring was utilized in Ukraine to assess
the impact of mine closures and downsizing and to evaluate the effectiveness of
mitigation policies.41

As described above, tracer studies tracking retrenched workers can provide key inputs to
the design of social mitigation measures. Retrenchments usually have an uneven impact
on men and women due to differences in types of jobs, contracts and seniority.42 In some
cases, there have been gender differences in the percentage of retrenched workers in self-
employment43 which would affect the type of retraining or job search assistance to be
provided. Similarly, lower-skilled and lower-educated workers have been found to have
greater difficulty in finding jobs.

Social Mitigation Measures

Globally, the amount of World Bank lending providing labor assistance to workers
redundant due to enterprise restructuring doubled between 1994 and 1998. The most
frequently used have been 1) training, 2) labor market mobility and 3) severance pay
programs.44 In ECA, the mining sector, in particular, has been a focus for operations that
have used the monitoring of social impact to assist with the design of social mitigation
measures. OED’s evaluation of the experience with extractive industries45 has been
reflected in the design of subsequent operations in that labor redeployment was
recommended to be the focus of a separate operation.

Macedonia has addressed the various elements of SOE restructuring with the support of
a number of operations over time (see footnote 49); the Social Support project46 was
linked to the Financial Sector Adjustment Loan 2 with a component designed mitigate
social risks from SOE restructuring. The project also provided assistance to long-term
40
   A majority (67%) of workers who remained employed at privatized firms considered privatization as
having had a positive impact on their company.
41
   Haney and Shkratan, Mine Closure.
42
   Susan Razzaz, “Minimizing Negative Distributional Impacts of Public Sector Downsizing� PREMnote
#84, April 2004. This note focuses on public sector downsizing, but some of the findings and
recommendations are applicable to enterprise restructuring.
43
   Studies of retrenched workers from SOEs in Zambia indicate that women are more likely to be self-
employed than men. The PSIA of mining sector reform in Romania found greater impact on women of
downsizing and less willingness to work in the informal economy.
44
   Yi Chen, “The World Bank and the Provision of Assistance to Redundant Workers: Experience with
Enterprise Restructuring and Future Directions�
45
   OED, Extractive Industries and Sustainable Development, Volume II- World Bank Experience, p.36.
46
   ICR rating- Highly Satisfactory.


                                                  7
unemployed in addition to those affected by retrenchments, an element that could be
applied in Croatia, given the presence of long-term unemployed. OED’s findings on
earlier operations in Turkey’s SOE restructuring program stressed the necessity for
realistic timelines in privatization and stressed the importance of assessing the political
risks early in the project cycle, including possible sources of resistance.

As in Macedonia, related operations addressed the various aspects of restructuring in
Serbia, but these were facilitated through adjustment lending, supplemented by technical
assistance grants and loans. Serbia’s Private and Financial Sector Adjustment Credits 1
(May 2002) 47 and 2 (June 2003) to privatize and restructure socially-owned enterprises
were complemented by the Social Sector Adjustment Credit (April 2003) focusing on
reforms in pensions and labor, while enhancing the focus on social assistance and poverty
monitoring.

Summarizing common factors in the range of ECA experience with managing the social
impact of enterprise restructuring indicates the following issues to be addressed in
designing mitigation measures:

     1. Determining the extent of back wages/ wage arrears.
     2. What has been promised in terms of severance payments (including statutory
        obligations)
     3. Loss of non-wage benefits provided by enterprises (social assets and services) and
        identification of who has taken over these services

As to the specifics of program, social mitigation responses vary along the continuum
between adapting existing institutions and policies to address restructuring to
development of ad-hoc or situation specific programs. The use of measures in
combination, as in Macedonia, has been found to increase effectiveness.48 Options
utilized are:
     ï‚· Employment services (job search assistance, placement, counseling)
     ï‚· Training, retraining and on-the-job training

47
   The first loan was rated highly satisfactory by OED, and the second loan was rated highly satisfactory
during the QAG quality at entry review. The second loan was a continuation of the reform program begun
under the first. The project was considered outstanding for building directly on the progress of first
PFSAC, pursuing the same objectives to further closure and taking advantage of the institutional foundation
laid by PFSAC I. The measures in PFSAC2 streamlined entry of new companies, contributing to the
increase of employment at SMEs at the same time as SOEs were being restructured.
48
   See David Fretwell, “Mitigating the Social Impact of Privatization and Enterprise Restructuring� Social
Protection Discussion Paper #405, March 2004. A study by the ILO found that use of measures in
combination tended to increase their effectiveness. N. Meager and C. Evans, Evaluation of Active Labor
Market Measures for Long-Term Unemployed. Within ECA, the provision of social services through
enterprises also raised additional issues for restructuring. The policy implications of social service
provision has diminished for Croatia as the enterprises have in general ceased to supply such services,
whether or not they have been restructured. Simon Commander and Mark Schankerman in “Enterprise
Restructuring and Social Benefits� (EBRD) examine Russian and Ukranian experience in this regard. They
conclude that effective reform must delink provision of benefits from incumbency in a firm, therefore
reconfiguring access to benefits. Access to certain benefits (health and child care) could be addressed
through provision of transitional income support which would decline over time as wages adjust.


                                                    8
     ï‚·   Job creation (temporary employment, public service/public works, short-term
         employment guarantees, wage/employment subsidies)
     ï‚·   Small business advisory services/business incubator /micro-enterprise
         development
     ï‚·   Income support (severance payments, preferential allocation of shares)
     ï‚·   Social welfare/ social benefits/ unemployment benefits (altering to address
         anticipated retrenchment, improving targeting)

Country factors such as a rapidly growing economy, well-established private sector, and
the presence (or absence) of social safety nets have influenced the choice and
effectiveness of social mitigation measures.49 In ECA, an evaluation of active labor
market programs in four countries found variation in impact within and between
countries. Different categories of workers have tended to benefit from different programs
with training programs having slightly higher positive impact on women and younger and
middle-aged workers. Less-educated workers tend to benefit from short-term transition
employment. Participants in wage-subsidy programs have tended to younger, less-
educated, and often long-term unemployed, but the impact of such programs is not
uniformly positive.50 An evaluation of a larger group of public works programs shows
that they can serve as short-term safety nets, but there is little evidence that such
programs improve long-term employment prospects.51

For example, in the Macedonia Social Support project,52 various options were used in
combination with components addressing severance payments, labor redeployment
(employment counseling and retraining), and improved targeting of social benefits. In
addition, the project specified evaluation of the impact of the labor redeployment services
and supported capacity-building for the Ministry to undertake monitoring. For Croatia,
this is a key lesson learned from this project, as small business advisory programs, micro
enterprise development, and business incubator programs have been undertaken, some of
them donor-supported, but impact monitoring has not occurred systematically.

Of the various social mitigation measures available, caution should be particularly
applied in the use of severance packages to induce workers to leave SOEs voluntarily.
Traditional severance packages tend not to be related to losses that workers experience,

49
   Sunita Kikeri, “Privatization and Labor: What Happens to Workers When Governments Divest?� World
Bank Technical Paper No. 396, 1998.
50
   D. Fretwell, J. Benus, and C. O’Leary, “Evaluating the Impact of Active Labor Programs: Results of
Cross Country Studies in Europe and Central Asia,� Social Protection Discussion Paper #9915, 1999. The
impact analysis used a quasi-experimental design to examine programs in Czech Republic, Poland,
Hungary and Turkey designed to expedite redeployment of labor.
51
   Gordon Betcherman, Karina Olivas, and Amit Dar, “Impacts of Active Labor Market Programs: New
Evidence from Evaluations with Particular Attention to Developing and Transition Countries� Social
Protection Discussion Paper Series #402, January 2004
52
   The objective of the Macedonia Social Support project was to mitigate the negative social and economic
impact from SOE restructuring occurring under the Financial and Enterprise Sector Adjustment Loan.
Earlier projects, specifically the Social Reform Technical Assistance project supported the government in
SOE restructuring through actions to reduce the individual and community impacts of worker dislocation.
(OED rated the project as likely to be sustainable and have substantial ID impact.) Currently under
implementation, the Macedonia Social Protection Project addresses pensions and targeting of cash benefits.


                                                    9
but are usually calculated on wage and seniority, which tend to disadvantage women.
Workers at a particular industry or enterprise may seek to be treated as a special case and
request higher compensation.53 Social impact analysis has found that severance
payments have tended to go into consumption, and without parallel policies to encourage
labor market re-entry, workers tend to seek social protection after funds from the
severance packages have been used. Another set of issues with severance packages in
SOE restructuring are similar to the risks faced by their use in public sector reforms:54 1)
adverse selection, 2) overpayment, and 3) moral hazard.55 Within ECA, severance
payments have ranged from three months pay to 24 months. A related concern with
respect on valuation occurs if SOEs have “ghost� or “fictitious� workers- workers who
remain formally as employees to gain certain benefits (including pension), but have
found employment elsewhere, usually in the informal economy. Options to address some
of the concerns with the use of severance packages would be to spread these over time or
to re-cast them as entitlements that are conditional on entering into some new productive
activity.

Although take-up on self-employment, business incubators and micro-enterprise
development services is usually quite low, they can be effective when well-targeted. In
particular, older better-educated workers tend to benefit more from these programs. It is
likely that labor retrenchment will affect this social group (as these are the workers
currently protected under existing labor legislation), and stock-taking on the existing
programs 56could be combined with impact evaluation.

In ECA more generally, the choice and design of operations has depended on the stage of
restructuring and the specific nature of the challenges. Sequencing has emerged as key in
developing social mitigation measures. Labor redeployment programs should be in place
prior to any large-scale lay-offs.57 The selection of implementing agencies is crucial in

53
   See for example, the discussion in Jolanta Hess “Recommendation and Proposed Strategy for Trepca
Restructuring� Kosovo Sectoral Employment Strategy, June 2004.
54
   This includes severance pay during retrenchment in public enterprises either wholly or majority owned
by the state. See the Operational Memorandum on Financing Severance Pay in Public Sector Reform.
55
   Adverse selection occurs when severance packages encourage well-performing staff to leave, and moral
hazard refers to workers who have received severance packages being rehired by the same enterprise or
entity. The Operational Memorandum notes that in the case of public enterprises being restructured for
privatization, the rehiring of previously retrenched workers is unlikely. The Operational Memorandum on
financing severance pay in public sector reform specifically states the necessity of addressing social impact
and the use of analytic work in designing mitigation measures. Paragraph 9 states as follows “Social
Impact. Public sector reform operations financing severance pay address the social costs of retrenchment.
The design of such operations should take into account the effect of layoffs on groups that may suffer
disproportionately and be rendered vulnerable as a result of layoffs, and, where appropriate, ensure that
mitigation measures are adopted as part of, or in parallel with, the operation. Appropriate analytic work to
identify and assess the potential social costs of retrenchment should inform the design of operations
financing severance pay and any relevant mitigation measures. Operations will generally contain change
management strategies to ensure that affected public sector staff are made aware of what is happening and
are informed about the options available to them.�
56
   The Territorial Socio-Economic and Institutional Assessment prepared for the Croatia Social and
Economic Recovery Project provides some overview of programs in the Areas of Special State Concern.
57
   Chen, “World Bank and the Provision of Assistance to Redundant Workers: Experience with Enterprise
Restructuring and Future Direction.�


                                                     10
ensuring that fast acting and targeted support reach vulnerable unemployed workers. In
Hungary, the rapid response model (based on North American experience) provided
services to workers prior to retrenchment, including both information and planning
services.58

Recommendations and Next Steps

Measuring Social Impact and Facilitating Strategy

Croatia’s experience indicates that social impact analysis would be appropriate in
determining mitigation measures. Evaluation of active labor market programs underline
the necessity of disaggregating affected workers prior to designing new measures. Issues
to be examined would be: worker age, gender, education, skill profile, wage arrears,
hours worked, secondary job holding, and other coping mechanisms. Data sources for
this analysis could include the Labor Force Survey, Business Environment Enterprise
Survey (BEEPS), and official employment and wage data. Surveys of enterprises,
including perhaps the BEEPS could aid in determining the skills in demand and the
extent of mismatch between retrenched workers’ skills and those needed in the labor
market. In addition, a tracer study may prove useful in determining mitigation measures
through identifying the outcomes for those who have already left SOE employment in
earlier retrenchments. The Regional Development and Living Standards study will also
provide inputs for this process.

Findings from social impact analyses in other contexts indicate that “restructuring� is
often viewed by workers and affected community members as a synonymous with
enterprise closure,59 indicating one of the areas to be addressed is that of improved
dialogue. Such dialogue may be time-consuming, but ultimately efficacious.
Implementation of the Slovenia Enterprise and Financial Sector Adjustment loan was
affected by the slow pace of legislative reform, but OED concluded that consensus-
building (and the time necessary to achieve it) was a key to such fundamental reforms.
Significantly, the Slovene experience generated a broad public debate and awareness on
the necessity to address issues such as wage growth, pension reform, and privatization of
SOEs.60 Experience in Croatia indicates that union concern in railway restructuring, for

58
   Gary Hansen and Maria Heidkamp, “Innovative Approaches to Worker and Community Adjustment in
Hungary� in Regional Development and Employment Policy. Findings indicate that it takes nearly half the
time for workers to find new jobs when they have access to early intervention services prior to severance.
59
   Haney and Shkaratan, p. 29.
60
   The OED Evaluation Memorandum stated it in the following terms, “The lessons derived from this
operation [Enterprise and Financial Sector Adjustment] reinforce those derived from other operations in
other transition economies: (i) coherent macroeconomic, enterprise, and banking sector reforms are
essential; (ii) fundamental reforms generate wide-ranging debate on policy, institutional, and legislative
issues, and time is required to develop a consensus and to adopt the required legislation; and, (iii) it is
important to establish professionally staffed institutions with clear mandates to develop and implement
transition policies and programs.� The ILO stresses Johannes Linn in his introduction to Labor,
Employment and Social Policies in the EU Enlargement Process notes that one of the key lessons of
downsizing industries is that the social costs must be addressed and “an effort must be made to explain the
approach to the affected population groups.� p.19. This echoes the experience in OECD countries which
has derived the following steps: 1) consult and negotiate with labor groups at the earliest possible stage, 2)


                                                      11
example, were lack of communication by the management, lack of a concrete action plan,
and disproportionate percentage of layoffs among non-managerial staff versus
managerial. The findings from upcoming PSIAs for the Brodotrogir and Uljanik shipyard
restructuring can be used as a basis for identifying such issues and opening a dialogue on
response. The apparent “prosperity� of these subsidy-dependent enterprises may obscure
the need for restructuring to worker and other stakeholders. Workers are often unaware of
the programs of support already available; older workers who will lose both income and
status are particularly

In Russia, the World Bank assisted the Russian government in developing a strategy to
address the impact of coal sector restructuring. The strategy process was initiated by
mapping the resources, policy and information flows and ensuring the input of regional
and local governments, and a stakeholder analysis that unpacked the various diagnoses of
the key problems. Differences in proposed policy preferences had followed from these
perspectives. To facilitate the process of reform, an interagency commission was
established with representatives from different levels of government, including
municipalities and ministries and agencies. Mapping served to identify institutional
capacity, mechanisms and procedures for reducing adverse impacts. For example, in
Russia, the role of the Treasury Department was expanded to encompass a new system
for accountability and benefit delivery. The expertise of an agency within the Ministry of
Energy was tapped in terms of calculations of benefits, based on anticipated
retrenchment. In Croatia, some of elements of this process have begun under the PAL
program, but the ongoing challenge is to build in-country constituency and capacity.

Given the political economy of coalition government, the World Bank can play a critical
role as a facilitator and a catalyst in enabling the GoC to use existing institutions more
effectively to address the remaining “hard cases� of SOE restructuring and privatization.
Consolidating the various initiatives and coordinating interventions is a key challenge for
the GoC, but lessons learned from other experiences indicate that some form of working
group on enterprise restructuring with representatives from the various ministries,
agencies, and local authorities can be effective. Given the need to address the protective
labor legislation systematically and the difficulties in the political economy of such
reform, broad coalition participation in any working group would be critical.

Reform and Social Mitigation Measures

Based on the lessons of evaluations and previous operations, the preferred option in
development of social mitigation measures is a package of services with a focus on labor
redeployment. Services should be linked to labor demand and actual workplaces and be
designed with careful targeting. The GoC should be realistic about goals and allocate
resources based on a cost-effectiveness basis (rather than on stakeholder strength and
advocacy), bearing in mind that a growing economy creates jobs not labor programs, but


identify issues at the onset, 3) establish a framework for mitigation against adverse effects, 4) communicate
the benefits of privatization and 5) explain government plans for mitigation. OECD, Privatizing State-
Owned Enterprises, p. 16


                                                     12
that labor programs can facilitate the process of responding to a new economy.
Addressing the stringent employment protection legislation needs to occur in parallel.

A general multi-sector, multi-region approach61 would be advisable in Croatia. This will
help prevent higher costs due to the stronger SOEs in a more privileged sector(such as
shipyards or railways) negotiating a more favorable severance package. State concerns
about regional disparities and inequities could be addressed in through tailoring the
specific emphasis in labor redeployment services provided, based on enterprise analysis.

Integrating local authorities into the planning process can identify opportunities in the
regional economies that could be developed and absorb some of the laid-off workers. The
planning process at the županija level with the development of regional operational
programs (ROPs) could be integrated into planning for SOE restructuring. ROPs identify
regional resources and opportunities. Not all SOE assets are necessarily attractive for
investors during privatization. In many cases (such as agrokombinats), land and unused
buildings could be used for other productive uses.

Once strategic directions have been defined, and institutional actors have been identified,
both in terms of existing and potential roles, the next steps are defining the specific mix
of options to be utilized. In planning labor redeployment services,62 the characteristics of
the retrenched workers should be taken into account in determining the extent of job
placement services (generally younger workers with needed skills), institutional and on-
the-job training (older workers with skills not in demand) and redeployment services that
would respond to lack of labor demand (business incubators, small business technical
assistance). Some of these programs already exist in Croatia and could be more
effectively utilized. Data on skill profiles and enterprise surveys can be used to determine
the extent of retraining needed. As discussed above, tracking system and evaluations are
necessary to determine the efficacy of given interventions; flexibility to adapt programs,
particularly with respect to regional variations in the economy in Croatia. Tailoring to
regional variations should also take into account incentives that local governments have
to create environments for investment.

The remaining hard cases of SOE restructuring are likely to involve significant social
impact, in some respects amplified by the time delay in transition. Mitigating this social
impact will require consensus-building and careful design in response.




61
  Fretwell, “Mitigating the Social Impact,� p. 6.
62
  The Global Labor Toolkit for Labor Issues in Infrastructure Reform received a highly satisfactory from
QAG in the first evaluation of non-lending technical assistance. The QAG panel noted that it was a
comprehensive source of information that is useful to Bank’s client countries and in restructuring exercises.


                                                     13