Once the broad strategy and options are determined, the next step is to develop the main elements of the labor program. Specific approaches to labor restructuring are bound to vary from one country and enterprise to the next, depending on local circumstances. But labor programs typically include four main components: severance payments, pension payments, retraining and redeployment support, and employee share ownership plans.
In the absence of unemployment insurance and other social security arrangements in many developing countries, severance packages are typically the primary source of income support during the transition period to alternative employment. As such, they are a central element in any labor program.
Severance packages typically include some or all of the following components, each of which is discussed in detail in module 4 of the Toolkit:
In addition to providing immediate income support, severance payments facilitate labor support and allow PPI to happen, and mitigate the social impact of layoffs in the absence of unemployment insurance systems. The financial and economic returns can also be high, with short payback periods and increases in the marginal productivity of redundant staff redeployed to productive activities elsewhere in the economy.
But the design and implementation of severance payments is one of the more challenging areas in labor restructuring. Four main issues typically arise:
A severance package has a number of elements. Each of these may be determined by different legislation, regulations, rules, or agreements.
For the reasons outlined above, many governments have resorted to voluntary departure programs by providing severance payments that exceed legally mandated requirements. The size of the payments has varied widely among countries and within countries by enterprise, depending on legal and contractual obligations, the negotiating strength of labor unions, and prior precedents.
In some cases an established severance arrangement may already be in place; in others, the existing scheme is not considered sufficient or there simply is none, and a new scheme may need to be developed. Usually, severance plans are based on a multiple of years of service and salary, taking into account legal or contractual obligations and, in some cases, prior experiences or precedents in the state enterprise sector more broadly. Such formulas are easy for managers to use (table 1.1) and are widespread in both the public and private sectors. As module 5 shows, severance formulas have generally ranged from 1 to 3 months of salary per year of service, with a typical average payment of 1.5 months per year of service.
| Advantages | Disadvantages |
|---|---|
|
|
In developing a severance scheme the main challenge lies in devising severance payments that are both attractive for workers and financially affordable and sustainable. Severance has to be attractive enough for workers to leave voluntarily. At the same time, governments cannot afford to overpay because overpayment leads to problems of cost and financing and to problems of adverse selection.
In practice, overpayment has often occurred because generous payments are seen to be politically and socially attractive. In Pakistan, for example, an agreement with the unions resulted in a package equivalent to five months' pay for each year of service, which neither the government nor the firms could afford to pay and which subsequently led to delays in implementation. If an overly generous package is offered to all workers, there is the added risk of adverse selection. In the rail sector in Argentina, for example, the across-the-board offer led to the loss of key staff, which hurt the performance of the newly privatized firm. There is considerable evidence from evaluation of downsizing programs that adverse selection can result in the rehiring of workers who have received compensation payments, and that leads to problems of moral hazard and inefficient use of scarce public funds (see, for example, Haltiwanger and Singh 1999).
One way to contain the risk of excessive payouts and to minimize adverse selection is to set minimum and maximum levels of payment, where a minimum floor can be seen as fair to everybody while a maximum cap would ensure that workers and managers with longest seniority are not overpaid. Another way is to make workers who are within a few years from retirement, and thus have the least to lose in terms of future income, ineligible for voluntary schemes or for their severance payments to decline as they approach retirement.
A third approach is to better tailor severance packages to workers' characteristics through a lossbased method that takes into account factors such as seniority, gender, and education. Unlike standard severance formulas that compensate workers mainly for past service–where workers with higher seniority benefit disproportionately compared with workers having fewer years of past service and more years of denied service–a tailored approach that takes into account factors such as education and gender can predict the welfare loss of each worker and compensate the worker accordingly.
The loss-based severance method is more a complement to any existing method than a purely alternative severance method. Its advantages are that it serves as a benchmark to assess possible overpayment, helps induce the right self-selection, and helps contain costs. The approach aims to:
The approach has been piloted in Guinea-Bissau, Madagascar, and Tanzania, and is described in Chong and Rama (2000). The data requirements and methodological issues are covered in further detail in module 5.
Another way to avoid adverse selection, reduce the risk of rehiring, and contain costs is to identify the work activities and subsequently the work force cadres to be separated, and then to target the severance offer only to workers whose jobs have been identified as redundant (through benchmarking studies or functional analysis, for example), rather than to offer severance to all employees.
There are other ways to improve targeting and selection as well. These are discussed in greater detail in module 5 and include:
Rehiring is an indicator of program failure and poor design.
Where governments are starting a program of work force restructuring across a number of infrastructure enterprises or organizations, a critical decision arises. Should government adopt a single, uniform approach to compensation that will apply to all enterprises, or should severance packages be negotiated on a case-by-case basis?
A uniform approach may be preferable where there are strong trade unions and where a series of case-by-case negotiations can result in very high costs to government. Such an approach would avoid a situation where each new award raises the minimum severance level for the next negotiations and creates a ratchet effect that leads to increasingly higher severance levels that eventually become unsustainable.
A key decision is whether to adopt a uniform or a caseby- case approach to ex gratia severance packages.
At the same time, some flexibility might be needed to take into account the particular circumstances of the enterprise (for example, based on levels of overstaffing or financial performance). One approach would be for the government to develop severance policies and guidelines within which enterprises are allowed flexibility. In such an approach, the government might finance the cost of severance according to the guidelines, and any severance beyond the guidelines would be financed by the enterprise. This might allow enterprise managers in more profitable firms to restructure their work force more quickly by offering higher payments that can be financed from the enterprise's own resources but within an overall framework that prohibits excessively high compensation payments that the rest of the public sector could not afford.
Work on securing financial commitments must start early.
There are six sources of funds for labor programs.
The cost of severance payments can be high and funding arrangements need to be put in place early in the process to assure workers that timely payments will be made. In the absence of such arrangements government credibility can be at stake.
There are several sources of finance for severance and for labor programs more broadly:
Asset disposal is important but can take some time to implement. There is, however, nothing to stop the PPI manager from starting the process by, for example, commissioning initial valuations of assets.
"In the past the World Bank was not allowed to directly finance severance pay because it was not considered a productive investment. There were also concerns about the effectiveness of retrenchment schemes and the Bank's vulnerability to accusations of supporting and financing unemployment. But a number of factors led the Bank to decide in February 1996 to allow direct Bank financing for severance pay as part of investment operations. These included the importance of large-scale restructuring and privatization, the potential obstacles arising from lack of financing for labor shedding prior to sale, the growing evidence on the economic and financial returns to severance pay, and the limitations of adjustment lending. As a result, severance pay financing can now be provided for individual state enterprises or groups for enterprises throughout the reform process that is, from corporatization to restructuring prior to privatization" (Kikeri 1998, p. 35).
Some of the more complex issues in PPI arise from the way pension arrangements are addressed. A poorly conceived pension strategy can make an otherwise viable transaction effort untenable. The cost of past obligations or future commitments of an existing pension scheme are commonly among the more significant considerations in determining potential investors' willingness to participate in a PPI initiative.
Public sector employees often benefit from generous pension schemes that may far exceed those available to private sector workers, especially in developing and transition economies where pensions of any type are often unavailable to the majority of the work force. The value of pension benefits can be the largest single component of the compensation package for many workers and the only form of "savings" accessible to them. The current and future costs of sustaining these arrangements, however, are often so great that the way in which they are structured and financed becomes a key factor in whether an enterprise is financially viable over the long term.
Pensions also represent a potentially powerful tool for the restructuring of labor. They may be an effective means to lower labor costs through early retirements or voluntary separations and a powerful incentive to attract and retain highly skilled or essential workers.
The implementing agency must address three closely related but distinct matters in dealing with pension issues in PPI:
Each of those issues is complex and its resolution depends on the specific type of pension scheme in place. The issues must also be addressed in ways that satisfy the differing interests of employees, the prospective PPI investor, and government. Module 5 presents detailed guidance on each of these tasks for different types of pension schemes.
Given the importance of pension provision to workers' welfare, it will be essential to communicate, consult, and often negotiate with employees, trade unions, pension plan trustees, government (ministry of finance, labor, or social protection), and other stakeholders on changes in pension arrangements.
Pension liabilities are increasingly recognized as a potential risk to enterprise restructuring.
Where national or multiemployer pension plans may be required to commit additional resources to meeting the future obligations for pensions, it is essential that the relevant government ministry or the government actuary is consulted before communicating with unions and workers. Especially when these plans are actuarially bankrupt, or where government is about to reform the pension system generally, the implementing agency will need to be very careful about raising the expectation of workers' representatives about what is– and is not–possible.
Pensions issues will be a critical component of any communications plan.
Pension plan trustees or supervisory board members are important actors and stakeholders in pension issues in PPI. They will be able to inform the implementing agency about the structure of the plan, past precedents in interpretation of the plan, plan assets, and financial circumstances. As box 1.8 illustrates, pension plan arrangements may be both technically and politically difficult, and the implementing agency and trustees may find themselves on different sides of the negotiating table. It is therefore essential to consult with pension trustees. Their primary duty, however, is to the beneficiaries of the plan, and they may well have their own source of independent professional advice. The implementing agency may also identify a need to recruit specialist professional advice on behalf of government.
The reform of Johannesburg's water supply included efforts to implement more equitable and more financially sustainable pensions. The Johannesburg Water Company (JWater) was formed on November 21, 1999, and it inherited staff who were contributing to 12 different (defined-benefit) pension plans originally set up by the City of Johannesburg. Not only was this arrangement complex to administer, but the plans were not equitable, and generally favored white employees and senior managers. Moreover, the unfunded liabilities of the plans were large and growing, a situation exacerbated by high administration costs.
Restructuring of the company's pensions was essential. However, union representatives on the pension plans' boards of trustees had been generally opposed to the reforms of the city's water supply. Trustees had not formally recognized JWater as an employer because the rules of the plans did not generally provide for admission of nonlocal government employees. Moreover, trustees of the two largest (City of Johannesburg) pension plans were reluctant to make rule changes, and they used the need for such changes as a bargaining tool to obtain other concessions from the city.
These problems were not unique to the water sector; other city pension schemes had similar difficulties. In December 2001, in an attempt to resolve citywide pension problems, the city administration unilaterally closed all pension plans to further contributions. A new provident fund was created. This was a defined-contribution (accumulation) plan with a 15 percent contribution by employers and a 7.5 percent contribution by employees, plus life insurance and disability benefits. The city's unilateral action was legally challenged in the courts, and an interim order granted relief to the two largest plans in March 2002. Other schemes subsequently launched similar applications. If all are successful, then pension provision will essentially be unchanged and the implementation of pension restructuring will be dealt a severe blow or at least delayed.
With hindsight, there had been too little consultation with plan trustees, unions, and workers. The court challenge to the city was based on the grounds that (a) the interests of some pension plan members had been prejudiced, (b) the consultation process had been inadequate, and (c) earlier guarantees given by provincial legislation and the employer during the reorganization had been infringed. Those pensions issues are not yet resolved, but key lessons for labor adjustment in PPI are:
Source: Personal communication from staff in Johannesburg Water, 2002.
The implementing agency will usually need to enlist specialist advice on pensions.
For some PPI implementing agencies national pension reform will be of little relevance, either because the reform has already happened or because pension reform is not in the pipeline.
If pension systems are in the midst of reform, this may well affect the way the PPI manager will package or restructure pensions within the PPI transaction.
Where government is in the midst of system reform, however, strategic decisions may need to be made. Wider pension reforms can affect specific decisions on pension issues for a power, telecommunications, rail, or port PPI, or may set precedents for other state-owned enterprises. Until 1981 most public pension plans were defined-benefit, pay-as-you-go arrangements, potentially unsustainable financially. The most well-known early reforms were those of Chile in 1981, which introduced a system based on individual accounts with fully funded, fully vestable, and fully portable benefits, plus minimum pension guarantees for workers on low wages and interrupted contributions. Chile's reform was followed by reforms in several other Latin American countries. In Bolivia's capitalization program (box 1.9), pension reform and privatization of state enterprises (including utility and infrastructure companies) were intimately linked because the state's shareholdings were used to establish new pension schemes.
PPI work force restructuring may also be affected by the timing of systemwide pension reform. For example, labor adjustment in Mexico rail took place as pensions were being reformed, which meant that workers knew that the new private companies would provide pensions under new mandatory, fully funded arrangements (López- Calva 2001). In Brazil many workers opted for early retirement rather than severance because of pension reform (box 1.10). This response is not unusual; several countries have seen surges in applications for retirement when pension systems are under review and workers are uncertain of the outcomes. Depending on circumstances, therefore, PPI labor adjustment may be helped or hindered by pension system reform.
Consultation with government is important to ensure consistency of approach and policy coherence between pension system reform and PPI implementation, particularly where:
Pension system reform can encourage more workers to opt for early retirement.
Bolivia's 1994 capitalization law is a unique example of a combined privatization and pension system reform program. Major state enterprises, including gas, telecommunications, railways, airlines, and electricity generation and transmission, were capitalized through a capital increase by private investors of up to 50 percent of the companies' capital. During 1995 and 1996 international and domestic investors acquired shares in these enterprises through competitive bidding processes. Government transferred the remaining 50 percent of the shares in the enterprises into two new privately managed pension plans that were also mandated by the capitalization law.
The Bolivian approach meant that there were no privatization proceeds to the state budget, but that the investment went directly into these infrastructure assets. At the same time, the utilization of pension plans helped provide for development of the local capital market and meet the social protection needs of the population.
Source: Guislain 1997.
The PPI implementing agency's plans for tackling pensions may run counter to government's overall policy.
In planning for labor adjustment in Brazil rail, it was expected that about 5,000 workers would take early retirement and 13,000 would choose voluntary separation. In fact, almost the reverse occurred: 11,771 opted for early retirement and 5,886 for separation. The reason was a proposal in Brazil's Congress for radical reform of the pension system, which implied that eligibility to retire no longer would be based on the number of years worked but on age. Fear that changes in the social security law would mean that workers would have to work many more years to receive similar benefits or that the changes would jeopardize the retirement income of older workers led to an unexpected increase in the number of applicants for early retirement.
Source: Estache, de Azevedo, and Sydenstricker 2000.
Many governments have supplemented severance and pension packages with redeployment support to help workers regain productive incomes– whether through formal employment, self-employment, or informal livelihood activities. Such programs also offer an additional incentive to encourage voluntary departures and help win support for politically difficult restructuring. They are aimed at facilitating the shift of economically unproductive workers from infrastructure sectors to more productive sectors of the economy. One by-product of retraining and redeployment programs may be a general shift in attitude among the work force at large, away from a perceived reliance on public sector employment toward private sector jobs and self-employment.
Redeployment is a tool of active labor market policy.
In some cases redeployment programs have been developed explicitly for PPI employees; in others, redeployment support is provided as part of broader active labor market programs for the unemployed. Such programs differ from passive labor market programs that act more as a safety net for the most vulnerable people (box 1.11).
There are five main types of redeployment support.
The main types of redeployment support are:
Redeployment support services linked to enterprise restructuring programs are a particular subset of "active" labor market policy. "Active" labor market policies or programs are those that directly prepare or assist the reintegration of workers in the labor market through activities such as job search assistance, job placement plans, training programs, and employment subsidies. These can be contrasted with "passive" labor market policies, which support incomes, usually through financial transfers. Examples of passive policies are unemployment insurance, worker disability payments, and–relevant to the context of PPI labor adjustments–severance pay. Some have used the metaphor of a trampoline to characterize the concept behind active labor programs (to lift workers back into work) compared with the safety-net concept of passive programs.
Counseling, training, and job search lie at the core of most redeployment programs (see figure 5.2 in module 5). Community-based approaches, public works, and employee enterprise are supplementary elements that can be appropriate in some circumstances. In designing programs, there is often opportunity to involve unions, local government and, where relevant, wider community and civil society groups in consultation processes.
On the whole, redeployment programs have had mixed results. Most evaluations have focused on the experience of industrialized countries. These experiences generally show that retraining programs resulted in modest gains in reemployment probabilities, but wage changes were negligible or negative. The same evaluations found that the costs of retraining are two to four times higher than job search assistance but are no more effective. By contrast, placement and counseling efforts tend to show positive results and are generally more costeffective. Systematic evaluations of the impact of such programs in developing countries have been few, although anecdotal evidence suggests that retraining programs often fail because of timing delays, weak institutional capacity, low education levels, and the lack of employment opportunities for retrained workers. The programs frequently have been more driven by supply than by demand.
Nonetheless, emerging experience from a wider range of circumstances can inform the design and implementation of redeployment programs and help workers with several remaining years of productive life to acquire gainful new livelihoods. Better results are more likely if wider economic and labor market policies aimed at creating sustainable employment growth are already in place (as outlined in the second section of this module). At the program level, effectiveness can be improved by:
The implementing agency can benefit from experience elsewhere when planning redeployment.
Module 5 discusses in detail the different types of programs and the key steps involved in designing and implementing redeployment programs. These programs and steps include the use of surveys as building blocks for the program, designing incentives to ensure that programs are demand-driven, costing out the services and securing funding arrangements, setting up labor funds or other implementation arrangements, identifying and commissioning service providers, and developing systems for monitoring and evaluating results.
Share transfers are best seen as an addition to rather than a substitute for voluntary departure or early retirement.
In work force restructuring, employee share ownership plans can be used in three ways:
In such arrangements, governments have reserved shares, ranging anywhere from 3 to 10 percent depending on the size of the transaction, for employees in the PPI process, often at discounted prices and with special financing arrangements. Financing share ownership schemes is the major concern in developing countries. Some countries (for example, Bolivia and Chile) overcame this concern by allowing workers to use their end-of-service benefits to invest in the share program, with the guarantee that the value of the shares would not fall below their entitled benefits at the time of retirement. As a result, in many of the enterprises more than 80 percent of employees participated in the program. In addition to financial gains from such arrangements, ownership programs can help give employees a direct stake in the performance of the company and thus help improve labor relations and labor productivity.
Module 5 summarizes experiences in the use of employee share ownership mechanisms, and concludes that share transfers are probably best viewed as a supplement to rather than a substitute for severance payments.
