Warning: This document contains many graphics and tables and may therefore take a long time to download!

Chapters 1 to 5

6. Labour market development

Centrally planned economies were organized as full employment societies. Unemployment had been unknown throughout the Soviet Union since before the second World War. Being out of work was even considered a "parasitic walk of life," and could be punishable by law. At regular intervals the State Planning Committee allocated labour by sector and industry branch, down to the individual enterprise, as well as set remuneration levels and social security benefits. The result for workers was job security, which brought with it income security, access to education and training, social services, a pension and other benefits. However, the system also resulted in production rigidity and inefficiency. Enterprises were not free to determine employment levels, a constraint that was tolerable since they also had guaranteed "markets" for allocated levels of output.

Because life-long employment within a single state enterprise was the norm, governments had little experience in devising and implementing labour market policies to speed re-deployment of workers between jobs. The magnitude of the workforce size reductions which have accompanied the transformation of state-directed enterprises requires the establishment of a different sort of labour allocation and safety net system. It must be designed to protect workers between jobs from falling into long-term unemployment and poverty and to prepare them for different jobs.

This section examines evidence gathered by the surveys on the demand side of the labour market, including information on how employers recruit and retrench workers, how terms of employment have adjusted to the new conditions of economic uncertainty, and how wages are determined. These indications of emerging labour markets provide some indication of what measures may be needed to improve the ability of labour markets to perform new tasks in the decentralized economies. The first part summarizes respondents' labour market activities, looking at how privatized establishments recruited and retrenched workers and whether they turned to the national employment services for help with either task. The second part describes how wages are determined, comparing the influence of "market" forces, individual and team performance, and regulations. The third part assesses the impact of restructuring and labour market practices on wage payments and social service provision. Section 7 continues the discussion on labour markets by examining changes in the occupational structure of employment and the implications for worker training.

6.1 Labour market activityThe surveys show that despite the 10 to 15 per cent net downsizing over the previous year, a surprisingly large share of respondents had also hired some workers. Nearly 90 per cent of the establishments in Russia reported hiring some workers, as did about two-thirds of the establishments in Armenia and Georgia (figure 6-1). However, newly hired workers represented at most 6 per cent of the surveyed workforce, in Russia and Georgia, and fell to less than one per cent in Ukraine. Similarly small magnitudes were reported for current openings: the vacancy rate was as high as 2 per cent only in Ukraine, and hovered around 1 per cent in the other surveys. (See Appendix table 9.)

Figure 6-1 Share of establishments reporting vacancies and recruitment (p. 85)

Exits far surpassed hires as a share of current workforce levels in all countries (figure 6-2). The ratio of job-leavers to recruits was highest in the Kyrgyz Republic and in Ukraine, where the surveyed establishments had hired hardly any workers. Over the same time period, the number of exits, workers that had separated from their employers for whatever reason, was equivalent to 28 per cent of the remaining workforce in the Kyrgyz Republic, 10 per cent or more in each of the other FSU countries, and 8 per cent in China. (See Appendix table 9 for data used in figure 6-2.)

Figure 6-2 Hires and exits as percentages of current employment (p. 85)

In this context "exits" refers to all manners of job separations: retirement, retrenchment, voluntary quits, and discharges for unsatisfactory performance on the job. The restructuring period began with employment institutions and regulations that made it difficult for workers to separate from employers, even when they had been placed on unpaid leaves or reduced work schedules: Regulations restricted workers' ability to move to other districts in search of employment, employment services could not provide adequate job-search assistance, formal separation from employers could entail loss of job-related subsidized housing, unemployment benefits were either nonexistent or below minimum needs, lack of training and retraining programmes outside of employment made it difficult for displaced workers to qualify for new types of work, and small-credit institutions to provide startup capital for small business creation did not exist. Although many of those counted as "quit voluntarily" may have been given incentive to leave by being placed on reduced work schedules or unpaid leaves, the group of job-leavers classified by employers as "retrenched" or "released" is the clearest measure of downsizing by establishments for economic reasons.

6.1.1 Retrenchment: Extent and practicesThe distribution of job leavers by method of leaving their employer could be calculated for four of the surveyed countries. Workers released by their employers for economic reasons as a percentage of all job-leavers was highest in Georgia, where retrenched workers accounted for one fourth of all job leavers, and was lowest in Russia, where they accounted for about 8 per cent (figure 6-3). Privatized establishments reported a higher share of retrenchment among job leavers. The differential was highest in Ukraine, where retrenchments accounted for 13 per cent of job leavers in state enterprises compared to 22 per cent in privatized establishments.

Figure 6-3 Retrenched workers as a percentage of job leavers, by property form (p. 86)

Across the two rounds of the Kyrgyz surveys, the share of workers released out of all job-leavers doubled from 7 per cent in 1993 to 14 per cent in 1995. The share of job-leavers "dismissed" (for poor work performance, absenteeism, breach of work rules, etc.) tripled, from 7 to 21 per cent. The distribution of job-leavers by reason was very similar across size classes of establishments and industries (Evans-Klock, 1998). The report based on the China survey also documented greater initiative by managers and workers. Although retrenchment for economic reasons could not be singled out (as it was combined with other rationales permitting employers to release workers such as their unsuitability for the job or discipline problems), dismissals and quits accounted for a larger share of job exits in the town/village enterprises and in the private sector than in the state enterprises (Lim, et al., 1996).

The evidence was mixed in terms of whether privatized establishments were more likely than state enterprises to release workers. In Georgia, 21 per cent of private establishments had released workers compared to 7 per cent of state enterprises. In Russia, the margin was only slightly smaller, 36 per cent of privatized establishments had retrenched employees compared to 17 per cent of state enterprises (table 6-1) . However there was no significant difference in the likelihood of making retrenchments between the two groups in Ukraine and the Kyrgyz Republic. The low rate of retrenchment is also shown in table 6-1: In Russia, less than 1 per cent of the workforce in either the privatized or state sector had been formally released by employers over the previous year. The retrenchment rate was as high as 4 per cent only in Kyrgyz state enterprises.

Table 6-1 Retrenchment in state-owned and privatized establishments

Country	State-owned establishments			Privatized establishments
	Share of establishments that retrenched workers	Retrenched workers as per cent of total job-leavers	Retrenched workers as per cent of total work force#	Share of establishments that retrenched workers	Retrenched workers as per cent of total job-leavers	Retrenched workers as per cent of total work force#
Georgia	6.9%	20.8%	1.6%	20.7%***	27.1%	3.7%
Kyrgyzstan	42.3%	16.3%	4.1%	40.1%	13.6%	3.1%
Russia	16.9%	6.0%	0.6%	35.8%***	8.4%	0.7%
Ukraine	58.5%	13.5%	1.2%	59.6%	22.2%	2.1%
*** The difference in the share of establishments that retrenched workers between privatized and state establishments is statistically significant at the 1 per cent level. # Per cent of total employment in survey, including in establishments that did not retrench workers.

Workers formally released by their employers are entitled to a severance payment from their former employers, usually equivalent to three months wages (see text box). This is a legal entitlement for displaced workers, but not for workers placed on unpaid leave, an alternative method of adjusting workforce size. Although most employers cited concern for workers' well-being and hopes that the full workforce could be employed productively in the future as their principal reason for retaining surplus workers, severance payment requirements do make permanent separations a more costly way of reducing the workforce. The severance payments are of enormous social concern. In the absence of government-provided unemployment benefits, even the temporary income provided by lump sum severance payments may be vital to keep families out of poverty, following job loss.

If adequate unemployment benefits were available, resisting formal departures (through extended leaves, for example) in order to avoid severance payments only increases the share of restructuring costs borne by individual workers. In their absence, maintaining formal employment relationships may enable surplus workers to retain their rights to some enterprise-based services, add working time to their pension eligibility, and keep their place in the queue for future work. On the other hand, resisting formal separations when downsizing is unavoidable postpones restructuring and recovery for the enterprise.

China was the first of the centrally-planned economies to introduce an unemployment insurance scheme. The programme was designed to relieve state enterprises of the principal burden of redeploying workers, as well as to provide a safety net for workers that would enable them to be more mobile and more accepting of the policy to promote state enterprise restructuring. The scheme is administered locally along national guidelines. The FSU countries have also instituted state-administered unemployment benefits, but these programmes have been underfunded due to government budget deficits and have generally been inadequate in terms of level of support and coverage, as shown in the following box summaries (information collected in field interviews and from background studies).

6.1.2 Employment services

Apart from their role in determining eligibility and distributing benefits, the public employment services (PES) serve as a clearinghouse of information to help match job applicants to employers' vacancies and thus promote labour mobility between firms. The surveys collected information on the extent to which employers made use of the public employment services by informing released workers that they should register there, by posting their vacancies with the PES, and by recruiting through them.

The reliability of unemployment statistics and thus estimates of the need for unemployment assistance depend on displaced and unemployed persons registering with the PES. Only in two of the five FSU countries did even half of the employers that had retrenched workers encourage released workers to register with the employment service (Russia and the Kyrgyz Republic). Notification rates were lower in Georgia, where only one out of five managers that had released workers notified them to register (figure 6-4).

Figure 6-4 Use of the Public Employment Services, FSU countries (p. 86)

As also shown in figure 6-4, Russian and Kyrgyz managers were more likely to advise released employees to register with the PES than they were to either post their own vacancies or to recruit from the PES. Still, more than half of employers with vacancies had posted them at the PES, as had about half of the employers looking for applicants in Armenia and just over a fourth of those in Georgia. There was no significant difference between private and state enterprises in the likelihood of their reporting vacancies or encouraging released workers to register with the PES.

As shown above (figure 6-1), a majority of respondents in each country (except Kyrgyz manufacturers) had hired at least one employee in the previous six to twelve months. Among these establishments, the variation in the proportion that recruited through the public employment service varied greatly between countries but only marginally between privatized and state-owned enterprises within countries. For example, about 40 per cent of the Russian establishments that had hired workers had turned to the employment service to help fill vacancies, with the share rising to 50 per cent of the open joint-stock companies (figure 6-5). Only a very few of the Georgian respondents that had hired workers had turned to the employment service, as had about one out of ten of the Armenian respondents. In the Kyrgyz survey, however, state-owned establishments were more likely than privatized establishments to use the employment service.

Figure 6-5 Recruitment through public employment services, by property form (p. 87)

China's public employment services appeared to be the most useful in serving the needs of non-state establishments, with nearly 60 per cent of the joint-stock and joint venture establishments reporting having recruited through the official employment services. However, in general, managers in China have had less freedom to select and recruit workers. For example, some 40 per cent of new recruits had been assigned to state enterprises by local authorities and another 23 per cent had been transferred from other enterprises. Just over half of the recruits in the labour surplus enterprises were, not surprisingly, surplus workers that had been transferred to them from parent state enterprises. Managers in the village/township enterprises enjoyed much more autonomy in selecting workers (Lim, et al., 1996; p. 24).

The poor performance of governments in the FSU to provide services to unemployed workers through these public employment agencies results in low reporting rates, in turn reducing their utility to employers. For example, just 13 per cent of the entire sample of Kyrgyz employers surveyed reported that they would turn to the employment service for assistance were they to recruit new workers. Some 70 per cent said recruits could be found from among general applicants or those responding to specific advertisements. The limited interaction between employers and the PES in terms of hiring may continue as long as employers can easily increase labour by calling back workers placed on leave or short work schedules. However, as long as only a small portion of released workers register, unemployment rates will be underestimated, leading to inadequate budget funds devoted to unemployment insurance and other emergency social services for displaced workers. The cycle continues as inadequate benefits further discourages workers from registering.

6.1.3 Employment termsAnother indicator of the emergence of market forces in labour distribution is the extent of non-permanent types of employment arrangements. The measure of flexible employment terms for which data was collected in each of the surveys was whether the establishment hired any workers on a part-time basis (as distinguished from the practice of placing full-time workers on part-time schedules as a means of managing labour surplus.) The survey findings show that part-time employment arrangements have become quite common in these transition economies (figure 6-6). The practice was most popular in Ukraine, where more than half of the survey respondents employed part-time workers, followed by Russia and Armenia (with about 40 per cent) and the Kyrgyz Republic (about one quarter).

Figure 6-6 Employment of workers on part-time basis, by property form (p. 87)

However, as also shown above, the likelihood of establishments using part-time employment contracts varied more by country than by type of establishment. Frequency of part-time contracts (not to be confused with workers placed on reduced working hours as discussed above) were markedly similar across private and state-owned enterprises in Russia and Ukraine (uniformly high) and Georgia (uniformly low). In China, where privately-owned or managed enterprises out-performed state enterprises, part-time work arrangements were found more frequently among the lagging state enterprises. The findings were reversed in Armenia, where 45 per cent of privately-owned enterprises but 37 per cent of state enterprises reported part-time work contracts.⁽¹⁾

Among employers already using part-time employment arrangements, only in the Kyrgyz survey did a majority indicate that the share of workers on part-time arrangements had increased over the previous year (figure 6-7). At least half of the employers with part-time employees in each of the other countries said that part-time employees' share of the work force had not changed.

Figure 6-7 Change in the share of part-time workers over previous year (p. 88)

On the one hand, the limited emergence of part-time employment arrangements could provide evidence that the quality of employment was not deteriorating for those workers that had retained their jobs. However, in the context of continuing labour surplus within firms, these findings could also indicate a lack of flexibility in adjusting the work force to the amount of work available in ways that might be more acceptable to workers than unpaid leave, involuntarily reduced work schedules, or retrenchment.

6.2 Wage determinationUnder the command economy, wages were determined largely by the State, with relatively little influence of plant or individual performance, and relative uniformity across industrial sectors and regions. The wage tariff system was a complex set of wage differentials, determined by the central government and based on a statutory minimum wage. As part of the privatization process, employers had more flexibility in determining not only wage rates but the criteria upon which they would be decided. In addition, they were pushed to break away from the wage tariff system by inflation and rapidly rising costs of living which had not been adequately compensated in the minimum wage levels and official wage schedules:

• In 1995, fewer than half of the state enterprises and only one quarter of the private firms in Russia were using the state tariff wage system. Where used, the official wage tariff rate was supplemented because the occupation and other differentials had been set on the base of the minimum wage which had been held down far below subsistence levels of income (Standing, 1996).
• Only one third of the Armenian respondents reported using the state tariff wage system in 1995. Among private firms, the rate was about 11 per cent. As in Russia, the tariff system was used less frequently by smaller firms (Samorodov and Zsoldos, 1995).
• An even smaller portion of firms surveyed in Georgia used the state wage setting mechanisms, about 20 per cent of all establishments and about 9 per cent of private establishments (Samorodov and Zsoldos, 1998).
• In Ukraine the nominal minimum wage lost meaning as a wage setting mechanism. The minimum wage constituted 7.8 per cent of the average wage in January 1994, compared to 53 per cent in 1991 (ILO-CEET, 1995).

One measure of responsiveness to market pressure is the extent to which wage levels reflect individual and plant performance. Comparisons are difficult to draw from the surveys because the wage-setting information was collected in two different ways. In some countries, the questionnaire asked respondents to identify the main determinant of wages for production workers (Armenia, Georgia and Russia) while in others, respondents were asked the determinants for only the performance-based portion of average wages (Kyrgyz Republic, Ukraine and China).

Figure 6-8 shows the share of establishments which reported that individual performance was the main criteria in setting wages for production workers. The left-side panel shows that from 16 per cent of the Georgian respondents to 26 per cent of the Armenian respondents cited individual performance as the main wage determinant. Similarly, individual performance was cited as the main determinant in performance-based pay (as opposed to plant or unit performance) by 10 per cent of the Kyrgyz respondents and 25 per cent of the Chinese respondents (right-side panel). However, allowing respondents to select a combination of wage determinants shows that actually a larger share of managers took individual performance into account, along with plant or unit performance, in setting wages: 40 per cent or more of the Russian and Armenian respondents identified individual performance as one key determinant in wage determination.

Figure 6-8 Individual performance in wage determination (p. 88)

Figure 6-9 shows a similar comparison for state and privatized establishments within countries, again with the caveat that the comparisons made are more similar between Armenia, Georgia and Russia and between the Kyrgyz Republic, Ukraine and China. The performance of the plant was the key determinant of workers' wages. However, in Armenia, respondents were equally likely to consider individual performance as a main factor. Rating of factors was markedly consistent between privatized and state enterprises. However in three of the countries, Georgia, Kyrgyzstan and China, a larger share of the privatized establishments based wage rates on individual performance, with the differential being largest in China.

Figure 6-9 Pay determination criteria, by property form (p. 89)

Bonus payments were also common among the surveyed establishments. At least 40 per cent of the establishments reported having bonus payment systems, except in Georgia were they were relatively uncommon (figure 6-10). Within countries, there was not a marked increase in the likelihood of distributing bonuses by the privatized firms.

Figure 6-10 Incidence of bonus payments, by property form (p. 90)

Although the distribution of establishments that pay bonuses cannot be broken down by profitability levels of the establishment, the figures seem high considering the poor overall performance of most establishments. The explanation may be that rather than represent extra earnings in reward for good performance, the bonuses are a management practice to try to boost workers' earnings when wages are particularly low. Some support for this interpretation is provided by the high share of wages represented by these bonuses. As shown in figure 6-11, among firms that paid bonuses, these payments represented from one fifth to over one half of the average wage. Again, there were no marked differences between privatized and state establishments. The fact that bonuses were not of relatively greater importance among the privatized establishments, where profit and productivity incentives might be assumed to be greater, lends credence to the interpretation that these payments may be driven more by income support concerns than as rewards for productivity or firm profitability.

Figure 6-11 Bonus payments as percentage of wage earnings, average by property form (p. 90)

6.3 Changes in wage payments and social service provisionsThe manufacturers surveys not only documented the job loss and precariousness of employment for surplus workers, but also documented the precariousness of earnings for those fully employed. Most of the country surveys collected data on non-monetary or in-kind wages, wage arrears accumulated by employers, and changes in social services provided to employees.

6.3.1 Wages and payment arrearsWage practices in Russia were interpreted as showing extreme flexibility (Standing, 1996). While establishments retained workers in excess of the number that could be deployed productively, wages adjusted downwards in real terms and were tolerated at basically zero: failure of employers to pay their workers. Wage arrears were tolerated -- by legal authorities, by unions, individual employees -- because the probability of being paid in future by the same employer was higher than the probability of finding paid employment elsewhere, because side-jobs, moonlighting, small service provision activities were relied on by workers to supplement job earnings, and because it was generally accepted that employers could not pay wages if the enterprises themselves were not earning cash.⁽²⁾

Half or more of the establishments in each of the FSU surveys reported having difficulty paying wages, defined as one or more periods of wage arrears in the previous year (figure 6-12). The portion of respondents reporting wage difficulties ranged from a low of 47 per cent in Russia to a high of 70 per cent in the Kyrgyz Republic. Thus, half or more of the survey respondents in each country owed workers back wages. There was no clear indication from the cross-survey comparison whether privatized establishments resorted to wage arrears oftener than did state enterprises. In Russia, a higher portion of the state enterprises reported difficulties meeting their payroll (59 per cent compared to 46 per cent of privatized establishments). In Armenia, privatized establishments were more likely to report this problem (58 per cent compared to 44 per cent of state enterprises).⁽³⁾

Figure 6-12 Share of establishments reporting wage arrears, by property type (p. 91)

For the workers affected, the period of not receiving wages could be substantial. Among establishments that reported periods of not paying wages, the average duration was ten weeks in Armenia and twelve weeks in the Kyrgyz Republic (figure 6-13). In most of the countries, the average duration of unpaid wages was longer among state enterprises than among privatized ones.⁽⁴⁾

Figure 6-13 Average duration of wage arrears, by property type (p. 91)

Wage arrears can be interpreted in many ways - employers taking advantage of workers, weakness of national institutions to impose penalties on employers for failing to pay workers, weakness of unions vis-a-vis management. Wage arrears joins unpaid leaves and reduced working time as one more means of cutting the costs of retaining surplus workers. They may, in fact, be due more to liquidity problems and supply and market network disruptions, than to employers' capriciousness. There are limited survey findings to support this interpretation: Over 60 per cent of the employers in each survey that had put some workers on extended leave also reported some period of wage arrears.

Some employers also resorted to making payments in-kind to their workers. This option may be attractive to employers when they can find no buyers for their products, or no buyers with cash, but give part of production to workers in lieu of monetary wages rather than releasing workers or declaring bankruptcy. Employees, in turn, become the sales personnel or distributors, trying to sell or barter what they have produced in exchange for the consumer goods they need. As shown in table 6-2, this practice was common only in Armenia (data was not collected in the Kyrgyz and China surveys). Nearly one out of five survey respondents reported having given workers part of their wages "in-kind." Among these establishments, in-kind payments were equivalent on average to 16 per cent of total establishment output. The practice was used relatively rarely elsewhere, by fewer than 10 per cent of establishments. There were no significant differences between the share of privatized and state enterprises that followed this practice.

Table 6-2 Output distributed to workers as remuneration, by property form

Country	All establishments		State-owned		Private-owned
	Per cent responding Yes	Of these, average per cent of output distributed to workers	Per cent responding Yes	Of these, average per cent of output distributed to workers	Per cent responding Yes	Of these, average per cent of output distributed to workers
Armenia	18.0	16.2	23.0	18.7	19.5	15.3
Georgia	8.5	11.9	0.1	29.0	9.5	9.5
Russia	7.6	11.5	4.6	18.0	8.1	11.8
Ukraine	8.5	5.9	5.5	6.4	9.9	6.6

6.3.2 Provision of social servicesWages were kept low in part because of the widespread provision of social services by employers -- traditionally considered an important function of enterprises under central planning. Enterprises provided an all-embracing system of social protection, including subsidized housing, garden plots, subsidized meals in company cafeterias, kindergartens, vouchers for buying durable goods, family allowances, vacation facilities, as well as more traditional employment-related benefits such as health insurance and pensions.

Enterprises' "earnings" provided by state agencies were calculated to cover such provisions, and substituted therefore for other public means of providing social services. In the command economies, enterprise-based social service provision may have been efficient: Unemployment was unknown, so universal coverage was assured; state enterprises were large, so economies of scale in terms of facility usage and staff could be gained. In the transition period, however, firms have been encouraged to shed service facilities as non income-earning activities which can be ill-afforded in turning each plant into a profit-oriented enterprise.

Movement in this direction, however, has been obstructed by the lack of alternative public provision of basic services. Local government agencies have lacked the means to collect taxes and fund services. Furthermore, low earnings stymy the emergence of private providers due to lack of effective demand - workers' ability as well as willingness to buy. Failure to build alternative mechanisms of providing social services has had two very significant effects: The well being of workers and their families is jeopardized and the functioning of the labour market is crippled as workers try to maintain ties to employers still able to provide some social services but no work.

As shown in figure 6-14, the share of establishments providing social services varies greatly across countries and according to type of service measured. Most noticeable is that employer provision of the full range of social services remains the norm for the Chinese establishments. Employers in Armenia and Georgia were least likely to provide a wide range of services.

Figure 6-14 Provision of social services, by property form (p. 92)

It was expected that privatized establishments would shed social service functions more quickly than would state enterprises. First, a re-definition of enterprise responsibility would motivate them to cut costs by reducing benefits and services. Second, if large establishments were broken up through privatization into smaller units, the economies of scale would be lost, making it more expensive for employers to provide the same range or quality of services. Across the countries there is mixed evidence on whether social service provision has dropped off more significantly among the privatized establishments. In China, the Kyrgyz Republic and Russia, provision of social services was often higher among the privatized establishments; in Armenia, Georgia and Ukraine there was usually no difference in the frequency of provision by privatized and state enterprises:

• In the Kyrgyz survey the provision of each of the principal types of social services was higher among joint-stock companies than among state enterprises. For example, 37 per cent of the joint-stock companies provided housing to their workers, compared to 27 per cent of state enterprises. The contrast was even more striking for medical care: 30 per cent of the joint-stock firms provided clinics compared to 17 per cent of the state enterprises. The privatized establishments were also more likely to provide child care facilities and farm plots and to subsidize workers' transportation costs (Evans-Klock, 1998).
• In China there were no significant differences between the state sector and the non-state sector in terms of the "social burden." The joint ventures and joint stock companies appeared to have conformed to the practice in China of offering workers a wide range of benefits, and in fact had the highest frequency of provision of housing and medical care. Although there is some question as to whether provision of such services was mandated by local governments, the survey did not support the contention that workers hesitated to take jobs in the non-state urban sector because private firms did not provide housing. However, the village/town enterprises generally provided the lowest level of social benefits, for mainly rural workers. The labour service enterprises provided fewer services than did their state-enterprise parents, which may have continued to provide some services to their employees (Lim, et al., 1996).
• Enterprise-based social services were substantially less common in Armenia and Georgia. In Georgia not a single major category of social services was provided by as many as one out of ten of the employers surveyed. In Armenia, health service, which encompasses both health insurance and clinics, was the most often provided service, but by a meagre 14 per cent of respondents.
• Extending loans to employees was the most frequently provided social service in the four FSU countries where it was included on the provided list of social services on the survey questionnaire. This may indicate acknowledgement that wage income is insufficient, or that employers are constrained in the wage they offer workers and yet have funds available to lend. It also indicates the lack of alternative institutions to provide consumer loans. A third possibility is that the loans are used to finance spin-off businesses, redeploying personnel into self-employment and in whose small businesses parent employers retain a financial stake.

There was more variation in providing social services by size of establishment than by type of property form. In the Kyrgyz Republic, for example, almost 90 per cent of establishments with more than 1000 employees provided housing to their workers, compared to 50 per cent of medium-sized establishments and under 10 per cent of those with fewer than 100 employees. Clinics were provided by three out of four of the largest establishments, but by only one out of five of those with fewer than 250 employees and by none of those with fewer than 100 employees.

From the perspective of either wellbeing or labour market functioning, provision of subsidized housing is of special importance. The linkage between housing and employment may go far to explain why workers maintain formal ties to employers when placed on unpaid leave or despite non-payment of wages. As shown above (figure 6-14), the share of respondents that gave workers housing subsidies or provided enterprise-owned housing varied widely: from ten per cent or fewer establishments in Georgia and Armenia, to 16 to 20 per cent in Ukraine, and 30 per cent in the Kyrgyz Republic. In Armenia and Georgia, establishments were more likely to provide their own housing to workers than to give them housing subsidies (about 2 per cent of respondents in each survey), whereas in Russia, establishments were more likely to provide subsidies than to own workers' apartments (8 per cent).

Linking housing to employment is most pronounced in China, where nearly 70 per cent of the survey respondents provided housing to their workers. "Housing has been a critical element of workers' compensation in China. Although capital investment by enterprises in housing has, no doubt, helped to improve housing conditions of urban families, this practice tends to limit labour mobility. Workers are reluctant to leave their jobs because alternative housing is difficult to obtain. In addition, the building and maintenance of apartments represent substantial financial burdens on companies" (Lim, et al., 1996, p. 30).

Although the surveys in some countries found a surprisingly high rate of social service provision by establishments, provision of social services by enterprises had declined between survey rounds:

• The fifth round of the enterprise surveys in Russia documented a decline in the share of firms offering social services to their workers. Particularly widespread erosion had occurred in the provision of subsidized meals, financial assistance for retirees, and subsidised kindergartens. Although social benefits were important supplements to low wages, the probability of workers receiving specific benefits tended to be greater in the firms paying above-average wages (Standing, 1996).
• Among Kyrgyz employers, on-site social service facilities had decreased the most, such as subsidized cafeterias, clinics and child care. Housing continued to be the most frequently provided benefit, by 30 per cent of all respondents in 1995, down only slightly from 1993.

However, the decline in provision of social services was not uniform. A surprisingly large, although minority, share of both privatized and state-owned establishments reported that they had increased provision of some essential social services. The country reports interpreted this counter tendency to employers' attempts to supplement falling real value of wages, to favour non-monetary forms of remuneration to avoid high tax rates on money wages (Standing and Zsoldos, 1995), and to the absence of alternate sources of social services due to either the lack of development of private markets (housing, medical insurance) or to the failure of government agencies to take over enterprise facilities (clinics, kindergartens).

• The Chinese enterprises were more likely to expect to increase provision of social services than to decrease them. Almost 30 per cent of the respondents said they planned to increase housing facilities for their workers in the upcoming year, as compared to less than 4 per cent that planned a decrease.
• Nearly half of the Russian respondents had added some benefits over the previous year (45 per cent), the most popular being medical insurance or medical services and subsidised food. The share of privatized establishments that had increased social services was significantly smaller, 28 per cent.
• In Ukraine, only paid vacations and financial assistance for retirement were provided by a larger share of firms in 1995 as compared to 1994. The increased popularity of these employment benefits reflected the growing need to reduce employment. Not counting employment (as opposed to social) benefits, only 40 of the 565 respondents had added any type of social benefit.
• In Armenia, 24 per cent of the privatized establishments had increased provision of some type of social service, compared to 15 per cent of the state enterprises. The most popular service to add was subsidized food and "financial assistance," clear indications of workers' need driving provision of basic services. Only 5 per cent of the respondents had dropped any social service.

Information collected in some of the surveys lends credence to the rationale that enterprises maintained or even increased social services because local government agencies were unable, or unwilling, to take them over.⁽⁵⁾ As shown in figure 6-15, in only Russia did as many as one out of ten respondents report that they had transferred social service facilities to local authorities. In Armenia and Georgia fewer than one out of 25 establishments had done so. There was no uniform pattern between state-owned and privatized establishments: In Georgia and Ukraine a larger portion of state enterprises had transferred social service facilities to local authorities, while in Russia privatized establishments were more likely to have done so, and in Armenia there was no difference.

Figure 6-15 Transfer of social service facilities to local authorities, by property form (p. 93)

Summary Findings

• Enterprise size restructuring would be aided by better unemployment support, both services to displaced workers seeking new jobs and temporary income support. Successful restructuring implies more than adjusting work force levels at the enterprise level, it also requires the development of labour markets through which laid off workers can search for new jobs and businesses can signal their need for worker skills and qualities.
• Labour market flexibility on the part of workers and employers requires new forms of providing housing, medical care, pensions and other essential social services. As long as state agencies are not able to undertake these responsibilities, workers will resist leaving employers and enterprises will continue to struggle with the labour surplus problem inherited from the previous command economy.
• Social protection has been placed in jeopardy by the way in which restructuring has taken place. Displaced workers have lost more than their immediate livelihood, but also guaranteed health care, child care for working parents, holiday facilities, and even their homes.
• Half or more of the survey respondents in each country reported that they owed workers back wages. A less "flexible" wage policy is needed, including enforced mandates for enterprises to pay wages, and to decrease "paying" workers with their own products, and a meaningful wage floor or minimum wage. New institutions may be needed to help otherwise viable establishments, strapped for cash because of overall liquidity problems and inter-enterprise indebtedness in the economy, meet their payrolls.

Figure 6-1. Share of establishments reporting vacancies and recruitment

Figure 6-2. Hires and exits as per cent of current employment

Figure 6-3. Retrenched workers as percentage of job leavers, by property form

Figure 6-4. Use of the public employment services, FSU countries*

Figure 6-5. Recruitment through public employment services, by property form*

Figure 6-6. Employment of workers on part-time basis, by property form

Figure 6-7. Change in the share of part-time workers over previous year

Figure 6-8. Individual performance in wage determination

Fig. 6-9 Pay determination criteria, by property form

Note: Pay criteria is for production worker wages in Armenia, Georgia and Russia and for the performance-based portion of wages in the Kyrgyz Republic, Ukraine and China surveys. Establishment shares sum to more than 100% in Armenia, Georgia and Russia because respondents could select more than one response.

Figure 6-10. Incidence of bonus payments, by property form

Figure 6-11. Bonus payments as percentage of wage earnings, average by property form*

Figure 6-12. Share of establishments reporting wage arrears, by property type

Figure 6-13. Average duration of wage arrears, by property type*

Figure 6-14. Provision of social services, by property form

Note: Housing is for subsidized rent or establisment- owned housing, whichever is higher.

7. Human resources development

This section continues the discussion of labour market development by assessing changes in the occupational structure of employment. It examines the evidence on whether job loss has been constrained primarily to less-skilled occupations or whether managerial and professional occupations have been as severely affected. Second, it looks at the implications of these changes on the need for training and re-training workers, and assesses how investment in human capital has changed during enterprise restructuring and privatization. Finally, it uses occupational wage differentials to assess whether wage rates had begun to take on the role of signalling shifts in labour demand among skill levels or occupations and of giving workers incentive to invest in their own education and training.

7.1 Changes in occupational mix of employmentThe surveyed establishments benefitted from a highly skilled workforce. Skilled production workers accounted for two-thirds or more of total employment covered by the surveys in each of the FSU countries (table 7-1). Unskilled workers made up a relatively small portion of the workforce, from 9 per cent in Ukraine to 15 per cent in the Kyrgyz Republic. In each country, technicians and professional workers, classified as "specialists," accounted for about one tenth of the workforce. The share of skilled workers in the Chinese establishments was lower but still high: 40 per cent of employees were skilled production workers, 29 per cent were unskilled, and 8 per cent were specialists.

Table 7-1 Distribution of employees by occupation

Country	Share of workers by occupational group (in per cent)
	Managers	Specialists	General service	Skilled	Unskilled	Total
Georgia	10.7	11.7	4.6	61.2	11.7	100
Kyrgyzstan	6.6	9.1	2.4	67.3	14.7	100
Russia	8.1	11.8	2.4	65.2	12.5	100
Ukraine	8.2	10.7	2.0	70.1	9.0	100
China	10.7	8.4	7.4	40.3	29.0	95.8
Note: *Total not equal to 100 because 'others' has been excluded from the table. Armenia is excluded due to incomplete data by occupation.

The distribution of production workers between skilled and unskilled had not yet diverged between state enterprises and privatized establishments in most of the FSU countries. It could be expected that privatized establishments would shed lower skilled, and thus less productive, workers at a faster rate than would state enterprises. However, the average share of unskilled workers in privatized establishments was not noticeably lower than in state enterprises, except in the Kyrgyz Republic (figure 7-1). In Georgia, privatized establishments had higher shares of unskilled workers on average than did state enterprises, while in the other countries, privatized establishments had only a slightly lower share of unskilled workers.⁽⁶⁾

Figure 7-1 Average share of workforce in unskilled occupations, by property form (p. 103)

The rate of employment change was fairly constant across occupations in some countries (Ukraine and Kyrgyz Republic) and markedly uneven in others (Russia and Georgia). In Ukraine, the range of job loss across occupations was very narrow, from 6 per cent of managers to 10 per cent of skilled workers (figure 7-2). In Russia, by contrast, employment change ranged from a slight gain by managers to an 18 per cent loss by general service staff. Somewhat surprisingly, in most countries the rates of job loss for skilled and for unskilled workers were similar; but again the exception was Russia where, even more surprisingly, the job loss rate for skilled workers (13 per cent) was twice the rate experienced by unskilled workers (6 per cent).

Figure 7-2 Employment change, by occupation (p. 103)

Occupations' share of employment and of job loss are juxtaposed in figure 7-3. In most countries, occupations' share of job loss were roughly equal to their share of employment. For example, in the Kyrgyz survey, skilled workers accounted for 67 per cent of employment and 69 per cent of job loss; unskilled workers accounted for 15 per cent of both employment and job loss. There were a few notable exceptions to this balance. In Russia, skilled workers lost a disproportionate share of jobs: in 1994 skilled workers accounted for 65 per cent of employment in the surveyed establishments but accounted for nearly 90 per cent of the total job loss over the following year. Clearly, skilled workers have not been sheltered from the size restructuring of establishments; although there seems to be some evidence that managers experienced disproportionately low displacement rates.

Figure 7-3 Distribution of employment and job loss by occupation (p. 104)

Managers were asked about their expectations for changes in the occupational structure of their workplaces in the near future. Respondents in the FSU countries seemed less certain about upcoming changes than did their counterparts in China. Some two thirds of the Chinese enterprise directors expected changes in the occupation distribution of their workforce, while less than one third of the managers in each of the FSU countries anticipated such changes (table 7-2 ). The share of managers that said that they did not know whether the occupational mix of their workforce would be substantially altered was high in Ukraine (18 per cent) and the Kyrgyz Republic (12 per cent). Reluctance to answer may indicate their general uncertainty about business expectations, but may also imply that these business directors were not yet "managing" their human resources to the extent that they could anticipate or plan for changes.

Table 7-2. Expected changes in occupational structure of employment

Country	Percent of establishments that expected change	Percent of establishments that expected no change	Did not know or did not answer	Number
Armenia	28.2	70.3	1.5	330
Georgia	19.7	75.0	5.3	284
Kyrgyzstan	23.3	64.2	12.5	231
Russia	23.7	72.9	3.4	469
Ukraine	28.7	53.4	17.9	543
China	67.8	30.1	2.1	292

Survey respondents were asked which occupations they expected would increase or decrease in the next year.⁽⁷⁾ Even though these answers indicate expectations for the upcoming year, information which may not be as reliable as changes documented over the previous year, their responses indicate that work skills would provide some advantage to workers and job applicants when labour demand picked up. Given the depressed economic conditions, naturally relatively few establishments could expect to increase employment in any category, but the question was designed to elicit expectations about the mix of occupations within the establishment.

With the exception of Ukraine, managers in the FSU countries consistently selected skilled workers as the occupation category most likely to increase in absolute size, selected by 15 per cent or more of the Russian, Armenian and Georgian managers, or in relative size, selected by 8 per cent of the Kyrgyz managers (figure 7-4). The finding that skilled workers would be the first category of worker to be hired by the Russian enterprises seems to contradict the data presented above showing that the job loss rate by skilled workers was nearly twice the rate for unskilled workers. The apparent inconsistency could indicate that the skill composition of the workforce no longer meets the skills needed by many of the surveyed firms.

Figure 7-4 Occupations expected to increase (p. 105)

The finding that skilled workers would be at some advantage as compared to unskilled ones in competing for future jobs is surprising only in that the advantage does not appear larger. To a great extent this can be attributed to the relatively small portion of survey respondents that expected any change at all in occupation mix. For example, only 15 per cent of all the Russian managers surveyed expected to increase the number of skilled workers, but out of the one in five managers that expected any increases, 76 per cent anticipated increasing skilled workers. In the Kyrgyz Republic, 86 per cent of those expecting any change in occupational mix thought skilled production workers would increase relative to other occupations.

Managers of privatized and state enterprises shared very similar expectations of change in occupation mix. Privatized establishments were no more likely than were state enterprises to expect to increase demand for skilled workers or managers, although they were somewhat more likely to expect to increase employment in professional categories. However, the large majority of privatized establishments, like state enterprises, indicated that no significant alteration in occupation mix was expected. (See Appendix table 8 for breakout of data by private and state enterprises).

7.2 Training needs and practicesInvestment in human capital is often waived in short-run survival strategies, or regarded as a non-essential activity during periods of high unemployment and relative ease of hiring workers with specific skills or experience. The case of these transition economies, however, does not exactly meet these expectations. While the survey findings in some of the countries confirm that training activities were provided by relatively few firms, and while none of the surveys provided evidence of training activities gaining in importance, worker training has remained a surprisingly prevalent activity for establishments in most of the countries surveyed.

As shown in table 7-3, there was a large difference across countries in the incidence of training provided by establishments, but hardly any difference between state and privatized establishments within countries. For example, in Ukraine, nine out of ten establishments reported providing some initial training or retraining to workers, but just one out of twelve establishments in the Kyrgyz Republic provided any training to their workers. In three of the countries, China, Russia and Ukraine, more than three-fourths of the surveyed establishments reported providing training to new recruits or retraining to improve the productivity of their current employees. Incidence of training activities was nearly identical between private and state enterprises in each country.

Table 7-3 Incidence of training activities, by property form

Country	Establishments offering either initial training or retraining
	No. of establishments	% of all establishments	% of state-owned establishments	% of privatized establishments
Armenia	153	46.0	50.0	43.8
Georgia	93	32.5	29.3	32.9
Kyrgyzstan	35	15.1	14.1	15.8
Russia	357	75.6	70.8	76.6
Ukraine	517	91.3	92.4	91.1
China	245	83.1	81.7	87.3

As expected, there was more variation in training activities by establishment size than by property form (figure 7-5). In each country, frequency of training activities increased with establishment size. The variation was most pronounced in Georgia, where one fifth of the small establishments compared to four-fifths of the largest establishments provided training. In Ukraine, however, over three-fourths of the small establishments offered some training, as did nearly all of the medium-sized and large establishments. Thus in these transition economies, as elsewhere, large establishments were more likely than smaller ones to offer training. This may be due to greater variation among jobs and more opportunities for advancement in large firms, as well as larger firms' ability to spread fixed costs of training across more workers (such as hiring trainers, providing facilities, or rotating workers to avoid idling machinery).

Figure 7-5 Training activities, by size of establishment (p. 105)

Training activities were reported only slightly more frequently by open than by closed joint-stock companies, and less frequently by other forms of non state-owned establishments, such as joint ventures and private or family-held firms (figure 7-6). As noted, nearly all of the surveyed establishments in the China survey provided training, which may be explained partly by the predominance of very large establishments in the sample survey. Training was provided by about 90 per cent of establishments in each category of property form except labour service enterprises, of which only 60 per cent reported providing some sort of training.

Figure 7-6 Training activities, by disaggregated property form (p. 106)

The survey findings did not clearly indicate a relationship between property form and training - but whether this was due to lack of resources to invest in training, to a lack of perceived usefulness of improving skills, or to a mis-match of skills needed with those of workers or of job applicants, could not be determined from the survey questions. However, there was no higher incidence of training among establishments that reported expecting the number of skilled production workers to rise than among the vast majority of establishments which did not.⁽⁸⁾

Information about the type of training establishments provided was collected in most of the surveys. Establishments provided initial training to new recruits or entry-level workers somewhat more frequently than they provided retraining to improve job performance (figure 7-7). Incidence of a third type of training, to upgrade workers skills in order to promote employees to higher positions, was nearly the same as for improving job performance (ability to be more productive in the same job, or in a different job at the same level). Of the FSU countries, only the Kyrgyz survey collected information on apprenticeship training. One-quarter of the privatized establishments and one-quarter of the state-owned enterprises reported having apprenticeship programmes, but the total number of their apprentices had fallen by 17 per cent from 1994 to 1995.

Figure 7-7 Type of training provided by establishments (p. 106)

The preferred method of improving workers' skills was on-the-job training. Over 80 per cent of the Russian and Ukrainian establishments that provided training to new recruits did so through on-the-job training; the share was over 90 per cent in Armenia and Georgia (data was not collected in the Kyrgyz survey). About 75 per cent of the Russian and Ukrainian establishments that provided retraining did so through on-the-job training, and again the share was over 90 per cent in Armenia and Georgia. In China, the question was asked differently, allowing respondents to list all the ways in which training was provided, rather than asking them to select only the main method of providing training. Training and retraining was provided within the establishment by 83 per cent of the respondents, at the parent company by 54 per cent, and at specialized training institutions by 71 per cent.

Establishments were asked about changes in training activities over the previous year or two. Even in the countries with very high incidence of training activities among establishments, only a small minority reported that they had increased training. About 8 per cent of respondents in Russia and Ukraine reported that they had increased training activities over the previous twelve months. In Russia, almost all the establishments that had increased training were privatized establishments, while in Georgia, state-owned enterprises were more likely to have increased training (table 7-4).

Table 7-4 Share of establishments that had increased training, previous 12-24 months

Country	Establishments that had increased training activities
	No. of establishments	% of all establishments	% of state-owned establishments	% of privatized establishments
Armenia	13	3.9	5.9	2.7
Georgia	7	2.5	6.9	1.4**
Kyrgyzstan	10	4.3	5.6	4.0
Russia	42	8.9	1.5	10.1*
Ukraine	47	8.3	8.2	7.5
Note: There is no relevant data for China Differences in the percentage of privatized establishments versus state enterprises that had increased training were statistically significant in Georgia and Russia.

In the Armenian, Georgian and Russian surveys, managers were asked to explain why training activities had increased or decreased over the past year. More than a handful of answers were gathered only in Russia, where needing a more qualified workforce or responding to changes in technology or products were the most popular reasons given for why training had increased (figure 7-8). Among the larger sample of establishments that had decreased training, a majority of respondents in all three countries attributed less training to downsizing; and if responses of "stabilised workforce" are included, then workforce size adjustment was the motive for decreasing training in two-thirds or more of the respondents (figure 7-9). One out of five respondents in the Armenia and Russia surveys cited high training costs, or no resources with which to pay for training, as their main reason for having decreased training.

Figure 7-8 Why training activities had increased (p. 107)

Figure 7-9 Why training activities had decreased (p. 107)

Training activities were somewhat more frequent among establishments that had adopted new technology over the past year in Russia and Ukraine. Figure 7-10 shows the share of establishments that had adopted a new technology or expanded the range of products produced (the height of the bars) and of these, how many had provided either initial training or retraining (the height of the fill patterns within each bar). In Russia and Ukraine at least four out of five establishments that had adopted new technology had also provided training. In Armenia and Georgia only about half of establishments with new products or technologies had provided training, and in the Kyrgyz Republic only 35 establishments out of the entire survey (15 per cent) had provided any training and only five of those had adopted new technology.

Figure 7-10 Training and adoption of new technology (p. 108)7.3 Occupational wage differentialsSmall wage differentials may have served societies' objectives of maintaining low income disparity in the FSU countries in the past. Typically, top managers earned only three times as much as did unskilled production workers, as set by national wage control. In market economies, wage differentials serve other functions: Wage fluctuation for similar work across industries or regions signals shifts in demand for labour and motivates worker mobility. Wage differentials across skill levels or occupations also show shifts in demand in occupation mix or demand for new skills and give workers incentive to invest in their own training. The evidence on real values of wages and wage arrears was assessed in Section 6 on labour market development. This section compares wage differentials across occupation groups as an indicator of "market" influence on human capital development.

In each establishment, the average wage reported for each occupation was re-calculated as a percentage of the average wage for unskilled workers. These differentials were then averaged across establishments in each country survey, and the results (as shown in figure 7-11) measure the difference between wages paid to workers in higher-skilled occupations and wages paid to unskilled workers. The differentials were fairly uniform across countries, with the average wage paid to skilled workers twice the wage paid to unskilled workers. The return to skill training was smallest in Georgia, where skilled production workers were paid on average only 50 per cent more than unskilled workers.⁽⁹⁾

Figure 7-11 Occupational wage differentials (p. 108)

There was little if any difference in average wages between skilled production workers and specialists (non-managerial professionals). In Armenia and Georgia, wage rates for specialists were essentially the same as average wages for skilled workers; in Ukraine and Russia, the differential for specialists was less than 10 per cent and in the Kyrgyz Republic it was about 15 per cent. Differentials jump sharply for managers; in Russia their average wage was four times that of unskilled workers, in Ukraine and the Kyrgyz Republic it was more than three times higher. (The categories for managers and for specialists grouped enterprise-level and shop-floor levels together.)

Thus the evidence shows that differentials from unskilled production workers to managers remain much narrower than the norm in western market economies. There seems to be a higher return to managerial than to technical training and professional education. This may reflect the scarcity of business know-how in the transition process which pushes up their wages, or it may simply indicate that managers profit from the ability to exercise greater control over salaries, including their own.

The pattern of wage differentials does not appear to have been affected by privatization. Across the five countries, there were only two instances where wage differentials differed significantly between state and private establishments: Higher differentials for skilled workers were found in Armenia (average wage was 2.3 times the average for unskilled workers in privatized establishments but only 1.9 times higher in state enterprises) and lower wage differentials for specialists in Georgia (average wage was 2.0 times the average wage for unskilled workers in state-owned enterprises but only 1.5 times higher in privatized establishments⁽¹⁰⁾). These two isolated incidences indicate that returns to skill levels, or occupation, has not grown in general as a result of change of ownership form.

This conclusion is further supported by longer time comparisons that could be drawn from the Kyrgyz survey. The wage differentials between unskilled workers, skilled workers and managers in 1995 was nearly identical to the differences reported by the same establishments in the first enterprise survey in 1992. There was no evidence that monetary returns to skill acquisition had increased during the four years of enterprise restructuring (Evans-Klock, 1997).

However, it should be noted that the surveys provided no information about other components of income, and thus no conclusions can be drawn about overall changes in income distribution.

Summary findings

• Job loss was fairly proportional to share of work force across occupations. However, evidence on future employment changes indicate that managers expected to increase the share of skilled production workers when demand for product increased.
• Occupational shifts have not been a noticeable part of establishment restructuring in most countries, with the important exception of China. Privatization was not accompanied by increased investment in human resource development as part of survival strategies or long-run plans to improve productivity.
• Training activities have been sharply curtailed in some countries, and establishment support for skill formation seems to be continuing primarily in the form of on-the-job training. There is some evidence that establishments are reducing support for outside training institutions which could have long-run consequences in weakening the institutional basis for worker training.
• The pattern of occupational wage differentials has not changed substantially from the command economy system, nor does the pattern of wage differentials across occupation, appear to have been affected by changes in property form.

Figure 7-1. Average share of workforce in unskilled occupations by property form

Figure 7-2. Employment change, by occupation

Figure 7-3. Distribution of employment and job loss by occupation

Note: Share of employment is for latest year, share of job loss is over the previous year.

Figure 7-4. Occupations expected to increase

Note: See footnote 2 for explanation of differences in information across surveys.

Figure 7-5. Training activities, by size of establishment

Figure 7-6. Training activities, by disaggregated property form

Figure 7-7. Type of training provided by establishments

Note: Data not collected for China or Kyrgyz Republic.

Figure 7-8. Why training activities had increased

Figure 7-9. Why training activities had decreased

Figure 7-10. Training and adoption of new technology

Note: * Includes establishments that had adopted a new technology or had expanded its product range.

Figure 7-11. Occupational wage differentials

>

8. Employment impact of restructuring by age and gender

A special concern during massive enterprise restructuring and downsizing is whether the adjustment burden may fall particularly heavily on specific groups of workers. Certain groups of workers may be more vulnerable than others for reasons unrelated to job performance when managers choose which employees to release, put on leave, train, or hire. The enterprise surveys collected a wide array of data by gender of workers and basic employment statistics for young workers, older workers, and workers with disabilities, in order to examine the distribution of employment effects across these groups.

8.1 Age and disability factors in employmentEach of the surveys provided some descriptive information about the age distribution of the workforce. The definition of young workers was under the age of 25. In most of the surveys, the definition of older workers was above the normal retirement age, which is 55 for women and 60 for men. In the Kyrgyz and China surveys, however, older workers referred to all workers over the age of 50. There are two ways to compare the effect of restructuring and privatization on these two groups of workers. The first is to compare their shares of the current workforce in privatized and state enterprises. The second is to compare their rate of job loss during the restructuring period to the overall job loss rates.

The findings show that privatization itself seems to have had little impact on the age distribution of the work force. Within countries, young workers shares' of the workforce in privatized enterprises was about the same as in state enterprises (table 8-1). The differentials were largest in China and Ukraine, where young workers accounted for a slightly larger share of employment in the privatized establishments. Older workers tended to hold a slightly higher share of jobs in state enterprises than in private establishments (except in Armenia), but again the differences were within two percentage points, slightly wider in Ukraine.

Table 8-1. Distribution of employment by age group and property form (in per cent)

Country	Young workers			Older workers
	All	State	Private	All	State	Private
Armenia	8.7	7.6	9.4	6.2	7.2	5.7
Georgia	10.5	11.6	10.4	5.2	4.2	5.6
Kyrgyzstan	20.2	21.7	19.8	12.3	13.1	11.9
Russia	15.6	15.3	16.0	11.8	11.7	11.6
Ukraine	14.8	12.6	16.8	15.8	17.4	14.8
China	14.6	14.2	17.0	9.2	9.4	7.8

As also shown in table 8-1, there was much larger variation in these groups' shares of employment across countries than between sectors within countries. For example, young Kyrgyz workers held one out of five employment slots, but only one out of ten in Armenia and in Georgia. The share of workers under the age of 25 was about 15 per cent in the other three countries. Similarly, older workers' shares of employment was two or more times higher in Ukraine, Russia and the Kyrgyz Republic than in Armenia and Georgia.

Older workers usually accounted for a smaller portion of the work force than did young workers. Their lowest low share was 5 per cent of employment in Georgia (compared to 11 per cent for youth.) The differential was largest in China, were 9 per cent of workers were over the age of 50 compared to 15 per cent under the age of 24. Employment shares for workers at both ends of the age spectrum were markedly lower in Armenia and Georgia than in the other surveyed countries.

The privatization process itself does not appear to have affected age distribution of the workforce. In most countries the average share of workers under the age of 24 among privatized establishments was higher than among state enterprises (figure 8-1), but the differential was significant only in Russia (14 per cent in privatized establishments compared to 11 per cent on average among state enterprises.) The trend for older workers was even less clear (figure 8-2). In Georgia, China and especially Armenia, their average share of employment was higher among state enterprises but elsewhere their share was slightly lower.⁽¹¹⁾

Figure 8-1. Young workers' average share of establishment employment, by property form (p. 117)

Figure 8-2. Older workers' average share of establishment employment, by property form (p. 117)

Although the comparisons between privatized and state enterprises show no systematic change in age profile of workers, few of the establishment surveys tracked employment by age groups over time, which made it more difficult to judge whether young or older workers have suffered disproportionate job loss during the economic recessions. The analysis of the two rounds of the Ukrainian survey found a slight rise in older workers' share of total employment, attributed to workers of pensionable age needing to remain actively employed (Standing and Zsoldos, 1995). Employment by age groups was tracked over a four-year time span across the two rounds of the Kyrgyz surveys, from 1992 to 1995. As in Ukraine, the job loss by workers over the age of 50, while severe, was proportional to their share of the work force. However, young workers' share of employment had dropped substantially: They had made up one fourth of the work force in 1992 but only one fifth of the work force in 1995 (Evans-Klock, 1998). The data provided by the CIS Statistical Committee showed that one third of the workers officially registered as unemployed in Russia were under the age of 29; in Georgia over half of the unemployed were young workers.

Workers that had acquired disabilities were traditionally retained by their employers. This was the defacto provision of disability insurance. The surveys asked employers what share of their employees had disabilities in order to see whether privatized establishments had a higher tendency to shed these workers. As shown in table 8-2, the proportion of workers classified as disabled (according to official eligibility definitions) ranged from a high of just under 2 per cent in Russia to a low of one fifth of one per cent in Georgia. Only in the Kyrgyz Republic did it appear that privatized establishments were less likely than state enterprises to employ workers with disabilities: 0.5 per cent of employees were so classified in the privatized establishments compared to 1.6 per cent in the state enterprises. Privatized and state enterprises had similar average share of workers with disabilities, except in the Kyrgyz Republic.⁽¹²⁾

Table 8-2. Disabled workers' share of employment, by property form (in per cent)

Country	All Establishment	Ownership Type
		State	Private
Armenia	1.22	0.37	0.59
Georgia	0.21	0.01	0.28
Kyrgyzstan	1.13	1.58	0.47
Russia	1.90	0.71	1.10
Ukraine	1.78	1.68	1.79
China	0.93	0.91	0.91

8.2 Gender factors in employmentMuch more of the data collected in the surveys was stratified by gender, which allows us to look for differential treatment of women across several aspects of employment, beginning by comparing their share of employment and job loss, but also looking for evidence of gender bias in hiring and training decisions.

8.2.1 Change in quantity of jobsAs was pointed out in Section 4, while women accounted for half or more of employment in the surveyed establishments in each country, their share of job loss varied substantially (Table 8-3; See also figure 4-1 above).

• Over the most recent two year period covered by the surveys in Armenia, Russia and Ukraine, women held about half of the jobs at the beginning and had sustained an equal share of the job cuts by the latest reporting period.

• In Georgia, women's share of employment was slightly lower than in the other countries, 48 per cent, but their share of total job loss was nearly 60 per cent.

• Kyrgyz women had also held slightly more than half of the jobs initially (53 per cent) but accounted for nearly two-thirds of the total job loss.
• In China, women had a lower share of employment than in the FSU countries, 42 per cent. (Employment change by gender could not be calculated for China.)

Table 8-3. Women's share of total employment, by property form (in per cent)

Country	All Establishments		State-Owned		Private-Owned
	Previous Year	Latest Year	Previous year	Latest Year	Previous Year	Latest Year
Armenia	51.4	51.5	48.7	48.8	51.0	51.0
Georgia	47.4	45.4	48.3	46.6	47.3	45.0
Kyrgyzstan	54.7	52.9	50.9	45.3	56.3	55.7
Russia	53.9	53.9	50.8	50.2	54.3	54.4
Ukraine	48.3	48.3	41.8	42.4	53.6	52.9
China	-	41.7	-	42.3	-	39.2

In the formerly central planned economies, women's labour force participation rates tended to be higher than in other industrialized countries. As the surveys related employment changes from only the demand side of the market, the extent to which lower shares of employment in some countries were due to women choosing to exit the labour market cannot be determined.

An alternate hypothesis is that with the removal of state control, gender bias may have increased, leading to lower rates of employment for women in the privatized establishments. The findings from the survey evidence are mixed. The above table shows that women's share of employment is much lower in privatized establishments in Ukraine and the Kyrgyz Republic, and slightly lower in China and Russia. But these differences could be due to different distribution of industries in these two groups, and reflect industry and occupation segregation rather than the effects of increased managerial discretion.

More telling information may be gleaned by comparing women's share of retrenchments - the method of employment loss that is most directly tied to management decision. Among establishments that released any workers, women's average share of retrenchments did not vary significantly between state and privatized establishments (figure 8-3). Women appeared to be less likely than men to be retrenched in Kyrgyz enterprises (on average about one third of released workers were women) and more likely to be retrenched in Russia (on average 60 per cent of released workers were women). However, at both ends of the spectrum across countries, women's average share of retrenchments was not significantly higher in privatized establishments. In Georgia, women's share of retrenched workers appeared to be higher in state enterprises than in privatized establishments, but the difference was not statistically significant.

Figure 8-3. Women's average share of retrenched workers, by property form (p. 118)

The survey findings in most countries provide clearer evidence of gender bias in hiring decisions. On average, women were hired for just one out of four new jobs in Georgia and in the Kyrgyz Republic (figure 8-4). In Russia, on average one out of three new jobs went to women. There appeared to be no such bias in China, where women's share of new hires was about the same as their share of overall employment. There was not a consistent story across countries as to whether privatized establishments were more or less likely to hire women. In Armenia, where women's share of overall hires was fairly high, their average share of new jobs in privatized establishments was significantly lower. The findings were reversed in Georgia: women were less likely to be hired in general, but accounted for a higher share of new jobs in the privatized establishments.⁽¹³⁾ There were no significant differences in hiring preferences between privatized and state enterprises in the Kyrgyz Republic and Russia.

Figure 8-4. Women's average share of recruits, by property form (p. 118)

The static comparisons between privatized and state-owned enterprises show that the process of privatization itself has not drastically changed women's share of employment, although hiring trends give cause for concern. However, women's share of hidden unemployment should also be taken into account in assessing employment effects of privatization and restructuring. Data identifying surplus workers and workers placed on leave or reduced work schedules was not collected separately for women. We do know, however, that women were more likely to be among the hidden unemployed because employers relied on extended maternity leaves to reduce surplus labour (encourage quits) or reduce its costs (unpaid months of leave). Furthermore, CIS statistics for 1996 showed that up to three quarters of the workers officially registered as unemployed in Armenia were women. In Russia and Ukraine women accounted for about two thirds of the unemployed.

8.2.2 Change in quality of jobsThe survey findings show that women were less likely than men to receive training from their employers. At best, women's average share of trainees was 40 per cent, in Georgia and Ukraine, and at worst just 10 per cent, in Armenia (figure 8-5). In the Kyrgyz Republic, one out of two employees were women but on average only one out of four trainees were women.⁽¹⁴⁾ Again, the comparisons between property types varied by country. In the Kyrgyz Republic, women's average share of trainees was significantly lower in privatized establishments (22 per cent) than in state enterprises (42 per cent). In Russia, women in the privatized establishments were more likely to participate in training, accounting for 36 per cent of their trainees on average, compared to 26 per cent in state enterprises.⁽¹⁵⁾

Figure 8-5. Women's average share of trainees, by property form (p. 119)

Women's share of employment in each occupation category is shown in table 8-4. Comparing their shares of employment in each of the occupation categories to their share of overall employment (as shown in the final column) shows that women were under represented in management occupations and over-represented in general services and, in a couple countries, among unskilled workers: Women's share of management positions was lowest in Georgia (just 24 per cent compared to 45 per cent of overall employment), and was also 15 percentage points lower in Ukraine. Women appeared to be over-represented in the unskilled labour category in Ukraine (accounting for 62 per cent of these jobs compared to 48 per cent of jobs overall) but not elsewhere. Not surprisingly, women held most of the clerical and other administrative jobs in the general services categories in each country, with the degree of over-representation ranging from 10 percentage points in China to 30 percentage points in Russia.

Table 8-4. Women's share of employment, by occupation (in per cent)

Country	Managers	General service	Specialists	Skilled workers	Unskilled workers	All employees
Georgia	23.5	56.9	55.3	47.5	40.2	45.4
Kyrgyzstan	42.1	66.5	69.0	51.5	51.8	52.9
Russia	43.0	82.7	68.5	50.9	57.3	53.8
Ukraine	31.7	73.6	67.3	44.9	62.4	48.3
China	35.0	51.8	32.0	42.5	41.4	41.7
Note: Armenia is excluded due to incomplete data by occupation.

There were no systematic differences in women's share of occupations between privatized and state establishments. In Georgia, women in the privatized establishments held a higher share of unskilled production jobs on average (41 per cent versus 26 per cent in state enterprises). In Russia, however, women held a higher share of the skilled production jobs on average in the privatized establishments (45 per cent versus 35 per cent in state enterprises). In Ukraine, there were several occupation categories in which women's average share of employment was higher in the privatized establishments: 43 per cent of managers, versus 31 per cent in state enterprises; 73 per cent of specialists, compared to 67 per cent in state enterprises; and 48 per cent of skilled production workers, compared to 40 per cent in state enterprises. In China, the only occupation category in which women's average share of employment differed significantly was unskilled labour, where women accounted for 46 per cent of unskilled labour jobs in state enterprises but 39 per cent in private establishments.⁽¹⁶⁾

These countries have had a fairly equitable distribution of jobs by gender, as evidenced in the overall high shares of skilled labour and specialized training positions held by women. As in industrialized countries, women workers tend to occupy a much higher share of administrative jobs and a much lower share of management jobs. Although occupational segregation is not more pronounced in the privatized sector, women's apparent more limited access to training may jeopardize their future ability to retain a fair share of skilled production, specialized and professional jobs. In particular, their more limited experience in management positions may make self-employment and entrepreneurship more remote re-employment options for displaced women workers than for their male colleagues.⁽¹⁷⁾

8.2.3 DiscriminationGender preferences in managers' employment and training decisions are more credible when revealed through documented practices. However, the questionnaires in Armenia, Georgia and Russia also asked managers to express and explain their preferences for employing men, and also for employing women. Their responses, for both employees in general and production workers in particular, are summarized in figures 8-6 and 8-7.

Managers were far more likely to admit gender preferences for production workers than for all employees, categories which include professional and management positions. This is noted in the Y axis labels as the number of establishments expressing a preference. Three to four times more managers expressed preference for hiring men as production workers than in overall employment; for example 124 Armenian managers gave a reason for preferring male production workers compared to 40 that had gender bias in non-production workers.

The most common rationale for gender bias in favour of men was the product-specific nature of work. This refers to workers' preferences for the type of work, or likelihood of receiving appropriate training, or "socialization" into expecting or accepting certain types of work, and was noted apart from the rationale of the "physical demands" of the job. Only in Armenia did a sizeable share of respondents (24 per cent) justify bias towards men on the basis that men needed work more than women did. From 10 per cent (in Russia ) to 20 per cent (in Armenia) of the respondents said that men had better skills. Worker productivity, quality of work and reliability were relatively infrequent answers for production workers. However, "better skills" was one of the most frequent rationales for bias towards hiring men for professional or managerial positions (from 20 per cent in Armenia to 30 per cent in Russia).

Fewer than half as many managers said they had a preference for hiring women, for either production or professional jobs, as had preferred male employees. The most popular rationale was "the specific product" being produced. This confirms findings presented in several of the country reports of the continued segregation of employment by industry, with women dominating employment in the food and textile/apparel industries. Labour laws appeared inconsequential, as they were cited by only one or two managers in each of the surveys as a rationale for bias in favour of hiring women workers.

Figure 8-6. Main reason for preferring men employees (p. 119)

Figure 8-7. Main reason for preferring women employees (p. 120)

Summary findings

• A static comparison of age distribution of workers between privatized and state enterprises did not reveal evidence of bias against older or younger workers. However, where time trend data was available (from the Kyrgyz survey) there was some evidence that young workers' share of employment had fallen during the lengthy recession.
• Women held about half of the jobs covered in each survey. Their average share of the workforce was if anything usually higher in the privatized establishments - an indicator that they were not discriminated against as managers acquired more latitude in making employment decisions.
• The surveys revealed cause concern about women's future status and opportunities in the labour market. Establishments on average hired women for 20 to 40 per cent of new jobs. Women accounted for 10 to 40 per cent of trainees on average. Only one or two out of every five jobs for which establishments recruited new workers went to women. Women may also account for a disproportionately large share of workers in precarious employment because of the use of extended maternity leaves to reduce labour supply.
• The most obvious indicator of women's disproportionate burden of adjustment costs is their vulnerability to losing their jobs informally through extended maternity leaves.

Figure 8-1. Young workers' average share of employment, by property type*

Note: * Workers under the age of 25.

Figure 8-2. Older workers' average share of employment, by property type*

Note: * For definitions of older workers, see text.

Figure 8-3. Women's average share of retrenched workers, by property form

Figure 8-4. Women's average share of recruits, by property form

Figure 8-5. Women's average share of trainees, by property form

Figure 8-6. Main reason for preferring male employees

Figure 8-7. Main reason for preferring women employees

9. Respondents' main business and employment problems

The preceding sections of this report have documented and described job loss and changes in employment security and remuneration. This section reports how enterprises managers or directors viewed their business problems and prospects, in particularly identifying threats to business survival or constraints on business growth. Their views of the external economic and business climate within which they were operating shows the debilitiating effects so far of market disruption and poorly developed capital and labour markets. Particular attention is paid to the views expressed by managers of joint-stock companies, as stability and growth in the privatized sector is crucial for stemming job losses and creating new employment opportunities in these economies.

9.1 Main business problemsAsked to identify their main business problem, apart from employment difficulties, survey respondents most frequently cited difficulty finding customers, arranging financing (including credit access and customers' arrears), and obtaining raw materials and other inputs (table 9-1).

Table 9-1. Distribution of establishments by main business problem (in per cent)

Country	Finding markets	Raw materials	Financial problems*	Equipment/ technology	High Taxes	Electric energy	Other	Total
Armenia	31.5	37.3	8.2	1.8	8.0	10.7	2.5	100
Georgia	66.9	12.7	4.3	0.4	0.7	14.4	0.6	100
Kyrgyzstan	21.3	40.5	24.1	0.9	9.9	2.2	1.2	100
Russia	29.3	16.9	24.7	0.8	27.1	0.4	0.8	100
Ukraine	33.2	19.7	35.9	0.4	8.9	1.4	0.4	100
China**	21.7	26.8	13.2	5.4	3.7	--	29.2	100
* Responses in this category include high interest rates, inability to access credit, and problems pertaining to financial settlements with customers or suppliers. In China enterprise debt was also included in this category. ** The "main problem" question in the China questionnaire included responses related to employment, which were asked separately in the FSU surveys. These were included in "other" for this table: 12.6 per cent of the respondents identified surplus labour, high wages or lack of skilled workers as their main problem.

Difficulty finding markets for products was most severe in Georgia where 67 per cent of respondents said it was their most pressing business constraint. However, the problem was prevalent everywhere, selected as the main problem by about one third of the respondents in Armenia, Ukraine and Russia. The pressure to find new buyers reflects both the recessionary conditions, particularly evident in Georgia, and loss of traditional customers through the dissolution of the Soviet Union. Non-economical trading patterns under centralized planning had collapsed, leaving directors scrambling to find new buyers.

Difficulties securing reliable supplies of raw materials and intermediate inputs, or their rising prices, can be attributed to the same problem: disruption of supply networks that had formerly spanned the Soviet Union, irrespective of transport costs. This was the most frequently cited problem by Kyrgyz and Armenian manufacturers (41 and 37 per cent, respectively.) These problems were also the top ranked business problem in China.⁽¹⁸⁾

Lack of access to credit, poorly-functioning banks, lack of liquidity resulting in financial settlement problems with suppliers and customers were each identified as aspects of problems in financial markets--selected as the main business problem by one third of the Ukrainian respondents and one quarter of the Russian and Kyrgyz respondents. In the Kyrgyz Republic, respondents most often complained about lack of access to investment capital (19 per cent). In Ukraine, lack of liquidity most often took the form of problems settling payments with customers (21 per cent) or with suppliers (14 per cent). In Russia, 18 per cent cited settlement problems with customers or suppliers and another 6 per cent complained about problems financing investment. All of these forms of financial problems stem from the poor development of financial institutions necessary for a market economy. Liquidity problems are also an outcome of the strict monetary policies followed by governments to bring down triple-digit levels of inflation. This problem, difficulties settling accounts with customers, is another aspect of the liquidity constraint that was noted above in connection with wage arrears to employees.

Tax rates appeared to be a problem primarily in Russia, chosen by over one fourth of the respondents. Taxes were also the main problem of about 10 per cent of Armenian, Kyrgyz and Ukrainian respondents. The rather low frequency could be attributed to the fact that relatively few enterprises had "taxable" profits, or that tax avoidance was easy. Other analyses have criticized high taxation rates on businesses, as inhibiting enterprise and employment growth, and difficulties in collection and tax avoidance, as contributing to government budget deficits and delay in substituting public for enterprise-based social services.⁽¹⁹⁾

In Armenia and Georgia, one tenth or more of the respondents cited difficulties maintaining supplies of electricity as their main problem. These smaller countries had been dependent on other parts of the Soviet Union to provide electrical power and businesses, as well as households, and had suffered repeated outages and rationing.

Compared to these other business survival problems, very few establishments had the luxury of considering updating technology as their principal problem. Problems relating to equipment and technology had the highest response rate in China, but still accounted for only 5 per cent of responses.

The story that emerges from how enterprise directors or managers viewed their own business prospects is one of trying to survive in economies without functioning product and capital markets. Each of the most commonly encountered problems is one aspect of the widespread disruptions in the command economy distribution of inputs and outputs and of the limited development of replacement distribution networks and of supporting financial institutions (see box).

Some of the questionnaires tried to gauge the severity of these business problems by asking whether the enterprise was threatened with bankruptcy in the next year. In both Ukraine and Armenia, one fourth of the respondents perceived bankruptcy as definitely or somewhat likely. In these countries there was no significant difference in the share of managers of privatized versus of state enterprises that viewed bankruptcy as likely. Fear of bankruptcy was less pronounced in Russia (by 15 per cent of respondents) and Georgia (10 per cent), and again state enterprises seemed as likely to sense this outcome as did privatized establishments.

The business problems identified by the managers of the privatized establishments did not differ substantially from those identified by directors of state enterprises. In the cross-country comparison there were relatively few differences in the frequency of difficulties identified as main business problems between privatized and state enterprises, indicating that privatization, or change in property form, had provided scant means by which establishments could respond effectively to the serious industry-wide difficulties (figure 9-1).

There was some indication that finding buyers was more difficult for joint-stock companies than for state enterprises. For example, in Georgia, 80 per cent of the closed joint-stock companies and 73 per cent of the open joint-stock companies cited lack of sales as their principal problem, compared to 58 per cent of state enterprises. The difference between state-owned and privatized establishments was also prominent in Armenia, but not in the other FSU countries. Difficulties collecting payments from customers or finding investment capital were cited somewhat more frequently by the state enterprises in Georgia, Russia and Ukraine.

In China, problems finding customers and materials were the most frequently selected main problem by establishments in each category of property form. Taxes appeared to be particularly burdensome to the labour surplus enterprises.

Figure 9-1. Main business problem, by property form (p. 126)

9.2 Main employment problemsCompared to other types of business problems, employment-related difficulties were low priorities. Indeed, a large portion of respondents in each survey indicated that they had no employment problem, from 17 per cent in Ukraine to 56 per cent in the Kyrgyz Republic (figure 9-2).

Figure 9-2. Distribution of establishments by main employment problem (p. 127)

Relatively few managers cited surplus workers as their main employment difficulty: as few as 4 per cent in Georgia to as many as 16 per cent in Ukraine. Managers were more likely to fault low wages as their main employment problem: 48 per cent in Ukraine, 26 per cent in Russia, and 20 per cent in Georgia. By contrast, at most 5 per cent of respondents (in the Armenia survey) complained of wages being too high. Managers appeared to be complaining that pay levels provided insufficient livelihoods to workers, resulting in low morale and productivity. Lack of skilled workers and difficulty retaining skilled workers was the most frequently identified problem in Russia (29 per cent), Armenia (27 per cent) and the Kyrgyz Republic (16 per cent). It came in a close second in Georgia, selected by 19 per cent of respondents. This problem could also be closely related to poor worker quality or productivity which was cited by 15 per cent of respondents in Armenia, but elsewhere accounted for 5 per cent or fewer responses.

Again, in most countries there was remarkable consistency between privatized and state enterprises in terms of which employment problems were considered most important. However, in Russia, privatized establishments were more apt to regard lack of skilled workers as their primary employment difficulty, while state enterprises were more likely to consider low wages the source of their problems (figure 9-3). In Ukraine, low wages were the overwhelming problem, selected by 60 per cent of the state enterprises and 40 per cent of the privatized establishments. Although the margins were small in some countries, a larger share of privatized than of state establishments selected problems finding or retaining skilled workers as their main employment problem in each survey. As was documented above under training practices, as these economies emerge from recession, unskilled workers are apt to find it more and more difficult to compete successfully in the labour market.

Figure 9-3. Main employment problem, by property form (p. 128)

In China, 13 per cent of respondents identified employment-related problems as their most severe business difficulty. Dealing with surplus wages, increasing wage rates and lack of skilled workers were the most frequent difficulties. Managers of joint ventures were particularly concerned with poor quality of workers. The state enterprises and labour service enterprises were most concerned with dealing with surplus workers, especially with the lack of financial support from government agencies to meet their salaries or help find them jobs elsewhere. Fewer of the privately-owned enterprises (7 per cent) identified labour problems as the most serious difficulty, compared to 17 per cent of the labour surplus enterprises and 14 per cent of state enterprises.

Summary findings

• Survey respondents in each of the FSU countries identified inadequate aggregate demand and disruption in supply chains for their raw materials and other inputs as their principal business problems. Difficulty securing investment capital and payment arrears from buyers, or to suppliers, was also a prevalent problem. Steps to resolve these problems will have to be taken outside of the enterprise, in the area of macroeconomic and financial policy.
• Problems finding, or retaining, sufficient numbers of workers with appropriate skills was the most common employment problem. Efforts to provide a liveable wage were also frequently cited as the most serious problem relating to workers.
• There was a strong similarity of responses to questions concerning both overall business problems and employment problems by managers of privatized establishments and directors of state enterprises. The responses from both sets of managers suggest that policy efforts to improve development of market-enhancing institutions (working and investment capital, trade relations, wage policy, training) would have a substantial impact on their ability to stay in production and maintain employment.

Figure 9-1. Main business problem by property form

Note: Within category answers do not sum 100% because "Other" responses are not shown.

Figure 9-2. Distribution of establishments by main employment problem

Figure 9-3. Main employment problem, by property form

Note: Within category answers do not sum to 100% because "Other" responses are not shown. 10. Conclusions and Policy Recommendations

10.1 Summary of findingsEach of the enterprise surveys documented the substantial extent of property form transformation in the transition economies reviewed. The share of surveyed establishments that had been privatized ranged from a low of 40 per cent in Armenia to a high of 85 per cent in Russia. In the China survey, about half of the surveyed establishments were traditional state-owned enterprises, another 30 per cent were labour service enterprises or village/town enterprises, and 20 per cent were joint ventures or joint-stock companies.

The performance of privatized establishments was compared to that of state-owned establishments on the criteria of sales, employment and continuing surplus labour. Although privatization has been extensive and there is some potential for change in management accountability, most of the establishments remained in precarious position at the time of the surveys. Most reported flat or still declining sales, many continued to rely on barter trade for disposing of their products, and the vast majority were operating at inefficient levels of output.

The surveys also documented a loss of about 10 per cent of employment over the most recent twelve-month period in the Russian, Ukrainian, Armenian and Georgian surveys. Job loss was proportionally highest in the Kyrgyz Republic, where employment in the manufacturers surveyed dropped by nearly 15 per cent. The surveyed establishments in China reported a small net loss.

Despite these continued job losses in the third to fifth year of economic recession and transition for the FSU countries, at least one third of the establishments in each survey reported that they had more workers than were needed for current and expected levels of production. The share of total workforce covered by the surveys considered redundant, and thus in a vulnerable employment situation, ranged from one tenth (in the Kyrgyz Republic) to one quarter (in Armenia).

The relative magnitude of downsizing and of retained surplus workers appears to bear little relation to property form. Government-owned and joint-stock companies in the FSU states reduced employment at similar rates, even when the comparison takes into account differences in initial size, industrial sector, urban location, share of output exported, percentage of workers in unskilled occupations and method of selecting senior management. The "best" record of privatized establishments was in Russia, where although total employment continued to fall, the average decrease in employment was 13 per cent lower than the average for state enterprises. Similar multivariate analysis of labour surplus showed that privatized establishments did not, on average, retain a lower share of surplus workers than did state enterprises in any of the FSU countries surveyed.

Both state and privatized establishments adopted a variety of measures to reduce the costs of retaining surplus workers--such as placing workers on unpaid leave, reduced work schedules, or extended maternity leaves--while waiting for economic conditions to improve or for government policies to release them from the burden of providing income or social services for too many employees. A majority of employers in Armenia, Georgia, Ukraine, and nearly half of the respondents in Russia cited social responsibility to their workers as the main reason for not releasing surplus workers.

Establishments identified depressed product markets and difficulties obtaining raw and intermediate inputs as their most pressing business problems. In several countries, difficulties obtaining credit and receiving payments from buyers were also identified as significant problems. Apart from these overriding constraints, the problem of dealing with labour surplus ranked low among employment problems. The employment problem identified most frequently was lack of skilled workers, or difficulty retaining skilled workers. This was followed in frequency by complaints about low wages. From one fifth of the respondents (in Georgia) to one half (in Ukraine) cited low wages as their main employment problem; by contrast, at most 5 per cent of respondents complained of wages being too high. There was very little difference between privatized and state enterprises in the ranking of business and employment problems within countries.

Inadequate social protection, in particular lack of unemployment insurance and of alternative access to basic social services in the face of severed ties to employers, was found to limit labour mobility. Relatively few cases of enterprises transferring social service facilities to public authorities were reported, and there was no higher frequency of such transfers from privatized establishments. Government agencies' inabilities to fund and manage social service provision confronts establishments with the difficult choice of decreasing the wellbeing of their workforce, and communities, or of accepting the financial drain imposed by maintaining such services themselves.

Occupational shifts had not yet been a noticeable part of establishment restructuring in most countries, with the important exception of China where the non-state sector had lower average portion of unskilled workers. However, in addition to identifying lack of skilled workers as a major employment problem, managers expected that their future workforce would have a larger share of skilled production workers: In Russia, of those managers that expected any employment increase, 76 per cent anticipated increasing demand for skilled workers. In the Kyrgyz Republic, 86 per cent of the respondents expecting any change in occupational mix said that demand for skilled workers would rise. Expectations about the future composition of the work force did not differ materially between privatized and state enterprises, nor did the incidence of training activities at the establishment level.

Job loss was fairly proportional to the distribution of the work force by occupation and by gender. However, in a couple of the smaller economies, Georgia and the Kyrgyz Republic, women accounted for nearly two-thirds of the job loss compared to about half of employment in the initial reporting period. The surveys revealed other areas of concern: Only one or two out of every five jobs for which establishments had recently recruited new workers went to women. If this trend were to continue, women's share of employment could decline drastically over the next decade. Within establishments that provided training to employees, women accounted for only 10 per cent of trainees (in Armenia) to 40 per cent (in Ukraine). Reduced access to skill advancement may put at risk the quality of women's future employment opportunities. Some of the restructuring practices followed by enterprises placed a disproportionately high burden on women, such as extended unpaid maternity leaves, loss of medical care and facilities during pregnancies and for children, and tendency to cut child care centres from employment-based social services.

10.2 Policy recommendationsThe cross-country comparisons between FSU countries (Russia, Ukraine, Georgia, Armenia, and the Kyrgyz Republic) have been drawn in order to contrast the performance of privatized establishments with that of state enterprises, particularly in the outcome of the privatization process on employment. The principal differences between the privatization experience in the FSU countries and that of enterprise restructuring in China are the strong macroeconomic growth that coincided with enterprise restructuring and the more proactive role of public policy to re-deploy redundant workers from state enterprises. These two factors are closely related. Overall growth creates alternative jobs and reduces social disruption stemming from downsizing state enterprises. Growth also improves revenue collection, thus relaxing budget constraints on governments' ability to implement new measures to promote new businesses and job growth.

Enterprise restructuring in the countries of the former Soviet Union has been driven by two imperatives: One, decrease state decision-making (on what is produced, by how many workers and to whom it is delivered) in favour of enterprise-based decision making. Two, decrease state budget deficits by removing loss-making enterprises from government responsibility and gain revenues by selling enterprises to private investors. The latter has been countered by the desire to transfer what were in essence public-owned production facilities back to a wide spectrum of the population rather than sell them to the few persons with sufficient assets to be able to buy them--hence the popularity of voucher-based transfer mechanisms, especially for small and medium-sized enterprises.

This stage of the privatization process is largely accomplished. Policy priorities are now turning to issues beyond those of establishment ownership and governance. The principal policy challenge at this juncture is to improve the environment within which entrepreneurship and market forces can lead to better social consequences. Market-enhancing institutions and policies in the areas of credit and investment, small business promotion, social services and labour market functions are now the preeminent concerns.

In China, the policy priority has been the "commercialization" of industry, meaning improving efficiency and competitiveness of the economy by submitting enterprises to more binding budget constraints (reducing state subsidies), allowing them more discretion in setting output and employment levels, and relating directors' remuneration to enterprise performance. In 1997, top officials in China announced renewed determination to increase market pressures on state enterprises and acknowledged that massive worker displacement will follow. Although China has taken a much more proactive stance towards redeploying workers displaced from state enterprises and towards providing income support, the particular challenge will be to expand programmes throughout the country and to extend performance requirements to more of the state enterprises and to their labour surplus enterprises.

10.2.1 FSU CountriesAnalysis of the findings from the surveys conducted in Armenia, Georgia, the Kyrgyz Republic, Russia and Ukraine points to three principal labour market policies that are needed:

(i) Anti-poverty measures, for workers displaced from enterprises and for workers whose earnings are below the poverty level;

(ii) Improved labour market institutions; and

(iii) Job creation efforts, including macroeconomic policies to spur growth.

(i) Anti-poverty measures

The evidence on labour surplus and lack of substantial job creation underscores the need to improve unemployment benefits to displaced workers, including job search assistance, retraining, and temporary income maintenance. New social safety net measures need to protect both the traditionally disadvantaged and the casualties of restructuring, privatisation, and recession.

The trap of needing tax revenue to fund such services while reducing budget deficits has certainly delayed the process. The World Bank, in a report on the Kyrgyz Republic but which is widely applicable throughout the region, acknowledged that "the sustainability and the pace of reform will depend on how effectively the Government can establish a social safety net, including unemployment compensation, and benefits targeted at the truly needy" (World Bank, 1994).

• Increasing the amount and coverage of unemployment benefits should be a high priority. This is important first of all as a means to decrease the costs borne by individual families for changes that, eventually, are to benefit the entire economy. But it is also needed to improve labour mobility and enterprises' eventual ability to create new jobs.
• Benefits must be indexed to inflation and the capacity of the public employment services to distribute benefits to all eligible workers, especially in rural areas, must be improved.

The growing poverty of employed workers requires a new approach to wage policy, concerning both rates of pay and penalties for non-payment. Real wages must be allowed to rise in order for effective demand to increase and create markets for consumption goods and services that were formerly provided by employers. Policies targeting employers with wage arrears will be effective in some cases, and their implementation is important for signalling support for workers in dire conditions. However, "sticks" may not be sufficient policy instruments since many employers reporting wage arrears are themselves owed payment for goods delivered, especially by their own governments. Governments must set a higher standard for paying inter-enterprise debt by themselves discharging debts owed to now-private enterprises supplying them with goods and services.

Wages must also be allowed to adjust to increase incentives for workers to invest in their own skill development. Income equality objectives should not be discarded, but they must be pursued in ways that do not decrease incentives for productivity and training. In the market economies ahead, unskilled workers are likely to see their standard of living fall precipitously relative to those with better and more modern skills. Rather than maintain wage differentials across occupations at artificially low levels, policy emphasis should be on upgrading workers' skills.

• The minimum wage must be brought up to non-poverty living standard levels. This is important for setting realistic wage rates for unskilled workers and for creating a meaningful benchmark for unemployment insurance benefits. By allowing the minimum wage to sink below subsistence levels governments have lost an important policy tool for combatting poverty throughout the economy.
• To confront growing poverty among the employed population, a "less flexible" wage policy is needed. Demands that enterprises pay current and back wages must be enforced with severe penalties on those that do not. Effective collective bargaining will be required, along with clearer laws and stiffer penalties, to reduce nonpayment of wages. This is the most immediate measure needed to provide minimal levels of protection for workers.
• The feasibility and effectiveness of a revolving credit scheme could be investigated for helping enterprises meet their payroll when they are "selling" their products but not receiving cash payments. The possibility of acquiring low-interest short-term operating capital when urgently needed because of lack of liquidity throughout the economy (e.g. buyers unable to pay for merchandise received because they in turn are owed by their own customers) would help relieve the burden that workers are paying now for mounting inter-enterprise debts. Rather than allow enterprises to accrue wage arrears, or go into bankruptcy, access to a revolving credit scheme would help protect otherwise viable manufacturers from the multiplier effect of inter-enterprise debts. This "carrot" policy would have to be narrowly targeted to avoid abuse by non-viable enterprises.

Enterprises have been encouraged to bring the services they offer to employees more in line with what they can afford. The priority now is to replace comprehensive, employer-based systems of income and job security with social support structures compatible with a market economy. What is needed is not more policy incentives for enterprises to cut social services, but their replacement at the public level.

Enacting such a policy requires a change in emphasis of fiscal policy, and coordination with labour market policy, covering both tax revenue collection methods and expenditure priorities. Loss of inter-Union transfers following the dissolution of the Soviet Union and the decline in domestic tax collection due to recession have resulted in drastic cuts in government expenditures. New government initiatives to confront emerging poverty were originally thought unnecessary because market liberalization and privatization would, after a temporary "shock", lead to faster growth. The pain, in unemployment and income insecurity, could be borne for the duration. The duration has spanned more than half a decade and, moreover, the micro-level data provided by the enterprise surveys revealed more cause for concern than optimism for immediate employment prospects in the manufacturing sector.

Social protection has been placed in jeopardy by the way in which restructuring has taken place. And the absence of alternative "public" social services to replace what was in essence enterprise-based public distribution has contributed to the erosion of much of the social-economic equality that it helped protect. Displaced workers have lost more than their immediate livelihood, but also guaranteed employment-based health care, child care, holiday facilities, and even service housing.

• A social consensus is needed as to what services should be purchased by individuals as consumers from private businesses and what basic services should be made available to all, especially the vulnerable and unemployed. This should be a social consensus and not a decision driven only by immediate exigencies to decrease government spending or for privatized establishments to shed unprofitable service units.
• The institutions for sound fiscal policy must be strengthened in order for public agencies to finance and manage delivery of public services. In particular, revenue collection systems must be implemented to make up for the large share of budgets that had been provided through inter union transfers.

(ii) Improve labour market institutions

Removing some of the social burden from enterprises is also needed to improve labour mobility and thus contribute to enterprises' viability. Both managers and production workers may consider arrangements to "share unemployment" by rotation as more equitable than large retrenchments as it allows more of the workers to maintain access to some medical benefits and to add work months to their pension eligibility. However, targeted assistance programmes, in particular unemployment assistance and provision of basic social services regardless of employment status, would be a more effective strategy to mitigate the poverty effects of continuing restructuring and to speed enterprise adjustment.

The prospects for enterprise growth would be improved by releasing employers of their social obligations to more workers than they can employ productively. If enterprises are to recover and be able to create new jobs, they must be able to redeploy what funds they can accumulate into investment. Their success, in turn, will lead to fewer workers requiring public assistance. It will also enable enterprises, many of which are now largely employee-owned, to deliver a realistic set of benefits and services to workers.

• Labour market flexibility on the part of workers, for example moving into open unemployment from "hidden" unemployment, requires separating employment benefits from the larger sphere of social services.
• Replacement of enterprises as providers of basic social services requires better taxation policy (lower nominal rates, higher collection rates, effective penalties).
• Enterprises will be more likely to give up providing social services to their employees and communities when tax revenue returns to their communities for utilities and transportation infrastructure, clinics and schools and when local government agencies have the capacity to manage them effectively and fairly.

The Public Employment Services will increasingly be called upon to monitor labour market developments. The establishment surveys yield essential information about the demand side of the labour market. And although they are sufficiently comprehensive to allow us to conclude that displaced workers were not finding replacement jobs in manufacturing, it does not tell us whether displaced workers exited the labour market, continued looking for work, or found re-employment in the service sector, petty trade or other "informal" sectors. Without this type of information, establishment surveys alone cannot document the growth of poverty resulting from job displacement.

In practice, the official unemployment rates measure the portion of the urban unemployed who have registered for unemployment benefits, which is far from an accurate count of persons available and willing to work but unable to find remunerative work. Thus they cannot be used as a base against which to measure employment changes because they do not capture the surplus labour (unemployed "hidden" within the enterprise) nor the majority of displaced workers. Higher unemployment figures in the future could be the result of improved data collection or improved benefit distribution, rather than a measure of change in employment and unemployment rates. The use of extended leave and reduced work schedules may be justified by employers as the best they can do to meet their responsibilities to their workers. But these workers are absorbing the brunt of the transition shock without even being counted as displaced and without being eligible for unemployment benefits.

The Public Employment Services are beginning to take on the role of a job information exchange. This is an abrupt change for institutions familiar with directing rather than facilitating labour mobility. By collecting vacancy postings from employers and applications for work from unemployed and underemployed workers, public employment services help workers and employers find each other. According to the survey evidence, such employment services are not available or are used by relatively few establishments.

Finally, the public employment services have also been tasked with the challenge of improving workers' access to training in order to improve their reemployment prospects. The surveys indicated that low-skilled workers were the most likely to lose their jobs and least likely to be re-employed. Difficulties in training their workers and retaining well-trained employees was identified as the main employment difficulty by more managers than was any other aspect of employment problems. Attracting investment in the long-term will require upgrading the skills of the least trained and re-training others in new machinery and technology.

• Labour or household surveys are needed to track displaced workers and should be taken more into account, along with PES registration, in estimating unemployment rates and family poverty. Policy-makers tasked with devising social safety nets need such data in order to form realistic estimates of the costs of providing sufficient unemployment benefits to those in need, to plan training programmes and to formulate programmes targeting young and older workers.
• As the economies begin to recover, efforts to improve the role of public employment services as a job information exchange could benefit both job-seekers and employers by reducing search time and duration of unemployment.
• In all aspects of the work of the PES - distributing benefits, tracking labour markets, providing training - direct measures should be taken to resist erosion of women's status and opportunities in the labour market.

(iii) Job creation

Resource constraints are undoubtedly very tight in these countries, and policy wish lists are easy to create. But re-deployment of displaced unemployed and redundant employed workers will not accelerate without good policies and commitment of adequate funds to implement them. The survey evidence showed unabated rate of job loss among manufacturers, as well as sizeable shares of the workforce considered "surplus" by their employers, beyond the number needed to produce current or expected output, who should be counted as the underemployed or hidden unemployed. Viewing policy interventions as unaffordable luxuries stems in part from ignoring the costs of the current strategy: Workers bear the burden in terms of unemployment and unpaid work.

At this juncture, the priority of public policy must be to leverage the economy out of depression by increasing effective demand and promoting job-creating private domestic and international investment. The principal constraint on job creation and improving work conditions for employees is the recovery and growth of the national and regional economies. The following areas of macroeconomic policy to stimulate growth could be particularly effective in helping enterprises:

• Manufacturers may not be able to take advantage of the potential of further trade liberalization until market failures in the credit and financial sectors are resolved.
• Policies to open new markets and help establishments improve technology or receive injections of foreign capital will need to target specific industries. In some cases, the objective must be to help establishments develop new products because previous production was sustainable only because of subsidized procurement of foreign-sourced materials. In the smaller Republics especially, successful adjustment in the agricultural sector is crucial to prospects in manufacturing.

• Transparency, consistency and fairness of tax policies are as important to business managers as tax rates. Uncertainty in tax liability reduces managers' willingness to commit resources to endeavours which if fruitful will be taxed at unknown rates. Reducing nominal tax rates while increasing efforts to collect taxes would probably increase state revenues by decreasing incentives to cheat. Although "flat" tax rates appear to be "fair" they are actually regressive, i.e. more burdensome to small, startup establishments, and should be replaced with graduated tax rates.

While efforts have succeeded in bringing inflation under control, through controlled monetary policy and reduced government expenditures, tax policy has not yet succeeded in balancing state revenue needs with incentives to start and maintain small businesses. As the large state enterprises permanently reduce the size of their operations, policies are needed to promote development of small and medium-sized manufacturers. The prospects for new, small businesses to acquire startup capital, in economies where such enterprises in the manufacturing sector were previously unknown, are remote. Development of new credit institutions conducive to entrepreneurship and small business growth will be key determinants. Such institutions as rotating credit funds, small-business cooperatives and government subsidized small business loans are all beginning in these transition economies. Because over capacity was seen as a common problem, enterprises must be able to invest in new equipment in order to downsize plants to economically efficient scale, appropriate to new smaller markets.

• Development of new credit institutions conducive to entrepreneurship and small business growth will be a key determinant of the success of programmes to promote small and medium-sized businesses. Examples of such institutions that merit consideration include rotating credit funds, small-business cooperatives, and government-subsidized small business loans.

• Self-employment promotion and small-business management training should target women. The same forces that have increased their vulnerability in the labour market may place them at disadvantage in competing for scarce startup-business credit.
• Management training programmes could yield great benefits by preparing managers of medium-sized companies to attract foreign partners, as investors or customers, and to access credit. Because disrupted supply and market networks were consistently ranked as the most serious business problem, management training should particularly emphasize market information and development.

Public investment in transportation and other infrastructure would both create jobs in the near term and leverage private and foreign investment for future job growth. Public investment in providing basic social services would also create employment opportunities. Public works schemes have a pejorative reputation in the FSU countries. The idea carries a connotation of demoralizing and unproductive "make work". It also has a connotation of grandiose projects, some of which provided great assistance to agriculture or transportation but others turned into environmental disasters. However, the needs are great to restore or modernize transportation infrastructure in order to improve productivity in manufacturing and industry, and a new understanding of appropriate public investment must be encouraged.

• Use of foreign investment or aid in infrastructure development, accompanied by scrutiny against corruption and for maximum job creation, could create near-term employment opportunities and improve the productivity and eventual job creation capability of manufacturers.

• Infrastructure projects could also serve the important objective of bringing economic opportunities to remote regions where economic distress is greatest, by improving transportation to national and regional commercial centres.

10.2.2 China The Chinese strategy has emphasized releasing the constraints of centralized planning on enterprises gradually, rather than by shock, and increasingly subjecting them to market forces. One principal means of this has been through encouraging competition in the form of foreign investment, village/township enterprises and joint-stock companies. Innovative enterprise structures can be effective in improving productivity and creating jobs, within an overall framework of gradual release from centralized planning. This is one lesson drawn from the survey findings from China.

Second, the China experience shows the effect of "commercialization" separately from adjustment to overall economic dislocation, as has been the case in the FSU countries. Enterprises had a better chance of benefiting from the paradigm shift away from centralized decision making because of the healthy GDP growth rate. Enterprises could view the external economy as a source of potential growth of demand, rather than as one more constraint or business difficulty.

One of the clearest differences between the findings from the China survey and those from the FSU countries is that enterprises that were less controlled by centralized governments demonstrated better performance. This finding was not limited to the categories of private ownership and joint ventures with foreign investors, but more importantly was also documented for the village and township enterprises in terms of capacity utilization, employment, and profits. By contrast, the state enterprises and the labour surplus enterprises established to absorb their surplus workers reported continuing labour surplus and heavy social burdens.

The Chinese strategy has been to couple incentives for enterprise reform with providing resources to enable managers to undertake change and encourage workers to be more flexible. Chinese enterprises undertook massive restructuring with more extensive government support for dealing with the problem of labour surplus, including more effective job exchange services from the public employment service, more adequate income support for dislocated workers, and, especially, establishing labour surplus enterprises to simultaneously absorb workers and relieve state enterprises from the burden of carrying too many workers. This type of enterprise has been criticized as unproductive, make-work. And the survey findings did point to the need and desirability to increase productivity requirements, technology and training to improve their performance. However, in contrast to the laissez-faire policy in the FSU states, or allowing workers to go unpaid as part of the hidden unemployed, the policy seems both more equitable and effective.

The country report made several recommendations about how to improve both the effectiveness and coverage of China's restructuring policies (Lim, Sziraczki, and Zhang, 1996). The recommendations emphasized extending the autonomy and accountability that had been granted to the joint ventures and village/township enterprises to the enterprises remaining in the state sector. While withdrawing subsidies from state enterprises, state support to underwrite investment credits and decrease social security requirements for village/township enterprises would enhance their job-creating capacity. The report also acknowledged that demands on the public employment service would increase as it played an active role in redeploying redundant workers and improving mobility between low-growth rural and the high-growth urban centres. The report also called for measures to nurture smaller, independent establishments as a potential source of future job creation and to decrease the heavy burden borne by enterprises to provide a wide range of social services to both productive and surplus labour. The most pressing challenge for China may be the magnitude of increasing job displacement ahead. The question for Chinese policy is whether resources commensurate to the need can be devoted to business investment and to worker redeployment programmes and unemployment insurance.

10.3 Conclusion Enterprise restructuring is not completed with the tasks of retrenchment and reducing surplus labour. Efforts to reduce labour surplus at the level of the enterprise must be matched by national efforts to redeploy retrenched workers, including providing adequate social services and unemployment benefits, resources and incentives for retraining, and the institutional framework to support the creation and growth of small firms. The fact that enterprises retain more workers than they can use productively, make limited recourse to national employment service agencies, and continue to provide expensive social services but to shrinking beneficiaries all underscore the need for public policies to aid restructuring.

Questions about the "affordability" of such measures to leverage economies out of recession must be matched with questions about the affordability of the alternative: stymied growth, increasing poverty and growing inequality, and social unrest. The goal of balanced government budgets must be pursued through efforts to restore economic growth and job creation. It is very difficult in the FSU countries to separate failures due to privatization methods from those caused by general economic crises. In essence, privatization has not been given a fair chance to succeed. The immediate results documented by the surveys are not surprising: employment in manufacturing continues to fall, and the livelihoods of many of the employed workers are also at risk due to the extensive use of various forms of leave time, cuts in social services, and low or unpaid wages.

One purpose of this report, and indeed of conducting the enterprise surveys, was to assess the impact of changes in ownership structure, or property form, on employment. The obvious answer is that no ownership restructuring alone would create sufficient conditions for surmounting the constraints imposed by disruption in input and output markets and the incomplete development of macroeconomic policy conducive to business growth. The business problems identified by business directors or managers through the establishment surveys indicate that the most urgently needed public policies are trade promotion to help rebuild market and supply networks, new credit institutions targeting small and medium-sized businesses, and improving skills training for workers.

References

Boycko, Maxim, Andrei Shleifer and Robert Vishny, 1995; Privatizing Russia, (Cambridge: MIT Press).

Commander, Simon and Richard Jackman, 1997; "Firms and government in the provision of benefits in Russia," in Martin Rein, Barry L. Friedman and Andreas Worgotter, eds, Center for Economic Policy Research, Enterprise and Social Benefits after Communism (London: Cambridge University Press).

Deng Shii, 1995; "The Present situation and tasks of enterprise reform in China," paper presented at the National Seminar on Enterprises Survey Data Analysis in China, organized by the Ministry of Labour, Beijing, 23-25 August 1995.

Earle, John and Saul Estrin, 1996; "Privatization versus competition: Changing enterprise behaviour in Russia," Centre for Economic Performance, London School of Economics and Political Science, Discussion Paper no. 316.

Economist Intelligence Unit, Country Profile of Armenia and Georgia, 1997; Country Profile of the Kyrgyz Republic, 1997; Country Profile of Russia, 1997; Country Profile of Ukraine, 1997; Country Profile of China, 1997.

Estrin, Saul and Robert Stone, 1996: "A taxonomy of mass privatization," Transition, vol 7, number 11-12, November-December, pg. 8-9.

Evans-Klock, Christine, 1998; "The continuing employment crisis in Kyrgyz manufacturing; Analysis of findings from the second round of the Kyrgyz Republic survey of manufacturing establishments, 1993-1995," ILO Department of Employment and Training, Employment and Training Paper.

Hoffer, Frank, 1997; "Traditional trade unions during transition and economic reform in the Russian Federation," ILO, Interdepartmental Action Programme on Privatization, Restructuring and Economic Democracy, Working Paper no. 8.

Hoeven, Rolph Van der and Gyorgy Sziraczki, 1997; Lessons from privatization; Labour issues in developing and transitional countries, (Geneva: ILO).

Honkkila, Juha, 1997; "Privatization, asset distribution and equity in transitional economies," UNU/WIDER, Working Paper No. 125.

ILO, 1997; World Employment 1996/97; National policies in a global context.

ILO, CEET, 1995; Central and Eastern European Team, The Ukrainian Challenge: Reforming Labour Market and Social Policy, (Budapest: Central European University Press in association with ILO-CEET).

Jermakowicz, Wladyslaw W. and Julian Pankow, 1995; "Privatization in the Kyrgyz Republic," Russian and East European Finance and Trade, volume 31, number 6, November-December, pp. 31.72.

Lim, Lin Lean, Gyorgy Sziraczki, and Zhang Xiaojian, 1996; "Economic performance, labour surplus and enterprise responses. Results form the China Enterprise Survey," ILO Department of Employment and Training, Labour Market Papers No. 13.

Martin, Brendan, 1997; "Social and employment consequences of privatization in transition economies: Evidence and guidelines," ILO, Interdepartmental Action Programme on Privatization, Restructuring and Economic Democracy, Working Paper no. 4.

OECD, 1995; Trends and Policies in Privatisation; Vol. II, No. 1.

Republic of Russia, 1991; Law of Russian Federation No. 1032-1 "On Employment of Population" of 19.IV.

Republic of Russia, Statistical Committee of the CIS, 1997; "Social and Economic Situation of Armenia, Georgia, Kyrgyzstan, Russia and Ukraine in 1992-1996," Unpublished manuscript, ILO Department of Employment and Training.

Samorodov, Alexander and Laszlo Zsoldos, (forthcoming); "Economic transformation and enterprise restructuring in Armenian industry; Armenian enterprise labour flexibility survey, 1995," Unpublished manuscript, ILO Department of Employment and Training.

Samorodov, Alexander and Laszlo Zsoldos, 1998; "Economic transformation and enterprise restructuring in Georgian industry; Georgian enterprise labour flexibility survey, 1997," ILO Department of Employment and Training, Employment and Training Papers.

Sokil, Catherine, 1997; "Background report: The employment impact of privatization in selected transition countries," Unpublished manuscript, ILO Department of Employment and Training.

Standing, Guy, 1996; Russian unemployment and enterprise restructuring; Reviving dead souls, (London: Macmillian Press).

Standing, Guy, 1996; "The "Shake-out" in Russian factories: The RLFS fifth round, 1995," ILO Department of Employment and Training, Labour Market Papers no. 14.

Standing, Guy and Laszlo Zsoldos, 1995; "Labour market crisis in Ukrainian industry: The 1995 ULFS," ILO Department of Employment and Training, Labour Market Papers no. 12,.

Sziraczki, Gyorgy and Anthony Twigger, 1995; "Employment policies for transition to a market economy in China", ILO Department of Employment and Training, Labour Market Paper no. 2.

UN, 1997; United Nations Department for Economic and Social Information and Policy Analysis, Report on the World Social Situation, 1997.

UN, 1996; United Nations Economic Commission for Europe, Economic survey of Europe in 1995-1996.

Uvalic, Milica and Daniel Vaughan-Whitehead, 1997; Privatization surprises in transition economies; Employee-ownership in Central and Eastern Europe, (Geneva: ILO).

Windell, Jim, Richard Anker and Gyorgy Sziraczki, 1995; "Kyrgyzstan: Enterprise restructuring and labour shedding in a free-fall economy, 1991-1994," ILO Department of Employment and Training, Labour Market Papers no. 5.

World Bank, 1996; World Development Report 1996, "From plan to market." "br1">

World Bank, 1994; Kyrgyz Republic Economic Report. Report No. 12942-KG, 13 May 1994.

World Bank, 1993; "Armenia. Country Economic Memorandum", Report No. 11274-AM, volume II.

World Bank, 1993; "Ukraine. Country Economic Memorandum", Document. No. 10029-UA, volume I.

World Bank, 1993; "Georgia. Country economic memorandum. From crisis to recovery: A blueprint for reforms" Vol. I, Report No. 11275-GZ, May 1993.

World Bank, 1993; "Ukraine, employment, social protection, and social spending in the transition to a market economy", Report No. 11176-U A, Washington DC .

World Bank, 1993; "Kyrgyzstan: Social Protection in a Reforming Economy," World Bank Country study.

1. The differences in the likelihood of employing part-time workers between private and state enterprises were statistically significant (at the .01 level) in Armenia and China. Data on other forms of flexible employment arrangements, such as short-term employment contracts and shared jobs, was not collected uniformly across surveys.

2. Most governments had instituted or strengthened laws requiring employers to pay workers in the years following the surveys. Beyond highly publicized instances of union protests against non-payment of wages, usually by government agencies, no information is available about enforcement of these laws, nor about their interpretation in the face of bankruptcy of firms with wage arrears.

3. The differences between the shares of privatized and state enterprises reporting wage difficulties were statistically significant only in Armenia (.01 level) and Russia (.05 level).

4. The differences in the average duration of wage arrears between state enterprises and privatized establishments were statistically significant in Georgia and Ukraine (at the .05 level).

5. See Commander and Jackman (1997) study of social service provision in Russia, which explains the complexity of problems in divesting social services and housing provision from commercial enterprises in the absence of market reforms (particularly in housing) and new local government finance systems.

6. The average share of workers in unskilled occupations was 20 per cent in state enterprises and 15.5 per cent in privatized establishments in the Kyrgyz survey, a difference that was statistically significant at the .05 level. Differences in the share of unskilled workers between open and closed joint-stock companies were not statistically significant in any of the countries.

7. The questions on occupational mix were asked in a variety of ways. In Armenia, Georgia and Russia, managers were asked to choose which occupation was most likely to increase or decrease. In Ukraine, managers were asked to give their expectation for each occupation individually- was it expected to increase or decrease. In the Kyrgyz survey, managers were asked which occupation would increase or decrease in relative size, compared to other occupations regardless of absolute growth or decline. Answers to these three methods of eliciting information reveal which occupations may be favoured or at risk within countries, but numerical values are not comparable across countries.

8. The percentage of establishments that provided any type of training to workers was not statistically higher among those that expected to increase skilled workers, in absolute or relative terms, than in those that did not in any of the surveyed countries, and was lower in Russia (.05 significance level in a chi-square test).

9. In each occupation category, wage levels more than 4 standard deviations above the median were treated as outliers and not included in the calculation of average wage rates.

10. In both cases (Armenian skilled workers and Georgian specialists), the difference between state-owned and privatized establishments in average wage differentials was significant at the .11 level.

11. The difference in the average share of young workers between privatized and state enterprises was significant in Russia just below the .1 level; the difference in the average share of older workers was significant in Armenia at the .05 level.

12. The average share of workers with disabilities across state enterprises was 3.5 per cent, compared to 1.3 per cent across the privatized establishments. The difference was statistically significant at the .1 level.

13. The difference between privatized and state enterprises in women's average share of new hires was statistically significant at the .05 level in Armenia and the .1 level in Georgia.

14. Women appear to be over represented among trainees in state-owned Georgian establishments, but the figure may not be representative because only five state enterprises provided training to any employees.

15. The differences between privatized and state enterprises in terms of women's average share of trainees were significant at the .1 level in Russia and the Kyrgyz Republic.

16. The differences in the mean percentages between privatized and state enterprises were significant at the .1 level or better in each of these comparisons.

17. It was not possible to compare remuneration rates by gender because of lack of data for most countries.

18. In the China questionnaire, the proffered list of business problems included a number of employment-related choices, which were excluded from the question in the FSU questionnaires.

19. ILO, World Employment Report 1996/97.

International Labour Organization (ILO): Contact us | Site map |

Copyright and Permissions 1996-2009 International Labour Organization (ILO) - Disclaimer