World Bank, Armenia: Strategies for State Spending, unpublished paper for Consultative Group meeting, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, May 30, 1996).
World Bank, Armenia: Poverty Assessment, unpublished paper for Consultative Group meeting, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, May 30, 1996).
World Bank, Estonia: Staff Appraisal Report, Health Project, Report No. 13297-GE, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, January 19, 1995).
World Bank, Georgia: Staff Appraisal Report, Health Project, Report No. 15069-GE, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, April 2, 1996).
World Bank, Kazakstan: The Impact of Transition on Budgetary Expenditures in Health and Education, unpublished mimeo, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, February 14, 1996).
World Bank, Kyrgyz Republic: Staff Appraisal Report, Health Sector Reform Project, Report No. 15181-KG, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, April 22, 1996).
World Bank, Latvia: Local Government Expenditures and Resource Transfers, Report No. 14470-LV, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, July 20, 1995).
World Bank, Tajikistan: Country Economic Memorandum, Report No. 12692-TJ, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, August 12, 1994).
World Bank, Turkmenistan: Rationalizing the Health Sector, Report No. 14861-TM, (Washington, World Bank, Country Department IV, Europe and Central Asia Region, January 19, 1996).
Data and other information were provided by European II Department desk economists. Comments were given by John Odling-Smee, Vito Tanzi, Sanjeev Gupta, Henri Lorie, Leif Hansen, David Owen, Adrienne Cheasty, Gerd Schwartz, Dennis Jones, Mark De Broeck, Chris Lane, Nita Thacker and Thanos Arvanitis. Research assistance was provided by Sepideh Khazai, Mandana Dehghanian and Alex Keenan.
Excessive capacity and overstaffing can be seen from various ratios. For example, in 1994 Kazakstan had 8.8 students per teacher, while Moldova had 13 students per teacher in 1995; by comparison, Germany, Turkey and the U.K. had ratios of 17:1, 29:1 and 20:1, respectively, in 1994. Ratios of physicians per 1,000 persons in 1994 far exceeded the OECD average of 2.5 in 1993 or ratios of 2.1 and 0.9 in Poland and Turkey (1992), respectively; these included Estonia (3.4 in 1993), Georgia (5.1), Kazakstan (3.7), the Kyrgyz Republic (3.1), Latvia (3.2 in 1993), Moldova (4.0 in 1995), Russia (4.7) and Turkmenistan (3.5). The number of physicians has been declining throughout the region, however, due to emigration and insufficient remuneration. Indicators for hospital beds per 1,000 persons (12.0 in Latvia, 12.2 in Moldova, 11.5 in Turkmenistan in 1994, as compared to an OECD average of 9.3 and ratios of 5.3 in the U.S., 4.3 in Holland and 4.5 in Canada) and inpatient length of stay also exceed those in the OECD and in countries of comparable development (16.2 days in Estonia, 17 days in Latvia, 17.6 days in Moldova and 15 days in Turkmenistan, as compared with 7-9 days in upper income countries). Excessive health staffing may lead to excessive costs and capacity through unnecessary appointments and treatments and prolonged medical stays ordered by underemployed staff.
A recent World Bank report for a health sector project in Georgia stated that in the Soviet period, “there were significant improvements between 1950 and 1970. Infant mortality was greatly reduced, and preventable, curable diseases were rare. However, [in the] mid-1970s…the health status stagnated and started to deteriorate in [the] late 1980s. This dynamic [was] due to emerging economic difficulties, [and] also a result of failing to recognize new health problems, in particular, deteriorating adult health.” Source: “Staff Appraisal Report. Georgia. Health Project.” April 2, 1996 (Report No. 15069-GE).
It should be noted that data from the Baltic states, Russia, and the other countries of the former Soviet Union continue to be characterized by inconsistencies and other problems (e.g., under-reporting, methodological questions), which bear consideration when making firm conclusions about developments.
Real government expenditures are obtained using general GDP deflators, based in turn on consumer price indices (CPI). Use of a more narrowly defined deflator, for example based on prices of services, might indicate a somewhat different path for real expenditures. As service prices have increased more rapidly than general consumer prices throughout the transition period, declines in real expenditures may have been more severe than indicated, while increases in real expenditures in some countries of region (Estonia, Latvia)—as discussed below—may have been less.
For Estonia, the period under consideration is 1992-94 (Table 7); for Latvia, the period is 1993-95. As discussed below, these additional real resources may not be spent in the most cost effective manner if, for example, rationalizing and restructuring of local health and education systems have not yet begun to take place.
Aside from the Caucasus countries, expenditures on health and education in Russia, the Baltic states and the other countries of the former Soviet Union are mostly in the range of 20-30 percent of total expenditures, which, as mentioned, is comparable to the proportion of spending in industrial countries, some East Asian countries (e.g., Korea, Thailand) and in Turkey (Table 2).
Data are not available on health and education spending in Azerbaijan prior to 1995; however, given the 60 percent decline in overall government spending in that country during 1992-95, it appears likely that the pattern of reduced spending in health and education during a period of military conflict—severe difficulties in maintaining tax revenues and possibly, diversion of resources to military spending—also holds for that country.
The data on employment should be interpreted with some caution, as official statistics may capture the number of positions, rather than the number of persons actually employed, i.e., it may be possible for one individual to hold more than one position, in order to receive additional remuneration. In such circumstances, a reduction in the number of staff may not be reflected in a decrease in the number of positions, or alternatively, a program to reduce in the number of positions may not lead to an expected reduction in the number of persons employed.
The labor force in Turkmenistan grew by 7 percent during the same period.
Kazakstan is a peculiar case—as is Moldova, where staffing levels decreased by 22 percent—in that out-migration of ethnic Russians and Germans has sharply reduced the labor force in Kazakstan. The sharp decrease in Moldova appears to reflect in part exclusion of workers in the breakaway Transnistria region.
Wage arrears, which have been significant in the region including in health and education, may impact assessment of real wage developments, as would additional, unreported income earned by health and education workers through side payments, filling of multiple staff positions, or other informal arrangements.
As shown in Table 6, much of the decline in real wages in Lithuania was concentrated in 1992, with real wages growing by over 20 percent since end-1992.
The same pattern seems to have held for Turkmenistan, although data on health and education spending during 1992-93 are somewhat limited, while in the other major energy-producing country of the region, Azerbaijan, real wages appear to have been compressed severely relative to real government spending, again perhaps due to pressure related to military conflict.
As shown in Table 7, health and education employment in Armenia, as well as relative health and education wages, were compressed, so that the wage bill is likely to have decreased as a proportion of total health and education expenditures.
It is possible that real health and education spending decreased by more than total general government expenditures also in Azerbaijan, however, data were available for Azerbaijan only for 1995. In Latvia, expenditures on health and education increased in real terms by somewhat less than total government spending increased.
For Estonia, the comparison is for 1992-94, while for the Kyrgyz Republic and Latvia, the comparison is for 1993-95.
It is difficult to ascertain the relative effects of arrears in the health and education sectors, as both wage and utility arrears are likely to be built up, as workers and utility companies are unlikely to suspend services to hospitals and schools. Installation of gas meters has helped to ease pressures from excessive consumption of gas at higher prices in some countries.
In Moldova, however, vocational education accounted for 6.5 percent of total education spending in 1995 (preliminary) down from 15.3 percent in 1992.
General government expenditures on pre-schools have increased substantially in some countries of the region, reflecting largely enterprise divestiture, while capital expenditures have decreased. By the same token, pre-school enrollment has decreased (as noted in Box 1), due in part to increased unemployment and children remaining at home with parents.
Although provision of private medical and education services has certainly increased, public sector staffing and expenditures have not been adjusted to account for this.
This pattern has been followed somewhat in Kazakstan and Ukraine, where there have been numerical declines in health and education labor forces, although these are more likely due to emigration (Kazakstan) or departure for the private sector, which is less well covered by official statistics (Ukraine), than to retrenchment programs. Emigration was also a factor in the decline in employment in health and education in the Kyrgyz Republic, and increases in relative wages in that country led to increasing instances of wage arrears.
It would be important to make provision for these additional or new expenditures in cost-effective ways. For example, in developing textbooks and teaching materials, the World Bank has suggested the use of materials from other countries, in cases in which locally developed materials are not absolutely required. Greater cost recovery may be sought, for example by requiring purchases of textbooks, meals or materials by some students who do not presently pay, although as noted elsewhere, such cost recovery has already been applied to a great extent in some countries of the region.
This may occur, for example, if there is substantial retrenchment of unproductive staff (e.g., including removal from the employment rolls of any workers with second jobs who still collect state salaries), with increased provision of supplementary materials from a portion of the savings on wages.
A form of private medical service—side payments—has long been common in the health sector throughout the region; the World Bank staff reports that side payments to teachers have now become common. Out-of-pocket payments for drugs are common, covering an estimated 80 percent of drug purchases in Moldovan hospitals and all outpatient drug expenses. Patients in hospitals throughout the former Soviet Union are called on to pay for supplies and drugs—to the extent that they are available—and to bring food and bed linens from home. Total private expenditure on health care has been estimated by the World Bank staff to account for 1.5-2.0 percent of GDP in Moldova and up to 75 percent of all health spending in Georgia. In Georgia, privatization of health facilities is being undertaken, with regional hospitals to remain publicly owned, and other facilities to be sold or auctioned, in some cases with requirements for continued provision of services. In some capitals, the wealthy and well-connected may have access to special clinics or private schools with more substantial resources than public facilities. Such access is only one form of growing inequality in the financing and provision of health and education services among segments of the population and across regions. With local differences in revenue capacity and in the ability to pressure the central government for subventions, previous regional equity may be undermined, without or prior to a reformed system being put in place.
One suggestion is to provide schools and hospitals with global budgets, within which managers could allocate resources as they see fit, as long as their institutions met quality and performance standards and provided agreed services. The state budget would not reimburse losses, but could allow managers to keep savings. In Latvia, for example, local governments have introduced both fee-for-service and capitation financing mechanisms, replacing the former system of budgets and salaries. The fee-for-service system is based on allocation of points according to services, which are pre-assigned a set number of points according to a relative scale of service intensity; a fixed amount of budgetary funds are allocated on a quarterly basis according to the total number of points generated. In the capitation system, family doctors are paid a fee per registered patient and are required to prescribe care—and resources—within the overall fees. In both systems, hospital reimbursement is set on the basis of expected durations for different ailments and per diems; per diems are paid at the 60 percent level when expected durations are exceeded. A bonus is paid for shorter-than-expected-duration stays.
Recognizing that low student/teacher ratios unnecessarily increase educational expenditures directly through the maintenance of too many teachers on the payroll and indirectly by requiring that too many teachers be trained and too many classrooms operated, some countries of the region have attempted to address excessive staffing through changes in sectoral legislation. For example, education regulations in Moldova now stipulate a targeted class size of 20 or 25 students, depending on the level and type of education. Other governments have resisted measures, arguing that low student/teacher ratios reflect the need to post teachers in rural areas, where costs to transport students to “magnet” schools would outweigh benefits from cutting staff.