TVET is expensive—more expensive than general education. Across-the-board reductions in public financing have cut into TVET in Sub-Saharan Africa. This experience over the past decade raises the following questions:
• How have TVET institutions managed in a climate of reduced public financing for skills development?
• Payroll taxes have been introduced to diversify TVET financing in many countries, but how well have they worked?
• What has been the experience with other measures used to diversify the financing of skills development?
Background
Confronted by budgetary pressures, many governments continue to find it difficult to provide adequate, stable financing for public training institu-tions and skills development in general. Pressures on public budgets have forced a search for alternative sources of funding. When funds from tradi-tional state resources are insufficient to cover requirements, two avenues are open: (i) improve the efficiency with which existing resources are used, (that is, pursue more cost-effective training solutions), and (ii) mobilize new sources of funds outside government. The latter has been part of a trend over the past decade in efforts to diversify financing for TVET.
Sources
There are five options for mobilizing additional training resources: payroll levies on employers; tuition and other fees paid by trainees and their fami-lies; production and sale of goods and services by training institutions;
community support and donations; and, indirectly, expansion of non-government training provision. These options for diversification can be used simultaneously.
Financing of TVET is usually based on the principle that those who ben-efit are those who should pay. Those benben-efiting from skills development through increased productivity (employers) and higher incomes (individu-als) have an interest and an incentive to help shoulder the costs of training.
Application of this principle, however, is impeded where poverty and abil-ity to pay are issues and where capital markets are not fully developed for borrowing against future income streams.
Training Levies
The training levy has become a popular tool for augmenting training finance. Earmarking levies on enterprise payrolls is the most widely adopted funding mechanism for training, both in public training institu-tions (usually under the aegis of a national training fund or authority) and in enterprises. These levies are central to training finance policies in many countries (Tanzania, Zimbabwe, Mauritius, South Africa). Other countries are examining the benefits and feasibility of introducing payroll levies to finance training (Botswana, Lesotho, Zambia).
TWO TYPES OF LEVY Discussion of the levy-exemption scheme is reserved for later in this chapter, since it is essentially a transfer mechanism.
Alternatively, the use of a levy to generate revenue produces proceeds that typically are used to finance public training or contract for training pro-cured competitively from government and nongovernment providers. Latin American levies on wages finance public training organizations that are governed by public administrators, workers, and employers. These levies are an important source for diversifying the financing of TVET and have become more popular in Sub-Saharan Africa. (See table 7.1.)
UNIFORMAPPROACH The scope for levying payroll taxes is well estab-lished by international experience. Almost all countries that have intro-duced payroll levies have adopted an approach that sets a standard national levy rate in the range of 1 to 2 percent of company payroll, mostly at the lower end of the range.
SECTORALLEVIES Some countries (for example, Mauritius, Kenya, and South Africa) have introduced sector-level training taxes, usually (but not
Table 7.1. Revenue-Generating Payroll Taxes in Sub-Saharan Africa Training tax as
Country percentage of payroll Used for training?
Gambia n.a. No
Mali 0.5 Yes
Mauritius 1 Yes
Senegal 3 Very little
Tanzania 2 Yes
Zaire 1 Yes
Zambia (planned) 2a n.a.
n.a. not applicable.
a Suggested.
Source: Ziderman 2003, table 6.1, p. 95.
always) based on payrolls. Sales or turnover can also be used as the basis.
The sector levy permits tailoring of the levy format to the characteristics and needs of the sector in question. The narrow focus of sector levies, however, obviates an integrated, national approach to financing and planning skills development and makes it difficult to shift funds from mature to emerging sectors that have different skills needs. South Africa collects a uniform 1 percent levy on enterprise payrolls and distributes 80 percent of that levy to various sectoral boards (SETAs) for use in sector-specific training programs.
ADVANTAGES ANDDISADVANTAGES The strengths and weaknesses of pay-roll taxes need to be weighed against their possible dangers and limitations.
Among their advantages, payroll levies
• Diversify the revenue base for financing training, by mobilizing additional revenues
• Can provide a stable and protected source of funding for national training provision, which is particularly important in the context of unstable national budgets
• Can be viewed as “benefit taxation,” if earmarked for training
• Can serve as a vehicle for cross-subsidization of training, especially from the formal to the informal sector, when collected from formal sector employers.
Limitations of payroll levies include the following:
• Given their particular training needs, many companies, particularly small ones, will not benefit from the scheme. This exclusion breeds resentment and opposition and compromises the status of training levies as “benefit taxation.”
• Earmarked taxation does not conform to the principles of sound pub-lic finance and weakens attempts to unify national tax systems.
• Under fiscal pressure, government may combine training levy pro-ceeds with general public tax revenues. Levy propro-ceeds may be diverted to nontraining uses. This diversion seems to have occurred frequently in Sub-Saharan Africa (for example, in Côte d’Ivoire, Gabon, the Gambia, and Togo).
• Payroll levies may constitute an oversheltered source of funding, leading to unspent surpluses, inefficiencies, and top-heavy bureau-cracies. Two examples of this can be seen in Zimbabwe and Mauritius.
The Zimbabwe Manpower Development Fund generated a surplus of 28 percent, which was invested in property. Property assets tripled even though property ownership was not in the mandate of the fund.
• Payroll levies raise the cost of labor to the employer, possibly dis-couraging employment.
• Employers may shift the incidence of the levy onto workers in the form of lowered wages. In this case, workers would bear the burden of the tax.
Issues
Issues affecting the use of payroll taxes include cross-subsidization, applica-tions of the funds, and choices in design and implementation.
CROSS-SUBSIDIZATION Payroll levies can be used to finance training out-side the enterprises being taxed. This is called cross-subsidization. Twenty percent of the South African levy is earmarked for use in training outside sectors that pay the levy. In Côte d’Ivoire, the Vocational Training Develop-ment Fund allocated between 18 and 20 percent of its budget (financed through a 1.2 percent payroll tax on formal sector enterprises) for training that benefits the informal economy (Johanson 2001, pp. 2, 3, 8). This cross-subsidization financed training in areas where informal enterprises could not raise their own funds. It contradicts the principle of beneficiary financ-ing and succeeded only because formal sector enterprises cooperated. Such cooperation can appear where enterprises are seeking to improve the capac-ity of suppliers of intermediate goods and services.
APPLICATION OF THELEVY PROCEEDS In some countries, the levy pro-ceeds finance exclusively or mainly the training centers run by the financing authority. Tanzania uses levy proceeds to finance its own centers, with 4,000 graduates a year, while other providers—with 60,000 graduates—receive little or no financing. In part this is because the levy proceeds are limited, but it also indicates the preferential treatment given to institutions run by the Tanzanian VETA (Atchoarena and Delluc 2001, p. 269). Similarly, in Mauritius, as a priority, the levy finances the centers owned and operated by the IVTB. This policy is in the process of being changed.
DESIGN ANDIMPLEMENTATION Different considerations affect the design and implementation of payroll levies:
• Levy rate.Levy rates should be subject by law to periodic review to avoid the accumulation of surpluses, as has happened in Zimbabwe and Mauritius.
• National or sectoral levy rates. A standard, national payroll levy is pre-ferred to a sector levy for its greater ability to permit funds to be allo-cated where training needs are greatest.
• Sectoral coverage.Levy coverage should be as wide as possible across economic sectors and should include public enterprises.
• Company size.Very small companies can be exempted from levy pay-ment, on both efficiency and equity grounds (they tend not to cap-ture the benefits).
• Levy collection.Levy collection is best placed in the hands of effective agents; self-collection by a funding organization should be avoided.
In Tanzania, collection by the social security administration did not work properly. Collection was recently transferred to the national
revenue collection agency, and revenues subsequently increased dramatically.
• Security of levy proceeds.Special attention needs to be given to guard-ing levy revenues from raidguard-ing by the government (especially where tax authorities act as the collection agent), by placing the proceeds in special, closed accounts. This has worked successfully in Côte d’Ivoire despite persistent efforts to divert the proceeds to other purposes.
• Employer buy-in.Employers need to be involved in the formulation and execution of payroll levy policy. The introduction of a levy is usually controversial and so is bound to face difficulties without employer support, as was the case in Tanzania (Atchoarena and Del-luc 2001, p. 18).
• Premature introduction. Payroll levies may be inappropriate where levy income-generating capacity is weak—because of either the lim-ited size of the formal sector or the administrative and organizational difficulties of levy collection. In Zambia, for example, a levy was adopted but never implemented.
Overall, the record of success of training levies in Sub-Saharan Africa is lower than in other regions. Côte d’Ivoire is considered a success, but there are few other unambiguous examples. The expectation that levy income complements other government financing and provides additional financ-ing for trainfinanc-ing has not always been realized in practice. In Tanzania, levy income completely displaced government spending on formal skills devel-opment. In other countries (the Gambia, Togo), earmarked training taxes were absorbed into general government revenues instead of being used to finance public training. In some cases, funds have proven excessive, leading to an inefficient use of resources (Zimbabwe and Mauritius, for a time.) Finally, the scope for establishing enterprise levies is limited by the small size of the modern sector in many countries of the region.
Raising Revenues at the Level of the Training Institution
Financial pressures over the past decade have forced almost all countries to promote income-generating activities at the institutional level (Atchoarena and Delluc 2001, p. 1). Income generation generally takes two forms—fee income and sale of goods and services—but it may also include donations and contributions from the local community. Madagascar is characterized by relatively low dependence on state subsidies (47 percent and 66 percent at lycées and vocational education centers, respectively). Fee income accounts for 12 percent of income at lycées and 14 percent at vocational edu-cation centers. Voedu-cational eduedu-cation centers receive 21 percent of their rev-enues from the sale of products and services (Atchoarena and Delluc 2001, pp. 93–94). Most institutions depend on multiple sources of revenue for their operations, as in Tanzania (table 7.2).
Donations, particularly from abroad, can be important sources of rev-enue to tide training institutions over rough economic periods. Nongovern-ment institutions naturally are reluctant to disclose such sources of support.
Cost-Sharing through Fees
Fees for training have been introduced or raised in many countries as public financing declines.1 Countries including Malawi, Madagascar, Mauritius, Nigeria, Tanzania, Zambia, and Zimbabwe have introduced fees for train-ing courses.2 The rationale for charging fees is beneficiary financing:
Trainees can be expected to bear at least a part of the cost of training when skills acquisition is seen as an investment in human capital, with the promise of higher labor market earnings and a greater probability of sus-tained employment.
SHARERECOVERED Fees are often set at a symbolic level only. For exam-ple, fee income accruing to the IVTB in Mauritius accounts for only between 1 and 2 percent of revenue. In some cases, however, fees are more substan-tial but still far from sufficient for full cost recovery. In public training cen-ters in Tanzania, fees represent about 15 percent of recurrent costs; in Madagascar, 27 percent. Cost sharing has also begun in formal apprentice-ship training in Kenya. One auto mechanics company has gradually intro-duced a policy of charging apprentices for training. By 1993, all apprentices were paying KShs 80,000 (about $1,200), but this rate was still nowhere close Table 7.2. Tanzania: Sources of Incomes and Training Costs, Selected
Church-Owned Training Centers
Training Training Sources of income costs/year fees/year (percentage of total)
Institution (TSh 000) (TSh 000) Fees Production Donations Other
Kasasha Village TC n.a. n.a. 60 28 10 2
Kalwande VTC 734 70 7 62 — 31a
Hai VTC 375 150 42 n.a. n.a. n.a.
Kilimanjaro YTTC 270 154 61 ¨ææææææææ39ææææÆ
Don Bosco VTC n.a. 65 n.a. — — n.a.
Mafinga Lutheran VTC 450 100 20 n.a. n.a. —
Tushikamane VTC 330 50 15 50 3 32a
Kisa Homecraft Center 240 100 42 32 16 10b
n.a. not available.
a. Business activities unrelated to training undertaken for income generation.
b. Income and distributions from the diocese.
Notes:TC, training college; VTC, vocational training center; YTTC, youth technical training center.
Source:Haan 2001, p. 88.
to the company’s total investment per apprentice in training (Grierson 2002, p. 18).
APPRENTICESHIP FEES Collection of fees from traditional apprentice-ship training in the informal economy is the norm in many countries. In Yaoundé (Cameroon) 70 percent of the informal sector entrepreneurs indi-cated they paid for their skills training (Haan and Serriere 2002, p. 115). In Ghana, one study of traditional apprenticeship found divergent practices of cost-recovery. About 60 percent of the enterprises charged apprentices (averaging $70 for up to three years of training) and 40 percent of the firms did not. The latter tended to be larger enterprises that expected compensa-tion in the form of work for reduced wages. The apprenticeship fee is usu-ally paid up front at the start of training; however, in some cases a down payment is made and some 20 percent to 40 percent of the fees are paid at the end of training. The deferment of payment gives the apprentices some security that they will in fact receive the training to which they are entitled.
When the employer prefinances the skills training and expects to recover these costs through lower wages, the apprentices have incentives to leave the firm before repaying. This “incentive” can be countered by paying sub-stantially higher wages in the postapprenticeship period as an inducement for graduates to stay (Haan and Serriere 2002, p. 34).
TUITIONFEES Compared with practices related to training levies, prac-tices related to tuition are much less uniform. The feasibility of setting fees in relation to unit costs of training is a function of diverse factors that vary from location to location with countries. They include the type and costs of the training, the price elasticity of trainee demand for different training courses, political constraints, and policies pertaining to equity of access.
Issues in Charging Fees
Deciding whether to charge for training involves more than an analysis of the break-even points: At what price will fees scare off students? What will happen to students who cannot afford to pay? This is a Catch-22, a perverse cycle. The scope for charging additional fees may be limited, and attention to the capacity of trainees to pay is important. Some critics contend that training fees have about reached their maximum. However, in the absence of fee increases, institutions may not be able to invest in inputs to raise the quality of training; quality will deteriorate, and fewer trainees will enroll, further reducing fee income. Tanzania cut its contributions to the Folk Development Colleges in 1994, limiting them to cover only salary costs. The colleges introduced fees, but their revenue from fees was inadequate to pur-chase new tools and equipment. College management is unable to invest in and improve training quality, without which their underutilization will increase further, thereby further lowering fee income.
CENTRALLY OR LOCALLY ESTABLISHED A central issue in fee policy is whether to impose standard, nationwide compulsory fees or to allow indi-vidual training institutions to set and vary their fees. In Tanzania, fees for public training institutions are set centrally. Institutional autonomy in fee setting seems the best approach, but it may not be feasible in otherwise cen-tralized training systems. Standard, compulsory fee-setting may be an inflexible tool, unlikely to reflect local market realities, but it is generally acceptable as a second-best measure. Because of the various factors involved, no prescriptions can be written for tuition fees.
EQUITYIMPLICATIONS The positive financial benefits from cost-recovery need to be weighed against the potentially adverse effects on equity. Here the tradeoff is clear. Higher, realistic fees may exclude from training those who cannot afford to pay, while comfortably low fees may not contribute enough for the provider to recover costs. Negative impacts on access to training opportunities for the poor, minorities, rural populations, and other disadvantaged groups are likely to ensue. The level of fees can also have an impact on dropout rates. The case of two Zambian community-based trade schools is illustrative (box 7.1). Since training fees were increased, the very underprivileged can no longer afford training (Haan 2001, p. 176).
Governments can offset the adverse impact of fees on equity by using some of the savings realized from fee income to provide targeted scholar-ships to low income groups.3 Theoretically, at least, increased fees could lead to increased equity of access because with the savings the government can afford to finance the enrollment of more low-income students. The weakness of this argument is the limited ability of most trainees in Sub-Saharan Africa to afford increased fees.
The equity implications of charging fees underscore the widely recog-nized need to introduce subsidies targeted to at-risk groups, in the form of scholarships and fee discounts. However, targeting the most needy individ-uals within these groups is not easy, particularly in Sub-Saharan Africa.
USE OF FEES Local institutional initiative in generating income from fees will be stunted if this income does not contribute to institutional bud-gets. In Zambia, vocational training institutes were unable to retain fee income (World Bank 2001). In Namibia, the public Vocational Training Cen-ters are similarly unable to keep any income generated by the institutions.
Such income must be transferred to the Treasury. In Kenya, the manage-ment of the Departmanage-ment of Industrial Training is considering advertising courses and charging fees to any member of the public who wishes to enroll.
However, current legislation does not allow the department to retain the fee income for refurbishing and essential maintenance (Grierson 2002, p. 20).
A public institution can also earn fee income from expanding fee-paying enrollments where demand exists. In Tanzania, the VETA’s institutions are not allowed to increase fees for individual trainees, but they can increase fee income by expanding enrollments, including adding a second session. In
Mali, the establishment of Units for Training and Support to Enterprises permitted the training centers to generate additional income through con-tinuing education (Atchoarena and Delluc 2001, p. 102). An appropriate share of the proceeds from extra (evening) classes is expected to accrue to the training institution, for reinvestment in facilities and equipment. Sene-gal also provides an interesting example (box 7.2).
In the context of high social demand, there is no guarantee that the courses correspond to market needs. Parents or trainees may be willing to pay the fees to obtain credentials regardless of the market relevance of the
Box 7.1. Zambia: A Tale of Two Community-Based Trade Schools and Their Fee Policies
The Chilenje and Dzithandizeni Trade Schools in Lusaka are interesting exam-ples of nongovernment community-based training centers. Founded in the late 1970s, both received small but crucial contributions in the form of techni-cal equipment and financial assistance from international donors until a few years ago. Both trade schools are essentially self-sustainable in terms of oper-ating costs. Initially, they provided training free of charge but gradually intro-duced elements of cost-sharing. First they asked students to pay for their own training materials and later raised the fees from a token commitment to a level of substantial cost-recovery. Chilenje Trade School has gone farther in this direction than Dzithandizeni. Chilenju charges ZK 20,000 ($7) per month for its courses ($126 for the 18-month course). Dzithandizeni is asking only ZK 125,000 ($40) for its 18-month training period, against an estimated actual training cost of ZK 1.3 million ($433) for one trainee over 18 months. This means that the training fees paid cover more than 25 percent of the total costs at Chilenje and less than 10 percent of the total costs at Dzithandizeni.
Dzithandizeni Trade School makes some additional money by letting informal sector operators use the center’s equipment (for ZK 1,000 per hour). The school’s graduates get a 50 percent reduction in the equipment fee.
All this means that training is still funded by the trade schools’ produc-tion units. Both have offered training-cum-producproduc-tion from the beginning and have been successful with it, partly through the appealing designs that volun-teers brought in from Europe. With the opening of the Zambian economy, a number of new and more modern furniture shops have opened up in Lusaka and competition has increased markedly.
In Dzithandizeni, the dropout rate is low (5 percent, mainly for nonfinan-cial reasons) although a few years ago when it was decided that the (tailoring) students had to buy their own materials, a number of the students left. The markedly higher training fees in Chilenje have led to substantially higher dropout rates (up to 50 percent). The trainees are said to reason that after they have obtained sufficient basic skills to use the equipment after a few months, there is no further need to spend ZK 20,000 per month to continue their training.
Source: Haan 2001, pp. 127–31.