AN ASSESSMENT OF THE TAX REVISIONS OF THE 1995 BUDGET OF BOTSWANA Astrat Tsegaye Abstract Various tax revisions concerning individual income tax, company tax and the sales tax were featured in the 1995 Budget Proclamation of Botswana. This paper examines these tax revisions and their implications for individuals and businesses. It is concluded that, although the reductions in the individual income tax appear to benefit some individuals more than others, they do not generally erode the progressivity inherent in the tax structure. Consequently, the burden of the tax continues to fall more on higher income groups than on lower income groups. With regard to the company tax concessions, such concessions, which are substantial by any standard, could serve as a complementary factor to other strategies used to encourage investment. Finally, with regard to the sales tax changes, despite the broadening of the tax coverage and the raising of the tax rate applicable to certain products, selectivity of the tax is maintained, with the result that the income distribution effect of the tax remains not regressive overall. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 2 Introduction The 1995 Budget of Botswana, which was presented to the Parliament by the Vice President and the Minister of Finance and Development Planning on the 13th February, 1995, contains fiscal legislation pertaining to taxation. Specifically, some tax revisions have been proposed" ...to generate revenues for government ...to improve the efficiency of resource allocation ..." as well to "...be equitable and simple" (Mogae, 1995 :25). These revisions concern mainly individual income taxation, company taxation, and sales taxation. The objective of this paper is to consider the implications of the tax revisions for individuals, businesses, as well as, of course, the economy at large. As a prelude to this, the paper will initially provide a background information on the tax reform process in Botswana and then focus upon each tax change separately. Background to the Tax Revisions Historically, the tax system of Botswana has encountered various changes which took place at different times. These changes concern primarily the income tax which, at independence, was based on the 1959 Proclamation which closely reflected the South African tax provisions and principles. This feature of the tax, together with the lack of fiscal discretion over indirect taxes due to the Customs Union regulations, suggested that" ...the tax structure was totally foreign and inconsistent with the national objectives of independent Botswana" (Seloilwe, 1982:13). To overcome this situation, the Government of Botswana in 1970 established a Team of Fiscal Experts to study and propose reforms to its tax structure. Following this, a new set of comprehensive Income Tax Legislation was introduced in 1973. The significance of the 1973 Legislation was that it differentiated for tax purposes the treatment of individuals from that of companies. Thus, individuals were subjected to a steep progression of marginal tax rates that rose up to 75% rate, while companies were taxed at a flat rate of30% which was later raised to 35% in 1976 and remained at this level for II years. Since this same Le~is.lation foresaw taxation both as a means of incentive provisions to mdlvlduals, as well as a tool for investment promotions and economic expans~on, it eff~ctively became the basis for all other subsequent legislations (Swami and Taklrambudde, 1988: 1.10-1.11). Thus, following the Income Tax (Amendment) Act of 1987, various modifications were made in the taxation of individuals to simplify and reduce the burden of the personal income tax Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 3 (Mmusi, 1987: 13; Bhuiyan, 1987: 138). To this end, education, insurance and medical allowance entitlements were no longer considered separately but treated as part of a higher and uniform tax exemption threshold of P5000, which was later revised upwards (P6000 in 1989, P9000 in 1990, P 10000 in 1994 and P15000 in 1995). The top marginal tax rate was also reduced tTom 75% in 1973 to 60% in 1981 and 50% in 1987, with each spouse being treated as a single person (Mmusi. 1987: 13). These rates were reduced further to 40% in 1990,35% in 1994 and 30% in 1995. Such successive revisions in the income tax rates of Botswana apparently are in conformity with the generally held view that "economic recovery will be stimulated by reducing disincentives to work and investment if tax rates are reduced; and rate reduction simplifies tax rules and reduces the 'high marginal direct and opportunity costs of both compliance and avoidance'. Accelerating integration of regional and world economies had rendered it difficult for countries to ignore the regional and global trend to reduce rates" (Takirambudde, 1995:389-90). In effect, Botswana is following the regional and world-wide trend to reduce the top marginal income tax rate, and at the same time to flatten out the progressiveness of the rate structure by raising the tax threshold. It is instructive to note that Botswana has now succeeded in resetting its top marginal income tax rate at the lower end of the 30-50% top rate margin or below the 40% rate recommended by the World Bank and IMF missions, respectively (World Bank, 1991:6; World Bank, 1988; Tanzi, 1991:169; Goode, 1993:42; van Rensburg, 1990:4). This is unlike Zimbabwe and South Attica which had set their respective top marginal rates for 1994 at 45% and 40-43%, Swaziland at 39%, while Zambia and Malawi had managed to set theirs at the same rate as that for Botswana in 1994 (Coopers & Lybrand, 1995). Turning to company income tax, this tax has developed over the last 22 years, rising tTom a flat rate of 30% imposed in 1973 to 35% in 1976 and to a maximum rate of 40% in 1987, and then falling down towards the current rate of 25%. The earlier development reflected largely the recommendation of the Taxation Review Committee to reduce the gap between the company rate and the top individual rate and thereby reduce the incentive to engage in tax avoidance manipulation (Mmusi, 1987: 14). It was also thought that the rise in the rate to, for instance, 40% "will be more in line with corresponding rates in the Southern AtTican region than the very low rate of 35% which has remained unchanged for II years" (Mmusi, 1987:14). Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 4 Effective July I, 1990, further changes were made to the company income tax, along with other taxes, ''to follow the established policy trend of neutrality, equity and simplicity" (Mogae, 1990:26). Accordingly, a new company income tax structure was set based on a two-tier system, consisting of a basic company tax and an additional company tax. For the tax years running from 1990 to 1993, the basic rate of company income tax was 30%, and the additional tax was 10%, giving an effective company rate of tax of 40%. It should be noted that, apart from company income taxation, a withholding tax of 15% on dividends paid is applied, but this tax can be used to offset the 10% additional company tax due in the corresponding year. Furthermore, provisions have been made to impose the taxes to companies only, thus excluding the taxation of dividends at the shareholder level (Coopers & Lybrand, 1993 :B.51). Turning fmally to sales taxation, a single-stage sales tax was introduced in September 1982 to be applied to petroleum (gasoline and diesel) and alcoholic beverages (beer, spirits and wine) at the import or manufacturing level. The coverage of the tax was first revised in 1984 to include cider and other brands of alcohol and exclude diesel. As was stated in the Budget Speech for 1988, "the broadening to the sales tax will be done gradually over a number of years, as the country develops the ability to administer effectively, and as revenue requirements dictate" (Mmusi, 1988:24). In line with this commitment, the base was broadened further by the sales tax regulations effective March 1, 1989, when the tax became known as the General Sales Tax which was imposed at 10% ad valorem rate on cigarettes, cigars and tobacco, and at fixed rate on all other products. In the subsequent revisions of the tax carried out in 1992 and 1994, the tax base has been widened to include most consumer goods and maintain the exclusion of basic food stuffs and other essentials namely, medicines, books and stationery. Implications of the Individual Income Tax Changes The 1995 budget proclamation pertaining to the individual income tax states that: a) the top marginal rate will be reduced from 35% to 30% b) the tax threshold will be increased from PI 0,000 to P 15,000, and Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 5 c) the income tax bands will be widened from P12,500 to PI5,000. Before assessing the implications of such changes to the taxpayers, it is useful to look at the previous year's income tax structure. As stated before, the individual income tax in Botswana, as in most other countries, is progressive. Such a progressivity in the tax is achieved by having different tax rates apply to different income brackets, with the top marginal rate lying within or below the World Bank/IMF recommended margins to avoid disincentives. As can be seen from the upper portion of table I, which shows the previous year's tax schedule along with the 1995 schedule, the fIrst PIO,OOOwas free from taxes the next P12,500 was subject to a 5% tax the following three P12,500 were subject to 10%,20% and 30%, respectively. Finally, the excess of over P60,000 was liable to 35% tax. Table 1: Schedule of the Previous and 1995 Individual Income Taxes A. Previous Individual Income Tax Taxable Income But not Tax Payable exceeding PO 10000 NIL 10000 22500 5% of excess over PI 000 22500 35000 P625 + 10% of excess over 22500 35000 47500 P1875 + 20% of excess over 35000 47500 60000 P4375 + 30% of excess over 47500 Over 60000 P8125 + 35% of excess over 60000 Source: Based on Budget Speech, 1994 Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 6 B. 1995 Individual Income Tax Taxable Income But not Tax Payable exceeding PO 15000 NIL 15000 30000 5% of excess over PI5000 30000 45000 P750 + 10% of excess over 30000 45000 60000 P2250 + 20% of excess over 45000 Over 60000 P5250 + 30% of excess over 60000 Source: Based on Budget Speech 1995 In this sense, additional incomes over the tax threshold were subject to marginal tax rates, which rose progressively from 5% to 35%. These rates, however, reflected the burden of the tax on people whose incomes fell under a given income bracket rather than on people earning differential income levels. In order to determine the burden of the tax on people at different income levels, we need to compare the proportion of their incomes which were taken in tax, that is, their average tax rates. This is shown in Table 2 and Figure I. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 7 Table 2: Structure of the Previous and 1995 Income Tax Rates Marginal Tax Tax Payable (P) Average Tax Rate Rate (%) (%) Taxable Previous 1995 Previous 1995 Previous 1995 Income (P) 10,000 ° °° 0 °° ° °° 15,000 5 250 1.67 20,000 5 5 500 250 2.50 1.25 22,500 5 5 625 375 2.78 1.67 25,000 10 5 875 500 3.50 2.00 30,000 10 5 1,375 750 4.58 2.50 35,000 10 10 1,875 1,250 5.36 3.57 40,000 20 10 2,875 1,750 7.19 4.38 45,000 20 10 3,875 2,250 8.61 5.00 47,500 20 20 4,375 2,750 9.21 5.79 50,000 30 20 5,125 3,250 10.25 6.50 55,000 30 20 6,625 4,250 12.05 7.73 60,000 30 20 8,125 5,250 13.54 8.75 65,000 35 30 9,875 6,750 15.19 10.38 70,000 35 30 11,625 8,250 16.61 11.79 80,000 35 30 15,125 11,250 18.91 14.06 90,000 35 30 18,625 14,250 20.69 15.83 100,000 35 30 22,125 17,250 22.]3 17.25 120,000 35 30 29,125 23,250 24.27 19.37 150,000 35 30 39,625 32,250 26.42 21.50 Source: Based on Budget Speeches, 1994 and 1995 Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 Figure 1 Previous and 1IN Average and Marginal Tax Rates (%) 30 ./ 21 -- Previous MTR • . . {I. 1991 MTR / -.- Previous ATR 00 --0-- 1991 ATR 10 I 0 0 0 0 I I 8 I 8 I 0 0 8 0 ~ I 0 ~ ~ 8 I 0 8 8 8 gI I 8 0 to 'PO 'PO 2" ::I to .... a :sf ~ ~ ~ i :g S :S ~ i i 'PO 2" ~ 'PO 'PO Taxable Income (P) 9 Table 2 and Figure I also indicate that the marginal tax rates rose from 5% to 35% as income increased. As a consequence of this, the average rates of tax rose as income increased. This structure of taxation, whereby the average rates rises as income increases is progressive. (The rise in average tax, by necessity, lagged behind that in marginal tax, because the initial income brackets were taxed at a lower rate than the later brackets). If we apply the 1995 individual income tax changes to the previous tax schedule, we obtain the revised tax schedule, shown also in the lower portion of Table I. It is evident that the new tax schedule retains the progressiveness of the tax system, as seen from the point of view of rising marginal tax rates, from 5% to 30%. In order to determine the burden of the revised tax schedule at different income levels, rather than at different income bands, average tax rates have been calculated for different income levels and then graphed, as shown also in Table 2 and Figure I. It is apparent from the inspection of Figure I that the average tax rates exhibit the same growth pattern as before, re-establishing the progressiveness of the tax structure. If the tax system, such as this, is progressive, it means that the burden of the tax falls more on higher income groups than on lower income groups. In order to determine the precise effects of the tax changes on all tax payers, we need information on the percentage distribution of the taxpayers at different income levels during a particular fiscal year. Such information is vital for the purpose of working out the average tax rates and the share of tax charges for different income groups, and then judging whether or not the tax changes are beneficial to all. In the absence of such information, some calculations of both nominal and real benefits at different taxable income levels have been made (see Table 3). The following observations can be deduced from Table 3: a) All taxpayers with income above PIO,OOO benefit from the changes, both in nominal as well as in real terms. b) Some tax payers appear to benefit from the effect of the rise in tax threshold, from PI 0,000 to P 15,000 alone, while others appear to benefit from both the tax threshold effect as well due to the widening of the income tax band. For instance, under the previous tax system, individuals earning taxable incomes of between P15,000 and P22,500 Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 10 Table 3: Taxable Nominal and Real Benefits arisiDl~from t hIT Tax Payable e ncome ~. ax Red ucations 1995 Nominal Benefits Real Benefits' Income (P) Previous 1995 (P) Amount Nominal as % of Amount Realas%of (P) (P) Previous Tax (P) Previous Tax 10,000 0 0 0 0 0 0 10,500 25 0 25 100 16.60 66.40 11,000 50 0 50 100 33.20 66.40 12,000 100 0 100 100 66.40 66.40 12,500 125 0 125 100 83.10 66.50 14,000 200 0 200 100 132.90 66.50 15,000 250 0 250 100 166.90 66.80 19,000 450 200 250 55.60 166.10 36.90 20,000 500 250 250 50.00 166.10 33.20 21,000 550 300 250 45.40 166.10 30.20 22,000 600 350 250 41.70 166.10 27.70 22,500 625 375 250 40.00 166.10 26.60 24,000 775 450 325 41.90 215.90 27.90 25,000 875 500 375 42.90 249.20 28.50 28,000 1,175 650 525 44.70 348.80 29.70 30,000 1,375 750 625 45.50 415.30 30.20 32,000 1,575 950 625 39.70 415.30 26.40 34,000 1,775 1,150 625 35.20 415.30 23.40 35,000 1,875 1,250 625 33.30 415.30 22.10 37,000 2,275 1,450 825 36.30 548.20 24.10 39,000 2,675 1,650 1,025 38.30 68I.IO 25.50 40,000 2,875 1,750 1,125 39.10 747.50 26.00 42,000 3,275 1,950 1,325 40.50 880.40 26.90 44,000 3,675 2,150 1,525 41.50 1,013.30 27.60 45,000 3,875 2,250 1,625 41.90 1,079.70 27.90 47,000 4,275 2,650 1,625 41.90 1,079.70 25.30 47,500 4,375 2,750 1,625 37.10 1,079.70 24.70 48,000 4,525 2,850 1,625 37.00 1,113.00 24.60 49,000 4,825 3,050 1,775 36.80 1,179.40 24.40 50,000 5,125 3,250 1,875 36.60 1,245.80 24.30 52,000 5,725 3,650 2,075 36.20 1,378.70 24.10 55,000 6,625 4,250 2,375 35.80 1,578.10 60,000 8,125 23.80 5,250 2,875 35.40 65,000 9,875 1,910.30 23.50 6,750 3,125 31.60 70,000 11,625 2,076.40 21.00 8,250 3,375 29.00 80,000 15,125 2,242.50 19.30 11,250 3,875 90,000 18,625 25.60 2,574.70 17.00 14,250 4,375 100,000 22,125 23.50 2,907.00 15.60 17,250 4,875 120,000 29,125 22.00 3,239.20 14.60 23,250 5,875 150000 39,625 20.20 3,903.60 13.40 , 32.250 7,375 .. 18.60 4,900.30 12.40 1995.by the Cost of Llvtng Index of" All Items" for March 1995 (Central Statistics Office, as deflated March Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 Figure 2 Nominal and Real Benefits as % of Previous Tax 100 80 10 70 l 80 ---- Nominal aa a % of Previoua Tax I: 30 -D--- Real aa a % of Prevloua Tax 20 10 o Taxable Income (P) 12 opportunities for citizens. As also stated in the Budget Speech of 1995, the tax reduction is intended to "increase retained earnings, making investment and expansion more affordable ...enable them to absorb a larger part of their operating costs so that prices would increase more slowly" (Mogae, 1995:25- 26). Thus, following the Tax Legislations of 1994 and 1995, the effective company tax rate, constituting the basic company tax and an additional company tax, has been reduced from 40% to 35%, and 25%, respectively. Furthermore, for "manufacturing fIrms", specifIcally those that are not engaged in packaging, blending, bottling or assembling, but produce new products, the current effective tax rate is only 15%. This lower rate also applies to the Botswana Meat Commission. In addition, provision has been made to exempt the grants under Financial Assistance Policy effective 1st July, 1995. It is, nearly a universal practice for governments elsewhere in Africa to also provide such and other incentives for investors, particularly to foreign investors. This can be seen from Table 4 which shows the trend in the effective rate of company income tax, as well as various incentives and grants provided in a number of African countries including Botswana. As can be observed from Table 4, there appears to be a global trend for lowering company income tax rate, presumably intended to foster investment. For instance, in 1986, seven of the eleven countries considered, had an average top tax rate of 49%, with Zimbabwe, South Africa, Malawi, Tanzania and Zambia, having rates in excess of the average. By 1994, virtually all had managed to reduce their tax rates to as low as 35%, with Zimbabwe and Swaziland being the only two countries having rates which were above the average. Zimbabwe, however, had set its tax applicable to approved manufacturing projects in growth areas for a period of 5 years at reduced rate of~O% for 1993 and 10% for 1994. Botswana and Senegal, on the other hand, which had the lowest rates of tax of35% and 33.4% respectively in 1986, had the same rates as those for others in 1994. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 13 Table 4 reveals an element of inter-country competition which also exists in the provision of incentives and grants, apart from lower tax rates, to investors particularly to foreign investors. The most common type of incentive is exemption from income and other taxes. This is carried out by granting tax holidays, as in Tanzania, Nigeria, Cote d'Ivoire, Swaziland, and Senegal, and/or tax free grants and investment allowances, as in South Africa, Botswana and Swaziland. Another type of incentive is a package of incentives designed Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 Table 4 Company Income Tu Rates, Incentives & Grants in Selected African Countries Company Tu Rates (%) Country Incentives & Grants 1986 1992 1993 1994 Zimbabwe 52.9 42.5 42.5 40 Sales tax refunds, a tax-free 9% export incentive bonus. 50 48 40 35 Income tax exemption on certain investments, depreciation and depletion allowances, various export South Africa incentives. Malawi 50 35 35 35 Various investment allowances, 12% export allowances. Tanzania 50 40 35 35 Various tax exemptions, depreciation and depletion allowances, tax-holdings. Zambia 50 40 35 35 Reduced tax rates on incomes derived from farming, rural manufacturing and non-traditional exports; job credits. Kenya 45 35 35 35 Various incentives on investment in export processing zones, accelerated depreciation allowances. Nil!eria 45 35 35 35 Various income tax exemptions. Cote D'ivoire 40 35 35 35 Various tax exemptions, accelerated depreciation allowances. Swaziland 37.5 37.5 37.5 37.5 Income tax exemptions, special grants and allowances, tax holidays, deduction for training expenses. Botswana 35 40 40 35 Tax relief grants for certain projects, fmancial assistance for new projects, deductions for approved training expenses. Senel!al 33.4 35 35 35 Various tax exemptions. For Zimbabwe, figures exclude a drought relief levy of 5% imposed effective fiscal year 1993, and 20 % and 10% tax rates for 1993 and 1994 Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 15 to promote exports, especially non-traditional exports, as in Zimbabwe, Malawi, Zambia, Kenya and South Africa. A further type of incentive is the provision of depreciation and capital allowances, as in South Africa, Kenya, Cote d'ivoire and Swaziland. Apart from such incentives, there are some other factors, which are perhaps more important than the tax concessions themselves, to influence foreign investors in their investment decisions in developing countries. These include, the desire to obtain access to the host country's domestic and regional markets, the desire to establish themselves inside tariff or other protective barriers, the nature of the foreign exchange regime, as well as the willingness of the host government to undertake complementary infrastructural investments, like the provision of roads, water and power (Killick and Ayisa, 1986:24-25; Bleijer & Khan, 1984; Chelliah, 1986; Shalizi and Squire, 1988: 15). Given that these factors have been addressed by the Botswana Government, one expects more foreign investment to flow into the country, and therefore the tax concessions stated in the recent Budget Proclamations could act as a complementary factor to enhance this expectation. Implications of the Sales Tax Changes Following the 1995 Budget Speech, the sales tax coverage has been extended to include canned and processed foodstuffs. At the same time, the rate of this tax applicable to the so called "demerit" products, specifically, beer made from malt, wines and ciders, spirits, liquors, tobacco, cigarettes, cheroots and cigarillos has been raised from the standard 10% to 15%. These changes reflect partly the need to minimize the revenue effects of reduced income taxes on the fiscal balance through increased sales taxes, and partly the global trends to move away from dependence on direct taxes to indirect taxation (Mogae, 1995:26). An argument, often advanced against sales taxation, is that it tends to be regressive, in the sense that, since consumption tends to be a declining proportion of increasing' income (i.e. the average propensity to consume becomes lower as income becomes greater), the burden of the tax will fall more on lower income groups than higher income groups. As a result, the application of such a tax on consumer expenditure will offset the progressivity imparted to the tax system through the taxation of individual incomes, and Pula: Botswana Journal of African Studies Vol. 10 No 2,1996 16 therefore should not be imposed. But this argument applies only in a situation where the fonn of the tax is broad based rather than narrow based. The general sales tax of Botswana is a narrow based tax which applies to selected, largely luxury products, and exempts the taxation of certain products, such as food (except canned and processed ones mentioned earlier), medicine, books and stationery, which may be regarded as essential for all. This reduces its regressiveness. This effect is further strengthened by the introduction of a differential and higher rate of tax on specific "demerit" products which are consumed largely by higher income groups. In fact, if it is believed that products falling under sales taxation are consumed more by higher income groups than by lower income groups, then, the incidence of this tax will fall more on the fonner than on the latter. This is quite beneficial from the point of view of achieving vertical equity in the redistribution of incomes, in addition to that which may be achieved through the progressive individual income tax system. Conclusion In general, the tax revisions contained in the 1995 Budget of Botswana are intended to fulfil various objectives. The reductions in the individual income tax seek largely "to improve the disposable income of wage earners". It is observed that while these changes appear to benefit some tax payers more than the others, they do not erode the progressiveness of the income tax structure. As a result, the burden of the tax will continue to fall more on higher income groups than on lower income groups. The reductions in the company taxation, on the other hand, is aimed primarily at stimulating domestic and foreign investment, improving the competitive stand of the nation in the region, and thereby creating more jobs for citizens. Finally, the widening of the sales tax coverage as well as the raising of the tax rate applicable to certain products is meant by and large to establish balance between direct and indirect taxes. In the process, however, such a move will not only raise additional revenue but it will also contribute towards the diversification of the revenue sources since the reliance on a single revenue source, namely mineral revenue, ~akes the revenue system to be vulnerable to external shocks. Pula: Botswana Journal of African Studies Vol. 10 N02, 1996 17 Bibliography Bhuiyan, M.N. 1987 Selected Papers on the Botswana Economy, Gaborone: Bank of Botswana, Research Department. Bleijer, M.1. and Khan, M.S. 1984 "Government Policy and Private Investment in Developing Countries", IMF Staff Papers, 31(2): 379-403. Chelliah, R.J. 1986 Report of the Commission of Inquiry into Taxation, Harare: Government Printer. Coopers & Lybrand 1993 1993 International Tax Summaries: A Guide for Planning and Decisions, New York: John Wiley, Coopers & Lybrand International Tax Network. Coopers & Lybrand 1994 1994 International Tax Summaries: A Guide for Planning and Decisions, New York: John Wiley, Coopers & Lybrand International Tax Network. Coopers & Lybrand 1995 1995 International Tax Summaries: A Guide for Planning and Decisions, New York: John Wiley, Coopers & Lybrand International Tax Network. Goode, R. 1993 "Tax Advice to Developing Countries: An Historical Survey", World Development. 21(1):37-53. Killick,T. and Ayisa, C. 1986 "Some Fiscal Experiences of Post Independent Africa: Lessons from Zimbabwe", Zimbabwe Journal of Economics, 1: 18-35. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 18 Mmusi,P.S. 1987 Budget Speech, Delivered to the National Assembly, Republic of Botswana, ~ February, 1987, Gaborone: Government Printer. Mmusi,P.S. 1988 Budget Speech, Delivered to the National Assembly, Republic of Botswana, 2r February, 1988, Gaborone: Government Printer. Mogae,F.G. 1990 Budget Speech, Delivered to the National Assembly, Republic of Botswana, JCjh February, 1990, Gaborone: Government Printer. Mogae,F.G. 1994 Budget Speech, Delivered to the National Assembly, Republic of Botswana, 7'h February, 1990, Gaborone: Government Printer. Mogae,F.G. 1995 Budget Speech, Delivered to the National Assembly, Republic of Botswana, 13'h February, 1995, Gaborone: Government Printer. Shalizi, Z. and Squire, L. 1988 Tax Policy in Sub-Saharan Africa, Policy, Washington DC: The World Bank, Research Paper Series, (2). Seloilwe, G.S. 1982 Taxation and Inequality in Botswana: A Challenge to Tax Policy in An African Growing Economy: University of Bath, Project for Diploma in Fiscal Studies. Swami, B.N. and Takirambudde, P. 1988 Income Tax Law and Practice in Botswana, Second Trial Edition, Gaborone: University of Botswana. Takirambudde, P. 1995 "Financial, Co-operate and Tax Harmonization for Capital Market Development in Southern Africa", Development Southern Africa, 12(8):379-402. Tanzi, V. 1991 Public Finance in Developing Countries, Aedershot, England: Edward Elgar. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996 19 Van Rensburg, B.P.J. 1990 "Tax Reform Issues in South Africa in the 1990s", South African Journal of Economics, 58(1): 1-20. World Bank, 1991 Lessons of Tax Reform, Washington DC: World Bank. World Bank 1988 World Development Report 1988, New York: Oxford University Press, Washington DC: World Bank. Pula: Botswana Journal of African Studies Vol. 10 No 2, 1996