Discussion Paper No. 2001/97 Taxation and Fiscal Reform in Ghana Tony Addison1 and Robert Osei2 September 2001 Abstract To finance basic pro-poor services, the Government of Ghana must mobilize more public revenue. But tax reform has been highly controversial in Ghana. An attempt to introduce VAT in 1995 failed after widespread protests. Although a second attempt to introduce VAT in 1998 succeeded, strong resistance to tax reform remains, and the total tax base remains narrow and over-dependent on petroleum taxes. This paper argues that slow progress in public expenditure reform, in particular in raising spending on basic services, has impeded the public's willingness to pay taxes, since many people see little benefit. This is especially the case in the poorer regions, which in addition have borne the brunt of the rising petroleum tax. The paper concludes that much more attention must be given to the political economy of fiscal policy in Ghana, if lasting improvements to the fiscal system are to be achieved. Keywords: fiscal policy, taxation, Ghana, sub-Saharan Africa JEL classification: O23, O55, H20 Copyright ã UNU/WIDER 2001 1 World Institute for Development Economics Research of the United Nations University (UNU/WIDER), Katajanokanlaituri 6 B, FIN-00160 Helsinki, Finland; addison@wider.unu.edu 2 University of Nottingham, UK; robert.osei@nottingham.ac.uk This study has been prepared within the UNU/WIDER project on New Fiscal Policies for Growth and Poverty Reduction which is directed by Tony Addison. UNU/WIDER gratefully acknowledges the financial contribution to the project by the Government of Italy (Directorate General for Development Cooperation). Acknowledgements Paper presented at a UNU/WIDER project meeting on 'New Fiscal Policies for Growth and Poverty Reduction', Helsinki, 17-18 November 2000. This project focuses on public expenditure management, taxation, and the macro-economic effects of fiscal policy in mainly (but not exclusively) low-income countries. Further details of the project can be obtained from WIDER's web site (www.wider.unu.edu) or from the project director, Tony Addison (addison@wider.unu.edu). We thank participants at the project meeting for their comments. This paper represents work in progress. Further comments are welcome. UNU World Institute for Development Economics Research (UNU/WIDER) was established by the United Nations University as its first research and training centre and started work in Helsinki, Finland in 1985. The purpose of the Institute is to undertake applied research and policy analysis on structural changes affecting the developing and transitional economies, to provide a forum for the advocacy of policies leading to robust, equitable and environmentally sustainable growth, and to promote capacity strengthening and training in the field of economic and social policy making. Its work is carried out by staff researchers and visiting scholars in Helsinki and through networks of collaborating scholars and institutions around the world. UNU World Institute for Development Economics Research (UNU/WIDER) Katajanokanlaituri 6 B, 00160 Helsinki, Finland Camera-ready typescript prepared by Adam Swallow at UNU/WIDER Printed at UNU/WIDER, Helsinki The views expressed in this publication are those of the author(s). Publication does not imply endorsement by the Institute or the United Nations University, nor by the programme/project sponsors, of any of the views expressed. ISSN 1609-5774 ISBN 92-9190-016-8 (printed publication) ISBN 92-9190-017-6 (internet publication) Introduction The design and implementation of public-expenditure management and taxation systems is one of the most important tasks facing low-income countries. This is especially the case in sub-Saharan Africa, where weaknesses in fiscal policy have contributed to slow growth and limited human development. Yet improvements in public expenditure management and taxation remain fraught with difficulty, not least because of their political dimensions (Addison 2000). Within the low-income group, Ghana presents a particularly interesting case; over two decades of economic stagnation and decline began to be reversed with the start of economic reform in 1983 and the adoption of the Economic Recovery Programme. Ghana has successfully stabilized its economy, thereby reversing most of the distortions and macro-economic imbalances of the period before 1983. The economy has achieved respectable rates of real GDP growth—averaging over 4 per cent—in the 1990s, compared to the negative average growth rate over the period 1970-1983 (Table 1). The average inflation rate fell to about 19 per cent in 1998 relative to 122 per cent in 1983. However, poverty remains widespread and it is still predominantly a rural phenomenon. The 1998/1999 Ghana Living Standards Survey (GLSS3) shows that the incidence of poverty was about 29 per cent per cent nationwide and 36 per cent in rural areas; the Rural Savannah and Rural forest areas account for almost 70 per cent of total poverty in the country. In fact the incidence of poverty in the rural savannah has increased from about 56 per cent in 1991/1992 to over 58 per cent in the 1998/1999 period (Table 2). The need for a closer look at equity issues in Ghana cannot be over-emphasised; about 23 per cent of the population live in the rural savannah region and contribute over 40 per cent to national poverty. The incidence of extreme poverty is therefore still high. Most of the reduction in national poverty has only been experienced by those close to the poverty line, particularly those living in the urban areas. Table 1 Selected macroeconomic indicators for Ghana, 1970-1999 GDP growth Inflation Dom. interest rates Terms of trade Debt: GDP Debt serv.: GDP Debt serv.: Exports Aid: GDP 1970–1983 -0.004 50.0 13 160.7 31.1 2.2 10.5 3.0 1984–1991 5.4 28.3 20.3 119.2 58.1 6.1 36.6 8.9 1992 3.9 10.0 19.4 95.6 70.2 4.9 28.2 9.6 1993–1995 4.1 41.4 31.4 95.5 90.9 6.1 25.5 10.2 1996 4.6 46.6 41.6 88.5 88.6 6.9 27.1 9.4 1997 4.2 27.9 42.8 88.5 86.9 7.3 29.8 7.2 1998 4.6* 19.3 34.3 98.1 8.5 31.1 1999 4.4* 13.8* 26.4 Notes: * Data obtained from the government budget statements. The inflation figure quoted from the budget is the end of period rate 1 Table 2: Poverty incidence by location (Ghana, 1991/1992 and 1998/1999) Upper poverty line = 900,000 cedis Lower poverty line = 700,000 cedis Poverty Incidence Contribution to Poverty Incidence Contribution to Total Poverty Total Poverty rd 3 Ghana Living Standard Survery – 1991/1992 Accra Urban Urban Urban Rural Rural Rural 22.4 28.3 25.8 37.9 49.7 60.8 72.1 3.6 4.8 5.6 4.0 13.8 35.4 32.8 11.6 14.9 12.9 27.0 30.7 45.1 55.9 2.7 3.6 4.0 4.0 12.2 37.4 36.2 All Urban All Rural All Ghana 27.5 62.4 50.8 th 18.0 82.0 100 15.3 45.8 35.7 14.2 85.8 100 Accra Urban Urban Urban Rural Rural Rural 4.7 26.8 24.8 42.2 46.3 41.4 70.5 0.9 5.3 5.9 4.4 16.1 31.6 35.6 2.4 17.1 15.1 29.7 30.1 24.4 58.2 0.6 5.1 5.2 4.5 15.1 26.9 42.6 All Urban All Rural All Ghana 22.8 51.6 42.6 16.7 83.3 100 14.5 36.2 29.4 15.4 84.6 100 4 Ghana Living Standards Survey – 1998/1999 Source: GOG, 2000 The government’s response to this severe poverty problem is set out in the Medium Term Development Plan drafted in 1995. In broad perspective, the plan's objective is for Ghana to become a middle income country by 2020 (the plan is therefore referred to as Vision 2020). To secure faster poverty reduction, Ghana needs to achieve sustained economic growth—certainly more than the 4.4 per cent in 1999. The government also needs to raise public spending on basic services and infrastructure of most value to the poor—in particular primary education, basic health care, sanitation and safe water and economic infrastructure (for example better transport infrastructure in the north). The task for policymakers is therefore to ensure that fiscal policy contributes to broad-based (poverty-reducing) economic growth and the direct improvement of human development indicators. This paper explores these issues, focusing on the mobilisation of more public revenue through tax reform. Ghana needs to raise revenue, in order to achieve higher levels of development expenditure (and match aid-financed capital expenditures with sufficient recurrent spending). But revenues must be mobilised in ways that do not unduly distort incentives, or patterns of private investment. Taxation reform has already proved to be highly controversial in Ghana, the 1995 introduction of VAT failed—after widespread protests—and was finally brought into operation in December 1998. 2 In section 1 we look at the trends and structure of public expenditures and revenues as well as the implications of trends in the fiscal deficit for the macroeconomy. Section 2 outlines the political economy of taxation in Ghana paying particular attention to tax reforms after 1983. We also discuss the incidence of petroleum taxes and its implication for poverty reduction in the poorer regions of Ghana. We conclude the paper by emphasising the importance of revenue mobilization to Ghana. However, we caution that until government expenditure policies become more credible there is a need to give more consideration to equity issues in taxation in Ghana 1 The fiscal position in Ghana 1.1 Public spending: trends and structure Table 3 shows the structure and trend of recent public spending. Ghana has managed to restore spending in real terms, although little progress has been made in shifting spending to development and/or pro-poor priorities. Government expenditures ratios (expressed as a per cent of GDP) increased over the post–ERP period; the most significant jump being in 1992 (the year Ghana returned to constitutional rule) when the expenditure/GDP ratio increased by over 4 percentage points. Comparatively, the share of public expenditures in GDP was over 4 percentage points higher than the SubSaharan average of about 23.2. Moreover, while the average for the sub region seems to have declined in the 1990s, the same cannot be said for Ghana (Table 4). A look at the components of public expenditure shows that the public wage bill accounts for the lion's share: it averaged over 27 per cent of the total in the pre-ERP period, reaching about 34 per cent in 1992 (Table 3).1 Although the share of wages in total expenditures has declined in more recent years, it is still higher than the pre-reform average. Interest payments have also become an important component of public spending especially during the 1990s, increasing from about 12 per cent of total expenditure in the late 1980s to about 32 per cent in 1997. This is a significant portion of government expenditures, and needs to be given serious attention if faster progress is to be made in reducing poverty. Two reasons account for this increase in interest payments. First, high domestic interest rates which have averaged over 30 per cent in the 1990s, significantly more than the early 1980s average of about 13 per cent. Second, Ghana’s external debt ratio of about 87 per cent of GDP in 1997 is almost 3 times the average for the pre-reform period (Table 1). This has resulted in the debt-servicing ratio increasing from an average of about 11 per cent of GDP over 1970-1983 to about 31 per cent of GDP in 1998. Despite the duration of economic reform, the pattern of public expenditures in Ghana still does not appear to be ‘pro-poor’—or at least it is no more pro-poor than it was in the pre-ERP era. Although there have been slight increases in both health and education expenditures (as a per cent of GDP) their importance relative to other expenditure items 1 As Amoako Tuffour et al. (1996: 7) point out, this is probably an under-estimate since government transfers to subvented institutions include personnel related expenditures excluded from the government wage bill. We expect that this ‘additional public wage bill’ will have decreased in more recent years with the increased momentum of the divestiture program. 3 has declined—the share of public health spending in total expenditures decreased from an average of about 9 per cent over 1984–1991 to about 7 per cent in 1997. Public spending on education relative to total government expenditures also decreased from about 19 per cent in the pre-ERP era to about 15 per cent in 1997. This is particularly worrying given that expenditures in Ghana have been traditionally biased towards the urban areas and also towards university education and tertiary health care. A World Bank poverty report on Ghana in 1993 reported that only 25 per cent of the total health budget was earmarked for primary health and preventive care. This was markedly better in education where the proportion spent on basic education was about 62 per cent in 1989 compared to about 44 per cent in 1984. Even if the share of spending on primary education and health care as a proportion of the respective sector totals has increased over the years, it still does not warrant a reduction in the share of education and health expenditures in the total budget. In a 1995 World Bank poverty assessment report on Ghana it was also noted that social spending was not well targeted to the poor. Public expenditure on health was too low and the existing spending was urban biased (World Bank 1995: 2). In summary one can say that although some progress has been made in reducing the size of the public-sector wage-bill, it has yet to be translated into increased resources to priority sectors such as health and education. Rather resources have been taken away from these priority sectors to finance mainly the servicing of public sector debt (both domestic and foreign). In the government’s Interim Poverty Reduction Strategy paper the projected shares of spending on primary health and basic education in their respective sectoral totals shows a significant improvement, when compared to the shares in the early 1990s. Also the projected shares of both health and education spending in total public expenditure are expected to increase in the medium term. However these projections will depend on whether the projected reduction in the spending on interest payments can be achieved. Fortunately Ghana does not have any pressing security concerns (assuming that Côte d'Ivoire does return to stability). Therefore military spending has been low and can be expected to stay that way, thereby avoiding the burdens experienced in more insecure sub-regions of SSA (Addison and Murshed 2000, Ndikumana 2000). Table 3 Trends and structure of public expenditures in Ghana, 1970-1998 Expenditure (% of GDP) Capital: Total Exp. Goods: Total Exp Interest: Total Exp Wages: Total Exp Military: Total Exp Health: Total Exp Education: Total 1970–1983 16.2 18.5 49.9 12.4 27.8 ---- ---- 1984–1991 13.1 18.1 56.0 11.6 31.4 4.8 1992 17.8 19.7 48.2 12.3 34.3 1993–1995 21.6 19.5 40.9 18.7 1996 22.2 26.0 34.0 1997 20.6 18.7 35.6 19.1 Health: GDP ---- Education: GDP 2.9 9.3 20.6 1.3 2.8 4.6 7.9 16.6 1.4 3.0 26.5 4.1 7.0 13.8 1.5 3.0 23.0 24.4 3.0 7.0 13.6 1.5 3.0 31.5 25.8 2.4 6.9 14.6 1.4 3.0 1998 3.0 1999 Notes: Expenditure ratios are obtianed from the World Bank and the IMF (i.e. 1994 onwards is from the IFS-IMF). Military expenditure data were obatined from the WDI of the World bank. Although they differ slightly from the SIPRI data the general trend for both data sets indicate a decline in this expenditure item.We report the World bank figures because they cover a longer period. Education expenditure data were obtained from the the World bank (WDI2000). They were expressed as a per cent of the GDP and so we divided them by the total expenditure to GDP ratio to get them as a per cent of the total public expenditure. The same was done for the health expenditure ratios. 4 Table 4 Comparison of the Structure of Publc Expenditure in Ghana and the Sub-Saharan Africa averages Period Gross Public Total Interest Wages and Capital Investment Payments Salaries Expenditure (% of GDP) (% of total) (% of total) (% of total) 1990-1995 10.8 11.0 21.3 41.9 1996 13.3 17.2 18.2 44.8 1997 12.4 22.9 18.8 40.9 1998 11.3 22.9 19.5 39.4 1990-1995 7.4 17.4 20.0 26.5 SSA (Excl. 1996 South Africa) 1997 6.4 15.6 20.6 28.0 6.6 14.0 21.1 28.7 1998 6.1 13.3 24.1 23.5 1990-1995 7.3 22.4 22.7 29.5 1996 6.6 18.7 20.8 33.5 1997 7.4 17.1 19.7 35.8 1998 6.6 14.5 22.4 25.8 1990-1995 6.9 11.5 25.5 21.3 1996 5.9 11.8 27.0 19.6 1997 6.1 11.2 26.7 19.9 1998 6.3 16.6 29.2 17.6 Ghana West Africa Africa 1.2 Revenue generation: trends and structure Revenue mobilization has risen under successive IMF supported adjustment programmes, with the share of tax revenues in GDP increasing from an average of about 11 per cent over the decade preceding the ERP (1983) to about 19 per cent in 1998 (see Table 5). In comparative terms, this is higher than the SSA (excluding South Africa) average by about 1 percentage point although still lower than the African average of about 24.5 per cent in 1998. Box 1 gives details of revenue reforms since 1984. The components of tax revenues (in Table 5) shows that trade taxes—which have been the major source of revenue for governments in Ghana—have declined from an average of about 40 per cent (of total government revenue excluding grants) during the pre-ERP era to about 28 per cent in 1998. This is an encouraging trend given that over-taxing international trade acts as a disincentive both to manufacturing as well as export promotion. Two factors account for this declining importance of trade taxes in Ghana. 5 Box 1 Chronicle of tax reforms in Ghana 1984 (June). Tax rates on cigarettes and beer are increased, whilst those on personal income tax are lowered. 1985. National Revenue Secretariat (NRS) is established, as well as two major revenue organisations (CEPS and the IRS) as autonomous institutions outside the civil service. The NRS is responsible for supervising the activities of CEPS and the IRS as well as recommending revenue policy to the government.a An investment code that provides a range of tax incentives for investors (mainly in agriculture, manufacturing, construction and tourism) is introduced. 1987. Corporate tax rate on manufacturing concerns reduced to 45 per cent from 55 per cent; special taxes on cigarettes, beer, alcoholic, and non-alcoholic beverages; excise duties imposed on all locally produced goods except petroleum products; 10 per cent sales tax on domestic electricity consumption is abolished; abolition of import duty and purchase tax on all commercial vehicles; deduction of duty on basic raw materials and capital goods. 1988 (January). Reintroduction of sales tax clearance certificate to enforce timely payment of sales tax and excise duties collected on behalf of the government by manufacturers. 1990 (January). Introduction of super sales tax on luxury goods, ranging from 50 per cent to 500 per cent. 1991. The super sales tax is reduced to a new range of 10 to 100 per cent; corporate taxes applicable to agriculture, manufacturing, real estate, construction and services are lowered to about 35 per cent; the NRS is relocated under the MFEP although it retained some degree of independence; the personal income tax threshold is raised from the 126,000 to 150,000 cedis. 1993. The contract for the design and implementation of the VAT is signed. Increase in petroleum taxes to counteract the pay increases awarded to civil servants and other public service organisations in the election year. Petrol, kerosene, and gas oil prices increased by about 60 per cent, whereas LPG increased by about 20 per cent. 1994. Debt collection Unit established under the MFEP in January. In December the VAT bill was passed into law, to be operational in March 1995. 1995. VAT became operational in March at a flat rate of 17.5 per cent compared with the sales tax of 15 per cent. In June VAT was withdrawn and sales tax reintroduced at a rate of 15 per cent after mass demonstrations in almost all the regional capitals in the country. 1996. The petroleum sector was deregulated. This involves a process whereby the procurement of crude oil and finished products needed as a top-up of the Tema Oil Refinery capacity will be through competitive tendering between all the oil marketing companies and the Ghana National Petroleum Corporation (GNPC). 1998. Reintroduction of the VAT at a rate of 10 per cent adopted by parliament in February. Taxpayers Identification Numbers are introduced to make assessment and collection of taxes from small businesses and market traders easier. Note a This arrangement followed the Canadian system (Seth Terkper 1995). 6 First the falling terms of trade especially in the 1990s has meant that the export tax base has been reduced; the terms of trade fell from an average of about 119 over the period 1984-1991 to about 98 in 1998. Second, the government has made a conscious effort to promote exports, which has been predominantly cocoa. There has been a significant increase in the real producer price of cocoa in the 1990s compared with the late 1980s which has resulted in increased output (Figure 1) Table 5 Trends and structure of government revenues in Ghana (1970-1999) 1 Total Revenue (% of GDP) Grants (% of GDP) Tax Rev (% of GDP) Direct tax (% of total) Indirect tax (% of total) G&S Tax (% of total) Trade tax (% of total) Petroleum taxes (% of total) 1970–1983 10.9 0.05* 6.5 20.7 --- 27.3 40.4 --- 1984–1991 12.7 2.3 11.3 21.5 66.9 29.5 37.4 12.02 1992 11.9 3.3 10.8 18.6 71.8 49.2 22.6 19.2 1993–1995 18.0 3.8 14.7 18.0 64.1 40.1 24.0 20.1 1996 17.6 2.6 15.1 21.7 63.9 36.7 27.3 17.0 1997 17.3 1.9 14.7 24.8 59.8 34.1 25.8 15.2 1998 18.7 2.7 16.2 22.9 63.6 35.3 28.4 14.1 1999 13.3 Note: * The average does not span the entire period. Figure 1 Plot of real producer price of cocoa in Ghana (at 1990 constant prices) 160 140 100 80 60 40 20 0 19 70 19 71 19 72 19 73 19 74 19 75 19 76 19 77 19 78 19 79 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 Cedis per Head load 120 Real producer price of coca 7 Table 6 Structure of government revenue in Ghana compared to the Sub-Saharan Africa averages P eriod Total Tax receipts (% of total) D irect Taxes (% of total) International Trade taxes (% of Total) Total G rants (% of G D P ) Tax : G D P 1990-1995 79.3 18.9 26.2 3.4 13.2 1996 81.7 21.7 27.3 2.6 15.1 1997 82.0 24.8 25.8 1.9 14.7 1998 83.5 22.9 28.4 2.7 16.2 1990-1995 90.6 20.1 17.9 2.1 15.9 SSA (E xcl. 1996 91.2 21.0 21.5 1.8 16.0 South A frica) 1997 90.8 20.8 21.1 1.8* 16.3 1998 89.2 20.7 20.3 1.7 16.0 1990-1995 93.6 13.9 17.8 1.7 12.7 1996 94.8 13.8 20.0 1.6 13.4 1997 93.6 14.5 19.0 1.5 13.6 1998 90.7 14.4 19.1 1.4 13.7 1990-1995 87.8 32.1 13.5 1.2 20.5 1996 87.0* 31.3 14.3 0.7 21.1 1997 87.0 28.6 13.3 0.7 21.2 1998 85.9 30.8 13.0 0.7 21.2 G hana W est A frica A frica ratio Compared to the pre-1983 period receipts of taxes on goods and services have generally increased in Ghana, although the more significant increases have occurred during the 1990s. By 1992 the share of revenue from taxes on goods and services in total revenue accounted for about 50 per cent (almost twice the pre-1980 average). Although this ratio fell to about 35 per cent in 1998, we expect to see a gradual increase after the reintroduction of the VAT in 1998; the provisional out-turn for VAT receipts exceeded the projected figure by over 13 per cent in 1999 (GOG 2000). Taxes on goods and services have therefore become the most important source of revenue in Ghana, certainly in the 1990s. The direct tax component of total revenue does not seem to have changed very much from the pre-1983 period. In fact compared to the adjustment period it declined from about 21 per cent over the decade preceding 1983 to about 19 per cent in 1992. It has however picked up in more recent years increasing to about 23 per cent in 1998. 1.3 The macroeconomic framework: recent developments Figure 2 shows the overall fiscal deficit over time. Prior to 1983 public expenditures persistently outstripped revenues resulting in large fiscal deficits which peaked at about 11 per cent of GDP in 1976. After 1983 the public expenditure-revenue gap narrowed and by the late 1980s revenues had outstripped expenditures translating into a surplus which was maintained until 1991. Most of the surplus that was achieved in the late 1980s and early 1990s was due to the large inflows of foreign aid. As Figure 2 shows deficits excluding grants achieved a surplus in only two years; 1991 and 1994. In 1992 there was a re-emergence of the large fiscal deficits that had been witnessed over the 1970s. This coincided with Ghana’s return to democratic rule. The large deficits reappeared in 1996 (another election year) after recovering from the 1992 level, although with a much smaller ratio. These troughs in the budget deficits are in a sense consistent with the fiscal illusion hypothesis in political business cycle literature. This 8 hypothesis asserts that politicians follow expansionary fiscal policies in pre-election years since voters are, in the terminology of Alesina and Perotti (1995; 8-10). ‘fiscally illuded’ and don’t fully understand the implications of expansionary policies for postelection years. In the case of Ghana, however, the emphasis is on consumption spending—increases in wages and other direct cash transfers—and so the expenditure increases coincide with the election years. Figure 2 Plot of government budget deficits in Ghana, 1970–1997 (per cent of GDP) 4 2 0 -2 % -4 -6 -8 -10 -12 19 70 19 71 19 72 19 73 19 74 19 75 19 76 19 77 19 78 19 79 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 -14 Deficits excluding grants 9 Deficits Including grants Table 7 Budget deficit and sources of deficit financing, Ghana (1970-1999) 1970 – 1983 Deficits Including Grants: GDP -6.0 Deficits excluding Grants: GDP -6.6 Primary Foreign deficits: GDP financing of deficits: GDP --0.6 Domestic Financing of deficits: GDP 5.4 1984 – 1991 -0.1 -1.0 -1.1 0.3 -0.3 1992 -5.2 -6.3 -7.3 0.01 5.1 1993 – 1995 -1.0 -1.0 -4.3 -0.3 0.1 1996 -3.0 -3-6 -4.4 -1.7 4.7 1997 -2.1 -2.6 -2.7 -3.0 5.2 1998 0.1 1999 In 1992 virtually all expenditure items exceeded their programmed levels, with the public wage bill being the worst ‘culprit’. The share of wages in total expenditure increased by more than 10 per cent between 1991 and 1992. Coupled with that, there was a general slackening in tax collection in that year (GOG 1993: 2). Again a similar excuse was given for the deficits that occurred in 1996; ‘… a higher than programmed wage-bill, and the non-disbursement of foreign inflows’ (GOG 1997; 3). A careful look at the numbers in the 1997 budget and, in particular the difference between the programmed surplus and the deficits that emerged, suggests that other factors were at work. Specifically, as in 1992 there was a slackening in the revenue collection effort; non-tax revenues in particular fell short of their programmed level. This suggests that within Ghana's electoral cycle, revenue slowdown is possibly as important as expenditure increases. Prior to 1983, and especially in the late 1970s, both public expenditures and revenues were declining and consequently the deficits that occurred were mainly because revenues declined proportionately more. Declining revenues epitomised the economic crisis that the country faced. By 1982 the economy (and society) had become split between a formal sector serving mainly a small elite and an informal sector within which most economic activity took place. Consequently, the base for both direct and indirect taxation seriously eroded. Frimpong-Ansah (1991: 111) writing on this period notes that: ‘the remaining limited economic activity had withdrawn into havens of anarchy beyond the reach of state authority and state taxation’. With the expenditure-revenue gap rising faster than aid inflows, and a limited (and shrinking market for government paper), the government resorted to inflationary mechanisms for financing the rising fiscal deficit—by borrowing from the domestic banking system (mainly the central bank). By 1982 the government’s share of borrowing from the domestic bank sector was about 86 per cent (Kusi 1998:10). This monetisation of the deficit resulted in the crowding out of private investment, acceleration in inflation, currency overvaluation and, consequently, reduced the incomes of net exporters (mainly cocoa farmers). 10 In the World Bank/IMF assisted adjustment programme (which started in 1983) emphasis was placed on the need to reduce fiscal deficits through increased revenue mobilisation and a reduction in government expenditures. Exchange rate devaluation and the removal of price controls had the effect of realigning the informal and formal sectors of the economy resulting in an increase in the tax base and consequently tax revenues. This has translated into a closure of the expenditure-revenue gap even though both series have been on the increase over the post-adjustment period (see Figure 3). Net foreign borrowing mainly financed fiscal deficits over 1983-1991. Huge deficits in 1992 coupled with a shortfall in aid flows resulted in the government resorting to domestic sources mainly the central bank for finance. This no doubt contributed to the increase in domestic interest rates and inflation in the years that followed. In more recent years divestiture receipts have also become a significant financing item (Dordunoo 2000: 100-103). However, these non-revenue sources of financing are by their nature temporary, and improving the tax system is crucial to macro-economic stability and growth. The oil price increase in 2000 will have adverse effects on Ghana's economy (and therefore the fiscal position), as is the case in other non-oil producing developing countries. The strategy to improve the fiscal position set out recently by the IMF only takes into account the adverse trade shock in 1999 (IMF 2000). Even then it is noted that the most efficient way of addressing the projected revenue loss (resulting from the economy's slowdown) will be to increase the VAT rate to 12.5 per cent in 2000 (which has already been implemented) and subsequently to 13 per cent by 2003. The oil price shock in 2000 calls into question these projections, and will put pressure on the government to raise taxes at a faster rate than originally planned—unless more aid is forthcoming to offset the negative fiscal effects of the terms of trade shock. Figure 3 Plot of public expenditure and revenue in Ghana, 1970-1997 25.00 20.00 15.00 10.00 5.00 Govt. Expenditure 11 97 19 95 96 19 94 19 19 93 92 19 19 91 90 19 19 89 88 Govt. Revenue 19 87 19 19 86 85 19 19 84 83 19 19 82 81 19 19 79 80 19 78 19 19 77 76 19 19 75 74 19 19 72 71 73 19 19 19 19 70 0.00 The government has consistently failed to meet its fiscal targets in recent years, and this calls into question the realism of the projections in Vision 2020; considerable catch up is needed to remain on target to meet them. First, as previously noted, the divestiture receipts are short-term inflows for the government and are therefore expected to dwindle in the not too distant future. Second, foreign aid flows—which had been one of the principal means for financing the fiscal deficit over most of the adjustment/post adjustment period—have seen a decline in recent years. The aid/GDP ratio in 1997 was about 7.2 per cent compared with an average of about 8.9 per cent over 1984–1991 (Table 7). The tax effort must be improved if a return to monetisation of the fiscal deficit and the accompanying inflation are to be avoided. Certainly the introduction of the VAT is a welcome step in this direction, although one that was not politically easy to take. 2 The political economy of taxation in Ghana 2.1 The choice of taxation instruments The dependence on trade taxes has been reduced and this has been compensated for by an expansion of indirect taxes and non-tax revenues. Tax reforms over the post 1983 period have mainly been geared towards achieving two goals. The first stage reforms aimed to restore the tax base and strengthen production incentives that had been eroded by persistent over-valuation and the divergence between prices in the official and parallel markets. Accordingly, the foreign-exchange market was liberalized and controls on prices of goods and services were removed. The investment code (passed under PNDC law 116) in 1985 as well as the new minerals law in 1986 provided a wide range of incentives for investors in priority sectors of the economy. These incentives included company tax allowances, for example 40 to100 per cent of the cost of capital investment could be set against tax (Kusi 1998). Second-stage tax reforms sought to improve the efficiency of tax administration and to tackle issues of equity within the system. In line with these objectives the national revenue secretariat (NRS) was formed in 1986 with ministerial status and charged with the duty of overseeing the Custom, Excise and Preventive service (CEPS) and the Internal Revenue Service (IRS) and also advising government on revenue measures.2 The formation of the NRS coincided with measures to make the operations of the IRS and CEPS independent of the civil service structure. However, in 1991 the NRS was relocated under the Ministry of Finance and Economic Planning but was still allowed some measure of autonomy (Terkper 1995: 5-7). In line with steps to improve the administration of indirect taxes, the government abolished excise duties on all products except petroleum beverages and tobacco in 1987, and instead increased the sales tax rate from 10 to 25 per cent by 1988. This rate has, however, been systematically reduced in recent years and was 15 per cent by 1998. 2 Both CEPS and the IRS were previously under the Ministry of Finance and Economic Planning. 12 The most radical tax reform in Ghana have been the introduction of the VAT. The VAT was adopted for several reasons. First, it brings Ghana into line with the ECOWAS protocol that makes it mandatory for members to adopt the VAT system by the end of 1999 (Osei 2000: 258). Second, this system of taxation was thought to be more efficient, less burdensome in terms of its incidence, and its overall impact more equitable than the sales tax (GOG 1994: 32). Despite these advantages, the first attempt at introducing the VAT in 1995 failed due to widespread public opposition. In part, this was because of the haphazard manner in which it was implemented reflecting inadequate investment in institutional capacity. Moreover, little attempt was made to educate businessmen or the public more generally about the rationale for the tax (Osei 2000). It was reintroduced in December 1998 and is now operational in Ghana. Petroleum taxes are very controversial and are one of the tax instruments most hotly debated in Ghana today. Two reasons account for this. First there has been a persistent increase in the prices of petroleum products over this period, and this has mostly been associated with increases in the petroleum tax. Second, the average Ghanaian doesn't have much interest in changes in the fiscal deficit, but they do clearly understand the effects of petroleum tax increases on their cost of living. Persistent increases in the prices of petroleum products have given rise to speculative activity, especially among retailers who have periodically made large windfall gains. The government, in order to get its share of the windfall gains, converted any such gains that resulted from a rise in the world price of oil into a specific excise duty on petroleum products (Kusi 1998). Petroleum price increases have also been used to discourage its smuggling to neighbouring countries (GOG 1991: 20). In 1996, attempts to reduce speculation related to petroleum products led to the deregulation of the petroleum sector. This involved a competitive tendering process between the oil marketing companies as well as the Ghana National Petroleum Company (GNPC) which was previously the only importer of crude oil into the country. The tender process then determined the costs (cif) of crude oil, which constitutes about 90 per cent of the cost in the refinery activity as a whole (GOG 1996: 28-30). In 1998 taxes on petroleum products were re-structured so that they consisted of two components; an ad valorem component to the ex-refinery price (applicable to all products) and a specific rate applicable to road transport fuels (mainly gasoline). Direct taxes have also seen some reforms over the post 1983 period. To reduce evasion, the basis for assessing corporate taxes was changed in the 1980s from profits to income. Corporate tax rates have also been unified and reduced for all sectors; these rates fell to about 35 per cent by 1993 compared to 55 per cent in 1986 (Kusi 1998). The tax free bracket for personal income tax has been persistently increased whilst marginal rates have been lowered so as to reduce the average effective rates. Moreover, to reduce tax evasion by individuals shifting incomes between corporate profits and personal taxes, the marginal tax rates of the top brackets on the personal income scale have been made equal to corporate tax rates. 2.2 Taxation and its incidence Should the taxation instrument be directed to the objective of poverty reduction? The standard argument is that taxation should be directed to raising revenue in the least distorting way possible, and that objectives of equity (poverty reduction and, perhaps 13 additionally, the reduction of inequality) should be addressed by other means (for example pro-poor public spending and targeted safety nets). But is this argument completely applicable to Ghana, or do there exist some second-best arguments for modifying the structure of taxation in order to address equity concerns? The current IMF view on the distributional effects of adjustment programmes is that expanding the tax base for property and income taxes will have positive distributional effects (i.e. they are broadly pro-poor) whereas indirect taxes, particularly on goods and services, tend to have adverse distributional effects (Tanzi 1997). Ghana’s tax reforms have not necessarily followed this pattern. The base for property and income tax has increased. However, for indirect taxes both the base and the rate have been increased. As we noted earlier, petroleum taxes have become particularly important in the category of indirect taxes since they provide a convenient way of raising revenue (Younger 1996: 234). For instance in 1993 the ex-pump prices of petroleum products increased by about 60 per cent.3 The reason for the increase as quoted in the budget was as follows: government revenue will have to be increased in order to pay the present wage-bill, and the only immediate way for this to be done is to increase the prices of petroleum products. (GOG 1993:8) Younger (1996) analyses the incidence of taxes in Ghana using household data and finds petroleum taxes to be progressive but taxes on alcoholic beverages and tobacco products regressive. He consequently argues that complaints about gasoline price increases falling disproportionately on the poor are unfounded (Younger 1996: 251).4 If tax incidence analysis alone is used to make inferences about welfare then the policy recommendation implied by Younger’s conclusion will be valid. However, tax incidence analysis of this kind only looks at the first-order impact and this may not be appropriate in welfare analysis. Indirect taxes may have spillover effects that may not be captured by incidence analysis alone. We attempt, in a very ad hoc way, to trace some of these spillover effects of a petroleum price rise. Say we divide households in Ghana into five groups: public servants,5 transport owners, traders, cocoa farmers and other food crop and livestock farmers. The last group produce for domestic consumption and basically represents the poorest households in Ghana. With a rise in price of petroleum products all the households are affected by the resulting increase in the cost of transportation and its consequent inflationary effects. Of course this will also vary both within- and between-groups; for instance within the group labelled public servants, the petroleum tax is likely to be regressive when just the consumption of transport services is considered.6 Our main interest here is to do with 3 The only exception was the price of liquefied petroleum gas (LPG) which increased by 20 per cent. Even then, the average for all petroleum products was over 50 per cent; Calculations are based on the 1993 budget statement. 4 This analysis is based on consumption of gasoline and transport services. 5 This can be defined to include those with white- and blue-collar jobs. 6 Some of those in the upper echelon of this group either get free fuel or get petrol and car maintenance allowance from their employers, usually the government. 14 what happens to those in the last group (food crop and livestock farmers) relative to the other groups when petroleum prices increase. Most of the farmers that fall in this category don’t sell their produce in the major markets where they are most likely to benefit from price increases. Most of their produce is sold to market traders (i.e. the ‘middlemen’) who have some power over farmgate prices since the supply of these goods is highly inelastic. It is not uncommon for the middlemen to pay lower prices to the farmers, citing increases in transport costs as an excuse. The implication of this will be that whereas all households are likely to be affected by the increased transport costs and inflation, the rural poor in Ghana are in addition likely to experience a decline in nominal income. For these households this decline could be substantial and can consequently lead to increased poverty within that group. Tax incidence analysis will not capture these adverse spillover effects on the food crop and livestock farmers. The conclusion of the above analysis—that such a broad based tax may have adverse distributional effects—is consistent with Colatei and Round (2000) who find, using a SAM-based CGE model, that structural adjustment may not be benefiting all segments of Ghanaian society. More specifically, they find that poverty in the savannah farm households (the poorest group in Ghana) cannot be totally eradicated using transfers financed from an indirect tax imposed on the other households.7 It must be emphasised that in the CGE analysis although all the resources generated from indirect taxation are transferred to the targeted households, poverty is still not eradicated within that group. Given the trends and structure of public expenditures in Ghana (which as we saw have been disappointing from a poverty reduction perspective), these regular increases in petroleum taxes do not seem to have been returned to the poor in the form of better public spending. Certainly the argument that the non-poor in society consume more energy products—and therefore a ‘first-order incidence analysis’ of an increase in petroleum taxes will be progressive—cannot be discarded. However the indirect effects of these taxes on the poor could be substantial and may actually worsen the distribution of income. Therefore further increases in petroleum taxes will be justified only if the extra resources generated are channelled towards specific pro-poor programmes. 3 Conclusions This paper has explored some of the hard choices that Ghana's policy makers face if they are to generate sufficient revenues to achieve public spending targets, including those for pro-poor spending. We also discussed progress to date in public expenditure management. The following observations were made: First, by and large, public expenditures in Ghana over the last decade have not been any more pro-poor than they were in the 1970s—despite all the rhetoric. The planned expenditure policies in Ghana’s Vision 2020 seek to reverse this trend. However, this can only be achieved if the public debt problem is brought under control and the share of interest payments in total expenditure is reduced, and more revenue is mobilized. 7 The authors calculate the amount of transfers necessary for eradication of poverty for the different groups of households. 15 Second, the structure of taxation in Ghana is moving away from heavy reliance on trade taxes to indirect taxes on goods and services. This is a step in the right direction as dependence on these taxes has been found to increase the degree of instability in government tax revenues (Bleaney et al. 2000). Third, the trend in the fiscal deficit in Ghana over the 1990s is starting to follow an electoral cycle. Admittedly, this observation is based on a small sample (only two elections to date). Nevertheless it is almost certain that Ghana, like many developed and developing countries that have competitive multi-party elections, will exhibit an electoral cycle in its fiscal variables (including a slackening in the tax collection effort in addition to the expansion of public expenditures in election years). The first election was certainly accompanied by a serious deterioration in fiscal control, the second less so. We can hypothesise that in new democracies where democratization is sustained, the amplitude of the electoral cycle in fiscal variables will decline over time as institutions that act as agents of restraint are developed, and as politicians begin to see that it is in their long-term political interest to avoid excessive manipulation of expenditures or taxation for party-political purposes. In summary, this paper has shown that Ghana must raise revenues if it is to achieve the sustained high growth that is necessary for poverty reduction. However revenue generation considerations should take issues of equity into account. This is particularly so given that expenditure patterns do not appear to have been geared towards tackling poverty reduction. It is heartening to note that the interim poverty reduction strategy of the government in the Vision 2020 document (endorsed by both the IMF and the World Bank), emphasises the need to deepen access to basic social services and infrastructure available to the poor, but it is clearly proving very difficult for the government to deliver on these promises (a situation aggravated by the recent and adverse terms of trade shock). To end, we make a couple of concluding points. First, given the importance of the road transport sector to the Ghanaian economy, it is possible that increases in petroleum taxes may hurt the poor, more than a first-order incidence analysis suggests. In order to achieve the pro-poor growth envisaged in the Vision 2020, there is a need to raise revenues to finance projected expenditures. Expenditure policies have, however, not been particularly credible in Ghana especially in the 1990s, because of both internal and external shocks. Therefore, further increases in petroleum taxes may only be justified if the generated revenue is earmarked as additional spending on specific pro-poor programmes over and above the already projected spending. In other words, policy makers should give more consideration to equity issues in taxation if the objective of broad-based growth is to be realised. Second, investment in democratic institutions is important to tax reform, since tax payers will be unwilling to comply with tax law unless there is some mechanism for ensuring that their money is used legitimately. Historically, effective tax systems have resulted from the exchange of resources for institutions: governments have only been able to sustain their tax effort if they provide effective institutions (such as law and order), including limitations on their own powers (Mahon 2000). Ghana has advanced with regard to both democratisation and the construction of appropriate fiscal institutions, which are now more transparent. Nevertheless, much more needs to be done, including the possibility of giving the legislature more power. This may be inconsistent with findings in the political business cycle literature, where countries with 16 ‘hierarchical’ budgetary procedures are noted to achieve more fiscal discipline. However, for a country such as Ghana—in which democracy is still an infant—it may be better to give the legislature more powers to act as a check on the executive until such time as the budget process becomes truly transparent. References Adam, C.S. and S.A. O'Connell (1999). ‘Aid, taxation and development in Sub-Saharan Africa’. Economics and Politics, 11 (3): 225-253 Addison, T. (2000). The Political Economy of Fiscal Policy in Low-Income Countries. Paper prepared for the UNU/WIDER project meeting on ‘New Fiscal Policies for Growth and Poverty Reduction’, Helsinki, 17-18 November 2000. Helsinki: UNU/WIDER. Addison, T. and S.M. Murshed (2000). Fiscal Policy in Conflict and Reconstruction, Paper presented at the UNU/WIDER project meeting on ‘Why Some Countries Avoid Conflict While Others Fail’, Helsinki 20-21 October 2000. Helsinki: UNU/WIDER. Alesina, A and R. Perotti (1995). The Political Economy of Budget Deficits, IMF Staff Papers, Vol. 42, Number 1, pp. 1-31 Amoako-Tuffour, J. (1999). ‘Ghana government fiscal deficits: how small or how large?’ Journal of African Economies, Vol. 8, pp. 1-30. Amoako-Tuffour, J., J.S.D.Brean and C. Youngblood (1996). Public Finances in Ghana (1983-1995): Trends and Issues, Centre for economic Policy Analysis (CEPA) Research/Working Paper, August 1996. Aryeetey, E., Harrigan, J. and Nissanke, M. (2000). Economic Reforms in Ghana: The Miracle and the Mirage. Oxford: James Curry. Bleaney, M., N. Gemmil and D. Greenaway, (2000). ‘Tax revenue instability in SubSaharan Africa: causes and consequences’, in Bevan et al. (eds), Trade and Fiscal Adjustments in Africa. London: Macmillan. Bolnick, B. and J. Haughton (1998). Tax Policy in Sub-Saharan Africa: Examining the Role of Excise Taxation. African Economic Policy, Discussion Paper Number 2. Colatei, D., and J.I. Round (2000). Poverty and Policy: Experiments with a SAM-based CGE Model for Ghana. Paper presented at the XIII International Conference on Input-Output Techniques, 21–25 August, Macerato, Italy Cowen, M. Laakso, L. Virtanen, S. and Yrjolä, R. (2000). Testing for Political Business Cycles in Africa. Processed, University of Helsinki, Department of Political Science. Dordunoo, C.K., (2000). ‘Fiscal trends: 1970-1995’, in Aryeetey, E., J. Harrigan, and M. Nissanke (eds), Economic Reforms in Ghana: The Reality and Mirage , pp. 88144. London: James Curry Frimpong-Ansah, J.H., (1991). The Vampire State in Africa: The Political Economy of Decline in Ghana. Trenton: African World Press. 17 Government of Ghana (1991). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1991. Accra: Ghana Publishing Corporation. Government of Ghana (1993). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1993. Accra: Ghana Publishing Corporation. Government of Ghana (1994). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1994. Accra: Ghana Publishing Corporation. Government of Ghana (1995). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1995. Accra: Ghana Publishing Corporation. Government of Ghana (1996). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1996. Accra: Ghana Publishing Corporation. Government of Ghana (1997). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1997. Accra: Ghana Publishing Corporation. Government of Ghana (1998). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1998. Accra: Ghana Publishing Corporation. Government of Ghana (1999). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 1999. Accra: Ghana Publishing Corporation. Government of Ghana (2000a). The Budget Statement and Economic policy of the Government of Ghana for the Financial year 2000. Accra: Ghana Publishing Corporation. Government of Ghana (2000b). Interim Poverty Reduction Strategy Paper: 2000-2002. Ministry of Finance and Economic Planning, June. IMF (2000). Ghana: Selected Issues. Staff Country Report, Number 2. Kusi, N.K. (1998). Tax Reform and Revenue Productivity in Ghana. AERC Research Paper, No. 74, March Edition. Mahon, J.E. (2000). Globalization and the Exchange of Institutions for Resources. Processed, Department of Political Science, Williams College. Paper presented at the 18th World Congress of the International Political Science Association, Quebec, August 1-16 (www.williams.edu/PoliSci/mahon/index.www.html). Ndikumana, L. (2000). Fiscal Policy, Conflict and Reconstruction in Burundi and Rwanda. Paper prepared for the UNU/WIDER project on ‘New Fiscal Policies for Growth and Poverty Reduction’, published as WIDER Discussion Paper 2001/62, Helsinki: UNU/WIDER. Osei, P (2000). ‘Political liberalisation and the implementation of value added tax in Ghana’. The Journal of Modern African Studies, Vol. 28, Number 2, pp. 255-278. 18 Tanzi, V., (1997). Asia and the IMF: The changing Role of Fiscal Policy in Fund Policy Advice. IMF Seminar Paper, September. Terkper, S.E. (1995). Ghana Tax Administration Reforms (1985-1993), Harvard Institute for International Development, Development Discussion Papers. World Bank (1993). 2000 and Beyond: Setting the Stage for Accelerated Growth and Poverty Reduction. World Bank Poverty Assessment Summaries. World Bank (1995). 2000 and Beyond: Setting the Stage for Accelerated Growth and Poverty Reduction. World Bank Poverty Assessment Summaries. Younger, S. (1996). ‘Estimating tax incidence in Ghana using household data’, in D.E. Sahn (ed.), Economic Reform and the Poor: 231-253. Oxford: Clarendon Press. 19 View publication stats