Inequality and Fiscal Policy
Chapter

Chapter 21. A Path to Equitable Fiscal Consolidation in the Republic of Congo

Author(s):
Benedict Clements, Ruud Mooij, Sanjeev Gupta, and Michael Keen
Published Date:
September 2015
Share
  • ShareShare
Show Summary Details
Author(s)
Maximilien Queyranne, Dalia Hakura and Cameron McLoughlin 

Introduction

This chapter addresses the issue of fiscal consolidation design in an oil-dependent country, the Republic of Congo, with a limited remaining lifetime of oil reserves and high income inequality and poverty. The scarcity of statistical data makes the analysis, as well as the formulation of policy recommendations, especially challenging. Within these constraints, the chapter presents a narrative of fiscal policy in Congo, including comparisons with similar countries, and its potential impact on poverty alleviation.

The Republic of Congo’s economy faces short-term upside risks with the sharp decline in oil prices since mid-2014, and significant challenges in the medium term associated with the expected depletion of oil reserves. Therefore, room for active fiscal policy to reduce poverty and inequality has already narrowed, and is expected to drastically diminish in the coming years. So far, fiscal policy has proven excessively accommodative, with a limited impact on poverty and inequality. In 2013, the overall fiscal balance on a cash basis was in surplus (13.9 percent of non-oil GDP), but major adjustment is required in the medium term because the country runs a very large non-oil primary fiscal deficit (-61.2 percent of non-oil GDP).1 Ensuring an orderly fiscal adjustment, while simultaneously reducing poverty and inequality, will prove particularly challenging given that income inequality tends to rise significantly during periods of fiscal consolidation (Agnello and Sousa 2012).

This chapter is structured as follows: First, an overview of inequality and poverty in Congo is provided and compared with that in similar countries. The current composition of the public budget in Congo is then discussed; this budget is insufficiently geared toward the reduction of inequalities and poverty. The need for fiscal consolidation going forward in Congo is examined, and a strategy that could mitigate its adverse effect on inequality and poverty is proposed.

Overview of Inequality and Poverty in Congo

In 2011, poverty in Congo was significantly higher than that in countries with similar GDP per capita2 (Figure 21.1), and Congo’s poverty scores are similar to or higher than those of other sub-Saharan African countries with significantly lower incomes. Although the poverty rate declined from 50.7 percent in 2005 to 46.5 percent in 2011, the number of poor people increased from 1.8 million to 1.9 million.3 In addition, poverty is widespread in rural areas (75.6 percent) compared with urban areas (29.4 percent in Brazzaville). With regard to income inequality, the Gini coefficient for disposable income declined in Congo between 2005 and 2011 and is now equivalent to the average for sub-Saharan Africa countries (0.44).4 But the Gini remains higher than that of countries with similar income levels (0.39). And Congo’s United Nations Human Development Index score is significantly below the average of countries with similar GDP per capita and has improved more slowly since 2005.

Figure 21.1Republic of Congo: Selected Poverty and Human Development Indicators

Sources: World Bank, World Development Indicators database (panel 1); United Nations Development Program, Human Development Index (panel 2).

Note: Comparators are developing countries whose per capita GDP (in purchasing-power-parity U.S. dollars) is between 25 percent above and 25 percent below that of the Republic of Congo.

Composition of Public Expenditure and Revenue

Total government revenue and spending are particularly high in the Republic of Congo. Total government revenue represented 120.5 percent of non-oil GDP in 2013, significantly higher than in most oil-exporting low-income countries (Figure 21.2). Oil revenue accounted for almost 75 percent of total government revenues. As a result, total government spending was also significantly higher than for comparators, with the exception of Equatorial Guinea.5

Figure 21.2Fiscal Aggregates in 2013

(Percent of non-oil GDP)

Source: IMF staff estimates.

However, the low level of tax revenue significantly reduces the redistributive role of tax policy. Tax revenue accounted for only 6.7 percent of GDP in 2010, much lower than the average for sub-Saharan Africa countries (17.9 percent) and countries with comparable income levels (23.4 percent). Tax policy can, therefore, play only a marginal role in achieving redistributive goals in Congo. In addition, the tax structure favors consumption taxes, which are less progressive than income and wealth-related taxes. Income tax and property tax revenues (classified as “other taxes”) are particularly limited, even compared with other sub-Saharan Africa countries (Figure 21.3).

Figure 21.3Composition of Government Revenues

Source: IMF staff estimates.

Recent personal income tax reforms have not strengthened progressivity. The government reformed the personal income tax system in 2011 and 2013 by reducing the rate applicable to each bracket by 5 percentage points (except the first bracket, which was maintained at 1 percent) and increasing the size of the four lowest brackets.6 The IMF staff estimates that this reform is mostly regressive because its benefits accrue more to higher-income families and single people with intermediate incomes (Table 21.1). These results reflect that (1) the impact of the tax rate reduction will be larger than the impact of the widening of the lowest brackets, which benefits more low-income households; and (2) the large tax deductions on salary income and the family tax system benefit higher-income taxpayers more, because they are proportional to income.

Table 21.1Decrease in the Average Tax Rate Due to the 2013 Personal Income Tax Reform(Percentage points)
Taxable Income Levels (CFAF)Change in the Average Tax Rate for a Single PersonChange in the Average Tax Rate for a Family with Two ChildrenChange in the Average Tax Rate for a Family with Four Children
232,000000
732,0005.800
2,000,0006.26.15.6
5,500,0006.26.47.2
15,000,0004.86.47.2
Source: IMF staff estimates.Note: CFAF = Communauté Financière Africaine Franc.

Social spending was marginal in 2010 and was largely crowded out by significant energy subsidies. Social spending was much lower than in most sub-Saharan Africa countries and countries with similar income levels (Figure 21.4). In Congo, fuel subsidies were higher in 2010 (3.6 percent GDP) than aggregate spending on education, health, and social protection (2.5 percent GDP) and significantly larger than in the sub-Saharan Africa region as a whole (1.4 percent). Large fuel subsidies may reflect the desire to share the country’s oil wealth through the provision of petroleum products at prices below those in the international market (Arze del Granado, Coady, and Gillingham 2010), even if refined oil is imported.7 Posttax subsidies, which take into account both revenue needs and negative consumption externalities, are among the highest for oil-exporting low-income countries, but remain below those of the Middle East and North Africa oil-exporting countries.

Figure 21.4Composition of Spending

(Percent of GDP)

Source: IMF staff estimates.

Note: In panel 2, pretax consumer subsidies are estimated as the difference between international prices, adjusted upward for transportation and distribution margins, and domestic consumer prices. MENA = Middle East and North Africa.

Fuel subsidies distort market incentives, lead to overconsumption of petroleum products, and favor higher-income groups. In the Republic of Congo, domestic petroleum prices are fixed by ministerial decree below supply costs. As a consequence, market participants consume more petroleum products, reducing the overall efficiency of the economy. And the overconsumption adds to the total cost of the subsidies. Moreover, low prices relative to neighboring countries create incentives to smuggle petroleum products out of the country. Finally, fuel subsidies are usually poorly targeted, and benefits accrue mostly to higher-income groups because they consume the most. Except for kerosene, the distribution of fuel subsidies is skewed toward the top two quintiles, which receive, on average, 62–81 percent of the benefits.8

The government has stepped up its investment sharply to address infrastructure gaps. Public capital spending rose from 6.1 percent of GDP in 2006 to 18.8 percent in 2010 (Figure 21.5). As a result, public capital stock in the Republic of Congo has increased since 2006 and was significantly higher than in comparators in 2011, including oil-exporting low-income countries9 (Figure 21.5). According to the National Development Plan, the vast majority of capital spending for the period 2012–16 is to be allocated to infrastructure (51.7 percent in 2014) and economic development (16.7 percent). However, the share of capital expenditure allocated to social development was expected to increase from 11.7 percent to 16.7 in 2014. According to the 2014 Budget Act, social ministries are to receive about 14 percent of government capital expenditure.

Figure 21.5Capital Expenditure and Public Capital Stock

(Percent of GDP)

Sources: Center for International Comparisons; OECD; and IMF staff calculations.

Education spending and service provision have recently increased significantly. Public resources allocated to the education sector were limited in 2010 compared with other sub-Saharan Africa countries and countries in the same income range (Figure 21.6). In addition, the composition of education spending was somewhat regressive. Primary education tends to be more progressive, because lower-income groups have greater access to this level of education. But the share of expenditure allocated to secondary and tertiary education in the Republic of Congo was higher than in countries with comparable income levels. In addition, although Congo was performing comparatively well according to the gross enrollment rate, the average class size in 2010 was significantly higher than in countries with comparable GDP per capita, raising questions about service quality. In recognition of these weaknesses, the government has progressively increased its education spending. The share of the budget allocated to the education ministries rose from 6.1 percent in 2012 to 8.9 percent in 2014, with a nominal increase of 57 percent during the same period, but the impact on education headline indicators is yet to be measured.

Figure 21.6Education Spending and Outcomes, 2010

Source: World Bank.

Health spending and in-kind services have not compensated for large income and geographic inequalities. In 2010, health spending in Congo was among the lowest in sub-Saharan Africa (Figure 21.7). As a result, the country’s reliance on out-of-pocket spending, at 64 percent of total health financing, was among the highest in sub-Saharan Africa. Service provision is insufficient and access to health care professionals limited, particularly in rural areas, which are also the poorest. Health inequality is high, with households in the lowest 20 percent of the income distribution suffering from significantly higher child mortality than the richest 20 percent.

Figure 21.7Health Spending, Service Provision, and Inequalities

Sources: World Health Organization (panels 1 and 2); Government of the Republic of Congo, Demographic and Health Survey, 2012 (panels 3 and 4).

The government has committed to implementing a system of universal health care insurance. Whereas budget allocations have increased for the education sector, they decreased for the Ministry of Health between 2012 and 2014, both in nominal terms (by 1.6 percent) and as a share of total budget spending (by 1.3 percentage points). However, with World Bank support, the government is expected to spend US$100 million between 2015 and 2020, with the aim of implementing universal health coverage. This program would include fee waivers for the poorest households, as well as free service provision. How it will be financed has not yet been determined and will depend on funding from household contributions and government subsidies to the poorest households.

The Republic of Congo has also started implementing social safety net programs. With increased donor support, safety net spending in Africa has increased since 2005, following food and financial crises (Monchuk 2014). Congo got a late start compared with other sub-Saharan Africa countries, but is preparing temporary income-generating activities for unemployed youth and labor-intensive, self-employment, and rural employment programs. In addition, a pilot conditional cash transfer program (the LISUNGI project) is being rolled out for 5,000 poor households and 1,000 elderly people. The cost of expanding this program on a national scale is estimated to be about 1 percent of GDP (Box 21.1).

Box 21.1Initiating a Social Safety Net through Cash Transfers

To address the challenges of high poverty and inequality, the government of the Republic of Congo is focusing on the development of a social safety net for poor and vulnerable groups. In its initial phase, the four-year social safety net program (the LISUNGI project) will make conditional cash transfers to 5,000 poor families and 1,000 people older than age 60 in three areas (Brazzaville, Pointe-Noire, and Cuvette). Eligible households are those living below the food poverty line and with at least one child (between the ages of 0 and 14 years) or an elderly person. Transfers to the households will be conditional on continuous schooling of children and regular health checks for all household members. The program will be monitored and evaluated regularly. Depending on its effectiveness, the program’s coverage will be expanded to all eligible households in 2016.

The conditional cash transfer program is expected to improve the Republic of Congo’s social and economic indicators in three ways. First, the transfers are expected to reduce current poverty, reaching families that do not directly benefit from economic growth. Second, because they are conditional on education and health access, the transfers should have a positive impact on school enrollment rates and on child nutrition, thereby improving human capital. Finally, household productivity should improve because the collateral provided by the cash transfers will help families invest in economic activities and gain access to microcredit. Higher employment and social cohesion should ensue.

The LISUNGI project mimics similar programs in other countries, such as Brazil, Colombia, Ghana, Kenya, Mexico, and Niger. The success of these programs is directly related to the quality of the management information system and direct and regular payments to recipients.

An impact analysis carried out by the World Bank suggests that the cash transfer program could have a significant impact on poverty and inequality. It is expected that the poverty rate would decline from 46.5 percent in 2011 to 38.9 percent—3.9 percentage points higher than the Millennium Development Goals Initiative target (35 percent by 2015)—if the program were to operate nationwide. The Gini index of inequality would also drop between 8.0 and 11.8 percent. Implementing the program on a national scale—with payments of CFAF 20,000–25,000 on average a month to the poorest households with children or elderly members (or both)—would cost about 1 percent of GDP or about 2 percent of public expenditure.

Strategy for Equitable Fiscal Consolidation

Estimates of proven oil reserves suggest that oil production is to peak in 2017 and will then decline decisively after 15 years.10 This pattern will have a major impact on government revenues, which are expected to decline by about 80 percentage points of non-oil GDP during the next 20 years (Table 21.2). As a consequence, spending will have to be sharply reduced, by about 60 percentage points of non-oil GDP during the same period, to cut the non-oil primary deficit to about 30 percent of non-oil GDP by 2019, with a further reduction in the medium to long term (IMF 2014c). This reduction in the non-oil deficit should be achieved by decreasing and reallocating spending, while progressively raising tax revenues and increasing the progressivity of the tax system.

Table 21.2Medium-Term Fiscal Consolidation(Percent of non-oil GDP)
2013201420152019202520332034
Revenue and Grants111.7102.694.267.445.232.531.8
Oil Revenue82.172.364.236.213.42.21.9
Non-oil Revenue28.628.829.130.331.33029.7
Total Expenditure91.49183.762.338.231.230.5
Capital Expenditure57.756.449.432.615.713.212.9
Non-oil Primary Balance−61.2−60.5−53.6−31.1−6.5−2.2−1.5
Debt-to-GDP Ratio38.238.737.335.231.418.817.4

Fiscal consolidation should be based on progressive tax and spending measures to protect vulnerable households during adjustment. In low-income countries, fiscal adjustment can have an adverse effect on employment and inequality in the short term, but this effect may be reversed in the long term. Inequality and unemployment may even decline in the longer term if fiscal adjustment helps bring down inflation—which is damaging to the poor—or corrects macroeconomic imbalances that are hindering growth. And because spending in developing economies is generally not progressive, cutting such spending can ensure fiscal consolidation while avoiding a surge in inequality (IMF 2014a). Hence, fiscal policy should aim to balance the provision of much-needed public services with fiscal sustainability, through tax revenue mobilization and prioritization of spending. The government should not resort to across-the-board spending cuts, which can hurt low-income groups. It should instead focus on improving the composition and efficiency of spending to prevent spending restraint from affecting the quantity or quality of basic services.

A better-functioning personal income tax system would enhance tax progressivity. Strengthening the personal income tax yield can raise the tax ratio while strengthening progressivity (OECD 2006). Implementing a zero bracket for the lowest-income brackets would both simplify revenue administration and enhance tax progressivity. Tax deductions must also be reduced because they accrue disproportionately to the rich and lead to significant revenue losses. Large tax deductions on salary income should be eliminated and further reduced for professional expenses. The family tax system benefits based on the number of dependents (quotient familial) is a major hindrance to income tax equity because it favors large, high-income families; it should be replaced by a fixed tax credit that is the same for all taxpayers. In addition, there is no need for a tax incentive to increase the fertility rate in the Republic of Congo, which is among the highest in the world. Deductions for mortgage interest and capital income should also be eliminated: only high-income households are able to borrow from banks and receive financial earnings.

Property taxes should be developed. Taxes on residential property and on excess returns or rents, particularly in resource-rich economies, are considered the least distortive for growth (IMF 2014b). There is large scope for increasing the residential property tax in the Republic of Congo. This reform could have a significant redistributive impact. To make it progressive, the tax could exclude the permanent residences of those below a certain threshold to prevent taxation of low-income households. A property tax could be implemented gradually given that it requires a reliable land registry and the administrative capacity to manage it.

The use of reduced value-added tax (VAT) rates and exemptions should be minimized. Achieving redistributive objectives through consumption taxes usually proves to be costly. The rich generally spend more in absolute terms, so they tend to benefit more from exemptions or reduced VAT rates. The Republic of Congo uses these instruments extensively for necessities, in particular on a large variety of food products and agricultural inputs. There is also an exemption for electricity and water consumption, which is particularly regressive given that only 37.8 percent of the population had access to electricity in 2011. Reducing the list of goods exempt or benefiting from reduced VAT rates would help raise revenues that could be used to increase targeted social transfers, administrative capacity permitting.

With regard to expenditure efficiency, the government should aim to cut fuel subsidies to scale up social spending. Reducing or eliminating fuel subsidies would create fiscal space for social spending. It would improve the progressivity of public spending, since higher-income households consume more petroleum products. However, it would have a negative impact on the poorest households, because energy consumption represents a large share of their total consumption. Implementing this reform gradually and compensating vulnerable households would be critical to its success (Alleyne and Hussain 2013). In particular, interventions targeted to vulnerable households through conditional cash transfers would be needed. But designing such programs requires significantly improved data transparency and reliability for informed policy decision making. Other mitigating measures could, for example, take the form of subsidies for public transportation.

Containing the public wage bill and spending on goods and services could provide additional fiscal space for social spending. Purchases of goods and services (9.2 percent of non-oil GDP) are much higher in the Republic of Congo than the average for oil-exporting low-income countries (6.9 percent) and other Central African Economic and Monetary Community (CEMAC) countries (excluding Gabon). Some expenditure reform is needed in this area (Figure 21.8). The wage bill in the Republic of Congo (9.8 percent of non-oil GDP) is also higher than in most other CEMAC countries. Progressive containment of the wage bill would help create fiscal space for social spending. The strategy should aim for adequate recruitment in the health and education sectors and reduced hiring in nonpriority sectors through natural attrition.

Figure 21.8Selected Current Expenditure, 2013

(Percent of non-oil GDP)

Source: IMF staff estimates.

A reduction and a change of the composition of public investment would also provide room for additional pro-poor and pro-growth infrastructure spending as well as social spending. The capital stock accumulated in the Republic of Congo by 2011 was significantly higher than in countries with similar income levels and almost twice the average of oil-exporting low-income countries. In addition, capital spending pressures will increase with the organization of the 2015 All Africa Games in the Republic of Congo. The proposed long-term path aims at stabilizing the capital-stock-to-GDP ratio at its 2011 level. The composition of public investment could be improved to maximize the impact on growth and human development. This applies particularly to public investment in infrastructure. Through better prioritization of public investment, space can be created to expand government social sector spending to improve service provision. This approach will require evaluating current expenditure needs associated with the increase in education and health infrastructure.

Improving access to in-kind services by low-income groups would help reduce poverty and inequality in the medium term and boost growth in the long term. In the education sector, the government should prioritize spending on primary schools and improve service quality, in particular through recruitment of additional teachers to close large staffing gaps (estimated to be 14,000 teachers by the Primary National Education Council). In the health sector, the fiscal costs of the introduction of universal health coverage should be carefully assessed. The implementation should be sequenced to increase household coverage progressively while mitigating the impact on fiscal aggregates. Plans should be made in particular to determine the level of noncontributory coverage to protect population groups unable to afford insurance, while containing fiscal costs. Progressivity should also be embedded in the design of social taxes to finance the health care system—a flat price would weigh disproportionately on those who are less well off. And to reduce poverty and inequality, the system should focus on expanding access to a broad package of essential health services for poor households; reducing health copayment and user charges for low-income households; and ensuring access to health care facilities and professionals, particularly in rural areas.

New social safety nets should be adequately designed. The Republic of Congo is in the initial steps of designing new safety net programs, so it should be mindful of design and implementation shortcomings that have been observed in other low-income countries. In particular, the Republic of Congo should avoid fragmenting and duplicating small social programs and guard against bad targeting, which can cause substantial leakage for non-poor households and result in insufficient benefits for the most vulnerable people.

Conclusion

Government spending in the Republic of Congo is comparatively high, benefiting from very large resource revenue. However, it is poorly targeted, with large fuel subsidies that crowd out social spending and ambitious investment plans that may prove to be unsustainable and inefficient.

The expected decline in oil revenue in the medium term is a major challenge to the Republic of Congo’s fiscal policy stance. Compensating for the shortfall in oil revenue through tax revenue mobilization will be difficult even in the medium to long term and may negatively affect the poorest, given the current low level of the tax-to-GDP ratio. Reducing the non-oil deficit will also require a significant reduction in spending, which may affect pro-poor spending.

To ensure that the fiscal consolidation is equitable, the government should aim to increase progressive taxation, sharply reduce fuel subsidies, and improve the efficiency and quality of capital spending by prioritizing public sector infrastructure investment that maximizes the growth and human development impact to create space to reallocate spending toward social sectors.

References

Defined as total revenue (excluding investment loans and grants) minus total expenditure (excluding interest payments and foreign-financed investments). In 2013, non-oil nominal GDP (CFAF 2,796 billion) represented about 42 percent of nominal GDP (CFAF 6,657 billion).

This analysis considers all countries whose per capita GDP (in purchasing-power-parity U.S. dollars) is between 25 percent above and 25 percent below that of the Republic of Congo. The analysis is based on internationally comparable data, for which the latest observations are 2011 or 2012.

The Republic of Congo’s population grew at an annual average rate of 3 percent during the same period.

The latest available international Gini coefficient is for 2005. For comparison purposes, the 2011 figure was estimated by applying the percentage point improvement between 2005 and 2011 from the national household survey, in which the Gini coefficient went down from 0.42 to 0.39 over this period.

Part of the oil revenues generated by the proceeds of the Republic of Congo’s oil sales to China (equivalent to more than one-third of the Republic of Congo’s total annual oil exports) is kept in an escrow account in the Export-Import Bank of China as a guarantee for the concessional loans granted by that bank to the Republic of Congo.

Tax brackets are as follows (2014 Budget Act): 1 percent (below CFAF 0.46 million), 10 percent (CFAF 0.46–1 million), 25 percent (CFAF 1–3 million), 40 percent (CFAF 3–8 million), and 45 percent (more than CFAF 8 million).

Only about 5 percent of domestic production is processed for domestic consumption by the national refinery; this amount meets about 70 percent of domestic demand for petroleum products. As a result, the national oil company imports additional refined products.

IMF staff estimates based on the 2005 household survey.

Cameroon, Chad, Sudan, Vietnam, Yemen.

Average production in 2012 and 2013 was 93 million barrels a year. After peaking at 118 million barrels in 2017, oil production is estimated to decline to about 18 million barrels by 2030.

    Other Resources Citing This Publication