Abstract
Selected Issues
Macro-Structural Policies and Inequality1
1. Benin shows high levels of poverty and inequality, with different patterns in terms of urban-rural populations. The highly informal nature of the economy and low productivity— particularly, in the agricultural sector—are the main causes of the lack of inclusiveness. Despite the drop of the overall non-monetary poverty in recent years, the data shows a high incidence of poverty in rural populations (north part of the country) compared with urban populations (concentrated in the south). At the same time, low tax revenues constrain the government’s ability to achieve social objectives.2 Based on a dynamic general equilibrium framework developed by Adrian et al (2017) and Fabrizio et al (2017), this paper discusses the medium-term macroeconomic and distributional impacts of structural policies and reforms being undertaking in Benin. The reforms considered center on measures to mobilize domestic resources.
2. The dynamic general equilibrium framework used builds on key structural characteristics of low income developing countries (LIDCs). The model is a small-open economy featuring two areas—rural and urban—and four sectors—domestic agricultural, exporting agricultural, manufacturing, and informal service. Value added taxes (VAT) are imposed on consumption of domestic agricultural and manufacturing goods (Box 1). The model is calibrated to match salient macro and distributional features of the Beninese economy, including share of different sectors in consumption and production, tax structure, and rural and urban income Gini indicators.
3. The framework generates an endogenous distribution of households that respond to changes in government policy in Benin. Changes in households’ behaviors when aggregated, help to predict the potential macroeconomic and distributional impacts of the policy reforms under consideration. The framework also captures both inequality across sectors and inequality within sectors. On the one hand, inequality across sectors depends on the mobility of workers across areas of activities. On the other hand, within-sector inequality is explained by the fact that, although households of a given type and location may be ex ante identical, their individual productivity is subject to shocks over time, thereby affecting their income in any given period. As a result, households end up with different incomes. Furthermore, government policies and financial sector features affect different groups of the economy differently, driving both macroeconomic performance and distributional outcomes.
A. Reforms for Enhancing Domestic Resource Mobilization
4. To address large macroeconomic imbalances and accommodate an ambitious investment plan, Benin launched a reform strategy in 2017 centered on domestic revenue mobilization. Benin faced a difficult macroeconomic situation characterized by the following observations: growth had slowed significantly, the fiscal accounts had weakened, and the public debt-to-GDP ratio had increased by 43 percentage points over three years, reaching 47 percent of GDP in 2016. Persistently low tax revenue constraints the size of the budget and, hence, limits the government’s capacity to pursue social objectives. The reform package sought to boost tax revenues through an increase in the VAT rate and cut non-priority recurrent spending to contain the accumulation of public debt. A quantitative assessment is required to assess the interaction of VAT and the distribution prospects of the economy (Box 2).
General Structure of the Model
✓ The model is a small-open economy with two areas (urban and rural), and four sectors (domestic agricultural good, exporting agricultural good, manufacturing, and informal service). *
✓ Households are heterogenous in their productivity, supporting an endogenous distribution of consumption, and income. Lower productivity rural households work for exporting farms which produce domestic and exporting agricultural goods, while higher productivity rural households work on their own plots, producing domestic agricultural good. On the contrary, lower productivity urban households produce informal service goods, and higher productivity urban households work for formal manufacturing firms.
✓ Households save through a risk-free asset, which is all turned into capital used by manufacturing firms. Government collects taxes through VAT, PIT, and CIT. Taxes can be used either in a non-productive way, or for cash transfer program and infrastructure investment. Both the international account and the government account is balanced in the model.
✓ In the equilibrium, all households and firms optimize, taken prices and government policies as given. Prices, wages, and interest rate clear all markets.
5. Simulation results of an increase in tax revenue equaling two percent of GDP through VAT (the main simulation) suggest that the reform leads to a direct negative impact on consumption and output. This is due to the fact that tax revenues are assumed to be used in a non-productive way. The impact on investment, however, is positive due to an increase in the relative price of saving to consumption, crowding in private savings, which are channeled to capital used by firms. The negative impact on consumption and output can be mostly mitigated, if the additional tax revenue collected are used for infrastructure investment that boosts the productivity in all sectors of the economy (Box 3, upper-left/right panel).
Quantitative Assessment of Tax Incidence: Why is it Needed? *
LICs have four prominent features that make VATs particularly effective and attractive to them than other tax instruments:
compared to developed countries, LICs usually are characterized as a “dual-economy” with large agricultural and informal sectors;
LICs often show low level of capital accumulation.
a large informal sector reduces the effectiveness of labor income taxes, and
low capital accumulation provides a very small tax base for capital return taxes.
As a result, the implementation of a policy recommendations to achieve fiscal consolidation in LICs, usually show pros and cons:
Due to the large incidence of informal sectors, tax evasion becomes a a typical characteristic, as most of the transactions are executed through means of cash transactions and alike.
Consequently, higher VATs create extra wedges between prices of different goods, distorting resource allocation towards informal sectors, which eventually leads to an informal sector that is larger than otherwise predicted by a frictionless model.
Furthermore, by definition, VATs tax people with higher consumption heavier. Given the large share of groups of poor households in LICs—who spend a large share of their income on food consumption—VATs impose a relatively higher tax burden on these poor households measured as percentage of total income.
Hence, the distributional impacts of the VATs could worsen situations of inequality. And this has a more pronounced incidence in LICs. On the other hand, tax revenues collected can be channeled to build infrastructure that benefits long-run economic growth, protect the poor, and finance redistributive social welfare programs to reduce inequality and poverty rate.
The interactions between VATs and specific features of LICs lead to channels that affect the growth and distribution prospects of the economy both positively and negatively. Thus, the overall effect is theoretically ambiguous, and needs to be investigated quantitatively.
*/ Based on Xin Tang (2017).6. A comparison of simulation results of revenue mobilizations equaling 2 percent GDP through VAT, corporate income taxes (CIT), and personal income taxes (PIT) suggests that VAT is the least distortive in terms of aggregate output. The reason is specifically because the crowd-in effect of private investment (Box 3, lower-left panel). The use of personal income taxes (PIT) or corporate income taxes (CIT) leads to higher reduction in economic activity. CIT reduces the return of investment, which leads to a large decrease in economic activity. The PIT reduces the disposable income of the richer households, leading to a reduction of the aggregate savings of the economy and consequently to lower investment and economic activity. Thus, due to its neutral impact on investment, VAT taxes are the least distortive tax on this economy.
7. The VAT reform results in higher income inequality in urban areas. In urban areas, poor households work in the non-tradable sector, which is informal, while richer households work in the manufacturing sector. The VAT reform reduces aggregate demand and the prices of non-tradable goods leading to a reduction on the income of the urban poor. The urban rich households observe an increase in income due to the increase in the investment in the economy, which leads to relative higher wage in the manufacturing sector (Box 2, upper-left panel).
8. The VAT reform leads to lower income inequality in rural areas. The decrease of aggregate demand leads to a reduction of agricultural prices affecting rural households. The richest rural households are affected more because they sell a largest share of their production, which leads to lower levels of income inequality in the sector (Box 2, upper-left panel).
9. The implementation of universal basic income (UBI) reduces both consumption inequality and income inequality. Inequality is reduced because the UBI boosts the income of poor households directly and indirectly.
directly due to the fact that the transfers represent a larger share of poor households’ income, and
indirectly through a 4 percent increase in the relative price of non-tradable goods that are produce by the poor (Box 2, lower-right panel).
10. Scenarios. The framework is simulated for five different scenarios using the benchmark equilibrium as the base.
| Scenario | Description |
|---|---|
| 1 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used in non-productive governmental expenditure. |
| 2 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used for uniform cash transfer. |
| 3 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used to finance infrastructure investment that boosts overall productivity by 1.74%. |
| 4 | Additional tax revenue equaling 2 percent of GDP by CIT, additional tax revenues used in non-productive governmental expenditure. |
| 5 | Additional tax revenue equaling 2 percent of GDP by PIT, additional tax revenues used in non-productive governmental expenditure. |
| Scenario | Description |
|---|---|
| 1 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used in non-productive governmental expenditure. |
| 2 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used for uniform cash transfer. |
| 3 | Additional tax revenue equaling 2 percent of GDP by VAT, additional tax revenues used to finance infrastructure investment that boosts overall productivity by 1.74%. |
| 4 | Additional tax revenue equaling 2 percent of GDP by CIT, additional tax revenues used in non-productive governmental expenditure. |
| 5 | Additional tax revenue equaling 2 percent of GDP by PIT, additional tax revenues used in non-productive governmental expenditure. |
Macroeconomic and Distributional Impacts of Tax Reforms
Citation: IMF Staff Country Reports 2018, 002; 10.5089/9781484336540.002.A004
* All results are percentage changes of the corresponding x-axis variables using the initial equilibrium (the status quo) as the benchmark.Macroeconomic and Distributional Impacts of Tax Reforms
Citation: IMF Staff Country Reports 2018, 002; 10.5089/9781484336540.002.A004
* All results are percentage changes of the corresponding x-axis variables using the initial equilibrium (the status quo) as the benchmark.Macroeconomic and Distributional Impacts of Tax Reforms
Citation: IMF Staff Country Reports 2018, 002; 10.5089/9781484336540.002.A004
* All results are percentage changes of the corresponding x-axis variables using the initial equilibrium (the status quo) as the benchmark.B. Policy Implications.
11. Success of Benin’s revenue mobilization reform depends, highly, on the nature of the tax instrument chosen and how the government revenue is used. The analysis indicates that for Benin, VAT is the least distortive instrument to raise revenue when compared to PIT and CIT. Specifically, model simulations indicate that increasing the VAT rate would be slightly progressive and would raise revenue with a smaller negative impact on economic activity than PIT and CIT hikes.
12. The presence of a large informal sector in Benin would shield the income of producers of informal (agricultural) goods, thus turning the tax slightly progressive.3 The model’s results are consistent with the analysis made for many LIDCs that have a small share of the population working for formal wages, a large unproductive agricultural sector, and scarce capital goods. If the government revenues are used on unproductive government consumption, revenue mobilization will lead to a contraction in economic activity. However, if they are used to finance efficient infrastructure investment, revenue mobilization can even boost economic growth if the investment has high return and it is efficient. In practice, it could be done through investment in roads, electrification, irrigation, agricultural research and development, and on agricultural services.
13. An implementation of UBI reduces the negative impact of revenue mobilization on economic activity. UBI also contributes to a reduction the inequality and poverty in both rural and urban areas. The increase in poor households’ consumption offsets some of the negative impact of VAT on output, making cash transfer program a desired policy instrument for inclusive growth.
References
Adrian Peralta-Alav, Marina Mendes Tarvares, and Xuan S. Tam, 2017, The Distributional Implications of Fiscal Consolidation in Developing Countries, Manuscript.
Bollinger, Christopher R. and Barry T. Hirsch, 2013, “Is Earnings Nonresponse Ignorable?” Review of Economics and Statistics, May, 95(2): 407–416.
Bollinger, Christopher R. and Barry T. Hirsch, 2006, Match Bias from Earnings Imputation in the Current Population Survey: The Case of Imperfect Matching, Journal of Labor Economics, July, 24, 483–519.
Fabrizio, Stefania, David Furceri, Rodrigo Garcia-Verdu, Bin Grace li, Sandra V. Lizarazo, Marina Mendes Tavares, Futoshi Narita, and Adrian Peralta-Alva, 2017, Macro-Structural Policies and Income Inequality in Low-Income Developing Countries. SDN/17/01. January
Institut National de la Statistique et de l’Analyse Economique, 2015, Enquete Modulaire Integree sur les Conditions de Vie des Menages. Octobre.
Medina, Leandro, Andrew Jonelis, and Mehmet Cangul, 2017, The Informal Economy in Sub-Saharan Africa: Size and Determinants, International Monetary Fund. WP/17/156
Tang, Xin, A Tutorial of the Toolkit for Solving a Multisector Heterogeneous Agents General Equilibrium Model, August. Manuscript.
Prepared by Benin team in close collaboration with Marina Mendes Tavares and Xin Tang (both SPR).
See INSAE (2015) for the most recent analysis of poverty in Benin based on a survey conducted in 2015, encompassing also a comparative analysis of inequality (2011-15).
