IMPACT OF VAT ON POOR HOUSEHOLDS IN BOTSWANA

1
IMPACT OF VAT ON POOR HOUSEHOLDS IN BOTSWANA

• by Achieng Okatch

2
Poverty and Inequality

• Botswana has experienced a decline in poverty
  since 1985/86.
• HIES of 1985/86 poverty rate was 59
• HIES of 1993/94 poverty rate was 47
• HIES of 2002/03 poverty rate was 30
• Although poverty has declined significantly,
  income inequality has worsened over the period
  1993/94 to 2002/03, with the Gini coefficient
  increasing from 0.537 to 0.573.
• The present level of the incidence of poverty is
  still inconsistent with the national development
  objective of building a Compassionate, Caring
  and Just Nation envisaged in Botswanas Vision
  2016.

3
The Problem

• Government Revenue
• Sixteen years prior to 1998/99, there were no
  budget deficits
• Budget deficits were realised in a number of
  years starting in 1998/99.
• Deficits can be attributed to the limited
  prospects of substantial increase in Government
  revenues outside the mineral sector,
• Growth in Government revenue has also reduced by
  trade agreements as SACU-USA Trade Agreement, the
  European Union-South African Free Trade Agreement
  and SADC Free Trade Protocol

4
Objectives of the Study

• The study seeks to
• the examines tax liability of current VAT rates,
  using micro-simulation techniques,
• propose new VAT rates that will strike balance
  between reducing poverty and income inequality,
  while, at the same time, ensuring an adequate
  collection of Government revenue.

5
Methodology

• The study uses or assumes
• HIES 2002/03 dataset 6053 Households and 23 823
  individuals
• Poverty datum line constructed by the Central
  Statistic Office - based on a theoretical
  consumption basket, which is assumed to
  correspond to a minimum amount goods and services
  required by households in Botswana to attain an
  acceptable minimum standard of living. The
  poverty datum line takes into account factors,
  such as region, price and composition of the
  household.
• Assumes constant expenditure - a change in VAT
  will not change the household expenditures but
  rather the quantity purchased.
• Static micro-simulation
• Distributional Characteristic derived from the
  Theory of Marginal Tax Reforms (TMTR)

6
Methodology (cont)

• Distributional Characteristic
• Derived from TMTR (Which stipulates that the
  social welfare effect of a marginal price change,
  due to tax, is given by the weighted sum of each
  households consumption of that good)
• Measures how heavily the consumption of a good is
  concentrated on the poor
• Used to identify goods candidate for having
  reduced tax rates on equity grounds
• Takes into consideration social weights of the
  households and consumption of the good by
  households

7
Current VAT Rates

• Two VAT rates in Botswana,
• the standard rate (10)
• the zero rate
• exports
• international transport services (passengers or
  goods),
• maize and sorghum (in all forms)
• petrol and diesel.
• Exempted goods
• accommodation
• international financial services
• education
• specified drugs, as indicated in the Drugs and
  Related Substances Act.

8
Limitations

• Its difficult to estimate the VAT on Medical
  expenditure
• The dataset does not record the household
  expenditure on the exempted drugs separately
• Poorer households use Government clinics and
  hospitals, which charge a standard consultation
  fee of P2, irrespective of the service provided
  and this rate is not subject to VAT.
• Richer households use private medical services,
  which are subject to VAT
• Individual covered by medical insurance only pay
  between 0 to 10 of the medical cost and all the
  VAT.
• Its difficult to VAT estimate on transport

9
Tax incidence of VAT by Decile, Gender of
Household Head and Strata
10
Budget Shares
11
Budget Shares
12
Budget Shares
13
Budget Shares
14
Budget Shares
15
Distributional Characteristic Results

• To calculate, expenditure were aggregated into 40
  groups
• Goods that should be zero rates
• (1) Sorghum and maize,
• (2) Sugar,
• (3) Oils and fats,
• (4) Other cereals (grains) and bread,
• (5) Other fuel (paraffin, menthylated spirit,
  charcoal, gas, coal and firewood)
• (6) Dairy products and eggs,
• (7) Vegetables,
• (8) Meats,
• (9) Tobacco,
• (10) Fish,
• (11) Personal care,
• (12) Non-alcoholic drinks,
• (13) Other foods,
• (14) Water
• (15) Fruits.

16
Simulations

• VAT Simulations
• Alternative scenario 1 (AS1)
• Zero rate - petrol and diesel,
• Exempt rentals
• 15 goods that are highly consumed by the poor
• Including Tobacco
• Imposing a VAT of 10 on the other 23 goods
• Including medical expenditure, construction,
  education
• Alternative scenario 2 (AS2)
• Same as AS1 except that instead of a 10 VAT, a
  14 is imposed on the 23 goods

17
Incidence of VAT for the Alternative VAT
Scenarios
18
Number of Gainers and Losers at the Two Proposed
Alternative Scenarios
19
Comparison of Current VAT Rates with the
Alternative Scenarios
20
Policy Recommendations

• Zero rating of the 15 goods whose consumption is
  heavily concentrated by the poor leads to more
  progressive VAT regimes. This is the case for
  both scenarios when 10 and 14 VAT rates are
  applied to goods that are not heavily consumed by
  the poor.
• The 14 VAT rate is preferred, and hence
  recommended on goods that are not highly consumed
  by the poor, as it decreases Government revenue
  by a smaller amount and at the same time also
  reduces poverty.