Transcript 2. Burundi: tax side

FERDI – CIFD – CERDI – Université d’Auvergne
12 juin 2014 – Clermont-Ferrand
Integrating the pro-poor agenda
into fiscal performance
measurement in LICs
A case study of Burundi
Jérôme Sansonetti
ODI Fellow, Office Burundais des Recettes
Introduction
Linking revenue and
poverty reduction
2
Introduction
Linkages between tax and poverty
o
In a country like Burundi, the pro-poor agenda is rarely included in fiscal
monitoring schemes
o
Poverty reduction remains a central policy objective
•
o
Fiscal performance should be measured and designed in a way that monitors propoor impact
Surprising, as it is a challenge to mobilize revenues in LICs
3
Introduction
Linkages between tax and poverty
o
The linkages between revenue reforms and poverty reduction are not
appreciated at every echelon on the ground
o
There is no direct link between revenue increase and poverty reduction
Increased
revenues
Progressive
taxes
Pro-poor
spending
tax side
spending side
Poverty
reduction
Pro-poor impact
Worst-case scenario?
4
Introduction
Increased risk of regressivity in LICs
o
LICs feature of higher risk of regressivity, all other things equal
•
•
o
Tax progressivity:
•
o
Narrow income tax bases
Large returns to capital (classical growth theory)
A tax is progressive if the tax payments of richer households accounts for a larger
proportion of their incomes than those of poorer households
Consumption taxes with a flat rate can be regressive
•
Poorer households consume a larger fraction of their incomes
5
Introduction
Research questions
1. Do revenue reforms in LICs
incur regressivity?
o
The nature of the linkages is
not fully appreciated
o
Raising revenues does incur a
risk of regressivity
2. Are fiscal performance
indicators fitted to monitor
pro-poor impact?
o
Indicators used in LICs fall
short
6
Outline
Introduction – the linkages between revenue mobilization
and poverty reduction
1. Literature review
2. Case of Burundi: tax side
3. Case of Burundi: spending side
4. Suggested indicators
7
Section 1
Literature review
8
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Overall fiscal structure in LICs
IFI consensus
Strengthen VAT revenues
Reduce trade taxes
Reduce corporate rate
9
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
VAT
o
IFI consensus (IMF FAD, 2011)
•
•
•
Broad-based: few exemptions
Single rate
High registration threshold
Regressivity
o
o
Progressivity
In South Africa, poorest
deciles feature higher VAT
burden (Go et al., 2005)
o
Average number of VAT rates
o
(IMF FAD, 2011)
•
•
LICs: 1,28
Rich countries: 2,52
VAT less regressive than
former sales taxes (Zolt & Bird,
2005)
More efficient to treat
progressivity on spending
side than by relaxing VAT
•
Case of Chile (Engel et al.
1999)
10
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Trade taxes
o
IFI consensus: reduce import duties and trade taxes (Marshall, 2009)
•
•
1980s: 4% - 5% of GDP
Late 2000s: 3% of GDP
Regressivity
o
Export taxes are arguably
progressive, so removal is
regressive
o
Replacement with VAT
creates aforementioned
issues
Progressivity
o
Removing import taxes is
progressive if poor
households consume a larger
fraction of imports
o
In South Africa, trade
liberalization was found to
be progressive (Daniels &
Edwards, 2006)
11
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Corporate income tax
o
IFI consensus (Marshall, 2009)
•
•
Decrease statutory rates to attract investors
Effect of Doing Business Indicators for “paying taxes”
Regressivity
o
Decrease in statutory rates in
SSA (Keen & Mansour, 2009)
•
•
o
1990: 44%
2005: 33%
Progressivity
o
General equilibrium effect
after tax shifting can be
progressive (Harberger, 1962)
Increase in exemptions in
SSA (Keen & Mansour, 2009)
•
•
1980: 40%
2005: 80%
12
Section 2
Burundi – tax side
13
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Burundi’s tax reforms since 2009
Tax administration
o
Creation of a Revenue
Authority in 2010 (OBR)
•
•
•
Compliance
Taxpayer services
IT systems
Tax policy
o
VAT created (2009)
o
VAT reform (2013)
•
•
o
Corporate income tax (2013)
•
o
More exemptions
Intermediary rate at 10%
Statutory rate reduced from
35% to 30%
Personal income tax (2013)
•
•
Threshold increased
Statutory rates decreased
14
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Burundi’s tax reforms
since
2009
Burundi's total tax revenues
600
30%
559,51
526,64
500
471,71
BIF bn
450
400
25%
20%
16,29%
13,66%
15,08%
14,14%
14,12%
15%
362,76
350
300
Tax/GDP
550
10%
301,21
5%
250
200
0%
2009
2010
2011
OBR collection (A)
2012
2013
Tax/GDP
15
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Overall distributional impact
As % of collection
2010
Taxes on revenues, profits and capital gains
2011
2012
2013
2010
2011
2012
2013
30,0% 26,6% 29,6% 25,7%
4,2%
4,3%
4,5%
3,6%
Total personal income taxes
Total corporate income taxes
11,7% 10,7% 12,7% 7,7%
14,3% 14,7% 15,7% 16,9%
Taxes on goods and services
53,4% 53,9% 54,3% 58,8%
VAT (domestic + customs)
Excise (domestic and customs)
36,5% 36,5% 36,5% 37,7%
16,0% 14,5% 15,3% 19,6%
Tax on international trade and transactions
Import duties
Export taxes
Non-fiscal products
TOTAL REVENUES
As % of GDP
10,8% 12,5% 9,4%
8,2%
0,0%
5,8%
8,4%
0,0%
7,0%
8,3%
0,0%
6,6%
9,0%
7,5%
0,1%
6,5%
100% 100% 100% 100%
1,6%
2,0%
7,6%
5,2%
2,3%
1,5%
1,2%
0,0%
0,8%
1,7%
2,4%
8,8%
5,9%
2,4%
2,0%
1,4%
0,0%
1,1%
1,9%
2,4%
8,2%
5,5%
2,3%
1,4%
1,3%
0,0%
1,0%
1,1%
2,4%
8,3%
5,3%
2,8%
1,3%
1,1%
0,0%
0,9%
14,1% 16,3% 15,1% 14,1%
16
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Data available to assess regressivity
o
QUIBB 2006: expenditure survey, household level
o
Construct a “total expenditure” variable
•
•
o
77 food and non-food expenses
Only health and education missing
Quartiles of wealth based on syncretic wealth score
Weighted aver. total exp. (BIF/year)
Weighted aver. total exp. (USD/year)
Quartile 1
178 946
119
Quartile 2
231 798
155
Total observations
Well surveyed observations
Observations for total exp. (BIF/mois)
Quartile 3
293 304
196
Quartile 4
898 268
599
7832
7042
5812
17
1. Literature review
2. Burundi: tax side
3. Burundi: spending
VAT – exemptions
Financial operations
Insurance operations
International transportation
Piped water and electricity private consumption
Post stamps
Rental of unfurnished private housing
Minibus transportation (14+ seats)
Health care
School and higher education material
Untransformed agricultural goods
Piped water consumed by HH
Electricity consumed by HH
Housing rental
Travels and transportation
Post services
4. Suggested indicators
Exempted?
Impact of
Pre-2013 Post-2013 exemption
Yes
Yes
Regressive
Yes
Yes
Regressive
Yes
Yes
Regressive
No
Yes
Regressive
Yes
Yes
Regressive
Yes
Yes
Regressive
No
Yes
Regressive
Yes
Yes
Regressive
Yes
Yes
Progressive
Yes
Yes
Progressive
Quartile 1 Quartile 2 Quartile 3 Quartile 4
0,03%
0,06%
0,07%
0,23%
0,02%
0,004%
0,04%
0,16%
0,07%
0,09%
0,16%
0,97%
0,40%
0,70%
1,13%
2,12%
0,01%
0,02%
0,01%
0,04%
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
VAT – intermediary rate
o
A 10% intermediary rate was implemented on 12 staple foods
•
Cassava, maize, wheat, etc…
Products taxed at 10%
Quartile 1 Quartile 2 Quartile 3 Quartile 4
52,02%
44,84%
42,34%
33,83%
o
Direct effect is progressive
o
Indirect effect is unclear, as retailers may not have shifted the VAT
decrease
•
•
Inflation for 10%-taxed staples (Jul.-Dec. 2013): 1.75%
Inflation for other food items (Jul.-Dec. 2013): -1.69%
19
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Excise
o
A 10% intermediary rate was implemented on 12 staple foods
Bottled water
Import of old vehicle (BIF)
Sodas (BIF/hL)
Beer (BIF/hL)
Sugar (BIF/Kg)
Wine and liquors
Tobacco
Fuel (BIF/L)
Sodas
Beer & alcoholic drinks
Sugar
Tobacco
Fuel
2010
0
0
18 000
17 820
400
50%
83%
NA
2011
18 000
19 049
165
2012
2013
10%
10%
1 500 000 1 500 000
30 000
30 000
36 000
36 000
400
600
70%
70%
120%
200%
24
21
Impact?
Progressive
Progressive
Progressive
Progressive
Progressive
Progressive
Regressive
Regressive
Quartile 1 Quartile 2 Quartile 3 Quartile 4
0,22%
0,23%
0,31%
0,73%
0,30%
0,55%
0,96%
2,21%
0,49%
0,71%
0,69%
1,63%
1,21%
1,31%
1,12%
0,64%
0,01%
0,04%
0,04%
0,26%
20
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Personal income tax – general
o
The reform triggered a dip in the contribution of PIT to total revenues
•
•
•
2010 to 2013 : 11.7% to 7.7% of total revenues
Higher taxable threshold
Lower statutory rates
Before reform
Bracket
Lower
Upper
40 000
40 001
48 350
48 351
56 650
56 651
65 000
65 001
73 350
73 351
81 650
81 651
164 950
164 951
248 250
248 251
331 550
331 551
Tax rate
0,0%
27,0%
31,0%
35,0%
40,0%
41,0%
43,0%
47,0%
55,0%
60,0%
Amount to
deduct
329
1 002
2 259
2 579
3 398
8 381
26 023
40 602
After reform
Bracket
Tax rate
Lower
Upper
150 000
0,0%
150 001
300 000
20,0%
300 001
30,0%
21
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Personal income tax – scenarios
o
Distributional impact mostly regressive
•
•
From 40 001 to 150 000, impact is unclear
From 150 001 upwards, impact is clearly regressive
Scenario
(i)
(ii)
(iii)
(iv)
(v)
(vi)
Monthly
taxable
income
20 000
50 000
100 000
200 000
350 000
950 000
Before 2013
As of 2013
Taxed
Effective
Taxed
Effective
amount
tax rate
amount
tax rate
0,0%
0,0%
2 771
5,5%
0,0%
22 402
22,4%
0,0%
66 819
33,4% 10 000
5,0%
145 398
41,5% 45 000
12,9%
505 398
53,2% 225 000
23,7%
22
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Corporate income tax
o
Two tax breaks for corporations, assumed to be regressive
•
•
o
Statutory rate decreased from 35% to 30%
Minimal 1% tax on sales in case of negative profits
Exemptions on corporate income tax are regressive loopholes
Fiscal losses (BIF bn)
% total revenues
% GDP
2009
1,94
0,64%
0,09%
2010
3,27
0,90%
0,13%
2011
11,21
2,38%
0,39%
2012
18,49
3,51%
0,53%
2013
9,19
1,64%
0,23%
23
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Burundi’s tax reforms since 2009
Regressive (assumed)
Progressive (assumed)
o
Corporate income tax
o
VAT intermediary rate
o
Personal income tax
o
Most excise fees
o
VAT exemptions
24
Section 3
Burundi – spending
side
25
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Pro-poor spending
o
Pro-poor spending growing much slower than tax revenues
As % of total revenues
2010
Fiscal revenues
Pro-poor spending
2011
2012
2013
As % of GDP
2010
2011
2012
2013
100% 100% 100% 100% 14,1% 16,3% 15,1% 14,1%
89,3% 78,8% 67,2% 66,0% 12,6% 12,8% 10,1% 9,3%
Absolute figures (BIF bn)
2010
Fiscal revenues
Pro-poor spending
o
2011
2012
2013
362,8 471,7 526,6 559,5
323,8 371,5 353,7 369,2
Pro-poor spending in Burundi:
•
•
All expenditures by social ministries (i.e. education, health, etc.)
“social projects” in other ministries
26
Section 4
Fiscal performance
indicators
27
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Indicators currently used in Burundi
o
Very few fiscal indicators are fitted to capture the pro-poor impact
•
•
Government: pro-poor aspect entirely absent
IFIs: mostly capture the spending side
PRSP
Government
indicators
Doing
Business
'Paying Taxes'
Total revenues (% PIB)
Ratio of yearly revenues to yearly target
Ratio of yearly revenues to state spending
Tax payments for a manufacturing company
Time required to comply with 3 major taxes
Total tax rate (% of profit before tax)
Fiscal Policy
Efficiency of Revenue Mobilization
CPIA
Quality of Budgetary and Financial Management
Equity of Public Resource Use
Business Regulatory Environment
IMF
IMF chapter 4 (2012)
publications 4th review under credit facility Feb. 2014
Pro-poor agenda
Spending
Tax side
side
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
Yes
Yes
No
No
No
Yes
No
Yes
28
1. Literature review
2. Burundi: tax side
3. Burundi: spending
4. Suggested indicators
Suggested indicators
o
Suggested indicators with two views:
•
•
Follow up on pro-poor impact of tax reforms (tax side and spending side)
Be available subject to data scarcity in a country like Burundi
2010 2011 2012 2013
Set 1 -- Indirect taxes: VAT, excise
1.1 VAT revenue a s a s ha re of tota l revenue
1.2 VAT revenue l evi ed on l ower ra te a s % of tota l VAT revenue
1.3 Exci s e l evi ed on progres s i ve products a s % tota l exci s e revenue
36,5%
36,5%
36,5%
37,7%
N/A
N/A
N/A
N/A
72,7%
73,1%
86,5%
82,9%
14,3%
14,7%
15,7%
16,9%
0,9%
2,4%
3,5%
1,6%
Set 2 -- Direct taxes: corporate income (CI) tax
2.1 CI ta x revenue a s a s ha re of tota l revenue
2.2 Fi s ca l l os s es on CI exempti ons a s a s ha re of tota l revenue
Set 3 -- Direct taxes: personal income (PI) tax
3.1 PI ta x revenue a s a s ha re of tota l revenue
11,7%
10,7%
12,7%
7,7%
3.2 Effecti ve ta x ra te for monthl y ta xa bl e i ncome of BIF 350 000
41,5%
41,5%
41,5%
23,7%
89,3%
78,8%
67,2%
66,0%
Set 4 -- Pro-poor state spending
4.1 Pro-poor s ta te s pendi ng a s % revenue
29
Conclusion
30
Conclusion
o
In an LIC like Burundi, there is a partial disconnect between revenue
reforms and poverty reduction
o
Tax side: the risk of regressivity is relatively higher
•
o
o
Narrow income tax bases
Spending side: impractical to compensate for tax regressivity with
increased pro-poor spending
•
Revenue reforms are not precisely coupled with allocation reforms
•
The case of Burundi illustrates this difficulty
Measurement indicators are not fitted to take into account the pro-poor
impact of revenue reforms
•
Neither tax side, nor spending side
Thank you!

32
Bibliography
o
Bastagli, F., Coady, D., & Gupta, S. (2012). Income Inequality and Fiscal Policy. IMF.
o
Daniels, R., & Edwards, L. (2006). The Benefit-Incidence of Tariff Liberalisation in South Africa.
o
Engel, E., Galetovic, A., & Raddatz, C. (1999). Taxes and income distribution in Chile: some unpleasant
redistributive arithmetic. Journal of Development Economics, 155-192.
o
Go, D., Kearney, M., Robinson, S., & Thierfelder, K. (2005). An Analysis of South Africa’s Value Added
Tax. World Bank.
o
Harberger, A. (1962). The Incidence of Corporation Income Tax. The Journal of Political Economy, 215240.
o
IMF. (2012). Article IV Consultation and First Review Under the Extended Credit Facility.
o
IMF. (2014). Fourth Review Under The Extended Credit Facility Arrangement.
o
IMF Fiscal Affairs Department. (2011). Revenue Mobilization in Developing Countries.
o
Keen, M., & Mansour, M. (2009). Revenue Mobilization in Sub-Saharan Africa: Challenges from
Globalization. IMF Working Paper.
o
Marshall, J. (2009). One size fits all? IMF tax policy in Sub Saharan Africa. Christian Aid.
o
World Bank. (2011). CPIA 2011 Criteria.
o
Zolt, E., & Bird, R. (2005). Redistribution via Taxation: The Limited Role of the Personal Income Tax in
Developing Countries. UCLA Law Review, 1627–95.
Appendix 1 – Total expenditure
Pagne
Frais de correspondance postale Choux
Robe
Paie des domestiques
Tomate
Chemises
Transferts à d’autres ménages
Carottes
Pantalon
Autres Dépenses
Oignons
Menteau
Haricot
Poireaux
Veste
Patate douce
Aubergine
Tricot
Banane a biere
Autres legumes
Tissus
Banane legume
Ananas
Chaussures
Pomme de terre
Avocats
Autres (vêtements et chaussures) Manioc tubercule
Citrons
Loyer
Manioc farine
Oranges
Réparation de maison
Riz
Mangues
Dépense d'ameublement
Petit pois
Autres fruits
Eau
Maïs
Boissons traditionnelles
Electricite
Sorgho
Bière et boissons alcoolisés
Dépenses en clôtures
Ble
Limonades
Cinéma
Sucre
Bois pour cuisine / charbon
Sports
Huile de cuisine
Pétrole
Lecture
Sel
Bougie
Dot et mariage
Colocase
Pile
Naissance et baptême
Pain
Savon
Autres cérémonies diverses
Ndagala et autres poissons
Allumettes
Frais de déplacement ou voyage
Viande
Brosse à dent
Carburant, lubrifiant et entretien Lait
Tabac
Téléphone, Fax, Télex, Internet
Miel
Restaurant
Radio, TV
OEuf