The impacts of external funding for district governments on tax - - PowerPoint PPT Presentation

The impacts of external funding for district governments on tax collection and public goods provision in Ghana

Edward Asiedu (University of Ghana) Dan Pavelesku (World Bank) Ryoko Sato (World Bank) Tomomi Tanaka (World Bank)

Research Questions

1) Do external revenues crowd in or crowd out internal revenues? Do the effects depend on poverty rates of districts? 2) When revenues rise, which category of expenditures do district governments increase? 3) Does an growth in revenues/expenditures lead to increased provisions of public goods?

Summary Statistics - Revenues

• Poor districts: smaller

revenues.

• Gap between poor & rich

districts:

• Narrowed for External revenue.
• Widened for Internal revenue.
• Rich districts increased internal

revenue and became less dependent on external revenue compared with poor districts.

500 1000 1500 2000 2500 3000

1995 2007 1995 2007 External Revenue Internal Revenue

Revenues in 1995 and 2007

Richest Poorest

Summary Statistics - Expenditure

• Poor districts: smaller

expenditure.

• William (2016): 29% of capital

expenditure is wasted; many district projects are never completed. Capital expenditure declined, esp in poor districts.

• Personal expenditure

increased, esp in poor districts. Could be due to ghost workers in the public sector.

100 200 300 400 500 600 700 800 900 1000

1995 2007 1995 2007 1995 2007 1995 2007 1995 2007 1995 2007 Personal Transport General Repairs Misc Capital

Expenditure over time

Richest Poorest

Identification Strategy for Impacts of external revenue change on internal revenue collection

• Specification (Mogues and Benin, 2012; growth model)
• ∆[𝑚𝑜(𝐽𝐻𝐺𝑗𝑢)] : change in the log of internal revenue (IGF) from the time t-1 to the time t
• 𝐹𝑌𝑈𝑗𝑢−1 : lagged external revenues for district i
• 𝐹𝑌𝑄𝑗𝑢−1 : vector of lagged expenditures at time t-1
• 𝐸𝑢 : time effects
• ηi + εit : error terms

∆[𝑚𝑜(𝐽𝐻𝐺

𝑗𝑢)]

= ln(𝐹𝑌𝑈𝑗𝑢−1)𝜀 + ln(𝐹𝑌𝑄𝑗𝑢−1)𝛿 + ∆𝑚𝑜(𝐽𝐻𝐺

𝑗𝑢−1)𝜚 + 𝑆𝑗𝜇 + 𝐸𝑢𝜐 + 𝛽 + 𝜃𝑗 + 𝜁𝑗𝑢

Impacts of external revenue change on internal revenue collection

• The increased external revenue leads to

reduced internal revenue from market fees (from market traders) & business licenses: crowding-out

• In the poorest districts: collection of

internal revenue does not respond to the increase in external revenue

• 0.06
• 0.05
• 0.04
• 0.03
• 0.02
• 0.01

Effects of Lagged External Revenues on Internal Revenues

• 0.2
• 0.15
• 0.1
• 0.05

Non-poorest Poorest

by Poverty Rate

Rates Lands Fees Licenses Rental Invest Misc. IGF Total

Impacts of revenue change on expenditures

Increased revenues raised expenditure in

• capital expenditure (could be

wasteful; William, 2016)

• Personnel expenditure (could be

wasteful due to a large number of ghost workers in the public sector)

50 100

Internal Revenue External Revenue

Effects of Change in Revenue on Change in Expenditure

Personal Transport General Repairs Misc Capital

100 200 300

Internal Revenue External Revenue Internal Revenue External Revenue Non-Poorest Poorest

by Poverty Rate

Personal Transport General Repairs Misc Capital

Impacts of revenue change on public goods provision

• No effect of expenditure on public

goods provision (no difference by poverty rate) : could be due to inefficient allocation of resource.

• Increase in external revenues, but

not internal revenues, increases some public goods provision.

• Non-poor districts: liquid waste

infrastructure

• Poorest districts: piped water

0.01 0.02 0.03 0.04 0.05 0.06 Internal Revenue External Revenue Internal Revenue External Revenue Non-poorest Poorest

Effects of Revenues on Pubic Goods Provision by Poverty rate

% of HH with piped water % HH with solid waste infrastructure % HH with liquid waste infrastructure

Conclusion

• Large disparity between poor and rich districts in their ability to raise internal

revenue. Rich districts were able to develop tax bases in the private sector, while poor districts were not.

• External revenue crowds out internal revenue in general, but not among poor

districts. Policy implication: Allocating more external funding to poor districts could be more effective.

• External revenue is more effective in increasing public goods provision than internal

revenue.