Does Money Work Better When It Is Spent Closer to the People?

1. The question behind the money

Every government budget eventually becomes something physical.

A classroom.
A road.
A water pipe.
A drainage system.
A bus route.
A hospital ward.
A waste truck.
A streetlight.

But the person using that thing usually has the least influence over the budget that created it.

That is the central puzzle of decentralization.

The province may know the road is needed.
The district may know exactly where.
The local government may know which street floods first.
The citizen may know why the previous project failed.

Who controls the money?

That question is more important than simply asking how much money government has.

2. Pakistan's local-government spending share is shrinking

The World Bank's 2026 fiscal-federalism review reports that local governments accounted for roughly 10% of total government spending in 2005 but less than 5% in 2024.

That is not a tiny adjustment.

It means the share of public expenditure directly associated with the local tier has at least roughly halved.

The number matters because local governments are responsible for many services that people experience at street level.

If their share of public spending falls, either:

1. responsibilities are also falling;
2. higher tiers are performing local functions;
3. local services are under-financed;
4. or some combination of all three.

Those are very different institutional situations.

A proper reform programme needs to identify which one exists for each service.

3. The province is often the middle of the problem

Pakistan's post-2010 system is often described through the federal-provincial relationship.

The 18th Amendment and the 7th NFC Award did indeed move substantial resources and responsibilities toward provinces.

But the World Bank's 2026 analysis makes a crucial observation: provincial revenue increased substantially after devolution, while federal expenditure did not fall commensurately, and the devolution of spending responsibilities has not always been matched by effective fiscal and institutional arrangements at lower levels.

This creates a paradox.

The state successfully moved some power:

Islamabad → provincial capitals

but did not complete the next movement:

provincial capitals → local governments.

That may be why citizens can still experience a state that feels centralized even in a constitutionally decentralized federation.

4. The last mile of government

Think about plumbing.

Water entering a city from a reservoir is important.

But a household doesn't care how elegantly the reservoir is managed if the last pipe is broken.

Public finance can work the same way.

The federal government may transfer money.
The provincial government may budget it.
A department may approve it.
A development scheme may exist on paper.

But the citizen experiences only:

Does water come out of the tap?

That is why the “last mile” of public finance matters.

Local government sits closer to that last mile.

5. Provincial budgets reveal another problem: too much of the money pays for the system itself

The World Bank reports that more than 80% of consolidated provincial expenditure in FY2023 was recurrent.

Recurrent expenditure is not necessarily waste.

Teachers need salaries.
Doctors need salaries.
Police need salaries.
Hospitals need electricity.
Government buildings need maintenance.
Pensions and other obligations must be paid.

But when most available spending is locked into recurrent costs, the room for new infrastructure and investment becomes limited.

That makes the quality of allocation even more important.

If local governments have little predictable investment funding, even a competent mayor can do very little.

6. International comparison: what does “local” mean in richer countries?

OECD data provides a useful benchmark.

Across OECD countries, subnational governments account for approximately:

• 40% of total public expenditure
• 55% of public investment
• more than half of public staff expenditure

Again, this is not proof that decentralization causes countries to become rich.

Most OECD countries also have:

• much stronger institutions;
• larger tax bases;
• more mature civil services;
• better data;
• stronger municipal systems.

So comparing percentages alone can be misleading.

But the comparison establishes something important:

Large-scale fiscal decentralization is compatible with an effective modern state.

The question is whether Pakistan can build the institutional conditions that make it work.

7. A local government is not automatically democratic in practice

A local council can be elected and still be powerless.

Suppose citizens elect a mayor.
But the mayor cannot control:

• the budget;
• staff;
• procurement;
• planning;
• water utilities;
• waste contracts.

Then elections exist without much governing authority.

This is one of the key distinctions between:

electoral decentralization
and
functional decentralization.

Article 140A is important because it refers not merely to elections but to devolution of political, administrative and financial responsibility and authority.

The three parts belong together.

8. Money without authority is not enough

Now reverse it.

Suppose a municipality receives money but cannot:

• hire technical staff;
• plan infrastructure;
• enforce building rules;
• collect revenue;
• maintain assets.

The money will not automatically translate into services.

So there are actually three pieces:

authority
capacity
money

A fourth is:
accountability.

Without accountability, local spending can simply become local patronage.

A fifth is:
information.

A local authority cannot plan effectively if it does not know what it owns, where assets are failing or what residents need.

9. Punjab's intermediate cities offer a useful real-world case

The World Bank's Punjab Cities Programme is one of the most valuable current examples because it moves beyond theory.

It covered 16 intermediate cities and reports that more than 4.5 million people benefited, with women representing about 49% of beneficiaries.

The programme supported:

• infrastructure upgrades;
• three-year rolling development plans;
• GIS-based municipal asset inventories;
• improved municipal management;
• own-source revenue reforms;
• performance monitoring.

The reported performance score across participating cities averaged 90 against a target of 70.

Own-source revenue nearly tripled from roughly US$6.2 million in 2017–18 to US$17.6 million in 2023–24.

This is not evidence that every local-government programme works.

It is evidence that local institutional capacity can improve when cities receive structured support, planning tools and fiscal reforms.

10. The significance of the revenue increase

Own-source revenue is important because dependency changes incentives.

Imagine two municipalities.

Municipality A

Receives almost all its money through discretionary transfers.

Municipality B

Collects a meaningful share through local revenue.

B has an incentive to care about:

• the tax base;
• businesses;
• property records;
• service quality;
• compliance;
• visible results.

Citizens also have a more direct question:

I pay this municipality; what am I getting?

That can improve accountability.

But local taxation can also be regressive or politically unpopular.

So the answer is not:

“Make municipalities tax everything.”

It is:

Give local governments enough predictable own-source revenue and transfers to match the responsibilities they actually hold.

11. The transfer problem

The World Bank argues that local transfers in Pakistan need to be more predictable.

This is critical.

Suppose a municipality knows it will receive:

Rs5 billion every year under a transparent formula.

It can plan.
It can hire.
It can contract.
It can maintain assets.

Now suppose:

“The province may release funds later.”

The municipality cannot plan the same way.

Projects become delayed.
Procurement becomes fragmented.
Maintenance gets postponed.

The same amount of money can therefore produce different outcomes depending on whether it arrives predictably.

12. The paradox of development schemes

A higher-level politician can take credit for a road.

But the local government may be responsible for maintaining it.

This creates a mismatch:

credit at one level
cost at another.

It becomes politically rational to build a new thing and less attractive to pay for maintenance.

Local governments often have the opposite problem.

They inherit:

pipes + roads + drains + markets

but lack enough money to maintain them.

This is why maintenance funding should be protected from the glamour of new projects.

13. Spending closer to people can improve information

Local officials have one major advantage:

proximity.

A provincial department may know that:

District X needs drainage investment.

A local authority may know:

Street 14 floods every monsoon because the culvert is blocked and the road was resurfaced without correcting the slope.

Information improves when the responsible institution is close enough to observe the problem.

This is one of the strongest theoretical arguments for decentralization.

It is also why decentralization should not mean:

“everything local, no standards.”

Central government can set standards.
Local government can adapt implementation.

14. The case for a hybrid model

The best system is probably not:

all federal
or:
all provincial
or:
all local.

It is a hierarchy of responsibilities.

Federal

• defence;
• national taxation;
• macroeconomic stability;
• national infrastructure;
• national standards.

Provincial

• secondary/higher services where scale matters;
• large hospitals;
• regional highways;
• universities;
• provincial regulation.

Local

• sanitation;
• local roads;
• waste;
• water distribution;
• municipal planning;
• local markets;
• street-level infrastructure.

The exact boundaries require empirical study.

But the principle is:

assign each function to the lowest level capable of delivering it efficiently while preserving national standards and fiscal equalisation.

15. Why small local governments can also fail

Criticism of decentralization is not irrational.

A municipality may have:

• too few taxpayers;
• weak staff;
• poor procurement systems;
• elite capture;
• corruption;
• political instability.

Two neighbouring municipalities may duplicate expensive services.

A small tax base may not sustain a modern hospital.

This is why decentralization needs scale-aware design.

Some services should be shared across municipalities.
Others should remain provincial.

The answer is not to make every government unit completely independent.

16. The best local system may be networked

Imagine:

Municipality A
Municipality B
Municipality C

sharing:

• waste treatment;
• hospitals;
• specialist technical staff;
• procurement;
• water reservoirs.

Each city still controls its local responsibilities.

But expensive facilities operate at larger scale.

This is common in mature public systems: decentralization does not have to mean isolation.

It can mean:

local decisions + shared regional infrastructure.

17. What Pakistan's own city reforms teach us

The Punjab Cities Programme highlights something important that is often ignored in political debate.

Local government needs systems, not only money.

GIS-based asset inventories matter because a city cannot maintain what it does not know it owns.

Three-year rolling plans matter because annual political cycles encourage short-term projects.

Performance scores matter because funding without measurable outcomes is difficult to evaluate.

Revenue systems matter because dependency weakens financial planning.

This is a complete institutional package.

18. The citizen sees results, not constitutional clauses

A local-government reform should eventually be visible as:

fewer flooded roads;

faster garbage collection;

more reliable water;

faster permits;

cleaner markets;

better-maintained streets.

A constitutional amendment matters because it changes the machinery behind these outcomes.

But the citizen judges the machinery through everyday life.

That is why decentralization research must remain connected to service outcomes.

19. What happens when money stays far away?

The opposite problem is concentration.

If all meaningful development decisions remain above the local level:

information travels upward
decisions travel downward
time is lost
local priorities compete with broader political priorities

and the person closest to the problem has the least control.

This creates the same administrative-distance problem examined elsewhere in the PPRI programme.

Money can be technically “in the province” while still being administratively distant from the neighbourhood.

20. Why the political incentives are difficult

Giving local governments more money also gives local leaders more political visibility.

A provincial politician may ask:

Why should I finance a mayor who will receive the credit?

That incentive does not prove every provincial government deliberately blocks local government.

But it is a real institutional question.

If a system is designed so that higher-level politicians receive the political reward while lower levels carry the implementation burden, decentralization will remain fragile.

Good institutional design has to align:

responsibility
with:
authority
with:
money
with:
political accountability.

21. Pakistan's fiscal chain should be visible

PPRI proposes a:

Pakistan Public Money Chain

For every district/city where data permits:

Federal revenues
↓
NFC / federal transfer
↓
Provincial receipt
↓
Provincial allocation
↓
District/local transfer
↓
Actual expenditure
↓
Project/service
↓
citizen outcome

This can expose where money is:

• allocated;
• released;
• delayed;
• reallocated;
• spent;
• under-spent.

The chain is much more useful than a single provincial budget number.

22. The difference between allocation and spending

A common mistake in public debate is:

“Rs100 billion was allocated.”

That doesn't mean:

Rs100 billion was spent.

And even:

Rs100 billion was spent

doesn't mean:

Rs100 billion produced Rs100 billion worth of public benefit.

We need:

allocation
→ release
→ expenditure
→ physical completion
→ service use
→ outcome.

That is what good fiscal research follows.

23. The geography of public spending

The World Bank's 2026 analysis finds that per-person public spending remains substantially higher in some provincial capitals than in other districts:

• Quetta: roughly Rs57,000 vs Rs12,000 elsewhere in Balochistan;
• Lahore: roughly Rs31,000 vs Rs7,000 elsewhere in Punjab;
• Peshawar: roughly Rs35,000 vs Rs10,000 elsewhere in KP;
• Karachi: roughly Rs25,000 vs Rs14,000 elsewhere in Sindh.

These differences should not be mechanically labelled waste or favoritism.

Provincial capitals host:

• secretariats;
• teaching hospitals;
• universities;
• major roads;
• police headquarters;
• provincial institutions.

Some spending serves people from outside the capital.

But the pattern raises an important empirical question:

How much of the difference is explained by functions concentrated in capitals, and how much represents geographic imbalance in access to public investment?

That is precisely the question PPRI should investigate.

24. Local government and the province debate

Here is where the argument becomes interesting.

If citizens in a region feel ignored, there are at least three possible institutional responses:

Option A

Create a new province.

Option B

Create smaller administrative regions within the province.

Option C

Empower local and regional governments.

There may be circumstances where one is more appropriate than another.

A new province can create:

• a new political centre;
• new budget authority;
• stronger representation;
• regional institutions.

But local government can address problems much closer to the service-delivery point.

The policy mistake would be to treat these as mutually exclusive.

25. What if Pakistan pursued both?

One possible reform architecture is:

fewer giant provinces
+
strong local governments
+
regional cities with fiscal authority
+
transparent equalisation transfers.

Then:

FEDERATION
     │
PROVINCE
     │
REGION/CITY
     │
LOCAL GOVERNMENT
     │
CITIZEN

Each level has a defined purpose.

The aim is not maximum fragmentation.

It is minimum unnecessary distance between responsibility and the problem.

26. What the international evidence actually permits us to say

OECD countries demonstrate that substantial subnational spending and investment can coexist with prosperous, functional states.

The Punjab Cities Programme demonstrates that better municipal systems can generate measurable improvements in Pakistan.

Neither proves:

“Decentralization causes prosperity.”

The reasonable conclusion is narrower and more useful:

Strong local government is compatible with effective governance, and Pakistan has domestic evidence that improved local capacity can produce better municipal performance.

That is enough to justify serious reform research.

27. The wrong question

The wrong question is:

Should Pakistan decentralize?

That is too broad.

A better set of questions is:

Which functions should be local?
Which need provincial scale?
Which require federal scale?
How much money should follow each responsibility?
What should the transfer formula be?
Which taxes should be local?
How should equalisation work?
How should citizens monitor spending?
What happens when local officials fail?

Those questions can be answered empirically.

28. A PPRI framework: the Responsibility-Money-Accountability test

For every function, PPRI proposes testing four things.

Responsibility

Who is expected to deliver?

Money

Who controls the funds?

Authority

Who can make the decision?

Accountability

Who can the citizen hold responsible?

A system is structurally weak when:

A is responsible.
B has the money.
C has the authority.
D gets the blame.

That is when citizens hear:

“It isn't our department.”

Good government requires fewer such dead ends.

29. A second test: proximity versus scale

For each service, ask:

Does proximity matter?

If yes, local government may be appropriate.

Does scale matter?

If yes, provincial/regional provision may be more efficient.

Does equity matter?

If yes, national/provincial equalisation may be necessary.

Does national standardization matter?

If yes, federal/provincial standards should remain strong.

This gives a practical framework rather than an ideological one.

30. The fiscal settlement Pakistan needs

Pakistan's debate often focuses on how much money should go from the federation to provinces.

The next settlement should also answer:

How much goes from provinces to local governments?

And:

By what formula?

And:

Can local governments rely on it?

The World Bank specifically recommends a more predictable local transfer system and better alignment of finance with assigned responsibilities.

PIDE's recent fiscal-federalism work reaches a similar practical problem from another angle: provincial governments remain heavily dependent on transfers while local governments remain financially constrained.

The missing piece is therefore not simply “more money.”

It is:

predictable, transparent money attached to clearly assigned responsibilities.

31. A local-government budget should be boring

This may sound strange.

But good finance is boring.

Citizens should be able to open a dashboard and see:

Rs X received
Rs Y spent
Rs Z committed
Rs A remaining

and every large project.

If the budget requires a political insider to understand it, accountability is weak.

A boring budget is a beautiful thing.

32. A future PPRI Local Finance Observatory

PPRI could eventually create:

Local revenues

property tax, fees, charges.

Transfers

federal/provincial/local.

Expenditure

recurrent/development.

Projects

location, cost, contractor, completion.

Outcomes

service indicators.

Citizen feedback

complaints, satisfaction.

Then compare cities across Pakistan.

This would turn fiscal decentralization from a constitutional argument into something citizens can inspect.

Conclusion: money does not become local just because the province received it

Pakistan's fiscal debate often stops at the provincial boundary.

That is too high in the system.

A province may receive billions from the federation while a municipal government struggles to repair a drain.

A province may control a large development budget while a district cannot reliably plan three years ahead.

The question is not whether provinces deserve resources.

They do.

The deeper question is:

How much public money should remain at the provincial level once the people who actually deliver local services are identified?

International evidence shows that substantial subnational fiscal responsibility is compatible with effective government.

Pakistan's own Punjab Cities Programme shows that municipal institutions can improve when planning, asset management, revenue and performance systems are strengthened.

And the World Bank's 2026 assessment shows a warning sign in the opposite direction: local governments' share of total public spending has fallen to under 5%.

The answer therefore is not:

“Give everything to local government.”

It is:

Put money, authority and accountability at the same level wherever practical.

Where scale matters, keep scale.
Where standards matter, keep standards.
Where proximity matters, move the decision closer to the citizen.

A government that lives far above the problem will eventually need more paperwork to discover what the person living beside the problem already knows.

The purpose of decentralization is to shorten that distance.

The question Pakistan now has to answer is not whether it can afford local government.

It is:

Can it afford not to have one that works?