Can the New Intergovernmental Structure Work in Pakistan? Learning From China
December 27, 2013
1.
Introduction
The 18th
Amendment to the 1973 Constitution of Pakistan disentangles overlapping
spending responsibilities between the federation and provinces in a wide range
of functions, devolving them to the latter. The legislation was also a reaction
to relatively poor service delivery and living standards that had fallen continuously
behind those in other countries in South Asia, and, indeed, are now lower than
sub-Saharan Africa in most respects.
The Musharraf
government had used this argument for its own decentralization
effort—delegating power to the districts and bypassing the political centers of
power in the provinces. The 18th Amendment reasserts the provinces’ power
and the associated political centers of power. It is designed to weaken the
center, and correspondingly make it less attractive for the military to assume
power by moving against an elected Prime Minister, as it has done periodically
in Pakistan’s history.
But will this
major reform work effectively and ensure higher living standards for all people
in all the provinces? To what extent is the need for a national identity
important in ensuring that the decentralization does not cause the federation
to unravel or the overall delivery of public services to deteriorate and lead
to greater exclusion of the poor? These are important issues and could well
determine the fate of the 18th Amendment as well as social stability
in Pakistan.
Section 2 outlines
developments in theory linking governance and the decentralization process. The
links between the two are critical.[1] The main
question is whether decentralized service provision can better provide for the
poorer sections of society by utilizing information that may be available at
the local level in tailoring services to local preferences and making access
easier. How are these responsibilities financed? Does the process impede closer
economic integration between the federating provinces? This chapter argues that
positive approaches to intergovernmental reforms, as exemplified by the
People’s Republic of China, are perhaps more important for countries such as Pakistan
that face significant structural challenges.
Section 3 focuses
on the 7th National Finance Commission (NFC) Award and key elements
of the 18th Amendment. Their components are examined in relation to
the main criteria for good governance, utilizing positive approaches to
institutional and multilevel governance reforms. The section argues that the
failure of tax reforms poses serious difficulties both for the NFC and the
stability of assignments arising from the 18th Amendment. These
accentuate the danger of Pakistan becoming a “failed state”, highlighted by
commentators prior to the NFC award or the 18th Amendment (see
Haque, 2009).
In order to
prevent dire consequences arising from the NFC award and the 18th
Amendment in the presence of failed tax reforms and opaque governance, Section 4
outlines an agenda for urgent action that revisits archaic and inefficient
revenue assignments dating back to the Government of India Act 1935. The
chapter concludes with some proposals that might be considered in the context
of the next NFC award, and possibilities for another constitutional amendment on
the revenue side to parallel that on the spending side. Joint action on tax
policy and assignments, as well as the complete overhaul of the country’s tax
administration framework are needed.
2.
Normative or Positive
approaches to Federalism and Decentralization?[2]
This section sets
the stage by reviewing the normative and positive approaches to
decentralization, and examining theory and evidence, including the example of
China. It also draws some lessons regarding preconditions for good governance
in a decentralized framework, and links these to the Pakistan context.
2.1.
Developments in Theory
The post-Second World War normative literature on fiscal
decentralization has been much influenced by the experience of the US, and the
work inter alia of Musgrave (1959) and Olson (1969). These were based on the assumption
that governments are benevolent. This reflected the views of Montesquieu, and
of Hamilton and Madison in the Federalist Papers, that a government should be
small and its functions separated, with the center responsible for issues that
affect all lower levels of government, such as defense and monetary policy. The
assumption has been, particularly on the part of some bilateral and
multilateral agencies, that decentralization leads to more efficient service
delivery, higher growth, and poverty reduction.
Experiences
outside the US, particularly in the European Union (EU) and especially in
developing countries, have led critics to question the normative approach,
spawning a surge in the “political economy” literature (see surveys in Ahmad &
Brosio, 2006; Oates, 2008; Lockwood, 2009). This reflects an earlier debate
associated with De Tocqueville and John Stuart Mill, which focused on the
actual workings of government and an evaluation of the pros and cons of
“decentralized” operations. The main difference is that the assumption of
“benevolent” government is dropped, and that incentives facing politicians and
bureaucrats become important as does the role of institutions and information
flows.
Bardhan and
Mookherjee (2000) write about the possibility of “capture” by vested interests
(see also Ahmad & Brosio, 2011). Besley and Case (1995) introduce the
concept of “yardstick competition” in which voters evaluate the performance of
their local government in relation to the results achieved in neighboring
jurisdictions. Given increasing mobility and information flows, the yardstick
competition idea has recently been extended to relate to countries, as citizens
in one country examine what results are achieved in other countries with which
they are familiar (Salmon, 2010; Besely & Case, 1995).
The building
blocks of both the normative and positive traditions are similar—spending and
taxation assignments, design of transfers, debt management, and information
flows and instruments for implementation. However, the sequencing and mix of
these instruments can vary, as discussed below.
2.2.
Decentralization Trends
The impetus to decentralize has differed in many cases. In Latin America,
the shift from one-party or military rule has led to a resurgence of interest
in decentralization as a means of consolidating political gains, whereas China—a
large, one-party, unitary state—has actually been quite decentralized. The
cross-country push to decentralize, supported by international agencies, is in
line with the normative approach to decentralization on the grounds that it would
lead to better service delivery and poverty reduction. However, the evidence on
this is at best mixed (see Ahmad, Brosio, & Tanzi, 2008, for a discussion on
the evidence in OECD countries). Table 8.1 gives some trends from Ahmad and
Brosio (2009).
Table 8.1: Main traits of recent
intergovernmental reforms in selected countries
|
Country
|
Main characteristics of intergovernmental relations
|
Comments
|
|
Australia
|
Federal system
|
Center
administers VAT on behalf of states; reforms introduced in early 2000s.
|
|
Belgium
|
Federalization based on
linguistic divisions
|
Transformed
from unitary to federal state
|
|
Bolivia
|
Three-layered unitary
system
|
Municipalities’
powers considerably increased. Election of governors of departments, some
demanding substantial but asymmetrical powers—associated with natural
resources.
|
|
Brazil
|
Federal
system based on three layers of
government
|
National
reform and coordination of VAT is an urgent priority, although proposals for
reform since the late 1990s have not been acted on.
|
|
Canada
|
Federal system
|
Asymmetric
federation (special treatment for Quebec)
|
|
China
|
Highly
decentralized system within a unitary constitution; operates like a quasi-federation
|
Taxing
power recentralized (1994)
|
|
Colombia
|
Three-layered unitary
system
|
Extensive
devolution of resources to provinces (departments); movement toward a
quasi-federation.
|
|
Denmark
|
Unitary system with strong
municipal governments
|
Recentralization
of higher education and health since 2006
|
|
France
|
Regional system
|
Regulatory,
fiscal, and political decentralization initiated
|
|
Germany
|
Federal system with
extended concurrent responsibilities
|
Reforms
to the federal structure initiated in a wide-ranging set of issues, but
little change effected as a result of two commissions.
|
|
Indonesia
|
Unitary state
|
Extensive
decentralization of spending powers to district-level administrations after
the fall of the Suharto administration, accompanied by a new revenue-sharing
arrangement.
|
|
Italy
|
Unitary, with asymmetric
arrangements
|
Fiscal,
regulatory, and political decentralization initiated with a new constitution
|
|
Mexico
|
Federal system with high
political and low fiscal decentralization
|
Fiscal
and regulatory decentralization since late 1980s, with basic education (1992)
and healthcare (1996) devolved to states, although revenues have been
effectively centralized since early 1980s.
|
|
Pakistan
|
Federal constitution
with interludes of military rule
|
Deconcentration
to districts in the early 2000s by
Musharraf.
Overlapping
responsibilities on the spending side
unwound with the 18th Amendment—most spending powers fully
assigned to provinces. Relatively little
subnational reliance on own-source revenues.
|
|
Peru
|
Unitary state—moving toward
a quasi-federation?
|
Election
of governors of regions—sharing of natural resource revenues. Overlapping
responsibilities with relatively limited spending or revenue devolution.
|
|
Poland
|
Unitary
|
Political and fiscal
decentralization with emphasis on the local level
|
|
South Africa
|
Post-Apartheid constitution
introduced a quasi-federal system
|
Devolution of extensive responsibilities
for education and health to provinces
|
|
Spain
|
Regional, quasi-federal
system
|
Transition toward a federal
system. Fiscal equalization with own-source
revenues at subnational level. Asymmetric assignments for some
regions.
|
|
Switzerland
|
Federal system
|
Equalization transfers from
federation to cantons
|
|
UK
|
Regional
|
Introduction of regional
government in Scotland and Wales
|
Source: Ahmad and Brosio (2009).
As in
Pakistan, many Latin American countries have experienced some movement toward
decentralization over the last two decades, often as a reaction to periods of
one-party or military rule. This has been more marked on the spending side than
on the revenue side. With respect to the latter, the trend has taken the
opposite direction: countries have established more or less centralized systems
for implementing value-added tax (VAT)—sometimes with the help of international
agencies and particularly the International Monetary Fund (IMF)—often replacing
myriad subnational taxes at the state and local levels.
Despite the rhetoric, the approach on the spending side—particularly in
the Latin American countries—has entailed mixed and overlapping
responsibilities that have not been adequately addressed. This partly reflects
the centralized tendencies of the past together with a paternalistic approach,
including by donors who do not trust subnational governments to make the right
choices for their citizens in their area of competence (including education and
social policy in general), or who feel that the lower levels lack the capability
to manage their affairs effectively.
The evidence
on the effects of decentralization regarding improvements in service delivery
in the OECD countries is mixed (see Ahmad et al., 2008). The evidence for
developing countries is not much more conclusive (see Ahmad & Brosio,
2009). The links between decentralization and preference matching and with growth
are often examined together. The studies confirm that any relationship, if it
can be established, is at best weak and tenuous.
Perhaps the
greatest lacuna in the decentralization processes of developing countries is
their lack of attention to adequate own-source revenues at the subnational
level. This may be due to the normative approaches that suggest focusing first
on the spending side, especially at the intermediate tier of
government/states/provinces/departments.
2.3.
Political Economy in Action: China
In the 1990s Chinese context of murky spending responsibilities—where state-owned
enterprises (SOEs) at different levels of government carried out a lot of
social spending—in a legal unitary state with no central tax collection other
than customs, the center had very limited ability to levy taxes. Economic
reforms in the 1980s had moved from a system of 100 percent profits taxation targeting
largely SOEs (collected by local governments on behalf of the center) to a more
moderate level of taxation. However, this caused the tax-to-GDP ratio to fall from
more than 22 percent to about 12 percent by 1993; more alarmingly, the central
government’s share of collection fell from just under 60 percent in the early
1980s to under 30 percent by 1993. This severely constrained the center from pursuing
macroeconomic and redistributive policy goals.
The debate at the time was whether the normative model of federal
reforms should be followed, i.e., to clarify spending responsibilities and then
adjust tax assignments accordingly—the “big bang” model also being used in
Russia. Alternative approaches supported the Chinese administration’s view that
it would be preferable to bolster central finances by establishing a state
administration of taxation (for the first time in Chinese history) responsible
for collecting modern taxes, particularly VAT (see Ahmad, Gao, & Tanzi,
1995). This view was accepted by the leadership, which was keen to avoid the difficulties
that were apparent by then in Russia, following the collapse of the Soviet
Union—another example of international yardstick competition.
The new tax-sharing system with a central
administration operated from 1994, and spending assignments were to be
addressed over time as the SOEs were gradually reformed.[3]
The VAT reforms in particular were spectacularly successful, raising the
central government’s share immediately and helping to bring the tax-to-GDP
ratio up toward 20 percent of GDP (see Figure 8.1). The interests of the local
governments in the tax reforms were protected by a “stop-loss” provision that
ensured that all local governments would get the amounts they had received in
1993 and the new system would be phased in. All local governments shared
incrementally as the new taxes were implemented (rich regions benefitted
concomitantly). Political economy concerns were protected by an equalization
system together with a Chinese innovation—“revenue returns”.
Figure 8.1: China: Total tax revenue,
local government revenue, and central government share of total revenue, 1985–2011
Source: Ahmad, Rydge, and Stern (2013).
A new
equalization transfer system was established similar to the most advanced in
the world (Ahmad et al., 1994; Ahmad, 1997, 2013) but its operations were
phased in over time. In the short run, a “revenue returned” system was
constituted that “returned” resources to the regions generating them (over and
above) the revenue-sharing arrangement. While this was criticized as increasing
inequality, in reality it was an essential component of “protecting” overall
growth and investment. With the freeing up of the labor market (there were now
150 million migrants working in the coastal areas), it was essential to ensure
full employment and reduce poverty (Lou, 1997). This is again an example of a
pragmatic “positive” action to meet the specific circumstances of a country in
rapid transition, for which the normative models are of limited utility.
Reforms of the
budget, treasury, and reporting systems were also set in motion in the late
1990s in a sequence of measures to prepare for the operations of a modern
economy. A second phase of the reforms is now needed to clarify spending
responsibilities at the lower levels of government, and also to examine
own-source revenues and debt in a way that optimizes land and local resource
use.
The Chinese
reforms of 1993/94 are an excellent example of the positive approach to
intergovernmental issues in action, and the importance of a new tax
administration as well as a nondistortive tax, such as VAT. There was no
concern that VAT would either affect the poor or hurt investment or growth. In
fact, the form of VAT that was in operation for the first 15 years after implementation
was the “investment” type that does not provide credits or refunds for capital
purchases. This did not, however, seem to affect either investment or growth, which
were spectacularly high during this period. The move to a more normal consumption-type
VAT was initiated only recently as the need for efficiency became more pressing
and the scope for raising revenues efficiently to around 25 percent of GDP became
more difficult.
3.
Preconditions for Good
Governance
3.1.
Overall Strategy for
Sustainability
Normative
approaches to fiscal federalism emphasize the sequencing in which “finance
follows function.” There is considerable validity in this proposition,
especially when marginal changes are envisaged. This “recommendation” is
designed to avoid an unsustainable expansion in overall spending and with a
view to maintaining macroeconomic stability. It is also an argument that
countries should begin decentralization reforms by starting with a devolution
process on the spending side first, followed by the reassignment of revenues.
The problems
occur when finance does not follow function or when it encourages access to
irresponsible financing mechanisms, such as borrowing without controls or
accountability, or running up arrears. In such cases, a macroeconomic crisis is
likely because local governments have every incentive to borrow and pass the
buck to others. It is usually the central government that has to pick up the
pieces, as in Latin America in the 1990s. In other countries, such as Nigeria,
the absence of adequate local own-source revenues has meant that there was no
incentive for local governments to pay teachers under the devolution stipulated
by the new constitution, and the functions had to be moved up to a higher
level.
Additionally,
under certain political economy circumstances, as in China in the early 1990s,
it may make sense to start with the revenue side first to ensure that there are
adequate overall revenues to match the spending needs of general government,
i.e., of the central and subcentral governments and associated public sector
undertakings. The 1994 reforms that facilitated substantially decentralized spending
over the following two decades were predicated on an effort to consolidate
central revenues, including though VAT, accompanied by automatic redistribution
mechanisms such as revenue sharing, equalization, and revenue returns, all of
which served different purposes. Thus, a careful redesign of the transfer
system is critical if a major structural reform on the revenue side is to be attempted
in a multilevel or federal country.
It is
generally accepted that, in order to meet the Millennium Development Goals (MDGs),
a tax-to-GDP ratio of around 18 percent is necessary for all levels of
government or general government (Ahmad, 2013). India and China have worked
very hard to increase their tax-to-GDP ratios to around 20 percent of GDP, but
given their substantive investment needs in education and physical
infrastructure, and for a more environmentally friendly growth strategy, a tax-to-GDP
ratio in the range of 25 percent is more likely required (see estimate for
India in Figure 8.2; IMF, 2013).
Figure 8.2: General government revenue and GDP per capita, 2012
Note: The figure excludes oil exporters
and microstates.
Source: International Monetary Fund
(2013).
Pakistan’s tax-to-GDP
ratio was 14.5 percent in the early 1980s, and had declined to 10 percent by
2008 at the onset of the macroeconomic crisis that led to a mega-loan from the
IMF. This was predicated on fixing the tax system, especially the moth-eaten general
sales tax (GST) replete with exemptions and “holes” for 65-year-old infant
industries (to use a turn of phrase popularized by a prominent Pakistani
economist, Dr Nadeem Ul Haque) and for friends and relatives of those in power.
The tax reforms failed, leading to the suspension of the IMF program in 2011.
As the tax-to-GDP
ratio slid below 9 percent in 2009, Haque (2009) correctly pointed to the
dangers of Pakistan becoming a failed stated. This level of tax effort barely
finances debt servicing and defense, and leaves precious little for public
services or investment at any level of government. Under these circumstances, a
major structural shift involving a significant decentralization of spending to the
provincial governments—unbundling the parallel responsibilities of government—is
of little more consequence than shifting deck chairs on the Titanic.
3.2.
Spending Assignments
A useful
typology of spending responsibilities showing how different countries approach
these issues is given in Figure 8.3. It addresses the subsidiarity principle,
which states that assignments should be devolved to the lowest level capable of
effectively providing them. This is a general principle of the EU’s legal
framework, constraining the supranational level from legislation to areas where
action at the national, regional, or local levels is insufficient (see
“Consolidated Version of the Treaty”, 2002).
The concept has both legal and political ramifications. The focus is on
scale as well as effects, including externalities, on other jurisdictions; this
has given rise to actionable cases where there is a legal connotation, as in
the EU.[4] In
political terms, the concept of subsidiarity is often taken beyond the
multilevel government connotation to include the boundaries between the private
sector and the role of the state (at any level). The assumption, especially by
conservative commentators in the US, is that the private sector should be
encouraged to provide public services as far as possible because this is
expected to be more efficient than public provision.
Figure 8.3
shows the differing trends regarding the centralization/decentralization debate
in different countries or regions. Arguments for the decentralization of
functions are based largely on accountability and effective provision, given
the subsidiarity principles. However, it is not enough to legislate the
assignments—the lower levels have to have the capability as well as the
incentive to provide services. Both are linked closely to the financing issue
as well as incentives for effective provision. Thus, the argument that local
governments lack “capacity” is not strictly binding if they have the financial
resources to hire skilled workers.
Figure
8.3: Modified subsidiarity principles
Source: Adapted from Dafflon (2006).
An important
hypothesis governing accountability comes through the electoral process when
voters are able to assess the performance of their “elected” rulers in relation
to standards in neighboring jurisdictions (see Salmon 1987, 2006; Besley &
Case, 1995).[5]
Again, the incentives are critical and voters are more likely to be responsive
if, at the margin, local governments rely on own-source revenues over which
they control rates or bases.[6]
Offsetting the
decentralization trends are concerns that limit subsidiarity—mainly
externalities such as spillovers (including those with environmental
considerations), congestion, and economies of scale. Moreover, decentralization,
especially of resource bases, can exacerbate inequalities across regions and
limit the extent of interpersonal redistribution that might be feasible. In all
cases, the federal, central, or supranational agencies have a role to play in
coordinating and harmonizing essential policies.
The US and certain
other federations maintain a unified economic space facilitated by a “commerce
clause.” In the EU, the common external tariff and harmonization of
country-level VATs (see the EU Sixth Directive) ensure a common economic space to
minimize harmful competition. Thus, a combination of legal and regulatory
frameworks is essential to ensure equal treatment and opportunity. Again, for
this to work efficiently, full information is needed on who spends what and on the
buildup of assets and liabilities; as the recent EU experience illustrates,
inadequate attention to the standardized flow of information could jeopardize a
common economic space.
3.3.
Full Information on
Transactions, Including the Uses and Flow of Funds
A key element in
accountable governance is timely information on the sources and uses of funds
at all levels of government. This is critical for establishing benchmarks
against which the performance of governments—federal, provincial, and district—should
be evaluated. Typically, it involves using the IMF’s (2001) government financial
statistics (GFS) manual standards for economic classification (wages, social
contributions, interest, operations and maintenance, etc.), and the UN’s
Classification of Functions of Government (COFOG) for education, health, and
other functions. This should provide an indication of what revenues were
generated, what was budgeted, and what was spent. In principle, this
information for Pakistan should have been generated by the Project to Improve
Financial Reporting and Auditing (PIFRA) project, which has been implemented
over ten years at a cost exceeding USD 100 million (under a World Bank loan).
Most
governments provide economic and functional data for each level of administration
to the IMF’s GFS yearbook. Pakistan’s GFS page for the most recent yearbook only
covers information on the budgetary central government. Given that most
spending is at the subnational level or is carried out by agencies associated
with the central government, the data in the GFS yearbook is less than useless
for policy purposes. With declining outcomes in education and healthcare—where
Pakistan is now falling behind sub-Saharan Africa—and donors’ focus on the
social sectors, the absence of readily available information[7] in these
critical areas is a serious problem. This ensures that the electorates of the
districts and provinces, indeed of the country as a whole, are comparatively
uninformed about relative spending by the public sector in areas of key policy
importance.
In addition to the financial information on public spending that forms
the basis for evaluating governments at election time, it is important to have
information on the outcomes of spending by the public sector. Again, this is a
critical element in the operation of “yardstick competition” and in the
operation of the electoral process to discipline governments.
The cash-basis
of the budget process in Pakistan also poses problems: it permits “game-play”
by the respective governments in terms of pushing liabilities into the future
and bringing forward credits, e.g., the securitization of revenues or asset
sales that inflate short-term revenues. A typical mechanism for hiding
liabilities is to shift them on to SOEs. Astonishingly, in the early 1990s the
IMF agreed to ignore the liabilities being generated in the SOEs (see Ahmad &
Mohammed, 2013). This would not be appropriate under the IMF’s (2001) GFS
manual framework and may have contributed to the buildup of circular debt in
the country.
Public-private-partnerships
(PPPs) are an increasingly convenient vehicle for “kicking the can of
responsibility” down the road. This is true of developing and developed
countries alike and contributed significantly to the fiscal problem in the EU. It
has led to tighter accounting rules, especially the recognition of “public”
liabilities in PPPs. Consequently, the accounting rules regarding PPPs were also
tightened, leading to a need for provisioning to prevent mechanisms to
circumvent liabilities.
For full
accountability, it is not sufficient to be able to track and report on budgeted
amounts, eventually also focusing on the results of the spending. Equally
important is tracking government cash. Typically, countries maintain treasury
single accounts (TSAs) into which all public funds flow and from which all
spending is authorized. Although in certain cases, commercial bank accounts may
be needed to facilitate payments or receive revenues, these should not contain
balances. These zero-balance accounts are linked with the TSA for overnight
deposit of revenues or the reimbursement of authorized payments.
At the time
the government was considering an IMF program in spring 2008, Finance Minister
Dar asked for the government balance sheet to be drawn up. This indicated
government balances of around USD 10 billion in commercial bank accounts at low
or zero interest. Although the IMF’s 2008 program had a provision for the
establishment of a TSA, this has not happened—both military and political
governments are disinclined to be subject to the discipline and transparency of
a TSA. As for the banks, they are quite happy to lend the same money back to
the government as the deficit increases and is financed by bank borrowing.
With this
opaque system, it is hard to impose accountability at any level of government.
There is no information on what should be spent, is actually spent, or what is
happening to public funds. Yardstick competition is impossible with a poor
governance structure. As important as the tax reform has been in China, the
establishment of a GFS 2001-compliant budget framework, and of TSAs at the
central and provincial levels, has been equally important in instituting an
effective system of decentralized investment and governance.
3.4.
Own-Source
Revenues, Transfers, and Access to Credit
A critical part of the story of accountable governance at the subnational
level concerns “own-source” revenues. This relates to the ability of a lower-level
government to raise revenues by varying the rate of a reasonable tax base.
Thus, in North America, state and local governments are able to set the rates
of state or local income taxes using the federal tax base. This ensures that
additional revenues can be generated in case of need. There need not be a state
or local tax administration, and the federal tax administration could be used
to do the “heavy lifting” in relation to IT, cross-referencing information, and
audits. It can be thought of as a “piggy back” or co-occupancy of a tax base,
in case there are parallel administrations.
Note that
shared revenues are not strictly own-source revenues and are closer to
transfers, since subnational governments can do little to influence the rate or
base and are merely recipients of the revenues. Additionally, taxes administered
by the central tax administration can be considered own-source if the local
government is able to vary the rate. Analogously, if it is difficult to vary the
rate of a subnational tax base (such as the GST on services), that tax base is
not an effective “own-source” revenue.
Without an
effective own-source revenue handle, it is not possible to hold a subnational
government responsible for its debt or buildup of liabilities (Ambrosiano &
Bordignon, 2006). This would weaken hard budget constraints, if any, and reduce
local accountability. One of the biggest macroeconomic problems in Latin
America during the 1990s was countries’ uncontrolled subnational borrowing,
often from their own banks. Following the Brazilian lead in the late 1990s,
many countries have constituted fiscal responsibility legislation at the
subnational level. This too can, however, be oversold—such legislation is only
as good as the systems to monitor and report on the buildup of liabilities. Moreover,
hard budget constraints are critical and require, in turn, effective own-source
revenues.
Even if a
country has an assigned own-source revenue handle, lower levels of government
may have little incentive to use it if they have access to badly designed
transfers or credit for which the liability can be shifted to others. Thus, if
central transfers are a function of actual deficits at lower levels (called “fiscal
dentistry” in India; see Rao, 1998), they will have no incentive to use
own-source revenues or spend efficiently.
It is thus clear from theory and practice that good governance at the
subnational level is a complex set of policy measures where interaction matters
in terms of generating appropriate incentives for accountability. It is likely
that isolated reforms, such as for spending assignments in isolation, may not
work as anticipated and might even make matters worse.
4.
Challenges for
Pakistan: Stalled Reform Agenda
Although
Pakistan has always been a federal country, extensive periods of military rule
have led to perceptions of dominance by the center, despite attempts to
“decentralize”. In the 1960s, Field Marshal Ayub Khan’s administration
experimented with a system of “basic democracies”—setting up an electoral
college at the local level that also formed the basis of development activities
in their regions. This effort at political “deconcentration” was abolished
under the 1973 Constitution, which restored the rights and functions of the
provinces—the main subnational unit of governance under the Government of India
Act 1935 (the basis for both India and Pakistan’s constitutions after
independence).
General
Musharraf’s administration also promoted a form of so-called “decentralization”
in the post-9/11 period. While it was ostensibly a mechanism to move services
closer to the people and elected local officials, there was little attempt to
adjust spending assignments or financing arrangements. Although the process was
clearly an attempt to bypass the established political parties and power
centers in the provinces, bilateral donors and multilateral banks rushed to
support the process along with the Federal Board of Revenue (FBR)’s
institutional reforms and government financial information systems at all
levels of government. Each of these reforms had failed or was in significant
difficulty by the time Musharraf left office in 2008 (see Ahmad & Mohammed,
2013).
The impetus for the 18th Amendment was primarily a reaction
against a decade of military rule. It also came at a period of economic
distress, after food and oil price shocks had severely affected the stability
of an economy that had relied on capital inflows to generate growth and
neglected domestic resource mobilization. The government’s approach to the IMF
in 2008 was predicated on tax reforms—principally fixing the holes in the GST.
At the same time, the NFC met to work on the 2010–14 award, keeping in view the
provinces’ deplorable levels of spending on the social sectors, principally
education and healthcare. The Finance Division’s Poverty Reduction Strategy
Paper (PRSP) envisaged significant progress toward meeting the MDGs, for which
additional resources were to have been allocated to the provinces. This section
argues that all three sets of reforms were closely intertwined, and that the
failure of the tax reforms has seriously jeopardized both the NFC award and the
18th Amendment.
4.1.
Tax Reforms
The reform of
the tax administration has been recognized as a priority since the early 1980s
and the report of the Tax Reforms Commission headed by Qamar-ul Islam, which
had called the then Central Board of Revenue a hotbed of corruption and rent
seeking. A GST was introduced in 1990 under an IMF-supported program (but
brought in through the back door, when the entire sales tax act was replaced as
part of the finance bill). It was administered very arbitrarily, with the tax
administration treating it like a production excise (Ahmad, 2010b), setting
reference prices and continuing to give exemptions and preferences through a
system of administrative orders (SROs) that provided ample opportunity for the rent-seeking
and corrupt practices to continue. The ability to give preferences and exemptions
and reward specific groups, while threatening to punish others without
reference to Parliament, provided convenient handles to politicians of
successive weak administrations to make friends and influence people.
At the end of
the 1990s, a committee led by former World Bank official Shahid Husain
recommended the creation of an integrated revenue administration, using the
modern principles of self-assessment, an arm’s-length functional administration
with minimal contact with taxpayers, and consequently limited opportunities for
rent seeking. This was supported by a large World Bank loan to create the new
FBR on the Argentine Revenue Authority model.
By the spring
of 2008, the World Bank had classified the project as “unsatisfactory”. A functional
organization structure had not been created and an IT system was prepared
in-house that largely automated the old procedures. Additionally, key
productive structures had been removed from the GST net with domestic
zero-rating, largely to offset delays in refunds and ease pressure on these
sectors from an overvalued exchange rate in a manner that would not attract the
World Trade Organization’s attention. In order to appear “investor-friendly”,
the audit system had been effectively abandoned in 2004/05. It is no wonder
that the GST failed to raise revenues, as had been expected under the strategy
to replace tariffs by the GST (the plan had been to replicate the Singapore
strategy that had very effectively used this method). An attempt to revive the
project under the IMF’s 2008 program also failed, as discussed below.
By 2009, the GST’s
efficiency in Pakistan had fallen to around 0.26 (as measured by the
C-efficiency ratio; see Ahmad, 2010b, for more details), and collection had
declined to 3.1 percent of GDP from 3.9 percent in the 1990s (see Table 8.2).
If Pakistan were to achieve the C-efficiency of Sri Lanka (from around 2004 at
the height of the civil war), it would more than double the collection or reach
around 7 percent of GDP with a 15 percent rate. It is worth noting that the
taxation of goods and services in China generates around 9 percent of GDP
(Ahmad, Rydge, & Stern, 2013).
Table 8.2: GST productivity—declining and low in comparison with
competitors
|
Country
|
|
Standard rate
|
Revenue/GDP
|
Productivity
|
|
Pakistan
|
(1990s)
|
15
|
|
0.39
|
|
Pakistan
|
(2005)
|
15
|
3.4
|
0.30
|
|
Pakistan
|
(2009)
|
16
|
3.1
|
0.26
|
|
Sri
Lanka
|
|
15
|
6.7
|
0.47
|
|
Philippines
|
|
12
|
4.3
|
0.45
|
|
Turkey
|
|
18
|
7.1
|
0.48
|
|
Lebanon
|
|
10
|
5.1
|
0.50
|
|
Jordan
|
|
16
|
10.1
|
0.62
|
|
Korea
|
|
10
|
6.7
|
0.67
|
|
Singapore
|
|
5
|
1.8
|
0.63
|
|
New
Zealand
|
|
12.5
|
8.9
|
0.93
|
Source: International Monetary Fund,
various country papers.
4.2.
The Stabilization Program
The economic
crisis of 2007/08 led to a significant rise in the budget deficit and overall
debts—leading to a hemorrhaging of record high reserves. A government
stabilization plan in September 2008 was based on raising the tax-to-GDP ratio by
five percentage points, and its key element was the reform of the VAT that
formed the basis of the submission to the IMF. The argument was that the
government needed roughly two years to revive and implement the Shahid Husain
plan to restructure the FBR, and that the IMF monies would be a “bridging loan”
while this reform took effect. [8]
The revised
VAT law was meant to remove distortions in the GST—especially the domestic
zero-rating and exemptions that were largely designed to benefit special
interest groups and, pari passu, consumers of luxury textiles and oriental
carpets. The other main objective had been to create the basis for an arm’s-length
tax administration based on self-assessment and effective audit, minimizing the
problems of direct contact between the tax administration and taxpayers and
also the difficulties with the issue of refunds, which had created considerable
rent seeking. A critical additional objective was to remove the tax
administration’s ability to confer benefits on the chosen few through the
notorious SRO system; the new law required any such change to be submitted to
Parliament and that the FBR would be stripped of this power.
The
“streamlined” VAT law would also have replaced multiple rates (from 17 to 26
percent) and the cascading associated with reference prices by a single rate
and considerable simplicity, including the elimination of SROs, but it was
badly sold to the public and to Parliament. This was partly due to opposition
from the vested interest groups that had benefitted from the holes in the GST,
and partly due to the tax administration’s reluctance to relinquish its “rent-seeking
powers” and the loss of the SRO handles. Although the Senate passed a corrupted
version of the VAT bill (retaining some draconian powers for the FBR), there
was enough opposition to the bill in the lower house to stall it on the absurd
grounds that it would “crush the poor”, without empirical or analytical
support. In reality, the poor would have been largely unaffected by the GST,
but will surely be crushed by the resort to deficit financing and borrowing
from the banking system.
In order to
“rescue the IMF program”, the government proposed a Plan B in March 2011 to
remove the main “exemptions” under the GST, but without the full overhaul of
the law. This was to remove by administrative order the SROs that had led to
the exemptions. This option faced no legal difficulty. It would not have raised
much additional revenue and may even have led to less revenue in the short run,
but it would clearly indicate that the authorities intended to tackle vested
interests seriously. The reform lasted less than a fortnight as the vested
interests coalesced, and the proposals were replaced with a far worse situation
with the SRO283, issued on 1 April 2011.
SRO283
provided all sorts of exemptions and lower rates on all manner of final and
intermediate goods—184 items in all—and recreated the “cascading” that is the
antithesis of a GST. Finally, item 185 stipulated that any other exemptions
that might be needed in the future would be included without having to issue an
additional SRO—this is SRO making ad absurdum. More problematic is that the FBR
has effectively abandoned the logic of the arm’s-length administration that was
the basis of the Shahid Husain proposal. Indeed, the use of third-party
information from the National Database and Registration Authority (NADRA) is
being accessed and selectively used. Without adequate safeguards, there will be
enhanced scope for reinvigorated rent-seeking in an administration reliant on
the SRO culture. It is extremely dangerous to leave tax making powers in the
hands of an unaccountable tax administration, and it negates the basis of a
parliamentary system.
The SRO powers
also reduce the trust of the federating provinces in the FBR since they effectively
reduce the overall revenues that should accrue to the provinces through the
divisible pool. This is also the subject matter of the NFC, which is addressed
next.
4.3.
The 7th NFC Award
The provinces
have long been responsible for the bulk of spending on health and education. Given
long-standing budgetary constraints, however, this spending has fallen far
short of what is considered necessary to meet minimum standards, let alone the
MDGs—indeed, outcomes in both sectors have fallen behind all other South Asian
countries. Total spending on health and education in 2007/08 was 5.4 percent of
GDP, which the PRSP aimed to increase to 6.8 percent of GDP by 2011 (Pakistan,
Finance Division, 2010, p. 331). The main vehicle for this was an increase in the
provinces’ share of the divisible pool, given that they did not have adequate
own-revenue sources.
The
calculation made by the finance minister at the time was that, with the
proposed tax reforms, an increasing total pie would leave sufficient resources
in the hands of the center even as the provincial share was increased from
under 47 percent in 2008/09 to 56 percent in 2010/11 and 57.5 percent thereafter.
The logic was understandable, even as Pasha, Pasha, and Imran (2010) warned
that the projections might be optimistic.
As it happens,
the collapse of the 2008 tax reform proposals, taken by the government to the
Friends of Democratic Pakistan and then to the IMF, proved to be calamitous for
the NFC award. First, it opened up a gap almost immediately between the provinces’
expected and actual revenue-sharing transfers—from 1.3 percent of GDP in
2010/11 to 2.9 percent by 2013 (Table 8.3). This gap is in relation to the spending
assignments that were in place in 2009, and does not factor in the 18th
Amendment (which is discussed in the next section).
Table 8.3: NFC projections 2010–14 (percentage of GDP)
|
Item
|
2009/10
|
2010/11
|
2011/12
|
2012/13
|
2013/14
|
2014/15
|
|
Divisible pool (actuals)
|
3.8
|
4.8
|
|
|
|
|
|
Total provincial resources
|
4.56
|
6.2
|
|
|
|
|
|
Federal tax collections
projected by NFC
|
10
|
11
|
12
|
13
|
14
|
15
|
|
NFC provincial shares
expected
|
|
6.16
|
6.90
|
7.48
|
8.05
|
8.63
|
|
Tax collections (actual)
2010/11
|
|
8.7
|
8.5
|
8.5
|
9.0
|
10.1
|
|
Revised NFC divisible pool
|
|
4.87
|
4.89
|
4.89
|
5.18
|
5.81
|
|
Provincial funding gap
|
|
-1.29
|
-2.01
|
-2.59
|
-2.88
|
-2.82
|
|
Education and health (PRSP
II)
|
|
6.79
|
|
|
|
|
Source: Government of Pakistan (2010); and author’s own
calculations.
The second
difficulty was that the failure of the tax reforms left far too little in the
hands of the federal government. Thus, for 2010/11, the share of the divisible
pool in the hands of the federation was around 4 percent of GDP. Debt servicing
alone was 5.6 percent of GDP in the same year (IMF, 2011).
It had been clear that the reforms promised under the 2008 IMF program
involved an integrated GST, building on the Musharraf government’s arrangement
to ensure a common administration but removing the exemptions and zero-ratings
that had been introduced by the previous regime. At the time the NFC award was
being finalized, a Ministry of Finance team had worked on drafting a revised
GST/VAT law that was to be presented to Parliament by end-December 2009. Yet
the NFC award finalized in December 2009 reiterated that the GST on services was
a provincial subject and that collection could also be provincial. Although
there was an attempt to paper over the gaping cracks and ensure that the FBR
would continue to administer those services that entered inter-industry
transactions, affecting cross-provincial ones—especially banking and insurance,
telecommunications, and trade-related services—the proposals did not stick and
the whole structure collapsed.
The failure of
the tax reforms implies that there is no conceivable way of reaching the 15
percent of GDP target for the overall tax-to-GDP ratio by 2014/15; the budget
strategy paper issued in 2012 brought the target down to 10.1 percent of GDP.
This is catastrophic for the federal government since the increasing cost of
borrowing alone will far exceed the federation’s share of the post-7th
NFC divisible pool. This can only hasten the collapse of the federal government
forecast by Haque (2009).
Unfortunately,
the situation in the provinces is no better. As mentioned above, a shrinking
resource pie relative to expectations puts the pre-18th Amendment
goals out of reach, e.g., as enunciated in the PRSP-II (see Pakistan, Finance
Division, 2010). The 18th Amendment merely adds to the unfunded
mandates, which can only lead to further erosion in public services and gaps vis-à-vis
the MDGs.
4.4.
The 18th Amendment
The devolution process that has begun with the 18th
Amendment presents a great opportunity to change the way that public policy is
formulated in Pakistan, and hopefully to make it more responsive to the needs
and desires of the population. However, if the tax reforms do not succeed,
given the vociferous opposition by the vested interests that have benefitted from
exemptions and zero rating,[9] the entire
devolution process will run into trouble, as has the current NFC award. This
seriously risks the implosion of the existing intergovernmental fiscal system.
Shah (2012)
provides a very comprehensive assessment of the benefits and challenges arising
from the 18th Amendment. The chapter focuses here on the issue of
unfunded mandates, which could lead to an implosion of public institutions and
services, as well as the very real dangers posed by increasing barriers to
interprovincial trade. Both are extremely damaging to the concept of an
integrated federation, and each is considered in turn.
4.5.
Unfunded Mandates
The unfunded
mandates have been exacerbated by the new responsibilities added to the pre-18th
Amendment spending assignments in relation to the assigned own-revenue bases
and shared revenues and transfers. As discussed above, the provinces and local
governments lacked the necessary resources, prior to the 18th
Amendment, to effectively provide for their responsibilities at that time.
Without the tax reforms, the NFC award is just a mirage in the desert.
4.6.
Subsidiarity and Spending
Assignments
The 18th
Amendment eliminated the Constitution’s concurrent lists, giving provinces sole
powers in a number of areas, including health and education. The speed at which
the spending functions were devolved meant that inadequate attention was given
to the role of “subsidiarity”, the role of regulations, and the coordination of
functions with associated externalities, such as primary healthcare, university
education, climate change and environment, and natural disasters.
Almost immediately, the provinces discovered that they could not finance
the very heavy expenditures that had been incurred by the Higher Education
Commission, and made a reference to the Council of Common Interests to return
the function to the federation. Unfortunately, financing for the function was
no longer available, and the federation attempted to move the function under a
line ministry rather than an independent commission, as had been the case prior
to the 18th Amendment. Some commentators suspect that this might
have been linked to the Higher Education Commission’s refusal to certify the
questionable educational credentials or degrees of a large number of lawmakers
(of all major parties).
In other areas
too, such as preventive healthcare and pharmaceutical standards, once a
function with widespread externalities is dismantled and handed over to
subnational governments, it is likely that there will be conflicting or
confusing standards that could increase the likelihood of epidemics. If the
provinces cannot handle the function, it is exceedingly difficult to
reestablish the previous institutional arrangements at the national level, not
to mention that the federation now has to borrow to meet its current
expenditures and is caught in a debt trap.
The 18th
Amendment unwound the steps taken by the Musharraf government to constitute a
third tier of government, passing virtually all powers to the provinces and
leaving it to provincial assemblies to decide whether or not to devolve
further. This reaction to a desirable reform by the military government (made, albeit,
for the wrong reasons) is understandable, but not well thought out. The
functions and operations of the third level of government should be clearly
delineated, as well as the role of the federation in keeping with the
subsidiarity principles outlined above.
In addition,
more work is needed to improve effective service delivery at the district or
local levels, and to counter the possible inadequacy of local incentives in
providing for the most vulnerable, e.g., the aged without extended family
support, single women, and minorities. It is likely that the local or
provincial government has less interest in providing services or protection to
minorities, such as the Hazaras in Balochistan, who are nonetheless full
citizens of the country and entitled to the same privileges and safety as any
inhabitant of Islamabad or the exclusive neighborhoods of Karachi or Lahore.
This could lead to significant miscarriages of justice
and equity in the future; indeed, it is happening with alarming
frequency—reminiscent of North, Wallis, and Weingast’s (2009) warning of
failure in the context of limited access states.
The
expectations raised by the 1973 Constitution bear no relation to what can be
financed. It guarantees:
- “Compulsory and free” education till secondary level [#37(b)]
- “Access to technical and higher education for all on
merit” [#37c]
- The ”basic necessities of life, such as food, clothing,
housing, education and medical relief for all such citizens, as are permanently
or temporarily unable to earn their livelihood on account of infirmity,
sickness or unemployment” [#38d].
The social
benefits for the unemployed or incapacitated are very Bismarckian. The
guarantees are very clear, with no additional targeting or score cards that are
open to “capture” or “clientelism.” These are constitutional basic rights and
actionable in court, but in order to finance these rights, a revenue-to-GDP
ratio commensurate with the more advanced developing countries, such as Chile
or Brazil, is needed—i.e., 25 percent or more. This is also the goal that China
is pursuing. However, it appears almost completely out of reach under the
present configuration of policy and administration in Pakistan.
4.7.
Barriers to Trade and
Integration
On the spending
side, as pointed out by Shah (2012), the absence of a national standard setting
capability might be a severe constraint to establishing a common integrated
market. On the revenue side, the sales tax on services, if applied at one of
the few ports by the respective province without providing credit to purchasers
in other provinces, could become an effective import duty on cross-provincial
trade. This is a potential conflict with the commerce clause in the
Constitution.
4.8.
Revenue Reassignments
A fundamental problem lies in the absence of effective own-source
revenues at the provincial or local levels. As seen in the Latin American and
East Asian cases, this is the Achilles’ heel of the devolution process in many
countries, leading to a loss of accountability and responsibility for local
service delivery. A share in the divisible pool or the unstable assignment of
the GST on services does not count as effective own-source revenues.
In keeping with the Government of India Act 1935, Pakistan’s current
constitution maintains the concept of split revenue bases, both for sales as
well as incomes. This has opened up vast avenues for tax avoidance and
evasion—such as the abuse of the agricultural income exemption. Moreover, with
the rent seeking in the FBR, there is little confidence in its ability to keep
the provinces’ interests in mind in performing its functions.
The 18th
Amendment reiterates the right of the provinces to administer the GST on
services, if they so desire—the revenues belong to them in any case. This makes
it very difficult for a provincial government to vary the rate structures
without making their GST almost impossible to implement for fear of it degenerating
into an instrument for “provincial tax wars” or impediments to trade. There is
also a danger with credit invoices issued by one province to be honored by
others or the federation—this would be akin to the “invoice sightseeing” that
has become a serious problem in Brazil, and would magnify the “flying” invoices
that are already a serious problem with zero-rating in Pakistan’s case (and
that is with a single administration).
The split base
of the GST relating to goods and services is unique to the Subcontinent, and has
its origins in the Government of India Act 1935 that assigned the sales tax on
goods to the states/provinces. After independence, the goods part was taken
over by the federal government in Pakistan, and the more difficult element on
services was left to the jurisdiction of the provinces, reiterated in the 1973
Constitution. As there was no GST or VAT at that time, the complexity of this
assignment was not realized. Thus, Pakistan finds itself in a unique position
as being the only country in the world trying to implement a GST on services at
the subnational level, without the administrative machinery to do so. Even if
it had the administrative machinery, this would be a herculean task.
A cooperative
solution would have been to permit the FBR to function on behalf of the federation
and the provinces (as had been initiated by the Musharraf administration) to
collect an integrated GST for the federation and all the provinces, close
loopholes, and deliver a larger pie to the provinces directly as well as
through the common divisible pool. After all, this was the basis of the NFC
award. However, one province rejected this arrangement, given the severe trust
deficit associated with the FBR. A complex alternative mechanism was proposed
to work around this difficulty, with the current FBR effectively operating the
crediting and refund mechanisms associated with the GST—the only agency capable
of doing so. But, as discussed above, vested interests opposed fixing the
loopholes in the GST and there is very little confidence in the current FBR’s
ability to operate on an arm’s-length basis.
By now it is clear that the current system underpinning the 18th
Amendment is not sustainable. A more stable solution is needed that provides the
provinces with significant tax handles that also generate greater
accountability for subnational spending. Marginal changes to tax rates will not
do. Also, as the post-NFC discussion on the GST illustrates, agreements are not
easy to reach and are unstable. Ideally, a new arrangement should be sought and
another constitutional arrangement on the revenue side introduced to preserve
the thrust of the 18th Amendment and positive elements on the
spending side. Both policy options and new administrative arrangements need to
be examined—this is a significant research agenda that should also involve the
next NFC award.
It is not
possible to initiate sensible reforms in the tax policy agenda without an arm’s-length
tax administration that has the trust of the federating units. Similarly, such
an administration cannot be conceived without overhauling the tax policy
framework and the associated assignments to different levels of government.
This is as complex a task as the 18th Amendment, but should be more
carefully designed before being rushed through Parliament. However, there is
nothing like an economic crisis to concentrate minds and create the political
will to carry out serious reforms, so if there is the opportunity, it should be
taken. Some of the lines of reform can be gleaned from the successes in China
and the difficulties faced in other federations such as Brazil and India, who
are also stuck with inefficient split bases and find it hard to overcome the
vested interests that coalesce around the benefits conferred.
4.9.
Policy Options
A fundamental
principle guiding the tax policy agenda should be that the major tax bases
should be consolidated, and that both the provinces and districts should be
assigned tax handles that allow flexibility to set rates at the margin. This
flexibility is the crux of own-source revenues and the foundation for
accountability in both revenue generation and spending.
At the
national level, the major tax bases could be consolidated along the following
lines:
-
Income taxes.
All sources of income are income and should be treated equally to avoid
distortions, tax shelters, and handles for rent seeking and corruption. This
implies that the following:
o Personal income
tax. All sources of income should be
included in the tax base, including agriculture, property, and foreign source
income. The provinces and districts should be allowed to “piggyback” on the
full base; this will give them more revenues than at present and does not
require separate administrations, just the setting of rates.
o Corporate income
tax (CIT). The holes and preferences
should be closed, and the rate reduced to 25 percent, as in China and many
other countries. A gross assets tax should be constituted in the short run to
plug the gaps, and this should be creditable against CIT liability. The CIT
could continue to go into the divisible pool.
-
GST. This should
be treated as an integrated tax and administered as an arm’s-length agency. It
should have a single rate and no exemptions other than on unprocessed food. The
sharing arrangements could include the following:
o Fully entering
the divisible pool. This would allow
57.5 percent of the hopefully better-performing GST to go to the provinces,
thereby fulfilling the objectives of the 7th NFC award.
o
Australian model. In Australia, the VAT is
centrally collected but 100 percent is returned to the states through an equalization
system run by the Commonwealth Grants Commission on which the states are
equally represented. This option closes both horizontal and vertical gaps and
is compatible with maintaining incentives for subnational efficiency. If this
model is chosen, the divisible pool could be scrapped since the personal income
tax piggyback (and similar arrangements for the CIT as in the US) would provide
for own-source revenues.
o Sharing
arrangements. This is the Chinese
model, which gives rich provinces an opportunity to share in the country’s growing
revenue generation. However, it needs to be accompanied by an equalization
framework in order to provide incentives for the poorer provinces that do not
generate revenues, given the lower volume of transactions.
-
Carbon tax.
This should replace the petroleum levy, as originally designed, and should also
be shared with the provinces (see Ahmad & Stern, 2009).
-
Excises. Some excises
could be established at the national level and feed into the divisible pool.
Others could be purely provincial, provided they do not affect interprovincial
commerce.
-
Property taxes. These should be assigned to the
districts/municipal governments as far as rate setting is concerned, although the
provinces could be responsible for the cadaster, land register, and valuation.
-
User charges and tolls. These should be largely local, although the old
octroi should not be resurrected.
The above options provide provinces with both additional revenues as well
as own-source tax handles, without the need for separate administrations. In
the Pakistan context, however, it would be necessary to establish an arm’s-length
tax administration that was acceptable to the provinces.
4.10.
Administration
The trust
deficit vis-à-vis the FBR/Central Board of Revenue has only grown since the
warnings of the Qamar-ul Islam Commission in the 1980s. The failed attempt to
implement the Shahid Husain report under the World Bank’s USD 135 million Tax
Administration Reforms Project (TARP) highlights the incentive problems
inherent in the current structure. The World Bank’s proposal to throw more
money at the problem through the USD 300 million TARP II is likely to meet the
same fate since neither the authorities or the donors recognize the incentive
incompatibility of the existing arrangement, and the political economy
difficulties in making the current model work.
TARP and the
rescue of TARP in 2009 could not reform the existing FBR to make it operate on
an arm’s-length basis. With its ability to use NADRA data, the FBR has become
truly intrusive and quite dangerous given that rent-seeking opportunities have
been magnified out of all proportion. An FBR that can issue SROs at will,
overriding tax laws without reference to Parliament and in the absence of any
consultation with the provinces whose revenue shares are compromised, is a
nonstarter for the sort of tax administration that needs to underpin a new
intergovernmental framework in Pakistan.
It may be
necessary to start on the tax administration afresh, with a completely new
staff (as was the case in Peru in the early 1990s). The entity may need to be
detached from the Ministry of Finance, with its own responsible and accountable
minister (as in China), and placed under a board with representation from the
provinces. As the Peruvian example shows, a new administration can be
constructed quite quickly as long as the roles and modus operandi are clear
from the outset.
Given the
implosion of public services and growing failures to keep minorities and other
citizens secure, the intergovernmental and governance framework needs to be
subject to urgent review and action. There is much to learn from China. Significant
additional work is needed to recalibrate spending responsibilities and,
particularly, to completely redesign the tax assignments and administration for
the 18th Amendment to work effectively without unraveling the federation.
References
Afonso, J.
R., & de Mello, L. (2002). Brazil: An evolving federation. In E. Ahmad
& V. Tanzi (Eds.), Managing fiscal decentralization.
London, UK: Routledge.
Ahmad, E.
(Ed.). (1997). Financing decentralized
expenditures: An international comparison of grants. Cheltenham, UK: Edward
Elgar.
Ahmad, E.
(2008). Tax reforms and the sequencing of intergovernmental reforms in China:
Preconditions for a Xiaokang society. In S. Wang & J. Lou (Eds.), Public finance in China: Reform and growth
for a harmonious society (pp. 95–128). Washington, DC: World Bank.
Ahmad, E.
(2010a). Improving governance in Pakistan: Changing perspectives on
decentralization (Allama Iqbal Lecture). Pakistan
Development Review, 49(4),
283–310.
Ahmad, E.
(2010b). The political economy of tax
reforms in Pakistan: The ongoing saga of the GST (Discussion Paper No.
95948). Bonn, Germany: University of Bonn, Centre for Development Research.
Ahmad, E.
(2013). Generating inclusive and
sustainable growth. Washington, DC: Intergovernmental Group of Twenty Four.
Ahmad, E.,
& Brosio, G. (Eds.). (2006). Handbook
of fiscal federalism. Cheltenham, UK: Edward Elgar.
Ahmad, E., & Brosio, G. (Eds.). (2009). Does decentralization enhance service delivery and poverty reduction?
Cheltenham, UK: Edward Elgar.
Ahmad, E.,
& Brosio, G. (2011). Effective
federalism and local finance. Cheltenham, UK: Edward Elgar.
Ahmad, E.,
& Mohammed, A. (2013). Pakistan, the
United States and the IMF: A new great game or a curious case of Dutch Disease
without the oil? (Working Paper No. 57). London, UK: London School of
Economics and Political Science, Asia Research Centre.
Ahmad, E.,
& Tanzi, V. (Eds.). (2002). Managing
fiscal decentralization. London, UK: Routledge.
Ahmad, E.,
Amieva-Huerta, J., & Thomas, J. J. (1994). Peru – Pobreza: Mercado de trabajo y politicas sociales [Peru –
Poverty: Labor market and social policies]. Washington, DC: International
Monetary Fund.
Ahmad, E.,
Brosio, G., & Tanzi, V. (2008). Local service provision in selected OECD
countries: Do decentralized operations work better? In G. Ingram & Y. H.
Hong (Eds.), Fiscal decentralization and
land policies. Cambridge, MA: Lincoln Institute of Land Policy.
Ahmad, E.,
Craig, J., & Mihaljek, D. (1994). China:
On the determination of a grants system. Washington, DC: International
Monetary Fund.
Ahmad, E.,
Craig, J., & Searle, B. (1994). China:
Formulating and estimating grants. Washington, DC: International Monetary
Fund.
Ahmad, E.,
Gao, Q., & Tanzi, V. (1995). Reforming
China’s public finances. Washington, DC: International Monetary Fund.
Ahmad, E.,
Lee, K.-S., & Kennedy, A. (1993). China:
The reform of intergovernmental fiscal relations, macroeconomic management, and
budget laws. Washington, DC: International Monetary Fund.
Ahmad, E.,
Li, K., & Richardson, T. (2002). Recentralization in China? In E. Ahmad
& V. Tanzi, (Eds.), Managing fiscal
decentralization. London, UK: Routledge.
Ahmad, E.,
Rydge, J., & Stern, N. (2013). China:
Structural reforms lead to tax reforms lead to structural reforms. Paper
presented at the China Development Forum, Beijing.
Ahmad, E.,
Brixi, H., Fortuna, M., Lockwood, B., & Singh, R. (2003). China: Reforming fiscal relations between
different levels of government. Washington, DC: International Monetary
Fund.
Ambrosiano,
F., & Bordignon, M. (2006). Normative versus positive theories of revenue
assignments in federations. In E. Ahmad & G. Brosio (Eds.), Handbook of fiscal federalism.
Cheltenham, UK: Edward Elgar.
Bardhan, P.,
& Mookherjee, D. (2000). Relative capture of government at local and
national levels. American Economic Review,
90(2), 135–139.
Besley, T.,
& Case, A. (1995). Incumbent behavior: Vote-seeking, tax-setting, and
yardstick competition. American Economic
Review, 85(1), 25–45.
Bird, R. (1984).
Intergovernmental finance in Colombia.
Cambridge, MA: Harvard Law School.
Burki, S.
J., & Perry, G. (Eds.). (2000). Decentralization
and accountability in the public sector. Washington, DC: World Bank.
Consolidated
Version of the Treaty Establishing the European Community. (2002). Official Journal of the European Communities,
pp. 33–184. Retrieved from
http://eur-lex.europa.eu/en/treaties/dat/12002E/pdf/12002E_EN.pdf
Dafflon, B.
(2006). The assignment of functions to decentralized government: From theory to
practice. In E. Ahmad & G. Brosio (Eds.), Handbook on fiscal federalism (pp. 271–305). Cheltenham, UK: Edward
Elgar.
Devarajan, S.,
Khemani, S., & Shah, S. (2009). The politics of partial decentralization.
In E. Ahmad & G. Brosio (Eds.), Does
decentralization enhance service delivery and poverty reduction?
Cheltenham, UK: Edward Elgar.
Ebel, R.,
& Yilmaz, S. (2002). On the measurement
and impact of fiscal decentralization (Policy Research Working Paper No. 2809).
Washington, DC: World Bank.
Gadenne, L.
(2012). Tax me, but spend wisely: Public
finance and government accountability (Mimeo). University College London
and Institute for Fiscal Studies, UK.
Gurgur, T.,
& Shah, A. (2002). Localization and corruption: Panacea or Pandora’s box? In
E. Ahmad & V. Tanzi (Eds.), Managing
fiscal decentralization. London, UK: Routledge.
Haque, N.
(2009, August). Pakistan: A failed state.
Presentation to the International Food Policy Research Institute, Washington,
DC.
International
Monetary Fund. (2001). Government finance
statistics manual 2001. Washington, DC.
International
Monetary Fund. (2011). Pakistan: Staff
report for the 2011 Article IV consultation. Washington, DC: Author.
International
Monetary Fund. (2013). India: Staff report
for the 2013 Article IV consultation. Washington, DC: Author.
Jin, J.,
& Zou, H.-F. (2002). How does fiscal decentralization affect aggregate,
national, and subnational government size? Journal
of Urban Economics, 52(2), 270–293.
Lockwood,
B. (2009). Political economy approaches to fiscal federalism. In E. Ahmad &
G. Brosio (Eds.), Does decentralization
enhance service delivery and poverty reduction? Cheltenham, UK: Edward
Elgar.
Lou, J. (1997).
Constraints in reforming the transfer system in China. In E. Ahmad (Ed.), Financing decentralized expenditures: An
international comparison of grants. Cheltenham, UK: Edward Elgar.
Martinez-Vazquez,
J., & Richter, K. (2008). Pakistan:
Tapping tax bases for development (Tax Policy Report). Washington, DC: World
Bank.
Musgrave,
R. (1959). The theory of public finance.
New York, NY: McGraw-Hill.
North, D. C.,
Wallis, J. J., & Weingast, B. R. (2009). Violence and social orders: A conceptual framework for interpreting recorded
human history. Cambridge, UK: Cambridge University Press.
Oates, W.
(2008). On the evolution of fiscal federalism. National Tax Journal, 61(2),
313–334.
Olson, M.
(1969). Strategic theory and its applications – The principle of “fiscal
equivalence”: The division of responsibilities among levels of government. American Economic Review, 59(2), 479–487.
Pakistan,
Ministry of Finance. (2008). Economic
stabilization plan: Reinvigorating hope and ameliorating people’s livelihoods.
Islamabad, Pakistan: Author.
Pakistan,
Finance Division. (2010). Poverty
reduction strategy paper (PRSP) II (Country Report No. 10/183). Washington,
DC: International Monetary Fund.
Pasha, H.,
Pasha, A. G., & Imran, M. (2010). Budgetary consequences of the 7th
NFC award. Pakistan Development Review,
49(4), 375–385.
Rao, M. G.
(1997). Intergovernmental transfers in India. In E. Ahmad (Ed.), Financing decentralized expenditures: An
international comparison of grants. Cheltenham, UK: Edward Elgar.
Salmon, P.
(1987). Decentralization as an incentive scheme. Oxford Review of Economic Policy, 3(2), 24–43.
Salmon, P.
(2006). Horizontal competition among governments. In E. Ahmad & G. Brosio (Eds.),
Handbook of fiscal federalism.
Cheltenham, UK: Edward Elgar.
Salmon, P. (2010,
September). Improving performance by
political decentralization (or centralization)? Paper presented at the SIEP
Conference in Pavia, Italy. Retrieved from
http://www-3.unipv.it/websiep/2010/201012.pdf salmon
Shah, A. (2010).
Empowering states and provinces or unshackling local governments: Does it
matter for peace, order, good governance, and growth? (Gustav Ranis Lecture). Pakistan Development Review, 49(4), 333–357.
Shah, A.
(2012). Making federalism work: The 18th
constitutional amendment (Policy Paper No. PK03/12). Washington, DC: World
Bank.
Silvani, C.,
Biber, E., Crandall, W., Grant, W., Reos, O., & Seymour, G. (2008). Pakistan: A tax administration review (Final
report). Washington, DC: World Bank.
Tanzi, V. (2002).
Pitfalls on the road to fiscal decentralization. In E. Ahmad & V. Tanzi
(Eds.), Managing fiscal decentralization.
London, UK: Routledge.
Tanzi, V. (2010).
Revenue sharing arrangements, options and relative merits (Mahbub ul Haq
Lecture). Pakistan Development Review,
49(4), 311–331.
* The author is Senior Fellow at the Center for
Development Research (University of Bonn) and Asia Research Centre (London
School of Economics and Political Science).
[1] See Burki and Perry (2000). See also Shah (2010), who
also draws attention to the very important lessons that can be drawn from the
Chinese experience.
[2] This section draws on Ahmad (2010a) and Ahmad and
Brosio (2006).
[3] See also Ahmad, Li, and Richardson (2002); Ahmad, Lee,
and Kennedy (1993); Ahmad et al., (1995); Ahmad, Craig, and Mihaljek (1994);
Ahmad, Craig, and Searle (1994); Ahmad (1997); Lou (1997); and Ahmad, Brixi,
Fortuna, Lockwood, and Singh (2003).
[4] An interesting example is the European Court of Justice’s
rejection of a case brought by the German government against the EU Directive
on Deposit Guarantee Schemes (Case C-233/94).
[5] A recent extension by Salmon posits that cross-country
comparisons may be even more important for voters.
[6] See Ambrosiano and Bordignon (2006) for a discussion
on the general issues, and Gadenne (2012) for an interesting assessment based on the case
of Rio de Janeiro.
[7] For example, on the websites of the Ministry of
Finance, the Federal Bureau of Statistics, or the State Bank of Pakistan.
[8] The former head of the Argentine Revenue Authority was
hired by the World Bank to prepare a plan to enable a reformed VAT to be
implemented in a reformed FBR by summer 2010—the key date under the IMF program
(Silvani, Biber, Crandall, Grant, Reos, & Seymour, 2008). By 2011, this
effort had been abandoned. The original USD 135 million loan was not fully
drawn. The World Bank now plans to revive the project with a USD 300 million
loan. The problems lay in the incentive structures facing officials and
politicians and not the financing constraints.
[9] This has been couched in
“populist” terms as affecting the interests of the poor—in fact, a properly
functioning tax system would reduce the government’s borrowing requirements and
the current inflationary pressures. It is also unlikely that relative prices
would change adversely for the poor with the removal of these extraordinary
benefits for the pampered sectors and a downward revision and consolidation of
the GST’s rate restructure.
Labels: Pakistan, Pakistan Economy, Pakistan: Moving the Economy Forward, Publications
posted by S A J Shirazi @ 12/27/2013 11:39:00 AM,
City Campus
104 - C, Gulberg III,
Lahore, Pakistan.
Phones: 92-42-35714936, 38474385
Fax: 92-42-36560905
Main Campus
Intersection Main Boulevard Phase VI
Burki Road
Lahore, Pakistan.
Phones: 36560935, 36560939
