The Role of The State and Public Finance in The Next Generation

ISSN 1608-7143
OECD Journal on Budgeting

Volume 8 – No. 2
© OECD 2008
The Role of the State and Public Finance

in the Next Generation
by
Vito Tanzi*
This article discusses the economic role of the state as it evolved during the
20th century, starting with how current tax systems developed and how fiscal
termites can weaken the foundations of tax systems, examining the spending side
of the government role, and speculating on future developments, particularly in the
Latin American context.
* Professor Vito Tanzi is the former Director of the Fiscal Affairs Department of the International
Monetary Fund. This paper was presented at the 20th Regional Seminar on Fiscal Policy of ECLAC
(United Nations Economic Commission for Latin America and the Caribbean), Santiago de Chile,
28-31 January 2008, www.eclac.cl.
1
THE ROLE OF THE STATE AND PUBLIC FINANCE IN THE NEXT GENERATION
1. Introduction
This article discusses the economic role of the state as it evolved during the
20th century and speculates on how it might evolve in future decades. Because of
availability of statistical information, there will be a greater focus on advanced countries.
The article will also address developments in Latin America recognising the much greater
heterogeneity among countries’ per capita incomes and economic developments in that
region. The wide scope of the topic makes the discussion of it inevitably broad-brush and
somewhat impressionistic. A discussion of the future scope of public finance must
inevitably start with a review of past developments. The past is always a prologue for the
future and there is always a lot to be learned from studying it. We shall start with how
current tax systems developed and then move to the spending side of the government role.
In the last section, we shall recognise that the role of the state can be played also with tools
other than public spending and taxes.
Modern tax systems developed largely in the period between 1930 and 1960, a period
characterised by: a) major restrictions on trade erected during the Great Depression and
during World War II; b) limited movements of portfolio capital; c) little cross-country
investment, except for direct investments in natural resources; d) little international
mobility of people, except for emigrants after World War II; and e) almost no cross-country
shopping by individuals. In Latin America, this was the period when import substitution
policies, at the time strongly promoted by CEPAL (Comisión Económica para América Latina y
el Caribe, the Economic Commission for Latin America and the Caribbean) and by Raul
Prebisch, became popular. During these decades, governments had not yet been expected
to assume the broad social and economic responsibilities that they would assume in later
decades although they were already being pushed, by the prevailing intellectual winds, in
that direction. Tax burdens were generally under 30% of the industrial countries’ gross
domestic products (GDP) until around 1960, and well under 20% of GDP in developing and
Latin American countries.
Between 1930 and 1960, two important “technological” innovations were introduced in
the tax area. These were “global and progressive” income taxes and the introduction of the
value-added tax (VAT). These two developments, together with social security taxes on the
growing shares of wages and salaries in national income in industrial countries that
characterised those decades, would account for most of the rise of their tax levels which,
by the 1990s, in many OECD countries, would exceed 40% of GDP and surpass 50% in a few
countries. In Latin American countries, however, with the exception of Argentina, Brazil,
Uruguay and some other countries, the tax levels remain today below 20% of GDP.
In an influential book, published in 1938, Henry Simons, then a professor at the
University of Chicago, made a strong case for taxing all sources of income of individuals as
a whole rather than as separate parts (the so-called global income) and for taxing this total
with highly progressive rates. This was a radical departure from past practices. Some
German economists, such as Georg Schanz, had made similar recommendations (see
2 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

Musgrave, 1998). It was argued that this approach would better satisfy revenue and equity
objectives at a time when the income distribution was becoming a growing concern while
the disincentive effects of high marginal tax rates were still dismissed as unimportant.
Having been proposed during the Great Depression (soon after Roosevelt’s New Deal) and
just before World War II, the global personal income tax with highly progressive rates
became very popular in the United States and helped finance the Second World War. It
soon came to be seen as the “fairest” tax. It remained popular until the 1970s.
Given the American influence in the world after World War II, the global income tax
was quickly exported to other countries. After the war, and for a couple of decades,
American tax consultants promoted this tax in both developed and developing countries.
In the 1960s in Latin America, this tax was pushed by the so-called “Joint Tax Program”
created during the Kennedy years by the Organization of American States, the Inter-
American Development Bank and the United Nations. However, in Latin America the
results were less productive in terms of revenue generation than in developed countries.
The other “technological” innovation, the value-added tax, originated in France. It
quickly replaced the turnover (cascade) taxes on transactions that had been common in
most European countries, including in the six members of the European Coal and Steel
Community that would in time blossom into the European Union. The VAT was welcomed
by the members of that Community because it allowed the zero-rating of exports and the
imposition of imports, thus eliminating discord between trading partners while still
leaving countries with the freedom to impose whatever rates they wished. The countries
were free to impose the VAT rate that they liked or needed, presumably without interfering
with international trade flows. This feature made the value-added tax a useful instrument
for countries belonging to customs unions. The value-added tax has proven itself to be a
major revenue source for most countries. Latin America was quick to adopt this tax in
Brazil, Uruguay and some other countries. It quickly spread to other countries.
In industrial countries, the two developments mentioned above, together with social
security taxes on labour income imposed to finance public pensions, made it possible for
the tax systems of many countries to finance the large demands for public revenue that the
growing functions of government, especially in the so-called welfare states, were creating
(see Tanzi and Schuknecht, 2000). However, Latin American countries were much less
successful, until more recent years, in raising substantial levels of taxation that would
allow their governments to play larger roles in the economy through public spending. The
consequences were twofold: first, the use of bad taxes to attempt to raise more revenue;
second, reliance on less efficient tools than public spending to pursue social goals. This
issue is discussed in the concluding section.
2. Globalisation and taxes

In recent decades, and especially since the 1980s, important developments have been
changing the economic landscape that had characterised earlier decades. These
developments have potentially great implications for tax systems but also for expenditure
policies. The most important among them are:
● The opening of economies and the extraordinary growth of international trade. Import
substitution theories and policies are no longer fashionable. The world economy has
become much more integrated than it had been in the past. Both developed and
developing countries have contributed to this growth. For Latin America, this trend
OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 3

toward globalisation represents a truly fundamental change from the policies of import
substitution of the 1950s and 1960s.
● The phenomenal increase in cross-border capital movements. This increase has been
promoted by the removal of obstacles to capital mobility. The removal has been
facilitated by new policies and by technological innovations that have made
communication cheap and rapid. There has been an extraordinary growth in the amount
of financial capital that now crosses frontiers on a daily basis. This capital finances
direct investment, feeds portfolio investments, covers current accounts imbalances, and
provides needed foreign currency to international travelers. It has thus relaxed the
correlation that existed in the past between a country’s saving rate and its investment
rate, a correlation stressed by Feldstein and Horioka. The great flow of capital has also
made it easier for governments to finance larger fiscal deficits because they no longer
must rely on domestic savings.
● The importance of multinational corporations has grown enormously both in the
financing of direct investment (for both the production of outputs from natural
resources and for the production of manufactured goods) and, especially, in promoting
trade among related parts of the same enterprises located in different countries. The
time is long past when most enterprises produced and sold their output in the same
country or even in the same city or region where they were located. Trade among related
parts of enterprises, located in different countries, has become a large and growing share
of total world trade. It now accounts for more than half of total world trade.
● These international activities, accompanied by growing per capita incomes, sharply
falling costs of transportation for both goods and people, increased informational flows
that instantly inform individuals about changing relative prices and opportunities
created by them, and more liberal policy, have also led to a high mobility of individuals,
either in their role as economic agents or simply as tourists and consumers. A large and
increasing number of individuals in both industrial and developing or emerging markets
now earn all or part of their incomes outside the countries where they were born and
where they may still have their official residence. At the same time, a large and growing
number of individuals spend part of their income outside the countries where they
officially live. In conclusion, markets have become more global.
The implications of these developments for the countries’ tax systems and the
economic role of the state are still not fully understood by policy makers or economists.
The clear and limited role of the state that was identified a hundred years ago by classical
economists is giving rise to a much more complex and much less well-defined role.
Increasing evidence suggests that the developments described above are also creating
growing difficulties for the tax administrators of many countries and opportunities for a
few of them. As a consequence, they are raising questions about the optimal role of the
state in the current and especially future and more globalised economies. We shall first
deal with the tax implications and then with the implications for the optimal role of
the state.
Because of the developments described above, a country’s potential tax base is now no
longer strictly limited, as it was in the past, by that country’s territory but, to some extent,
has been extended to include parts of the rest of the world. The reason is that a country can
now try to attract and tax fully or partly: a) foreign financial capital; b) foreign direct
investment; c) foreign consumers; d) foreign workers; and e) foreign individuals with high

incomes, including pensioners. These possibilities did not exist in the past and they are
fuelling “tax competition” among countries because, at least in theory, each country can try
to take advantage of these new possibilities. Tax competition implies that, to some extent,
a country’s tax burden can be exported at least in part. Especially, a small country may now
be able to “raid” the tax bases of other countries in ways that were not possible in the past.
Like the ocean and the atmosphere, the “world tax base” is thus becoming a kind of
“commons” – a common resource without clearly established property rights that, to some
extent, all countries can try to exploit to their advantage and to the potential detriment of
other countries. The Latin American countries are not immune from this problem.
Tax competition is in part related to the importance of taxation for location and of
location for taxation. By lowering the burden of taxes on some sensitive activities, tax
competition aims at making certain locations (say, Costa Rica or Ireland or Luxembourg)
more attractive to some investors and for particular activities than other locations. This
issue is particularly important when it comes to tax incentives used specifically to attract
capital to one country and away from competing countries. The attraction of a location
depends on several elements such as: a) statutory tax rates on the income of enterprises; b) tax
practice (administrative and compliance costs); c) predictability of the tax system, or “tax
certainty” over time in both rates and administrative requirements; d) legal transparency, that
is, clarity of the tax laws; e) use of tax revenue, that is, the services that the residents or the
enterprises get from the government in exchange for the taxes paid; f) fiscal deficits and
public debt, because these may predict future tax increases; and, more generally, g) the
economic or investment climate of the country which is much influenced by regulations,
corruption, crime, rule of law and similar factors.
When people face high tax rates or an unfriendly tax climate in today’s environment,
they may: a) “vote with their feet”, thus moving to a friendlier fiscal environment, as long
as the ceteris paribus condition holds; b) “vote with their portfolio” by sending their
financial assets abroad, to safer and lower tax jurisdictions; c) remain in the country, but
exploit more fully any opportunities for tax avoidance; and d) engage in, or increase,
explicit tax evasion. Globalisation and tax competition are making it easier for individuals
and enterprises to exploit these options. They have raised the elasticity of tax bases with
respect to tax rates. These actions affect the role that the state is expected to play or is able
to play.
Is tax competition a positive or a negative global development? On this question, views
diverge sharply. Some, and especially theoretical economists and economists with a public
choice bent, tend to see tax competition as a clearly beneficial phenomenon. Ministers of
finance, directors of taxation and policy-oriented economists tend to see it more as a
problem.
The main arguments in favour of tax competition are the following:
● It forces countries to lower their high tax rates, especially on mobile tax bases such as
financial capital and highly skilled workers.
● By reducing total tax revenue, tax competition forces governments to reduce inefficient
public spending. This “starve the beast” theory was promoted by Milton Friedman and
became popular during the Reagan administration in the United States in the 1980s.
● It presumably allocates world savings toward more productive investments.
● Because of lower tax revenue, it forces policy makers to make the economic role of the
state more focused and more efficient.

● It leads to a tax structure more dependent on immobile tax bases lowering the welfare
costs of taxation.
Against these arguments, there are others that find tax competition damaging. The
main ones are:
● Because public spending is often, politically or legally, inflexible downward, tax
competition may lead to higher fiscal deficits and public debts, and eventually to
macroeconomic instability.
● When governments are forced to cut public spending because of tax competition, they
will not cut inefficient public spending which may have strong political constituencies
that protect it, but rather capital spending or spending for operation and maintenance.
● Tax competition may lead to what is called “tax degradation” – that is, governments may
try to maintain public revenue by introducing bad taxes to replace lost tax revenues.
● The shift of the tax burden from mobile factors (financial capital and highly skilled
individuals) to immobile factors (largely labour income) makes the tax system less fair.
● The increased taxes on labour income stimulate the growth of the underground
economy and tax evasion and promote informal activities.
● Tax competition (and reactions to it) can make tax administration and tax compliance
more costly and difficult. Growing tax complexity is a frequent consequence of tax
competition because tax administrators try to fight tax competition by introducing more
complex rules. For this reason, tax systems are becoming progressively more complex
(see Tanzi, 2006b).
It is difficult to assess the quantitative impact of globalisation on tax revenue. This has
led some observers to dismiss its impact. However, close observation can help identify
some impact and can point to growing future difficulties for high-tax countries:
● In the OECD countries taken as a group, the ratio of taxes to GDP stopped growing in the
1990s, even though large fiscal deficits in many countries would have called for higher tax
revenue. In an increasing number of OECD countries, the average tax ratio has fallen in the
current decade. In contrast, in Latin American countries, recent years have brought higher
tax revenue in several important countries, facilitated by the favourable cycle.
● The rates of both marginal personal income taxes and corporate income taxes have been
reduced substantially in most countries in the past two decades, in part because of tax
competition. However, because of some widening of the tax bases, and because of the
increasing share of enterprise income in national income in several countries, corporate
income taxes have not fallen as shares of GDP.
● The rates of excise taxes on luxury products were sharply reduced in most countries in
the past two decades, leading to substantial falls in revenue from these taxes. This fall
has been made up by increases in value-added taxes and, in several countries, in taxes
on petroleum and tobacco. The reductions in the taxes on luxury products are in part the
consequence of increased foreign travel by taxpayers and the resulting possibilities for
shopping in places where excise taxes on expensive and easy-to-carry items are lowest.
Internet shopping has also contributed to this result.
● The “global income tax” has been losing popularity. The dual income taxes introduced by
the Scandinavian countries and by some other countries, including Uruguay, are an
example of the losing attraction of global income taxes. The dual income tax is a de facto
return to the schedular approach to income taxation that had prevailed in the past.

● There is a growing interest in flat-rate taxes and in “consumption-based taxes”.

However, few countries have so far moved toward the introduction of these taxes.
In some papers written over the past decade, I discussed the rise of what I called “fiscal
termites”. These “termites” result from the interplay of globalisation, tax competition and
new technologies. As their biological counterparts can do for wood buildings, fiscal
termites can weaken the foundations of tax systems, making it progressively more difficult
for countries to raise high levels of taxation and to maintain the tax structure that they
would prefer. I will list some of these termites without providing much elaboration. For
more elaboration, see Tanzi (2001).
The first of these termites is electronic commerce. Electronic commerce has been
growing at a very fast rate, both within countries and across countries, for consumer goods
and services and for trade in inputs of intermediate and capital goods. Its growth has been
accompanied and facilitated by the growing shift, in the countries’ gross domestic
products, from physical to digital products, including intangible capital. This kind of
commerce leaves fewer traces than the previous invoice-based commerce that, for
example, could be inspected by customs officials, and is much more difficult to tax.
Electronic commerce is creating great difficulties for tax administrators and legislators
who at times seem to be at a loss on how to deal with it. Revenue from value-added taxes
is clearly affected.
A second termite is electronic money (credit cards, other forms). Real money is
progressively being replaced by electronic money embedded in chips of electronic cards. A
“purse” software may be purchased through deposits in foreign banks or from secret bank
accounts, making it more difficult to trace and tax various transactions. The use of
electronic money may also reduce the revenue from “seigniorage” that countries get from
the emission of paper money.
A third important termite originates in transactions that take place between different
parts of the same multinational enterprises (i.e. intra-company transactions) located in
different countries. Because these transactions are internal to a multinational company,
they require the use of “transfer prices” – that is, of prices at which one part of the
enterprise, located in a given country, “buys” products or services from other parts of the
same company located in other countries which have different tax systems and tax rates
on the incomes of the companies. Being inputs for final products, the products or services
bought and sold may not be traded in the open market. Therefore, there may not exist
market or “arm’s-length” prices that can be used as references. Problems arise especially
with: a) inputs that are made specifically for a final product (say, a particular jet plane);
b) the use of copyrights, trademarks and patents for which a value must be determined;
c) the allocation of headquarters R&D or other fixed costs; and d) interest on loans made
from one part to another part of a multinational corporation for which a determination of
a market rate is difficult. The determination of these costs or of the prices of the goods and
services traded within the enterprises is often difficult and arbitrary. It lends itself to
manipulation by enterprises aimed at showing more profits in countries where nominal
tax rates on enterprise profits are low (say, Ireland) and less profit in countries where the
rates are high (say, Germany). The strategic use of “transfer prices” by enterprises can
significantly reduce the total taxes paid by multinational enterprises, creating major
problems for tax administrators.

Another termite is the existence and continued rapid growth of offshore financial
centres and so-called tax havens. Total deposits in these tax havens have been estimated
to be huge by both the International Monetary Fund and the United Nations. The
distinguishing characteristics of these tax havens are: a) low tax rates, to attract foreign
financial capital; b) rules that make it difficult or impossible to identify the owners of the
deposits located in these countries (no-name accounts, banking secrecy, etc.); and c) lack of
regulatory powers, or lack of information on these deposits, on the part of the countries
where the owners of the deposits reside. These tax havens make it possible for individuals
and enterprises from the countries where the capital originates to receive incomes that are
difficult for national authorities to tax.
Another important termite consists of new, exotic and complex financial instruments
that have been continually entering the financial market in recent years. The day is long
past when a normal citizen could understand and easily choose from the financial
instruments in which he/she invested savings. New financial instruments are designed by
extremely clever and highly paid individuals and, at times, are specifically designed to
avoid (if not evade) paying taxes. In the United States, this has allowed some billionaires to
pay tax rates on their incomes that are much lower than the rates paid by their drivers. As
a consequence, it is becoming more difficult for the employees of tax administrations (who
have normal training and modest salaries) to keep up with these developments.
The developments described above and others not mentioned will have a
progressively larger impact on tax revenue, tax structures, and the use of particular tax
bases. This impact will naturally be larger for some countries and less significant for
others. Because the role of the state played through public spending over the longer run
depends on the countries’ capacity to raise taxes and particular types of taxes, that role
will also be affected.
All countries will be affected by the existence of these fiscal termites. However, we
might speculate that high-tax countries, such as various European countries and a few
Latin American ones like Argentina, Brazil and Uruguay, would be more affected. Transfer
prices are a clear concern for all countries, and so are electronic commerce and the
possibility that more and more investments in Latin America may be financed through
loans originating from tax havens and not through equity capital.
Latin American countries suffer from another problem: the share of national income
that goes to wages and salaries is much smaller than in industrial countries. This means
that, to generate high revenue, either very high tax rates must be imposed on wages and
salaries or non-wage incomes must be subject to reasonable taxes. The problem with the
latter is that incomes that are not wages and salaries derived from large enterprises or
from the public sector are difficult to tax because: some of these incomes derive from the
informal sector or from self employment; and some are returns to capital (interest,
dividends, capital gains, rents, some forms of profits) that may be difficult to ascertain and
that are often lightly taxed for fear that the capital that generates these income might fly
out of the country. The result is an unusual situation whereby the top income deciles, that
receive an overwhelming share of personal income because of the high Ginis that prevail
in Latin America, pay little taxes, thus putting a strong downward bias to total tax revenue.
The move toward flat-rate taxes would not help with this problem. According to various
sources, the (non-weighted) level of taxation in Latin America has hardly changed in recent
decades and has remained below 20% of GDP (see Lora, 2007a).

3. The role of public spending

The last half century has witnessed major developments in the role that governments
have played through public spending in both industrial and developing countries,
including the countries of Latin America. This section describes some of these
developments. A later section attempts to pierce the veil of the role that governments
might play in the future.
The tax levels of many industrial countries are close today to their historical high and
sharply higher than they were a century ago. In 1870, a group of 18 currently advanced
countries for which data are available had public spending and tax levels of only about 13%
of GDP. The United States had even lower levels (see Tanzi and Schuknecht, 2000). These
statistics are much lower than even the low levels that have prevailed in Latin America in
recent decades. As a consequence, public spending at that time was limited and largely
focused on “core” or essential functions such as defence, protection of individuals and
property, administration, justice, and large public works. These were broadly the functions
described by Adam Smith in 1776 in his book, The Wealth of Nations. Because of wars, tax
rates in Latin America were higher than in Europe at that time.
In the 20th century, public attitudes vis-à-vis the economic role of the state started
changing. Governments and especially democratic governments with universal suffrage
were pressured by their citizens to widen their economic role to include some social and/or
redistributive functions. The pressures led to the phenomenal expansion of public
spending that took place especially in the second half of the 20th century. Public spending
started to grow during World War I but its growth was slow until about 1960. The great
acceleration came in the period between 1960 and 1990 when many countries, and
especially most European countries, created public programmes aimed at the economic
protection of individuals “from the cradle to the grave”. These programmes included public
pensions, public health, free public schools, subsidies to large families, unemployment
compensation, support for the disabled, public housing, and so on. As a consequence of
these programmes, in several European countries public spending approached or exceeded
50% of GDP toward the end of the 20th century.
The countries of Latin America were not able to raise their public spending to the
European level because of their inability to raise their tax levels. Those most exposed to
European influences, such as Argentina and Uruguay and later Brazil, tried to raise their
public spending and the taxes needed to finance it. However, these countries failed to raise
enough taxes to avoid macroeconomic difficulties even though the level of public spending
did not reach European levels.1 Brazil was a latecomer because, until the 1970s, it had a low
tax and spending level. However, in later years, especially promoted by spending for
pensions, public spending (and taxes) grew dramatically, approaching European levels in
recent years.
Not being able to significantly raise their tax levels, but being pressured to play larger
roles in the economy, the Latin American countries made a growing use of tools other than
public spending to achieve similar objectives. We shall return to this issue in a later
section.
There is some debate on whether the large increase in public spending (as
distinguished from the growth in per capita income over the period) that occurred in
industrial countries, and especially in European countries, contributed to a genuine
improvement in the welfare of the majority of citizens, or whether the citizens would have

been better off with a lower growth in that spending that would have left them with more
money in their pockets (because of lower taxes) but with less governmental services. The
increase in public spending often went towards paying for the social services mentioned
above. Because public sector intervention often displaces existing charitable or non-profit
institutions, or private mutual assistance organisations, it does not necessarily or
automatically add, on a net basis, to the informal arrangements for social protection that
citizens had been receiving, or could have received, through private programmes. In some
countries, there had been extensive social networks that informally provided significant
social protection to those in real need.
It can be assumed realistically that the welfare of citizens is linked to the values of
certain socio-economic indicators – such as life expectancy, infant mortality, educational
achievements, literacy rates, growth in per capita incomes, inflation and others – that
governments want to influence through their public spending policies (see Tanzi and
Schuknecht, 1997 and 2000). Evidence collected by Tanzi and Schuknecht has shown that
there has been little relationship, if any, in recent decades in advanced countries, between
changes in the countries’ shares of public spending in GDP and changes (in the desired
direction) of these socio-economic indicators. Countries that allowed their public spending
to grow significantly more than other countries (the “large government” countries) did not
show, on average, better quantitative results for these indicators than countries that kept
their governments smaller and leaner.
The conclusion reached by Tanzi and Schuknecht (2000) is strongly supported by the
estimations of the “Human Development Index” (HDI) prepared by the UNDP.2 The last year
for which these HDIs have been prepared is 2005.
The HDIs can be mapped against the share of public spending in GDP for the same
year. If more public spending promotes higher levels of “human development”, the
countries that have higher spending levels should have higher levels of human
development. Table 1 provides, for 19 advanced countries, the shares of public spending in
GDP and the ranking of these countries in the HDI for 2005. Figure 1 provides a visual
representation of the relationship.
The remarkable result is the absolute lack of a positive relation between public
spending levels and HDI rankings. High-spending countries do not have better ranks. For
example, the four countries with the highest HDI ranks – Norway, Australia, Canada, and
Ireland – have average spending levels of 37.6% of GDP while the four countries with the
highest spending levels – Sweden, France, Denmark, and Finland – have an average HDI
rank of more than 9. Their average spending is 53.5% of GDP. The ten countries that spend
between 44.7 and 56.6% of GDP have an average rank of 12.7 while those that spend
between 34.4 and 42.3% of GDP have an average rank of 7. Thus, at least for this group of
highly developed countries, with per capita incomes and development levels that are not
too different, there is at best no positive relation between public spending and welfare, as
measured by the HDI. At worst there seems to be a negative relation.3 After some level of
public spending is reached, which for advanced countries seems to be around 40%, more
public spending does not seem to improve welfare – at least as measured by the HDI.
Before leaving this group of countries, it may be worthwhile to mention that several of
the best performers, that had had very high levels of public spending in the 1990s, had
sharply reduced public spending without apparently suffering any serious consequences
(see Table 2).

Table 1. Public spending and indices of human development in industrial

countries in 2005
Public spending
HDI Rank
% of GDP Rank
Sweden 56.6 1 5
France 54.0 2 9
Denmark 52.8 3 13
Finland 50.4 4 10
Austria 49.9 5 14
Belgium 48.8 6 16
Italy 48.3 7 18
Germany 46.9 8 19
Netherlands 45.5 9 8
United Kingdom 44.7 10 15
Norway 42.3 11 1
Canada 39.3 12 3
New Zealand 38.3 13 17
Japan 38.2 14 7
Spain 38.2 15 12
United States 36.6 16 11
Switzerland 35.8 17 6
Australia 34.6 18 2
Ireland 34.4 19 4
Sources: Public spending data from OECD (2007), OECD Economic Outlook No. 81, Volume 2007/1, June, OECD, Paris;
indexes of human development (HDI) from UNDP (2007), Human Development Report 2007/2008, United Nations,
New York.
Figure 1. Public spending and the Human Development Index

HDI
20
19 Germany
18 18 Italy
17 New Zealand
16 16 Belgium
15 United Kingdom
14 14 Austria 13 Denmark
12 12 Spain
11 United States
10 10 Finland
9 France
8 8 Netherlands
7 Japan
6 6 Switzerland
5 Sweden
4 4 Ireland
3 Canada
2 2 Australia
1 Norway
0
30 35 40 45 50 55 60
% GDP
New York.
The data in Tables 1 and 2 seem to support a conclusion that public spending of, say,
around 35% of GDP should be sufficient for the government of a country to satisfy all the
objectives that are realistically expected to be achieved by the spending action of a public
sector in a market economy. If public spending is efficient and well focused, an even lower
spending percentage might be possible. Unfortunately, in many countries public spending

Table 2. Spending levels in selected countries (percentage of GDP)

Public spending
HDI Rank Difference
1992 2007
Norway 1 55.7 41.0 –14.7

Australia 2 38.6 34.0 –4.6
Canada 3 53.3 39.1 –14.2
Ireland 4 45.1 34.4 –10.7
Sweden 5 71.1 54.1 –16.7
New York.
is neither efficient nor well focused. The result is that more public spending provides no
guarantee that social welfare and the well-being of the masses will be improved.
When we leave the advanced countries of Table 1 and move to the Latin American
countries, we are faced by the realisation that in Latin America there seems to be an
apparent greater need for public sector intervention, because of widespread poverty,
because of very uneven income distribution, and because of the need to improve
institutional and physical infrastructures that in many countries remain inadequate. At
the same time, we must face the fact that the Latin American countries’ public sectors are
likely to be less efficient than those of the countries in Table 1. Furthermore, their capacity
to raise revenue and spend money efficiently is much more limited.
The above dilemma is reflected in the responses to survey questions by the citizens of
Latin American countries. Quoting from a recent OECD report (OECD, 2007b, p. 37): “most
Latin Americans say that the quality of basic public services in their country is not good.
According to Latinobarometro surveys of public opinion, 92% of Latin Americans express the
view that their government should spend more on basic education, and 75% that it should
spend more on social security” (emphasis added). The OECD also reports that a small
proportion of the population (15% in 2003 and 21% in 2005) trusts that taxes are well spent
and believes that fiscal policy in Latin America has done little to improve the distribution
of income.4 Thus, we are faced with the classic situation of a customer in a restaurant who
complains about small portions and bad food. Most Latin Americans want the government
to spend more on health, education and social security but most believe that the spending
will do little to improve things. Because more spending requires more taxes or more public
debt, it seems questionable whether more tax money should be spent unproductively.
In the Human Development Index, the rating of the Latin American countries also
seems to bear little relation to the level of public spending. Table 3 gives the relative positions
in the index of various Latin American countries. For these countries, a complicating factor
is the large divergence in per capita incomes and economic development that inevitably
influences the ranking, because richer countries tend to have higher HDI scores regardless
of the action of their governments. Still, the position of Brazil is striking because of its low
HDI rank and the very high spending levels. We might add that focused social spending can
generate a much higher ranking than expected from a country’s per capita income, as
indicated by Cuba’s ranking.
The higher taxes needed to finance high public spending reduce the disposable
income of the taxpayers that pay them, thus restricting their economic freedom and their
ability to buy what they wish from the market. Most likely, over the long run, high tax levels

Table 3. Rankings in the Human Development Index

for various Latin American countries, 2005
Rank
Argentina 38
Bahamas 49
Barbados 31
Belize 80
Bolivia 117
Brazil 70
Chile 40
Colombia 75
Costa Rica 48
Cuba 51
Dominica 71
Dominican Republic 79
Ecuador 89
Grenada 82
Mexico 52
Panama 62
Paraguay 95
Peru 87
St. Lucia 72
Suriname 85
Trinidad and Tobago 59
Uruguay 46
Venezuela 74
Source: UNDP (2007), Human Development Report 2007/2008, United Nations,

New York.
may also have a negative impact on the efficiency of an economy and on economic growth,
especially if the taxes are collected inefficiently and the money spent is used
unproductively.
An obvious question for higher-spending countries is whether the level of public
spending (and, consequently, of taxation) should be reduced if this could be done without
reducing public welfare and without hurting the poorer population. That is to say, if public
welfare is not reduced, on any objective criterion, by reduced public spending, then public
spending and tax revenue could be cut. This would allow most individuals to have
discretion over a larger share of their pre-tax incomes, giving them more access to
privately provided goods. In other words, the citizens would decide how to spend this
money, not the government.5 Of course, this argument is not relevant in countries where
tax levels and public spending are too low to even provide the minimum resources needed
for essential public goods. While public spending can be too high, it can also be too low.
The theoretical reasons advanced by economists to justify the spending role of the
state in the economy, including especially the need to help the truly poor, could be satisfied
with smaller shares of public spending in GDP than is now found in many countries if
governments could be more efficient and focused in the use of their tax revenue.6 Much
current public spending “benefits” the middle and higher classes. At the same time, much
of the tax “burden” is also likely to fall on the same classes. Putting it differently, the
government taxes these classes with one hand and subsidises them with the other, playing
the classic role of an intermediary. This intermediation on the part of the government

inevitably creates disincentives and inefficiencies both on the side of taxation as well as on
the side of spending.
Before going on with our discussion, let us consider some statistics related to social
spending in several Latin American countries. The spending is allocated among the five
quintiles (see also OECD, 2007b, p. 40). Table 4 refers to education. Table 5 refers to health.
Table 6 refers to social security. The tables tell us what we already know, but they do it in a
striking fashion. The main lesson from Table 4 is that, while spending for primary
education helps almost everyone – and it seems even to help the poorest 20% of the
population more than the richest percentiles, who may send their children to private
schools – as we move toward secondary and especially tertiary education, the spending
share moves up toward the richest quintiles. It is the richer quintiles that benefit the most
from this spending. This seems to characterise all countries for which there are data and is
most pronounced for tertiary education. In Guatemala, a full 82% of the spending for the
tertiary education goes to the top quintile. In Brazil, the percentage is 76%. In Paraguay, it
is 56%. It is difficult to justify a spending role of the state that subsidises the top 20% of the
population. It is also difficult to make a case that the government is more efficient than the
private sector in providing higher education.
Spending for health seems to be more evenly distributed, creating a stronger case for
public spending also because of the greater difficulty for the private sector to provide an
efficient market for health that would be affordable on the part of the poor. Of course this
highlights the need for efficiency in this spending.
Before discussing social spending for social security, it may be useful to add the
observation that the data in Tables 4 and 5 allocate spending among quintiles and not
benefits. There has been a habit among economists of identifying spending with benefits.
However, we should realise that the two are different concepts and may diverge
significantly. The spending is often received not by the citizens who use the services, but
by those who deliver them, such as schoolteachers, school administrators, doctors or
nurses. The benefits are assumed to be received by those who use the services, such as
school children, patients, and so on. In many cases, the providers of services come from
higher income quintiles than the beneficiaries of the services. In some cases, the spending
may not become a “benefit” for the recipient, especially when inefficiency, incompetence
or corruption are present. Therefore, the allocation by spending may exaggerate the
distribution of the benefits to the poorer groups. In some situations, those who deliver the
services may appropriate most of the benefits in the form of high salaries (see Tanzi, 1974).
This, of course, does not occur with cash benefits such as pensions.
Table 6 gives a clear impression of the extent to which social security benefits are
appropriated by higher income classes. For the countries included in the table, the bottom
40% of the population received anywhere between a maximum of 24% (Argentina) and a
minimum of 2% of the total (Colombia). On the other hand, the top 20% of the population
received between 80% (Colombia) and 30% of the total (Argentina). Those covered by public
pensions are a relative minority and are not the poorest citizens who reach the pensionable
age. It is thus difficult to justify a public role for pensions on the basis of social needs
because the poor, who are often in the informal sector and do not have regular jobs, do not
benefit from these programmes.
The bottom line is that the so-called “social public spending” in Latin America which,
including social security, has averaged about 15% of GDP in recent years and has been

Table 4. Distribution of social spending by income quintiles for education

in selected Latin American countries, circa 2000
First Second Third Fourth Fifth
Argentina (1998) 21 20 21 20 18
Bolivia (2002) 17 17 21 22 23
Primary 25 25 23 18 10
Secondary 15 18 24 24 19
Tertiary 3 5 17 30 45
Brazil (1997) 17 18 18 19 27
Primary 26 27 23 17 8
Secondary 7 12 28 33 19
Tertiary 0 1 3 22 76
Chile (2003) 35 27 19 13 6
Colombia (2003) 24 23 20 19 14
Primary 37 28 19 12 4
Secondary 24 27 23 19 8
Tertiary 3 8 17 31 42
Costa Rica (2000) 21 20 19 21 19
Primary 32 25 19 15 10
Secondary 18 21 22 22 17
Tertiary 3 8 14 30 45
Dominican Republic 25 26 24 16 9
Ecuador (1999) 15 20 20 22 23
Primary 35 26 20 13 6
Secondary 15 24 25 22 14
Tertiary 3 13 16 28 40
El Salvador (2002)
Primary 27 25 23 17 8
Secondary 11 20 26 25 18
Guatemala (2000) 17 21 21 21 21
Primary 21 25 23 21 10
Secondary 3 12 23 31 32
Jamaica (1997)
Primary 31 27 21 15 6
Secondary 10 15 25 30 20
Mexico (2002) 19 20 19 23 19
Primary 30 26 20 16 8
Secondary 14 20 21 26 19
Tertiary 1 7 15 33 44
Nicaragua (1998) 11 14 20 21 35
Paraguay (1998) 21 20 20 20 19
Primary 30 26 21 15 8
Secondary 14 18 25 24 19
Peru (2000) 16 18 19 21 26
Uruguay (1998) 28 23 19 16 15
Source: Adapted from information collected at the Inter-American Development Bank from various official sources. See
also CEPAL (2006), Panorama Social de América Latina, Comisión Económica para América Latina y el Caribe, Santiago.
growing in the past two decades (see ECLAC, 2005; and Lora, 2007a), together with tax
systems that are broadly proportional or even regressive (because of the low taxes on
personal incomes and on both real and financial wealth) do little to improve the income
distribution that continues to be characterised by Gini coefficients that are the highest in
the world. The OECD (2007b, p. 31) and the IMF (2007, p. 30) have called attention to the

Table 5. Distribution of social spending by income quintiles for health

Argentina (1998) 30 23 20 17 10
Bolivia (2002) 11 15 14 25 35
Brazil (1997) 16 20 22 23 19
Chile (2003) 30 23 20 17 9
Colombia (2003) 18 19 19 22 22
Costa Rica (2000) 29 25 20 15 11
Ecuador (1999) 19 23 23 24 11
El Salvador (2002) 26 23 21 18 12
Guatemala (2000) 17 18 23 25 17
Honduras (1998) 22 24 24 17 14
Mexico (2002) 15 18 21 23 22
Nicaragua (1998) 18 23 22 19 18
Source: Adapted from information collected at the Inter-American Development Bank from various official sources.
See also CEPAL (2006), Panorama Social de América Latina, Comisión Económica para América Latina y el Caribe,
Santiago.
Table 6. Distribution of social spending by income quintiles for social security

Argentina (1998) 10 14 20 27 30
Bolivia (2002) 10 13 14 24 39
Brazil (1997) 7 8 15 19 51
Colombia (2003) 0 2 5 13 80
Costa Rica (2000) 12 12 12 18 45
Ecuador (1999) 4 7 21 22 46
Guatemala (2000) 1 3 5 15 76
Mexico (2002) 3 11 17 28 42
Uruguay (1998) 3 7 15 24 52
Source: Adapted from information collected at the Inter-American Development Bank from various official sources.
See also CEPAL (2006), Panorama Social de América Latina, Comisión Económica para América Latina y el Caribe,
Santiago.
marginal impact that fiscal policy has had in Latin America in reducing Gini coefficients,
normally by no more than one or two percentage points in the whole region and around
4.5 percentage points on average in Central America mainly because of Panama. At the
same time, an argument can be made that the attention paid to (inefficient) social
spending has distracted governments from their basic role in providing institutional and
real infrastructures that are needed by a modern society, and from focusing major
attention on the truly poor.
In spite of some progress reported for several Latin American countries in various
state reforms (see Lora, 2007b) (including political reform and reform of the judiciary, the
public administration, the tax systems, the fiscal decentralisation arrangements, the
regulatory framework, pensions, and so on), there is still considerable confusion about
what economic role the state should play in Latin America. Because tax revenues do not
seem to have increased in many countries in the past two decades (except mainly in
Argentina and Brazil) but social spending has increased, there remains the concern that
public resources have been diverted from financing fundamental public goods towards
social programmes that, for the most part, have not been focused on the truly poor – say,

on the bottom quintile of the income distribution.7 In recent years, more efforts have been
made to make some public transfers more focused. These transfers have been combined
with particular incentives for those who receive them, thus making the transfers
conditional. Examples of such programmes are the Chile solidario, the Bolsa Familia in
Argentina, Brazil, Panama and Peru, Progresa in Mexico, and the Hambre Cero in Nicaragua.
These are important programmes but, as long as the tax incidence does not change and as
long as much social spending continues to significantly benefit the higher quintiles, the
impact of these programmes on Gini coefficients will be moderate.
4. Fiscal instruments for state intervention

Before moving to the final section that discusses the future role of the state in
industrial and Latin American countries, it may be worthwhile to briefly describe various
instruments that have been and will be available to governments to promote various goals
such as allocation of resources, stabilisation of the economy, redistribution of income, and
economic growth. It will be noticed that economic growth has been listed as an additional
objective to the trilogy made famous by Musgrave (1959). The reason is that many policies
now pursued by governments do not easily fall within any one of the famous three
Musgravian objectives but are aimed at promoting economic growth rather than letting it
be the automatic result of the allocation of resources to promote public goods.
There has been a tendency in thinking of “fiscal policy” exclusively in terms of taxes
and public spending. This myopic approach has failed to acknowledge that fiscal policy can
be promoted with instruments other than the traditional ones of taxes and public spending.
Depending on the times and the circumstances, some of these instruments have been used
and preferred over others. It must be realised that, at times, while the instruments have
changed, the main governmental objectives have remained the four listed above. Let me
provide a more complete list of what I would call “fiscal” instruments. The list, though
more extensive, may still miss some important ones.
● Government spending is of course the most traditional and obvious instrument. Both
the level of public spending and its structure or composition are important and can be
considered as separate instruments.
● Taxes are the other obvious instrument. But taxes comprise at least four potential and
separable instruments, such as the level of taxation, the structure of taxation, “tax
expenditures” and tax incentives. All of these have been used, but some countries rely
more on levels and structures (Scandinavian countries) while others have relied more on
“tax expenditures” (Anglo-Saxon countries) and on tax incentives (Asian countries and
many developing countries).
● Nationalisation and privatisation of enterprises. At different periods and in different
countries, nationalisation or privatisation policies have been used as instruments to
promote government objectives. For many countries, the years after World War II were
periods of nationalisation. The 1980s and especially the 1990s, when the so-called
Washington Consensus became popular, became periods of privatisation.
● Expropriation of particular assets, such as land, has been at times, especially in Latin
America, a powerful instrument of fiscal policy. In today’s world, this instrument is less
used.
● Conscription. The government can tax some citizens, say the young, through the
process of forcing them to serve in the military or in other public services without (or

with little) compensation. Conscription was very important in the past, when people
were forced to provide their labour to build roads and public buildings, fight in wars, and
so on. This instrument, again, is less used today.
● Certification. The government may require that particular economic agents are certified
by the government, or by agencies authorised by the government, to be able to exercise
some economic activity. In some countries, certification is needed to engage in many
activities. At times certification has become an instrument to create “positional rents”
for groups of individuals.
● Regulations can easily replace taxing and spending with similar effects (see Tanzi, 1998).
Regulations have often replaced spending and taxing. Regulations can be pursued
through instruments such as multiple exchange rates, monetary repression and policy
loans, price and wage controls, preferential hiring quotas, and so on. In all cases, the net
result is to (implicitly) tax some economic activities and to subsidise others. Regulations
have played an overwhelming role in Latin American countries over past decades.
Finally, we must mention two relatively new instruments that are becoming
progressively more important, including in some Latin American countries. The first is
contingent liability – that is, the assumption on the part of the government of
responsibility for liabilities that might arise in the future in connection with the activities
of particular groups of economic operators. These contingent liabilities cost nothing when
they are assumed but may become very costly in future years if particular developments
occur.
Various examples of contingent liabilities are the following: a) insurances provided to
airlines by the government for terrorist acts; b) assumption of risk for low rates of returns
for investments made by private enterprises in connection with public-private
partnerships; c) implicit or explicit government protection for losses connected with
natural catastrophes such as floods, droughts, earthquakes, hurricanes, tidal waves;
d) liabilities for future expenses connected with the ageing of the population (pension
liabilities, health expenses, care for the very old, etc.); e) liabilities for fiscal deficits of
subnational governments, especially when connected with unfunded mandates;
f) liabilities for banking crises; g) potential liabilities for global warming; and so on. The
main problems with these contingent liabilities are that they can become very costly to the
government but they are not shown in the budgets when the government assumes them.
However, through them, the government may influence the behaviour of the economy. For
example, it can have some infrastructures built by private operators.
The other instrument, one that is still in its infancy but that might become an
important tool for government policy in the future, is the allocation of future assets to
specific categories of citizens. For example, a government could legislate that, in place of
social programmes, the government will open an account with a given amount of money
in it for every newborn child or for every person that, say, reaches a given age (16? 18?) and
let the individuals buy from the market the desired protection against particular risks such
as illnesses, unemployment, etc. Thus, flow expenditures connected with public spending
become stock allocations, and the government reduces its involvement in providing social
services. The allocation could be based strictly on demographic information.
All these instruments have played and are likely to continue playing their part in the
economic role that the state assumes in the economy. However, their relative importance
is likely to change in the future.

5. The future role of the state in advanced countries

We have argued above that, in advanced countries, public spending may be too high.
The real difficulty that would be faced by a government in reducing and changing the role
of the state in the economy is not that a state that spent less would necessarily imply a
reduction in economic welfare, but rather that it would face strong political reactions on
the part of those whose current or expected standards of living have come to depend on
the existing public programmes. Government spending programmes inevitably create
strong constituencies: pensioners, those close to the retirement age, school teachers,
public employees, those who receive public subsidies, and others. These constituencies
consider a reduction in public spending as a negative-sum game. Therefore, the evidence
that some countries with relatively low levels of public spending operate well cannot be
interpreted as an indication that high-spending countries could easily and painlessly
reduce their public spending. It only means that, after the short or medium-run costs of
reform have been paid, a country that had made the right reforms could continue to have
high socio-economic indicators (high social welfare) with significantly lower public
spending and more individual liberty. This has happened to Canada which, after a decade
of sharp reductions in public spending, has seen its human development index go up.
Levels of public spending at any one time tend to be set by past political trends and
policies rather than by informed decisions based on the best evidence of the day.8 Annual
budgets are typically incremental. They rarely address the question of whether an activity
should be continued or discontinued. For this reason, zero-base budgeting has not had
much success. At any given moment, the level of public spending depends substantially on
the entitlements and claims on the government created in past periods, often by previous
governments. It does not depend on well-thought-out analyses and considerations of what
the state could or should do in a modern and more sophisticated market economy. It rarely
matches the spending level that a modern government might wish to have if it had the
freedom and the courage to change the status quo.
For the reasons mentioned above, there is often no realistic possibility of a genuine
zero-base assessment of the optimal economic role of the state at a given moment in time.
However, if past mistakes, or past misguided actions, have determined the current high
level of public spending, that level cannot be assumed to be optimal or nearly optimal in an
economic or even a political sense. It is simply the result of political opportunism. It is thus
important to distinguish, at least analytically, what could be the optimal role of the state in
the long run from its current role.
Should the governments of today’s advanced countries simply accept the status quo?
Or should they put in motion radical reforms that in the long run – say, over a generation –
would bring the role of the state more closely in line with an ideal or currently
economically optimal role? Another way of putting the question is: what economic role
should the state play, especially in relation to public spending, in advanced industrial
countries in the 21st century? This is a difficult question to answer because, inevitably, the
answer must reflect political biases as well as the importance that one attaches to the
transitional costs of getting from where we are today to where we could be, say, 20 or
30 years from now. The greater the importance that one attaches to the transitional costs,
and especially to the political costs, the greater will be the inclination by policy makers to
maintain the status quo and the current spending programmes. It is natural that
governments want to remain in power rather than risk reforms that demand much

political capital. Let me focus on some essential elements to consider when dealing with
the above question.
The first of these elements is the recognition that, in a market economy, there should
be a relationship between what the market is capable of delivering and what the
government should do. After all, in a market economy, the state is supposed to correct the
mistakes made by the market, or compensate for its shortcomings, but not replace the
market. More efficient markets should require less government. In a society where the
market is underdeveloped for a variety of reasons, so that it is not capable of performing
some important tasks well, there will be more theoretical justification for the state to step
in to correct or complement the market in some of these functions. This was the main
argument that, over the years, led to the enormous expansion in the economic role of the
state, especially in the last half century. It is also the argument that is often made for more
market intervention in developing countries.
As markets develop and become potentially more efficient in performing various tasks
and in allowing individuals to satisfy various needs directly and not through the
intermediation of the government – including the need to buy protection against particular
events that could have economic consequences – the theoretical justification for
governmental intervention through public spending decreases. This should result in a fall
in public spending. A perfect market, if it existed, would of course dispense with the need
to have any government at all. However, a perfect market cannot exist. Furthermore, some
regulatory government role is needed to make or keep the market as efficient as it can be.9
A second important element is that, when in past decades governments entered a
given sector, they introduced laws and regulations that facilitated and justified their own
intervention in that sector. This inevitably made it more difficult, or at times impossible,
for the private sector to develop private alternatives in that sector. Governmental
involvement created public monopolies that eliminated the possibility of developing
private alternatives. In many European countries, public monopolies in energy,
communications, postal services, transportation, the provision of pensions, health
services, education and in several other activities prevented the market from developing
potentially efficient private alternatives to the public programmes that existed in these
areas. This created the belief, on the part of a large section of the public, that the public
sector must remain engaged in these areas if the welfare of citizens is to be protected.
A third element is that other factors are changing the conditions for providing the
services that citizens need: a) rapid technological innovations; b) the growing
sophistication of the market on a global scale; c) the development of global financial
services; and d) globalisation in general. The current role of the state in many countries
was developed mostly in the period after World War II when, for a variety of reasons, the
markets were far less developed than they are, or can be, today. The markets were also far
more closed. This was the period when the concept of a “mixed economy” that combined
elements of central planning and of market economies, and assigned a large and benign
economic function to the state, seemed natural and became most popular. At the time, it
must have seemed reasonable for governments to take over many new responsibilities.
The economic profession generally encouraged them to do so (see Tanzi, 2006a).
In spite of many obstacles imposed by governments on markets, and the existence of
many public monopolies, markets have become much more sophisticated over the years.
With the right governmental guidance, they could become even more sophisticated.

Various developments have made it possible for the private sector to replace activities that
had previously been public. Technological developments have destroyed the presumption
that there are “natural monopolies” in the generation of electricity, in various forms of
transportation (railroads, airlines), in communications (telephones, telegraphs), in postal
services, and in other areas. In several countries, the government has started to withdraw
from some of these activities, and relatively well-functioning private markets have quickly
developed in them. This is the case also for private pensions, financial services, and
transportation and communications. In most cases, the economic welfare of the average
citizen has not been damaged by these developments. On the contrary, and with
exceptions that often are much publicised, services have frequently improved in quality
while prices have fallen significantly.
Major developments in financial markets, including greater international capital
mobility, have removed the presumption that financial savings must be invested
domestically and that governments should be involved in the allocation of private savings
and credit. In financial markets as well as in the other areas mentioned above, there is,
however, a very important surveillance and regulatory function that governments must
perform. This function cannot, or should not, be left to the private sector. It is a function
that should be taken seriously by the government but that, so far, has not been because
governments have focused on their spending role. This regulatory function should be part
of the core activities of the state.
A fourth element is that globalisation, in its various aspects, is bringing major changes
to the way markets operate or could operate. Foreign competition can make domestic
markets more efficient by destroying or reducing the power of domestic private
monopolies and by offering alternatives. Globalisation is affecting and can affect public
sector activities in other ways. By eliminating frontiers, or making them less constraining,
globalisation is creating the potential for more options for both citizens and governments.
For example, educational and health services can now be obtained in other countries more
easily than in the past. In one sense, they have become tradable goods.10 Public sector
procurement can now benefit from foreign participation, thus reducing government costs.
Savings can be invested abroad. This access to foreign markets has created options beside
the ones traditionally available domestically and which were often available only from the
public sector.
The current public spending policies of many European countries are likely to prove
unsustainable in future decades because of the impact of demographic development on
public spending and of globalisation on government revenue. Demographic developments
with unchanged policies will dramatically push up various public expenditures, and
especially those for health, pensions, and care for the very old. This increase in spending
will come on top of already precarious public finances and high levels of taxation and
public debt in several European countries.
The impact of globalisation on government revenue and tax competition could make
it impossible for many European countries to compete with countries such as China, India,
Korea, Mexico, Vietnam and others while maintaining tax levels that are already very high
and, in several cases, not capable of financing even today’s public expenditure. The impact
of the baby boom on social spending is yet to be felt fully, and the impact of globalisation
and tax competition on tax revenue has just started to make itself felt. In the next ten
years, both could be in full force. To prevent major future fiscal difficulties, there is only

one way out: to try patiently, systematically and rationally to scale down the spending role
of the state in the economy while making a serious and competent effort to increase the
efficiency of the private sector as well as that of the public sector. This would make it
possible for the private sector to step in and replace the government role in covering some
important economic risks that citizens face, thus allowing the public sector to reduce its
spending.
The reduction in the spending role of the state should be based on three pillars. The
first should be the improvement in the working of the private market through the effective
use of the government’s regulatory power. In this role, the government will need to be
ruthless and efficient. In a market economy, this is surely the most important role of the
state.
The new government role in protecting individuals against risks with economic
consequences can be played in two ways. First, by requiring individuals to buy some
minimum protection directly from the market. Governments already force individuals to:
a) get insurance for their cars; b) get driving licenses; c) have fire alarms in some buildings;
d) build safe buildings; e) wear seat belts while driving; f) quit smoking in public places;
g) get vaccination against some diseases; h) stay in school until a given age; and i) take
other actions aimed at making individuals pay for, or avoid being damaged by, events that
might affect them as well as others. Why not apply the same principle vis-à-vis the
treatment for major illnesses, the payment of minimum pensions, or other similar needs?
The second pillar should be the progressive substitution of programmes with
universal, free or almost free access, toward programmes targeted for the poor and based
exclusively on ascertained and documented needs. Universal programmes (such as free
health services for all, free higher education for all, etc.) are easier politically but are
expensive. Targeted programmes can save a lot of money but are more demanding
politically and in terms of information. Problems connected with poverty traps must
receive specific attention. The difficulties in these changes cannot be minimised.
The third pillar should be the progressive exploitation of new opportunities offered by
globalisation for services not domestically available or available at high costs – such as
elaborated medical procedures, advanced technical training, relatively safe channels for
money saved for old age, and so on. These services can now be bought from foreign
providers if the domestic private market is unable to provide them at competitive prices,
and the government still has the obligation to provide these services to some citizens.
It is obvious that much thinking and much experimentation will be required over
future decades to bring out the progressive and efficient scaling down of public spending
and tax levels. It is also inevitable that mistakes will be made. But when the transformation
comes, it is likely to include the three pillars mentioned above. Without that
transformation, the public finances of several high-spending countries will become more
and more a public concern.
6. The future economic role of the state in Latin America

For many years, the countries of Latin America did not have well-developed markets
and had little capacity or will to raise high tax revenue. Repressed financial markets,
multiple exchange rates, high import tariffs, price controls, and politically influenced
public enterprises dramatically restrained the development of private markets. As we saw
earlier, the (unweighted) level of taxation increased little over recent decades. It was only

in the most recent years that some important countries, and especially Argentina and
Brazil, have succeeded in significantly increasing their tax level. However, while their
ability to raise taxes was restrained, the Latin American governments were not immune
from strong popular pressures to spend more and to play a larger social role in the
economy. Being unable to raise the tax level, they relied on regulations to play such a role.
This in turn made it more difficult for private markets to develop.
Repressed financial markets that favoured some loans, multiple exchange rates that
favoured some imports, high tariffs on some imports and low tariffs on others, public
enterprises that employed too many people and sold some of their services too cheaply,
controlled labour markets that made firing costly, minimum wages, controlled prices for
some basic commodities, rent controls on housing, export taxes to reduce the domestic
price of some goods, and so on, can be considered as an alternative way of exercising a
large government role in the economy – i.e. an alternative to high public spending. In effect,
the governments created primitive regulatory welfare states, at least in intention if not in
results.
The “Washington Consensus” was to a large extent a frontal attack on this government
role. That consensus aimed at removing many of the regulations that Latin American
governments had used in the decades before the 1990s to protect urban workers through
regulations and not through public spending. It is easy to criticise this role of the state
because it is clearly inefficient and an obstacle to economic growth, the point stressed by
the Washington Consensus. However, for many workers (and especially urban workers) the
protection appeared real and even helpful. The dismantling of this “regulatory welfare
state” may have led to the recent reactions, in several Latin American countries, against
the Washington Consensus. These reactions are evident from responses to questions
asked by Latinobarometro of Latin American citizens and from recent election results. In
principle, the removal of many of these regulations could have been replaced by public
spending. But the limitation in tax revenue made this impossible for many countries in the
region. Thus, urban workers lost some of the indirect social protection that they thought
they had. Some lost their jobs in public enterprises or in enterprises protected by high tariff
walls.
In the future, the governments of the Latin American countries where taxes are low
ought to try to make their tax systems more productive. However, this will be difficult as
long as personal income taxes continue to contribute little to total tax revenue.11 In
countries where Gini coefficients approach 0.60, flat-rate taxes are not likely to generate
the needed revenue and are not policies that should be contemplated by the Latin
American governments. Much of the potential tax base is in the top deciles of the income
distribution. This has to be recognised by policy makers and should be reflected in the
incidence of taxation.12
If more taxes could be collected, the higher revenue should be directed first to
modernising the state, by improving the quality of the basic services that it provides. Basic
education and essential health services should receive full attention, but so should
services related to personal safety, justice, public transportation and similar.
The state should reduce its involvement in activities that are costly but are used
mainly by higher income groups and where the services could be bought by these groups
directly from the market. Higher education would be one of these activities. Incomes for
old-age pensions could be another. However, an argument could be made for helping old

people who were too poor to save for old age and that reach old age without a pension
because they never had a regular job. The latter include a significant share of the
population of Latin America.
As we argued above, those who are covered by public pension systems are often not
the poorest among those who reach pensionable ages. In many developing countries,
available forecasts of the liabilities of public pension systems for future years (i.e. the
present value of the stream of future pensions promised to workers under current
legislation less the present value of the stream of future social security taxes) indicate
growing financial liabilities for many countries for public pension systems. These large
pension liabilities (or hidden pension debts) have been the main reasons why, in recent
years, several countries around the world introduced some version of the pension model
first introduced in Chile in 1981 that privatises all or part of the pension systems.
In the Chilean model, the government reduces its spending responsibility and
increases its regulatory responsibility. This represents a fundamental step toward a state
that would exercise its social responsibility not by taxing and spending but by requiring its
citizens to follow prescribed actions.13 The state can then focus its spending role on
providing truly public goods and on assisting the truly poor instead of those in middle or
higher classes who should be able to look after themselves. The state does not abandon its
social goals; it mainly changes the instruments through which it pursues them.
In a world in which many markets have or can become more efficient and more global,
perhaps with the push of international institutions, if governments became more forceful
and efficient in their regulatory role, it should be possible for them to reduce the
intermediary, spending and taxing role that they have played, especially in public pensions.
In the traditional public pension system, workers pay social security taxes (based on their
wages and salaries) to the government during their working life. When they reach the official
pensionable age, the government is expected to repay them with monthly pensions that bear
some but often not a close relationship to the taxes they had paid during their working life.
Thus, in some sense the government operates as a savings bank for each worker but without
a real guarantee that what they contribute to the bank will determine what they will take
out. The problem is that the bank is empty most of the time, and is often in the red. The
contributions of the current workers go quickly to pay the pensions of those already retired.
The workers often see their contributions as taxes and not as savings.
Often governments are forced to use general public revenue to be able to meet their
pension obligations because the contributions by current workers are not sufficient to meet
the obligations. This system is exposed to problems created by demographic changes that
increase the number of retirees with respect to the number of workers, by the inefficiency of
governments in using productively the taxes that they receive, by political pressures to
increase the level of pensions and, most importantly, by the fact that people who have not
contributed – because they were not part of the formal economy – are not entitled to receive
a public pension when they reach retirement age, regardless of their needs.
Given the characteristics of the economies of developing countries and emerging
markets, there are valid reasons to suggest that governments should change their basic
role that has largely been to protect those lucky enough to have had jobs in the formal
labour market while forgetting about those who had been in the informal economy. The
latter were often the poor majority. It can be argued that the basic role of the state should
require that it pay more attention to the truly poor. This can be done by seeing the public

role as providing a minimum income, or a minimum pension, to all citizens who reach a
given old age. In this system, age would be the sole criterion for receiving this income.
The minimum pension could be estimated as a fixed proportion (say, as an example,
25%) of the per capita income of the country in the most recent year for which this
information is available. Thus, the absolute level would change automatically, as the per
capita income changed. The pensionable age could be set as a constant proportion of life
expectancy, say at 80 or 90% of the average life expectancy of the country. Thus, it would
change automatically when the life expectancy changed. The pensions would be paid out
of general revenue and not out of payroll taxes. Therefore, there would not be the
disincentive effects of high social security taxes on the labour markets.
In sum, the variables needed would be: a) the country’s per capita income; b) the
country’s life expectancy; and c) the age of the pensioners.14 Such pensions would make a
significant impact on poverty reduction, because many who are old and poor have no
incomes or have incomes that are very low. Those receiving these pensions could continue
to be economically active if they so desired. They would still be entitled to the pension. The
administration of these pensions would be simple and the cost to the countries not very
high, because the pensionable age could be set at a level that would limit the number of
eligible people. The state would be performing one of its basic roles, that of assisting the
poor, and doing so without distorting the labour market or removing individual
responsibilities.15 This is suggested as an alternative to the current systems. But, of course,
for a long time those who have acquired rights in the present systems would continue to
receive the pensions from those systems.
In addition to this general basic scheme, the government could help the workers in the
formal labour market by providing a regulatory framework (and the needed information)
that would assist all citizens who wished to do so to invest part of their savings in income-
producing assets that would provide them with additional resources when they retired,
beyond the minimum pensions. If channeled to specific categories of assets, these
investments could benefit from a deduction from taxable incomes as, for example, IRA
accounts (individual retirement arrangements) do in the United States. This would create
a culture that would encourage individuals to take personal responsibility for their actions.
Obviously, specific transition problems would need to be solved, and these problems might
be difficult ones.
7. Concluding remarks
The role that the state plays in economic matters, in both advanced and developing
countries, has been much influenced by the ideologies of the past. They often do not reflect
modern thinking, modern needs, and modern possibilities. Discussions have often been
directed toward the instruments used and not enough to the goals to be achieved. For
example, public spending has been defended even when much evidence indicates that it is
less beneficial to the poor than generally believed.
Furthermore, public spending can be of an “exhaustive” kind – that is, the kind that
uses resources directly – or it can be in cash. The relative use of these alternatives should
depend on the efficiency of public sector institutions in performing some activities. When
the high-spending welfare states of Europe reduced their public spending in the past
decade, they generally preferred to reduce cash transfers rather than transfers in real
services. However, it is less clear that this would be the better policy for Latin America,

where the institutions that deliver social services are likely to be less efficient in general.
Thus, in these countries, a role of the state based on objective criteria (say, age) and using
cash transfers or the creation of earmarked cash allocations to particular groups might, in
particular circumstances, be the preferred alternative. We need to devote more analyses to
these possibilities.
Notes
1. For the Argentine experience, in its attempt to create a European-style welfare state, see Tanzi
(2007).
2. Quoting from the Human Development Report 2007/2008 (UNDP, 2007, p. 225): “The … HDI is a
composite index that measures the average achievements in a country in three basic dimensions
of human development: a long and healthy life; access to knowledge; and a decent standard of
living. These basic dimensions are measured by life expectancy at birth, adult literacy and
combined gross enrolment in primary, secondary and tertiary level education, and gross domestic
product (GDP) per capita in purchasing power parity US dollars.”
3. There is actually a correlation of 0.33 between higher spending levels and (poorer) HDI scores. See
the line in Figure 1.
4. The percentages of those who thought that taxes are well spent ranged from a low of 10% in Peru
to a high of 37-38% in Chile and Venezuela.
5. We reject here the view that private citizens are not able to make good decisions with the money
that they control.
6. The truth is that the amount of money spent on the truly poor (say, the bottom 20% of the income
distribution) is a small proportion of the total in most countries.
7. Actually the focus on tax revenue gives a distorted impression of the public revenue available to
Latin American governments. Many Latin American governments receive large public revenue
from the natural resources that they own. This is certainly the case for Bolivia, Ecuador, Mexico,
Venezuela, and several other countries. In 2005, total public revenue was about 28% of GDP; that is
high by international standards.
8. This section draws from Tanzi (2005).
9. As Adam Smith recognised as far back as 1776, without some government controls the private
sector tends to develop monopolistic practices. Thus, as Paul Krugman has put it, it is necessary
for government “to exercise adult supervision on markets running wild”.
10. The greatest British export today is educational services. International shopping for health services
is becoming common, and some hospitals have been set up specifically to attract foreigners.
11. Given the high concentration of income in most Latin American countries, personal income taxes
contribute shockingly little tax revenue.
12. Flat-rate taxes may be good tools for countries with low Ginis, such as several “transition”
countries.
13. It is a paternalistic role of the state based on mandates rather than on spending. Both differ from
a state based on individualistic responsibilities in which the state would play no role.
14. In countries where no records are available for some citizens, this would be a problem.
15. Note that the pensions received would be a large share of the incomes of the poorest people and
an almost insignificant share of the incomes of people in the high percentiles.
References
CEPAL (2006), Panorama Social de América Latina, especially Chapter II, “Gasto Social en America Latina
y el Caribe: tendencias recientes, orientación y efectos redistributivos”, Comisión Económica para
América Latina y el Caribe (Economic Commission for Latin America and the Caribbean), Santiago.
ECLAC (2005), Panorama Social de América Latina, Capitolo II, Economic Commission for Latin America
and the Caribbean, Santiago.

IMF (2007), Regional Economic Outlook, Western Hemisphere, November, International Monetary Fund,
Washington DC.
Lora, Eduardo (2007a), “Trends and Outcomes of Tax Reform” in E. Lora (ed.), The State of State Reform in
Latin America, Inter-American Development Bank and Stanford University Press, Washington DC.
Lora, Eduardo (ed.) (2007b), The State of State Reform in Latin America, Inter-American Development Bank
and Stanford University Press, Washington DC.
Moser, Caroline O.N. (ed.) (2007), Reducing Global Poverty: The Case for Asset Accumulation, Brookings
Institution Press, Washington DC.
Musgrave, Richard A. (1959), The Theory of Public Finance , McGraw Hill, New York.
Musgrave, Richard A. (1998), “The Role of the State in the Fiscal Theory” in P.B. Sorensen (ed.), Public
Finance in a Changing World, MacMillan Press, Ltd., Houndmills, Basingstoke, United Kingdom.
OECD (2007a), OECD Economic Outlook No. 81, Volume 2007/1, June, OECD Publishing, Paris.
OECD (2007b), Latin American Economic Outlook 2008, OECD Development Centre, Paris.
Simons, Henry (1938), Personal Income Taxation: The Definition of Income as a Problem of Fiscal Policy.
Smith, Adam (1776), The Wealth of Nations.
Sorensen, Peter Birch (ed.) (1998), Public Finance in a Changing World, MacMillan Press, Ltd., Houndmills,
Basingstoke, United Kingdom.
Tanzi, Vito (1974), “Redistributing Income through the Budget in Latin America”, Quarterly Review,
Banca Nazionale del Lavoro, March.
Tanzi, Vito (1998), “Government Role and the Efficiency of Policy Instruments”, Chapter 2 in
P.B. Sorensen (ed.), Public Finance in a Changing World, MacMillan Press, Ltd., Houndmills,
Basingstoke, United Kingdom.
Tanzi, Vito (2001), “Globalization, Technological Development and the Work of Fiscal Termites”,
Brooklyn Journal of International Law, Vol. XXVI, No. 4.
Tanzi, Vito (2005), “The Economic Role of the State in the 21st Century”, Cato Journal, Vol. 25, No. 3, Fall.
Tanzi, Vito (2006a), “Final Remarks” in T. Boeri et al. (eds.), Structural Reforms without Prejudices, Oxford
University Press, Oxford, United Kingdom.
Tanzi, Vito (2006b), Death of an Illusion? Decline and Fall of High Tax Economies, Politeia, London.
Tanzi, Vito (2007), Argentina: An Economic Chronicle, Jorge Pinto Books, Inc., New York.
Tanzi, Vito and Ludger Schuknecht (1997), “Reconsidering the Fiscal Role of Government: The
International Perspective”, The American Economic Review, May.
Tanzi, Vito and Ludger Schuknecht (2000), Public Spending in the 20th Century, Cambridge University
Press, Cambridge, United Kingdom.
UNDP (2007), Human Development Report 2007/2008, United Nations, New York.

The Role of The State and Public Finance in The Next Generation

Uploaded by

Copyright:

Available Formats

The Role of The State and Public Finance in The Next Generation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Role of The State and Public Finance in The Next Generation

Uploaded by

Copyright:

Available Formats

ISSN 1608-7143

OECD Journal on Budgeting

The Role of the State and Public Finance

2 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

2. Globalisation and taxes

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 3

4 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 5

6 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

● There is a growing interest in flat-rate taxes and in “consumption-based taxes”.

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 7

8 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

3. The role of public spending

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 9

10 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

Table 1. Public spending and indices of human development in industrial

Figure 1. Public spending and the Human Development Index

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 11

Table 2. Spending levels in selected countries (percentage of GDP)

Norway 1 55.7 41.0 –14.7

12 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

Table 3. Rankings in the Human Development Index

Source: UNDP (2007), Human Development Report 2007/2008, United Nations,

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 13

14 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

Table 4. Distribution of social spending by income quintiles for education

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 15

Table 5. Distribution of social spending by income quintiles for health

Table 6. Distribution of social spending by income quintiles for social security

16 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

4. Fiscal instruments for state intervention

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 17

18 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

5. The future role of the state in advanced countries

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 19

20 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 21

6. The future economic role of the state in Latin America

22 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 23

24 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 25

26 OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008

OECD JOURNAL ON BUDGETING – VOLUME 8 – No. 2 – ISSN 1608-7143 – © OECD 2008 27

You might also like