The drag of public
Recent policy reforms in many low-income countries have failed
to reduce significantly the dominance of the public sector in the economy. The
inefficiency of public enterprises and the spread of their losses throughout the
economy continue to drag down savings, investment, and economic growth.
The dominance of the public sector
In most low-income countries, the public sector is dominant in
infrastructure, heavy industry, agricultural marketing, and
finance—typically accounting for more than 10 percent of GDP, 20 percent of
employment, and 25 percent of investment. But public enterprises account on
average for 14 percent of GDP, 18 percent of employment, and 27 percent of
investment in Sub-Saharan Africa. In many countries public enterprises have
retained their shares of production and employment even after public sector
reform. This persistence reflects continuing investment in public enterprises as
well as the slow pace and limited scope of divestiture. With few exceptions,
divestiture programs in low-income countries have consisted mainly of
privatization or liquidation of small enterprises, with little impact on the
structure or performance of the economy.
Public enterprises dominate large industry, especially in Asia.
In China, stateowned enterprises are particularly dominant in capital goods,
such as steel, machinery-building, cement, and energy production, despite the
growth of COEs and private firms during the past decade. This pattern also holds
in other Asian economies that historically have adopted a mixed economy
approach. In India, the public sector owns 50 percent of industrial assets,
mainly in heavy industry; and accounts for 25 percent of industrial production.
In Pakistan, the integrated steel works is publicly owned, while the private
sector owns small arc furnaces, foundries, and ship-breaking. In the heavy
engineering sector, large state enterprises in machine-building and machine
tools exist alongside private engineering firms.
Agricultural development is still circumscribed by heavy state
intervention, mainly through direct involvement in the marketing of inputs,
outputs, and support services. The inefficiencies of these institutions have
muted the supply response of the agricultural sector since increases in the
price of tradables resulting from exchange rate reform often have not been fully
passed through to producers. Even when marketing systems have been liberalized,
private sector entry has been limited because of restrictive and discretionary
licensing systems and the limited, high-cost credit provided by a
state-dominated financial system.
The dominance of public enterprises in the economies of
low-income countries means, when they perform poorly, that they drag down
overall growth rates. In Kenya, for instance, total factor productivity in
majority-owned state enterprises declined 3 percent during 1986-91, compared
with a 5 percent increase in the private sector. If state enterprises had been
as productive as the private sector during this period, Kenya's GDP would have
grown by an additional 2 percent each year. Public enterprise inefficiency has
an even larger impact in countries with more dominant public sectors than
Kenya's.
For private firms, the inefficiency of public enterprises raises
the cost of doing business. In Sub-Saharan Africa, the poor quality of utilities
and infrastructure services adds 10-25 percent to firms' costs. It has been
estimated that the efficiency losses of the publicly owned petroleum sector in
Sub-Saharan Africa amount to about $1.4 billion a year (box 1.3)-more than the
Bank's annual disbursement of adjustment policy loans to the region, and about
two times the foreign direct investment in 1993.
Fiscal drain and financial sector leakages
The losses of unprofitable public enterprises are typically
financed by a combination of transfers from the fiscal budget, a tax on the
financial system through forced investments and portfolio requirements, and
domestic credit, which increases the cost and reduces the volume of credit to
the private sector. Public enterprise losses often are large relative to the
size of low-income economies and reduce the availability of resources to finance
private sector activity as well as alternative public investments. Often, the
lack of data on state-owned enterprises, their losses, and the distribution of
those losses throughout the economy makes it even more difficult to deal with
them.
Covering public enterprise losses with fiscal transfers forces
governments to finance larger fiscal deficits and increase tax revenues or
reduce public expenditures in other areas, or both. Increasing tax revenues to
finance the losses of state enterprises is difficult in many low-income
countries, where the tax base is narrow and tax administration weak. During the
1980s most reforming low income countries reduced their budget deficits by
cutting spending rather than by increasing revenues. Faced with strong political
pressures, governments often sacrificed public investments with long-term
benefits—including social and infrastructure investments—rather than
cut current expenditures. In Sub-Saharan Africa, for example, average capital
expenditures fell from 8.7 percent of GDP to 6.1 percent during the 1980s.
|
BOX 1.3 THE HIGH COST OF INEFFICIENT PUBLIC ENTERPRISES:
PETROLEUM IN SUB-SAHARAN AFRICA
Oil products in Sub-Saharan Africa represent 70 percent of total
commercial energy consumption, consume one-third of available hard currency
reserves, and account for two-fifth-sofindirect taxes. The petroleum products
industry in Sub-Saharan Africa is extremely inefficient. A 1992 study estimated
total annual potential savings of more than $1.4 billion (approximately $51 a
ton). The scale of these efficiency losses can be gauged from the facts that
they exceeded IDA's total disbursements in that year to Sub-Saharan Africa for
adjustment programs and they were twice the FDI inflows. The key areas of
inefficiency: Procurement ($690 million). Most countries suffer from lack of
foreign exchange and from monopolistic and inefficient government involvement in
procurement. Refining ($550 million). Most refineries are too small, lack
adequate technology, and face inadequate incentives for efficiency improvements.
Distribution ($180 million). Transportation infrastructure is run-down. The
study indicated that more than half the estimated savings could be achieved
without new investments— by changing policies and procedures, particularly
opening markets to competition and providing adequate pricing policies, and by
regional cooperation. The key investment require meets would be for
rationalizing and rehabilitating transport and storage facilities.
Governments in low-income countries often have controlled
interest rates to reduce the fiscal cost of accommodating budget deficits. But
reduced real deposit rates, often negative, have been a disincentive to savings
(encouraging savers to seek nonfinancial assets), fostered capital flight, and
encouraged over-borrowing by enterprises. As a result, financial depth is
shallower in low-income countries than in other economies. This is especially
true in Sub-Saharan Africa, where bank deposits are only 15-20 percent of GDP
(compared with 40 percent in Pakistan, 60 percent in India, and 80 percent in
China).
The financing of state enterprise losses through the domestic
financial sector—directly, through credit to state enterprises, or
indirectly, through credit to the government to finance budget
transfers—increases the cost and reduces the availability of credit to the
private sector. In Africa, the public share of domestic bank credit averages
around 40 percent and exceeds 80 percent in some countries. In countries where
banks are required to invest a certain portion of their assets in government
paper or other forced investments at belowmarket returns, the effect is a tax on
the financial sector that is usually reflected in higher intermediation margins.
Even in countries that do not require banks to invest in particular industries,
the allocation of credit often is not fully market-determined. Where state-owned
banks dominate the financial sector, as they do in most low-income countries,
they often channel resources to public enterprises rather than act as
independent financial intermediaries. In addition, where the public sector is a
dominant force in the real sectors, banks are under continual pressure to lend
to public enterprises to prop up the banks' own nonperforming assets and to prop
up their client firms that are linked with state enterprises (figure 1.9).
In many low-income countries, state enterprise losses are so
large that they are two to three times public expenditures on education and
health. Typically, a majority of these losses is attributable to a relatively
small number of large state-owned enterprises. These normally are utilities
(power, telecommunications, water, railways, ports), capital-intensive
industries (steels, fertilizers, chemicals, pulp and paper, cement), and
agricultural marketing boards. The fiscal drain of public enterprise losses is a
form of continual government dissaving that makes it difficult for low-income
countries to generate the total savings needed for accumulation and growth.
|
Figure 1.9 Public enterprises crowd
out private credit, 1986-88
In Africa, the drain on the fiscal budget and the leakages from
the financial system resulting from public enterprise losses can be as high as
12 percent of GDP, nearly as great as gross domestic savings (averaging 12.5
percent) and two to three times the spending on health and education. So, few
resources remain for investment, public or private.
In China, fiscal losses of SOEs (that is, not counting losses in
the banking system) are 3-5 percent of GDP. They are less significant in
crowding out investment since gross savings have been about 39 percent during
1987-93. Still, accommodating such losses threatens the viability of the
financial system, where it is estimated that 15-20 percent of the loans of state
banks are nonperforming. The fast-growing economies would thus have done better
but for the drag of the public sector on savings and
investment.
