In Favour of Financial Liberalisation

Maxwell J. Fry

The Economic Journal, 107 (May), 754 - 770

 INTRODUCTION

Many developing country governments find it virtually
impossible to satisfy their inter-temporal budget constraint
with conventional tax revenue
 Hence, they rely on revenue from the inflation tax and they
reduce their interest costs through financial repression.

This paper suggests that financial repression is a particularly
damaging quasi-tax from the perspective of economic growth.

A key stylied fact about financial systems in developing
countries
 Dominated by commercial banks
 Assets of insurance and pension companies are minuscule
in most developing countries.
 INTRODUCTION
 Development financial institutions such as agricultural and
development banks are also small compared with the
commercial banks.
 Commercial bond markets are typically thin and
government bond markets are often used only by captive
buyers obliged to hold such bonds to satisfy liquidity ratio
requirements or to bid for government contracts.
 Although equity markets are sizeable in several developing
countries, their role in the process of financial
intermediation between the household and business sectors
remains small.
 At best stock markets play a minor role; more often they
resemble gambling casinos and may actually impede
growth in developing countries (Singh, I 997).
I. THE FINANCIAL REPRESSION PARADIGM

McKinnon (I973) and Shaw (I973) argue that financial

repression - indiscriminate 'distortions of financial prices
including interest rates and foreign-exchange rates' - reduces
'the real rate of growth and the real size of the financial system
relative to nonfinancial magnitudes.

This strategy has stopped or gravely retarded the development
process.
I. THE FINANCIAL REPRESSION PARADIGM

The essential common elements of the McKinnon-Shaw model
are:
 (a) a saving function that responds positively to both the
real rate of interest on deposits and the real rate of growth
in output;
 (b) an investment function that responds negatively to the
effective real loan rate of interest and positively to the
growth rate;
 (c) an administratively fixed nominal interest rate that holds
the real rate below its equilibrium level;
 (d) inefficient non-price rationing of loanable funds.
I. THE FINANCIAL REPRESSION PARADIGM

In the McKinnon-Shaw model,
 banks allocate credit not according to expected productivity of
the investment projects but according to transaction costs and
perceived risks of default.
 Quality of collateral, political pressures, 'name', loan size, and
covert benefits to loans officers may also influence allocation.

Even if credit allocation is random, '
 the average efficiency of investment is reduced as the loan rate
ceiling is lowered because investments with lower returns now
become profitable.
 Hence, adverse selection from the perspective of social welfare
occurs when interest rates are set too low and so produce
disequilibrium credit rationing of the type described here.
I. THE FINANCIAL REPRESSION PARADIGM

Interest rate ceilings distort the economy in four ways.
 First, low interest rates produce a bias in favour of current
consumption and against future consumption. Therefore,
they may reduce saving below the socially optimum level.
 Second, potential lenders may engage in relatively low-
yielding direct investment instead of lending by way of
depositing money in a bank.
 Third, bank borrowers able to obtain all the funds they
want at low loan rates will choose relatively capital-
intensive projects.
I. THE FINANCIAL REPRESSION PARADIGM
 Fourth, the pool of potential borrowers contains
entrepreneurs with low-yielding projects who would not
want to borrow at the higher market-clearing interest rate.
 To the extent that banks' selection process contains an
element of randomness, some investment projects that are
financed will have yields below the threshold that would be
self-imposed with market-clearing interest rates.
I. THE FINANCIAL REPRESSION PARADIGM

Raising the interest rate ceiling towards its competitive free-
market level
 increases both saving and investment.

Changes in the real interest rate
 trace out the saving function in this disequilibrium
situation.

Raising the interest rate ceiling also
 deters entrepreneurs from undertaking all low-yielding
investments that are no longer profitable at the higher real
interest rate.
 Hence the average return to or efficiency of aggregate
investment increases.
I. THE FINANCIAL REPRESSION PARADIGM
 The output growth rate rises in this process, so further
increasing -saving. Thus, the real rate of interest as the
return to savers is the key to a higher level of investment,
and as a rationing device to greater investment efficiency.

The increased quantity and quality of investment interact in
their positive effects on the output growth rate.

The policy prescription for the financially repressed economy
examined by McKinnon and Shaw is to raise institutional
interest rates or to reduce the rate of inflation.

Abolishing interest rate ceilings altogether produces the
optimal result of maximising investment and raising still
further investment's average efficiency.
 II. THEORETICAL ADVANCES

A second generation of financial growth models incorporating
both endogenous growth and endogenous financial institutions
has emerged .
Financial intermediation is now modeled explicitly rather than
taken for granted or treated in simple deterministic terms, as it is in
the first-generation financial repression models.

Finance and financial institutions become relevant in a
world of positive information, transaction and monitoring costs.
If monitoring costs are high, a simple debt instrument may
dominate a more complicated state-contingent contract that
resembles equity.

The lender may reduce default risk by considering a potential
borrower's balance sheet and taking collateral, rationing the
borrower by providing less than requested or restricting the
maturity of the loan.
 II. THEORETICAL ADVANCES

Diamond-Dybvig model of financial inter-mediation
(Diamond and Dybvig, I983).
Individuals can choose between unproductive assets
(consumer goods-or commodity money) and an investment in
a firm.
The investment in a firm is illiquid because it takes time to
become productive.
However, the expected return from a two-period investment in
a firm is greater than the return from an inventory of
consumer goods or currency.
Uncertainty may force some individuals to liquidate or
abandon their investments in firms after only one period.
In such case, they would be worse off than had they held
solely an inventory of consumer goods or currency.
 II. THEORETICAL ADVANCES

Without banks, individuals must allocate their portfolios
between capital and currency to maximise expected utility.
 Although they know the probability of the event which
could make a productive investment worthless, their choice
will also be affected by their degree of risk aversion.
 Those with greater risk aversion will choose a higher
proportion of currency than those with less risk aversion.
Any productive investment bears some risk of becoming
worthless.
 II. THEORETICAL ADVANCES

Bencivenga and Smith (1991, I992), Greenwood and Smith
(I997), and Levine (I993) embed the Diamond-Dybvig
financial intermediation model in an overlapping-generations
model with production and capital accumulation.

With the introduction of banks, individuals can hold deposits
which banks then invest in currency and capital.
 By exploiting the law of large numbers, banks ensure that
they never have to liquidate capital prematurely.
 Banks can estimate deposit withdrawals which are
unpredictable individually but predictable for the economy
as a whole.
 Hence, banks avoid the uncertainty which leads to resource
misallocation by individuals.
 II. THEORETICAL ADVANCES
 By ensuring that capital is never wasted, financial
intermediation may produce higher capital/labour ratios and
higher rates of economic growth.
 By engaging in maturity intermediation, financial institutions
offer liquidity to savers and, at the same time, longer-term funds
to investors.

In so doing, they stimulate productive investment by
persuading savers to switch from unproductive investment in
tangible assets to productive investment in firms.

In contrary to Diamond-Dybvig model of financial inter-
mediation, Greenwood and Jovanovic (1990) stress the role of
financial intermediaries in pooling funds and acquiring
information that enables them to allocate capital to its highest
valued use, so raising the average return to capital.
 II. THEORETICAL ADVANCES

King and Levine (1993b) suggest that financial institutions
play a key role in evaluating prospective entrepreneurs and
financing the most promising ones: “Better financial systems
improve the probability of successful innovation and thereby
accelerate economic growth.”

Following Schumpeter (I9I2), they stress that 'financial
institutions play an active role in evaluating, managing, and
funding the entrepreneurial activity that leads to productivity
growth.

In all these models, financial repression in the form of
discriminatory taxes on financial intermediation reduces the
growth rate.
 II. THEORETICAL ADVANCES

Financial sector taxes are equivalent to taxes on innovative
activity, since they reduce the net returns that financial
intermediaries gain from financing successful entrepreneurs.

More generally, 'financial repression... reduces the services
provided by the financial system to savers, entrepreneurs, and
producers; it thereby impedes innovative activity and slows
economic growth' (King and Levine, 1993).
 III. PREREQUISITES

Several interest-rate liberalisation experiments have failed to
produce the results outlined above.

The basic problem lies in the perverse reaction to higher
interest rates by insolvent (or non-profit-motivated) economic
agents - governments, firms or individuals.
 By definition, an insolvent agent (one whose liabilities
exceed its assets) or 'distress borrower' is unable to repay
its loans. Hence, it is not deterred from borrowing by a
higher cost.
 It simply continues, if it can, to borrow whatever it needs to
finance its losses.
 Hence such agents exhibit loan demand functions that
respond positively to the interest rate.
 III. PREREQUISITES
 Pathologically high positive real interest rates, possibly
triggered by fiscal instability, indicate a poorly functioning
financial system.
 Inadequate prudential supervision and regulation enable
distress borrowing to crowd out borrowing for investment
purposes by solvent firms,
 so producing an epidemic effect.
 Funds continue to be supplied because of explicit or
implicit deposit insurance.
 The end result is financial and economic paralysis.
 III. PREREQUISITES

This international experience over the past 20 years indicates
that there are five prerequisites for successful financial
liberalisation
 (1) Adequate prudential regulation and supervision of
commercial banks, implying some minimal levels of
accounting and legal infrastructure.
 (2) A reasonable degree of price stability.
 III. PREREQUISITES
 (3) Fiscal discipline taking the form of a sustainable
government borrowing requirement that avoids inflationary
expansion of reserve money by the central bank either
through direct domestic borrowing by the government or
through the indirect effect of government borrowing that
produces surges of capital inflows requiring large
purchases of foreign exchange by the central bank to
prevent exchange rate appreciation.
 (4) Profit-maximising, competitive behavior by the
commercial banks.
 (5) A tax system that does not impose discriminatory
explicit or implicit taxes on financial intermediation.
 IV. THE STIGLITZ CONTROVERSY

In a series of papers (e.g. Stiglitz, I994), Stiglitz criticises
financial liberalisation on the grounds that financial markets
are prone to market failures.

He suggests that 'there exist forms of government intervention
that will not only make these markets function better but will
also improve the performance of the economy‘.

Specifically, he advocates government intervention to keep
interest rates below their market-equilibrium levels.
 IV. THE STIGLITZ CONTROVERSY

Given information imperfections, Stiglitz (I994, pp. 39-42)
argues that financial repression can improve the efficiency
with which capital is allocated.
 First, lowering interest rates improves the average quality
of the pool of loan applicants.
 Second, financial repression increases firm equity because
it lowers the cost of capital.
 Third, financial repression could be used in conjunction
with an alternative allocative mechanism such as export
performance to accelerate economic growth.
 Fourth, directed credit programmes can encourage lending
to sectors with high technological spillovers.
 IV. THE STIGLITZ CONTROVERSY

Criticism

Stiglitz has amazing faith in government. The government in
his papers is exemplary: disciplined, knowledgeable, long-
sighted, objective.
 It pursues economic objectives without deviating into the
many side alleys of patronage and sleaze.

The main doubt about actual governments pursuing his policy
is that, once they have an intellectual justification for
intervention, they will use it for purposes that would horrify
him.

According to Arestis and Demetriades (I997, p. 796): 'market
failure does not necessarily imply government success.’
 IV. THE STIGLITZ CONTROVERSY

Korea is often cited as an example of a country that has
prospered without full-blown financial liberalisation.

Korea is an outlier as economic policy-making in Korea
approximates Stiglitz's idealised world.

Unfortunately, one also has to agree that there are extremely
few other developing countries for which the same claim could
be upheld.
 IV. THE STIGLITZ CONTROVERSY

In a series of papers (e.g. Stiglitz, I994), Stiglitz criticises
financial liberalisation on the grounds that financial markets
are prone to market failures.

He suggests that 'there exist forms of government intervention
that will not only make these markets function better but will
also improve the performance of the economy‘.

Specifically, he advocates government intervention to keep
interest rates below their market-equilibrium levels.
 IV. THE STIGLITZ CONTROVERSY

Given information imperfections, Stiglitz (I994, pp. 39-42)
argues that financial repression can improve the efficiency
with which capital is allocated.
 First, lowering interest rates improves the average quality
of the pool of loan applicants.
 Second, financial repression increases firm equity because
it lowers the cost of capital.
 Third, financial repression could be used in conjunction
with an alternative allocative mechanism such as export
performance to accelerate economic growth.
 Fourth, directed credit programmes can encourage lending
to sectors with high technological spillovers.
 IV. THE STIGLITZ CONTROVERSY

Criticism

Stiglitz has amazing faith in government. The government in
his papers is exemplary: disciplined, knowledgeable, long-
sighted, objective.
 It pursues economic objectives without deviating into the
many side alleys of patronage and sleaze.

The main doubt about actual governments pursuing his policy
is that, once they have an intellectual justification for
intervention, they will use it for purposes.

According to Arestis and Demetriades (I997, p. 796): 'market
failure does not necessarily imply government success.’
 IV. THE STIGLITZ CONTROVERSY

Korea is often cited as an example of a country that has
prospered without full-blown financial liberalisation.

Korea is an outlier as economic policy-making in Korea
approximates Stiglitz's idealised world.

Unfortunately, one also has to agree that there are extremely
few other developing countries for which the same claim could
be upheld.
 V. EMPIRICAL EVIDENCE

According to economists of almost all persuasions, financial
conditions may affect the rate of economic growth in both the
short and medium runs.

Tobin's (1965) monetary growth model posits a negative
impact of a higher real return on money holdings in the
medium run but has nothing to say about the short run.

The McKinnon-Shaw school expects financial liberalisation
(institutional interest rates rising towards their competitive
free-market equilibrium levels) to exert a positive effect on the
rate of economic growth in both the short and medium runs.
 V. EMPIRICAL EVIDENCE

The neostructuralists predict a stagflationary (accelerating
inflation and lower growth) outcome from financial liberalisation
in the short run. In the medium run, there is the possibility that
the saving ratio will increase by enough to outweigh the negative
influence of portfolio adjustments.

In practice, neostructuralists, with the possible exception of
Buffie (i 984), view a dominant saving effect as unlikely.

A simple way of discriminating between the McKinnon-Shaw
school and others would be to examine episodes of financial
liberalisation and see whether or not these were accompanied by
higher or lower rates of economic growth.

In practice, however, most clear-cut cases of financial
liberalisation were accompanied by other economic reforms
(such as fiscal, international trade and foreign exchange reforms).
 V. EMPIRICAL EVIDENCE

In such cases, it is virtually impossible to isolate the effects of
financial components of the reform package.

Lanyi and Saracoglu (I983) implicitly address the causality
issue by dividing developing countries into three groups .
 Lanyi and Saracoglu give a value of 1 to countries with
positive real interest rates, 0 to countries with moderately
negative but 'not punitively negative' real interest rates, and
-1 to countries with severely negative real interest rates.

The cross-section regression indicates a positive and
significant relationship between the average rates of growth in
real gross domestic product (GDP) and the interest rate
dummy variable for the period 1971-8o.
 V. EMPIRICAL EVIDENCE

The World Bank (1989) uses the same methodology as Lanyi
and Saracoglu for a sample of 34 developing countries.
 First, it shows that economic growth in countries with
strongly negative real deposit rates (lower than - 10% on
average over the period 1974-85) was substantially lower
than growth in countries with positive real interest rates.
 Second, the World Bank also reports a regression showing
a positive and significant cross-section relationship
between average growth and average real interest rates over
the period 1965-85.
 V. EMPIRICAL EVIDENCE

For 53 countries over the period 1960-85, Roubini and Sala-i-
Martin (I992) also find that countries with real interest rates
less than -5% in the I970s experienced growth rates that
averaged 1.4 percentage points less than growth rates in
countries with positive real interest rates.

The global evidence suggests that Asian developing countries
may be more sensitive to real interest rate changes than other
groups of developing countries.

King and Levine (I993) show that each financial indicator is
positively and significantly correlated with each growth
indicator at the 99 % confidence level.
 V. EMPIRICAL EVIDENCE

De Gregorio and Guidotti (I995) claim that real interest rates
are not a good indicator of financial repression or distortion.

They suggest that the relationship between real interest rates
and economic growth might resemble an inverted U curve.
 'Very low (and negative) real interest rates tend to cause
financial disintermediation and hence tend to reduce
growth, as implied by the McKinnon-Shaw hypothesis....
On the other hand, very high real interest rates that do not
reflect improved efficiency of investment, but rather a lack
of credibility of economic policy or various forms of
country risk, are likely to result in a lower level of
investment as well as a concentration in excessively risky
projects‘.
 V. EMPIRICAL EVIDENCE

Fry finds also similar results. He concludes that the
relationship between the real interest rate and the growth is
non-linear and growth is maximised when the real interest rate
lies within the normal or non-pathological range of, say, -5 to
+ I5 %.

From the empirical work on financial repression by estimating
a simultaneous-equation system, Fry finds that saving and
investment ratios as well as export and output growth rates are
affected by financial conditions.

He also examine the effects of the excessively high real
interest rates that have been experienced after several financial
liberalisation experiments. These distorted financial conditions
appear to be just as debilitating as financial repression.
 V. EMPIRICAL EVIDENCE

Large positive and negative real interest rates exert the same,
presumably negative, effect on the saving ratio.

Further results indicate that financial distortions as measured
by the real interest rate squared and the black market exchange
rate premium reduce investment ratios and export growth. In
turn, lower investment ratios and export growth reduce output
growth rates.

The relationship between the real interest rate and the national
saving ratio resembles an inverted U. Both very low and very
high real interest rates reduce national saving mainly through
the effects of these interest rates on output growth.
V. EMPIRICAL EVIDENCE
Thanks