Tải bản đầy đủ

The Mckinnon-Shaw Hypothesis: Thirty Years on: A Review of Recent Developments in Financial Liberalization Theory


The Mckinnon-Shaw Hypothesis: Thirty Years on:

A Review of Recent Developments in Financial Liberalization Theory


Dr Firdu Gemech and Professor John Struthers
University of Paisley

The Mckinnon-Shaw Hypothesis, in its’ various forms, is now thirty years old. Over that
period literally hundreds of empirical studies have been completed examining the
hypothesis in many different contexts. Initially, the hypothesis focused on the effects of
so-called “Financial Repression” (low or negative real interest rates) on savings and
investment levels in developing countries. In more recent times, researchers have
extended the debate to consider other effects of financial repression on: economic
growth; financial crises and poverty (for example the effects of overvalued exchange

rates). Currently, significant research is being conducted on the potentially destabilizing
effects of financial liberalization (the converse of financial repression) on global
financial markets.

This paper attempts to survey the literature on the Mckinnon-Shaw Hypothesis and tries
to draw out some of the recurrent themes of this literature. The paper also highlights the
continuing relevance of the original hypothesis to on-going debates concerned with the
effects of financial liberalization.


Dr Firdu Gemech 00 44 141 848 3393;firdu.gemech@paisley.ac.uk
Professor John Struthers 00 44 141 848 3364; john.struthers@paisley.ac.uk
Division of Economics and Enterprise
Paisley Business School
University of Paisley
PA1 2 BE

* Paper presented at Development Studies Association (DSA) Annual Conference on
“Globalisation and Development”, Glasgow, Scotland, September 2003


An Overview of Recent Developments in Financial Liberalization Theory

The literature on Financial Liberalization policies in developing countries has a long
pedigree. This literature commenced with the seminal work of Mckinnon and Shaw in
1973 which focused on Financial Repression (see below) and the need for developing
economies to allow real interests rates (along with other financial indicators) to be
determined by market forces. Though originally focusing on interest rates, the Financial
Repression approach also incorporated the adverse affects of high reserve ratios and
government directed credit programmes, which together contributed to low savings,

credit rationing and low investment. In later years, the literature can be classified into:
First Generation approaches (represented by the work of Krugman (1979); Second
Generation approaches (Obstfeld, 1996) ; and Third Generation approaches again
represented by the work of Krugman (1998;1999) In this evolving literature, it is
possible to detect a clear lineage stemming from the original Mckinnon-Shaw
contribution, albeit one which represents an increasingly sophisticated theoretical and
empirical development of the original hypothesis. This paper will trace the development
of this body of thought as well as highlight possible further theoretical developments. It
will also highlight some of the future directions that this research might take and the
potential policy implications therein.

I. Theoretical Underpinnings

1. Liberalization as a catalyst for higher saving, (McKinnon-Shaw)

McKinnon (1973) and Shaw (1973), analysed the benefits of (if not eliminating)
Financial Repression, at least reducing its impact on the domestic financial system within
developing countries. Their analyses- (sometimes referred to as the Complementarity
Hypothesis)- concluded that alleviating financial restrictions in such countries (mainly by
allowing market forces to determine real interest rates) can exert a positive effect on
growth rates as interest rates rise toward their competitive market equilibrium. According
to this tradition, artificial ceilings on interest rates reduce savings, capital accumulation,
and discourage the efficient allocation of resources. Additionally, McKinnon pointed out
that Financial Repression can lead to dualism in which firms that have access to
subsidized funding will tend to choose relatively capital-intensive technologies; whereas
those not favored by policy will only be able to implement high-yield projects with short

Another effect of Financial Repression, to which the original hypothesis made only scant
reference, stemmed from the implicit “credit rationing” effect which results from the
Feast and Famine consequences of excessive government intervention in money and
credit markets in developing countries. Given that real interest rates are prevented from
adjusting to clear the market, other “non-market” forms of clearing have to take their
place. These can include various forms of “queuing” arrangements to “ration” the
available credit such as auctions, quantitative restrictions (for example quotas), as well as

different types of “bidding” systems which themselves may be open to nepotism or even
outright corrupt practices. In essence, these manifestations of Financial Repression mean
that not only is the quantity of savings (and investment) low, or at the very least
irregular; it also means that the level of activity which does occur is of poor quality. This
is really what the term Financial Repression entails. If the real interest rate is not allowed
to clear the money and credit markets, both the overall level as well as the quality of
savings and investment will be repressed. The quantity and the quality effects compound
each other. In a Feast and Famine environment, the typical borrower may borrow too
much (too little) and this very tendency will reinforce the Feast and Famine problem

The early hypotheses of McKinnon and Shaw assumed that liberalization, which would
be associated with higher real interest rates--as controls on these are lifted--would
stimulate saving. The underlying assumption is, of course, that saving is responsive to
interest rates. The higher saving rates would finance a higher level of investment, leading
to higher growth. Therefore, according to this view, we should expect to see higher
saving rates (as well as higher levels of investment and growth) following financial
liberalization. (see Appendix for the relevant equations of the McKinnon-Shaw model)

2. Liquidity constraints, credit channels, and financial liberalization ( Campbell -

As a further development of the Financial Repression literature Campbell and Mankiw
(1990) concluded that it is reasonable to assume that not all households have access to
credit markets, and hence, some households have no ability to smooth consumption over
time. Thus, for such liquidity-constrained households, consumption decisions are entirely
determined by current income. On theoretical grounds, it has been shown that a
relaxation of liquidity constraints will be associated with a consumption boom and a
decline in aggregate saving. More specifically, Campbell and Mankiw postulated that
there are two types of households in the economy: One type of household, λ , is liquidity
constrained and their consumption is entirely determined by the evolution of current
income, while the remaining type (1 − λ) , has free access to capital markets and can
smooth their consumption intertemporarily. Such a theoretical development led these
authors to challenge the implicit Mckinnon-Shaw assumptions that were based on a
homogenous household set in which it was assumed that all relevant households had free
access to capital markets within the domestic economy.

3. The role of subsistence consumption (Ostry and Reinhart)

This development was based on the Stone-Geary utility function where the intertemporal
elasticity of substitution (which determines the sensitivity of consumption to real interest
rates) is determined by permanent income and subsistence consumption. According to
this view, increases in real interest rates will affect consumption/saving decisions in
varying degrees. In countries where the representative household is close to subsistence
consumption, consumption(and saving) will not be sensitive to changes in the real rate of
interest. Only in wealthier countries would consumption decline (and saving increase)

following an increase in real interest rates. Hence, in this analysis the magnitude of the
increase in saving following the higher real interest rates associated with financial
liberalization will depend on the level of income (which was used as a proxy for how
close are actual consumption levels to subsistence levels).

II. The Empirical Literature

The broad empirical literature varies greatly in terms of both empirical approach and
country coverage. The McKinnon-Shaw hypothesis literally spawned hundreds of such
empirical studies across many different contexts, countries and time periods. The
empirical literature, in general, suggests that the relationship between saving rates and
real interest rates is at best ambiguous. Yet surprisingly, and somewhat perversely,
financial liberalization also has a mixed track record regarding saving rates. Indeed, in
the studies reviewed here, in most of the cases liberalization appears to lead to a decline
in the saving rate.

a) Saving and real interest rates

There is little consensus in the empirical literature on the interaction between saving and
the real rate of interest. Some researchers have been unable to detect much of an effect of
changes in real interest rates on domestic saving in developing countries. For example,
Giovannini (1985), who examined this issue for eighteen developing countries, concludes
that for the majority of cases, the response of consumption growth to changes in the real
rate of interest is insignificantly different from zero and that one should therefore expect
negligible responses of aggregate saving to the real rate of interest. In a model with a
single consumption good, Ostry and Reinhart (1992) confirm these findings. Only when a
disaggregated commodity structure that allows for traded and nontraded goods is
assumed, do these authors find higher and statistically significant estimates of the
sensitivity of consumption to interest rates. In a similar vein, Ogaki, Ostry, and Reinhart
(1996), present evidence that consumption in developing countries may be more related
to subsistence considerations -particularly in the case of extremely low-income countries
- than to intertemporal consumption smoothing. The rationale here is that if households
must first achieve a subsistence consumption level-allowing intertemporal considerations
to guide their decisions only for that portion of their budget left after subsistence has been
satisfied - then the intertemporal elasticity of substitution and the interest-rate sensitivity
of private saving will be close to zero for countries at or near subsistence consumption
levels, but will rise thereafter.

b) Liberalization and saving

1. Bandiera, Caprio, Honohan, and Schiantarelli (2000), construct an index of financial
liberalization on the basis of eight different components: interest rates; reserve
requirements; directed credit; bank ownership; prudential regulation; securities markets
deregulation; and capital account liberalization. Their data spans from 1970-94 for Chile,
Ghana, Indonesia, Korea, Malaysia, Mexico, Turkey and Zimbabwe. Among the key
findings of the estimation of their benchmark model is that, there is no evidence of any

positive effect of the real interest rate on saving. Indeed in most cases the relationship is
negative, and significantly so in the cases of Ghana and Indonesia. Furthermore, the
effects of the financial liberalization index on saving are mixed: negative and significant
in Korea and Mexico, positive and significant in Turkey and Ghana. The long run impact
of liberalization is, however, sizeable. Corresponding to the realized change in the index,
the estimated model indicates a permanent decline in the saving rate of 12% and 6% in
Korea and Mexico, and a rise of 13% and 6% in Turkey and Ghana.

Based on an estimate of augmented Euler equations (a la Campbell-Mankiw) , Bandiera
et al present some evidence of the presence of liquidity constraints. It was not possible,
however, to confirm whether financial liberalization removes these constraints. The Euler
equation results may suggest, at best, that financial liberalization has had little impact on
the amount of credit available to consumers through the formal financial sector. The
general conclusion that emerges from this study is that there is no systematic and reliable
real interest rate effect on saving; whilst the effects of liberalization have a mixed record.

2. Bayoumi (1993), examined the effects of financial deregulation on personal saving.
Within an overlapping generations framework, the author argues that deregulation
produces an exogenous short-run fall in saving, some of which is recouped over time.
Also, deregulation increases the sensitivity of saving to wealth, current income, real
interest rates and demographic factors. Using data on the eleven standard regions of the
United Kingdom, the author finds that household saving showed an exogenous decline
associated with financial innovation — saving also became more sensitive to wealth, real
interest rates and current income; though the results imply that much of the decline in
savings in the 1980's was caused by the rise in wealth. Financial deregulation also played
a significant direct role. In particular, the author concludes that an autonomous fall of
2.25% in the personal saving rate may be attributed to deregulation alone.

3. Jappelli and Pagano (1994), investigate the role of capital market imperfections on
aggregate saving and growth. The analytical framework of their paper is a simple
overlapping generations model, within the context of which it is shown that liquidity
constraints on households (but not on firms) can raise the saving rate; strengthen the
effect of growth and may increase welfare. Using a panel of OECD countries for the 1960
to 1987 period, the authors conclude that financial deregulation in the 1980's has
contributed to the decline in national saving and growth rates in the OECD countries and
expressed concern regarding the welfare implications of further liberalization within the
European Union.

4. Koskela, Loikkanen and Viren (1992), describe the institutional aspects of housing
markets and analyze the evolution of prices of owner-occupied housing and their
interaction with the household saving ratio in Finland in the 1970's and 1980's. The
volatility of house prices in relation to income can be traced to a large extent to major
changes in financial market conditions. The evidence they present suggests that financial
market conditions - as measured by the household’s indebtedness rate, the after tax rate
of return on housing, and the “thinness” of rental markets - have all had a positive effect
on housing prices. Yet, household saving was affected negatively by the rate of change of

real house prices, and positively by the after tax nominal interest rate. Taken together,
their findings imply that financial conditions, and the liberalization of the mid-1980's in
particular, contributed to the decline in the household saving ratio in these countries.

5. Loayza, Schmidt-Hebbel, and Serven’s (2000), produced results, which suggest that
the direct effects of financial liberalization are detrimental to private saving rates. The
real interest rate has a negative impact on the private saving rate. Its income effect
probably outweighs the sum of its substitution and human wealth effects. A 1% increase
in the real interest rate reduces the private saving rate by 0.25% in the short run.

The indicator of financial depth (M2/GNP) has a small and statistically insignificant
impact on the private saving rate. The flow of private domestic credit relative to income
has a negative and significant coefficient; relaxing credit constraints reduces the private
saving rate. When the flow of private credit rises by 1%, the private saving rate declines
by 0.32% on impact. The authors suggest that though they do not find direct positive
effects of financial liberalization on the saving rate, if financial reform has a positive
impact on growth, it has a potentially important indirect positive effect on the saving rate.

6. Reinhart and Tokatlidis (2001), in a study of 50 countries (14 developed and 36
developing) report that financial liberalization appears to deliver: higher real interest rates
(reflecting the allocation of capital toward more productive, higher return projects.);
lower investment, but not lower growth (possibly owing to a shift to more productive
uses of financial resources); a higher level of foreign direct investment; and high gross
capital flows. Liberalization appears to deliver financial deepening, as measured by the
credit and monetary aggregates--but, again, low income countries do not appear to show
clear signs of such a benefit. As regards saving, the picture is very mixed. In some
regions, saving increased following financial sector reforms; but in the majority of cases
saving declined following the reforms. Indeed, it would appear that what financial
liberalization delivers is greater access to international capital markets, although this
appears to be uneven across regions and income groups.

III. First Generation Models: The Financial Crises Literature

Financial crises are attributed to rapid reversals in international capital flows prompted
chiefly by changes in international investment conditions. Flow reversals are likely to
trigger sudden current account adjustments, and subsequently currency and banking
crises. A second generation of currency crises models (Krugman, 1979) explained the
collapse of exchange rate regimes on the grounds that weak fundamentals lead foreign
investors to pull resources out of the country, and as a consequence the depletion of
foreign reserves (see Appendix for a summary of the key equations of the Krugman

A Second Genration of models (Obstfeld, 1996) suggests that currency crises may also
occur despite sound fundamentals, as a result of self-fulfilling expectations, speculative
attacks and changes in market sentiments. More recently, there have been important
developments in this area of research, most of them using the “event analysis

methodology”. A summary of two studies is presented below (see Appendix for relevant

The Third Generation Models (Krugman, 1998, 1999) attempt to stylize the causal
mechanics underpinning the 1997 Asian currency crisis as the First and Second
Generation models did not fully explain these phenomena. One version of the Third-
Generation model attributes the crisis to implicit guarantees offered by domestic banks in
developing countries leading to a massive influx of short-term capital which turns out to
be unsustainable. This invariably results in an asset price bubble that is destined to burst
and reverse the capital inflows. Another version identifies the existence of ‘Fragile
Financial Institutions’ as the cause of the build up of unhedged short-term borrowing
denominated in foreign currency. A sudden change in market sentiment causes panic and
investor responses which bring about a reversal in these capital flows. This transforms an
illiquid asset into insolvency and ultimately a currency peg collapse (See Appendix for
relevant equations)

In a key empirical study Kaminsky and Schmukler (2001) examined the short-run and
long-run effects of financial liberalization on capital markets by constructing a
comprehensive chronology of financial liberalization in 28 developed and emerging
economies since 1973. They used three measures of financial liberalization: (a) capital
account liberalization (capital mobility); (b) domestic financial system liberalization
(regulations on deposit interest rates, lending interest rates, allocation of credit, and
foreign currency deposits) and, (c) stock markets liberalization (evolution of regulations
on the acquisition of shares in the domestic stock market by foreigners, repatriation of
capital, and repatriation of interest and dividends etc). The authors arrived at the
following broad conclusions:

1. While liberalization has been an uninterrupted process in most developed
markets, it has been characterized by reversals in emerging markets, in which
capital controls and restrictions are at times reintroduced. They also found that the
pattern of liberalization varies across regions, with developed countries
liberalizing first their stock markets and developing economies opening first their
domestic financial sector.

2. Although liberalization leads to excessive financial booms and busts in the short-
run, these booms and busts have not intensified in the long run. In fact, despite the
claim that financial integration leads to volatile capital markets around the world,
stock market cycles become less pronounced after liberalization. The short-run
effects of liberalization vary across developed and emerging markets. The
evidence from emerging markets reveals larger booms and crashes in the
immediate aftermath of liberalization. In contrast, the evidence from developed
markets, with larger bull markets but less pronounced bear markets in the
aftermath of deregulation, supports the view that liberalization is beneficial even
in the short run.

3. To explain the contrasting short and long-run effects of financial liberalization,
the authors collected information on the quality of institutions as well as data on
the laws governing the functioning of the financial system. The evidence suggests
that as the quality of institutions improves, financial cycles become less
pronounced. Perhaps due to lack of correct incentives, countries do not tend to
improve their financial systems before liberalization, disregarding the typical
policy prescriptions.

Baldicci, de Mello and Inchauste (2002) in a new approach considered the impact of
financial crises (defined as large scale nominal currency depreciation and successful
speculative attack) on the incidence of poverty and on the distribution of income in
Mexico. The authors identified the following channels through which financial crises
affect poverty and income distribution:

• A slowdown in economic activity following a crisis leading to a fall in
earnings of both the formal and informal-sector workers. Reduced working
hours and real wage cuts adversely affect the earnings of the poor.

• Relative prices change after a currency depreciation leading to an increase in
the price of imported food (and domestic food prices). This in turn adversely
affects poor individuals and households that are net consumers of food.

• Spending cuts (fiscal retrenchment) affect the volume of publicly provided
crucial social services and limits the access of the poor to these services at a
time when their incomes are declining.

• Changes in asset prices (wealth effects) following changes in interest rates and
real estate prices affect the wealth of the better off.

IV. Arguments for and against financial liberalisation policies

(a) For

In what appears to be a parallel world, many authors still praise the advantages of
liberalization. It is claimed that financial liberalization helps to improve the functioning
of financial systems, increasing the availability of funds and allowing cross-country risk

• Obstfeld (1998) argues that international capital markets can channel world
savings to their most productive uses irrespective of location.

• Stulz (1999) and Mishkin (2001) claim that financial liberalization promotes
transparency and accountability, reducing adverse selection and moral hazard
while alleviating liquidity problems in financial markets. These authors argue that

international capital markets help to discipline policy makers, who might be
tempted to exploit an otherwise captive domestic capital market.

The benefits of financial liberalization can therefore be grouped into increased access to
domestic and international capital markets, and increased efficiency of capital allocation.

(b) Against

• Critics of financial liberalization policies have argued that the efficient markets
paradigm is fundamentally misleading when applied to capital flows. In the
theory of the second best, removing one distortion need not be welfare enhancing
when other distortions are present. If the capital account is liberalized while
import competing industries are still protected, for example, or if there is a
downwardly inflexible real wage, capital may flow into sectors in which the
country has a comparative disadvantage, implying a reduction in welfare.

• If information asymmetries are endemic to financial markets and transactions, in
particular in countries with poor corporate governance and low legal protections,
there is no reason to think that financial liberalization, either domestic or
international, will be welfare improving (Stiglitz (2000)). Moreover, in countries
where the capacity to honor contracts and to assemble information relevant to
financial transactions is least advanced, there can be no presumption that capital
will flow into uses where its marginal product exceeds its opportunity cost.

• Stiglitz (1994) argues in favor of certain forms of financial repression. He claims
that repression can have several positive effects such as: improving the average
quality of the pool of loan applicants by lowering interest rates; increasing firm
equity by lowering the price of capital; and accelerating the rate of growth if
credit is targeted towards profitable sectors such as exporters or sectors with high
technological spillovers. However, these claims can be doubtful given that they
increase the power of bureaucrats, who can be less capable than imperfect
markets, to allocate financial resources.

• Focusing on the development of the domestic financial sector, capital account
liberalization that allows firms to list abroad has been identified as a factor
leading to market fragmentation, and can tend to reduce liquidity in the domestic
market thereby inhibiting its development.

• Finally, liberalization has also been linked to macroeconomic instability. The
financial reforms carried out in several Latin American countries during the
1970s, aimed at ending financial repression, often led to financial crises
characterized by widespread bankruptcies, massive government interventions,
nationalization of private institutions and low domestic saving (Diaz-Alejandro
(1985). Demirguc-Kunt and Detriagiache (1998), however have shown that the
likelihood of a crisis following liberalization decreases with the level of

institutional development in the country. In this sense, the arguments that Stiglitz
(1994) makes in favor of government intervention in financial markets in the
form of prudential regulation and supervision are convincing. The main argument
is that the government is, de facto, the insurer of the financial systems, and hence
a financial collapse can have significant fiscal repercussions.

V. Measurement problems

(a) Proxies for repression (and liberalization)

• Several empirical studies have tried to address the extent to which financial
liberalization affects growth. Researchers have followed two distinct empirical
approaches. One approach is to proxy financial liberalization with outcome
variables; the other approach focuses on explicit policy measures. Regarding
outcome variables, several measures have been suggested to proxy financial
repression. Early empirical literature focused on the value of real interest rates as
an indicator of repression.
The presumption was that countries with negative real
interest rates
were financially repressed, while those with positive ones were
liberalized. In short, it was found that countries with negative real interest rates
exhibit lower growth rates compared with those with positive real interest rates.
However, De Gregorio and Guidotti (1993) claim that real interest rates are not a
good indicator of financial repression, and that a better indicator of repression, or
lack thereof, is the ratio of credit to the private sector to GDP or a similar
measure of financial development (actually financial widening rather than
financial deepening).

• Similar concerns were expressed regarding the measurement of financial
deepening. Prior to international financial liberalization, broad money offers a
good indication of the banking system’s scope for credit expansion, since
domestic bank deposits are the main source of finance for bank lending. When
capital controls are abolished, however, capital inflows — in the form of deposits
made by foreign residents in domestic banks — add to the funds banks have
available for credit expansion but do not increase broad money (since they are
excluded from it by definition). Money based measures of financial deepening
may therefore be misleading when capital inflows are important.

• Capital flows are not the only reason why money and credit-based measures of
financial deepening may diverge. In general, government borrowing from the
banking system will, for a given level of broad money, reduce the amount of
credit available to the domestic private sector. If private sector activity is more
productive than government expenditure, then this crowding out of private
borrowing may have strong negative repercussions for economic performance that
would not, however, be reflected in the conventional measure of financial

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Tải bản đầy đủ ngay