Financial Liberalization and Investment

Kanhaya Gupta, Robert Lensink

• Publisher: Taylor & Francis
• Year: 1996
• eBook ISBN: 9781134772629
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

For two decades thinking on economic policy has been dominated by the idea of economic liberalization in general and financial deregulation in particular. This field has become both extensive and controversial, yet there is no single book which treats financial deregulation in a complete and coherent manner. This book rectifies the shortfall by foc

1
INTRODUCTION

The purpose of this brief introduction is to explain the motivation for our work and provide a brief outline of the chapters to follow.

Inspired by the influential works of McKinnon (1973) and Shaw (1973), and by the requirements of the IMF/World Bank sponsored structural adjustment programs, the effects of financial liberalization on investment in developing countries have drawn much attention. The major thrust of this literature has been to understand the mechanisms by which interest rate deregulation and the elimination of other forms of financial repression—for example, changes in reserve requirements—affect savings and investment. Broadly speaking, we can distinguish two sets of approaches in the literature: those based on non-optimizing models and those which involve optimizing frameworks. The former includes works which essentially build on the model suggested by McKinnon (1973). This body of literature is extensively summarized in Fry (1988) and Gibson and Tsakalotos (1994). The optimizing models draw their inspiration from the pioneering works of Romer (1986) and Lucas (1988) on endogenous growth models. This part of the literature is extensively surveyed by Berthelemy and Varoudakis (1994), De la Fuente and Marin (1993), Pagano (1993) and Schiantarelli et al. (1994b), but see also the appendix to this chapter.

Our work is in the tradition of the non-optimizing models. Instead of giving a survey of the existing literature, our approach is to discuss the relevant literature within the context of our model in each chapter. Consequently, in this introduction we only explain the motivation for our work and then give a brief review of the chapters to follow. In addition to the literature on the effects of financial liberalization, there is also much work on the role of informal credit markets in meeting the needs of households and firms. Other areas of important concern have been the role of foreign aid in the growth of developing countries, the implications of the credit needs of the governments to finance budget deficits, the implications of financial repression for raising government revenues, and the importance of wealth effects. However, in virtually no work on the effects of financial liberalization that we are aware of are all these different strands of thought brought together in an integrated framework. Of course, there are efforts which try to combine one or more of these issues. For example, Van Wijnbergen (1983a, 1983b), and others in the ‘neostructuralist’ school, examine the role of informal credit markets, but pay very little attention to the role of foreign aid, wealth effects, crowding out of private credit by the needs of the government to finance budget deficits and so on. Morisset (1993) examines the effects of crowding out of private credit caused by the government’s budgetary needs and a shift in the public’s portfolio caused by interest rate deregulation. But he treats government budgets as being exogenously given, ignores the role of foreign aid and informal credit markets and treats savings as being exogenously determined, thus eliminating all indirect effects of interest rate deregulation via changes in wealth.

The aim of this study is to provide a systematic analysis of the whole range of questions which have been raised in the literature with respect to financial liberalization. More specifically, our study is distinguished by the following features:

1. It analyzes three aspects of financial liberalization. These three aspects are the effects of interest rate deregulation on the quantity of investment, the effects on the allocative efficiency of investment and the effects of an improvement in banking efficiency on investment. It is only the first of these effects that has received considerable attention in the literature. This is curious since the other two are mentioned almost as often as the first one.
2. Its second distinction is that it takes into account many mechanisms by which a financial liberalization may affect investment which have hitherto not been incorporated in the existing models. In particular the role of wealth effects, effects of portfolio changes and crowding out by government budgetary behavior are allowed simultaneously. We accomplish this task by using an integrated model of portfolio selection and consumption—saving, which has not been done before. Similarly we endogenize government budget deficit, thus explicitly analyzing the effects of interest rate deregulation on government expenditures, revenues and interest payment. At the same time we also incorporate informal credit markets.
3. While there is a good deal of literature on foreign aid and economic growth in developing countries, as well as the conditionally of IMF/World Bank on the granting of concessional loans on the adoption of financial deregulation policies, there is, to our knowledge, no formal treatment in which the effects of financial deregulation and foreign aid on investment are modelled simultaneously. Our study tries to fill this gap by considering both sides of the structural adjustment program simultaneously.
4. While the above exercises fill many of the gaps in the existing literature and open up some new avenues, the fact still remains that our ability to derive analytical results from a model which combines all of the aspects mentioned above remains limited. One of the options is to formulate a general model and carry out simulations. This is precisely what we do. The model is based on the foundations laid out in 1, 2 and 3 above. Simulations are carried out under a variety of assumptions about the behavior of the economic agents involved in the model. These simulations help us shed light on a number of controversial issues in the literature on economic liberalization in general and financial liberalization in particular as discussed in the chapter outlines which follow.

Chapter 2 constitutes the foundation for the entire work. Assuming a given government budget deficit and a demand-determined world, we specify our model and explain how it differs from the existing works. It distinguishes three sectors: a consolidated private sector, a consolidated banking sector and a government sector. Unlike the existing works, we assume that the private sector’s decisions about portfolio selection and consumption—saving are integrated, in which wealth plays an important role and where the private sector is credit constrained. This part of the model draws on the work by Owen (1981) and others on portfolio selection modeling. In this chapter we then go on to derive the conditions under which an interest rate deregulation can lead to an increase in the supply of bank credit to the private sector in the presence of crowding out effects of government’s borrowing requirements from the banks, the crowding out being induced by a reallocation of the private sector’s portfolio consequent upon interest rate changes on deposits. The chapter then goes on to derive the conditions under which McKinnon’s well-known ‘complementarity’ hypothesis holds. It is shown that the conditions are far more stringent than is recognized in the literature so far. It is also shown that the outcome is greatly affected by our assumption that the private sector’s portfolio decision and consumption—saving decision are integrated.

The model in Chapter 2 assumed that government budget deficit was exogenously given and that it was entirely financed by domestic resources. In Chapter 3 we introduce two innovations compared to the rest of the literature in the field. First we endogenize the deficit. This we do by drawing on the literature on the fungibility of foreign aid. For this purpose we use an optimizing model of the government’s fiscal behavior. The second innovation is that we introduce foreign aid. While there is voluminous literature on the effects of foreign aid on the economies of developing countries, there is virtually none which tries to assess the implications of such aid for the success or failure of interest rate deregulation policies. This omission is rather curious in that in the structural adjustment programs of the IMF/World Bank a condition for the provision of concessional loans/ aid is often conditional on the adoption of reforms in the financial sector including interest rate deregulation. We try to model such an interaction using the basic model specified in Chapter 2. We show that an interest rate deregulation without foreign aid may lead to a sub-optimal situation. In the case where financial liberalization does not lead to a fall in government investment, probably the crowding out of the supply of credit to the private sector by the government will be large. Otherwise, in the case where the government tries to avoid these crowding out effects, a decline in government investment will probably be the result. We also show that foreign aid might neutralize the crowding out effects of private credit. However, it is not a guarantee for a halt in the decline in government investment. This depends on the pattern of usage of such aid, i.e. on the degree of ‘fungibility’ of such aid.

In the models discussed previously we have abstracted from informal credit markets. However, recently many authors have pointed out the role of the informal financial sector in financing the credit needs of the private sector. In the literature two views with respect to the financial intermediation of the informal sector are distinguished. Traditionally, the informal lenders are associated with monopolistic moneylenders who exploit the poor charging usurious interest rates and are stereotyped as being exploitative. In this view, informal financial intermediation is very inefficient and costs are high so that a substantial part of total savings goes to informal banks as a reward for the services supplied. However, many recent studies argue that the informal financial market is a highly efficient competitive market with developed linkages with the official markets. In Chapter 4 we incorporate the informal financial market in the Chapter 2 model and assess the impact of interest rate deregulation on private investment under different assumptions about the workings of the informal financial market. We show that most of the existing works are special cases of our more general model and, once again, that the way we model the private sector’s behavior about portfolio selection and consumption—saving has important bearing on the outcomes. We also go on to derive the conditions under which informal credit markets are helpful to the effectiveness of interest rate deregulation policies.

The analysis so far has dealt with the scale effects of financial deregulation, namely the effects on the quantity of investment. However, the literature also emphasizes another channel through which such deregulation may enhance growth. This has to do with the allocative efficiency effect. The idea is that higher interest rates, by inducing the selection of projects with higher rates of return, will raise the average productivity of investment and hence growth, even if the effect on savings and thus total investment was negligible.

The main difficulty in dealing with this issue is how to measure improvements in allocative efficiency which are induced by financial deregulation. Using a two-sector model, Galbis (1977) tried to analyze this problem. However as we explain his model is just an assertion of his results and that it cannot be used to answer the question being considered. Besides he ignores the informal credit market which apparently plays a crucial role in this debate. In Chapter 5 we develop a two sector version of the model in Chapter 4. Then we suggest how to assess the effects of interest rate deregulation on allocative efficiency. Finally, we show the conditions under which allocative efficiency may improve as a consequence of interest rate deregulation. It is shown that such deregulation does not always guarantee an improvement in overall productivity. The outcome depends on how the two sectors react towards the formal and the informal banking sectors as lenders and as borrowers which ultimately affects the portfolio behavior of the two sectors and consequently allocative efficiency.

In the previous chapters we have paid attention only to the consequences of changes in interest rates on deposits. Another important issue with respect to the effects of financial deregulation relates to the effects of changes in banking efficiency on private investment. It has been argued that a high spread between deposit and lending rates reflects high cost of financial intermediation and, therefore, lower banking efficiency. Although this issue has been raised in the literature, there is little formal analysis of it. In Chapter 6 we make an attempt to model the effects of improvement in banking efficiency on private investment. Again, this is done by using a variant of the model in Chapter 2. More specifically, we approach the concept of banking efficiency by distinguishing between effective and nominal rates of return on deposits and cost of loans. It is shown that a financial liberalization in the conventional way, i.e. an interest rate deregulation, leads to an increase in both the effective deposit and lending rates, whereas an improvement in banking efficiency leads to an increase in the effective deposit rate and a decrease in the effective lending rate. Further, it is shown that while the effects of an improvement in banking efficiency cannot be determined a priori, under several circumstances the possible negative effects of interest rate deregulation on private investment may be mitigated or even overcompensated by the positive effects of an improvement in banking efficiency.

There are three major limitations of the analysis of Chapters 2–6 (particularly Chapters 2, 3 and 4). These are the partial nature of the model used. Partial in the sense that in each case we ignored some of the features. For example, in Chapter 2 we treated budget deficits as being exogenous and completely ignored informal credit markets and foreign aid. In Chapter 3, while we endogenized the deficit and incorporated foreign aid, we still excluded the informal credit markets. In Chapter 4, we rectified this shortcoming but only at the cost of the assumptions about the deficit and aid in Chapter 2. These respective simplifications were introduced so that we could derive some analytical results to highlight the significance of the attribute under consideration. But obviously a more appealing approach would be to take all of these factors into account simultaneously. The second shortcoming is the assumption about inflation. We have assumed that it was given. This is not an unusual assumption in demand-determined models. But still it is unsatisfactory.

In Chapter 7 we formulate a simulation model which rectifies all of these shortcomings. But the basic structure of the model is derived from that used in Chapters 2, 3 and 4. Thus, this model is based on an integrated model of portfolio selection and the consumption—saving decision of the private sector with appropriate adding-up restrictions, with endogenous budget deficits, explicitly recognizing the role of foreign aid including the implications of the ‘fungibility’ of foreign aid as well as the role of informal credit markets without imposing any a priori restrictions on the degree of its intermediation capacity. Further, inflation is endogenized by explicitly treating the aggregate supply side. Finally, we explicitly introduce an external sector, albeit in a somewhat rudimentary form.

This model is simulated to ask the same question: does interest rate deregulation affect private investment? In order to answer this question we can simulate the model in a variety of ways. But we believe that the most illuminating way for our purpose is to concentrate on one sector at a time and then do what amounts to a sensitivity analysis compared to a baseline simulation. While the possibilities are far too numerous to be outlined or examined, we concentrate on those which highlight a number of contentious issues in this area. In brief, we simulate the effects of interest rate deregulation under seven alternative assumptions about the model. These are: (1) no substitution between deposits and capital: (2) only investment or only consumption is credit constrained; (3) consumption and therefore saving is not affected by real interest rates—the standard assumption in much of the literature; (4) higher wealth effects on investment; (5) higher reserve requirements in the formal banking sector; (6) higher foreign aid; and (7) complementarity between government and private investment. As we shall see these simulations allow us to shed light on many issues. For example, we get some idea about the importance of liquidity constraint. Recently Jappelli and Pagano (1994) have shown that if financial liberalization can eliminate or relax liquidity constraints for the household, then it may lead to a reduction in savings and thus in growth. Similarly, the simulations have something to say about the sequencing of liberalization reforms, a topic that McKinnon (1991) has much to say about, namely, should budgetary reforms precede financial reforms?

In Chapter 8 we add to a relatively new type of literature. So far we have concentrated on the effects of financial deregulation on the real economy. In this literature the focus is on public finance. It is beginning to be argued that financial repression can be, and is being, used by governments to extract more resources from the economy (see Sussman (1991) and Giovannini and De Melo (1993)). We deal with this issue by using the simulation model of the previous chapter. We assess the cost of financial repression in terms of its effects on inflation, private and public investment and the pattern of public consumption. The indicators of financial repression are those which are meant to raise additional revenues for the government. The measures of financial repression studied are: (1) the government levying a tax on interest income from government bonds held by the non-bank private sector; (2) the government borrowing from the banks at rates lower than those charged to the private sector; and (3) government borrowing from the banking sector by increased reserve requirements of the formal banking sector. While this approach does not provide a true cost-benefit analysis of the government’s attempts to raise revenues through financial repression, nevertheless we believe that it sheds significant light on the issues involved.

Chapter 9 sums up the main findings and limitations.

APPENDIX 1

There is by now a sizable literature on the optimizing models in this area. For example: Bencivenga and Smith (1991, 1992, 1993), Berthelemy and Varoudakis (1994), Bernanke and Gertler (1989, 1990), De Gregorio (1992, 1993, 1994), De la Fuente and Marin (1993), Gertler and Rose (1994), Greenwald and Stiglitz (1989), Greenwood and Jovanovic (1990), Greenwood and Smith (1993), Gylfason (1993), Jappelli and Pagano (1994), King and Levine (1993a, 1993b, 1993c), Levine (1991, 1992), Roubini and SalaiMartin (1991, 1992a, 1992b), Saint-Paul (1992a, 1992b), Sussman (1991, 1993), Sussman and Zeira (1993), Varoudakis (1992), among others.

Not all of these papers use the endogenous growth models as their starting point nor do they all suggest the same mechanisms by which financial growth affects real growth, but nevertheless we can get a feel for what they do by considering the framework used in Pagano’s (1993) useful survey of this area. Using the ‘AK’ model, he shows that

g = Aϕs – δ,(A.1)

where g is the real growth rate, A the social marginal productivity of capital, ϕ the proportion of savings not lost in the process of financial intermediation, s the private rate of saving and δ the depreciation rate. Assuming that δ is constant, financial intermediation or financial factors can affect real growth by affecting any of the other three parameters. Many of the papers cited above try to show which one of the three parameters and how it can be influenced by financial intermediation. For details the reader is asked to refer to the other survey papers or the original articles. We can only give a flavour here by referring to some of the recent works which are also used in our work above.

First note that anything that can affect banking efficiency, as discussed in our Chapter 6, can affect ϕ. Some discussion of what those factors might be can be found in Sussman (1993).

The role of financial intermediation in affecting A essentially relates to the allocational efficiency issue as discussed in our Chapter 5. In this literature, this issue has been discussed by Bencivenga and Smith (1991), Greenwood and Jovanovic (1990), Levine (1991) and Saint-Paul (1992a and 1992b), among others.

There is much literature on how financial factors may affect s. But here we mention two contributions which have a bearing on our model too as we point out. These are the contributions by De Gregorio (1994) and Jappelli and Pagano (1994). Both papers use the OLG model. Jappelli and Pagano show that financial liberalization, by relaxing or eliminating liquidity constraints from the household, may depress saving, thus affecting growth unfavourably. De Gregorio, however, shows that if the borrowings by the household are also for investment in human capital, then not relaxing the liquidity constraint can harm growth. In other words, once we consider both effects, the outcome is ambiguous.

Some of the papers also address the opposite question, namely, what causes financial growth. Here an interesting contribution is by Sussman (1993). Finally, some of the papers also present models in which financial and real growth take place simultaneously. Here three interesting contributions are: De la Fuente and Marin (1993), Greenwood and Jovanovic (1990) and Saint-Paul (1992a and 1992b).

2
THE BASE MODEL

In this chapter, as pointed out in Chapter 1, we develop the base model which serves as the main framework for the subsequent chapters. The distinctive feature of the model developed here is that, unlike the rest of th...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Figures
• Tables
• Acknowledgements
• 1 Introduction
• 2 The Base Model
• 3 Role of Foreign Aid and the Government
• 4 The Role of Informal Financial Markets
• 5 Allocative Efficiency and Financial Deregulation
• 6 Banking Efficiency and Private Investment
• 7 Some Simulation Results
• 8 Financial Repression and Fiscal Policy
• 9 Summing Up
• Notes
• Bibliography