Export instability and economic growth in asian developing countries

ABSTRACT EXPORT INSTABILITY AND ECONOMIC GROWTH IN ASIAN DEVELOPING COUNTRIES TONG MINH TUAN HCM Economic University · September 2002 In this study, we look at the relationship betw

VIETNAMESE- DUTCH PROJECT FOR M.A PROGRAMME IN

DEVELOPMENT ECONOMICS

EXPORT INSTABILITY AND ECONOMIC GROWTH

IN ASIAN DEVELOPING COUNTRIES

TONG MINH TUAN September 2003

Tong Minh Tuan

Date: September 91h, 2003

ii

ACKNOWLEDGEMENT

This thesis IS done under the Vietnamese-Dutch Project for Development Economics The author would like to thank the Netherlands for her Aid and Scholarship

A lot of praise goes to Dr Gabrielle Berman and Dr Y oudi Schipper for their comments for my thesis I also would like to thank Dr Haroon Akram-Lodhi and

Dr Karel Jansen for their worthy academic teaching and methodology in doing thesis

Many thanks are also given to all project teachers and staff in HoChiMinh city, especially Mr Tran Vo Hung Son, Michael Palmer, Ms Nguyet, Ms Chi and Ms Dieu for their helping during the time I studied at the project

I am deeply indebted to my supervisor, Dr Nguyen Khac Minh, who gave me instructions and comments during the writing of this thesis I also thank to Dr Vu Thieu for his help in providing me a good source of references from Vietnamese-Dutch Project library in Hanoi

I have benefited from the discussions with many people and friends, especially my best friends Mr Nguyen Xuan Nam, Dang Tai An Trang, who gave me many supports in doing the thesis I also would like to thank to World Bank's staffs, who helped me in searching and collecting necessary documents in WB library in Hanoi

Finally, I wish to express my gratitude and my love to my dad and mum for their spiritual encouragement and material support, they encourage me very much whenever I have difficulties in doing thesis

iii

ABSTRACT

EXPORT INSTABILITY AND ECONOMIC GROWTH IN

ASIAN DEVELOPING COUNTRIES

TONG MINH TUAN

HCM Economic University ·

September 2002

In this study, we look at the relationship between export stability and economic growth in Asian developing countries using panel data analysis The database includes annual data for seventeen Asian developing countries over the 1986- 2000 period In theory, this effect is ambiguous, and it is thought that there are no unique and systematic relationships in the causes or effects of export instability Our result found in the case of Asian developing countries

is that there is a negative effect of real export instability on the growth Specifically studying in composition of export commodity give us the conclusion that the biggest negative impact falls on countries which export manufacturing products and are relatively heavily concentrated in capital-intensive sectors, rather than on countries which mainly export primary products The impact of export instability on Investment is also examined, and found negative but ambiguous This result will bring some supports for the hypothesis that the negative effect of instability mainly works through reducing the export productivity of capital stock, rather than the level of investment

iv

TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION !

1.1 PROBLEM STATEMENT 1

1.2 SCOPE OF THE THESIS AND LIMITATION 3

1.3 OBJECTIVES OF THE STUDY 4

1.4 METHODOLOGY AND DATA COLLECTION 5

1.5 THE STRUCTURE OF THE THESIS 6

CHAPTER 2: LITERATURE REVIEW 7

2.1 THEORETICAL BACKGROUND OF EXPORT INSTABILITY 8

2.1.1 EFFECTS OF EXPORT ON ECONOMIC GROWTH 8

2.1.1.1 Export is one of the sources of economic growth 8

2.1.1.2 Effects of Primary and manufactured export in Developing countries 10

2.1.1.3 Empirical studies 12

2.1.2 THE ECONOMIC CONSEQUENCES OF EXPORT INSTABILITY IN DEVELOPING COUNTRIES 12

2.1.2.1 Macroeconomic consequences ofExport earnings instability 14

2.1.2.2 Micro consequences of Export earnings instability 22

2.2 EMPIRICAL ANALYSIS 25

CHAPTER 3: MODEL SPECIFICATION AND DATA BASE 29

3.1 SOME SELECTED MODELS 30

3.1.1 FOSU MODEL 30

3.1.2 OZLER MODEL 32

v

3.1.3 JAME LOVE AND SINHA'S MEASUREMENT OF INSTABILITY 33

3.2 MODEL SPECIFICATION 35

3.2.1 INDEX OF EXPORT INSTABILITY 35

3.2.2 MODELING 35

3.3 DATA DESCRIPTION 39

CHAPTER 4: THE EMPIRICAL RESULTS 43

4.2 DESCRIPTIVE ANALYSIS 43

4.2 THE ESTIMATED RESULTS OF THE MODEL 47

4.3.1 EXPORT INSTABILITY AND GDP GROWTH 47

4.3.2 EXPORT INSTABILITY AND INVESTMENT 51

CHAPTER 5: CONCLUSIONS ~ 54

vi

LIST OF TABLES

TABLE 1 SUMMARY OF THE EFFECTS OF VARIABLES IN THE MODELS 3 7

TABLE 2 SUMMARY STATISTICS OF REGRESSION VARIABLES 44

TABLE 3 CORRELATION COEFFICIENTS OF REGRESSION VARIABLES, 249

Chapter 1: INTRODUCTION

1.1 Problem statement

The issue of export instability m developing countries and less-developed countries has· always been a great field for theoretical and empirical studies Recent events have put the problem of export instability in an important concern Some economists even consider instability as a major item affecting growth in several countries In the past two decades, many development economists have been concerned about the negative effect that instability in export may have on the growth of developing countries (Koomsup, 1978) The sudden change of oil prices by the OPEC countries in 1973-7 4 causing balance

of payments difficulties in most non-oil producing countries was an example Hence the issue of export instability has increased its important in international forums such as UNCT AD IV and the Conference of International Economic Cooperation (Koomsup, 1978:5)

A major concern of developing countries is the problem of instability in their export of products (Begg, 1991 ) Many countries in Asia are developing countries, in which Vietnam is one It is often said that export instability in developing countries is more severe than in developed countries, because many developing countries are usually exporters of primary products which are traded in a very volatile and unpredictable world market (Massell, 1964) Primary products also have a tendency to decline in term of international price with increasing fluctuation in price over time (Begg, 1991 :636) Whether export instability has a negative effect on development in these developing countries? In addition, some countries have achieved high average ratio of export over GDP (Malaysia( 57%), Thailand(25%), HongKong(111 %), Korea(30%) in 1970-1994) (Ba,1996) So instability in export may lead to instability in GDP, and thus, in GDP growth in these countries James Love(1992) undertook his study on export instability in developing countries and his conclusion was that Export instability increased between two periods of

1960-1971 and 1972-1984 Whether such mcrease m Export instability deteriorates GDP growth in these countries?

Because Vietnam is one of Asian developing countries which export lots of primary products, then what would be drawn for these countries in studying of export instability could also be useful for Vietnam Furthermore, the adverse effects of export instability may be worse than for Vietnam than a developed economy since like a developing country, Vietnam lacks necessary tools and ability to deal the problem effectively If the damage of instability could be easily recognized in Asian developing countries as well as in Vietnam, governments can reduce or eliminate the problem by doing something such as diversifying their export commodities, or reducing their dependence on a few markets Some developing countries can also reduce the adverse effects of export instability on key economic variables by separating domestic price from fluctuations in world prices Recently Vietnamese government has followed a strategy of export-oriented in order to achieve the high economic growth rate like some successful NICs,(Ba, 1996), so this makes export play a more important role in developing the economy Hence there is a need to know what the relationship between export instability and economic growth is

Export instability creates the uncertainty about the export eaming, and according to Keynes (1938), it has a negative impact on investment decisions and technological improvements On the contrary, Friedman (1954) say that Export instability may lead to a decline in consumption and then, an increase in savings, thus economic growth may be higher Previous empirical studies so far also give different results: some studies find a positive relationship between export instability and economic growth, some other studies find a negative relationship between export instability and economic growth while some other studies find no relationship between export instability and economic growth (Sinha, 1999) While there is no consensus in generally finding of export instability effects on economic growth, it is therefore interesting to study this issue in Asian developing countries The result obtained for Asian developing

countries will help us to know (i) first: what is the relationship between export instability and economic growth in countries which export lots of primary commodities like Asian developing countries, (ii) if is found that export instability does have a negative impact on economic growth, so we will know the significance of export, and the government has to follow a policy whereby such fluctuations can be smoothed out, (iii) which theoretical school is appropriate for the case of Asian developing countries Therefore beside all of empirical studies listed above we need more research to know exactly the relationship in Vietnam and Asian developing countries

1.2 Scope of the thesis and limitation

Empirical studies have also checked the existence of a negative relationship between export instability and economic growth, and some have provided support for a positive relationship Almost all previous studies have used a cross-countries approach alone, though in some studies researchers have relied

on time-series approach In this study we use an instability index that varies over time as well as across countries We then investigate the impact of export instability on GDP growth by employing annual data for seventeen Asian developing countries over the period 1986- 2000 This research will focus on Asian developing countries and estimate over period of 1986-2000 to conclude that there is a negative effect which is sufficiently large for export instability to

be a serious problem for developing countries

Our study will then rely on panel data of 17 developing countries over the most recent period of 1986-2000 The study uses mostly secondary data collected from Word Bank or IMF statistics On general problem with cross-country appro11ch is that it estimates average relationship and does not provide much information on th~ specific countries The result of the study, therefore, will give only the general and common relationship which can be apply for Asian developing countries instead of a specific relationship in a given country like Vietnam It may be a limitation of this study However, it is reason to believe that the conclusions drawn in our study may provide general guidelines of

some use in development planning by generating a better understanding of certain fundamental economic relationship Furthermore it is still necessary to study because export performance of Vietnam is rather similar to those of some Asian developing countries in the early period (Ba,1996)

1.3 Objectives of the study

It is the relationship between export instability and economic growth in Asian developing countries that this study seeks to examine The purpose for this is to know and compare the result from the case of Asian developing countries to that of other countries The research on Asian developing countries is particularly interesting because there have been considerable primary products

in export of these countries

The results from previous studies on export instability and economic growth are so far conflicting They are different from study to study Glezakos (1973) conclude that instabilities in export earnings are harmful to growth and income when studying in LDC's, but MacBean (1966), Coppock (1977) and Knudsen

& Pams (1975) do not find any empirical evidence that there is a significant relation between export instability and such economic variables as investment, national income, and economic growth rate Moran (1983) uses cross-section data for 30 countries (18 of them in Lantin America) to study the relationship between export fluctuation and economic growth Using several measures of export instability, he finds that the results are very sensitive and no general conclusions can be reached

Hence, while there are no homogenous conclusion m prevtous empirical studies, our research will study the characteristics of expmi m Asian developing countries, by investigating the relationship between export instability and economic growth in Asian developing countries

Our research needs to answer the following important question:

Research question:

1 What is the relationship between the percentage of absolute value of the deviations of actual export earning from five-year moving average of export earnings and GDP growth in Asian developing countries in the period of 1986-2000?

2 What is the relationship between the percentage of absolute value of the deviations of actual export earning from five-year moving average of export earnings and Capital formation growth in Asian developing countries in the period of 1986-2000?

In our study we use the percentage of absolute value of the deviations of actual export earning from jive-year moving average of export earnings as an index to measure export instability

Research hypothesis

In many Asian developing countries, export of primary goods has still played

an important role in developing economy in some countries (Berg, 1991) Instability in price of primary good may lead to volatile export earnings and fluctuation in GDP growth in these countries We may expect that export instability indeed have a negative influence on economic growth Many empirical studies in developing countries also bring in this result So we formulate following hypotheses:

'An increase in the percentage of absolute value of the deviations of actual export earning from five-year moving average of export earnings lead to a decrease in GDP growth in Asian developing countries'

1.4 Methodology and Data Collection

To find the impact of export instability on growth we first measure export instability as the percentage deviation of export earnings from its five-year moving average, so this instability index varies over time for each country We

will investigate the impact of export instability on econom1c m Asian developing growth by pooling across 17 countries and over time of period 1986-2000 Regression on neoclassical production function will be estimated with instability as an explanatory variable The ordinary least square (OLS) method is a suitable tool for running regressions because we are investigating the economic growth and its influential factors among which expmi instability 1s one

The data used for this study are secondary data and come from the International Financial Statistics of the International Monetary Fund (IMF) All data are expressed in relative terms to correct for country size Sample size is about 300 observations obtained from panel data of 20 countries and 16 years (1986-2000)

1.5 The structure of the thesis

The thesis will consists of five chapters:

Chapter two is concerned with the theoretical framework of instability The chapter presents some literature review and empirical studies about expmi instability The chapter will discuss about the effects of export, and then export instability, directly or indirectly impacting on growth The chapter also reviews some empirical research finding the relationship between export instability and economic growth

Chapter three introduce models and data sources for the research

Chapter four is the main one, which presents the empirical results of model estimation through descriptive analysis and econometric regression

Chapter five presents the Conclusion The chapter concludes by summarizing the thesis and suggesting some policies that would reduce the harmful effect of export instability

Chapter 2: LITERATURE REVIEW

The issue of export instability in developing countries and less-developed countries has always been a fertile ground for theoretical and empirical investigation (Koomsup, 1978) But what is meant by instability? We will first make clear some basic concepts involving export instability before presenting the theoretical and empirical issues

Definition of instability

The issues of the choice of index raised by researchers relate directly to the question of defining what is meant by "instability" Many researchers have focused on separating out trend earnings from deviations around the trend This approach regards deviations from trend as comprising instability (Love, 1992) The trend value here can be viewed as 'normal' or 'anticipated' path of earnmgs

James Love(1992) as well as most of researchers regarded 'deviations from trends' as comprising 'instability'

Coppock ( 1977) argued that instability means "excessive departure from some normal level", and several authors have found to be more specific about the definition of the "excessive" components of deviations from "normal"

On a different way, Knudsen and Parnes (1975), based on the Friedman's hypothesis of permanent income, argued that unforeseen and temporary in export earnings embrace the concept of uncertainty, thus they comprise instability Hence, by adaptation of Friedman's permanent income hypothesis Knudsen and Parnes define instability as being caused by unforeseen components in the future

Although there has been no consensus on definition of instability, the choice of instability indices based on each definition does not matter (Love, 1992) A number of authors such as Coppock, Masse! and Glezakos found that there were high degrees of correlation among estimates of instability obtained using different indices Through some studies of many researchers using different

indices in their estimating instability, results often give the same answer (Love, 1992)

In this study, we will use the definition of instability as James Love do: instability is regarded as "deviation from trend" It is chosen because it was most widely used by many authors In addition, we chose this definition in order to be appropriate with the measure of instability which will be presented

in next chapters

As a result of this export instability can be considered as determining the

deviation of actual export earning around its trend

We will first discuss the impact of export on economic growth theoretically before we can go further This reason is due to the fact that export has gradually become an important factor, and people often consider it an input in neoclassical production function Today countries in the world are more open, trading between them then becomes indispensable Balassa (197), Kreuger (1993) and Hughes (1990) argued that openness to trade was a crucial source of East Asia's rapid growth and that government's principle contribution was to limit protection and ensure that incentives were largely neutral And in such a region of open trade, according to the study by World Bank (1993), rapid export growth played an important role in permitting East Asian economies to avoid foreign exchange constrains Therefore the study suggests that exports and export policies played a crucial role in stimulating growth It seems to be reasonable that today export may be seen a new input in ·neoclassical production function

2.1.1 Effects of Export on economic growth

2.1.1.1 Export is one of the sources of economic growth

According to Neoclassical trade theory, the two countries can benefit from each other through International Trade Because of the opportunity cost, the two countries will use its resources efficiently by producing and exchanging goods

to which they have relative advantages Because now each country specializes

m its own comparative advantage product, each country can use the most abundant resources to produce, so they can get higher output and can import many other goods at relatively lower price Each country tries to use most of the abundant resources to make it have comparative advantage (B.erg, 1991) In free trade both countries can enjoy more goods than they produce and enjoy without trade Early many economists studied the gains from trade in the view

of economic growth such as: Adam Smith, Ricardo, and many economists furthered later such as Balassa, Ram, Gillis They all agreed about the significant and positive effects of trade on economic growth (Robert, 1999) Through studies in development, economists found that economic growth was caused by following factors: (i) the increase in the factor inputs, (ii) the increase in the productivity or efficiency How to increase growth also means how to increase these factors Quality of inputs between countries can be improved by reallocating resources effectively But the process of reallocating resources between countries can be taken place through export and import (Robert, 1999) Productivity can also be increased by greater specialization in each country If it is possible to exchange, a country can specialize on producing commodities to which it has relative advantage Therefore export, which also means exchange, has contribution to economic growth (Ba, 1996) Productivity in each country involves much in the development of its technologies But export is a channel for learning and technological advancement Export is also an effective means of introducing new technologies both to the exporting firm in particular and to the rest of the economy For some new industrializing countries (NICs), exporting of manufactured goods was very successful and it stimulated much economic growth through the enhancement of productivity, reallocation of resource, economics of scale, better specialization (Ba, 1996) Many studies of export which leads to rapid growth were carried out by Balassa, Feder, Kruger In the early 1970's Balassa (1971) and others began exploring the links between trade and growth Over the next twenty years, a large number of studies found that

export growth and export level were highly correl~ted with GNP growth (Edwards, 1992)

2.1.1.2 Effects of Primary and manufactured export in Developing countries

For developing countries, early the export consisted almost entirely of primary commodities while the share of manufactured exports in total export was very small (Ba, 1996) From 1970, export expansion has been taken place, especially at manufactured export The rapid development of Asian NICs and ASEAN might be due to their outward-looking and market-oriented economic policies The growth rate of expoti was very high for these countries For example, in 1982-1983 the average ratio of exports to GDP in NICs and ASEAN was over 50%, while the ratios of Unites and some other developed countries were 8% Developing countries merchandise exports· in 1999, according to WTO News 2000 Press releases, expanded by 8.5%- about twice

as fast as the global average Remarkably, the growth of export was increasing

in relation to the growth of income in these countries It is why we will examine effects of primary and manufactured export in developing countries (Ba, 1996)

According to Hecksher-Ohlin, less developed countries can also export to developed countries In contrast with Ricardo, who just based his theory on labor opportunity cost, Hecksher-Ohlin bases on differences in factor endowments of countries and explains the participation of countries in international trade (Edward, 1995) By his theory, less developed countries can exchange their labor-intensive products for capital-intensive products from developed countries If a country has relatively abundant supply of the factors used in a commodity production then it can be said that this country has comparative advantage in the commodity Hecscher-Ohlin theory seems to focus on export from less developed countries to developed countries Obviously it expanded the application of comparative advantage theory, explaining that even a less developed countries can export to developed countries Because of it's higher ratio of labor relative to other factor, less developed or developing country will export goods that use their abundant

factors such as labor intensive products and natural resources intensive products intensively (Edward, 1995)

Although it explains why less developed countries can trade with developed countries, one of the weaknesses of Hecksher-Ohlin theory is that it deals only with demand-side factor The relative prices would not only be depended on supply side, but also on demand side

Export affects on growth through productivity gain from technology transfer

In the global trade, from 1970 to now, the trend of growth rate of manufactured exports increased much rapidly than that of primary commodity The significant demand of manufactured products make them able to be exported from a developing country to other developing one, and from a developing one

to developed one as well Export in manufactures has grown fastest between advanced industrialized countries (Ba, 1996) According to Linder (1961 ), for countries with similar factor endowments, trade can take place because of different varieties of products Raymond Vernon (1966) when discussing about technology gap argued that some corporations taking advantage of cheaper labor cost in developing countries might come there to invest and operate The kind of this activity is normally known as Foreign Direct Investment (FDI) Through this the host developing countries can get gains of more productivity from technology transfer through FDI NICs and ASEAN countries, for instance, could attract and absorb relevant technique in export sectors to make them able to export their products to other developed countries Therefore, by gains of foreign capital, foreign technology, foreign skill and foreign management from FDI in export sectors, domestic economic growth will be enhanced due to "endogenous technical change" (Karel Jansen, 1995)

Export also affect on growth through investment When FDI comes to host country, it will stimulate investment and consumer demand One of the most significant contributions of FDI is that it provides capital stock and technological progress to invest in host country, thus it will enhance domestic investment (Sundrum, 1994) According to R.M Sundrum (1994), growth of

export in domestic country will create more opportunities to invest He also found positive relationship between growth of export and growth of investment

in developing countries in the 1960s Export growth also in other way increases the efficiency of investment, resulting in a lower incremental capital output ratio (ICOR) Higher rate of export growth will stimulate investor to upgrade their techniques in order to compete with the world market

2.1.1.3 Empirical studies

There are many empirical studies have been conducted to the role of exports in economic growth A large number of studies found that expmi growth and export levels were highly correlated with GNP growth (Edward, 1992) R.M Sundrum(1994) uses data for 41 less developed countries (LDCs) and runs regression of the role of export He found that the relationship between export growth and economic growth is strong positive Choong Chee Keong (2003) studies the case of Malaysia, choosing six variables which according to author are reasonable affect economic output, and finds that most of the explanatory variables, including export growth variable, are statistically significant and have positive coefficients Robert (1999) examines the case of Japan and Korea, and finds that lower tariffs and higher import volumes would have been particularly beneficial for Japan during the period 1964 to 1973 His result also leads to the conclusion that Japanese export were a particularly impmiant source of productivity growth, this suggests that further liberalization by Japan and other ASEAN countries may result in future dynamic gains Hence, export expanston is indeed a part of national products that directly adds to GDP growth

countries

The issue of export instability

The issue of export instability m less developed countries (LDC's) and developing countries has always been a fertile ground for theoretical and empirical investigations Recently, due to some bad events such as the

downward tendency of primary products, the oil shocks which have happened

to export sectors in developing countries, the issue of export instability has assumed increasing importance in international conferences (Koomsup, 1978) Export instability, at first glance, seems to be detrimental to the economies of most LDC's and developing countries because of their primary exports which are traded in volatile and unpredictable world market Because it is argued that export earnings are a major source of income in most LDC's and some developing countries, export instabilities adversely affect their ability to import essential capital goods, domestic investment, national income, and government revenue and expenditure (Ozier, 1988:18) Furthermore, because of lack appropriate tools in LDC's and developing countries, export instability is expected to cause more damage in LDC's and developing countries than in developed countries Some factors such as the concentration on a small number

of export commodities and export market, the high proportion of primary products in their export, the low income price elasticity of demand and supply

in export commodities and the small economic size are the causes of export instabilities in LDC's (Koomsup, 1978) Today there are two international organizations which were designed to reduce the adverse impacts of short-term instabilities in export earnings of LDC's One of them is IMF organization which provides necessary financing facilities to compensate for shortfalls in export earnings for circumstances beyond the control of the country The other organization is ST ABEX, which was set up by European Economic Community to some LDC's in Africa Countries in this organization will be helped to stabilize their earnings by compensating and reimbursing for the difference between actual export earnings and the average of their export earnings over the past four years (Koomsup 1978)

While the trend of price of primary products was fluctuated very much, the price of manufactured was more stable According o Maizel's calculation, there exists a decline in term of trade of primary export relative to manufactures imported during the period of 1980-1991 (Maizel, 1968) And according to Singer (1991) and Matthias Lutz's (1994) studies, volatility of terms of trade

has a significantly negative effect on growth John Cuddington (1989) studied for all developing countries, and found that the terms of trade fluctuate considerably during the period of 1954-1988, so it creates instability in export earning Observation suggests that earning instability may severely harm these countries

Export instability can affect economic growth through a number of ways Theoretically, we will examine the macroeconomic consequences of export instability, first in the short term by discussing the Dutch Disease framework and some other analysis, and second its effects on growth in the long term We then examine the effects of instability from a microeconomic point of view

2.1.2.1 Macroeconomic consequences of Export earnings instability

2.1.2.1.1 Traditional development theories:

International Economic theory argues that developing countries benefit from specializing in producing primary goods due to their endowment of factors (Berg, 1991 ) The comparative advantage of this endowment of factors may promote foreign direct investment flows But this opinion which implies a high commodity dependence is criticized by some development economists who argue that such specialization would lead these countries to be much more affected by export earning instability (Araujo, 1999) In their view export instability does have a negative effect on economic development in developing countries One of the negative effects is that instability may create the uncertainty related to the cost of commodity prices unpredictability and may lead to unforeseen losses and gains Instability in price will lead to macroeconomic fluctuations which are normally expressed as the national income instability and thus may hatm the economic growth (Araujo, 1999) The traditional economic development theories strengthen the negative effect

of macroeconomic instability Myrdal (1958) argued that in the short term export instability could create inflation thus lead to a sluggish downward reaction of price As the result of this, fiscal deficit will emerge Myrdal also believed that commodity price instability was central in explaining developed

country inflation Hence, there exists a positive correlation between fiscal deficits and exports instability In the long term, according to Nurkse (1962), and particularly, to Keynes (1936), export instability creates uncertainty and thus it has a negative effect on investment ·decision and technological improvements But the concept of "uncertainty" may be difficult to calculate in scientific meaning

Keynes dearly saw an important role for uncertainty in his General Theory He argued that uncertainty actually has a major economic impact People want to store their wealth to be stable returns, and thus they increase their demand for cash and decrease the demand for investments But it is just one impact of uncertainty on investment The other impact is that in uncertainty environment, the increased demand for cash leads to higher interest rates and hence people are driven out of investment into bonds Thus investment may be deteriorated further because of uncertainty (Keynes, 1936)

Briefly, Myrdal (1958) and Keynes (1936) emphasized the negative effect of export instability on factors of development such as inflation and investment The restriction of their argument is that Myrdal analyzed this effect of instability just in the short term, while Keynes argued just for the long tem1 In addition, in his discussion about instability, Keynes thought that uncertainty is true where there is no scientific basis to which uncertainty can be calculated

(Araujo, 1999) He said : " By uncertain knowledge, I do not mean merely to distinguish what is known for certain from what is only probable For example European war is uncertain, or the price of copper and the rate of interest twenty years hence About these matters there is no scientific basis on which

to form any calculable probability whatever We simply do not know "

(Keynes, 1937) However, few contemporaries agreed with him, and they nearly clear the distinction between risk and uncertainty Many economists argued that uncertainty is economically impotent It only has effect in conjunction with some other feature of models These conc1usions of Keynes

might be right, but if so they can be justified without reference to Keynesian uncertainty (Shapiro, 1997)

On the contrary, Friedman (1954) argued that instability could lead to a reduction in consumption, according to his theory of permanent income His hypothesis originates from the basic intuition that individuals would wish to smooth consumption and not let it fluctuate with short run fluctuations in income Friedman argued that individuals base their consumption on a longer term considering about lifetime wealth or wealth over a reasonably long horizon Friedman's hypothesis thus can explain why income, in fact, is more volatile than consumption and why the long run propensity to consume out of income is higher than the short run one In this case the intuition for precautionary motives for saving may arise Therefore instability may lead to

an increase in saving rate Because people worry about their future, they want

to save more to be secure in the future By increases in savings for precaution, instability thus can stimulate economic growth (Meghir, 1999)

The ideal of positive effect of instability was also reinforced by Hirschman (1958) He argued that in the short term, shortcuts in exports earnings would restrain manufactured import, thus promote domestic production

Briefly, the permanent income hypothesis has been regarded as the most important theory on consumption decisions In Friedman's argument there might be a positive effect of export instability on economic growth through further saving But his argument based completely on his permanent income hypothesis, which in itself has some limits The lifetime of a household may not be infinite while his hypothesis bases on infinitely lived household assumption The loose definition of permanent income leaves open the question

of its measurement On the other hand, the permanent income hypothesis as stated by Friedman did not take a vety firm view on the appropriate horizon for consumer choice Although it provides a flexible framework for the study of consumption and savings, it is this flexibility that make hard to define the theoretical basis of the model, and thus its is hard to make more detailed

statements about policy (Meghir, 1999) Friedman thought that permanent income which determines people decisions rather than current income, so it is also his limit Furthermore saving is just one of factor determining growth Besides saving technological progress is also important determining of growth

In Hirschman's (1958) argument, instability may generate positive benefits from import substitution strategies on economic growth But this phenomenon,

in fact, appears to be highly improbable

2.1 2.1.2 Recent developments of economic theories on export instability

In this part we will discuss the effects of export earnings instability in the short term and the long term respectively In the short term, the discussion stresses the link between primary export instability and competitiveness, mentioning The Dutch Disease analysis In the long term, export instability does have an impact on economic growth

Short term effects

a The Dutch disease

When a small, dependent economy has an unexpected and temporary increase

in primary export earnings, its competitiveness of manufactured goods may be deteriorated Therefore, natural resources, contrary to what common sense would indicate, may be not a blessing for rich endowed countries This economic phenomenon is known as the Dutch disease (Wikstrom, 2003)

The name of this phenomenon is obtained from the experience that followed a sharp rise in Holland's in nature gas exports in 1970 (Davis 1995) The Dutch manufacturing experienced a loss of competitiveness, particularly for export sectors, because of appreciated currency due to booming in primary exports To explain the Dutch disease, it is important to consider the distortion created by primary export booming We can outline some responds from an export boom

in a developing country:

There exists an income effect: Originally, more inflows of foreign currency cause exchange rate to appreciate

primary exports boots the national income and domestic demand Labor demand increases and then wages, maybe leading to inflation Thus the combination of an appreciation of nominal exchange rate and domestic price inflation result in a rise in the real exchange rate, which reduces the country's competitiveness

There exists a resource effect: labor will flow from the traditional tradable goods, such as manufactured goods and agricultural products, to the non-tradable goods, such as construction and services, and to booming sector Because while wages increases, it will reduce profits in the traditional exports sector In addition, the real exchange rate appreciation tends to reduce relative prices of tradable goods relative to prices of non-tradable goods Therefore, the negative consequences of spending effect is reinforced by the resource effect, thus the Dutch Disease show a definite de-industrialization effect after the export boom (Davis, 1995)

Export of manufactured commodity decline and import mcreases The increase in imports and the decline in manufactured exports are often adjusted by Govemment, with imposing import restrictions and subsidizing exports Hut it will create more distotiions by attracting investment to high-cost production Higher manufactured goods prices deteriorate agriculture and further reduce the competitiveness of tradable goods in the expoti markets (Davis, 1995)

Briefly, the Dutch Disease theory was originally used to describe the negative effect of natural gas discoveries on Dutch manufacturing in the 1960s Hence,

it has been also applied to other countries and other periods of time to explain the negative effects on country's competitiveness But it still contains some limitations: it only concems a temporary increase in expoti eamings, in the short term In addition, the mechanism that explain the adverse effect on manufacturing have not been clear when the model is applied to empirical cases Corden (1984) states that the true Dutch Disease in the Netherlands was not the adverse effects on manufacturing of real appreciation but rather the use

of booming sector revenues for social service levels which are not sustainable, but which it has been politically difficult to reduce (Corden 1984:359) Therefore, it is not clear whether the mechanism is economic or political economic in nature

b The Role of the Public Agents

This part will focus on discussing the poor public management of expmis earnings This problem may result in some ineffective ways to usc and manage the export earnings obtained from a boom While benefited from a boom through the world market, the public agent may have some following reactions: The boom is normally considered temporary and thus it will be saved and invested on international capital markets But it is difficult to see the investment because of the permanent income theory (Combes, 1993)

According to Davis, the public agent may save the boom but increase domestic investments (Davis, 1983) But it may result in low returns on public investments

In a poor and imperfect capital market, there are poor investment opportunities, and thus investment may just occur in the building sector, which is non-tradable

Briefly, the role of public agent is also a problem in developing countries where the public agent is in fact responsible for deterioration of benefit gained from export booms According to Davis (1995), the Dutch Disease is also mainly the result of poor public reactions to exports earnings instability Some negative effects of Dutch Disease are that it generates lobbying activities with custom taxes, imports quota, wages rigidities, mainly to implement protection

Long term effects

When export earnings fluctuate frequently, investors may loose information about loss and gain Therefore, in this circumstance, export instability generates what was called Risk (Kemp, 1973) Risk is one aspect of instability for which we will discuss about its effects theoretically At first glance, risk

may affect the aggregate reaction of agents, thus it has an influence on economic growth

A dynamic analysis of the effect of risk was introduced by Brock (1991) Against some economists who used static analysis, Brock introduced his

dynamic analytical framework to analyze the effects of export earnings instability on growth In his analytical framework Brock followed some hypotheses: They are: an infinite lived representative, and a neoclassical technology (which are constant returns to scale, factors substitutability, etc.) One crucial characteristic of this model is that it has production activities in random because random shocks are exogenous Basing on Brock model of dynamic analysis, some economists have argued about the effects of instability

as well as export instability on some factors such as: savings, investments, and

the economic growth:

On savings: Risk can affect on savings under two terms of risk: Labor income risk and Capital income risk While the effects of labor income risk is quite clear, the effects of capital income risk is ambiguous The effect of labor income risk is that when confronting risk economic agent will save more to insure for effects of risk (Leland, 1968) This saving will act as a buffer-stock insuring their future when there is a risk of a fall in income Among important risks are unanticipated variations in income There are some "consumption shocks" such as health expenses, and length of agents' lifetimes People would expect that a sufficiently large stock of wealth would reduce uncertainty at least about future economic constraints It is also the purpose for holding wealth Therefore the effects of labor income risk on savings is positive

For Capital income risk, it will affect on savings through two opposite effects, namely income effect and substitution effect (Sandmo, 1970) Income effect will make economic agents increase savings, thus it has a positive effect on savings Substitution effect arises when agents reduce their exposition to risk Then they want to reduce savings Hence, capital income risk also has a negative on savings

On Investment: Savings are major sources for financing Investments, thus if

risk has a positive effect on savings, it also does on Investments But in open economy, savings are partly used to purchase non-risky foreign assets A simple portfolio analysis tells us that an increase in risk will prevent risk-averse agents to invest, so it shows a negative effect For example, risk impacts on imports of raw materials and capital goods which are used to invest in expmi sectors in most developing countries These imports take account for a large portion of investment in most developing countries Therefore, it is reasonable

to expect risk of importing raw materials has an negative impact on domestic investment However, portfolio analysis also justifies for positive effect by arguing that risky often promises a high return in investment than less risky which often bring low return investments (Araujo, 1999)

Therefore, what theory has to say about the relationship between risk or uncertainty and investment is ambiguous Different theories emphasize different channels, and some showing a positive relationship and some showing

a negative relationship Few empirical studies have been done For example, Brainard and Shoven (1980) studied this relationship in a version of the capital asset pricing model (CAPM) to assess the affects of a CAPM-based measure of risk on investment via average q, the ration of the market to the book value of the capital stock They used a sample of 187 firms from the years 1958 to 1977 Their results are mix, finding both positive and negative coefficients on risk Hence, although most economists probably find plausible idea that risk reduces investment, the above discussions demonstrate economic theory does not provide a clear conclusion And the empirical evidences, although not plenty enough, also show that We therefore believe that there is room to explore further

On Economic Growth: In Neoclassical growth theory, the steady state of

growth just depends on labor growth and technological progress Therefore risk does not affect on the long run trend (steady state) of growth, it just affects on the transitionary economic growth.(Barro, 1991 )

Briefly, Brock ( 1991) based his analysis framework on neocla~sical

assumptions, so it has some limits For example neoclassical theory said that growth is exogenous from risk because it depends only on population growth and technological progress It is an obvious limit of neoclassical growth theory Recent developments of theory have considered risk as endogenous item Because risk affects investment and investment also affects technological progress So technological progress is no longer to be exogenous, and neither is growth

2.1.2.2 Micro consequences of Export earnings instability

In microeconomic point of view, we will first discuss about the microeconomic consequences of risk, mainly focusing on producers behaviors And then we will consider the risk management behavior of producers

2.1.2.2.1 Microeconomic consequences of risk on producer behaviors

In this part we examine the effect of price instability, questioning whether the price instability generates income instability In price-taker countries, this effect seems to be obviously positive For example, food price instability has very important effects on consumers in LDC's and developing countries The costs of primary commodity price instability are reviewed and can be significant Price instability can be due to both supply or demand shock But no matter what it is caused by, the price and quantities instability indeed impact on producer income instability Producers nay respond to price changes by means

of offsetting changes in volume offered for sale Thus a rise in price may be met by a fall in volume, and vice versa However, in price-maker countries, for example countries exporting manufactured goods, the story may be different There are two different cases, the case of a demand instability and the case of a supply instability In the first case obviously price instability enhances the income instability, because a fall in demand of manufactured goods and the steady decline in commodity prices made commodity exports less valuable, and

a rise lead to contrary In the latter case, the effect of price instability is ambiguous and may stabilize income (Newbery and Stiglitz, 1981) Because in

some circumstances, instable income can be compensated by demand demand elasticities

price-We now focus on the effect of instability on producer's welfare price-We know that instability generates costs for producers The cost of risk is defined as the amount that producers are willing to pay to get rid of the instability by an insurance scheme Such schemes are considered by Newbery (1981) In practice, such schemes often seem a way not so much to stabilize but to transfer income across group For the risk-averse people, the cost of risk will be high because they like safety But it is not always easy to undertake such a stabilization procedure because of costs (Newbery, 1981 )

Some recent developments focus on analysis of effects of instability on agricultural supply On theoretical grounds, the effect of risk is ambiguous An

increa~e of the income risk may either lead to the substitution of work into leisure (substitution effect) or make the producer work harder in order to achieve a certain level of income (income effect) Normally the income effect will dominate the substitution effect in the case of very risk-averse producers

So finally risk has a negative influence on the agricultural supply This is especially true for the poorest producers (Newbery and Stiglitz, 1981) If we analyze the effects of risk by products, the effects of risk can be taken into account with a portfolio analysis: then producers can compensate for an increase in risk with higher yielding agricultural choices because risk investment often promise in a higher return

2.1.2.2.2 The management of income instability by households

We know so far that risk generates costs And according to Friedman, it induces a loss of welfare when it is a source of consumption instability (Friedman, 1954) In agricultural countries, there needs to have government interventions or insurance market because agricultural sector is the largest sector But this is difficult to be realistic for very poor countries The household

in these countries may then protect themselves by diversifying their activities

Because it is rather risky for a household to rely on few exports, fluctuations in earnings from one export are unlikely to be offset by those from others

In order to be secure in the future, through the permanent income theory, households will save their current income to compensate for future income In addition, households also want to save more for precautionary reasons due to daily uncertainties A necessary condition for existence of such saving is that the households are prudent (Kimball, 1991) But this precautionary savings is imperfect according to Deaton (1990), it is limited by households consumption And such a scheme also can not prevent the households from heavy income shortfalls The poorest households are most severe affected if they are permanent Furthermore, precautionaty savings is subject to the multiplicative risks (Combes, 1993)

The agricultural households that are risk-averse may diversify their activities Diversification in the type of exports makes them possible that a fa11 in some exports may be counterbalanced by a rise in the others But there some who argue that it is not at a11 clear that more expoti diversification necessarily reduces export earnings fluctuations (Kumsup, 1978) Knudsen and Parnes (1975) argue that more diversification could in some instances increase fluctuations in earnings of the existing individual exports by reducing the share

of the original exports in the world market, thus decreasing the possibility of negative correlation between prices and exported quantities of these goods Furthermore, that kind of diversification is costly, and to some extent, it reduces the specialization benefit from households' activities of diversification The empirical studies testing the hypothesis that export commodity diversification tends to decrease fluctuations in export earnings also give conflicting results and conclusions

All risk reducing and managing resorts are costly and not always easy to undertake since they reduce the consumption These costs are severe to poorest and thus government is expected to intervene

Conclusion

In briefly, all of our theory discussions mainly focus on commodity dependent economies Theoretical analyses so far tend show many controversial conclusions, both in macroeconomic and microeconomic views Two fairly distinct schools of thought have emerged: the first view emphasizes th~

negative impact of export instability on growth, and the second view argues that export instability may encourage growth The most likely explanation of the negative correlation relies on the lack of perfect insurance markets, particularly in developing countries In such economies, risk-averse investors are likely to reduce their investment A shortfall in primary exports also has a negative effect to country's competitiveness through Dutch Disease theory The argument for positive effect is that with risk-averse individuals, uncertainty about future income will have a positive impact on savings by increasing the precautionary demand for savings This in tum will eventually lead to higher investment and higher growth For Dev.eloping countries in Asia, arguments for negative affects have showed its strength Because there are many primary-producing countries which tend to depend more heavily on exports as a source

of income than do industrialized countries Therefore a given degree of expoti instability has a greater impact on the economy of a primary-producing country Theoretical advances in the comprehension of the effects of instability have also been greatly improved by the use of new modem analytical tools

There have been many empirical studies so far studying the effects of expoti instability And a majority of them use data from developing countries As discussed above, the results for the relationship between export instability and growth is ambiguous, and are far from conclusive As Muller-Sebastian (1998) remarks: "Three decades of research on export instability have resulted in a consensus on only one of the main areas of study, namely, that export instability is higher for LDCs than for developed countries" We can now list some of works:

Fosu (1991) uses the squared deviations from the best-fit trend index to measure instability and he finds that both export instability and capital instability have a negative on economic growth in Sub-Saharan Africa In his studying he uses a cross-sectional analysis involving 1967-86 for less developed countries of Sub-Saharan Africa

Knudsen and Parnes (1975) use a transitory index to measure instability and they find that marginal propensity to consume out of permanent income is negatively related to export instability In studying they use cross section data for 28 developing countries (average data for 1958-68)

Yotopoulos and Nugent (1976) use two measures of export instability: (1) the squared deviations from an exponential trend index and (2) an index in the spirit of permanent income hypothesis (a transitory index) And they find that when the transitory measure is used, the effect of export instability reduces the marginal propensity to consume out of permanent income, increases saving and higher growth But when they use the measure of squared deviations, it leads to the opposite conclusion: it is that export instability has a negative impact on economic growth In their studying they use cross section results from data for

38 developing countries

Ken en and Voivodas ( 1972) and Ozier and Harrigan ( 1988), studying in 26 developing countries, 1963-82, find a negative effect of instability on investment

Moran (1983) uses several measures of export instability when studying the relationship between export fluctuations and economic growth Data used is cross-section data for 30 countries (18 of them in Latin America) and for a single year, 1974-75 He finds that the results are very sensitive to the period and no general conclusions can be reached

Gyimah-Brempong (1991) uses data for 1960-86 for 34 Sub-Saharan African countries The same conclusion like Fosu for Sub-Saharan, Gyimah-Brempong uses the production function framework and he finds that no matter how export instability is measured, export instability has a negative effect on economic

growth He uses three different measures of export instability: (i) the coefficient of variation of export earnmgs, (ii) the mean of the absolute difference between actual export earnings and its trend value, (iii) average of the squares of the ratio of actual export earnings to trend earnings

In briefly, all above studies bases on cross-section data Usually researchers use econometric analysis on international data One general problem with cross-section data is that the studies using cross section data estimate average

relationships and does not provide much information on the specific countries (Araujo, 1999) Moreover, many studies measure the instability indices as the difference between the observed values and the trend of export earnings This measure is biased if the trend is stochastic The problem in which the results are not homogenous - a positive effect (Knudsen and Parnes, 1975), or a nil effect(Sinha,1999) or a negative effect (Moran, 1983, Yotopoulos and Nugent) may be due to a changing period coverage and often to the lack of an appropriate model, or very diverse samples

Some researchers use time series data to study such as J Love (1987), and most recently Sinha (1999)

Love uses data for developing countries which rely on primary goods and are subject to more fluctuations than the industrial goods export He uses absolute deviations from a five-year moving average as an index to measure instability His study focuses on the sample of 20 countries He finds that export instability causes income instability However, Love does not test for stationarity of the variables when estimating According to Sinha (1999), the variables are non-stationary in their levels and not cointegrated So if not test for this before estimation may lead to a doubt about the result

Sinha (1999) uses a time-series data analysis studying the relationship between export stability, investment and economic growth in nine Asian countries He also follows Sinha in using the five-year moving average index to describe instability He finds that in most cases, economic growth is found to be positively associated with domestic investment For Japan, Malaysia,