Trade and Economic Growth

Growth and Welfare Effects of Trade and Comparative Advantage

Bachelor Thesis 2010 64 Pages

Economics - Foreign Trade Theory, Trade Policy



1 Introduction

2 Trade and Economic Growth - The Empirical Relationship
2.1 General Empirical Evidence

3 Sources of Economic Growth
3.1 Capital Accumulation
3.2 Productivity
3.2.1 Solow-Model
3.2.2 Research & Development (R&D) - Endogenous Growth , , , ,
3.2.3 Learning by Doing
3.2.4 AK-Model

4 Openness in Growth Theory
4.1 Capital Accumulation - Trade and Neoclassical Growth
4.2 Trade and Endogenous Growth
4.2.1 Research and Development (R&D)
4.2.2 Learning by Doing

5 Static Comparative Advantage
5.1 Supply Side
5.1.1 Productivity Levels
5.1.2 Wages
5.2 Demand Side

6 Dynamic Comparative Advantage
6,1 A Model of the Dynamics of Specialisation
6.1.1 Infant Industry Argument
6.1.2 The Static Model
6.1.3 Learning by Doing and Productivity Dynamics- Endogenous comparative advantage
6.1.4 Dynamic analysis
6.1.5 Opening up to trade - Loek-in Effects of Trade Patterns ...
6.1.6 Expanding the Market Share
6.2 Welfare Analysis
6.2.1 The Static Model
6.2.2 Learning and Productivity Dynamics
6.2.3 Welfare Analysis
6.2.4 Static Welfare in Autarky
6.2.5 Static Welfare in Free Trade
6.2.6 Dynamic Welfare Effects in Autarky
6.2.7 Dynamic Welfare Effects in Free Trade

7 Conclusion

8 References

9 Appendix: Calculations
9.1 Solow-Model
9.2 AK-Model
9.3 Welfare Analysis

1 Introduction

“Economic growth (...) is the part of macroeconomics that really matters.” [1 ], [2 ]

“You could say that the study of international trade and finance is where the disci­pline of economics as we know it began."3

The following paper relates two of the most important eeonomie phenomena, namely eeonomie growth and international trade. The notion that international trade may be the engine to growth and welfare is not new, though it has not always been very popular. Going baek at least to Adam Smith’s discussion of specialisation, it has been subject to a good deal of research and discussions and the general opinion on trade has varied throughout history.

The second half of the last century was characterized by an unprecedented pro­cess of regional and global integration. Between 1960 and 1968 the volume of world exports annually grew by 7,3 percent on average and between 1968 and 1973 by 9,7 percent. The oil crisis led to a significant slowdown but exports still continued to grow between 2 percent and 5 percent per year until 1990, In 2007 the world’s to­tal production amounted to approximately 50 trillion US-Dollars at current prices. More than 30 percent thereof was sold across national borders. The growing in­terconnection of countries, companies and people does not only bring about the exchange of goods. Flows of factors of production such as physical capital, labour and human capital also have to be taken into consideration nor should the flows of technology, innovation, ideas and international aid be forgotten. The list of eeo­nomie interactions among countries could be continued.4

Considering that eeonomie integration is constantly increasing and trade has be­come a natural part of our world it is almost evident to ask for the relationship between trade and eeonomie growth, Eeonomie growth is of tremendous impor­tance for eeonomie welfare and the standards of living. Even small variations of growth rates can lead to vast differences over the years and can influence the stan­dard of living enormously. The following calculation illustrates the importance of growth rates in the long run. In 1870, the real gross domestic product (GDP) per capita of the United States amounted to $3,330, Until 2000, it increased tenfold to $33,330, which implies an average growth rate of 1,8 percent per year. Suppose, the United States had only grown by 0,8 percent per year since 1870, then its real per capita GDP would have been $9,450 in 2000, This corresponds to only 28 percent of its actual valne and is close to that of Mexico and Poland, India, Pakistan and the Philippines experienced growth rates of approximately 0,8 percent from 1900 to 1987, If the United States, on the other hand, had grown by 2,8 percent, i.e, one percentage point more per year, its real per capita GDP would have amounted to $127,000 in 2000, Should they continue to grow at 1,8 percent, the United States will not reach this level before 2074.5 “In fact, aggregate growth is probably the single most important factor affecting individual levels of income.. Hence, understanding the determinants of aggregate economic growth is the key to understanding how to increase, the. .standards of living of individuals in the. world and, thereby, to lessen world poverty."6

The following paper focuses on impacts of trade on the economy as a whole, thus on the effects on welfare and growth. The discussion of the distribution of gains and potential losses from trade between the inhabitants of a country does not go into detail. The fact that not all the inhabitants of a country necessarily profit from possible benefits from trade should, however, not be forgotten.

Before analysing the relationship between two economic phenomena in detail the study of empirical evidence is always of interest to to get an idea of the topic. Since both international trade and economic growth form natural parts of our world their empirical relationship has been subject to a good deal of research and discussion and a lot of studies have been conducted. The second section provides an overview of some of the most prominent empirical studies concerning the relationship between openness to international trade and economic growth in general. As most of them seem to have reached the conclusion that trade influences growth in a positive way the question for the reasons of this presumably positive relationship arises.

Before introducing various channels through which trade might have an influence on growth, factors which cause or influence economic growth in general need to be identified. This is the aim of the third section. It is dedicated to various sources of economic growth such as capital accumulation and productivity growth to stress their importance and analyse how they affect economic growth. The Solow-Model and the АК-Model are introduced in order to distinguish between long-run and short-run effects of capital accumulation, learning by doing and R&D on economic growth.

The forth section has a similar structure and the sources of growth, which have already been described in the third section, such as capital accumulation and produc­tivity, are dealt with again. After analysing their importance for economic growth in general in the third section a possible influence of trade on each of these factors is examined in a rather theoretical context and enriched by some empirical evidence.

The remaining analysis concentrates on one channel in particular, namely on how trade determines a country’s import and export structure. The importance of the range of products a country produces is enormous and affects economic growth and welfare. The fifth section introduces the static “Ricardian" model of comparative advantage in order to show how productivity levels dictate the patterns of trade and determine which products a country produces depending on static productivity levels at the time a country opens up to trade.

Since productivity levels do, however, not remain constant but are influenced by learning by doing and specialisation, dynamic effects of specialisation on com­parative advantage should not be neglected. For this purpose, a model of dynamic comparative advantage is introduced in the sixth section. It shows how compara­tive advantages which exist at the time an economy opens up to trade tend to lock in and determine trade patterns in the long run. The question is raised when an economy should open up to trade and a justification of the infant industry argument is provided on theoretical grounds. The paper is concluded by a welfare analysis, which tries to answer the question under which conditions free trade or protectionist policies are best suited for a country.

2 Trade and Economic Growth - The Empirical Re­lationship

The importance of economic long-term growth for human welfare and the standards of living is enormous. The consequences of economic integration and trade for economic growth and human welfare are of great interest not only from an academic point of view but also from an economic policy perspective. The topic has been subject to a lot of research and discussion. Before question if, how, why and through which channels trade influences growth, arises it is appropriate to introduce some general empirical evidence on the relationship between trade and economic growth. The following section gives an overview of some of the most prominent empirical studies in the field and their results.

2.1 General Empirical Evidence

In the 1990s economists devoted a lot of attention to the relationship between open­ness and economic growth and published various empirical cross-country analyses motivated by the improvement of growth theory, the availability of more complete and qualitatively better data and new econometric techniques. Relating different measures of “outwardness", “openness" and “trade" to growth rates of GDP or to­tal factor productivity (which in turn enhances the GDP growth rate as economic growth theory has shown), they tried to quantify the effects of trade on growth. This section picks a few of the most prominent works and describes their re­sults, Since neither “openness" nor “trade" can be measured directly economists have developed different indices capturing the two phenomena.

The first paper to be introduced is a study by David Dollar, Dollar’s (1992) paper may be the “most heavily cited empirical paper on the link between openness and growth."7 He created an openness index composed of a weighted average of an exchange rate under-/overvaluation index and an exchange rate volatility index. The index is based on the assumption that openness to trade needs stable and correctly valued exchange rates to keep incentives constant and uncertainty low. Outward- oriented countries, such as a lot of Asian countries, tend to have undervalued real exchange rates which encourage exports. Inward-orientation, on the contrary, is often reflected in an overvalued exchange rate as for example in Africa and South America, An overvalued exchange rate encourages the growth of the non-tradable sector. As far as the volatility of exchange rates is concerned it was less volatile in Asian countries than in African and South American countries, which were known to be more inward-orientated. Dollar assumes that the price level of a country’s goods depends on its factor endowments and that the law of one price (purchasing power parity) remains true in free trade. In order to eliminate the effects of individual country differences on the price levels he uses the residuals from a regression of the prices of a representative bundle of consumption goods on factor endowments for his index.

Dollar (1992) uses data for 117 economies (mostly developing countries) over the period 1976-1985 and the calculation of the countries’ openness indices shows that developed countries tend to be more open than developing countries and that countries that are defined as being more open experienced a better economic perfor­mance, Notably, Thailand, Malaysia and the Asian Tigers are among the 25 percent most open countries, whereas Ghana, Uganda, Nigeria and Tanzania form part of the 25 percent most inward-oriented of the sample, A regression of GDP per capita growth on this openness index controlling for investment ratios in order to set apart the effect of capital accumulation during convergence towards the steady-state shows that the investment rate and openness are positively associated with growth. In par­ticular, countries in the most open quartile grew on average 2,9 percent, whereas the most closed quartile shrunk by -1,3 percent. According to Dollar (1992) Latin America and Africa would have grown by 1,5 and 2,1 percentage points respectively more, had they been as open as the Asian countries.8

Sachs and Warner (1995) used data from 1965 to 2000 and defined countries as open or closed in different years depending on non tariff barriers, average tariff rates, government monopolies on exports, the black market exchange rate and the economic system. According to their results, countries that have been open for a longer period were more likely to be rich. Countries that were always open were on average 4,5 times as rich as countries that were never open. Between 1970 and 1989 the average per capita GDP growth rate of the developing countries classi­fied as open was 4,49 percent per year as opposed to 0,69 percent for the closed countries. Also within the group of developed countries the open ones grew by 2,29 percent, far more than the closed ones, which grew by 0,74 percent. The typical developing country started as being closed and liberalised later. However, Sachs and Warner (1995) identified fifteen countries who closed their economies during the pe­riod observed, Bolivia, Costa Rica, Turkey, Marocco and a few more countries in South or Middle America are among them. The average per capita GDP growth rate dropped significantly in the closed period in twelve of the fifteen countries,namely from 11,67 percent to 2,4 percent in Turkey, from 2,49 percent to -2,52 per­cent in Bolivia and from 3,88 percent to -0,45 percent in Venezuela only to name a few.9 This suggests that “the. decision to dose, the economy was generally not caused by slow growth during the open period, but rather by political and ideological shifts within each country.10 The results of a regression of per capita GDP growth rate between 1970 and 1989 on their binary openness indicator controlling for initial per capita GDP, political unrest and revolutions, school enrolment rates, government spending and population density show that “open economies grow, on average, by 2.f5 percentage, points more, than the. closed economies, with a highly statistically .significant effect."11

Edwards (1998) reacts to the problem that a unique measure of openness does not exist. Economists have defined openness differently and actually analysed differ­ent phenomena due to varions indices of “openness", which are imperfect correlates of each other.12 While some have concentrated on actual trade flows, measured for instance by the share of exports or imports in GDP, others have measured trade policies, sneh as tariff levels and other trade barriers.13 Edwards (1998) encourages researchers to leave the difficulties in finding an appropriate and satisfactory open­ness index behind and concentrate on testing whether economic results are robust to different indices. Using data from 93 countries, he regresses growth of total fac­tor productivity on nine different openness indices inclusive of Sachs and Warner’s Index, He controls for the initial GDP per capita and years of schooling and applies two different OLS methods. The relationship between the growth rates and the respective openness index is positive in the ease of all but one openness index (The negative sign of this particular index is, however, not significant.) and 13 out of the 18 results are significant.14

Edwards (1998) concludes that a “significantly positive, relationship between open­ness and productivity growth” 15 exists, which in turn indisputably influences eco­nomic growth positively.

Ann Harrison (1996) conducted a similar study. She regressed GDP growth on seven different openness indices using panel and cross-country section data and received the expected positive sign in most eases, though far from all of them are significant.

For further empirical literature on the subject Frankel and Römer (1999) and Dollar and Kraay (2002) for instrument variable approaches, Waeziarg and Welch (2003) for an extension of Sachs and Warner (1995) and Baldwin (2003) and Ro­driguez and Rodrik (1999) for critical reviews can be referred to among others.

Even though economists seem to agree that the relationship between openness to trade and growth is positive, the nature of this relationship is still controversial. Is it a causal relationship? If yes, in which direction does the causality run? Or does a third characteristic cause countries to be rich and open at the same time, which has been ignored up to now? Some economists even doubt that trade and growth are positively related under all circumstances: “We are in faet seeptieal that there is a general, unambiguous relationship between trade openness and growth waiting to be discovered. We suspect that the relationship is a contingent one, dependent on a host of country and external characteristics. Research aimed at ascertaining the eireumstane.es under which, open trade, policies are conducive, to growth, (as well as those, under which, they may not, be) and at, scrutinizing the channels through, which, trade policies influence, economic performance, is likely to prove more productive."16 A lot of research on the subject still needs to be done so that a consensus might eventually be found some day.

Irrespective of the use of different data and methodologies and the ongoing dis­cussion, the studies unanimously seem to obtain the same qualitative result, namely that a positive relationship between openness and growth is evident. Additionally, it should be noticed that no famous paper has reached the opposite conclusion, namely that trade affects growth in a negative way.17

3 Sources of Economie Growth

“If we. can learn about government policy options that have even small effects on long-term growth rates, we can contribute, much more to improvements in standards of living than has been provided by the entire history of macroeconomic analysis of countercyclical policy and fine-tuning.18

Studies that have investigated into the general empirieal relationship between trade and eeonomie growth have been the subject of the previous section. The general consensus seems to be that trade affects growth in a positive way. The interested reader might now ask himself why this is the ease since the reasons and different channels have been ignored up to now. Before analysing the reasons for the positive relationship between trade and growth as well as different channels through which trade might influence growth, another question should be asked. What are the main sources for growth? Economists have asked themselves this question at least since 1776, when Adam Smith published “An Inquiry into the Nature and Causes of the Wealth of Nations”.

Some of the main sources for eeonomie growth are described in the context of growth theory in the following section to illustrate their general importance for eeonomie growth. They are picked up again in the forth section, which has a similar structure and tries to answer the question how these individual factors affecting growth might be influenced by openness to trade. The following section consequently concentrates on the sources of eeonomie growth and prepares the reader for the forth section, which relates them to international trade.

3.1 Capital Accumulation

“It is by now incontrovertible, that increase, in per capita income cannot be. explained simply by increases in capital-labour ratio.19

Substantial eeonomie growth started in the course of the industrialisation in the 18th century. The industrialisation led classical economists such as Adam Smith to believe that capital accumulation (i.e, machines,,,,) was the source of growth. However, the question arises whether saving and investing in an economy’s capital stock can lead to eternal positive growth rates.

Cross country data from 1960 to 2000 reveal that the average growth rate of a sample of 38 Sub-Saharan African countries’ real GDP was 0,6 percent and their average investment ratio was only 10 percent, whereas nine East Asian “Miracle" countries had an investment ratio of 25 percent and grew by 4,9 percent. Before concluding that high investment rates lead to higher growth rates, attention should be drawn to the fact that the average growth rate of 23 OECD countries was 2,7 percent even though their investment ratio was 24 percent. Their growth rate was consequently lower although the investment ratio was almost as high as in East Asia, Thus, capital accumulation, even though it indisputably has positive effects, cannot be the sole engine to growth.20

Already the classical economists emphasized that decreasing returns to capital imply that growth driven by capital accumulation alone once has to come to an end. In 1848, John Stuart Mill stated that “(...) it would require but a short time to reduce profits to the minimum, if capital continued to increase at its present rate (..with,,) no circumstances having tendency to raise the rate of profit in the meantime.” 21 Falling profits slow down the incentive for investment and thus capital accumulation and growth. Even though, capital accumulation constitutes an important factor of growth, it cannot be the only one in the long run, “Economic growth that is based on the expansion of inputs rather than on growth of output per unit of input is inevitably subject to diminishing returns (..and,,) input driven growth is inevitably limited.” 22

What is it then that drives economic growth in the long run?

3.2 Productivity

“Productivity growth is not everything, but in the long run it is almost everything.”23

Economists assume that growth is unlikely to stop since most countries have had a positive average growth rate of around 2 percent for the last 200 years. The classi­cal economists were also aware of the fact that one of the “circumstances having the tendency to raise the rate of profit” 24 were technological progress and its impact on productivity.

Productivity is defined as output per unit of input, thus productivity of labour is output per hour of labour Y/L, where Y denotes output and L labour. People’s standard of living depends on their consumption and consumption per capita is naturally restricted by output per capita, thus by productivity (if L is the number of inhabitants of a country and under the condition of full-employment). Consequently, growth of per capita consumption is only enabled by growth of productivity in the long run.

After 1870, the economists’ centre of interest drifted away from growth theory until 1956, when Solow published a neoclassical growth model widely known as the “Solow-Model".25 The model sums up the classical ideas in a mathematical way, Robert Solow comes to the conclusion that growth caused by capital accumulation alone cannot continue eternally unless technological progress enhances productivity. Economists became aware of the importance of productivity as the only force that could stop the slowdown of growth due to decreasing returns to capital. Economists have dealt with technological progress in different ways. Two lines of theory have to be distinguished, namely neoclassical and endogenous growth theory.26

As already mentioned above, Solow’s model is a neoclassical growth model, which implies that he took technological progress as given exogenously by modelling its consequences but not its sources. The general level of productivity is assumed to grow at an exogenous and constant rate, without asking for the reasons. Output per capita grows due to both, rising capital intensity and technological progress until it reaches the steady-state. This is a condition where the capital-output ratio stays constant because decreasing returns to capital slow down investment, which is just high enough to offset depreciation and the negative effect of population growth. However, thanks to technological progress output per capita continues to grow at the rate of technological progress g. This is Solow’s important message: It is the rate of technological progress g, which determines growth of income per capita in the long run. Thus, long-term growth can only be influenced by productivity!27

3.2.1 Solow-Model

“All theory depends on assumptions which are not quite true. That is what makes it theory. The art of successful theorizing is to make the inevitable simplifying as­sumptions in such a way that the final results are not very sensitive.” 28

A more formal analysis will illustrate the aspects outlined in the previous section and introduce the Solow-Model. The Solow-Model is the most widely known growth model and explains growth as the result of capital accumulation, labour and tech­nological progress. Production is described by a simple Cobb-Douglas production function.

illustration not visible in this excerpt

К is the capital stock, L the labour force and A у 0 the level of productivity, which comes as “manna from heaven"29 and requires no input of production. Each of the variables depends on time t. The level of productivity is modelled as labour- augmenting or “Harrod-neutral”, Another possible Cobb-Douglas function would be Y = AF(K,L), which is called “Hicks-neutral”.30 The results of the Solow-Model are not influenced no matter which of the two kinds of technology is chosen. The share of capital is denoted by a. Since output depends positively on both capital and labour their exponents must be positive. It follows that a can only take values between zero and one (0 У a У 1), Otherwise 1 — a would be negative. The fact that a У 1 implies decreasing returns to capital and to labour separately but con­stant returns to scale to all input factors together. The level of productivity A and the labour force L are assumed to grow at the constant exogenous rates g and n:31

illustration not visible in this excerpt

If y = Y/L and k = K/L, the production function can be rewritten as

illustration not visible in this excerpt

Output per capita y is a function of the capital-labour ratio к and the produc­tivity level A. The growth rate of output per capita is calculated by taking logs and differentiating equation (3) which results in

illustration not visible in this excerpt

where gy and gk are the growth rates of y and к respectively and g is the growth rate of productivity. Growth of output per capita can result from two different sources, namely from growth of the capital-labour ratio or the level of productivity. It is a weighted average of their growth rates with the weights a and 1 — a. The interesting question that remains to be answered is what will happen in the long run. In the long run the economy will converge to a steady-state where к and у grow at the same rate since k must remain constant. If gy = gk, all three parameters, у, к and A must grow at the same rate.

illustration not visible in this excerpt

Consequently, output per capita grows at the same rate as the capital-output ratio and the productivity level. Put differently, it is productivity growth, which deter­mines growth of income per capita in the long run.32

For simplicity, the analysis of the steady-state is continued using the stock of capital per effective unit of labour, which is defined as = Akl and ^eve^ °f output per effective unit of labour fc = AL- Both, and к remain constant in the steady-state, since к and A grow at the same rate. The saving rate s equals the investment rate since the model describes a closed economy. The evolution of к can be described by

Abbildung in dieser Leseprobe nicht enthalten33

where the rate of depreciation is denoted by δ. Capital per effective unit of labour consequently grows at

Abbildung in dieser Leseprobe nicht enthalten34

In the steady-state the stock of capital per effective unit of labour does not grow(k= 0), Consequently, the following condition must be true:

illustration not visible in this excerpt

The steady-state is illustrated graphically in Fig, 1 as the intersection of the two k between two periods k.

illustration not visible in this excerpt

Fig, 1, Solow Diagram I35

Fig, 1 can be transformed by using the expression for the growth rate of capital per effective unit of labour in equation (7) as illustrated in Fig, 2, Since the growth rate of k must be zero (| = 0) the following condition must be met:

illustration not visible in this excerpt

The steady-state is illustrated graphically in Fig, 2 as the intersection of the two graphs. The difference between the two graphs is the growth rate of capital per effective unit of labour. The saving curve ska-1 is a falling function of k due to decreasing returns to capital.


1 Barro et Sala-i-Martiri 2004: 6

2 Some chapters begin with a quotation. Please don’t assume that I totally agree with all the quotes. Think of these quotes as provocations and hypotheses to be critically assessed.

3 Krugman et Obst fehl 2009: 1

4 see Krugman et Obstfeld 2009: 12ff and Ray 1998: 623ff

5 see Barro et Sala-i-Martiri 2004: Iff

6 Barro et Sala-i-Martiri 2004: 6

7 "Rodriguez et Rodrik 1999: 6

8 see Dollar 1992: 540ff and Rariiseger 2007: 2

9 see Sachs et Warner 1995: 33ff

10 see Pritchett 1996: 208

11 see Raniseger 2007: 5

12 see Edwards 1998: 391

13 Edwards 1998: 391

14 Sachs et Warner 1995: 35

15 Sachs et Warner 1995: 47

16 Rodrigucz et Rodrik 1999: 6

17 scc Rariiscgcr 2007: 4

18 Barro et Sala-i-Martiri 2004: 6

19 Arrow 1962: 155

20 see Barro et Sala-i-Martin 2004: 23

21 Mill 1848: Book IV, Section iv. 10

22 Krugman 1994: 03(1

23 Paul Krugman cited in Lutz 1997: 13

24 Mill 1905: 739

25 see Solow 1956

26 see Lutz 1997: 3ff

27 see Lutz 1997: 13ff

28 Solow 1996: 65

29 2BSorensen et Whitta-Jacobson 2005: 133

30 see Jones 2002: 36

31 see Mankiw et al. 1992: 408ff and Jones 2002: 23ff

32 soo Sorensen et Whitta-Jacobson 2005: 130ff

33 For calculation see Appendix 9.1

34 For calculation see Appendix 9.1

35 scc Jones 2002: 39


ISBN (eBook)
ISBN (Book)
File size
3.1 MB
Catalog Number
Institution / College
Vienna University of Economics and Business – Institut für Außenwirtschaft und Entwicklung
Trade Growth Comparative Advantage Dynamic Comparative Advantage Infant Industry Trade and Growth




Title: Trade and Economic Growth