'The More We Import From Developing Countries, The More They Will Import From Us'

This paper will use both qualitative and quantitative analysis to evaluate the proposition that the more we import from developing countries, the more they will import from us. For the purposes of this analysis, it will be assumed that Ireland aims to maximise its net exports, rather than its real income, quality of life, or any other measure of welfare. The many other arguments in favour of trade (such as the gains from specialisation, scale and international competition) will therefore be ignored.

While there are several mechanisms by which an increase in Ireland's imports from developing countries could theoretically lead to an increase in their demand for Irish goods, there are a number of qualifications to this theory. There is, in practice, only a very small link between Irish imports and future demand for Irish goods.

Section I of the paper will state the arguments in favour of the proposition. Section II will examine the problems with these arguments and Section III will run three simple regressions using trade data from the last 23 years to test the theory.

Because of capital flight, deteriorating terms of trade, high debt-servicing costs and the increase in protectionism in developed economies in the 1980s, most developing countries face perpetually high balance of payments deficits. Even direct foreign investment, which many economists regard as the solution to many of the developing world's problems, has been shown in a number of studies to have a negative net effect on the host country's balance of payments. This is because multinational companies tend to import inputs and repatriate profits, royalties, management fees and interest payments.

Since foreign exchange flows and balance of payments deficits are a major source of concern, many Third World governments attempt to control import levels. Thus by importing goods from developing countries, we can alleviate their balance of payments problems and make them more open to imports from us.

When we protect our economies from competitors in developing countries, we encourage those countries to become more inward-oriented. On the other hand, if we open up our markets to producers in less developed countries, those countries will become more outward-oriented.

This will probably have a direct influence on the amount they import from us because, as they become more export-oriented, they will have to compete with firms in developed countries. They will therefore need access to technology and inputs, which are currently available in the developed world alone.

More importantly, a shift towards export orientation will have an indirect effect on the amount less developed countries import from us, because outward-oriented economies tend to grow faster than inward-oriented ones, resulting in a greater demand for Irish produce in the future.

There are a number of reasons why outward-oriented economies grow faster than inward-looking ones:

• The most important way in which an outward orientation leads to output growth is by allowing developing countries to exploit their comparative advantage, thereby using their resources more efficiently;
• Indigenous companies in export-oriented countries are forced to compete with foreign firms, they thus have a greater incentive to innovate, reduce costs and eliminate inefficiencies. This leads to further productivity gains for the export-oriented economy;
• Interaction with developed countries facilitates technological advancement in less developed countries through the transfer of technology as well as managerial skills and business know-how;
• Export-oriented economies tend to have higher savings rates. The 1987 World Development Report offers three explanations for this:

The policy shift from inward to outward orientation increases real incomes. In the developed world, the marginal propensity to save is generally higher than the average propensity to save, so a rise in real incomes will cause average saving levels to rise.

Export growth leads to higher domestic savings than growth in domestic sales, because 'the entrepreneurs and firms engaged in exporting probably engage in higher than average saving'.

Capital markets tend to be very distorted and underdeveloped in the Third World - so much so that real interest rates may be negative in some cases. Capital markets in export-oriented countries perform much more efficiently than those in inward-oriented countries, so savings levels are higher.

• Inward-looking countries find it more difficult to attract direct foreign investment.

There is an abundance of empirical evidence supporting the theory that an outward orientation leads to better economic performance in developing countries. The World Bank conducted a survey of forty-one developing countries in 1987, in which it divided the countries into 'strongly outward-oriented', 'moderately outward-oriented', 'moderately inward-oriented' and 'strongly inward-oriented' economies. They found that the more outwardly-oriented an economy, the faster its output level tended to grow. In another study, Anne Krueger concluded that 'an increase in the rate of growth of export earnings of one percentage point annually was associated with an increase in the rate of growth of GDP of about 0.1%'.

Export promotion policies also help countries to recover from external shocks, according to a study of the responses of 43 developing countries to the 1973-79 period of shocks. It concluded that:

By removing all protection from developing world exports, we provide better incentives for them to adopt an outward-looking approach to economic policy. This will boost their growth and, in the long run, increase their demand for Irish products.

• An outward orientation may not lead to higher growth for developing economies.

While the World Bank, the IMF, the World Trade Organisation and most other international institutions agree that an outward orientation has beneficial effects on income growth, critics of these institutions are not convinced. They point to a number of reasons why export promotion may not be the best policy stance for developing countries.

Many less developed countries have, in the past, tried to improve their export performance by specialising in those sectors in which they have a comparative cost advantage (usually products which are intensive in the use of unskilled labour and natural resources). One problem with this, from the developing country's point of view, is that prices of these products tend to rise slower than prices of developing countries' imports leading to deteriorating terms of trade for the developing country. This means that developing countries have to export more and more every year (in real terms) simply to keep importing the same amount. The Neo-classical Heckscher-Ohlin model, which is essentially a static model, does not take this into account.

Furthermore, developing countries frequently find that their comparative advantage lies in cash crops such as tea, coffee and bananas. This leaves them extremely vulnerable to price fluctuations in world markets for these commodities. Such vulnerability and instability are not conducive to high rates of income growth.

Critics of export promotion argue that the empirical studies cited in Section I are either inconclusive or misleading. David Evans, for example, argues that the three 'strongly outward-oriented' countries in the World Bank survey - South Korea, Singapore and Hong Kong - are all 'either city states or exemplary nation states whose experience cannot easily be generalised'. A close analysis of the World Bank's data reveals that 'it is not possible to reach any strong conclusions based on the comparison of the moderately outward and the moderately inward-oriented countries'. Hans Singer argues that the correlation between export promotion and growth is strong only when external conditions are favourable. Paul Krugman criticises the subjective criteria used to classify states as outward- or inward-oriented and notes that 'the decision to classify South Korea as 'open', for example, has raised many doubts'.

Nicholas Stern echoes the views of many other critics when he criticises the World Bank survey as being too one-sided in its presentation of the results.

Furthermore, even if the Asian tigers' success is due to their outward orientation, it does not follow that other developing countries can do the same. Richard Pomfret calls this the fallacy of composition argument - that 'a simultaneous increase in supply by many countries differs in effect from a few countries' increased supply'. Pomfret concludes, however, that the 'empirical evidence for or against the fallacy of composition argument is inconclusive'.

• If developing countries' demand grows, they will not necessarily import more from us.

A more fundamental problem with the proposition is the fact that, although we may be able to boost developing countries' demand for imports, we have no guarantee that they will consequently import more from us. Most of the added demand is likely to be for goods produced in other developing countries, or indeed in other developed countries.

This argument can be expounded using game theory. Let us examine, for example, the hypothetical situation where Ireland attempts to increase Ghana's demand for Irish products by importing an extra $100 billion from Ghana. For the purposes of this analysis, we shall take the view that this increase in the level of Irish imports imposes costs on Ireland (in the form of lost jobs, for example). As Section I concluded this will eventually lead to an increase in Ghana's demand for foreign goods. However, only a small proportion of those goods is likely to be Irish.

Table 1 below shows the payoff matrix for this situation. The payoffs have been calculated assuming that an increase of $100 billion in imports from Ghana will ultimately lead to an increase of $150 billion in exports to Ghana. It is assumed that 2% of this increase (that is, $3 billion) will be bought from Irish producers.

Table 1

As Table 1 illustrates, no matter what the rest of the world does, it is never in Ireland's interest to import from Ghana in the expectation of higher future exports because the costs of such a policy to Ireland will always outweigh any corresponding benefits. This arises because the gains from the policy, while very high, are non-exclusive. The situation is similar in many respects to the problem of paying for public goods. If Ireland were to adopt a policy of importing from Ghana in the hope of higher future exports the rest of the world would free-ride and reap most of the benefits, to Ireland's cost.

Although it appears to be in the interest of the rest of the world to import from Ghana as they will always be better off importing that country's produce then protecting against it, the rest of the world will not pursue such a policy because it is not a single actor - rather, it consists of a large number of economies, each of which faces a payoff matrix roughly similar to Ireland's. For example, if we take the case of the United States, which accounts for 40% of Ghana's exports, Table 2 shows that it would never be in America's interests to import from Ghana.

In contrast, if we look at the OECD countries as a group, the results are very different. Assume, for example, that the OECD expects to import 70% of Ghana's future exports.

Table 3

In this case, it is in the OECD's interest to import from Ghana, because the gains (measured in terms of extra exports to Ghana) outweigh the costs.

This analysis shows that it is unlikely to be in the interest of any small economy to import from developing countries in the hope that they will be able to export more in future, because only a small proportion of the gains from such a policy are likely to accrue to the country which bears the cost of this policy. If a large group of countries (such as the OECD) pursues such a policy, however, it is more likely to be to their advantage.

It is important to note that our expected share of the rise in Ghana's imports will not simply correspond to our current share of imports to Ghana. A number of other factors need to be taken into account.

The most important of these factors is the income elasticity of Ghana's demand for goods produced in Ireland. Our success in building trade links with companies in Ghana will also have an effect. Porter's Diamond is useful here in attempting to calculate the proportion of Ghana's imports sourced in Ireland.

The relatively recent emergence of Regional Trade Blocs further complicates matters. Many development economists believe that Regional Trade Blocs offer 'better prospects for a balanced and diversified development than the current almost exclusive reliance on the very unequal trading relationships that they individually engage in with the developed nations'. Through regional trade blocs, less developed countries can avail of many of the gains from specialisation, scale economies and international competition without fear of deteriorating terms of trade or protection from developed countries. If Regional Trade Blocs become more prominent in the future, any attempt by developed countries to increase demand for their produce by importing more from developing countries is likely to fail, as countries like Ghana would be more likely to import from within their RTBs.

However, while Regional Trading Blocs and collective self-reliance were buzzwords of development economics in the 1970s, the 1991 World Development Report points out that the share of intra-union trade in developing countries' trade flows decreased practically everywhere during the 1980s.

In conclusion, economic theory suggests that, while importing from less developed countries will ultimately boost their demand for foreign goods, this will be less the case for a small developed economy than for a large developed economy or the OECD as a whole.

The hypothesis that has been developed so far in this paper will now be tested against empirical evidence. The relationship between imports from less developed countries and exports to less developed countries will be examined. The a priori expectations of this analysis, derived from the preceding theory, are that a strong correlation will exist. Furthermore, the analysis suggests that the relationship will be strongest when a large group of industrialised countries is examined; for the purposes of this paper the OECD will be taken. The relationship will also be stronger for a larger economy than for a smaller one; here, the United States and Ireland will be compared.

In order to test these predictions, three simple regressions will be performed. The first will attempt to measure the relationship between OECD imports from Africa and African imports from the OECD. The variables in the first regression are as follows:

The second model will measure the relationship between Irish imports from Ghana and Ghana's imports from Ireland. The variables are as follows:

The third regression will measure the relationship between the United States and Ghana. The variables for this regression are as follows:

The qualitative analysis in Sections I and II suggest that the relationships in Model 1 and Model 2 will be the most and least significant respectively.

The ordinary least squares estimation technique has been employed in this study. This has yielded a line of best fit corresponding to the data. The form of the model in this multiple regression is shown below:

Y = b ₀ + b ₁X₁ + b ₂X₂ +m

m represents the error term of the regression or the residual. The investigation will arrive at an estimate of the sign, size, and significance of the unknown parameters b ₀, b ₁ and b _2.

The estimation of the regression line was achieved using the SPSS econometric package, using 23 observations of annual data from 1972 to 1994. The results of the regression are as below. The lines of best fit have been estimated as:

Model 1: Y = 5151.491469 + 0.690107 X₁ + 2.159948 X₂

Model 2: Y = -2.795222 + 0.143713 X₁ + 0.002016 X₂

Model 3: Y = 34.437398 + 0.016403 X₁ + 0.254672 X₂

Model 1:

R² = 0.92575

Model 2:

R² = 0.85096

Model 3:

R² = 0.53586

The evaluation will deal with the results of the multiple regression, which will be compared with those postulated by the theory, and the derived t-statistics. The null hypothesis that there is no statistical relationship between the X and Y (H₀: b = 0) will be considered. A parameter estimate is deemed statistically significant if the t-statistic associated with it, at a particular significance level, causes one to reject the null hypothesis.

The correlation coefficient indicates that 92% of the variation in Y can be explained by the linear influence of the X variables. The estimate for b ₁ is statistically significant at the 5% significance level. However, there is quite a high level of multicollinearity present in this model, as a regression of X₁ on X₂ yielded an R² of 0.52155.

The correlation coefficient, R², indicates that 85% of the variation in Y can be explained by the linear influence of the X variables. The estimates for b ₁ and b ₂ are statistically significant at the 5% significance level. This supports the proposal that is being tested. However, it should also be noted that there is also a high level of multicollinearity present in this model - a regression of X₁ on X₂ yielded an R² of 0.65097.

The correlation coefficient indicates that 53% of the variation in Y can be explained by the linear influence of the X variables. The estimates for b ₁ and b ₂ are statistically significant at the 5% significance level. These results go against the a priori expectations of the study, as the influence of US imports from Ghana on US exports to Ghana is less than in the Irish case. There is effectively no multicollinearity present in this model as a regression of X₁ on X₂ yielded an R² of 0.02253.

The empirical evidence tends to support the hypothesis insofar as it shows a relationship between imports from developed countries and exports to developed countries, and that this link is greater for the OECD as a whole than for either of the smaller developed economies examined.

Curiously, the connection was not as strong for the United States as it was for Ireland. This went against the a priori expectations of the study because Ghana's imports from the US are about eight times its imports from Ireland. This aberration could probably be explained by an examination of the income elasticities of Ghana's demand curves for American and Irish products. It is quite possible that demand for Irish products (such as beef) is more income elastic in Ghana than demand for American products (such as high-tech manufactures).

The analyses in Sections I and II of this paper concluded that developed countries could increase less developed countries' demand for foreign goods by importing more from them. This would ease their balance of payments difficulties, encourage them to be more outward looking and possibly lead to export-led growth.

However, it was noted that this is not a viable policy from the perspective of an individual industrialised country. Any country which pursued a policy of importing from less developed countries for this motive would not necessarily reap the benefits.

The empirical study in Section III seemed to support the theoretical conclusions. A strong relationship was found to exist between OECD imports from Africa and African imports from the OECD. In accordance with the theoretical conclusions, a weaker link was found when individual OECD members were looked at. Curiously, the relationship was not as strong between the US and Ghana as that between Ireland and Ghana, even though a large proportion of Ghana's imports come from the United States. The explanation for this empirical anomaly inevitably lies with factors beyond the remit of this analysis and so the task of uncovering them is left to another study.