Until the mid-1960s, import substitution was the trade strategy followed by many developing countries. From that period on, however, more and more developing nations have looked to export-led growth for the solution to their development problems. A small group of East Asian nations, including Hong Kong and Korea, has been significantly more successful in following the export promotion strategy than the rest of the developing world. The study on which this article is based examined the experience of a particular country (Colombia) with a particular export commodity (clothing) to try to discover why this should have been the case. Price-related factors are found to have been of some importance, though not always those which might have been anticipated. Nonprice factors, too—in particular, quality control and punctuality of delivery—which are often neglected in discussions of the determinants of export success, are found to have been significant.
In 1977, Hong Kong exported US$2.9 billion of apparel, and Korea $2 billion, compared with Colombia’s sales of less than $50 million. Colombia’s clothing exports grew from less than $1 million in 1967 to $52 million in 1974, fell back to below $30 million in 1975, then rose gradually during 1976-78 without surpassing the 1974 level.
In 1975, the first recent year for which reliable data on destinations are available, half of Colombia’s clothing exports went to the United States and a further 12 per cent went to Europe. Venezuela received one quarter of these exports in the same year and the Caribbean took 9 per cent. Since 1975 there has been a dramatic change in the volume of exports to these destinations. Exports to the United States and to Europe have fallen, while exports to the more accessible markets in Venezuela and the Caribbean have increased dramatically—from about 33 per cent of Colombia’s clothing exports in 1975 to 72 per cent in the first half of 1978. This article will examine the reasons why the volume and destinations of Colombian exports changed and the implications for policy as well as for export prospects for other sectors and for other Latin American countries.
The main price-related causes of the changes in the volume and destinations of Colombian clothing exports are seen to be official exchange rate and export incentive policies; labor productivity; and the pricing of domestic fabric.
Until 1967, Colombian governments, like many others in Latin America, tended to tackle balance of payments problems by devaluing the currency, waiting until inflation had eroded the benefits of the step, and then devaluing again. This made exporting a risky business, since a businessman could never be sure what his returns in terms of local currency would be from one year to the next. In 1967, among other foreign sector reforms, a crawling peg exchange rate was introduced. During the next six years, the rate of crawl was such that the effective exchange rate rose more rapidly than local prices, or at least at an equivalent rate. This new policy encouraged exporters, and exports of clothing eventually began to rise. Then, from about 1973 onward, this trend was reversed: local prices rose more rapidly than the effective exchange rate. This fact helps to explain why Colombia’s clothing exports failed to maintain their earlier quite rapid rate of growth after the mid-1970s.
This article is based on World Bank Working Paper No. 368 Why the Emperor’s new clothes are not made in Colombia, available from the Bank.
Interestingly enough, “cheap labor” in competitive countries was found not to be the reason for Colombia’s inability to compete with the East Asian nations. On the contrary, wages in the apparel industry are by now higher in Hong Kong, for instance, than in Colombia, and Korean wages are at about Colombian levels. Costs of production per garment depend not only on hourly wages, however, but also on labor productivity. Here the East Asians have a clear advantage. Average hourly output per worker is 30 to 50 per cent higher in East Asia than in Colombia.
Differences in the abilities of management, both top-level and middle-level, seem to be of primary importance in explaining this difference in productivity, but cultural and social factors may play some role as well. The latter might include the degree to which workers are susceptible to organization and discipline, the protection that they have against being fired, and the place that money income or being seen to be “doing one’s duty” holds on their scales of values relative to the place of a relaxed or sociable work experience.
Tariff policy is another key factor why Colombian exports fell behind. In East Asia, garment exporters are assured duty-free access to top quality fabrics at world prices. Colombian garment exporters also have a similar privilege, at least in theory. Under the Vallejo Plan, an enterprise is exempt from paying tariffs and prior deposits on imports of raw materials or semi-finished goods that are to be used in producing commodities for export. However, administrative problems and delays reduce the scheme’s value. At least two weeks’ delay is incurred before an import request is approved, and at least two to four weeks more are needed to clear the goods, once landed, through customs. Given that the season in the clothing business is only 12 weeks long, these delays effectively nullify and often reverse the one clear advantage that Colombia has over East Asia in garment exporting: the ability to offer shorter lead and turnaround times. Garment exporters who wish to import from abroad fabrics that are also produced locally have to submit a letter of approval from the relevant local textile firm. In practice, exporters are understandably reluctant to request such a letter for fear that the textile firm will discriminate against them in future supplies. As a result of all these problems, Colombian garment exporters are often forced to buy fabric from domestic producers at prices 50 to 100 per cent above world levels. For cotton, the study found that four large textile firms use their oligopolistic power behind high protective walls to raise prices. For synthetics, the problem begins further back in the production chain, with plants for petrochemicals and synthetic fibers that are too small and that also enjoy high protection and produce at high unit cost. For reasons that are not entirely clear, the export processing zones (Free Zones), which ought to provide a solution to these problems, do not do so. (In these zones, inputs can be imported without having to go through customs, provided the final product is exported.)
Other price-related factors that do not help to explain the difference between Colombian and East Asian performance include government subsidies and transport costs. Government export subsidies are nonexistent in Hong Kong; in Korea they seem to be no greater than in Colombia. In transport, if, as is common, Colombians use air whereas East Asians ship by sea, it is the Colombians who have a significant advantage: they can land goods in New York for the same transport costs as East Asians, but at a saving of four weeks in transit time. The fact that transport and communications services tend to operate more frequently and more reliably in East Asia partly, but only partly, offsets this advantage.
Among the nonprice reasons for Colombia’s inability to compete with East Asia, differences in quality control and punctuality of delivery seem to be the most important. The best Colombian firms are as good on these two criteria as firms anywhere; but the less-than-satisfactory performance of a number of Colombian companies has given the country a bad name and has helped to contribute to a fall of export sales to the more competitive markets. Some managers apparently do not perceive how important quality control and punctuality are for sales to world markets. This may partly reflect the fact that quality control and punctuality norms for sales in Colombia are rather different from those in the United States and Europe; this, in turn, may partly result from the fact that Colombian garment producers have been sheltered from import competition in the domestic market. The unreliability of domestic textile suppliers (both in terms of the quality and the timing of deliveries), delays in importing inputs, and robberies in the ports aggravate the quality control and punctuality problems.
an Australian citizen, wrote this article while he was a consultant to the Development Economics Department of the World Bank. He is currently Associate Professor of Economics at Boston University (U.S.A.). His writings on economic development range from detailed analyses of specialized topics to wide-ranging reviews of development experience.
Garment buyers in the United States commonly point out that you can get anything you want in Hong Kong or Korea—any garment, in any material (cotton, wool, synthetics, fur, leather), at competitive prices, of acceptable quality, and delivered on time. In Colombia, by contrast, the range of garments and fabrics is limited, prices are generally higher, and quality and delivery are less dependable. As a result, many buyers maintain full-time purchasing offices in East Asia, but there is not a single permanent foreign buying office in Colombia that specializes in clothing. This, in turn, makes garment exporting a riskier business for Colombian clothing firms than it is for East Asians; so few Colombian firms specialize in exporting; so few foreign buyers come to the country, and the East Asia-Colombia gap continues to widen. Success breeds success; failure breeds failure.
How can exporting garments be made at least as profitable as selling the same goods at home? First of all, the effective exchange rate for clothing exports would have to be raised to offset increases in domestic costs, and the Government would have to convince private entrepreneurs that this policy would be maintained over an extended period. Protection against imported garments might be also lowered to induce domestic sellers to become more competitive.
Second, clothing exporters would need to be assured of duty-free, and speedy access to a wide range of top quality fabrics at world prices. Tariffs on imported textiles might be lowered, and the full range of fabric imports could be transferred from the licensing list to the free list. The Vallejo Plan drawback scheme might be streamlined so that approvals could be granted in 24 hours instead of two weeks; perhaps the requirement that fabric imported under this scheme be re-exported could be converted to a ton-for-ton basis, with suitable allowances for wastage. The administration of the ports and customs would need to be overhauled to ensure that, once goods arrive, they are cleared in a day or two instead of in weeks. The existing Free Zones might be studied to discover the reasons for their failure; new, improved Zones could then be established. An indirect export subsidy might be introduced; that is, domestic textile firms that sell fabric to garment makers who in turn use this cloth in their exports would be entitled to receive the export subsidy for such sales, and perhaps subsidized export credit as well. Indirect export subsidies of this type have been used successfully for certain important domestic products in East Asia.
Third, since management ability seems to be crucial in both price (labor productivity) and nonprice (quality control, punctuality) areas, the Government might try various strategies to improve the country’s management. Large numbers of top-level and middle-level managers and graduate students in business might be sent abroad for study and work experience and for exposure to U. S., European, and East Asian standards of quality control and punctuality of delivery. Foreign buyers could be invited to Colombia to give short courses for managers and would-be managers which would again emphasize the important, but often neglected, nonprice aspects of exporting.
It is difficult to rank these policy options in order of importance. Price, quality, and punctuality are all necessary conditions for export success in the clothing industry; the absence of any one of the elements of the package disqualifies the would-be exporter from serious consideration. If one set of factors had to be chosen as the most important, however, it would probably be ‘that relating to price; for if exporting were made more profitable, entrepreneurs and foreign buyers would have a strong incentive to overcome the quality control and punctuality problems on their own.
The clothing industry is similar to other industries in which developing countries have achieved export success: average wages are low, the share of wages in value added is high, economies of scale are unimportant, and marketing tends to be taken care of by buyers from developed countries.
Colombia’s exports of other manufactured goods appear to have followed a pattern over time that is similar to the experience in the clothing sector. The difference in order of magnitude between Colombia’s exports of other manufactured goods and those of Hong Kong and Korea is almost as great as for clothing. Further, Colombia’s share of total U. S. imports under the U. S. 807 offshore assembly scheme is significantly lower for all manufactured goods (it never reached half of one per cent during 1973-77) than for apparel (5 to 9 per cent during 1974-76).
There are, of course, important differences between apparel and other sectors. Perhaps the most significant one for the present discussion is that clothing is currently among the few industries in which developing country exports are subject to quota restrictions. Only in two narrowly defined product lines, however, has the lack of a quota hampered the growth of Colombia’s clothing exports in the past. Mostly, quotas have remained unfilled.
Because the distribution of quotas among exporting nations tends to be based on historical performance, Colombia’s garment quotas are only a fraction the size of those of Hong Kong and Korea. This means that, even if all obstacles to increased exports were removed tomorrow, Colombia could still look forward to only modest rates of increase in its exports of apparel to the United States and Europe. It seems at least possible that the increasingly protectionist developed countries will impose quantitative restrictions sooner or later on imports of other manufactured goods as well. These quotas, too, are likely to be distributed on the basis of past export performance. This possibility serves to underline the urgency of the need for policy reform in Colombia if exporting manufactured goods is to continue to be the central focus of the national development strategy.
Another difference between clothing and other industries is that seasons are more important and fashions change more frequently in the former. This means that short lead times and punctuality in delivering are more important in apparel than in most industries.
In spite of these differences, clothing shares many important characteristics with other industries in which developing countries have achieved export success to date. In addition, it seems that the key problems impeding Colombia’s clothing exports in the past are likely to have retarded the sales abroad of other Colombian manufactured products, too.
To what extent are the conclusions and policy implications of this study likely to be relevant to Latin American countries other than Colombia? Colombia is about as close to an “average” Latin American country as can be found. It is neither among the largest nor the smallest Latin American nations, nor is it among the most or the least industrialized; in terms of gross national product (GNP) per capita, it is poorer than many countries but it is by no means the poorest; the growth rate of its GNP per capita has been close to, if slightly below, the Latin American average for 25 years; and its distribution of wealth and income is typically skewed.
Colombia has been typical in the timing and nature of its policy changes: several other Latin American countries shifted from import substitution to export promotion strategies during the mid- to late-1960s. It has also been about average for Latin America in its degree of political and policy continuity: some countries have had more stable economic policies, but others (Argentina, Bolivia, Chile) have been much more unstable.
Many of the problems that have hampered Colombia’s exports of clothing and other manufactured goods seem likely to have been similarly important in other Latin American countries as well. In the face of sharply fluctuating World prices for natural resource-based exports and moderate to runaway domestic rates of inflation, few countries have managed to maintain competitive exchange rates, and hence constant returns in local currency to exporting, for any length of time. In most countries export subsidies have been simply grafted onto the existing import substitution structure; since domestic protection against imports remains high, the incentive to sell domestically is still greater than that to export. Many countries still hamper the access of local firms to imported inputs when domestically produced substitutes are available and the port, customs, transport, and communications problems of Colombia are by no means unique. Colombia’s low level of labor productivity in garment production is typical for Latin America—if anything, Colombian productivity may be 5 per cent above the Latin American average.
In sum, many of the problems that have impeded Colombia’s exports of clothing and other manufactured exports seem likely to have retarded the sales abroad of manufactured goods from other Latin American countries as well. Thus, other Latin American nations may be able to benefit from some of the policy suggestions made above and, in general, from the Colombian experience. F&D
Changing your address?
In order for us to change
your address, you must
return the mailing label from
this copy of the magazine.
If you cannot find the mailing
label, you must send us
both your previous address
and your new address.