Die Rolle der Parlamente in den lateinamerikanischen Regierungssystemen

Executive Summary

When the cycle of civil wars ended in Central America some ten years ago, the intense interest that the region had provoked in the outside world – particularly the United States and Europe – rapidly diminished. Among the reasons for this shift was the perception that economic development and democratic consolidation in the region could now proceed without the obstacles created by armed conflicts and the Cold War. Such optimism, it is now clear, was misplaced. Central America – despite significant advances in some areas – still lags far behind the rest of Latin America on most development indicators. This gap was reinforced in late 1998 by Hurricane Mitch, which reminded Central Americans of the vulnerability and fragility of their region and compelled the international community to reconsider the role it might play in the development effort.

This report is a product of the renewed international interest in Central America. Sponsored by the European Commission (EC) and the United States Agency for International Development (USAID) and with the institutional support of Florida International University in Miami, the Inter-American Dialogue in Washington, D.C. and the Institute of Ibero-American Studies in Hamburg, Central America 2020 outlines a regional development model for the next two decades.

Any report on Central America must begin by recognizing the significant economic, social and political progress made in the last ten years. Macroeconomic stability has once again become the norm and regional economic growth in the 1990s was positive. Exports, both inside and outside the region, have expanded and diversified. The transfer of power through the electoral process is now commonplace and demilitarization has been occurring throughout the region. Inter-state tensions may still exist, but the regional crisis of the 1980s has been left far behind.

These advances are welcome, but they are insufficient for two reasons. First, Central America has yet to reverse many accumulated problems from the past. Poverty is still widespread, the distribution of income and wealth is highly unequal, and powerful interest groups frequently block essential reforms. Health and education systems in some countries remain deficient and underfunded. The environment is extremely fragile and the pattern of development in the region cannot yet be described as sustainable. Public security has deteriorated in the face of common crime and violence, and the justice system has been slow to adapt. Democratic institutions continue to be circumscribed by limitations of citizenship and weak popular legitimacy.

Second, globalization is presenting Central America with a new set of challenges that it did not need to confront before. The liberalization of trade and capital flows throughout the world poses special problems for small countries. These problems can most effectively be addressed through a regional response. Yet, despite efforts in the 1990s to revive the Central American Common Market, regional cooperation falls far short of what is required. Investment in human capital, in particular through the educational system, has failed to adapt to the new demands of a global marketplace that puts ever-greater emphasis on training and skills. While the export sector in each country has been transformed, non-export sectors - where most people are employed - are often characterized by low productivity and low investment.

The nature of these difficulties is widely recognized inside and outside Central America. Governments, non-governmental organizations (NGOs) and international agencies have devoted considerable time and effort to these problems. The innovative character of some of these initiatives should be acknowledged; however, their focus is usually short term and lacks an integrated approach. This report, in contrast, adopts a long-term and multi-dimensional perspective in which the focus is the future of Central America – defined to include Belize and Panama - in the next two decades. That is why we have called it Central America 2020. Our recommendations are derived from a conception of development that includes sustainable economic growth, improved social welfare and expanded citizenship. Each of these dimensions is crucial for the future of the region and none should be privileged at the expense of the others.

The model of development outlined in Central America 2020 has several guiding principles. First, we emphasize regional integration as the most appropriate response to globalization (the external challenge) and the limitations of nationally based development strategies (the internal challenge). Second, we stress the need for a new approach to the management of environmental resources that is designed to increase protection of the environment from the damaging effects of current development practices and integrate the environmental dimension more fully into the development process. Third, we argue the case for substantial increases in human and social capital, emphasizing education, health, labor markets and transnational communities. Finally, we focus on the need for a mode of development that brings together state and societal actors around common objectives, with particular attention to removing barriers and expanding opportunities for the effective participation of all social groups. Each of these principles deserves further elaboration.

Regional Integration

To many, Central America already appears to be following a regional approach. The relaunch of the Central American Common Market in 1990 was greeted with much enthusiasm inside and outside the region. The rapid growth of intraregional trade in the 1990s heralded a breakthrough in the integration process. Yet it is now clear that regional integration in Central America is reproducing many of the weaknesses from the earlier phase. The export trade is concentrated in three countries (Costa Rica, El Salvador and Guatemala); regional decisions are regularly ignored; the common external tariff is undermined by bilateral treaties with third parties; and regional institutions lack credibility and resources. The Central American response has been to address these problems by seeking to widen the number of countries with which free trade might apply rather than tackling the problem directly by deepening the integration process.

We believe, however, that a deepening of regional integration is a necessary component of development on a region-wide scale. The benefits of regional integration include an increase in competitiveness, strengthening of negotiating capacity, improved access to the world’s capital markets, and much needed support for the process of democratic consolidation at the national level. These benefits can be considerable even for large countries, but are particularly relevant for small nations.

A long-run model of development for Central America must include all of the region’s countries. It is not difficult to design policies that will benefit the richer and more successful economies, but the poorest countries, Honduras and Nicaragua, now lag far behind the rest of the region in terms of income per head with little prospect of early convergence. If this gap widens, it will increase migration pressures, destabilize the region and enhance the negative perception of Central America by the outside world. Closing the gap must be central to any thinking about Central America in the next two decades.

In this report we outline a series of measures to strengthen regionalism in Central America. These include the completion of the customs union through the sharing of tariff revenues, the adoption of a monetary union based on a single currency, the reform of regional institutions, and the creation of opportunities for the increased participation of new social actors in the integration process. Other recommendations that address specific development issues, such as labor markets, migration, and crime, call for linkages between regional, national and local levels of action, thus providing a social dimension to the integration effort.

The Management of Environmental Resources

Economic development in Central America has traditionally shown little concern for the environment. The region’s natural resources, particularly its forests, have been seriously depleted. The quality of the soil has deteriorated and land yields for a number of important agricultural products have fallen. Marine resources have suffered from over-exploitation and coastal erosion, while the quality and quantity of water supplies have been put at risk by urban and agro-industrial developments. Urban areas increasingly suffer from pollution and contamination as a result of uncontrolled development and an evasion of environmental laws.

Several steps should be taken for development in Central America to become more sustainable. The first is a change in national accounting systems to reflect the rises and falls in the stock of natural resources. The second step involves the application of a system of indirect “green” taxes designed to establish a new set of relative prices that is more environmentally friendly. A third priority is speeding up the process of issuing titles and establishing rural property rights.

Central American countries also need to be sensitive to changes in the international trading system that are likely to occur in the next few years. The World Trade Organization (WTO) has struggled without much success to reconcile its mandate in the area of liberalizing international trade with its members’ interest in environmental safeguards. Many of the major disputes in international trade - from shrimp caught with nets that trap turtles to beef cattle injected with hormones - arise from the inability of the WTO to distinguish between the process and the product. It is increasingly clear that the least costly way to resolve this problem is through the use of labels on exports. Central America should seize the initiative to develop a regional system of eco-labeling to take advantage of the higher prices that consumers in developed countries are willing to pay for goods certified to have been produced in an environmentally friendly way.

The final step that Central American countries must take to prepare themselves for the prospect of international trade involves the emission and fixation of carbon dioxide (CO₂) and other greenhouse gases. The Kyoto Protocol (1997) established a Clean Development Mechanism (CDM) that will allow rich countries to meet their reductions in part through projects in poor countries that either reduce greenhouse gas emissions or increase carbon sinks (fixation). The potential of the CDM for Central America is enormous. However, much of this potential will be wasted if the Central American countries have not prepared themselves for the opportunity. The private sector in rich countries will not participate unless the projects in Central America can guarantee results. The absence of clear title to property rights will undermine schemes to increase reforestation, and extensions to national parks will be meaningless unless resources are committed by the public sector to ensure compliance by local populations.

Without a much greater degree of awareness of environmental issues by ordinary Central Americans and without a higher level of participation by grassroots organizations, the terrible damage to the environment in Central America will continue. Schools have a vital role to play in promoting understanding of the fragile nature of the Central American environment and the need to enforce measures to protect it. Increasing awareness of the links between poverty and environmental degradation is essential and the poor must be given incentives to change their practices. Grassroots organizations are a national – and, increasingly, regional – resource that can help to ensure that national laws and regional agreements are respected.

Human and Social Capital

The success of Central American development over the next two decades will be heavily contingent on the region’s ability to increase average levels of human capital (individual-level resources and capacities), while simultaneously finding ways to promote and deploy social capital (resources and capacities created through social relationships and networks) in pursuit of development objectives. In several countries this will require a concerted effort to overcome the legacy of the deficient social policies that characterized previous development models, and that have only begun to be addressed. Education and health are critical areas for human capital development because of their immediate impact on both quality of life and longer-term enhancement of productivity and growth. We recommend that government spending in these areas be at least maintained where it is relatively high and substantially increased where it is weak. Investment in the health sector can be an especially positive economic force inasmuch as it can generate a strong demand for goods and services, as well as employment opportunities for a wide range of skill levels.

Health sector reforms in the last decade have sought to promote institutional rationalization, changes in legal frameworks, decentralization, and increased involvement of the private sector and NGOs. These on the whole are positive developments. At present, attention to communicable diseases and other health problems affecting children, containment and treatment of the HIV virus, and deficiencies of pre- and post-natal maternal health care are priorities and are likely to remain so during the next decade. Over two decades, however, the relative growth of elderly population cohorts will require appropriate adjustments in health care systems as chronic and degenerative diseases assume more importance as leading health problems.

The target for increased support for education ought to be the quality of the educational experience in primary and secondary schools, with curricular emphasis on the inculcation of the learning and analytical skills needed to promote adaptability in the face of rapidly changing economic and social environments. In support of this objective, three specific areas deserve emphasis. The first is the need for a classroom environment that lends itself to higher levels of motivation and achievement for teachers and students alike. The second is the need for strengthening teacher training in the use of best pedagogical practices. Third, there is a strong need throughout the region to improve the educational system’s efficiency through the reduction of grade repetition and average years to degree.

Another component of the effort to increase human and social capital should be a strategy for upgrading the labor force to reduce vulnerability and increase capacity to respond to new employment opportunities, thereby strengthening the inclusionary quality of Central American development processes. Existing efforts have proceeded almost exclusively at the national level, embodied in reforms of national labor codes, periodic adjustments of minimum wages, and ratification of international labor accords. We contend that a more effective strategy is one that subsumes national-level measures into a regional effort to establish “labor citizenship.” Two lines of action are recommended: the promotion of employability, or initiatives to develop specific skills and competencies tailored to the exigencies of growth sectors of the economy; and the establishment of fair minimum labor standards at the regional level. The enactment of labor citizenship should also be locally organized, in recognition of the dynamics of new labor markets that operate at that level, and should not be exclusively centered on the initiatives of the state, as was the practice under past regulatory and clientelistic regimes.

Migration will remain a major factor in the shaping of Central American development through 2020. The remittance flows and social networks of transnational migrant communities constitute an important resource that can be better utilized for development purposes. Sustained by the communications and transportation innovations that have made transnationalism not only feasible but also commonplace, these networks have constituted an important motor of local development processes through the provision of access to both economic and social capital outside of formal national institutions. However, they are largely disconnected from national and regional development initiatives. A regional network of national migration councils should be established with broad participation of social sectors and constituencies. These councils could assist community-level organizations in identifying resources and practices to address local development needs, or to undertake initiatives at a higher level of aggregation. Regionally federated, these organizations would also offer a stronger link to emigrant community organizations and interests abroad.

State and Citizenship

The long-term stability and legitimacy of democratic political institutions will depend on effective mediating mechanisms by which the views and preferences of competing social actors (especially those emerging from social sectors that traditionally have been marginalized in the political arena) can be articulated and reconciled with respect to state policies. Political parties in Central America are generally performing poorly in this respect. We recommend that strategies be developed for promoting durable linkages between political parties and civil society. The principles on which these strategies are founded ought to include the encouragement of democratic values and practices within organizations (parties and interest groups alike); simultaneous attention to local, national and regional levels of organization and action; and special efforts to reach new social actors representing groups that have traditionally been excluded.

We believe that the state remains a crucial locus for developmental initiatives in Central America, but that effective action will require a different mode from that of the past: a mode that is constructed around organizational forms that bring state and societal actors together around common objectives. We offer a series of proposals - support for fiscal reform, violence prevention and expanded citizenship - as means towards a more thorough modernization of the state than has been achieved thus far and the strengthening of democratic practices in state-society relations.

The Central American countries have made substantial achievements in the reform of state structures, but the need remains for generating additional resources for public expenditure in at least three areas - social spending, infrastructure and public security. Fiscal reform is therefore needed in Central America to ensure that resources are adequate for the tasks at hand. The first priority for fiscal reform is to ensure that the low-tax countries increase their fiscal effort. In Central America the return on labor is taxed at a rate similar to the rest of Latin America, but the return on capital (interest, dividends and capital gains) is entirely untaxed in some countries. We recommend widening the tax net to include the return on capital, both outside and inside the region. The tax systems also need to adjust to changes in the structure of production. In the next two decades, the relative importance of tariff revenues may decline as a result of the fall in average tariffs and an increase in the number of countries with which Central America has free trade. Central American countries need to start shifting their tax systems towards activities that are expected to be fast-growing, most notably the service sectors.

A high priority must be assigned to strengthening the rule of law and enhancing public security. Present levels of violence and crime, combined with unresolved legacies of injustice, spell major trouble for all three development dimensions - economic, social and political. It is important that long-term strategies in this arena be formulated, debated and implemented. The reforms of police forces carried out in the 1990s will need to be sustained, while the more limited reforms of the administration of justice need to be deepened and accelerated. We do not believe, however, that shoring up state agencies charged with administering public security is a sufficient response. We therefore propose the concept of “integrated violence prevention” as a regional development strategy. This concept encompasses an emphasis on preventive policies versus purely reactive responses, a focus on violence as opposed to crime, and a design that is based on interagency, multisectoral initiatives rather than the disconnected policies presently pursued by international, state and societal actors.

Societies where citizens face obstacles or are insufficiently motivated to participate as citizens are societies that are vulnerable to the curtailment of rights and development reversals when confronted with sudden challenges. Central America’s long-term development outcomes will benefit from an expansion of citizenship that is both broader (dedicating attention to groups that have been traditionally discriminated against, such as women and indigenous populations) and deeper (moving from formal guarantees of civil and political rights to the active involvement of citizens in the exercise of those rights). Actions to incorporate gender perspectives into development policies, increase the presence of women in public and private leadership roles, and guarantee the integrity of indigenous cultures within pluricultural societies can all contribute to a strategy of increased civil society participation. Another important component of this strategy should be to reinforce the commitment to the decentralization of public administration, public services and other political structures, especially through increasing the revenue base of local government.

In preparing this report, we have benefited from the work of a team of Latin American, European and US consultants who prepared a set of studies on the following themes: globalization, regional integration, sectoral trends, labor markets, environment, migration, public security, education, democratic citizenship and state modernization. These studies incorporated the results of ten workshops held in the region during 1999, in which more than 100 specialists from the seven Central American countries participated. A full list of consultants, workshops and participants is provided in the Appendix to this report.

Central America 2020 is directed at two constituencies. The first comprises all those Central Americans who are exploring the avenues through which the region can best respond to the internal and external pressures the seven nations currently face. The second encompasses the external actors, including the sponsors of this report, who are responsible for shaping the design of cooperative assistance to the region. For both constituencies, the main priorities are often short term. However, development is a long-term process with no shortcuts. The most successful countries or regions tend to be those where a consensus has been built around long-term goals. This requires acceptance of a framework within which policies can be adopted. Our hope is that Central America 2020 will contribute towards the construction of such a framework and that our recommendations will be seen in this light.

1. Introduction

In the last 15 years, the global economy has been transformed through the emergence of new technologies, the increase in capital flows and the international recognition of intellectual property rights. These changes, almost without precedence in recorded history, have posed a challenge for all nations. Governments of even the richest and most powerful nations have felt it necessary to forge an appropriate response in order to meet the challenge of globalization. Political leaders have taken note of the prospects for improvements in social and economic conditions, but they remain fearful of the consequences of lagging behind in the struggle to prepare their countries for what lies ahead.

The problems facing the advanced countries in preparing for the challenge of globalization are even greater in the case of the developing countries. With little or no control over the international institutions that shape the agenda, without a strong position in the generation of technology and dependent on foreign capital, developing countries often are uncertain how to respond. Many political leaders in the developing world would clearly prefer a slower pace of change, while others have demagogically condemned the new agenda without offering a real alternative.

Central American leaders have begun to react to the challenges their countries face. All have recognized the opportunities that globalization can bring, while none have sought to evade their obligations. While many Central Americans are fearful of what the future holds, almost no one believes that the region can respond with the policies adopted in the past. A serious attempt has been made to address the issues in the region in the last five years and to build a consensus around a number of key policies. Even if that consensus does not yet exist, the debate has certainly begun and we are grateful for the opportunity to contribute to it.

In writing this report, we have benefited from the work of others. At the regional level, our task was made much easier through the recent publication by the United Nations Development Program (UNDP) of the Estado de la Región and by the Centro Regional de Investigación Económica y Social (CRIES) of the Enfoque estratégico centroamericano sobre reconstrucción y transformación desde la sociedad civil organizada nacional y regionalmente. We have also been helped greatly by the work of the Harvard Institute for International Development (HIID) and INCAE; although we disagree with some of their policy recommendations, their research program has been both imaginative and illuminating, leading to a qualitative increase in our understanding of Central American societies. At the national level, we have learned from the action plans developed by governments such as the recent Bases para el Plan de Nación in El Salvador. And throughout the region, research centers and universities have been experiencing a resurgence in the last ten years which has greatly enriched our understanding of the Central American reality at the local, national and regional levels.

This report is the outcome of a research project (Central America 2020) that began in 1998. The project received funding from the European Commission (EC) and the United States Agency for International Development (USAID), as well as the institutional support of Florida International University in Miami, the Inter-American Dialogue in Washington, D.C. and the Institute of Ibero-American Studies in Hamburg. In its first stage, the project held ten workshops in Central America organized around specific themes with the participation of regional specialists. The second stage was the completion and distribution to a wider audience of reports by a team of ten consultants on each of the themes. The third stage is the publication of this final report, which draws upon the work of our consultants and presents our main findings.

Our purpose in this report is to make recommendations that can lead Central America over the next two decades towards a pattern of development that includes sustainable economic growth, increased social welfare and expanded citizenship. This broad definition of development is designed to reflect the reality of a region where economic growth in the past has not been sufficient to avoid social and political tensions. Our vision is therefore inclusive, but it is also long-term and this has influenced our choice of countries. While the core countries of the region remain those that once formed the Central American Federation, we believe that globalization requires new forms of regional cooperation that will inevitably include Belize and Panama as well. Thus, our definition of Central America includes all seven countries, although we are fully aware that these seven nations do not yet constitute a region in the usual sense of the word.

2. Development in Central America in the 1990s

In the last ten years, Central America has recovered much of the ground lost during the 1980s. Although, as we shall see in the next section, this recovery still rests on weak foundations, it marks a welcome change from the stagnation or, in some cases, decline in living standards during the so-called lost decade. This recovery, furthermore, extends beyond macroeconomic performance to most social and political indicators. Advances at the national level have been followed by progress at the regional level with the development of the Sistema de Integración Centroamericana (SICA) to monitor the deepening and widening of the integration process.

The outstanding features of the last decade have been the ending of civil wars in El Salvador, Guatemala and Nicaragua, the establishment of democratic rule in Panama and respect for the electoral process in all countries of the region. If attention has now shifted more towards the weaknesses of the political party system, the absence of judicial independence and the promotion of human rights, this in itself is a measure of the advances achieved in the electoral sphere. The transfer of power from one civilian head of state to another is now commonplace, the press is much freer and the male stranglehold over the political process has been undermined with women now in evidence at all levels of the electoral process.

The recovery of the Central American economies since 1990 is made clear by the evidence in Table 1. While regional growth rates were stagnant or even negative in the 1980s, all seven countries registered positive rates of growth of Gross Domestic Product (GDP) during the 1990s. The regional rate of growth conceals considerable variation at the national level, but the divergence is not enormous: The slowest annual rate of growth in the 1990s (Honduras) is estimated at 3.1 percent[1] and the highest (Panama) at 4.7 percent. It is true that Costa Rica has enjoyed spectacular growth of GDP in the last two years. However, when the Gross National Product (GNP) is used in preference to GDP, thereby eliminating the income received by non-residents, the divergence is more modest. This is so because the growth of GDP in Costa Rica is distorted upwards by the investments made by the US multinational INTEL.

The growth rate of GDP needs to be adjusted for population. This is affected both by the natural rate of growth of population (the difference between the number of births and deaths) and by migration. The natural rate has fallen, but not by as much as in the rest of Latin America, so that the transition to a stable population is far from complete. The region as a whole has been experiencing net emigration, but intraregional migration has produced two cases - Belize and Costa Rica - where net migration has been inwards. The demographic picture is therefore complicated. The three poorest countries of the region - Guatemala, Honduras and Nicaragua - continue to have high natural rates of population growth. Emigration has reduced considerably the rate of growth of El Salvador, but immigration has increased demographic expansion in Belize and Costa Rica.

Table 1: GDP and GDP per capita in the 1990s

	Annual Growth Rate (%)		Real GDP per capita (1980 = 100)	Real GDP per capita in 2000 (1990 dollars) ^(c)
	GDP	GDP per capita	1999 ^(a)	Real GDP per capita in 2000 (1990 dollars) ^(c)
Belize	3.5	0.8	134.3 ^(b)	2.298
Costa Rica	4.1	1.2	110.6	2.316
El Salvador	4.4	2.3	113.1	1.341
Guatemala	4.2	1.5	91.3	1.035
Honduras	3.1	0.2	96.8	660
Nicaragua	3.2	0.3	63.9	523
Panama	4.7	2.8	119.5	2.880
CA7 ^(d)	4.1	1.6	104.2	1.640

(a) Estimated; (b) 1998; (c) Projected; (d) Weighted average for seven countries using 1995 GDP
Sources: derived from Zuvekas (2000); ECLAC (2000b); Inter-American Development Bank (1999) and authors’ calculations.

The result is a rate of growth of GDP per capita (see Table 1) that is very close to the rate achieved before 1980[2]. Thus, the regional economy has not yet achieved the acceleration necessary to overcome accumulated social and economic problems. Furthermore, some countries - Guatemala, Honduras and Nicaragua - have not yet surpassed the level of GDP per head recorded before the regional crisis (see Table 1). This is a striking indication of the challenges that face many of the countries in Central America.

Economic growth in the 1990s has been accompanied by a modest drop in some countries in the proportion of households living in poverty[3]. However, poverty remains widespread and the absolute number of the poor continues to grow. The rise in average living standards has also made little impact on the degree of inequality. This pattern of growth is widespread in Latin America, but the need for an improvement in income distribution is perhaps greater in Central America in view of the initial low levels of income per head and the sharp decline in living standards in the 1980s.

The economic recovery in Central America has been mainly due to the growth of the export sector. Both intra- and extra-regional exports have grown more rapidly than GDP. This has been true of all the economies in the region, although Costa Rica’s export performance has been particularly impressive. The completion of the two INTEL plants in Costa Rica has played a major part in this transformation. Exports now represent over 50 percent of GDP - an export ratio that is more than twice the Latin American average.

Just as important as the growth of the export sector has been its diversification. The traditional dependence on five primary products - coffee, bananas, sugar, cotton and beef - has declined significantly as new exports have emerged. These non-traditional exports include the output of maquila plants in export-processing zones (mainly textiles and clothing), as well as new, natural resource-intensive agro-industrial products. Service exports, mainly tourism, have also expanded rapidly. Mention has already been made of the INTEL factories in Costa Rica.

The value of intraregional trade surpassed its previous (1980) peak in 1994 and has continued to grow despite the institutional weaknesses of the Central American Common Market (CACM). This trade, previously limited to manufactured goods, has diversified and now includes a modest contribution from the agricultural sector. All five members of the CACM[4] have participated in the growth of intraregional trade, although intraregional exports are heavily concentrated in Costa Rica, El Salvador and Guatemala. Panama, and even more so Belize, still have only modest trade links with the rest of the region.

Export-led growth has been accompanied by a notable improvement in macroeconomic indicators. The superior export performance, combined with net capital inflows, has led to much greater stability in nominal exchange rates. This has made possible a reduction of annual inflation rates to single digits in most countries, with the prospect of further declines in the near future[5]. Fiscal deficits have been reduced and have ceased to be a major cause of inflation, and monetary authorities have been given increased autonomy. Central American countries are still some distance from the macroeconomic stability achieved in the 1950s and 1960s, but the performance represents a distinct improvement on the 1970s and 1980s.

Similarly, with respect to socio-political trends, the close of the 1990s signified the return of political stability after roughly a quarter-century of violent upheavals throughout much of the region. More remarkably, this achievement was not based on a return to the coerced authoritarian order that, with Costa Rica and Belize as democratic outliers, had preceded the turmoil, but rather rested on a regionwide pattern of freely contested, competitive elections. Notably, all countries in the region have experienced at least one electoral transition in which an opposition candidate has assumed office after defeating the candidate of the incumbent party in a free election[6].

The region has experienced an important process of demilitarization, marked by falling military expenditures as a proportion of GDP, declining size of standing armies and other military forces, and important advances towards the institutionalization of civilian authority over the military. Public security has also been enhanced by improvements in human rights guarantees, police forces and judicial institutions. Human rights abuses, once commonplace, have been reduced following the establishment of revised security force doctrines embracing respect for civil and political rights, and the strengthening of internal monitoring mechanisms. Fundamental police reforms in Guatemala, Honduras, El Salvador, Nicaragua and Panama have emphasized independence from military or partisan political control and greater professionalization through improved training, better pay and benefits, increased specialization and other measures. Judicial reforms, while much more limited, have resulted in somewhat greater autonomy and professionalization of judges, accompanied by steps towards modernized criminal codes and trial procedures.

A number of health and education indices have shown improvement over the last decade. Infant mortality rates have continued their long-term decline. Literacy rates have been rising, reaching between 65 and 80 percent of the population in the mid-1990s (except in Costa Rica and Panama, where literacy exceeds 90 percent). More impressive, however, is the increasing coverage of the education system, as measured in the percentage of school-age children enrolled and greater female participation[7].

The flow of remittances from emigrants outside Central America, which has averaged $1 billion per year in the 1990s, has remained at high levels and will probably continue to do so over the next two decades[8]. El Salvador and Guatemala have been the primary beneficiaries, but the volume has been growing in other countries of the region as well. Remittances are not only an important source of foreign exchange, but also have been converted into a resource that within local communities has helped to alleviate poverty, promote self-employment and encourage investment in human capital.

These achievements at the economic, social and political levels deserve recognition. They demonstrate that Central America has advanced considerably since the regional crisis of the 1980s. They have been made possible by cooperation between different social and political actors within the region along with support received from outside. However, achievements must be measured against needs, and it is by no means clear that the advances in Central America in the last ten years are sufficient in light of the region’s urgent priorities. It is to this point that we now turn.

3. The Limits to Recovery

The combination of democratic consolidation and economic recovery in Central America in the last ten years marks an important advance in the region’s development. However, there are still many weaknesses in the pattern of development and doubts with regard to its long-run sustainability. This section draws attention to some of the major limitations the region currently faces in its effort to accelerate economic and social progress.

First, the countries of Central America do not yet constitute a region in anything other than a geographic sense. Despite the relaunch of the CACM in 1990, the creation of SICA in 1991 and the participation of Belize and Panama in some regional institutions, Central America has not yet reaped the benefits that should be available to a regional actor. The countries do not speak with one voice in international fora, there has been no serious effort to exploit the cost savings available to a region, and Central America has not used its geographical advantage to improve its terms of trade[9].

Under these circumstances, the countries of Central America are still not perceived as a region by the rest of the world. Because the international capital market does not recognize Central America as a region, the inflow of capital is restricted. Many foreign investors continue to make their calculations based on the opportunities offered by a single nation as a basis for export to other countries outside Central America. Little new investment has been aimed at the regional market, in contrast to the situation in the 1960s.

This may seem a harsh judgment in view of the investment in regional infrastructure - telecommunications, transport and electricity - and the interest of Central America in signing free trade agreements with other countries. However, the widening of the scope for free trade - with countries as diverse as Mexico, Chile and the Dominican Republic - is to a large extent being used as a substitute for the integration of the region rather than as a complement. The benefits of regionalism cannot be maximized unless the countries of Central America are prepared to take the necessary steps to deepen rather than widen the integration process. Widening alone is not enough.

The second limitation on the region’s recovery is the continuing degradation of the environment. Deforestation continues at a pace that threatens both human development and biodiversity; the terrible impact of Hurricane Mitch in Honduras and Nicaragua in 1998 can be attributed in part to the deterioration of the environment and the lack of forest cover. The traditional emphasis on natural resource-intensive exports, coupled with the growth of populations that rely heavily on charcoal for energy, continues to create incentives for the destruction of the forests at a time when international concerns over global warming may be raising the value of protected forests throughout the world[10]. Environmental legislation has recently been enacted to address these questions, in some cases under the auspices of ALIDES[11], but the standards set have often been inadequate and enforcement efforts have typically been unsatisfactory.

Intense agroindustrial development has created other problems. The indiscriminate use of chemicals and pesticides has made yields dependent on ever-greater applications of these same inputs. The collapse of the cotton industry in Central America was in part a man-made environmental disaster. There have been serious problems in other branches of agro-industry as well. Not only has the health of many workers been affected, but neighboring populations have also suffered from the deterioration in the quality of air and water supplies.

Marine resources have suffered from over-exploitation in the region’s territorial waters and the destruction of mangroves. The run-off of chemicals and pesticides into the river systems has had a negative effect on both fresh- and saltwater fish stocks. All of this has had a damaging impact on the coral reefs, although the main problem has been global warming (for which the developed countries are mainly responsible). At a time when all Central American countries are keen to promote ecotourism, the failure to protect marine resources is a shortsighted policy with serious long-term implications.

The growth of urban populations in Central America has made improved management of water resources a matter of urgent priority. The level of the water table in many areas is falling and the costs of providing clean water have been rising. The water consumed by urban communities in one country often has its source in other countries, giving rise to the possibility of inter-state friction over the control of water resources in the future. As in many parts of the world, access to clean water will be a critical issue in Central America in the years to come[12].

The third limit on recovery is the vulnerability to events outside the region. Small, open economies, such as those in Central America, suffer disproportionately from the impact of external shocks, both positive and negative. The economic history of the region has demonstrated on numerous occasions the exposure of each economy to events over which the countries have no control. The integration of markets outside Central America through globalization has rendered this vulnerability even greater. New technologies and improved communications have reduced the time lag between external shocks and their local manifestation.

As Central American countries have expanded and diversified their exports, their openness has increased. As a result, the region is now more - not less - vulnerable to external events. A change in the US tariff codes for textile and clothing imports, for example, would have major consequences for the economies of the region; a rise in US interest rates could lead to a significant outflow of capital; a natural disaster could affect tourism for many years to come. Central America will have to live with the consequences of greater openness, but needs to find new ways of offsetting the damaging impact of negative shocks.

The fourth problem is low productivity and low investment in small and medium-size enterprises (SMEs), found predominantly in the non-export sector[13]. Despite all the changes of the last ten years, the non-export sector remains the most important part of the Central American economy. It contains a number of large and highly capitalized firms (e.g., in telecommunications), but it is also the home of many SMEs, for which the challenge of globalization is particularly severe. SMEs in the non-export sector are in general poorly integrated with the rest of the economy and have received little attention. While the export sector has thrived in the 1990s, expanding output and introducing new goods and services, the non-export sector continues to be held back by a vicious circle of low productivity and low investment. The SMEs in this sector, while accounting for a high proportion of employment, have been unable to reap the benefits from the rise in productivity and improved technology in the export sector.

The biggest obstacle faced by SMEs is the absence of low-cost finance. Financial institutions in each Central American country continue to apply huge spreads between lending and borrowing rates, and the real rate of interest on borrowing remains prohibitively high for many firms in the non-export sector. Compared with large firms, SMEs have lower entry barriers and modest profit margins. These are insufficient to generate the cash flows needed to finance expansion, making it difficult for such firms to grow through new investments. Too small to be vertically integrated, SMEs have to depend on inputs - such as transport - purchased from the formal sector, where prices are often inflated by oligopolistic practices.

The fifth limitation is the widespread nature of poverty in almost all the countries of the region. In no other part of Latin America does poverty affect such a high proportion of the population. Poverty restricts economic growth, generates social tensions and undermines efforts to promote political participation. The lack of purchasing power among a large sector of the population reduces the size of the internal market. This is true both at the national and regional level. A market of 35 million inhabitants - large enough to exploit economies of scale in many activities - shrinks to perhaps ten million when the limited purchasing power of the poor is taken into account. Both foreign and domestic investors therefore look to exports rather than regional sales to generate growth in earnings.

Poverty is a negative factor in the perception of the region by foreign investors. The lack of skills associated with those living in poverty adds to the costs of training for employers, while the public sector lacks the resources needed to complete the task that formal education failed to achieve. Efforts by the state to design programs intended to help the poor are often thwarted by the difficulty of reaching communities that are isolated and illiterate. No model of development in Central America can be judged a success until it has made deep inroads into poverty.

Sixth, despite some noteworthy advances in the region, the overall profile of Central American health and educational systems reveals major deficiencies as well as substantial inequalities among the countries of the region. In 1998, 29 percent of Central Americans still had no access to potable water, and roughly the same number had no regular access to health care[14]. These figures largely reflect the deprivation of rural areas, particularly in El Salvador, Guatemala and Nicaragua. Indigenous communities throughout the region are especially affected. Finally, even as infant mortality rates have been declining, maternal health indicators still show a pattern of neglect, with only Costa Rica escaping the regional trend (see Table 2)[15].

Table 2
Selected Maternal Health Indicators in Central America, 1995-97

Country	Attendance by Trained Personnel*		Maternal Mortality Rate** (per 100.000 live births)
Country	Prenatal Care (percent)	Births (percent)	Maternal Mortality Rate** (per 100.000 live births)
Belize	95	80	139
Costa Rica	92	97	29
El Salvador	56	62	60
Guatemala	54	35	190
Honduras	84	54	110
Nicaragua	87	87	124
Panama	89	86	84

* Figures for 1995.
** 1997.

Source: Pan American Health Organization (2000).

Deficiencies in health care are not just a social problem. They also impact negatively on labor productivity and macroeconomic performance. The low levels of productivity in Central America in comparison to other parts of the world are due to many factors, but poor health care is certainly one of them. Any attempt to raise productivity in Central America will have to take into account the quality of health services and look for new ways of delivering health care to vulnerable sectors of the population.

Turning to education, as of the 1990s only Costa Rica and Panama had achieved high literacy rates. In the other five countries, more than 20 percent of the population remained functionally illiterate. Primary school enrollment rates, which might otherwise herald the achievement of full literacy, were 85 percent or less of the respective age cohort in Guatemala, Honduras and Nicaragua in 1996-97. The situation for secondary education is much more serious and generalized in the region. Enrollment rates range from a paltry 19.6 percent in Guatemala to 67.1 percent in Panama (see Table 3). Throughout Central America, moreover, the public school systems are relatively inefficient, as shown by high rates of grade repetition and time to graduation; even in Panama and Costa Rica, students average more than seven years to complete six grades. Another set of problems applies to the teaching profession. In general, teachers are poorly paid, insufficiently trained and inadequately supported with pedagogical resources[16]. Finally, the region’s systems of higher education, greatly disrupted during the conflicts of the 1970s and 1980s, are now characterized by deteriorated, inefficient public universities amidst the proliferation of small private institutions, many of dubious academic quality.

Table 3
Selected Basic Educational Characteristics in Central America, 1996-7

Country	Net School Enrollment Rate		Grade Repetition Rate (percent)	Average Years to Complete Primary Grades
Country	Primary (percent of age cohort)	Secondary (percent of age cohort	Grade Repetition Rate (percent)	Average Years to Complete Primary Grades
Belize	-	-	10.3	-
Costa Rica	101.7	57.7	10.1	7.7
El Salvador	91.4	59.2	4.1	7.4
Guatemala	69.3	19.6	14.5	9.3
Honduras	85.4	26.3	9.7	-
Nicaragua	73.5	29.1	12.6	10.3
Panama	95.2	67.1	8.5	7.1

Sources: Estado de la Región (1999), Walter (2000)

A seventh problem is violence. In the 1990s, Central America was beset by a different kind of violence than that which had characterized the political strife of preceding decades. Seemingly without warning, common crime accompanied by violent aggression appeared to surge across the region. While comparable data on crime and violence are difficult to assemble cross-nationally, available statistics reveal a disturbing regional trend, albeit with significant country variations. A five-country comparison of homicide rates during the 1990s shows extraordinarily high figures for El Salvador, Guatemala and Honduras, with much lower rates for Nicaragua and Costa Rica; nonetheless, even in Costa Rica the homicide rate increased by more than 40 percent between 1991 and 1998 (see Table 4)[17]. Other manifestations of the problem include kidnappings, armed robberies, youth gang violence and crimes against property. Public opinion surveys across the region place crime and violence at or near the top of the problems that most concern citizens.

The sources of violence in Central America are numerous. One set of factors derives from the wars of the 1980s, including the widespread availability of arms, the demobilization of army and guerrilla combatants with little experience other than that of warfare, the lessened deterrence of police and legal systems undergoing much needed reforms, and a general socialization towards violence as a means of resolving disputes. Other contributing factors are the favorable location and permeability of Central America for international organized crime dealing in illegal drugs, illegal migration, arms trafficking and auto theft; highly unequal patterns of wealth and income distribution; and deficient judicial systems that tend to produce impunity more often than correctional justice[18]. Whatever the reasons, however, there can be little doubt that the rise of public insecurity is exerting a strong downward pressure on Central American development through direct human and financial costs, disincentives for investments and other economic transactions dependent on the rule of law, and the undermining of progress towards the consolidation of democratic political systems, among other effects.

Table 4

Violent Deaths in Central America, 1991-98*

	1991	1992	1993	1994	1995	1996	1997	1998
Costa Rica	4.1 (132)	4.9 (160)	4.8 (160)	5.3 (182)	5.2 (184)	5.2 (189)	5.7 (210)	5.8 (222)
El Salvador	-	-	-	164.5 (9.135)	149.7 (8.485)	139.0 (8.047)	145.1 (8.573)	82.4 (4.970)
Guatemala	63.7	67.5	63.5	69.8	68.3	66.9	75.2	76.9
Honduras	-	-	-	38.2 (2.202)	36.8 (2.192)	37.2 (2.287)	41.0 (2.520)	40.7 (2.505)
Nicaragua	18.3 (732)	20.0 (826)	17.8 (762)	16.5 (725)	15.6 (707)	15.6 (662)	15.0 (679)	13.3 (639)

Deaths per 100.000 population; total deaths in parentheses;
Source: Call (2000).

An eighth problem, and related to many of the first seven, is the fragility of democratic political institutions[19]. Increasing levels of voter abstention from national elections, even in Costa Rica, are one prominent symptom of distress. The legislative and judicial branches of government typically remain weakly developed relative to the executive branch, preserving a pattern of centralized presidential authority with a weak separation of powers. Political party systems have weakened in their capacity to provide effective mediation between civil society and the state. Informal political practices often circumvent the established procedures of formal institutions, in effect reproducing some of the exclusionary characteristics of past authoritarian regimes. It must be noted that these difficulties are by no means unique to Central America at the outset of the 21^st century, but in countries where electoral democracy is a relatively recent political achievement, the threat posed to the legitimacy of institutions is a severe one.

4. The Challenge of Globalization

4.1. The External Environment

No serious analysis of Central America over the long run is possible without consideration of the external environment. Throughout the independence period, the countries of the region have been deeply affected - economically and politically - by their relationship with the outside world. The region is vulnerable to external shocks - some positive, some negative - and the intellectual climate in Central America is very sensitive to the main currents in the United States and Europe.

In the next two decades, the dominant issue with which the region has to come to terms is globalization. This well-worn word refers to integration across the world of product and factor markets through the movement of goods, services, capital and even labor. Globalization, in this sense, has been under way since the Second World War, but it experienced a qualitative change with the liberalization of international capital markets in the 1980s. The increased flows of capital in the last ten to 15 years - direct and portfolio - have done more to break down national barriers than the liberalization of international trade since the creation of GATT in 1947.

Globalization poses a challenge for small, vulnerable countries such as those in Central America. A country that turns its back on globalization, restricting the import of goods through high tariffs and imposing strict capital controls, will lose out on the efficiency gains that increased trade and investment can bring. A country that embraces globalization without reserve, eliminating all restrictions on the current and capital account of the balance of payments, will face severe problems of adjustment and runs the risk of external shocks through a reversal of capital flows. It is not an easy dilemma to resolve.

In the wake of the suspension of the world trade talks in Seattle in November 1999, some observers have argued that globalization has reached its limit. They cite the hostile reaction of many developing countries to further liberalization, the pressure from trade unions to link the growth of trade to the adoption of core labor standards by the World Trade Organization, and the strong lobbying by non-governmental organizations (NGOs) in favor of environmental protection. We do not share this view. The failure of Seattle had much more to do with the absence of a common agenda between the United States and the European Union (EU) over the scope of a new round and does not imply that globalization has reached its limit. On the contrary, many studies suggest that globalization has still not reached the extent it had on the eve of the First World War, since trade and capital flows are smaller in relation to GDP than they were one hundred years ago[20].

Our view is that trade will continue to grow faster than GDP, as it has done in almost every year since the Second World War, and that the WTO will eventually recover the ground lost in Seattle[21]. However, it is important to remember that globalization is to some extent independent of the WTO. The surge in capital flows owes little or nothing to the WTO, and many trade liberalization initiatives are a result of regional agreements that are not negotiated under the auspices of this organization. While the WTO process is in limbo, these regional initiatives will gather strength. This trend is likely to be of some relevance for the countries of Central America in light of the negotiations to create a Free Trade Area of the Americas (FTAA).

The FTAA is supposed to begin in 2005, but the outlook does not look promising. The absence of fast-track authorization from the US Congress has meant that the negotiations lack urgency. Meanwhile, Brazil is pushing hard to promote its own version of hemispheric integration through a South American Free Trade Area (SAFTA). If the FTAA fails to materialize, the deepening of integration in North America - defined to include all countries north of Colombia - may prove to be a more attractive option for the new US administration that takes office in January 2001. The United States conducts most of its trade in Latin America with these countries (50 per cent is with Mexico alone)[22] and this is likely to be reinforced by the recent extension of the Caribbean Basin Initiative (CBI). This will extend further the close trade links between Central America and the United States, even if the CBI extension falls short of the “NAFTA parity” sought by the region’s states. Finally, Vicente Fox - the president-elect of Mexico - has stated that the deepening of NAFTA will be a priority when he takes office in December 2000.

Central American countries need to position themselves to take advantage of these changes in the next two decades. Not all of the changes will necessarily be favorable. Increased access by other countries to the US market will erode Central America’s privileges under the Caribbean Basin Initiative. The widening of the EU from 15 to nearly 30 countries may reduce the advantages of the Cooperation Agreement between the two regions[23]. The ending of the Multi-Fiber Arrangement in 2005, although it marks an important victory for developing countries as a whole, raises awkward questions about the future of textile and clothing exports from Central America in the next few years[24].

Central America must also prepare itself for a world in which the number of national currencies is set to shrink. The launch in January 1999 of the euro, which will completely replace 11 national currencies with a single unit of exchange in 2002, has meant that more than 95 percent of international bonds are now denominated in either dollars or euros. Capital account liberalization forces countries to reconsider exchange rate policies and eventually the nature of their currency regimes. In the next two decades, many national currencies will disappear as countries either adopt the dollar (e.g. Ecuador), the euro (e.g. Estonia) or regional currencies. Central America will not be immune from these pressures.

4.2. Population

At the heart of development in Central America are its people – the human capital. It is the populations of Central America that are supposed to be served by economic activity, although this has often not been the case, as the high levels of poverty and low rankings on the Human Development Index demonstrate. In the next two decades, the demographic profiles of the Central American countries are set to change significantly. This has to be taken into account in any model of development.

In the next twenty years (see Table 5), the rate of growth of population in all Central American countries is projected to fall below two percent per year and in one case (Panama) below one percent. This at first sight seems encouraging, but it is clear from Table 5 that lower population growth will still lead to a major demographic expansion in the region. Faster reductions are desirable in order to increase the capacity of governments to make inroads into the social deficit inherited from the past and provide adequately for the growing population of elderly people. Much can be learned from the fall in the birth rate in other Latin American countries, where increased educational and employment opportunities for young women in particular have been accompanied by a drop in fertility. However, even the modest projected decline in the rate of population growth means that Central American governments now have a better opportunity for people-centered development than they have had for several decades.

The task will still be difficult. The growth of the labor force in the next two decades will reflect population growth in the 1980s and 1990s and will therefore be higher than population growth in the next two decades. A balance will have to be struck between the needs of future cohorts, addressed primarily through educational spending, and the needs of the present labor force, for whom job creation is all important. Even if governments use all available resources to address future needs, relying on the private sector to create employment today, there will be limits on what can be done.

Table 5
Central America’s Demographic Profile, 2000-20.

	Estimated Population (‘000)		Annual Rate of Growth of Population (%)
	2000	2020	2000-5	2005-10	2010-15	2015-20
Belize	242	348	2.21	1.99	1.56	1.56
Costa Rica	4.023	5.592	2.03	1.73	1.49	1.33
El Salvador	6.276	8.534	1.82	1.58	1.39	1.35
Guatemala	11.385	18.123	2.58	2.44	2.26	2.01
Honduras	6.485	9.865	2.49	2.20	1.95	1.74
Nicaragua	5.074	7.997	2.67	2.37	2.15	1.90
Panama	2.886	3.622	1.43	1.26	1.10	0.97
CA7	36.372	54.081	2.29	2.06	1.86	1.67

Sources: CELADE (1999) and ECLAC (1999).

For this reason, we expect migration to be an important part of the demographic picture in the next two decades[25]. International migration now has two dimensions in Central America: the net flow to countries outside the region (primarily the United States) and the net flows within the region. Both dimensions will remain important, as we discuss in greater detail below.

4.3. Poverty

Two decades is a long period in terms of socio-economic development and it is as well to remember what can be achieved under the right circumstances. South Korea, for example, was transformed in the twenty years after 1960 from a poor agricultural economy to a major industrial power. Much the same was true of Taiwan. The People’s Republic of China, at the time of the market-friendly reforms of the late 1970s, was one of the poorest countries in the world, barely able to feed itself and with a ratio of exports to GDP of less than five percent. Today, China is one of the fastest-growing countries of the world with exports to rival many EU countries and an income per capita that has doubled in each of the last three decades[26].

We do not believe that Asian-style rates of growth of GDP are feasible in Central America in view of the present state of human and capital resources, nor - in environmental terms - are they desirable. Our most optimistic scenario allows for an annual rate of growth of GDP of six percent over the next two decades[27]. This would be a massive achievement in the context of recent experience. It implies that GDP per capita would more than double by 2020 compared with its level in 1997 and take the regional average up to the level currently enjoyed by Chile.

This high growth scenario implies a significant change in the structure of production, with agriculture (see Table 6) declining from 17.9 percent of GDP in 1997 to 12.1 percent in 2020. This is still above the average for Latin America as a whole today, but is consistent with the tendency all over the world for agriculture’s share to decline with rising incomes. The winner from agriculture’s relative decline, however, would not be the manufacturing sector, which -according to our simulations - would remain at 16 percent (i.e. growing in line with GDP). Instead, the big winner would be the service sectors, whose share of GDP would rise from 50 percent in 1997 to 58.4 percent in 2020. Although agriculture and manufacturing will still account for most exports, the service sectors will be increasingly responsible for the creation of employment. This has major implications for many public policies, including education and enterprise creation.

It is also important that growth favor the poorer countries - particularly Honduras and Nicaragua – in order to close the gap in living standards in Central America. As the experience of poorer countries in the European Union has shown, this does not mean that richer countries have to grow more slowly. Further divergence in income per capita in Central America will lead to an increase in intraregional migration, generating regional instability and undermining efforts to promote regional integration. This is in no country’s interest.

Table 6
The Structure of Production in 1997 and 2020 (% of GDP)

	1997	2020 Scenarios
	1997	Low	Base	High
Aggregate GDP	100.0	100.0	100.0	100.0
Agriculture	17.9	16.0	13.7	12.1
Manufacturing	16.0	17.9	16.0	16.0
Mining/Utilities/ Construction	7.0	7.8	7.8	7.0
Public Administration and Defense	9.1	8.1	7.3	6.5
Private Services	50.0	50.2	55.2	58.4

Source: Zuvekas (2000), Table 5, p.52

Closing the gap in living standards is a major challenge and one that in our view is impossible without a greater role for regional integration. In the European Union, convergence between the richest and poorest countries has been due to a series of regional instruments that Central America would find too costly to reproduce. Yet, options are available. Regional and international financial institutions need to give some weight to the need for convergence in their lending policies; donor nations need to consider it in their trade – not just their aid - policies; and Central American countries need to develop mechanisms that favor the poorer countries over the long term. We return to this point below in our discussion of regional integration.

A long-run approach to development in Central America cannot ignore the evils of poverty, income distribution and environmental degradation. These are problems for which there are no short-term solutions. Instead, development strategies are needed that can reasonably be expected to improve the situation over the next two decades. “Trickle-down” is not an option. It has not worked before in Central America and it is unlikely to work in the future. In Chile, a country where poverty has fallen substantially in the last 15 years, successive governments have not relied exclusively on fast economic growth; instead, social deprivation has been tackled through imaginative social policies targeted at the poorest. Despite the rhetoric of neoliberalism, Chile has one of the highest ratios of tax revenue to GDP to support these programs[28].

Most governments in Central America face a daunting task in designing public policies to reduce poverty, improve income distribution and end environmental degradation. The modernization of the state has lagged behind the modernization of the private formal sector[29]. The tax base is limited and evasion is common. Below a small and well-trained elite of public sector bureaucrats is a poorly trained and badly paid mass of middle-level functionaries responsible for implementing public policies. In the next two decades this has to change. The state cannot carry out its unavoidable functions if it is starved of resources and unable to attract its fair share of the most talented members of the labor force. Imaginative solutions will need to be found to this problem, including – if necessary – the pooling of resources at the regional level.

No public policy is more important than human capital formation. That is why we have attached great importance to educational policy in our work. More resources will be needed, but those that are already spent needed to be disbursed more efficiently. It is a complaint heard all over the world, but it is particularly relevant in Central America where, on the one hand, expenditure per pupil is often low and, on the other, resources are wasted through repetition of grades and classes. Yet, Costa Rica has shown what can be done over the long term without devoting an unsustainable share of public spending to education. The country is now reaping the rewards of having given a high priority to human capital formation in the past.

5. The Harvard-INCAE Project

Several projects in the 1990s have focused on the limitations of Central American development and have made recommendations for the adoption of different policies. In preparing this report, we have been conscious of those that have preceded ours. Many of these reports share common ground; however, there are also differences of emphasis. In this section we highlight one such approach, the Harvard-INCAE project[30], and explain the points of congruence before outlining the distinctive features of our model of development. We have chosen Harvard-INCAE because it is the best-known project and has received support at the highest level.

The Harvard-INCAE project draws its inspiration from the work of Michael Porter and, in particular, his 1990 book The Competitive Advantage of Nations. In this monumental work, drawing on the experience of the advanced capitalist countries together with South Korea and Singapore, Porter outlines a stages of growth theory in which nations proceed from factor-driven growth to investment-driven growth to innovation-driven growth. Each stage of growth is assumed to be superior to the last, and at each stage four considerations determine the degree of success or failure: 1) the quality and quantity of factors of production; 2) demand conditions in the home market; 3) the presence of related and supporting industries (usually known as clusters); and 4) firm strategy, structure and rivalry.

These four considerations are interdependent, but public policies have an influence on the relationship between them and can contribute to the success or failure of nations in each stage. The state therefore has an important role to play even if the main actors are firms and most decisions are resolved through the marketplace.

Porter rejects traditional theories of comparative advantage and argues in favor of a different approach in which firms - and nations - create competitive advantages through investment and innovation. It is this that allows nations to move from one stage of growth to another and, specifically, to escape from the factor-driven stage, which Porter regards as inferior to the other two stages. At the core of this approach is the concept of clusters, defined by Porter (in 1998) as “geographic concentrations of interconnected companies and institutions in a particular field”[31].

The Harvard-INCAE project adapts this intellectual framework to Central America, where it argues that the goal is the transformation of factor-driven growth into investment-driven growth through the development of four clusters: tourism; high-value agribusiness; textiles and clothing; and the manufacture of electronic components and services related to computers and software.

In order to promote these four clusters, the project identifies needed policy reforms in five areas (business competitiveness; environment; governance; legal reform; and macroeconomic reform) together with four priorities (a Central American logistical corridor; strengthening of the financial system; tourism sustainability certificates; and aggressive insertion in the market for clean development under the Kyoto Protocol). The project then concludes with a set of goals or targets for the next two decades, including an acceleration in the annual rate of growth of GDP per capita to five percent and a reduction of poverty to less than 15 percent of households.

The Harvard-INCAE project has many positive features. It marks the first time in forty years, since the launch of the Central American Common Market (CACM), that a regional project has been developed with the support of all governments[32]. Many of the recommendations are sensible and much of the research that has been commissioned is of a high quality. It has led to a serious debate about the growth model in the region and has generated interest in Central America outside the region.

5.1. Stages of Growth

Central America is assumed by Harvard-INCAE to be in a factor-driven stage of growth, in which factor endowments - defined broadly to include natural resources and infrastructure as well as land, labor and capital - are harnessed to generate increases in output without leading to significant increases in productivity. Factor-driven growth is described as if it implies low rates of investment and is therefore perceived as inferior to investment-driven growth. However, this distinction is more apparent than real. In the last 15 years, the Chilean economy has been the most successful in Latin America and has maintained high rates of investment, but Chilean growth has been factor-driven and based on natural resources. Thus, factor-driven growth can also be investment-driven.

Porter, and by implication, the Harvard-INCAE project, objects to factor-driven growth on the grounds that the accumulation of factor inputs will not produce a significant increase in total factor productivity (TFP). As all economists recognize, TFP growth - the increase in output adjusted for the input of all factors - is crucial to development in high-income countries, where capital accumulation on its own can easily lead to diminishing returns and wasteful investment (as has happened in Japan). However, TFP growth is not so important for poor countries, where the capital-labor ratio is very low and capital - natural as well as physical - can be accumulated for many years without running into diminishing returns. In any case, in poor countries capital accumulation is almost invariably associated with the transfer of technology, so that TFP may well be enhanced.

Central America’s growth has always been factor-driven and there is little prospect of this changing in the next two decades[33]. Indeed, traditional theories of comparative advantage, such as the Hecksher-Ohlin theorem, still apply in the Central American context, where exports tend to use intensively the factors of production in relative abundance (land and labor) and import-competing products tend to use intensively the factor of production that is relatively scarce (capital). It is therefore unlikely that Central America can escape in the next two decades from a factor-driven model of growth, nor is it necessarily desirable that it should do so.

The concept of competitiveness favored by Porter emphasizes macroeconomic conditions and micro-level decision-making by entrepreneurs. Other theories of competitiveness place greater emphasis on the need for political actors to develop strategies that remove the bottlenecks to development at the institutional and infrastructural level. We believe that this multi-dimensional approach is more appropriate in the context of Central America, where competitiveness needs to be interpreted broadly.

5.2. Clusters

Three of the four clusters promoted by the Harvard-INCAE project are in fact heavily dependent on factor endowments. Tourism, particularly the ecotourism projects favored by the project, is dependent on natural resources; high value agribusiness is dependent on land; and textiles and apparel rely on cheap labor. Only the fourth cluster (the manufacture of electronic components and services related to computers and software) is truly consistent with the stated aims of the Harvard-INCAE project.

The dependence of these activities on factor endowments is not for us a major criticism, as we have made clear above. However, we do question whether these activities are clusters. If we recall Porter’s own definition (“geographic concentrations of interconnected companies and institutions in a particular field”), it is difficult to claim that textiles and clothing are a cluster since the spectacular growth of this sector has been due to production in maquiladoras. Almost all maquiladora inputs are imported and exports are highly sensitive to the tax concessions provided under US tariff codes. Similarly, high-value agribusiness is an activity that often operates in relative isolation from the domestic market, and tourism - particularly ecotourism - is almost by definition not a cluster since it must avoid geographic concentration in order to safeguard the environment.

The fourth activity - electronic components and services related to computers and software - is certainly a cluster in Costa Rica, where INTEL has reaped benefits from the pre-existing software industry and has also helped to generate downstream operations. However, we question whether other Central American countries will be able to replicate the Costa Rican success in this area and we doubt very much whether the Costa Rican cluster will ever spill over into the rest of the region. The interest of high technology multinationals in Costa Rica is explained by the country’s educated labor force, stable political system and independent judiciary, which ensures that contracts are respected. These “endowments” are difficult to reproduce quickly elsewhere.

We also question the novelty of these four activities. Agro-industry has been the backbone of the Central American economies for centuries and an emphasis on “high value” has always been an objective. Tourism has been developing strongly in the region for many years and is now recovering from the setback during the regional crisis of the 1980s. Maquila exports have been growing strongly for at least ten years. The fourth activity - electronics and software - is new, but INTEL made the decision to invest in Costa Rica before the Harvard-INCAE project was launched. The choice of these four activities therefore appears to be a rationalization of the development model already under way in the region rather than an attempt to promote a new model.

True clusters take many years to create. Not only must many firms be involved, but also the rules under which they compete and cooperate take time to develop. It is these rules that help to drive down costs of production for all members of the cluster and lead to the sharing of information. The glass industry in Italy, for example, is a classic example of a cluster in which there is a constant process of innovation, technological change, and sensitivity to local and foreign demand conditions. It is unrealistic to expect clusters like this one to be formed in Central America in the next two decades.

The Harvard-INCAE project argues strongly in favor of regional integration in Central America and claims that the promotion of these four activities will enhance the integration process. We share the project’s commitment to integration, but we are skeptical about the impact of these activities on regional cooperation. Textiles and clothing exports from assembly operation are national activities aimed at the US market; not only do they not build links between the different countries, but they are also very sensitive to international demand conditions. With the abolition of quotas under the Multi-Fiber Arrangement in 2005, it is even possible that some of these exports will disappear as capital moves away from Central America when its quota - and perhaps tariff - advantages come to an end. High-value agribusiness may have better long-term prospects as an export activity, but there are few links at the regional level between firms in this sector. The high-technology cluster in Costa Rica, as already argued, is likely to remain isolated from the rest of the region. Only tourism can genuinely be represented as an activity that can promote regional integration through La Ruta Maya and other such initiatives[34].

5.3. The Actors

The Harvard-INCAE project places business groups at the center of its analysis in cooperation with governments and regional institutions. Civil society, NGOs, trade unions and peasant cooperatives have a very low profile in this project. Even among the business groups there is a clear preference for those firms that can contribute to the promotion of the four “clusters” outlined above.

This approach runs the risk of aggravating the social divisions that have afflicted Central American society for far too long. Furthermore, the business groups that are favored are all engaged in international trade, leading to a strong bias in favor of the export sector. Indeed, one of the objectives of the Harvard-INCAE project is to raise the ratio of trade (exports plus imports) to GDP to 150 percent by 2020[35]. This implies a degree of openness that would make the region even more vulnerable to external shocks.

In our view, growth should be inclusionary and give far more weight to the non-export sector, where the SMEs are concentrated. The problem in Central America is not so much the export sector, which has been - and continues to be - quite dynamic, with high productivity and high investment, but the non-export sector. The latter remains trapped in a vicious circle of low productivity and low investment. A model of development that ignores the non-export sector is unlikely to succeed in the Central American context.

These are the main differences between the Harvard-INCAE project and our own approach. However, there are also many points of convergence. Central America needs to increase the value of its exports in the next two decades and the Harvard-INCAE proposals for increasing competitiveness at the national level are important. The priority given to a logistical corridor is also valuable as this will create opportunities for new exports in each country and not necessarily only in the four clusters. Indeed, we think it is important for export promotion not to be too concerned with specific sectors. “Picking winners” is a dangerous exercise.

Missing from the Harvard-INCAE approach is a set of policies that will help to stimulate the sectors that are not involved in exporting to the rest of the world. Our emphasis on regional integration and SMEs goes beyond Harvard-INCAE to provide new opportunities for the non-export sector. Raising productivity in this sector, providing opportunities for exports to neighboring countries and lowering the cost of finance are also important aspects of a development model for Central America.

We recognize the valuable contributions that the Harvard-INCAE project has made in the field of sustainable development. The proposal to insert Central America in the market for clean development under the Kyoto Protocol is outstanding. We are more skeptical than Harvard-INCAE about the possibility of sustainable development in Central America in the next two decades, although we have our own ideas about how governments in the region might come closer to the ideal. However, we are convinced that sustainability is not purely a technical matter and requires a much greater degree of involvement by grassroots organizations. No amount of national legislation or inter-governmental agreements will work unless there is a high degree of awareness among ordinary Central Americans of the risks to the environment from current practices in the region.

The main focus of the Harvard-INCAE project is the business elite that is largely responsible for the promotion of non-traditional exports. The work the project has done with this group has been excellent and has helped to break down rather fatalistic attitudes towards the development of new exports outside the region. We recognize fully that there is a need to work closely with elites to achieve the goals of development. However, we are also conscious of the exclusionary nature of growth in Central America, where benefits are too often limited to a small group. Thus, we stress the need for investment in human and social capital and place a greater emphasis on participation.

These are the themes that we take up in the long-run model of development outlined below. The core of the model is a deepening of the integration process that goes beyond what is envisioned in the Harvard-INCAE project. This is followed by a section on the management of environmental resources in which we outline a set of policies that we believe will make development in the region more sustainable. We then turn our attention to human and social capital, where the emphasis is on strengthening the capacity of individuals to share more fully in the development process. We conclude with a section on state and citizenship, in which we address the need to increase participation.

6. A Long-Run Model of Development for Central America

We have spoken in general terms about the nature of development in Central America over the next two decades. It is now the moment to set forth our own ideas in more detail. These ideas are derived in part from the work of our team of consultants, but we go beyond that work in several crucial respects.

At the core of our thinking is an inclusionary model of growth that does not leave on the margin any major sector of the economy. We do not believe that development will be sustainable if priority is given only to a small number of activities. The linkages between the different branches of the Central American economies and between Central American countries are too weak to sustain this kind of approach. Of course, there will always be winners and losers in the growth process. What must be avoided is a growth model that systematically favors one sector over another.

We are therefore in favor of export-led growth, but not export growth alone. The distinction is crucial. Export-led growth implies an export sector whose net output is growing more rapidly than GDP, but where links to the non-export sector are sufficiently strong and mutually reinforcing to lead to a transfer of technology and productivity gains. Export growth alone means a rapid growth of exports that is accompanied by a rapid growth of imports and additional pressure on the import-competing sectors of the economy.

The first kind of growth is superior to the second, but the links between the export and non-export sector are still very limited in Central America. That is why a development model for the region must take the needs of the non-export sector explicitly into account. This sector sells primarily in the domestic market, but could easily be induced to expand into the regional market. It includes many small and medium-sized enterprises for whom financial and trade liberalization have so far brought few benefits.

A long-run model of development for Central America must include all countries. It is not difficult to design policies that will benefit the richer and more successful economies. What will happen to the other countries? Honduras and Nicaragua in particular now lag far behind the rest of the region in terms of income per capita (see Table 1) with little prospect of early convergence[36]. If this gap widens, it will increase migration pressures, destabilize the region and enhance the negative perception of the region by the outside world. Closing the gap must be central to any thinking about Central America in the next two decades.

Development has many dimensions. It is easy to assume that the economic dimension must take precedence over others. Yet social development is also crucial. This means that social indicators must be an integral part of how development is measured and not an optional extra. It also means that civil society needs to be properly engaged in the development process. A development model led only by the elite always runs the risk of being captured by special interests. This has been, and still is, a serious problem in Central America. Below we explore ways in which the development process can be widened to include broader participation.

Finally, development must strive to be as nearly sustainable as possible. We are skeptical of claims that development can be truly sustainable in a world of rapid growth of output and population. Sustainable development in the strict sense may have to wait until the world’s population has stabilized and new technologies have been invented. However, all countries and regions have a responsibility to seek a development model that minimizes the damage to the environment. This is particularly true of Central America, where the fragile ecosystem and the extent of biodiversity have been threatened by development in the last few decades.

6.1. Regional Integration

The importance of the nation state is shrinking in the face of globalization. This is true of both developed and developing countries. Decision-making is being transferred upwards to supra-national institutions such as the World Trade Organization and downwards to provincial and municipal governments. Even the United States, the one remaining superpower, is not immune to these centrifugal forces, despite the reservations that many US citizens have expressed on the streets and through the ballot box.

The transfer of sovereignty to supranational bodies is a voluntary act that is easier to accept politically where the states concerned exercise direct control over the institutions in question. This is one of the great attractions of regional integration schemes, since the member states - even small ones - retain influence over the decisions that have been transferred away from the nation state. The history of the European Union, now poised to increase its membership from 15 to a possible total of 29[37], is a good illustration of what can be achieved voluntarily through the pooling of sovereignty. The advance of the EU has not been easy, but at each step of the way the member states have relied on political will to overcome the numerous obstacles. In military terms there is only one superpower, but the EU is a match for the United States in many other areas.

Not all states are in a position to join regional integration schemes. Some countries are in dispute with their neighbors over territorial boundaries, mineral rights or access to water. In some cases, lack of trust and a history of suspicion make cooperation at the regional level difficult if not impossible. The countries of Central America, however, are in a privileged position[38]. The five core states - Costa Rica, El Salvador, Guatemala, Honduras and Nicaragua - formed the Central American Common Market (CACM) in the early 1960s; Panama has had an association with the CACM for many years, and even Belize has begun to cooperate in some regional fora with its Central American neighbors.

Despite this privileged position, Central America is still not a region except in the geographical sense. Intraregional trade expanded throughout the 1990s, but it accounts for less than 20 percent of exports and less than 15 percent of imports[39]. The goal of a common external tariff has still not been achieved and the regional institutions - weak and underfunded - are incapable of ensuring that executive decisions are converted into actions. Non-tariff barriers are widespread and the area is still far from being a single market, making it impossible for firms to exploit economies of scale.

These are traditional complaints[40], but they have acquired greater urgency with the arrival of globalization. The international capital markets do not perceive Central America as a region and the domestic capital markets are too small and underdeveloped to attract portfolio capital. Country risk premiums are high and no state has been able to avoid the perception that it is living in a “bad neighborhood.” The quantity and - even more important - the quality of capital flows are diminished through the failure of Central America to present itself to the outside world as a region.

6.1.1. Institutions

Every regional integration scheme requires a set of institutions to ensure effective management. There is considerable diversity in the institutional arrangements adopted by different schemes and there is no reason to believe that one set of institutions is inherently superior to another. The correct balance between supra-national and inter-governmental decision making will always be difficult. However, institutions must be capable of meeting the objectives of member states, ensuring that decisions are implemented and resolving disputes. In all three respects the institutional arrangements for regional integration in Central America fall short of what is required.

These deficiencies have long been recognized inside and outside Central America. At the Summit of Central American Presidents in 1995, the need for institutional reform was made a priority. A report prepared by a distinguished group of Central Americans[41] outlining the steps that needed to be taken was accepted by the presidents at their summit meeting in Panama in July 1997. Yet the basic weaknesses remain and the relevance of several institutions is being increasingly questioned not only by civil society, but also by the presidents themselves.

Some of this criticism is justified. It is not at all clear, for example, what role the Central American parliament is expected to play in the integration process. Furthermore, Central America has no institutional arrangements that would allow the region to speak with one voice in international negotiations. Thus, one of the main objectives of new regionalism – the ability of states to enhance their influence outside the region – is not being met. On the other hand, the hostility exhibited towards the Central American Court of Justice in some quarters is unjustified given the importance of a regional forum to settle disputes.

The first step towards improving the efficiency of regional institutions must be greater clarity about the objectives of regionalism. Governments in the region have very different ideas about the scope and purpose of integration. Two conflicting strategies have emerged. First, a consensus favors signing free trade agreements with countries outside the region, although the partner in question (e.g. the Dominican Republic) may be largely irrelevant in trade terms. Second, the governments show a preference for a multi-speed approach to integration that would allow some countries to advance more rapidly than others, even though this approach undermines many of the advantages that Central America might be able to enjoy as a region.

Defining the objectives of integration will make it easier to reform institutions. The following discussion outlines a model of integration for Central America that has many implications for institutional development. However, institutions need financing and the arrangements in Central America are far from satisfactory. Funds often arrive late, while several institutions are chronically underfunded and cannot therefore perform effectively. Thus, the second step in making institutions more efficient is a reform of the funding system. We propose a new method based on sharing the revenue from tariffs.

Sharing tariff revenues is the system used in both the Southern African Customs Union and the European Union, but it has not yet been adopted in the Americas. However, other practices in the Americas would repay careful study by Central American countries. Both CARICOM and MERCOSUR have developed effective machinery for coordinating a regional perspective in international fora. CARICOM’s Regional Negotiating Machinery (RNM) has been particularly successful in allowing small countries to “punch above their weight” in negotiations with extraregional states. We believe that Central America should explore the possibility of establishing its own version of the RNM to be used in negotiations in the CBI, the WTO and the FTAA, as well as with Mexico.

Regional integration schemes throughout the world often suffer from a lack of interest by civil society. Ordinary people all too often see integration as either irrelevant or as a threat. Even in the European Union, governments and supra-national institutions face a constant danger of running too far ahead of public opinion. The easiest way to overcome this challenge is to establish activities at the regional level with which the population at large can engage. In CARICOM, for example, the University of the West Indies links many member states through the integration of higher education. The existence of a regional cricket team is also a highly visible symbol of integration. Central American states need to support at least one regional activity – perhaps in education or popular culture – with which their electorates can easily identify.

6.1.2. Customs Union

Central America must take three steps if it is to convert a loose association of states into a single market. First, the customs union must be made a reality. A common external tariff is a necessary condition for a customs union, but it is not sufficient. Goods entering the region from outside must be free to circulate inside the region without the imposition of additional trade restrictions. At present a television imported from, say, South Korea incurs a tariff to the customs authorities of the port where it is landed, but would incur a further tariff if re-exported to another Central American country. This happens because the external tariff is still not the same in all cases and because each nation protects its tariff revenue jealously. The result is complicated rules of origin and endless delays at frontier posts as customs authorities establish whether a product qualifies for duty-free status or not.

The core countries of Central America - the five members of SIECA - are close to a common external tariff, but they are still very far from a mechanism for sharing tariff revenues. Indeed, this issue is not even on the agenda. As a result, regional institutions suffer long delays in the receipt of annual quotas from member states, which are in any case too small to allow them to carry out their functions properly.

The sharing of customs revenue is a radical step that no regional integration scheme in the Americas has yet adopted, although it is fundamental to the working of the European Union[42]. Its implementation is unavoidable - at least in part - if member states are serious about forming a single market. And a single market is an integral part of the challenge of globalization, allowing firms in the region to reap the benefits of a larger home market before launching themselves into the world market.

Central American governments are understandably nervous about embarking on such a step, and adequate safeguards would have to be built into the process of distributing resources. Thus, all states need to be sure that the revenue returned from tariff sharing would never fall below a fixed percentage of their share of total tariffs collected and that the absolute value would never be less than the amount collected in the base year. Provided that extraregional imports continue to increase for the region as a whole, these two safeguards should not be difficult to achieve.

Tariff revenue is still a large part of total tax income in Central America despite the fall in the average tariff. As Table 7 shows, tariff revenue represents approximately 15 to 30 percent of tax income for the five countries. This contrasts with the 0.3 percent of tax income that the five governments are currently contributing to regional institutions[43]. Thus, tariff revenue is far greater than is needed to fund the institutions, even taking into account their need for additional resources. This situation will continue for many years, although tariff revenue as a share of GDP and as a proportion of government revenue will decline if Central America follows the pattern of other developing countries.

States need to be sure that, after funding regional institutions, the money transferred from tariff revenue is returned in other forms. An important part of government expenditure consists of investment in regional infrastructure (e.g. roads) or the promotion of regional activities (e.g. tourism). States could also share the costs of foreign embassies and could ensure that Central America has a single voice in its relations with the World Trade Organization, the European Union, the United States and other international actors.

Table 7
Customs Revenues and Tax Income in Central America, 1993-7 (average)

	Customs Revenue as % of Tax Income	Customs Revenue as % of GDP	Tax Income as % of GDP
Costa Rica	20.5 ^(a)	3.0	14.7
El Salvador	14.7	1.7	11.3
Guatemala	19.5	1.6	8.2
Honduras	28.2 ^(a)	4.4	15.6
Nicaragua	22.6	4.6	20.5

a) Includes export taxes
Source: SIECA (1999)

It is unlikely that these expenditures will exhaust the revenue from tariffs. However, a single market in goods and services would have major implications for sea and airport development in Central America. The duplication of facilities in high-cost and inefficient points of entry and exit could lead to a process of rationalization, whereby transport services would be concentrated in a smaller number of ports (both sea and air) leading to a reduction in unit costs[44]. The benefits of this process for all Central America would be considerable, but the costs of adjustment would be eased if the states had access to a pool of resources that could be used to compensate the losers. Regional transfers could also be funded by tariff revenues disbursed at the regional level.

The pooling of revenue from tariffs would also give Central American countries a resource to promote convergence. The low levels of income in some countries - particularly Honduras and Nicaragua - act as a disincentive to regional cooperation. Central America cannot afford European-style transfers between states, but it can promote investment in regional infrastructure to favor the poorer countries.

Which regional institution would receive the revenue from the common external tariff? The institutional deficit at the center of the CACM needs to be addressed as a matter of urgency. Yet, a large part of the problem is a lack of resources. Additional funding is essential to improve the quality of the region’s institutions and an improvement in the quality of institutions will reduce the resistance to increased funding. There is therefore a problem of interdependency, which means that the institutional vacuum and the transfer of funds have to be tackled simultaneously. Whether the solution is the creation of a new supranational institution or the strengthening of existing institutions is a choice that only Central Americans can make, but the status quo is not an option. Either choice can be made to work provided that the political will exists and that the institution is accountable to the member states.

6.1.3. Monetary Union

A true customs union would bring many benefits for Central American producers and consumers, but no single market will exist without a single currency. The second step is therefore the creation of a monetary union to replace the national currencies with a single currency for the region. This bold move would do more than anything else to eliminate the non-tariff barriers that prevent Central Americans from reaping the advantages of a region and preparing themselves for the challenge of globalization.

The replacement of national currencies by a regional currency implies costs and benefits. The analysis of these costs and benefits has given rise to the concept of an optimal currency area[45]. The conditions for an optimal currency area are now well established. The benefits are associated with the elimination of transaction costs, the reduction in the scope for price discrimination, the reduction in the degree of uncertainty, and the elimination of the exchange rate risk. The costs are to be found in the difficulty of addressing macroeconomic disequilibria when the exchange rate is no longer an instrument of policy. The consensus is that the larger the share of trade between countries as a proportion of GDP, the greater the benefits. To the extent that nominal depreciation is ineffective, costs will be smaller[46].

Intraregional trade as a proportion of GDP is still modest in Central America, although it is growing (see Table 8). Furthermore, there is every reason to believe that the ratio of intraregional trade to GDP could increase significantly in the next two decades through the elimination of non-tariff barriers. On the cost side, there is strong evidence that the exchange rate is an ineffective instrument. There is a correlation between the rate of currency depreciation and the rate of price inflation[47], so that countries are forced to address macroeconomic disequilibria through other measures. In practice this is not difficult, since the labor market is flexible in most countries and migration within and from the region is a powerful adjustment tool.

Central America - at least the core five countries- therefore meets the criteria for an optimum currency area. However, additional reasons support moving towards a common currency. There is no more powerful signal to send to the rest of the world than the replacement of national currencies with a regional currency if the nations of Central America wish to be perceived as a region. The Caribbean Community (CARICOM) has already committed itself to this step and MERCOSUR members may be moving towards it despite Brazil’s initial misgivings. It would provide a huge boost to the tourist industry and be welcomed by multinational companies as well as domestic firms selling in the regional market.

Table 8
Intraregional Exports (IRE) in Central America, 1994 and 1999

	IRE 1994 ($000)	IRE 1999 ($000)	Annual Rate of Growth (%)	Share of GDP in 1994 (%)	Share of GDP in 1999 (%)
Costa Rica	285.852	612.212	16.5	3.4	5.6
El Salvador	341.892	626.000	12.9	3.6	5.2
Guatemala	474.993	774.745	10.3	3.6	4.3
Honduras	139.900	241.701	11.6	4.1	4.6
Nicaragua	83.901	139.519	10.7	4.6	6.2
CA5 ^(a)	1.326.538	2.394.176	12.5	3.7	5.0

(a) Weighted average using same weights as in Table 1

Source: derived from SIECA (2000).

A regional currency is not the same as dollarization. On the contrary, a regional currency is free to fluctuate against major international currencies, such as the dollar, providing a possible cushion against external shocks and a change in the net barter terms of trade[48]. A unit of account - the peso centroamericano - already exists and could become a unit of exchange through the conversion of national currencies at a prespecified date. From that moment, however, the peso would fluctuate against the US dollar and other international currencies so that residual inflation would not necessarily lead to overvaluation of the real exchange rate. Only when inflation had fallen throughout the region to US levels would it be realistic to expect Panama (a dollarized economy) and Belize (a semi-dollarized economy) to participate. At that point the single currency could easily be exchanged for the US dollar if that is what the member states desire. However, it is important to stress that a single currency among the core countries need not necessarily lead to dollarization.

A single regional currency has some support in Central America in view of the opportunities it presents for reducing the costs of financial transactions and merging stock markets. Dollarization finds favor among many members of the business community. However, a regional currency is resisted by all governments and dollarization by most. The reason is in part the absence of macroeconomic harmonization among the member states. Inflation rates have not yet converged, nominal interest rates are very different and fiscal policy varies enormously from one state to the next. In terms of the famous Maastricht criteria, used to determine the preparedness of European countries for entry into the euro zone, Central American states appear to be too divergent.

This divergence is more apparent than real. The difference in inflation rates is largely explained by the rate of currency depreciation. Thus, a single currency would very quickly - far more quickly than in the EU - bring about convergence of inflation rates. The nominal interest rates in the region also reflect the differences in inflation; a single monetary policy, coordinated by a Central American Central Bank with representation from each participating country, would soon lead to convergence of both nominal and real interest rates. Fiscal policy would remain unharmonized, but the European experience has shown that fiscal policy can remain a national preserve long after other policies have been harmonized at the regional level.

The value of a single currency will fluctuate in accordance with external shocks, such as a fall in the price of coffee, but not all shocks will affect each country equally. Hurricane damage, for example, tends to affect one or two countries at a time, and some countries are more dependent than others on certain primary commodities. This is another reason why the pooling of tariff revenue is so important. Regional institutions need to be able to transfer resources quickly, as loans or grants, to nations subject to adverse shocks. The building of a true customs union and the adoption of a single currency should be part of a twin strategy to build a region in Central America.

6.1.4. Regional Capital Markets

A development model for Central America needs to have at its heart the interests of consumers and producers. To many Central Americans, a customs union and a single currency appear to be arcane concepts of little relevance to their daily lives. Nothing could be further from the truth. The building a of a single market is in fact a crucial part of an inclusionary model of growth from which all sectors of society can benefit. In Central America, as in so many parts of the world, the key to the growth of output and exports may be large firms, but the key to the growth of employment and the reduction of poverty is to be found in small and medium-size enterprises (SMEs).

There are some 20.000 SMEs in the core countries of Central America, not including the micro-enterprises in the informal sector. These firms account for almost half of all jobs and their growth depends on access to credit. The cost of finance is extremely high, however, with the spread between borrowing and lending rates averaging ten percent and real interest rates among the highest in Latin America[49]. It is almost impossible for SMEs to borrow at these rates, since legitimate activities with such a high real rate of return on capital are difficult to find. Large firms face fewer constraints since they either have access to the international capital market at lower rates of interest or can rely on undistributed profits to finance growth. In any case, the borrowing rates they are offered by Central American financial institutions are usually lower than those charged to SMEs.

This bias in favor of large firms and against SMEs is a major weakness of the development model in Central America. Financial liberalization at the national level has done little to help - some argue that it has made the situation worse - and the entrance of foreign banks has not helped to reduce spreads. As a result, SMEs are unable to contribute much to the growth of the economy. Job creation is also held back, since SMEs are much more labor-intensive than large firms.

A single regional market would go a long way towards reducing the costs of borrowing for SMEs. Transaction costs would be much lower for lenders and exchange rate risk would disappear for many contracts. Financial institutions would be forced to compete and firms could borrow in pesos centroamericanos from any country. Inefficient banks would be forced to close or merge with others. The process of consolidation might be dangerous at the national level because of the risk of excessive concentration, but this is much less likely to happen at the regional level. At last, SMEs and perhaps even microenterprises would be free to participate more fully in the growth of output. This would lead to a big increase in jobs and a fall in poverty.

It is often assumed that even a regional market of 35 million people is too small to exploit economies of scale; that firms looking to export need to focus on the world market; and that Central America’s privileged access to the United States and the European Union makes exports to the rest of the region unnecessary. This may be true of large multinational companies, such as INTEL in Costa Rica, where economies of scale require investments that are much greater than even the regional market can justify. However, for most SMEs the dominant market is the domestic or national one. They lack the resources, knowledge and experience to export to the regional market - yet alone the world market. The lowering of non-tariff barriers, a single currency and cheaper finance are the ingredients needed to persuade these firms to export to neighboring countries. Some of those that do so will go on to export to the rest of the world.

The relaunch of the CACM in 1990 was greeted with much enthusiasm inside and outside the region. The rapid growth of intraregional trade in the 1990s heralded a breakthrough in the integration process. It is now clear, however, that regional integration in Central America is reproducing many of the weaknesses from the earlier phase. The export trade is concentrated in three countries (Costa Rica, El Salvador and Guatemala); regional decisions are regularly ignored; the common external tariff is undermined by bilateral treaties with third parties; and regional institutions lack credibility and resources. The Central American response has been to address these problems by seeking to widen the number of countries with which free trade might apply rather than tackling the problem directly through deepening the integration process.

These problems are common to regional integration schemes in developing countries, but other regions have been working hard to rectify them. Central America must respond appropriately if it wishes to compete as a region. The ideas we have outlined here - a true customs union, a single currency and an integrated financial market - are bold, but we believe that they offer the best chance for Central America to consolidate itself as a region and at the same time offer an opportunity for SMEs to participate more fully in the development process. For too long, development in the region has focused on those firms in high-productivity, high-growth sectors that are relatively divorced from the rest of the economy. Regional integration - not national clusters - provides the best opportunity for breaking down these barriers and creating an inclusionary model of growth that enhances competitiveness at the same time through the reduction in costs and the fall in the regional risk premium.

We have said very little so far about the social and political dimensions of the integration process. It would be a big mistake to imagine that the agenda we have outlined in this section could be implemented without a fundamental change in the participation of civil society, on the one hand, and the democratization of the integration scheme, on the other. One of the reasons why regional integration remains superficial in Central America is that the process seems remote to so many citizens. Changing this situation will require a revolution in attitudes that includes the educational system, the media and regional institutions. In this way, ordinary people will learn how regional integration works and will become more involved in the decisions that affect their lives.

6.2. The Management of Environmental Resources

Economic development in Central America has traditionally shown little concern for the environment. The definition of sustainable development given by the Brundtland Commission is “development that meets the needs of the present without compromising the ability of future generations to meet their own needs”[50]. This means that each generation must leave for its successor a stock of natural and physical capital per capita that is at least as great as the one it inherited itself. It is fair to say that according to this definition development has not been sustainable in Central America in the last few decades. The stock of natural resources, particularly the forests, has been seriously depleted and deforestation continues at an annual rate of 2.5 percent[51]. The quality of the soil has deteriorated and land yields for a number of important agricultural products have fallen. Marine resources have suffered from over-exploitation and coastal erosion, while the quality and quantity of water supplies have been put at risk by urban and agro-industrial developments. Urban areas increasingly suffer from pollution and contamination as a result of uncontrolled development and evasion of environmental laws. All of these problems have been exacerbated by rapid population growth.

Central America is prone to natural disasters from hurricanes, earthquakes and volcanic eruptions. Hurricane Mitch was the latest and most dramatic of the disasters to strike the region. Much remains to be done to mitigate the impact of natural disasters in Central America. We do not develop the topic in this report, but we are conscious of the urgency of measures to reduce the tragic consequences of such calamities. However, it is now clear that the impact of Mitch was exacerbated by the environmental damage of human activity. This substantially increased the number of deaths, displacements and destruction of property and emphasized the need for a fuller understanding of the management of environmental resources.

The main reason for environmental damage in Central America has been deforestation. The private return on forest conservation has been very low, while the private rate of return on alternative uses - cattle ranching, for example - has been much higher. In a market economy, deforestation will continue until the two rates of return are equal. However, if the social return on forest conservation is higher than the private return, this will lead to over-exploitation of the forests[52]. It is not difficult to argue that the social return is indeed higher than the private return, since the former reflects the value placed by society on greater biodiversity, water retention and the absorption of carbon dioxide (CO₂) emissions and the latter does not. A similar argument can be applied to the private and social returns from the exploitation of marine resources.

In the case of arable lands, the problem is not so much the gap between social and private benefits as the gap between social and private costs. The intensive use of inorganic fertilizers in many branches of agroindustry has led to deterioration in the quality of water supplies and raised the costs of providing clean running water for urban households. Farmers do not take these costs into consideration in choosing the level of their inputs, leading to excessive use of fertilizers and other chemicals in many cases. The shift in the pattern of demand in developed countries towards organic farm products, sold at a price premium to other agricultural products, has so far had little impact on farm practices in Central America.

6.2.1. Economics of the Environment

Several steps can and should be taken to address these issues for development in Central America to become more sustainable. The first is a change in the accounting system to reflect the rises and falls in the stock of natural resources. The United Nations, in its latest System of National Accounts, has developed a methodology that all member states are encouraged to apply. Few have done so and none in Central America. However, a number of unofficial studies show that the rate of growth of GDP has been overestimated as a result of the neglect of environmental resources. The methodology can be applied to countries in Central America and in most cases it will lead to a fall in the estimated rate of growth per capita[53].

The second step required in the region, therefore, is reform of the fiscal system to move social and private costs as well as social and private benefits more closely together. This will require the elimination of all subsidies - implicit and explicit - on those products whose use is excessive. It also involves the application of a system of indirect taxes to discourage the consumption of products whose social rate of return is below the private rate and encourage the use of goods and services whose social rate of return is higher than the private rate. This system of “green” taxes and subsidies is likely to differ significantly from the system currently in place in each country, but it need not increase or decrease the tax burden; rather, it is a redistribution of the tax burden designed to establish a new set of relative prices that is more environmentally friendly. Examples would be higher taxes for thermal energy and lower taxes for alternative sources of energy; higher taxes for inorganic chemicals and lower taxes for organic fertilizers; higher taxes for activities that are unable to dispose of their waste products efficiently and lower taxes for those that recycle the waste in environmentally friendly ways.

Altering relative prices in order to reflect more closely social costs and benefits is regarded by welfare economists as a “first-best” policy. In Central America, however, there are limits to what can be achieved by relative prices alone as a result of numerous market and government failures. One of these is the absence of clearly defined property rights in rural areas (the problem is less acute in urban centers). Occupants may not have an incentive to apply best practice since they cannot be sure that the increased value of the land will accrue to them. According to some estimates, 95 percent of titles in rural Guatemala are subject to dispute. This problem clearly has major implications for land use and the management of environmental resources.

The absence of clearly defined property rights is a major cause of the rapid rate of deforestation in Central America. Occupants who are uncertain of their legal title have an incentive to clear the land, plant crops and move on before the yield deteriorates. The absence of title will become an even more serious problem in the future if carbon emission trading (see below) begins in earnest. Only those who can demonstrate their ownership of the forests will be in a position to take advantage of any market that may arise in this area.

Thus, a third priority in Central America is speeding up the process of issuing titles and establishing rural property rights. This step is needed for many reasons. Long-term conservation measures will not be undertaken by those with insecure title. Access to bank finance will be denied to those who cannot prove ownership. And an absence of titles makes it difficult for governments to design a fair and efficient fiscal system based on property taxes. It should be stressed that many agencies - inside and outside the region - have emphasized the need for an adequate system of property rights[54], but progress has been painfully slow.

6.2.2. Trade and the Environment

Central American countries also need to be sensitive to changes in the international trading system that are likely to occur in the next few years. The WTO has struggled without much success to reconcile its mandate in the area of liberalizing international trade with its members’ interest in environmental safeguards. The WTO is built on two principles, one of which is equal treatment of national and imported products. This means that the WTO cannot take into consideration the environmental impact of the process used to make a product. It is this unwillingness to recognize the role of processes that has led to so many protests by environmentalists against the WTO.

The WTO position is understandable, since governments could easily use the manner of processing as a form of neo-protectionism. Yet, consumers rightly object when they are unable to differentiate between goods produced by different processes. Many of the major disputes in international trade - from shrimp caught with nets that trap turtles to beef cattle injected with hormones - arise from the inability of the WTO to distinguish between the process and the product. Indeed, the dispute over trade in genetically modified crops, although it raises many other issues as well, is also a controversy over the process of production.

It is increasingly clear that the least costly way to resolve this problem is through the use of labels. Thus, genetically modified (GM) crops or products whose inputs are based on GM crops could be labeled as such, giving consumers the opportunity to make an informed decision while allowing the WTO to ignore the process used in production in the resolution of trade disputes. If adopted, this system will go some of the way towards reducing international trade tensions. Similarly, labeling allows consumers to identify those products that have been certified for their environmental qualities, which is why it is often known as eco-labeling.

Labeling in general, and eco-labeling in particular, presents Central America with both opportunities and challenges. It is an opportunity, because consumers in developed countries are willing to pay a premium for goods certified to have been produced in an environmentally friendly way. In addition to this private benefit that accrues to producers, society also benefits from the adoption of more sustainable practices and the curtailment of damaging environmental practices. This is a typical win-win situation, from which all groups can gain.

However, there are also challenges. First, the coveted eco-label is only available to those who can demonstrate that their process of production is not ecologically harmful. This presupposes a degree of sophistication that may not be possible for small farmers, who typically concentrate on sales within the region rather than exports to developed countries. Second, the award of the eco-label is still a haphazard process in which public and private bodies in developed countries (including NGOs) have set up rival systems. The closest to an international standard is the ISO 14000 system, but this tends to be of interest to large enterprises with global sales.

Large Latin American countries have been able to tackle this problem through their own system of certification. The Forest Stewardship Council, for example, has accredited the Instituto de Manejo e Certificação Florestal e Agrícola (IMAFLORA) in Brazil as the body to certify that wood products meet the highest environmental standards[55]. Central America has no such organization. Thus, the fourth priority is the establishment of institutions with accreditation from developed countries capable of issuing eco-labels throughout the region.

A regional institution responsible for eco-labeling would do much to resolve the bias against small farmers inherent in the present system. It could also accelerate the switch of resources to organic farming if consumption patterns in developed countries justify the costs. In addition, the eco-label could help the region’s producers to access export markets, since the label itself is already a mark of quality. Similar institutions could be established at the regional level for fish and fish products, on the one hand, and forest products, on the other.

It is unlikely that developed countries will abandon their own efforts to promote labeling and eco-labeling. The European Union has had its own system in place since 1992, while many member states operate their own independent systems in addition to other systems of certification operated by NGOs and the private sector. Central American countries need to pay attention to the development of these systems and press to ensure that they do not operate in a way that discriminates against Central America in general and small farmers in particular.

The final step that Central American countries must take is to prepare themselves for the prospect of international trade involving the emission and fixation of carbon dioxide (CO₂) and other greenhouse gases. The Kyoto Protocol (1997) established in principle the reduction in emissions from 1990 levels in developed countries to be achieved by 2010. While international agreement has not yet been reached on the extent to which reduction can be achieved through the purchase of emission permits from other countries, it is clear that some trade will be allowed. In particular, the Kyoto Protocol established a Clean Development Mechanism (CDM) that will allow rich countries to meet their reductions in part through projects in poor countries that either reduce greenhouse gas emissions or increase carbon sinks (fixation).

The potential of the CDM for Central America is enormous[56]. Costa Rica has already shown in a limited way what might be achieved in this area. Projects that could qualify include a switch from thermal to alternative sources of energy; a reduction in deforestation; an increase in secondary forests; an extension of the national park system; and the protection of the coastal environment. Furthermore, provided that the cost per ton of abatement is lower in Central America than in the rich countries, this process can be driven by the private sector. Indeed, if the average cost per ton of emission reduction in rich countries is $50, the average price paid to Central America is $30 and the cost in Central America is $10, then the net gain per ton in Central America is $20. With an estimated 55 million tons per year to be traded in the region[57], this would yield a gross income of $1.65 billion (approximately 15 percent of today’s regional exports) and a net income of $1.10 billion.

However, much of this potential will be wasted if the Central American countries have not prepared themselves for the opportunity. The private sector in rich countries will not participate unless the projects in Central America can guarantee results. The absence of clear title to property rights will undermine schemes to increase reforestation, and extensions to national parks will be meaningless unless resources are committed by the public sector to ensure compliance by local populations. In the international market in carbon trading that is soon expected to come into existence, the best prices will be obtained by those projects that are backed up by solid environmental impact assessments (EIAs).

The agricultural sector in Central America is still dominated by small farmers producing basic grains for the domestic market. With low land and labor productivity, these farmers are struggling to survive in the face of trade liberalization and high real interest rates. Carbon trading provides a unique opportunity to lower the costs of adjustment for this group by providing it with a viable alternative to existing crops. A commitment to protect as little as ten hectares of forest could generate a gross income of approximately $1000 per year, or three dollars per day. Preparing the small farm sector to take advantage of this opportunity is no simple matter, but the work needs to begin now.

6.2.3. Society and the Environment

Our approach to the management of environmental resources has stressed to need to reconcile economic management with the pursuit of sustainability. However, we recognize that there are limits to what economics can do on its own. Without a much greater degree of awareness of environmental issues by ordinary Central Americans and a higher level of participation by grassroots organization, the terrible damage to the environment in Central America will continue.

Awareness is a function of education. Schools have a vital role to play in promoting understanding of the fragile nature of the Central American environment and the need to enforce measures to protect it. However, awareness should not be limited to the schools. Environmental damage has many causes, among them the stresses created by widescale poverty. If Central America waits until poverty is abolished, there will be no environment to protect. Increasing awareness of the links between poverty and environmental degradation is essential and the poor must be given incentives to change their practices.

Participation is the other side of the environmental coin. Some environmental abuses are carried out by large firms who do not feel constrained by national laws. Such abuses have always been a problem in Central America and have led to a widespread sense of cynicism. Enforcement is often impossible because of the costs involved. Grassroots organizations have a vital role to play in this area. Local and national politicians have become sensitive to the agenda of such groups, but more needs to be done. Such organizations are a national - and, increasingly, regional - resource that can help to ensure that national laws and regional agreements are respected.

Twenty years ago, grassroots organizations had no role in much of Central America. Their voice could not be heard. The opening up of the political system has created new opportunities and environmental groups now have a major role to play. Governments have understandable reservations about giving space to unelected groups and have a legitimate right to insist on responsible behavior by NGOs in environmental and other fields. However, preservation of the environment can be enhanced if the enthusiasm and local knowledge of NGOS can be harnessed to the environmental objectives set by the state.

6.3. Human and Social Capital

The success of Central American development over the next two decades will be heavily contingent on the region’s ability to increase average levels of human capital (understood as individual-level resources and capacities), while simultaneously finding ways to promote and deploy social capital (resources and capacities embedded in social relationships and networks) in pursuit of development objectives. In several countries this will require a concerted effort to overcome the legacy of the deficient social policies that characterized previous development models, and that have only begun to be addressed (see Table 9). At the primary school level, per capita expenditures on education in Guatemala, El Salvador and Nicaragua are less than a third of the levels in Belize, Costa Rica and Panama, and the situation for secondary schools is not much different. In health spending per capita, Nicaragua, Honduras and Guatemala are extremely low, while El Salvador has managed to increase its expenditures to a level closer to that of Costa Rica.

Education and health are critical areas for human capital development. Public expenditure in these two areas not only increases the opportunities open to individuals, but also has a major impact on productivity and future growth. Investment in the health sector can be an especially positive economic force, inasmuch as it can be expected to generate growing demand for technologically advanced goods and services as well as employment opportunities over a wide range of skill levels and open to women in particular[58]. We therefore recommend that government spending in these areas be at least maintained where it is strong and substantially increased where it is weak. This is likely to require increasing state fiscal revenues, a consideration we deal with later in this report.

Table 9
Indicators of Education and Health Spending in Central America, 1993-97.

	Expenditures as % of GDP		Education Expenditures (US$ per student)***		Health Expenditures (US$ per capita)**
Country	Education*	Health**	Primary	Secondary	Health Expenditures (US$ per capita)**
Belize	5.0	3.9	$348	$693	$106
Costa Rica	5.3	8.6	$350	$598	$224
El Salvador	2.2	6.8	$131	$107	$158
Guatemala	1.7	4.2	$105	$219	$56
Honduras	3.6