Intel opened the door, not the economy.

Costa Rica upgraded, but linkages remained weak.

Intel signaled change, not transformation.

Between 1990 and 2026, Costa Rica moved from assembly exports toward high-tech manufacturing, services, and ecotourism. The visible break came in 1996, when Intel chose the country for a semiconductor plant, giving it global credibility.

The central point of the blog is that Intel opened the door, not the economy. Public policy built a platform that attracted anchor firms, but it did not spread those gains across the domestic economy. This blog shows how Costa Rica upgraded exports and gained international credibility, while explaining why those gains did not produce a broad domestic transformation.

Exports upgraded, but spillovers stayed narrow.

Export composition changed sharply after 1990. Traditional agriculture and apparel assembly lost ground, while electronics, medical devices, and services expanded. Textile exports fell from 13.3 percent of goods exports in 2000 to 1.6 percent by 2014.

Foreign direct investment grew in parallel. Net inflows rose from $400 million in 2000 to more than $5 billion by 2024. The Free Trade Zone (FTZ) regime expanded from 56 firms in 1990 to 626 by 2024, raising its contribution from 1 percent to 15 percent of gross domestic product (GDP).

Intel’s arrival marked the first structural shift. The plant initially created more than 2,200 jobs and later exceeded 3,000, while driving a large share of export growth. By 2014, FTZ firms generated 53 percent of total exports.

Sectoral upgrading followed in waves. Electronics led the early phase. Medical devices then became dominant, moving from low-value disposables toward more complex instruments and therapeutic equipment. Services also expanded through information and communication technology and business processing, supported by bilingual labor and global demand.

Tourism provided a parallel export engine. By 2016, it generated $3.66 billion, exceeding the combined value of coffee and banana exports.

These shifts mark a structural break in exports and productive capabilities. However, most of that change stayed inside the FTZ system rather than spreading across the domestic economy.

Targeted upgrading raised credibility and resilience.

Anchor firms validated a new strategy.

The shift reflected a sequenced strategy with three mechanisms: anchor-firm targeting, institutional adaptation of skills, and environmental stabilization.

First, the Costa Rican Investment Promotion Agency (CINDE) screened sectors, identified anchor firms, and worked directly with executives. Intel’s 1996 investment created a demonstration effect, which validated Costa Rica as a high-tech location. That validation then supported expansion into medical devices and business services.

Second, the state turned broad education gains into firm-specific skills. The 1949 social compact created literacy and human capital, but general education did not meet specialized needs. Public universities, the Instituto Nacional de Aprendizaje (INA), and firms then coordinated new technical programs in electronics, engineering, and informatics.

Third, upgrading moved through sequential sector building. After Intel, the same targeting logic shifted toward medical devices. Firms increased product complexity and strengthened regulatory compliance, which opened markets beyond the United States.

Environmental policy stabilized the transition.

Fourth, ecotourism stabilized external accounts during industrial transition. The 1996 Forestry Law created the Payment for Ecosystem Services system. That policy paid landowners and reversed deforestation, raising forest cover from 21 percent in 1987 to 57 percent by 2017.

The through-line remains clear. Intel opened the door by proving credibility. Policy then used that credibility to move into other sectors.

However, the same mechanisms also reveal the constraint. Domestic absorption stayed weak. Specialized local inputs in high-tech sectors remained below 2 percent, signaling limited technological spillovers.

The state created platforms, not linkages.

The Costa Rican state did not retreat from development. Instead, it recast its role across five functions.

Public institutions lowered investor risk.

First, the state shaped markets. The FTZ regime offered tax exemptions, duty-free imports, and regulatory predictability. Trade agreements then connected the country to global markets.

Second, the state financed public goods. Investments in education, health, and energy created assets that multinational firms required. The renewable energy grid and skilled labor force became central to investor decisions.

Third, the state established key institutions. CINDE, the Ministry of Foreign Trade (COMEX), and the Costa Rican Foreign Trade Promoter (PROCOMER) coordinated investor attraction, regulation, and export promotion. These agencies lowered transaction costs and improved execution.

Fourth, the state adapted under stress. Intel’s 2014 manufacturing exit removed roughly one-fifth of goods exports. The economy absorbed that shock through growth in services and medical devices. This response showed that capabilities extended beyond a single firm.

Weak linkages exposed the model’s limits.

Fifth, the state accepted fiscal and policy risks. It financed incentives and environmental programs, often through narrow tax bases such as fuel taxes. Those choices supported growth but created long-term fiscal pressure.

The state’s exit remained incomplete. FTZ incentives did not include strong domestic-content requirements or technology-transfer conditions. Supplier development programs helped some firms, but they did not scale.

The result was uneven capacity. Export-promotion agencies performed well, but the broader innovation system remained fragmented.

Domestic absorption must come before scale.

The Costa Rica case highlights a binding constraint: domestic absorption. Attraction succeeded, but integration lagged.

Sequencing shaped the model’s long limits.

First, sequencing matters. Costa Rica built a credible export platform before it developed supplier ecosystems. That sequence created enclave dynamics that persisted over time.

Second, institutional trade-offs are unavoidable. Specialized agencies such as CINDE can speed investor attraction. However, parallel institutions can also isolate export sectors from domestic firms and workers.

Third, fiscal design shapes sustainability. FTZ tax exemptions attracted investment, but they also reduced revenue for public goods. The state financed education and infrastructure, while key firms remained partly outside the tax base.

Fiscal and skills constraints still matter.

Fourth, skills policy must be targeted, not generic. General education created a baseline, but specialized programs drove upgrading. Limited secondary completion and shortages in science, technology, engineering, and mathematics constrained inclusion.

Fifth, environmental policy can stabilize growth but also create new constraints. The Payment for Ecosystem Services (PES) program relied on a fuel tax, which will be weakened by decarbonization. That creates a direct funding challenge for conservation.

The lesson returns to the anchor. Intel opened the door, but the economy did not fully follow. Without domestic absorption, anchor-led growth can entrench a dual economy instead of producing convergence.

For policymakers in Latin America and the Caribbean (LAC), the message is clear. Anchor firms can open new sectors and signal credibility, but they do not guarantee domestic transformation in the region. That depends on whether public policy builds suppliers, skills, and fiscal capacity early enough to spread gains beyond the export enclave.