Promoting the Development of the Semiconductor Ecosystem in Costa Rica
📄 Read at OECD ↗Country Review · Apr 2026 · 179 pp
Summary
Costa Rica has 13 semiconductor firms employing approximately 5,000 workers, dominated by Intel (2,800 employees + 5,600 contractors) and Qorvo. The country’s advantages — near-100% renewable electricity, strong FDI framework, geographic proximity to the US, and an English-speaking technical workforce — are offset by transport infrastructure gaps, STEM pipeline constraints, and extreme concentration risk. In 2025, both Intel and Qorvo announced they would transfer their ATP plants out of Costa Rica, pushing the country back toward design, validation, and services. The OECD recommends integrating semiconductor goals into the National Strategic Plan (PEN 2050) and investing in local supplier development to reduce fragility.
Key Insights
13 firms, 5,000 workers — and Intel is most of it. Intel invested $600 million in 2021 to reactivate its ATP line in Costa Rica. At peak, Intel’s contribution was 0.9% of GDP and 27% of total exports. Both Intel and Qorvo announced ATP plant transfers out of Costa Rica in 2025. (Table 2.1, pp. 45–48)
The 2014 Intel exit cratered the ecosystem. Gross output fell from $794 million (2012) to $130 million (2015). SME share of employment collapsed from 19% to 3%, value-added from 24% to 5%. The sector grew only 9% in 12 years while medical devices saw fivefold growth. (pp. 52–58)
Near-100% renewable electricity — but climate-vulnerable. Costa Rica achieved 99.93% renewable electricity in 2020 (67% hydro, 12% geothermal, 10% wind). But El Niño caused rationing in May 2024 — the first in 17 years. Renewable share dropped to 91% in 2023. (pp. 140–145)
FTZ tax incentives are aggressive. Outside the Greater Metropolitan Area: 0% corporate tax for 6 years, then 5% for 6 years, then 15% for 6 years (vs. standard 30%). This is more generous than India’s ISM capital subsidy model, though structured as tax breaks rather than upfront grants. (pp. 85–90)
70%+ workforce overlap between medical devices and semiconductors. The most demanded semiconductor occupations overlap with medical devices — a sector where Costa Rica has 41,000 workers and $9 billion in exports. Countries with existing electronics manufacturing can cross-leverage that talent pool. (pp. 115–120)
Persistent semiconductor trade deficit. Average net imports: $122 million for chips, $47 million for equipment, $95 million for photosensitive devices. Costa Rica has Revealed Comparative Advantage only in chip exports, not in materials, equipment, or other segments. (pp. 60–65)
What This Means for India
India’s ISM is running the same anchor-firm playbook. Tata Electronics (Dholera fab), Micron (Sanand ATMP), CG Power (OSAT) — India’s semiconductor programme is built around a small number of large commitments. Costa Rica’s experience shows this works for initial ecosystem seeding but creates structural fragility. India needs a deliberate local supplier development strategy from day one, not as an afterthought.
The “missing middle” problem is directly relevant. Costa Rica’s SME share collapsed as Intel dominated the ecosystem. India’s ISM has no programme for semiconductor SME development. If Tata Dholera operates as an enclave — importing materials, using foreign EDA tools, exporting finished wafers — the multiplier effect on the Indian economy will be minimal, just as it was minimal for Costa Rica outside Intel’s own operations.
Costa Rica’s FTZ model offers a design comparison for Indian SEZs. The tiered corporate tax rates (0% outside metro, 6% inside) specifically incentivise geographic diversification. India’s ISM subsidies are location-agnostic in structure but concentrated in Gujarat in practice. A tiered incentive model could direct semiconductor investments to states with lower costs and existing electronics manufacturing — Tamil Nadu, Karnataka, Telangana.
Source
OECD (2026), Promoting the Development of the Semiconductor Ecosystem in Costa Rica, OECD Publishing, Paris, https://doi.org/10.1787/35fe6797-en.