Overview: a new external shock for Asia's accounts
Artificial intelligence (AI) is transforming global production and services in ways that go beyond domestic labor markets. For many Asian economies the risk is not only job disruption at home but also a sudden deterioration in external accounts — exports of services, remittance inflows, and capital flows that have supported current-account surpluses and FX reserves could all shift quickly. This article maps the channels by which rapid AI adoption in advanced economies can produce balance-of-payments (BOP) stress in Asia, illustrates the most exposed countries with data, and outlines policy options to prevent a disruptive adjustment.
Key facts:
- AI exposure is uneven across the region: IMF analysis finds about half of jobs in Asia's advanced economies are exposed to AI, versus roughly a quarter in emerging market and developing economies — but the types of jobs matter for services exports and remittances (IMF, Jan 2025) (https://www.imf.org/en/blogs/articles/2025/01/05/how-artificial-intelligence-will-affect-asias-economies).
- The Philippines' BPO sector generated roughly US$35.5 billion in revenues in 2023 and employs well over a million workers; that sector is widely cited as highly exposed to AI-driven automation (IMF working paper; industry reports).
- Remittances to low- and middle-income countries were an estimated US$656 billion in 2023 — a major source of external financing for many Asian economies (World Bank Migration & Development Brief 2024) (https://documents1.worldbank.org/curated/en/099714008132436612/pdf/IDU1a9cf73b51fcad1425a1a0dd1cc8f2f3331ce.pdf).
Transmission channels to the balance of payments
- Services exports (outsourcing, contact centers, back-office services)
Many Asian economies run large services export surpluses driven by business-process outsourcing (BPO), IT services and contact centers. AI tools — conversational agents, large-language-model driven automation, and end-to-end service orchestration — can replicate large parts of low- and mid-skill service provision remotely and at lower cost in advanced-economy firms or cloud providers. That creates three connected risks to the BOP:
- Demand reallocation: Buyers in high-income countries can substitute offshore human agents with AI-enabled domestic platforms, shrinking imports of offshore services. Industry reporting documents rapid AI deployment in global contact centers and rising automation of routine interactions (Bloomberg; Rest of World) (https://www.bloomberg.com/news/features/2024-08-27/philippines-call-centers-navigate-ai-impact-on-jobs; https://restofworld.org/2024/ai-reshaping-call-center-work-philippines/).
- Price compression: AI-driven service delivery can lower the price of certain services, squeezing revenue even if volume holds.
- Competitiveness shifts: Cloud-native AI firms based in advanced economies may capture higher-value service segments, leaving legacy offshore providers with declining margins.
Because services exports are often a concentrated share of current-account receipts in countries like the Philippines, Sri Lanka, and parts of South Asia, a meaningful drop in BPO/IT revenues would directly reduce exports of services and widen current-account deficits.
- Remittances and labor income
Remittances are a sizable, stable source of FX for many Asian economies. If AI reduces demand for lower-skilled migrant labor in destination countries (for example, routine customer service, low-complexity administrative roles), remittance volumes could moderate. The World Bank reports remittances remain large — US$656 billion to LMICs in 2023 — but notes downside risks from weaker demand for less-skilled migrant labor (World Bank Migration & Development Brief) (https://documents1.worldbank.org/curated/en/099714008132436612/pdf/IDU1a9cf73b51fcad1425a1a0dd1cc8f2f3331ce.pdf).
Countries heavily dependent on remittances for household consumption and external financing are therefore vulnerable to a double hit: lost BPO revenues and lower remittance inflows.
- Goods trade and competitiveness
AI will also reshape goods production through automation, smarter supply chains, and onshoring of advanced manufacturing. Over time, vertical reorganization could reduce demand for some intermediate imports from Asia, narrowing trade surpluses in manufacturing exporters. While these effects are slower than services automation, they can compound external adjustment pressure if simultaneous with a collapse in services receipts.
- Capital flows and market sentiment
A rapid external revenue shock can trigger portfolio outflows and currency depreciation. Investors price in policy and political risk; sudden drops in FX earnings (services + remittances) erode reserve adequacy and may force central banks to tighten domestic policy or use FX reserves to defend currencies — outcomes that raise sovereign financing costs and feed back into the BOP.
Who is most exposed? Country illustrations
Philippines — a frontline case
- The Philippines is the world’s leading contact-center/BPO hub; BPO revenues were reported at about US$35.5 billion in 2023 (IMF) and the sector employs well over a million people directly (IMF working paper; industry sources). Rapid adoption of AI-based conversational systems and automation in buyers’ countries threatens both revenue and jobs (IMF; Bloomberg; Rest of World) (https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025043-print-pdf.pdf; https://www.bloomberg.com/news/features/2024-08-27/philippines-call-centers-navigate-ai-impact-on-jobs).
- A sizeable contraction in BPO exports would translate fast into lower services receipts and diminished foreign-exchange inflows, pressuring the current account and the peso.
Other exposures across South and Southeast Asia
- Countries with large workforces abroad (e.g., Bangladesh, Pakistan, Nepal) depend heavily on remittances. If AI reduces demand for the types of roles migrants fill in destination markets, remittance growth could slow (World Bank migration brief).
- India’s IT services sector is more diversified and higher up the value chain; IMF analysis suggests advanced economies have more jobs complementary to AI, which could favor higher-skilled IT services — but parts of India’s services exports are also exposed to automation (IMF, Jan 2025).
Scenarios and potential scale
Quantifying the exact BOP hit from AI is uncertain: impacts depend on the pace of adoption in advanced economies, the kinds of services automated, and policy responses. Still, some rough scenario thinking is useful:
- Moderate scenario: gradual AI adoption reduces low-value BPO volumes by 10–20% over five years. For a sector contributing tens of billions in annual receipts (e.g., the Philippines), this implies a multi-billion-dollar annual loss in export receipts — enough to swing small current-account surpluses into deficits.
- Severe scenario: rapid reshoring and price compression reduce services receipts by 30%+ within 2–3 years. That could produce sharp currency depreciation, a spike in external financing needs, and pressure on reserves unless offset by capital inflows or policy buffers.
Because remittances are large and relatively stable, even a modest slowdown there would compound the stress from services revenue losses.
Policy options: prevent, cushion, adapt
Short-term — cushion the external adjustment
- Targeted FX management: maintain adequate reserve buffers, use swap lines, and, where prudent, temporary macroprudential measures to limit destabilizing capital outflows.
- Fiscal targeting: reorient temporary fiscal support to displaced workers and to preserve essential public spending without procyclical cuts.
Medium-term — manage the structural transition
- Reskilling and labor-market policies: scale reskilling programs targeted at moving workers from routine service roles into AI-complementary functions (supervision, quality assurance, AI fine-tuning, data labeling, higher-value IT services). IMF analysis highlights reskilling and social-safety nets as key to an inclusive AI transition (IMF, Jan 2025) (https://www.imf.org/en/blogs/articles/2025/01/05/how-artificial-intelligence-will-affect-asias-economies).
- Export diversification: accelerate movement up the services value chain (digital engineering, AI services exports, healthcare IT, fintech) and strengthen goods export competitiveness in niches less vulnerable to immediate AI substitution.
- Attract AI investment: encourage AI-related FDI and start-ups to locate data centers, model-training and annotation services, and AI operations in Asia — converting a potential job-loss channel into a new export opportunity.
Regulatory and cooperation measures
- Trade and data governance: negotiate commercial frameworks that allow cross-border data flows with strong privacy protections so service providers can remain competitive.
- International cooperation on migration: coordinate with destination countries on labor market adjustments for migrants and channels for upskilling.
Conclusion: preparedness matters
AI is neither an all-good nor an all-bad force for Asian economies. It offers productivity gains and the possibility of new high-value service exports. But the risk of a balance-of-payments shock follows directly from concentrated dependence on low- and mid-skill services exports and remittances — sectors that AI can replicate or reduce demand for in origin and destination markets.
Policymakers should treat the AI transition as a strategic external-risk episode: measure exposure (magnitudes of services receipts and remittances), build contingency financing and reserve buffers, and accelerate policies that convert displaced labor into AI-complementary roles and attract higher-value digital and AI services. Countries that prepare proactively — by blending macroeconomic resilience with aggressive human-capital and industrial policy — can avoid BOP crises and capture a larger share of the global AI-driven opportunity.
Sources and further reading
- IMF, "How Artificial Intelligence Will Affect Asia's Economies," Jan 5, 2025 (IMF blog): https://www.imf.org/en/blogs/articles/2025/01/05/how-artificial-intelligence-will-affect-asias-economies
- IMF, "Artificial Intelligence and the Philippine Labor Market" (working paper, 2025): https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025043-print-pdf.pdf
- World Bank, Migration and Development Brief 40 (remittances): https://documents1.worldbank.org/curated/en/099714008132436612/pdf/IDU1a9cf73b51fcad1425a1a0dd1cc8f2f3331ce.pdf
- Bloomberg, "The World's Call Center Capital Is Gripped by AI Fever — and Fear," Aug 27, 2024: https://www.bloomberg.com/news/features/2024-08-27/philippines-call-centers-navigate-ai-impact-on-jobs
- Rest of World, "AI is reshaping call center work in the Philippines": https://restofworld.org/2024/ai-reshaping-call-center-work-philippines/