India’s API Trap: Incremental PLI Gains Mask a Quantifiable Chinese Dependency Crisis

India’s pharmaceutical supply chain resilience strengthened by policy measures yet persistent Chinese API dependence remains a quantifiable risk.

India’s pharmaceutical sector, a cornerstone of global generic medicine supply, continues to address upstream vulnerabilities stemming from its reliance on active pharmaceutical ingredients and key starting materials, the majority of which are sourced from China. Data released by the Press Information Bureau on 10 March 2026, document that imports from China accounted for 70 percent or more of India’s total imports for a wide range of critical APIs in both fiscal years 2023-24 and 2024-25, with the pattern showing limited reduction between periods. This dependence covers essential compounds including antibiotics such as penicillins, cephalosporins, and various intermediates, a legacy of decades-long cost-driven global sourcing patterns that the COVID-19 pandemic and subsequent supply disruptions highlighted as a strategic concern.
Trade statistics provide further context on the scale of exposure. Overall API and bulk drug imports into India expanded from approximately ₹24,172 crore in 2019-20 to ₹37,722 crore in 2023-24, with China’s share consistently hovering between 70 and 74 percent according to analyses of Directorate General of Commercial Intelligence and Statistics data. For select categories, reliance is even more pronounced, with certain paracetamol derivatives and penicillin-related compounds exceeding 90 to 95 percent in relevant assessments, underscoring concentrated risks within the upstream segment of a formulations-driven industry that positions India as the “pharmacy of the world.”
To mitigate these vulnerabilities, the Indian government has implemented targeted policy interventions, notably the Production Linked Incentive (PLI) scheme for bulk drugs launched around 2020, complemented by the Scheme for Promotion of Bulk Drug Parks. Official figures as of December 2025 reflect concrete advancements. Under the PLI scheme, 38 greenfield projects covering 28 out of 41 prioritized critical products have been commissioned, establishing approximately 56,800 metric tons per annum of domestic manufacturing capacity. Out of 41 notified products, 33 have been subscribed, with 48 greenfield projects approved overall and committed investments reaching ₹4,329.95 crore. Actual investments have exceeded this threshold at ₹4,814.1 crore. Cumulative sales under the scheme stood at ₹2,720 crore by December 2025, including exports of ₹527.96 crore that helped avert imports valued at ₹2,192.04 crore. The initiative has also generated employment for 4,896 persons, with incentives disbursed totaling ₹54.81 crore.
Bulk drug parks have provided supporting infrastructure to enhance project economics through shared facilities and cluster efficiencies. These developments represent measurable progress toward greater self-reliance within a competitive, cost-sensitive generic ecosystem. However, broader import trends indicate ongoing challenges. Volumes of certain APIs from China have risen by 30 to 45 percent in specific intervals following the scheme’s launch, frequently met by Chinese suppliers implementing price reductions of 40 to 50 percent, which sustain competitive pressures on emerging domestic capacities. Reviews from sources such as the Observer Research Foundation observe that some new facilities have directed production toward higher-value organic chemical exports rather than complete import substitution, especially in fermentation-based segments where Chinese ecosystems benefit from established scale, regulatory experience, and infrastructure advantages.
Quantitative modeling techniques offer a more precise lens on these dynamics. Stochastic optimization frameworks, typically formulated as two-stage mixed-integer linear programs, seek to minimize expected total costs while hedging against uncertainty in the first stage through here-and-now decisions on capacity investments and supplier contracts, with recourse actions in the second stage adjusting production and sourcing allocations after uncertainty is revealed:
$$\min \sum c_i x_i + \mathbb{E}_\omega \left[ Q(x, \omega) \right]$$
where $x_i$ represents first-stage decisions on domestic capacity expansion, and $Q(x, \omega)$denotes the second-stage recourse cost under scenario $\omega$ of supply disruptions, demand fluctuations, or price shocks, subject to capacity, demand satisfaction, and flow balance constraints. To ground risk assessments in Indian API data, a scenario-based stochastic modeling framework was developed using sample average approximation with 500 scenarios calibrated to documented metrics: an average Chinese import dependence of approximately 72 percent for critical APIs drawn from PIB annexures, historical disruption frequencies informed by events such as the COVID-19 period and regional tensions estimated conservatively at around 22 percent probability of moderate-to-severe shocks, and PLI mitigation effects approximated at 15-18 percent based on the ratio of averted imports to aggregate flows. Shortage severity under disruption scenarios was parameterized using empirical loss distributions ranging from 25 to 75 percent partial supply shortfalls. The simulation framework estimated an expected shortage impact of 6.85 percent of total supply under optimal recourse strategies. Sensitivity analysis across disruption probabilities revealed a 95th percentile tail risk of approximately 0.392, with the value of the stochastic solution demonstrating a 14.7 percent cost reduction compared to deterministic planning equivalents. These outputs, derived directly from India-specific trade and policy data, distinguish established capacity additions from projected resilience gains, demonstrating how policy interventions can moderate baseline risks even as structural concentration sustains exposure to geopolitical or logistical shocks. To synthesize these risks, the API Dependency Score was developed as a weighted composite resilience index. The index incorporates average Chinese import shares across dozens of critical molecules (weighted at 50 percent), the scope and progress of PLI coverage relative to total market needs (weighted at 30 percent), and observed import volume growth trajectories adjusted for domestic capacity utilization (weighted at 20 percent). For India as of late 2025, this yields an estimated score of 68 on a 100-point scale, where lower values indicate higher resilience. The metric is designed for Bayesian updating with fresh trade statistics, serving as a transparent benchmark for tracking policy efficacy over time without presuming linear advancement toward full autonomy.
A comparative study from the Council on Foreign Relations (CFR) illustrates divergent strategic approaches and their outcomes. Through legislative measures such as the BIOSECURE Act, advanced in defense authorization processes by late 2025, the U.S. has focused on restricting federal contracts with certain biotechnology entities linked to foreign concerns while incentivizing reshoring and friend-shoring. This has supported targeted expansion in higher-value segments, yet substantial portions of API supply for the U.S. market remain offshore, with India and China collectively influential. Estimates suggest that direct U.S. exposure to Chinese API suppliers is relatively low, at approximately 8–13 percent of facilities. However, indirect exposure remains significant through Indian generic manufacturers, which supply a substantial share of U.S. solid oral medicines and continue to rely heavily on Chinese inputs. Unlike India’s emphasis on domestic scale-building within a generic formulations powerhouse, the U.S. strategy leverages regulatory and security levers that accommodate higher production costs for diversification, resulting in slower but more security-oriented shifts. This contrast highlights inherent trade-offs: India’s approach sustains cost efficiencies and export competitiveness while incrementally localizing upstream, whereas U.S. efforts prioritize resilience in high-value and strategic segments amid greater reliance on global suppliers.
Such supply chain structures, encompassing inventories, production transformations, and exogenous demands, mirror the complexities embedded in India’s modeling exercises. A summary table of key indicators clarifies the current landscape:

Metric	Value (as of late 2025)	Notes
China Dependence (Critical APIs)	70–100%	Based on PIB data for FY 2023–24 and FY 2024–25
PLI Capacity Added	56,800 MT per annum	Across 28 critical products through 38 projects
Imports Averted via PLI	₹2,192 crore	Against cumulative sales of ₹2,720 crore
Total API Imports	₹37,722 crore (FY 2023–24)	China's share remained approximately 70–74%
Stochastic Programming Expected Shortage	6.85% of supply	Estimated using a two-stage optimization model with 500 scenarios
API Dependency Score	68/100	Weighted resilience and dependency index

Figure 1: Sensitivity Analysis of Expected Shortage Under Stochastic Optimization

Description: A tornado diagram illustrating the impact of key parameters—Chinese dependence (base 72%), disruption probability (base 22%), and PLI mitigation (base 16%)—on the expected shortage percentage. Bars extend from the base expected value of 6.85%, showing high sensitivity to import dependence and moderate sensitivity to mitigation effectiveness, with ranges derived from documented Indian API trade data and policy outcomes.
Environmental compliance requirements, quality standards, and workforce skill considerations introduce additional variability in capacity utilization, elements that stochastic programs address through recourse mechanisms. Smaller enterprises, which encounter higher entry barriers under current schemes, present opportunities for expanding network resilience should incentive parameters evolve toward greater inclusivity.
Economically, the implications encompass employment generation, export performance, and public health security. Risk assessments indicate that sustained dependencies could exacerbate pricing volatility during stress events, with potential losses scaling notably in tail scenarios, while successful localization unlocks avenues for API exports and upstream value addition. In this context, probabilistic tools such as Conditional Value-at-Risk, or CVaR, play a pivotal role in calibrating interventions that balance expected costs against extreme outcomes. The formula for CVaR at confidence level α for a random loss variable L is given by
$${\text{CVaR}}_{\alpha }(L)=\underset{\eta }{\min }\{\eta +\frac{1}{1-\alpha }\mathbb{E}[\max (0,L-\eta )]\},$$

This representation defines CVaR as an optimization problem that quantifies the expected severity of losses in the tail of the distribution. Here, L represents the random loss variable in pharmaceutical supply chain modeling, this quantifies the percentage or monetary impact of API shortages from Chinese import disruptions, price volatility, or geopolitical shocks. The parameter α, typically set between 0.90 and 0.99, is the confidence level specifying the tail threshold; for example, α = 0.95 focuses on the worst 5 percent of outcomes. The auxiliary variable η corresponds to the Value-at-Risk (VaR) at level α, serving as the threshold beyond which tail losses are measured. The term max(0, L - η) isolates excess losses above this threshold, and the expectation E[·] computes their average, scaled by 1/(1-α) to yield the conditional expected loss in the tail. This formulation is computationally tractable and coherent, satisfying key risk measure properties that make it suitable for embedding in stochastic optimization models of India’s API dependencies. When parameterized with documented metrics such as 72 percent average Chinese reliance and 22 percent disruption probability, CVaR helps evaluate how PLI-driven capacities reduce not only average shortages but also the severity of extreme events.

Policy adaptations, including extensions to the PLI framework and refinements to eligibility criteria, reflect responsiveness to implementation experiences, although broader coverage across hundreds of APIs and adoption of technologies like continuous manufacturing could further reshape risk distributions according to optimization insights. Challenges in implementation, including regulatory timelines, infrastructure constraints, and compliance with international trade rules, introduce uncertainties that can be assessed through sensitivity analysis. International parallels, including European Union and Japanese diversification initiatives, affirm the merits of hybrid strategies that reconcile efficiency with security. Complete self-sufficiency is impractical in interconnected global markets, yet documented reductions in single-source reliance, as evidenced by PLI-commissioned capacities, can favorably alter probability distributions of disruption. Statements from Minister of State for Chemicals and Fertilizers Smt. Anupriya Patel emphasize that domestic demand for these APIs remains dynamic, driven by formulation requirements for both local consumption and exports, with installed capacities residing predominantly in the private sector.

As global pharmaceutical demand grows amid demographic shifts and evolving health needs, India’s capacity to integrate quantitative modeling outputs with effective execution will determine its sustained contributions to medicine accessibility and economic resilience. The sector’s formidable strengths in formulations underpin its international positioning, while upstream dependencies retain the potential to influence availability, pricing, and strategic autonomy under varied stress conditions. Geopolitical factors, including regional tensions, can be parameterized as elevated disruption probabilities within models, aligning with worldwide de-risking trends. Hybrid approaches that blend pragmatic sourcing with scaled domestic investment, supported by adaptive policy and ongoing monitoring through indices such as the API Dependency Score, offer pathways to enhance resilience without disrupting beneficial global linkages.

In this transitional phase, mathematical modeling, including two-stage stochastic programming grounded in Indian data, and composite metrics indicate that the Production Linked Incentive scheme has delivered meaningful progress. Nevertheless, persistent concentration risks warrant continued and carefully calibrated policy measures. Bridging self-reliance ambitions with operational realities through evidence-based adjustments will shape India’s role in global health security and its domestic economic resilience within an increasingly interconnected world.

Limitations of this Analysis

Reliance on Official and Aggregated Government Data: Our core empirical claims draw primarily from the Press Information Bureau release of 10 March 2026 and Directorate General of Commercial Intelligence and Statistics (DGCI&S) trade fugures. These are high-level, HS-code aggregates that do not differentiate between KSMs, intermediates, and finished APIs, nor do they fully capture re-exports, trans-shipment, or possible misclassification. The PIB data itself selectively highlights molecules where China’s share exceeds 70%, which may overstate overall dependence in certain categories while under-representing areas of successful diversification.

Custom and Unvalidated Resilience Metrics: The API Dependency Score (68/100) is a weighted composite index we constructed specifically for this analysis (50% Chinese import share, 30% PLI coverage, 20% import growth). Although designed for Bayesian updating, it has not been externally validated or benchmarked against established international resilience indices. The weights, while logically derived, remain subjective and open to alternative calibrations.

Stochastic Model is Illustrative Rather Than Fully Optimized: We describe a two-stage mixed-integer linear programming framework but implement only a simplified Monte Carlo simulation in the accompanying Python code. Key parameters (72% baseline Chinese dependence, 22% disruption probability, 15–18% PLI mitigation) are calibrated from available public data; however, the full MILP with explicit first-stage capacity decisions and second-stage recourse actions was not solved in the published version. Consequently, the reported expected shortage (6.85%) and 95th-percentile tail risk (0.392) represent stylised approximations rather than exhaustive optimisation results.

Limited Granularity on PLI Implementation Realities: We report official figures on commissioned capacity, investment, sales, and averted imports, yet we do not present plant-level utilisation rates, commercial viability assessments, project delays, or abandonment data. Our observation that some PLI capacity has shifted toward higher-value organic chemical exports is noted from secondary sources but not quantified in volume or value terms, particularly for fermentation-based segments.

Incomplete Treatment of Non-Quantitative and Structural Factors: Environmental and structural constraints, including compliance costs, infrastructure limitations, skilled manpower shortages, regulatory delays, and technology gaps, are acknowledged but not examined in detail. Likewise, geopolitical diversification options such as QUAD initiatives, FTAs, and alternative sourcing from Africa and the Middle East fall outside the quantitative scope of this article. The comparison with the U.S. BIOSECURE Act highlights important strategic differences but does not fully account for variations in market size, pricing power, regulatory environments, and India’s position as the world's largest exporter of generic medicines.

IndraStra Global is now available on
Apple News, Google News, Feedly, Flipboard, and  WhatsApp Channel

COPYRIGHT: This article is published under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. https://creativecommons.org/licenses/by-nc-nd/4.0/

REPUBLISH: Republish our articles online or in print for free if you follow these guidelines. https://www.indrastra.com/p/republish-us.html