Cross-Border Ownership and Input Access: Evidence from the Lithium Supply Chain
summary
The gist
How does cross-border ownership shape the lithium supply chain? The single most informative sentence about the paper's subject and finding is…
In short
The study models lithium production through a network of ownership and refining links to see how cross-border ownership affects supply chain dynamics. It finds that ownership restrictions create complex profit incentives for mines and producers, influencing sourcing decisions. Policy tests show divestiture changes ore purchases while export bans significantly reduce global lithium supply.
Key concepts
- Ownership Incentives
- When owners have stakes in both a mine and a refiner, they gain incentives to act in ways that benefit their portfolio. This can lead to mines restricting sales or refiners expanding production, creating competition among affiliated entities.
- Input-Foreclosure Incentive
- This occurs when a mine whose owners also hold refining interests might restrict sales of ore that would allow its affiliates' competitors to gain market share. This incentive is captured in the model as a term ($T_{Vm}$) reflecting the mine's profit changes due to ownership-linked orders.
- Divestiture Counterfactual
- Comparing divestiture of Chinese stakes with an export ban shows that divesting Chinese stakes increases spot ore purchases and ore prices, but aggregate lithium output remains nearly unchanged. This suggests trade access matters more than the specific ownership structure for overall supply.
- Export Ban Counterfactual
- A ban on exporting overseas ore to China substantially reduces global lithium supply by 25.36% and raises world prices by 77.19%. This demonstrates that restrictions on input access can severely limit the total amount of lithium available globally, regardless of where refining capacity exists.
Terminology used across episodes
This episode discusses
The paper
Cross-Border Ownership and Input Access: Evidence from the Lithium Supply Chain · Read on arXiv
Kei Irizawa
Yale University · Benchmark Mineral Intelligence · Yale Tobin Center for Economic Policy · Cowles Foundation
Transcript
Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.
Ruth: We're looking at the paper: "Cross-Border Ownership and Input Access".
Warren: Here's the gist of the paper. How does cross-border ownership shape the lithium supply chain?
Ruth: The single most informative sentence about the paper's subject and finding is: Ownership restrictions change affiliated sourcing advantages, mines’ incentives to supply spot ore, and producers’ interests in competing firms’ profits Model Structure The paper develops a model of mining and refining linked by an observed ownership network where producers maximize ownership-weighted profits This model represents lithium production from hard-rock deposits and brine, connecting producers both along the supply chain and across these competing sources of lithium The objective for an owner is to maximize a weighted sum of its owners’ portfolio profits where weights are proportional to equity stakes in that producer Ownership Incentives Ownership creates competing profit incentives, such as a refiner with a stake in a mine sharing in mining profits which can lower its effective input cost A mine whose owners also hold refining interests may restrict sales that would strengthen its affiliates’ competitors—an input-foreclosure incentive Horizontal interests among final lithium producers create incentives to internalize rivals’ profits, discouraging expansion that reduces those profits Vertical ownership links can give mines an incentive to restrict spot supply and refiners an incentive to expand production through reduced double marginalization Policy Counterfactuals The study compares divestiture of Chinese stakes in overseas producers with a ban on ore exports to China Divestiture increases spot ore supply but leaves lithium output nearly unchanged, while continued trade preserves Chinese access to ore The export ban reduces global lithium supply because refining capacity elsewhere cannot fully replace lost Chinese production Key Findings on Interventions Under observed portfolios, divestiture increases spot ore purchases and the ore price while leaving aggregate lithium output nearly unchanged The comparison under hypothetical pooling of Chinese refining interests suggests a larger output response to divestiture, driven primarily by brine expansion, even as spot ore purchases fall The ore-export restriction substantially reduces global lithium supply and raises lithium prices, despite lowering ore prices outside China Sourcing and Production Dynamics The effective marginal ore cost on route (r, m) is calculated as crm = (1 − κVrm)w + κVrmC M′m (xm) When w > CM′m (xm), a higher ore price therefore widens the refiner’s cost advantage over spot buyers while also raising its own effective sourcing cost The vertical block combines the mine’s profit change from ownership-linked orders with the downstream profit changes it internalizes The input-foreclosure incentive is captured in the term T Vm, which contains the input-foreclosure incentive Policy Implications Ownership restrictions change affiliated sourcing advantages, mines’ incentives to supply spot ore, and producers’ interests in competing firms’ profits Divestiture shifts ore purchases toward the spot market while allowing Chinese refiners to continue purchasing ore through international trade, with little change in global lithium supply The export ban prevents overseas ore from reaching Chinese refineries and lowers ore prices outside China, but existing refining capacity elsewhere cannot fully replace the lost production Ownership and trade restrictions can alter access to inputs, but they cannot substitute for the refining capacity needed to use those inputs Data and Estimation The panel spans 2015–2024 and draws primarily on Benchmark Mineral Intelligence (BMI) and S&P Capital IQ Pro (S&P) The unit of observation is the operator-site, distinguishing between hard-rock mines, hard-rock refineries, and integrated brine producers Production and capacity are expressed in tonnes of lithium carbonate equivalent (LCE), which converts each product’s lithium content into an equivalent quantity of lithium carbonate The joint estimation uses regional demand, measured mining costs, and refiners’ output optimality conditions to impose crossequation restrictions The joint demand elasticity is ηb = −0.511 (SE 0.172) The estimated hard-rock refining-cost slope is 4,847 The joint demand elasticity is informed by both regional demand and lithium producers’ first-order conditions The estimated brine AVC slope implies marginal-cost slopes of 9,894 and 4,781 per tonne LCE for a unit increase in utilization The residual omega˜BR rt captures unobserved carbonation costs The capacity of rival producers whose profits the brine producer does not internalize, and its interaction with log source-brine concentration, as excluded instruments Both instruments must be uncorrelated with unobserved extraction costs in the AVC equation and with the remaining cost residual in the full-site first-order condition The joint demand elasticity is informed by both regional demand and lithium producers’ first-order conditions The estimated hard-rock refining-cost slope is 4,847 Model Fit The mine optimality condition is excluded from estimation, so estimation does not require observed mine output to satisfy that condition Prediction R2 is 0.851, and RMSE is 10.86, and aggregate mine output is overpredicted by 17.25% This comparison assesses the production choices implied by the mine optimality condition, conditional on the estimated parameters and recovered cost shocks Prediction R2 is 0.851, and RMSE is 10.86 thousand recoverable-LCE tonnes Counterfactual Comparison Divestiture leaves global lithium output nearly unchanged, raising output by 0.06% The export ban instead reduces output by 25.36% and raises the world lithium price by 77.19% Divestiture substantially changes ore sourcing while leaving aggregate lithium supply nearly unchanged The export ban prevents Chinese refiners from obtaining overseas ore through either spot or affiliated purchases The comparison shows how these policies differ in their effects on ore and lithium supply Divestiture leaves global lithium output nearly unchanged, raising output by 0.06% The export ban instead reduces output by 25.36% and raises the world lithium price by 77.19% The ban prevents Chinese refiners from obtaining overseas ore through either spot or affiliated purchases The comparison matches Table 5 to the reported precision and is not repeated here The counterfactuals reveal a constraint on policies intended to reduce dependence on Chinese refining Divestiture changes who owns mineral resources, but continued trade allows Chinese refiners to purchase ore through the spot market An export restriction directly prevents overseas ore shipments to China, but existing refining capacity elsewhere cannot fully replace the lost production Investment in refining capacity outside China could therefore change the usefulness of these policies, allowing more redirected ore to support lithium production Ownership and trade restrictions can alter access to inputs, but they cannot substitute for the refining capacity needed to use those inputs The supply cost of the restriction depends on whether the refineries permitted to receive the ore have sufficient capacity to process it The counterfactuals reveal a constraint on policies intended to reduce dependence on Chinese refining Investment in refining capacity outside China could therefore change the usefulness of these policies, allowing more redirected ore to support lithium production References Asker, John, Allan Collard-Wexler, and Jan De Loecker (2019) “(Mis)Allocation, Market Power, and Global Oil Extraction,” American Economic Review, 109 (4), 1568–1615 Azar, Jos´e, Martin C.
Warren: Schmalz, and Isabel Tecu (2018) “Anticompetitive Effects of Common Ownership,” The Journal of Finance, 73 (4), 1513–1565 Backus, Matthew, Christopher Conlon, and Michael Sinkinson (2021) “Common Ownership in America: 1980–2017,” American Economic Journal: Microeconomics, 13 (3), 273–308 Barwick, Panle Jia, Myrto Kalouptsidi, and Nahim Bin Zahur (2025a) “Industrial Policy Implementation: Empirical Evidence from China’s Shipbuilding Industry,” Review of Economic Studies, 92 (6), 3611–3648 Barwick, Panle Jia, Hyuk-soo Kwon, Shanjun Li, and Nahim Bin Zahur (2025b) “Drive Down the Cost: Learning by Doing and Government Policies in the Global EV Battery Industry,” August, Working paper Berry, Steven T.
Ruth: and Philip A.
Paper summary: Ruth: So to recap this paper, it's examining how cross-border ownership shapes access to ore and the supply of lithium. The main thesis is that ownership creates competing profit incentives across mines and refineries in a way that can lower refiners’ effective ore costs, encourage mines to restrict spot sales, and soften competition.
Warren: They use global data on mines and refineries to test this idea, looking at how ownership connects these stages across borders.
Iris: The model specifically represents lithium production from hard-rock deposits and brine operations, connecting producers both along the supply chain and across these competing sources of lithium.
Ruth: The objective for an owner in their model is to maximize a weighted sum of all their owners' portfolio profits, where those weights are proportional to the equity stakes they hold in each producer.
Warren: This structure allows them to see how ownership portfolios link producers both along the supply chain and across these competing sources of lithium.
Iris: It models hard-rock mines choosing how much ore to supply to the spot market, anticipating downstream sourcing and production responses from refiners.
Ruth: Hard-rock refineries then jointly choose ore purchases and lithium output, sourcing from affiliated mines and the spot market, while brine producers make their choices on lithium output alone.
Warren: The paper is interested in how these decisions are made under different ownership scenarios because the ownership structure dictates the incentives they face.
Iris: It examines policies like divestiture of Chinese stakes in overseas producers and a ban on ore exports to China as examples of how governments try to manage this supply chain access.
Ruth: The paper's key finding is that these ownership restrictions change affiliated sourcing advantages, mines’ incentives to supply spot ore, and producers’ interests in competing firms’ profits.
Warren: That means the structure itself isn't just about who owns what; it actively changes the economic behavior of every participant in the chain.
Iris: It shows how vertical links can give mines an incentive to restrict spot supply and refiners an incentive to expand production through reduced double marginalization, which is a specific mechanism they model.
Ruth: So, when we look at the numbers, the study uses regional demand and mining costs alongside refiners’ output optimality conditions to set up those crossequation restrictions.
Warren: That’s how they get the joint demand elasticity of zero point five one one with a standard error of zero point one seven two.
Iris: Before we move on to the specific results, I want to stress that this identification is based on regional demand and lithium producers' first-order conditions, which is what makes this model work.
Ruth: Right, so we know the mechanism they are testing before we see if their specific assumptions about those people hold up in the data.
Warren: And then we look at how divesting those Chinese stakes or banning exports actually plays out when you plug in real-world data.
Conclusion: Ruth: So, looking at "Cross-Border Ownership and Input Access: Evidence from the Lithium Supply Chain," what this all boils down to is that ownership structure isn't just a legal thing; it’s an economic lever that dictates how much ore gets where and who benefits from the supply chain.
Warren: It shows that even when governments try to change who owns mineral resources, those changes don't automatically mean more lithium coming to consumers.
Iris: The implication is that policies aimed at reducing dependence on a single refining country might be constrained by the physical capacity of refineries located elsewhere.
Ruth: Precisely. Divestiture can shift ore purchases toward the spot market while keeping Chinese refiners buying through trade, meaning the overall global lithium supply doesn't really change much.
Warren: But if you try to impose an export restriction, it directly prevents overseas ore from reaching those Chinese refineries, and because existing refining capacity elsewhere can't fully replace that lost production, the global supply drops significantly.
Iris: So the practical lesson for policymakers is that they have to consider the downstream processing capability when designing these restrictions.
Ruth: That’s right. Ownership and trade restrictions can alter access to inputs, but they simply cannot substitute for the refining capacity needed to actually use those inputs effectively.
Warren: It's a constraint on policy design, not necessarily a failure of the ownership mechanism itself, but it tells us how much leverage these ownership structures really give you over global supply.
More episodes
- Daily Summary for 2026-11-19
- 2608.21274-Recommendation Quality and the Concentration of Consumption: Experimental Evidence from Netflix
- 2408.02757-A nonparametric test for diurnal variation in spot correlation processes
- 2306.16393-High-Dimensional Canonical Correlation Analysis
- 2605.17299-Geometric Brownian motion with intermittent entries and exits
- 2607.04753-Fooling Yourself: how narratives shape beliefs
- 2408.14671-Inference on Treatment Effects with Repeatedly Mismeasured High-Dimensional Controls
- 2201.06898-Difference-in-Differences Estimators for Treatments Continuously Distributed at Every Period
- 2501.15307-Influence Functions: A Unified Differentiation-Based Framework
- The day in conversation for 2026-11-19