● Tesla FSD Tsunami, 10 Billion KM, Europe on Edge

European RDW’s February 2026 FSD Target Schedule, Direct Hit on Chinese-made Model 3·Y, and the New Economics of Autonomous Driving Built on 10 Billion km of Data

The Core Points Contained in This Article

• European RDW has set February 2026 as the target for Tesla FSD approval, and this news article quickly summarizes the regulatory implications in a news format.
• It explains how the approval of FSD for Chinese Shanghai-made Model 3·Y could reshape the sales and subscription revenue structure.
• It highlights the true significance of Tesla Autopilot’s achievement of 10 billion km of real-world driving data and pinpoints the critical moment in AI training competitiveness.
• It reinterprets the choices available to Korean companies from a data and AI perspective, including Hyundai-NVIDIA collaboration and robot foundries.
• It organizes points that are rarely covered by other YouTube channels or news outlets (UNECE/AI Act/Cybersecurity·SOTIF·OTA conformity·Insurance/liability frameworks) in a separate section.
• It presents scenarios from a global economic perspective, outlining the implications when autonomous driving and electric vehicles intersect with investment strategies, productivity, and interest rate environments.

News Summary

• It has been reported that the Dutch vehicle authority RDW has proposed February 2026 as the target timeline for Tesla’s FSD certification process.
• Although RDW is not the agency that ‘decides’ the overall EU standard, its approval within the EU type approval (WVTA) system is practically significant as approval in one location is recognized throughout the region.
• According to Tesla’s official Chinese channel, Autopilot’s real driving data has surpassed 10 billion km, which is interpreted as experiential evidence to be submitted to regulators.
• Europe’s conservative stance on autonomous driving regulation is expected to shift towards relaxation and refinement after 2024, and around 2026, a commercialization pathway combining the AI Act, UNECE R155/156 (cybersecurity·OTA), R157 (ALKS), and SOTIF (ISO 21448) is anticipated to be concretized.
• If FSD becomes available even for Chinese Shanghai-produced Model 3·Y, Tesla’s software margins could significantly improve due to the simultaneous increase in both sales and subscriptions in Europe.
• Elon Musk has expressed skepticism about the idea of licensing FSD to traditional automakers, complicating the calculations for OEMs who are considering in-house development, partnerships, or partial adoption.

The Implication of RDW’s February 2026 Target Schedule

• RDW is one of the leading type approval agencies in the EU and has been a key channel that Tesla has utilized in the past.
• The disclosure of a ‘target timeline’ signals that the process has moved beyond document review to on-site inspections and verification roadmaps.
• When a type approval from one national agency is mutually recognized throughout the EU via the WVTA system, the adoption rate among member countries can be accelerated.
• However, detailed restrictions such as operational design domain (ODD), mapping, speed, weather, highway/urban areas may vary from country to country.

The Critical Moment of 10 Billion km Data: Why Now Is Important

• The enhancement of autonomous driving performance hinges more on ‘real-world driving data’ and ‘efficiency in AI training’ than on hardware.
• Achieving a scale of 10 billion km serves as a threshold that increases sample density for learning long-tail events, functioning as persuasive evidence of safety improvements for policymakers.
• In the end-to-end network (version 12 series), the surge in data is likely to lead to enhanced generalization and suppression of regression errors.
• For regulatory submissions, not only the ‘quantity’ but also the ‘quality’ is crucial. Data label quality, tracking of regression in failure cases, scenario coverage, and simulation evidence are all required.

Update of the European Regulatory Framework: 2024~2027

• UNECE R155/156: These impose mandatory requirements on cybersecurity and software updates (OTA), linking high-function OTA such as FSD directly to the conditions for maintaining type approval.
• UNECE R157 (ALKS): This covers highway-level 3 systems, with discussions on incremental expansion, though the degree of adoption will vary by country.
• ISO 21448 (SOTIF): Beyond functional safety, it addresses ‘safety of intended functions,’ systematically managing residual risks in data-based learning functions.
• EU AI Act: High-risk AI, considered a safety component, will be subject to obligations on conformity, data governance, logging, and transparency, with full enforcement expected around 2026~2027.
• In summary, 2026 could become a turning point where the RDW target timeline aligns with the requirements of the AI Act, cybersecurity·OTA, and SOTIF.

Impact of Chinese-made Model 3·Y: A Scenario of Simultaneous Surge in European Sales and Subscriptions Upon Approval

• The key point is that regardless of the country of manufacture, if the vehicle meets EU technical standards, type approval is possible.
• Chinese-made Tesla vehicles already hold a significant share of European sales, so approval could lead to a rapid expansion in subscription customer numbers.
• Combining vehicle sales (hardware) with FSD subscriptions (software) enhances price elasticity and can offset hardware margin pressures with software margins.

Business Model Adjustment: Subscription ARPU and Margin Leverage

• Hypothetical scenario: With a €100 monthly FSD subscription in the EU and an activation rate of 15–25%, based on a cumulative 2 million Tesla vehicles in Europe, annual recurring revenue could range between €360 million and €600 million.
• If the activation rate exceeds 30%, a strategy to boost market share by lowering hardware prices and recouping revenue through software sales will be effective.
• The gross margin on software sales is significantly higher than that of hardware, so the proliferation of autonomous driving could lead to a re-evaluation of company valuations (a software premium in stock markets).
• This contributes to valuation defensibility through enhanced “subscription-based visibility,” regardless of fluctuations in global interest rate environments.

Competitors’ Strategies: Licensing vs. In-house Development vs. Partnership

• Musk has expressed skepticism about licensing FSD to traditional automakers, while many OEMs are in a mode of either selectively adopting it or carrying out long-term reviews.
• In-house development requires creating what is effectively “another AI company” encompassing GPUs, data lakes, simulations, and edge computing, which is extremely challenging.
• Partnership models can quickly roll out region-specific services that rely on mapping (HD maps) or LiDAR; however, they face limitations in terms of universality, cost, and scalability.
• In conclusion, between 2026 and 2027, the market is likely to polarize between ‘universal end-to-end expansion’ and ‘region-specific/feature-limited services.’

Korean Perspective: Hyundai-NVIDIA, Robot Foundries, and Strategic Choices

• Hyundai’s exploration of a collaboration with NVIDIA and its plans for robot foundries are timely as they prepare for the “physical AI” era.
• The key decision is whether to internalize autonomous driving or to adopt an external stack.
• The advantage of internalization is securing data and software sovereignty, while the disadvantages include constraints on time, capital, and talent.
• An external stack offers speed and cost advantages but carries structural dependency risks as the vehicle’s “brain” and data flow externally.
• For the Korean ecosystem, a strategy that focuses on maintaining a global edge in specific value chains—such as mapping, simulation, verification (V&V), safety standards (SOTIF/cybersecurity), sensors, and edge AI chips—would be effective in integrating into the global supply chain.

Investment Strategy Points (from a Global Economic Perspective)

• Despite debates over peak interest rates, an increasing share of repetitive software revenue is likely to be awarded a valuation premium.
• It is possible to adopt a diversified investment approach by investing in AI infrastructure (NVIDIA-related GPUs, power & data centers, optics & storage devices) and essential autonomous driving components (cameras, electronics, cybersecurity).
• As price competition in electric vehicle hardware is expected to intensify, focusing on companies that integrate software and services (subscriptions, insurance, mapping, V2X) is a rational strategy.
• Given the significant policy risks, it is essential to monitor European regulatory milestones (R155/156 and the application timing of the AI Act) as well as recall/OTA issues.

Key Details Rarely Covered Elsewhere (Regulation, Evidence, and Liability)

• Conformity evidence package: It is essential to include metrics covering scenario coverage, the reproduction and regression testing of failure cases, simulation-to-reality gap analysis, and curves showing safety improvements with each update.
• Cybersecurity (R155) and OTA (R156): When functionality is altered through OTA, re-examination of the conditions for maintaining type approval is required, along with mechanisms for operating data logging and rollback procedures.
• SOTIF (ISO 21448): Residual risks in abnormal conditions (such as snow, rain, backlighting, or damaged signs) must be quantified, emphasizing the importance of securing diverse cases over sheer data volume.
• AI Act: With the expansion of obligations related to data governance, logging, transparency, and human override for high-risk AI, investments in explainability and debugging systems become essential.
• Insurance and Liability: The allocation of responsibilities differs between Level 2+ and Levels 3/4, so aligning country-specific insurance rates, exemptions, and accident investigation (EDR data) systems is crucial for commercialization speed.

Risks and Counterarguments

• Safety incidents and media issues directly impact the approval timeline.
• Performance discrepancies may remain in complex urban intersections, densely populated pedestrian areas, and adverse weather conditions.
• ODD restrictions and geofencing specific to individual countries may be essential initially and will determine the upper limits of subscription activation rates.
• Bottlenecks in GPUs, power supply, and data centers could constrain training speed.
• Underestimating competitors’ ‘region-specific’ services could result in losing market share in certain regions or fleets.

Timeline Scenarios (2025~2027)

• 2025: Limited expansion in select countries, refined compliance with OTA regulations, and strengthened evidence through data and simulations.
• First half of 2026: Conditional approvals and pilot expansions around the RDW target timeline, with country-specific applications of limitations related to ODD, speed, and road types.
• Second half of 2026 to 2027: Full implementation of high-risk requirements under the AI Act, integration of cybersecurity/OTA conformity, and a gradual rise in subscription activation rates.

What to Prepare

• For companies: Establish dedicated teams for responding to SOTIF, R155/156, and the AI Act; enhance data governance and logging infrastructure; and reinforce the simulation-to-real-world capture loop.
• For governments and regulatory bodies: Design phased ODDs and geofencing strategies; standardize accident investigation and EDR data; and expedite the finalization of insurance and liability guidelines.
• For individuals and investors: Compare subscription prices, feature ranges, and update cycles, and evaluate electric vehicle purchases using a “time returned per day” metric. Investment strategies should be linked to software share and regulatory milestones.

< Summary >

• The RDW’s February 2026 FSD target schedule holds significant practical implications within the EU type approval framework and could act as a catalyst for the proliferation of FSD even in Chinese-made Teslas.
• Tesla’s achievement of 10 billion km of data is highly significant in terms of regulatory persuasion and marking a critical point in AI performance; the period between 2026 and 2027 may become a major turning point in commercialization with the convergence of the AI Act, R155/156, and SOTIF.
• As hardware margin pressures are offset by software subscriptions, a new investment strategy emphasizing “electric vehicles + autonomous driving” is emerging.
• For Korea, whether to internalize or adopt an external stack, combined with a focus on niches such as safety, cybersecurity, and simulation, could be the key to achieving a global edge in specific value chains.

[Related Articles…]

• Key Summary of the EU AI Act and Changes in Autonomous Driving Regulations
• Profitability Analysis of Tesla’s FSD Subscription Business Model

*Source: [ 오늘의 테슬라 뉴스 ]

– “테슬라 FSD, 유럽이 움직였다… RDW의 2026년 2월 일정! 자율주행 판도가 완전히 바뀐다”

---

● TPU Coup Threatens Nvidia, Tesla’s Driverless Tsunami

AI Chip War Intensifies: A Snapshot of the Google TPU–Meta Alliance, Nvidia’s Counterattack, and Tesla Robo-Taxi Acceleration

It covers the reality behind Google TPU’s dominance narrative and how Meta’s plan to rent then purchase TPUs is creating shifts in investment strategies.
It interprets Nvidia’s universality strategy and the impact of TSMC’s semiconductor supply chain with supporting data.
It examines how rapidly Tesla’s robo-taxi fleet is expanding, alongside the timeline for removing safety drivers.
It bundles changes in mobile phone usage alerts mentioned in FSD v14.2, the teacher–student model compression strategy, and signals from Norway’s new record as well as insights from China.
It concisely covers the realization of the autonomous driving milestones foretold in Master Plan 2 nine years ago, Cathie Wood’s evaluation, and the implications of Elon’s social statements on corporate value.
It explains comprehensively how AI trends and digital transformation in the global economy are reshaping the landscape of technology investment, from data centers to the edge.

Today’s Key News Briefing: AI Chip War and the Market

Google’s next model, ‘Gemini 3’, emphasizes perceptible performance improvements over the previous generation and prominently features its proprietary ASIC, the TPU, as its learning infrastructure.
Meta announced its plan to rent then purchase Google’s TPU, and with this news, Alphabet and Meta saw gains while Nvidia experienced significant early-session drops followed by a recovery in volatility.
Nvidia reiterated its commitment to “universality, compatibility, and ecosystem support, enabling any model to run anywhere instantly,” in an official comment.
Despite TSMC being the key foundry producing both Nvidia GPUs and Google TPUs, it exhibited simultaneous early-session weakness, suggesting short-term inefficient pricing.
In summary, the scale advantages of general-purpose GPUs and the competitive total cost of ownership (TCO) efficiency of specialized ASICs are coming to the forefront, reshaping risk and opportunity allocation across the semiconductor supply chain.

Impact on Tesla: The Essence of Robo-Taxis and FSD

Observations indicate that the tracked number of robo-taxis in the Bay Area increased from 63 to 92 in a short period, aligning with the goal of operating 1,000 units by year-end and highlighting the accelerated expansion pace.
The current bottleneck lies in the availability of safety drivers rather than vehicle production, and if their removal is expedited, expansion could reach 100 to 200 units per day.
Regarding FSD v14.2, there are user reports suggesting that in-driving mobile phone usage alerts have been relaxed, although adherence to regional regulations and safety standards remains paramount.
Elon has reinforced the message that “the essence is not the vehicle but FSD,” shifting advertising focus toward autonomous driving while prioritizing maximization of revenue from proprietary services over licensing.
The data accumulation on the scale of “billions of miles,” foretold in Master Plan 2 from 2016, is now becoming a reality, with approval processes from various national regulatory bodies gradually underway.

The Key to Model Compression: The Teacher–Student Strategy

Knowledge distillation is in full swing, with colossal “teacher” models, which can only run in data centers, instructing compact “student” models designed for vehicle deployment.
The objective is to achieve inference efficiency that runs smoothly on HW4 and, with further compression, can extend to HW3.
This aligns with the digital transformation strategy that shifts the focus of AI trends from data centers to the edge, lowering both vehicle costs and OPEX.
Some communities have claimed that the next-generation ASICs from Tesla and xAI offer “the same performance at 10% of the cost” compared to Nvidia, although this remains unverified and should be approached with caution.

Demand and Performance Signals: Norway’s New Record and On-the-Ground Insights from China

In Norway, cumulative sales by the second month of the quarter are set to break historical records, and the brand secured the top spot in annual registrations.
In China, on-site videos from delivery centers in cities like Guangzhou suggest a strong flow of orders; however, until monthly statistics are formalized, these remain qualitative indicators.
The 2026 North and Central American World Cup is seen as a global event that could rapidly boost public awareness of robo-taxis, with discussions suggesting that value recognition might begin as early as mid-2025.

Key Market Storylines and a Tech Investment Checklist

Short-term stock trends are driven by news sensitivity and positioning, frequently leading to inefficiencies; therefore, multi-layer diversification across GPUs, ASICs, foundries, and services can be effective in a portfolio.
The interdependence among Nvidia, Alphabet, Meta, TSMC, and Tesla is high, and the optimal chip selection varies depending on model architecture, workload, and software stack.
The checklist should include the timeline for phasing out safety drivers, the pace of regulatory approvals, the TCO curves for chips, power, and cooling, bottlenecks in the semiconductor supply chain, as well as the alignment of insurance and liability frameworks.
Changes in interest rates, electricity costs, and CAPEX cycles within the global economic environment are key variables that influence the expansion pace of AI data centers and robotics.

Cathie Wood’s Assessment and Elon’s Social Statements: Reputation Risks

Cathie Wood described Elon as an “innovator and a leader with a benevolent vision” and emphasized his impact on the sectors of space, robotics, and energy.
Elon actively shares his personal views on controversial issues such as healthcare, the judiciary, and immigration, which could influence the brand’s polarization and the dialogue flow with regulatory authorities.
Claims made in the fields of health and law are subject to scientific scrutiny and ongoing policy debates, making it necessary for investors to separate facts from opinions when managing risks.

The Core Points Overlooked Elsewhere: Investor Insights

The rise of TPUs is not merely about chip replacement but about reconfiguring the TCO through software-hardware co-optimization, redrawing cost curves from the model design stage onward.
TSMC remains a player that attracts CAPEX regardless of the winner, and since product mix and process node utilization determine margins, volatility can present investment opportunities.
The real bottleneck for expanding robo-taxis isn’t vehicle production but rather the availability of safety drivers, insurance, local government regulations, and operational data; solving these four issues could lead to exponential growth.
Changes in FSD’s UI and notification policies may signal technological maturity, but they also serve as a trust test with regulators and insurance companies, so it’s crucial to closely monitor regional adjustment speeds.
Model compression isn’t about sacrificing performance, but rather achieving a three-pronged strategy of problem redefinition, data curriculum, and on-device optimization, reducing chip dependency risks and enhancing profitability.

< Summary >

The alliance between Google’s TPU and Meta is reshaping the AI chip landscape around TCO, while Nvidia is responding with its advantage in universality and ecosystem support.
TSMC, being the common denominator in the supply chain, is likely to continue reaping structural benefits despite short-term volatility.
Tesla is accelerating the expansion of its robo-taxi fleet and its FSD-focused strategy, while enhancing inference efficiency from HW4 to HW3 through model compression.
Demand signals from Norway and China are robust, and the 2026 World Cup provides a momentum for mass exposure.
Investors need to manage key risks such as regulations, safety drivers, TCO for power and chips, and insurance frameworks.

[Related Articles…]

AI Chip Economics Disrupted by TPUs: Reshaping the Cost Curve

Commercialization of Robo-Taxis and the Regulation Economy: Checking the 2026 World Cup Momentum

*Source: [ 허니잼의 테슬라와 일론 ]

– AI 칩 전쟁의 시작! 테슬라 영향은? 엔비디아 GPU vs 구글 ASIC TPU / 본격 로보택시수 급증! 테슬라의 대세 상승은 다음 월드컵이다?!

---

● Pension Offlimits, Won Collapse, Liquidity Shock Looms

[Immediate Analysis] A Summary of the ‘Real’ Issues Behind the Won’s Weakness: The Meaning of Ku Yun-chul’s “No Use of the National Pension Fund”, the Structural Reasons Why the Exchange Rate Remains Unstable, Policy Scenarios, and Even AI & Stablecoin Liquidity

This article covers 1) the signal “No use of the National Pension Fund” sends to the foreign exchange market, 2) the true structural reasons behind the persistently high won–dollar exchange rate, 3) the cards and limitations the government and the Bank of Korea can actually play, 4) three potential exchange rate scenarios, and 5) the ripple effects of liquidity from AI and stablecoins that are rarely addressed elsewhere.

It consolidates global economic variables and the microstructure of the foreign exchange market in one place so that it can be used for making decisions immediately today.

Breaking Summary: Deputy Prime Minister Ku Yun-chul’s Key Message and Its Market Implications

– “No use of the National Pension Fund (NPS) to stabilize the exchange rate.”

– The four-party working group (Ministry of Economy and Finance, Ministry of Health and Welfare, Bank of Korea, and National Pension Service) is not discussing short-term ‘market intervention’ but rather long-term institutional design to enhance the pension’s profitability and payment stability.

– If the National Pension Fund’s assets exceed 3,600 trillion won in the future, the impact of exchange rate risk arising from overseas investments will increase significantly.

– Amid an exchange rate reaching the 1,470 won level, speculative trading and price concentration are being monitored, and decisive action will be taken if volatility increases.

– Using the pension fund for purposes other than those prescribed by law and institutional objectives will not happen. Stabilizing the foreign exchange market is merely a ‘result’ within the scope of maximizing profitability, not an ‘objective’ in itself.

News-Style Immediate Analysis: Points the Market is Reading Right Now

– Short-term verbal intervention has already been executed.

– The willingness to manage the upper threshold (around the psychological line of 1,500 won) has been signaled to the market.

– However, an expansion of ‘direct intervention’ is being approached with caution. Any circumvention of intervention through the National Pension Fund is denied.

– The Bank of Korea’s rate hike tool is advantageous for exchange rate stabilization, but given concerns about the economy, household debt, and real estate, it remains a dilemma.

– Although encouraging corporate hedging of exchange rate risk or offering short-term incentives is “not under consideration at present,” such options remain open if needed.

The Real Reasons Behind the Won’s Weakness: Surface News vs. Structural Forces

1) Combination of Global Rate and Fiscal Risks.

– Uncertainty over the U.S. rate-cut trajectory and concerns over major countries’ fiscal deficits are strengthening the dollar.

– Volatility in commodity and energy prices shakes expectations regarding Korea’s terms of trade and trade balance.

2) Domestic Structural Demand for Foreign Currency.

– Continuous overseas investment by pension funds, insurance companies, and institutions creates a steady demand for dollars.

– Fund flows intensify KRW selling and dollar buying pressures during rebalancing periods.

3) Liquidity Gap (the Core Factor).

– The market is watching “who releases money faster.”

– In recent years, when the growth rate of Korea’s broad money (M2) and credit creation has lagged behind that of major countries, this gap has been observed to weaken the won’s fundamentals.

– The key point is not just one or two numbers, but that the “relative pace of liquidity growth” determines the trend direction of the exchange rate.

4) Market Microstructure.

– Price discovery is led by NDF (offshore foreign exchange derivatives), while deterioration in the forward exchange curve, exchange rate insurance, and swap points increases hedging costs for companies.

– Authorities’ smoothing operations (fine adjustments) are effective in mitigating large swings, but they cannot reverse structural factors like the liquidity gap.

Three Policy and Market Scenarios

– Scenario A: Defense within the upper range under managed volatility (e.g., 1,450–1,520 won).

Verbal intervention combined with passive smoothing to block sharp rises.

The Bank of Korea maintains its baseline assumption of keeping the benchmark interest rate unchanged.

– Scenario B: Gradual downward movement as external variables ease.

If the U.S. rate-cut trajectory becomes clear and global risk appetite recovers, re-entry into the 1,400 won range could occur.

A narrowing liquidity gap would enhance downward momentum.

– Scenario C: Testing the upper range again in the event of a recurrence of external shocks.

If energy prices suddenly surge, geopolitical risks intensify, China’s economic slowdown deepens, and U.S. fiscal instability worsens simultaneously, the 1,500 won level may be revisited.

In this case, the intensity of authorities’ interventions may increase.

Policy Options and Limitations: What Can Be Done and What Cannot

– Use of the Pension Fund is Not Permissible.

In terms of legal and institutional appropriateness, using the National Pension Fund directly to stabilize the exchange rate is prohibited.

– Available Options.

Strengthening verbal intervention, smoothing operations, fine-tuning foreign exchange soundness regulations, reinforcing corporate hedging guidelines, and targeting specific zones of concentration and speculation.

– Cautious Options.

Although raising interest rates immediately stabilizes the exchange rate, it can have adverse side effects on growth, debt, and real estate.

Short-term incentives for companies are “not under current consideration.”

– Structural Options (Core).

Gradually adjusting the pace of domestic liquidity growth and deepening the foreign exchange market to reduce distortions in price discovery are necessary.

Institutional enhancement of the framework for managing exchange rate risks for pension funds and institutions (e.g., easing procyclicality in evaluation currencies and rebalancing rules) is required.

What Others Rarely Address as the ‘Most Important Point’

1) Asymmetry between the Pension’s Evaluation Currency Risk and the Exchange Rate.

– The performance of the National Pension Fund is accounted for in the won.

– Even with the same rate of return, the impact on the pension’s finances differs greatly depending on the exchange rate level.

– Particularly when the timing of large-scale rebalancing coincides with periods of a strong or weak won, it can impart an asymmetrical shock on long-term finances.

– The true purpose of this four-party working group is to create a ‘rule’ that institutionally addresses this structural issue.

2) The Paradox and Moral Hazard of the ‘Upper Threshold Signal’.

– Signaling an upper threshold to the market helps achieve short-term stability.

– At the same time, repeated short covering and long building ahead of that line can exacerbate volatility.

– Authorities must minimize the predictability of zones and timing to curb front-running.

3) The Shadow of AI & Stablecoin Liquidity.

– Global AI trading, CTA, and high-frequency strategies react to interest rate and liquidity signals by simultaneously increasing positions.

– Issuance and redemption of stablecoins are peripheral indicators of dollar liquidity, yet they quickly reveal the pace at which risk appetite shifts.

– Korea’s high proportion of crypto trading means that changes in risk appetite can transfer to won sentiment.

– In other words, the synchronization of trades created by AI and the stablecoin cycle could amplify volatility in the foreign exchange market.

Checklist for Investors and Corporate Practitioners

– Corporate Finance.

Exporters: Manage upper range risks through diversification of forward rollover positions and option collar structures.

Importers: Consider partial hedging during periods of rising input costs and diversification of payment currencies.

– Individual Investors.

Diversify with currency hedging ETFs and multi-asset strategies; adjust dollar exposure based on scenario ranges.

Chasing short-term spikes in the exchange rate carries significant volatility risk.

– Data Watch.

Monitor the liquidity gap (M2 and credit) between Korea and the U.S., trends in Korea’s foreign exchange reserves, the divergence between NDF and spot rates, swap points, trade balance and oil prices, and the probability of the Fed’s rate path.

The Core of the Bank of Korea’s Dilemma

– From an exchange rate perspective, a rate hike seems rational.

– However, considering the risks to the economy, household debt, and real estate amid easing inflation, keeping rates unchanged remains the default option.

– Ultimately, balancing interest rates, the exchange rate, and liquidity is key, and stabilizing the foreign exchange market requires a combination of ‘interest rate, liquidity management, and structural reforms.’

One-Line Conclusion

– The declaration “No use of the National Pension Fund” signals the maintenance of long-term institutional consistency.

– The core of the won’s weakness lies in the liquidity gap and structural demand for foreign currency.

– In the short term, authorities will reduce volatility through verbal and fine adjustments, while mid-term efforts will focus on liquidity management, institutional reform, and market structure improvement.

– Considering the ripple effects of global liquidity movements driven by AI and stablecoins, the exchange rate is likely to remain in the upper boundary area under managed volatility.

< Summary >

– No Pension Fund Usage, Yes Long-Term Institutional Design.

– The core of the won’s weakness is the liquidity gap + structural foreign currency demand.

– Short term: verbal and fine-tuned intervention; mid term: liquidity management and market structure enhancement.

– For the Bank of Korea, balance is more of a challenge than interest rates.

– AI and stablecoins act as amplifiers of volatility.

[Related Articles…]

Won Weakness and the National Pension: Proposing Structural Solutions

Foreign Exchange Market Volatility and the Dilemma of Interest Rate Policy

*Source: [ 경제 읽어주는 남자(김광석TV) ]

– [속보] ‘원화약세’ 환율불안의 진짜 이유 : 구윤철 “외환시장 안정에 국민연금 동원 아니다” [즉시분석]

---