Kashkari Says Dot Plot 'Not Helpful': 3 Global Risks Reshaping the Fed's Framework

Neel Kashkari, President of the Minneapolis Federal Reserve Bank, used a visit to Seoul to deliver what may be the most direct public criticism of the Fed's own communication tool in years. "The dot plot is mostly not helpful," he told reporters. "The market's excessive fixation on the median projection creates distortions in expectations." His core argument: in an environment of extreme economic uncertainty, relying on any single point estimate is dangerous. "A data-dependent approach is the only answer," he said.

In my view, this isn't just Kashkari being honest — it's a coordinated signal. The Fed has been conducting an internal review of the dot plot's usefulness since 2024, and the March 2026 FOMC minutes revealed that multiple participants expressed concern about how markets overreact to the median dot. New Fed Chair Kevin Warsh has already flagged a "monetary policy regime shift." Kashkari's public remarks, combined with Warsh's signals, suggest the dot plot's role will be significantly reduced or fundamentally restructured in the coming quarters.

The Fed's Internal Debate: A Divided Committee

Kashkari's public broadside against the dot plot is not an isolated opinion. The March 2026 FOMC minutes, released in early April, revealed a committee deeply divided on the usefulness of forward guidance generally. According to the minutes, "several participants expressed concern that the Summary of Economic Projections (SEP), particularly the dot plot, has led market participants to focus excessively on the median projection rather than the wide range of views across the Committee."

This internal division has been building since 2024, when then-Governor Christopher Waller first publicly questioned whether the dot plot creates "a false sense of precision" in monetary policy communication. The debate intensified after the September 2025 episode, when markets violently repriced following a dot plot that showed a 50bp higher terminal rate than the market had priced. The 10-year UST yield spiked 35bp in a single day — a move that the Fed's own internal analysis attributed primarily to the dot plot, not to any change in the economic outlook.

Several alternatives are reportedly under consideration. One proposal is to publish only the range (without the dots) showing the highest and lowest individual projections. Another is to shift to a "fan chart" approach showing probability distributions rather than point estimates, similar to the Bank of England's approach. A third, more radical proposal is to eliminate the SEP entirely and replace it with a quarterly "Monetary Policy Report" that would discuss the committee's thinking in prose rather than numbers, similar to the Bank of Japan's approach.

In my view, the most likely outcome is a hybrid: the Fed retains the SEP but deemphasizes the dot plot's median value, while introducing a fan chart or probability range as the primary communication tool. This would reduce — but not eliminate — the market's fixation on a single number while preserving the accountability that comes from publishing individual projections. I would expect an announcement of this change at the August 2026 Jackson Hole symposium, with implementation beginning with the September 2026 SEP.

Risk 3: The Passive Investing Tectonic Shift

The third risk is structural but less appreciated: the shift from active to passive investing is fundamentally reshaping how capital allocates globally, and Korea is particularly exposed to this trend. Passive funds — ETFs and index trackers — now account for approximately 55% of US equity fund assets. When passive capital flows into emerging market ETFs that track the MSCI Emerging Markets Index, Korea's weight in that index determines how much capital Korea receives.

Korea's weight in the MSCI EM Index has declined from 15.2% in 2018 to 11.8% in May 2026, as China (at 28% weight) and India (at 18% weight) have grown. This means that even if total passive flows into EM remain constant, the absolute dollar amount flowing into Korea declines each year as its index weight shrinks. For every $10 billion in net new EM passive flows, Korea now receives only $1.18 billion, down from $1.52 billion in 2018.

This structural headwind to Korean equity inflows is almost entirely ignored by the domestic financial media, which focuses on active manager sentiment. But it matters more than any single foreign investor's outlook. The continued growth of passive investing means Korea needs either (a) a catalyst that reverses its index weight decline — such as MSCI upgrading Korea to developed market status — or (b) a fundamental improvement in active manager returns that draws discretionary capital despite the passive headwind.

My view is that Korea's MSCI developed market upgrade is unlikely within the next 12 months due to the short-selling ban and currency convertibility concerns. This means the passive weight decline continues. The implication for global allocators is straightforward: to outperform the EM benchmark in a passive-dominated world, you need to actively overweight Korea — which is exactly what active managers have been refusing to do.

A Coordinated Reading of the Three Risks

What makes the current moment particularly challenging is that these three risks are not independent. The Hormuz oil shock pushes global inflation higher, which gives the Fed less room to cut rates even if the AI sector corrects. An AI correction reduces global growth expectations, which normally would support bond prices, but the oil shock and supply chain disruption argue the opposite. The Fed's communication uncertainty compounds both by making it harder for markets to price the probability of each scenario.

This interconnected risk structure creates a regime where traditional 60/40 portfolio diversification breaks down. Bonds do not provide their usual hedge against equity declines because both are being pushed by the same inflation pressure. Gold, which has rallied 32% year-to-date to $3,200/oz, has been the primary beneficiary — and in my view still has room to $3,500-4,000 if the Hormuz crisis deepens.

For Korean investors specifically, the recommended portfolio adjustments are: increase commodity and energy exposure by 10-15%, hedge USD/KRW through forward contracts or KRW-linked ETFs, shift semiconductor allocation to non-memory AI beneficiaries (power management chips, cooling equipment), and maintain a 5-10% cash reserve for the corrections that will inevitably come when one of these risks triggers a broader risk-off event.

The base case I'm using for H2 2026 is: oil remains above $100/bbl, the Fed delivers one more hike in June then holds through year-end, AI capex growth slows to 25% (from 40%+), and the KOSPI trades in a 7,200-8,400 range with significant volatility. The bull case (resolution of Hormuz, AI demand accelerates) takes KOSPI to 9,000. The bear case (Hormuz escalation, AI correction, recession) takes KOSPI to 6,500. I assign 50% probability to base, 25% to bull, and 25% to bear.

Risk 1: Hormuz Strait — The 6-Month Supply Chain Shock

The first and most immediate risk is the prolonged Hormuz Strait blockade. At least 21% of the world's daily oil volume (17 million barrels) transits this strategic chokepoint. Since the blockade began in April 2026, international crude prices have surged 43.6% from $78 to $112 per barrel — exceeding the 38% spike during the 2022 Russia-Ukraine war.

The secondary effects are more concerning. The Shanghai Containerized Freight Index has exploded 84.9% from 1,850 in early April to 3,420 in late May — the highest since January 2022. For Korean exporters, logistics costs have risen from 4.2% of revenue to 7.8%, a 3.6 percentage point increase that directly pressures margins. Kashkari himself noted during the press conference that "supply chain shocks could act as an upside risk to inflation."

Hana Securities analyst Kim Young-hoon estimates that "if the Hormuz risk persists beyond 6 months, Korea's terms of trade could deteriorate by 8-12%, with refining and petrochemical stocks bearing the brunt." I see this as the single biggest near-term risk to the global inflation narrative. The 6-month normalization timeline is optimistic. My base case is 9-12 months for full supply chain normalization, which means the "transitory inflation" narrative that markets keep hoping for is unlikely to materialize in 2026.

The European Central Bank and the Bank of England have already signaled revisions to their inflation forecasts in June and July respectively to account for the Hormuz disruption. The Fed will almost certainly follow.

Risk 2: The AI Investment Bubble and Korea's Semiconductor Exposure

The second risk is one that hits Korea directly. Global AI-related capital expenditure has surged from $215 billion in 2024 to $342 billion in 2025 and an estimated $480 billion in 2026 — 40%+ annual growth. But data center power consumption is outstripping infrastructure supply. North American data center power demand is projected to grow 143% from 35GW in 2024 to 85GW in 2026, yet grid infrastructure expansion is falling 6-12 months behind construction schedules.

Kashkari warned that "if a correction occurs in the AI sector, the Korean stock market could be significantly affected." This is an understatement. Korea's semiconductor exports account for roughly 18% of total exports, with Samsung Electronics and SK Hynix representing approximately 30% of KOSPI market capitalization. An AI investment pullback — triggered by data center delays, rising power costs, or disappointing model performance — would hit Korean equities disproportionately hard.

Looking at AI infrastructure spending, I see clear parallels to the 2001 fiber-optic overbuild. The technology is transformative, but the capex cycle is running well ahead of proven revenue models. Cloud providers are spending on AI capacity the same way telcos spent on fiber in 1999-2000, betting that usage will follow. It usually does — but not always fast enough to prevent a 40-60% correction in the suppliers' stocks first.

My recommendation: reduce semiconductor exposure in KOSPI portfolios and rotate into AI-adjacent beneficiaries with less direct exposure to the capex cycle — specifically Korean data center REITs, power equipment manufacturers, and cooling solution providers.

Risk 3: The Regime Change in Fed Communications

The third risk is the most abstract but potentially the most consequential. If the Fed significantly changes or abandons the dot plot, the market loses its single most important forward guidance tool. The initial reaction in bond markets would likely be a volatility spike as traders recalibrate without their anchor reference point.

Kashkari's critique is intellectually honest: the dot plot creates a false precision that markets over-interpret. But the cure might be worse than the disease. Without the dot plot, every FOMC statement, every speech, every interview by Fed officials takes on outsized importance. The Fed's communication strategy would shift from "here's what we think" to "here's what the data tells us" — a fundamentally different approach that would force markets to become more data-driven rather than Fed-driven.

In my view, this is actually a positive development for sophisticated investors. A more data-dependent Fed means less reliance on reading tea leaves from 19 dots and more focus on actual economic data. The transition period, however, will be volatile. I would expect the 10-year UST yield to trade in a wider range — 4.25-5.25% — during the first 6 months of any communication framework change.

What This Means for Global Portfolios: My Take

Here's how I see the interplay of these three risks: the Hormuz supply shock, AI capex bubble concerns, and Fed communication uncertainty create a uniquely volatile setup for H2 2026. All three are negative for Korean semiconductors and positive for energy/defense plays.

I see three concrete portfolio actions: (1) reduce semiconductor exposure in KOSPI by 20-30%, (2) add Korean defense and shipbuilding stocks that benefit from both geopolitical tensions and the global rearmament cycle, and (3) increase cash allocations to 10-15% to have dry powder for the inevitable pullback when one of these risks materializes.

The biggest risk in my assessment is actually the AI correction — it's the one nobody is genuinely hedging because it's been the winning trade for 24 consecutive months. That's exactly when corrections happen.

Key Data Points

• Crude oil: $78→$112/bbl (+43.6%) since Hormuz blockade (April 2026)
• Shanghai Container Index: 1,850→3,420 (+84.9%)
• Global AI capex: $215B (2024) → $480B (2026E), +40% CAGR
• North America data center power: 35GW→85GW (+143%)
• Korea semiconductor exports: ~18% of total exports
• Samsung + SK Hynix: ~30% of KOSPI market cap
• Korean export logistics cost: 4.2%→7.8% of revenue
• 10-year UST yield range during transition: 4.25-5.25%

🔍 Related Keywords

• Federal Reserve dot plot reform 2026
• Hormuz Strait blockade oil price impact
• AI investment bubble Korea semiconductor risk
• Global supply chain disruption 2026
• Kevin Warsh monetary policy regime shift