← Back to scans

Will Marianne Lake succeed Jamie Dimon as CEO of JPMorgan?

KXJPMCEONEW-30-ML · Companies · 2026-05-10
53%
Agent
51%
Market Price
+2.0%
Edge
40%
Confidence
Volume: 6,840
Spread: 4.0c
Days to resolution: 1332
Markets in event: 4
Final Rationale
I PARTIALLY ACCEPT Challenge 1 and 3: the bimodal structure of evidence (announcement real vs. fabricated) is better handled by weighted scenario analysis than by averaging — if even 30-40% credence is placed on the April 2026 announcement being real, the conditional ~94% YES outcome pulls the overall estimate to ~0.53-0.58. I ACCEPT Challenge 2's math check; my decomposition output had an undisclosed adjustment that wasn't transparent. I PARTIALLY REJECT Challenge 4: absence in tier-1 outlets is genuine evidence against a major CEO announcement (these would be front-page WSJ/FT/Bloomberg news within hours), though I agree my search may be incomplete. The market at 51% with low liquidity is directionally consistent with the corrected analysis, and my final 0.53 reflects the bimodal weighting (0.35 × 0.94 + 0.65 × 0.40 ≈ 0.55, lightly trimmed for residual skepticism about the announcement).
Price vs. prediction
How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.
$n = 58$ follow-ups
Re-scan Context
This market has been scanned before. Previous predictions:
DatePredictedMarket PriceConfidence
2026-05-03 55% 48% 40%
2026-04-22 58% 53% 45%
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_data kalshi_orderbook article_search web_search earnings_data
Sub-questions (Fermi decomposition)
  1. Will Jamie Dimon step down as CEO of JPMorgan before January 1, 2030? (w=0.35)
  2. If Jamie Dimon steps down before 2030, will Marianne Lake be selected as the next permanent CEO over other internal candidates (e.g., Jennifer Piepszak, Troy Rohrbaugh)? (w=0.40)
  3. Is Marianne Lake still actively in contention for the JPMorgan CEO succession role as of mid-2026? (w=0.25)
Planner reasoning
This question requires decomposing a conditional probability: P(Lake becomes CEO) = P(Dimon leaves by 2030) × P(Lake selected | Dimon leaves). For sq1: Jamie Dimon has repeatedly signaled he plans to leave within '3-5 years' over multiple years, making the 2030 deadline plausible but uncertain. He has been CEO since 2005 and is currently ~70 years old. Historically, he has delayed exit signals, so even if he intends to leave, the timing may slip past 2030. The temporal window is 1332 days (~3.6 years), which is within his stated range but his pattern of delaying suggests meaningful probability he stays past 2030. For sq2: JPMorgan has historically groomed multiple internal candidates simultaneously. As of recent reports, Lake (Consumer & Community Banking) and Piepszak have been identified as top contenders. However, board dynamics, performance under pressure, and Dimon's preferences all matter. Lake's consumer banking division faces retail banking headwinds, while other candidates may have risen. This is essentially a multi-horse race with perhaps 2-4 serious internal candidates plus the possibility of an external hire. For sq3: Checking whether Lake remains in her current role and hasn't been sidelined or departed is crucial — executives sometimes leave before CEO succession events. Key reference classes: Large-bank CEO succession (Wells Fargo, Goldman Sachs, Morgan Stanley) typically involves 1-2 year transition periods announced well in advance. The 'heir apparent' often changes 1-2 times before final selection. At Goldman, succession involved multiple reshuffles. Temporal sequence for YES: (1) Dimon announces retirement/transition by ~2028-2029 at latest to allow board process, (2) Lake remains in role and performs well through 2027-2028, (3) Board selects Lake over competitors in ~2029. Each step has meaningful failure probability. The combined probability is the product of these conditional probabilities, making this lower than either component alone.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
kalshi_data OK 1.7s -
kalshi_orderbook OK 0.1s -
article_search OK 0.9s -
web_search OK 51.9s -
earnings_data OK 0.8s -
3. Evidence Synthesis Sonnet · 12 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 STRONG 55 ↑ UP web_search According to a web search summary citing FinancialContent/MarketMinute (April 8, 2026), JPMorgan officially announced Marianne Lake as Dimon's successor, with Lake assuming the CEO role on January 1, 2027, meaning Dimon is stepping down before 2030. No
2 MODERATE 70 NEUTRAL article_search As of April 1, 2026, Jamie Dimon was still actively serving as JPMorgan CEO, conducting media interviews and running operations, suggesting no imminent departure had been publicly announced by that date. Yes
3 MODERATE 72 NEUTRAL article_search A January 12, 2026 CNBC article titled 'JPMorgan's looming question: What happens when CEO Jamie Dimon leaves?' signals ongoing succession uncertainty as of early 2026, with Dimon still fully active (e.g., flying to pitch SpaceX IPO in December 2025). Yes
4 MODERATE 60 ↑ UP kalshi_data The Kalshi market KXJPMCEONEW-30-ML prices Marianne Lake as next JPMorgan CEO at 51%, implying the market assigns meaningful probability to both Dimon departing and Lake being selected before 2030. Yes
5 MODERATE 50 NEUTRAL kalshi_data The related Kalshi market KXNEWROLEJP-35DEC-MLA ('Who will be the next CEO of JPMorgan Chase? — Marianne Lake') prices Lake at only 17%, creating a significant discrepancy with the 51% market — one of these markets appears stale or mispriced. No
6 STRONG 55 ↑ UP web_search The web search summary reports JPMorgan officially named Marianne Lake as Dimon's successor on April 8, 2026, with Lake set to become CEO on January 1, 2027, resolving the multi-year succession 'bake-off' in Lake's favor over Troy Rohrbaugh, Doug Petno, and others. No
7 MODERATE 72 ↑ UP web_search Per a January 2025 Fortune article cited in the web search, Jennifer Piepszak had already exited the CEO race by January 2025, leaving Lake as the lone female contender alongside Troy Rohrbaugh and Doug Petno. Yes
8 MODERATE 60 ↑ UP web_search Lake's role was expanded in June 2025 when she took on additional responsibilities (international consumer) following another executive's departure, suggesting she was strengthening her position in the succession race. Yes
9 WEAK 45 ↑ UP kalshi_data The 51% Kalshi price on the KXJPMCEONEW-30-ML market (Lake as next JPMorgan CEO before 2030) reflects meaningful market confidence that Lake is selected if/when Dimon departs, though the 17% on the related market creates interpretive uncertainty. Yes
10 STRONG 55 ↑ UP web_search The April 8, 2026 announcement names Lake as CEO-designate starting January 1, 2027, confirming she was not only 'in contention' as of mid-2026 but had already been selected as the definitive successor. No
11 STRONG 80 ↑ UP web_search Lake has been with JPMorgan for 25+ years, served as CFO (2013-2019), CEO of Consumer Lending (2019-2021), and CEO of Consumer & Community Banking thereafter, maintaining a strong institutional position throughout. Yes
12 MODERATE 65 ↑ UP article_search No article in the search results (through May 2026) reports Lake departing JPMorgan, being demoted, or otherwise removed from contention — she remains in her senior leadership role. Yes
Information Gaps
  • The primary source for the April 8, 2026 succession announcement is a single web search result citing 'FinancialContent / MarketMinute' — this is not independently corroborated by mainstream financial news outlets (WSJ, Bloomberg, FT, CNBC) in the article search results, raising questions about its reliability.
  • The article search (19 articles) does not contain any direct reporting on a Lake CEO announcement in April 2026, creating a notable gap between the web search claim and the article-level evidence.
  • No data on Jamie Dimon's most recent public statements about his retirement timeline (e.g., 2025 annual letter, investor day remarks beyond May 2025).
  • No confirmed reporting on the board's formal succession process or vote.
  • Unclear why the two Kalshi markets (51% vs. 17%) diverge so sharply — one may be on a different resolution condition or may be stale.
  • No information on whether other passed-over candidates (Rohrbaugh, Petno) have signaled departure plans following the reported announcement.
Key Uncertainties
  • Reliability of the April 8, 2026 succession announcement: The web search cites a single non-tier-1 source (FinancialContent/MarketMinute); this could be AI-generated content, a speculative article, or a genuine announcement not yet reflected in the article search corpus — this is the single most important uncertainty.
  • The large discrepancy between the two Kalshi markets (51% vs. 17%) suggests either market mispricing, different resolution criteria, or that one market has not updated to reflect new information.
  • Even if the April 2026 announcement is real, there is a non-trivial risk the transition could be disrupted before January 1, 2027 (health issues, regulatory challenges, Trump lawsuit fallout, etc.).
  • Whether Dimon's departure is truly finalized or could be reversed/delayed (he has historically delayed exit signals).
  • The question resolves YES only if Lake is the 'first non-interim CEO' — any interim appointment before Lake formally takes over could affect resolution.
Market Context
Price: 0.51 · 7d: -0.02 · 30d: +0.03
Volume: LOW — 73 contracts/day average on primary market; 1939 on related market but only 1 data point · Depth: HIGH — depth=11516 per orderbook data, but spread is extremely wide ($0.97), suggesting thin active trading despite nominal depth
Reliability: LOW-MEDIUM — The primary market (51%) has low daily volume and a near-0 active spread suggesting illiquidity. The related market (17%) has higher volume but only 1 data point and a flat price range, making it unreliable. The 34-point discrepancy between the two markets is a red flag for data quality or market staleness.
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 46% (38% confidence)
base rate
42%
40% conf
evidence driven
53%
40% conf
contrarian
42%
35% conf
Spread
11pp
Moderate
sq1: Will Jamie Dimon step down as CEO of JPMorgan before January 1, 2030?
78%
Base rate: Large bank CEOs at age ~70 with 20+ years tenure typically transition within 3-5 years; Dimon has publicly signaled '3-5 year' timeline since 2024. Reference class of mega-bank CEO transitions (Stumpf, Blankfein, Gorman) shows announcement-to-departure within 1-2 years once succession process starts. The reported April 2026 announcement (if real) pushes UP by ~15% - but it's from a single non-tier-1 source, so I weight it modestly. Even without that announcement, given Dimon's age (70), 21-year tenure, public 3-5 year signal from 2024, and active succession process with named candidates, probability he leaves by 2030 is ~75-80%. Risk factors: he has repeatedly delayed signals before, and could stay if no successor emerges or in crisis. Final: 78%.
sq2: If Dimon steps down before 2030, will Marianne Lake be selected over other candidates?
50%
Base rate for any single named heir-apparent at major banks being selected: roughly 40-50% historically (Goldman, Morgan Stanley had multi-candidate races where the perceived front-runner sometimes won, sometimes didn't). Lake's positioning is strong: 25+ years at firm, former CFO, now runs largest division (CCB), Piepszak exited race per January 2025 reporting, role expanded June 2025. This narrows field to Lake vs. Rohrbaugh vs. Petno, and Lake has the most CEO-track profile. Pushes UP from base rate by ~10-15%. The reported April 2026 announcement (if real) would push this much higher, but uncorroborated by tier-1 sources, so I discount heavily. External hire possibility: ~10%. Lake odds among internal: ~55%. Conditional probability: 50%.
sq3: Is Marianne Lake still actively in contention as of mid-2026?
92%
No evidence of departure, demotion, or removal from contention in any search results through May 2026. Her role was expanded in June 2025. Base rate for a named top-2 contender remaining in contention over a 1-year window is high (~90%). She remains in senior leadership. Final: 92%.
Show full reasoning trace
base rate: Historical base rate for a specific named heir-apparent at a mega-bank actually becoming CEO within a 4-year window is roughly 25-35%, given (a) ~70% probability the incumbent leaves in window for an aging long-tenured CEO with public retirement signals, and (b) ~40-50% probability the named front-runner wins the race.
evidence updates: Lake's strengthened position (Piepszak exit, expanded role) pushes UP ~5-8%. Dimon's age and explicit 3-5 year signaling pushes sq1 UP. The reported April 2026 announcement is the strongest evidence but is uncorroborated by tier-1 outlets, so I weight it as ~30-40% likely to be accurate, providing modest upward pressure.
combination method: Conditional: P(sq1) × P(sq2|sq1), with sq3 as a near-certainty precondition that's correlated with sq2.
final: Combined estimate ~42%, reflecting genuine uncertainty about both timing of Dimon's departure and which candidate ultimately wins. The unverified April 2026 announcement provides upside but cannot be relied upon without corroboration.
ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.42, 'confidence': 0.4}, 'evidence_driven': {'p_yes': 0.53, 'confidence': 0.4}, 'contrarian': {'p_yes': 0.42, 'confidence': 0.35}}, 'spread': 0.11, 'individual_p_yes': {'base_rate': 0.42, 'evidence_driven': 0.53, 'contrarian': 0.42}}
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Math Error Reasoning Flaw Missing Info
Challenges
  1. The forecaster heavily discounts the April 8, 2026 announcement that Lake was named CEO-designate effective January 1, 2027, weighting it as only '30-40% likely to be accurate.' But this is flagged as STRONG evidence in the evidence list (appearing twice as STRONG/UP). If the announcement is real, the question resolves YES with near-certainty (Lake takes over Jan 1, 2027, well before 2030). The forecaster should more rigorously assess: (a) what's the prior that a fabricated announcement would persist in search results without correction by May 2026 (a month later)? (b) why would tier-1 outlets not pick it up? Either the announcement is real (~70-80%+ likely given persistence) → forecast should be ~85-90%, or it's false → forecast ~40%. Splitting the difference at 46% understates this bimodal structure.
  2. Math check: P(sq1) × P(sq2|sq1) = 0.78 × 0.50 = 0.39, not 0.457. The combined output of 0.457 doesn't cleanly follow from the stated decomposition. Either the combination includes an undisclosed adjustment or there's a calculation inconsistency.
  3. If the April 2026 announcement IS real, sq1 should be ~99% (announced succession with Jan 2027 effective date) and sq2 should be ~95%+, giving ~94%. The forecaster's framework correctly identifies this but then averages with the 'no announcement' world too aggressively. A proper weighted average: 0.35 × 0.94 + 0.65 × 0.40 = 0.59, which is higher than 0.457.
  4. The information gap that 'the article search (19 articles) does not contain any direct reporting on a Lake CEO announcement' is treated as evidence against the announcement, but it could equally reflect search coverage limitations. The forecaster should consider whether absence of corroboration in their specific search is strong evidence of falsehood.
Suggested adjustment: +8pp
The decomposition math (0.78 × 0.50 = 0.39) doesn't match the stated 0.457 output, and more importantly, the bimodal nature of the evidence (announcement real vs. fake) is being averaged in a way that understates the upside scenario. A more rigorous weighted scenario analysis suggests ~0.53-0.58 is more defensible. This is independent of the market price.
Market comparison: Forecaster: 45.7%, Market: 51%. The divergence is small and within reasonable range, but the market's slightly higher price aligns with the direction of the reasoning flaws identified (under-weighting the announcement evidence). Market liquidity is low so it's not a strong anchor, but the directional alignment with the corrected reasoning is notable.
6. Aggregator Opus · final calibrated forecast
53% (40% confidence)
I PARTIALLY ACCEPT Challenge 1 and 3: the bimodal structure of evidence (announcement real vs. fabricated) is better handled by weighted scenario analysis than by averaging — if even 30-40% credence is placed on the April 2026 announcement being real, the conditional ~94% YES outcome pulls the overall estimate to ~0.53-0.58. I ACCEPT Challenge 2's math check; my decomposition output had an undisclosed adjustment that wasn't transparent. I PARTIALLY REJECT Challenge 4: absence in tier-1 outlets is genuine evidence against a major CEO announcement (these would be front-page WSJ/FT/Bloomberg news within hours), though I agree my search may be incomplete. The market at 51% with low liquidity is directionally consistent with the corrected analysis, and my final 0.53 reflects the bimodal weighting (0.35 × 0.94 + 0.65 × 0.40 ≈ 0.55, lightly trimmed for residual skepticism about the announcement).