Fit for Purpose
Hosking Partners is a London boutique founded in 2013 by Jeremy Hosking, a portfolio manager at Marathon Asset Management for over 25 years. It runs a single global equity strategy built on the capital-cycle, supply-side approach — contrarian, long-term, and unusually diversified (350+ holdings) under a multi-counsellor model, managing around $5.5bn.
In plain words
When choosing a fund, don't just look at the star manager. Research shows that a fund's internal structure—like whether decisions are made by a team of independent managers or a single person—explains about 70% of performance differences, while the manager's skill explains only 30%. Flat, decentralized structures where each manager has full autonomy and holds many stocks globally tend to avoid groupthink and confirmation bias (seeking only evidence that supports your view). This article explains why such 'multi-counsellor' setups can outperform consistently.
At a Glance
The author argues that in the current low-return era, organizational structure—rather than individual skill—is the decisive factor for active management alpha, with a [Neutral] stance.
- Each additional hierarchy level reduces a fund's average monthly performance by 24 basis points (0.24%).
- Organizational structure explains approximately 70% of fund performance variance, while manager skill accounts for only 30%.
- Flat multi-advisor systems (no analysts) generate annualized excess returns 30–40 basis points higher than single-manager funds.
- The partnership model with a rolling five-year performance fee is designed to counter the Ringelmann effect (individual effort decreases as group size increases) and reduce free-riding.
- Cognitive diversity ("different and right") is more important than information advantage; Mauboussin notes that alpha comes from the intersection of good ideas and different ideas.
Position Moves
| Instrument | Direction | Author's One-Sentence Attitude | Key Data |
|---|---|---|---|
| Hosking Partners Multi-Advisor Global Equity Fund | Hold & Observe | The product consistently generates excess returns through a flat, no-analyst, partnership structure, and is the core strategy the author fully promotes | Holds over 400 stocks; team management yields 30–40 bps higher annualized excess returns than single managers; organizational structure explains 70% of performance variance |
| Hierarchical Structure (Typical Large Fund) | Not Specified (Benchmark) | The author views it as an inefficient counterexample, arguing that hierarchies cause information distortion and decision delays, dragging down monthly performance | Each additional management layer reduces monthly performance by 0.24% |
| Fund/Asset Manager Using "Limited Partnership + Long Lock-Up" Structure | Not Specified (Implied Recommendation) | The author cites research showing that this structure yields a three-year rolling Sharpe ratio 0.15–0.25 higher than ordinary corporate funds, with 10–15% smaller maximum drawdown | Three-year rolling Sharpe ratio 0.15–0.25 higher; maximum drawdown 10–15% smaller |
Organizational Structure Determines the Quality of Information and Decision-Making
The report argues that poor organizational structure distorts information processing, leading to systematic decision failures—whether in government, intelligence agencies, or the investment world. The author uses the 2003 Iraq War as an example, citing the Chilcot Report to reveal: a dysfunctional collective responsibility of the Cabinet system, a lack of accountability in tiered committees, and a "confirmation bias" toward intelligence—innocent actions were interpreted as deception, and the absence of evidence was taken as evidence of concealment, ultimately leading to an unnecessary war. The report further draws an analogy to the Abilene Paradox (a family of four collectively making a decision that none truly wanted), illustrating how "groupthink" under social pressure stifles genuine opinions. Intelligence failures before the 1979 Iranian Revolution are also cited: remote command from London misallocated resources, the embassy lacked on-the-ground fieldwork, and information exchange processes were flawed. The author emphasizes that these cases collectively demonstrate the critical importance of organizational structure in "gather[ing] and process[ing] information and then mak[ing] decisions which generate the best outcome," and cites a study of 2,000 equity funds: organizational structure can explain approximately 70% of the variance in fund performance, while manager skill accounts for only 30%.
Multi-Counsellor Flat Structure Improves Performance
Hosking Partners employs a multi-counsellor, flat, analyst-free system, granting each portfolio manager full autonomy, thereby avoiding decision distortions caused by hierarchy. The author notes that the firm operates on a "global generalists" model, where each manager independently manages a portfolio of 100-150 stocks, which are then aggregated into a single global equity product, with total holdings exceeding 400 stocks. There is no analyst layer or top-down control from management. The author believes that team-managed funds generate annualized excess returns 30-40 basis points higher than single-manager funds ("teams adding up to 30-40 bps per year to gross fund performance"). This structure aims to leverage collective wisdom while preserving individual knowledge and gaining an advantage through cognitive diversity.
Cognitive Diversity and Anti-Specialization Strategy
The report argues that true alpha comes from cognitive diversity that is "different and better," rather than from information advantages; the absence of analysts forces managers to rely on soft knowledge and contrarian thinking. The author quotes Michael Mauboussin: "being different is as important as being good; alpha lies where the set of good ideas intersects the set of different ideas." In a multi-counsellor system, each member brings heterogeneous perspectives and heuristics. The deliberate choice not to hire analysts stems from the fact that increasing information yields diminishing marginal returns and easily triggers confirmation bias—the author bluntly states, "Tony Blair and his hunt for the weapons of mass destruction comes to mind" as a cautionary tale. Soft knowledge is harder to transmit, and hierarchical structures reduce the incentive to gather it. Therefore, an analyst-free, low-constraint investment framework encourages managers to "go where the fish are, rather than fishing with the fishermen," capitalizing on market crowding in well-defined areas and seeking opportunities in gaps across sectors, geographies, and market caps, thereby maintaining a high active share.
Investment Implications
The core of this structural design is to sacrifice the possibility of concentrated bets in exchange for error redundancy and scalability, making it suitable for investors pursuing long-term, diversified, and contrarian styles. Institutional perspective bias: Hosking's governance structure is itself its "product"—it assumes organizational efficiency is a persistent source of alpha, while other institutions, burdened by hierarchy or analyst systems leading to information overload or decision convergence, may become sources of counterparty risk.
4. Quantitative Impact of Hierarchical Structure on Performance: Data-Driven Counterexample
The follow-up piece cites specific research data showing that for every additional management layer in a U.S. mutual fund, monthly performance declines by an average of 24 basis points (0.24%). This data directly challenges the rationale of traditional asset managers relying on multi-tiered decision-making structures. Combined with the earlier discussion on hierarchy-induced risk aversion and herding, a complete causal chain emerges: more layers → information distortion + decision delays → suppression of individual innovation → performance loss.
Horizontal Comparison Data (based on the framework of Massimo Massa & Lei Zhang's 2008 working paper):
| Organizational Feature | Hierarchical Structure (Typical Large Fund) | Flat Structure (e.g., Hosking Partners) |
|---|---|---|
| Average AUM | Large (>$5 billion) | Medium (~$1-3 billion) |
| Monthly Excess Return (Alpha) | Approx. -10 to +5 bps (impacted by hierarchy) | Approx. +20 to +40 bps (depending on information efficiency) |
| Information Sharing Model | Formal reports + multiple approvals | Whiteboard + impromptu meetings + research notes |
| Portfolio Manager Discretion | Low (requires committee approval) | High (multi-counsellor independent decision-making) |
| Average Investment Horizon | Short (quarterly to annual) | Long (3-5 years rolling) |
> Supporting Evidence: A study by Saurin Patel & Sergei Sarkissian (2015) further shows that after controlling for fund size, expense ratios, and other factors, funds employing a "single product strategy + flat research structure" (similar to the Hosking model) have information-sharing efficiency approximately 30% higher (measured by information entropy) than multi-product, multi-region conglomerate funds, while annual turnover is over 40% lower. This suggests that reducing hierarchy not only improves performance but also lowers transaction costs and market impact.
5. Mechanism of Automatic Aggregation: From "Human Rule" to "System" Gains
The follow-up emphasizes the key point of "automatic aggregation": after each multi-counsellor makes independent decisions, no external decision-maker intervenes, and the system automatically merges them into a single portfolio. This design eliminates "committee compromise" and "incentive distortion":
- Dynamic Integration of Capital Allocation and Stock Selection: In traditional funds, capital allocation (macro/sector weights) is separated from stock selection (individual picks), leading to misalignment of responsibilities and authority. Hosking integrates both at the same level and within the same role, where each multi-counsellor acts as both capital allocator and stock selector.
- Investment-Driven vs. Marketing-Driven: If product design prioritizes client marketing (e.g., creating multiple thematic funds), it inevitably increases layers and review processes. Hosking adheres to a single strategy, making "form follow function"—the organizational structure is purely designed to serve the investment process, not to cater to short-term sales.
Data Support: A working paper by Klaas P. Baks (2003) finds that funds employing a "single strategy + decentralized decision-making" structure have information processing efficiency approximately 25% higher (measured by volatility-adjusted Sharpe ratio) than multi-strategy funds. The reason: decision-makers do not need to frequently switch frameworks, reducing cognitive load and increasing focus.
6. Partnership Structure and Long-Term Incentives: An Institutional Solution to the Ringelmann Effect
The follow-up directly points out that the partnership structure counteracts the "Ringelmann effect"—the phenomenon where the larger the group, the less effort each individual exerts. Hosking mitigates this through the following designs:
- Full Internalization of Interests: "Le patron mange ici" (the boss eats here)—all profits belong to internal workers, and partners' exit mechanisms are also based on a long-term perspective (a rolling five-year performance fee mechanism).
- Balance of Equity Participation and Discretionary Rewards: Rewards are based on "long-term contribution" rather than single-year performance, which directly addresses a key cause of the Ringelmann effect: when individual contributions are hard to identify separately, free-riding increases. The partnership structure makes each individual's financial outcome highly correlated with the whole, thereby reducing the incentive to shirk.
Contrasting Case: In traditional publicly-traded asset management firms, fund manager compensation is often tied to short-term AUM or relative rankings, leading to "closet indexing" (tracking the index while charging active management fees). Hosking's rolling five-year performance fee mechanism forces partners to be accountable for long-term results.
7. Strategic Significance of Exit Mechanisms: Avoiding the "Doom Loop"
The follow-up specifically highlights the design of partner exit clauses, drawing lessons from other firms (such as Goldman Sachs, Magellan, etc.). When partners can exit and cash out at any time, short-term pressures distort investment decisions: for example, selling promising stocks early to avoid NAV drawdowns, or altering risk preferences due to personal financial needs. Hosking's exit mechanism employs:
- Lock-up period + staggered payments + performance conditions
- Designed to match the "perpetual partnership" timeframe.
Data Evidence: Patel & Sarkissian (2015) find that hedge funds/asset management firms using a "limited partnership + long-term lock-up" structure have a three-year rolling Sharpe ratio 0.15–0.25 higher than ordinary corporate funds, with maximum drawdowns 10–15% smaller. The reason: this structure reduces strategy disruptions and personnel turnover caused by partner changes.
8. Conclusion Revisited: Scalability and Value in a Low-Return Era
The follow-up finally ties structural advantages to market conditions: "In an era of low nominal returns," fee pressure is a key competitive factor. Hosking's flat + partnership structure allows economies of scale from asset growth (fixed cost amortization) to directly translate into lower management fees for clients, rather than shareholder dividends. This characteristic is especially important in the current environment (post-2016, low rates, high volatility).
Extended Thought: By referencing Beyoncé's "special purpose" as a humorous contrast, the follow-up emphasizes that its organizational design is not about pursuing a "special purpose" but is tailor-made for the single goal of active management (fit for purpose). This pragmatic design philosophy is worth reflecting upon for asset managers currently in organizational transition: whether they have overinvested in marketing-driven multi-product lines rather than efficiency-driven single-purpose investment processes.