Site Logo

STPrism

Geoffrey Vickers

5 min read

Based on the provided sources, Professor Michael Mainelli’s risk-case-study, which advocates for “Environmental Consistency Confidence” and “Scientific Management” in financial risk, can be interpreted through Peter Checkland’s Soft Systems ideas as an attempt to formalize the Appreciative System of an organization.

Here is an interpretation of the Mainelli case study through the lens of Checkland’s concepts:

1. KRIs as “Standards” in the Appreciative System

Mainelli’s central tool, the Key Risk Indicator (KRI), functions identically to what Checkland (drawing on Vickers) identifies as Standards or Norms within the appreciative system.

• Mainelli’s View: KRIs are metrics (e.g., trading volume, IT downtime) used to predict losses. He argues that many so-called indicators are merely “unvalidated opinion” until statistically proven to predict outcomes[1],[2].

• Checkland’s Interpretation: In Checkland’s model, standards are the criteria by which agents judge the “flux of events”[3]. By demanding statistical validation, Mainelli is attempting to rigidify these standards, moving them from tacit “readinesses to see”[3] into explicit, empirically tested regulators. Mainelli is essentially trying to “clean” the appreciative settings of the managers to ensure their “Reality Judgments” (what is happening) match the actual system behavior[4].

2. The IT Case Study: The Social Negotiation of Meaning

Mainelli recounts a case where an IT department, upset that “IT downtime” was flagged as a risk indicator, unilaterally changed the definition to “unplanned IT downtime” to improve their scores[5].

• Checkland’s Interpretation: This is a classic example of the social nature of the appreciative system. Checkland argues that standards and norms are not absolute but are socially negotiated and mutable[6]. The IT department engaged in a “value judgment” (we should not be penalized for planned work) and consequently altered the “reality judgment” (the definition of downtime) to maintain a favorable relationship with the organization[5].

• Systemic Insight: Checkland would view this not just as data manipulation (Goodhart’s Law), but as the system attempting to maintain stability by adjusting its internal standards to avoid negative judgments[3]. The IT department manipulated the “informational” component to ensure the “value” component remained positive.

3. The Trader Training Case: Correcting Reality Judgments

Mainelli describes a commodities firm where managers believed “more training leads to fewer errors,” but statistical analysis proved this false (training had no impact on error rates after the first six months)[7].

• Checkland’s Interpretation: This represents a breakdown in the managers’ Reality Judgment—their “hypothesis” about how the world works was incorrect[8]. The managers were operating with a “mental model” or Weltanschauung (worldview) that professional competence is a linear function of formal training.

• Learning Cycle: Mainelli’s statistical intervention acted as the feedback loop in Checkland’s learning cycle[3]. The “flux of events” (actual loss data) contradicted the “interpretation” (training helps). This forced a resetting of the Appreciative Setting: the managers had to abandon their old norm and accept a new reality—that competence is likely innate or established early, and further training is not an effective instrumental response to error[7].

4. “Scientific Management” vs. “Human Activity Systems”

Mainelli explicitly advocates for a “scientific approach” where the organization is viewed as a “black box” where inputs predict outputs[9],[10].

• Checkland’s Interpretation: Checkland might critique this as an attempt to treat a Human Activity System as a “Hard System” (like a machine)[11]. Mainelli acknowledges that “correlation doesn’t demonstrate causation” but argues it “should cause questions”[12].

• Synthesis: From a Checkland perspective, Mainelli’s statistical engine is not the decision-maker; it is a generator of mismatch signals[13]. These signals trigger the social process of appreciation. The statistics do not make the decision; they provide a more rigorous input for the managers’ dialogue, forcing them to confront whether their “internal standards” align with external reality.

5. The Limits of Prediction and “Black Swans”

Mainelli notes that his statistical method (Environmental Consistency Confidence) works well for normal distributions but fails with “Black Swans” (rare, extreme events) and complex human behaviors[14].

• Checkland’s Interpretation: This validates Checkland’s assertion that social reality is not governed by immutable laws like natural science[15]. Human systems are “autopoietic” (self-creating) and capable of changing their own structure and responses[16],[17].

• The Unknowable: Checkland would argue that Mainelli’s statistical approach can track past patterns of behavior (the “history of events”), but it cannot fully account for the “history of ideas”—the shifting values and intentions of the people within the system that can suddenly alter the system’s behavior (e.g., the IT department changing definitions or a market panic)[18],[5].

Summary

Interpreted through Checkland, Mainelli’s case study is an effort to introduce explicitness and empiricism into the Appreciative System. It attempts to replace tacit, unexamined assumptions (e.g., “training helps,” “downtime is bad”) with explicit, tested correlations. However, as the IT and Trader cases show, this process is still subject to the “flux of events and ideas”[3], where human actors will interpret, negotiate, and even subvert these metrics based on their own worldviews and interests.

Graph View