Herbert Simon: Decision-Making Theory
Herbert A. Simon (1916-2001)
An American political scientist, economist, sociologist, psychologist, and computer scientist who pioneered theories of decision-making, organization theory, and artificial intelligence. Winner of the Nobel Prize in Economics (1978) and the Turing Award (1975), Simon challenged classical economic rationality with his concept of “bounded rationality.”
Key Contribution: Developed the theory of bounded rationality and administrative behavior, revolutionizing our understanding of how decisions are made in organizations and challenging the perfect rationality assumptions of classical economics.
1916-2001
Module Introduction: The Revolution in Decision Theory
Herbert Simon’s work fundamentally transformed our understanding of decision-making in organizations. Moving beyond the unrealistic assumptions of perfect rationality, Simon introduced the concept of bounded rationality—the idea that decision-makers operate with limited cognitive resources, incomplete information, and time constraints.
This module explores Simon’s groundbreaking theories of administrative behavior, satisficing, programmed vs. non-programmed decisions, and the intelligence-design-choice cycle through interactive visualizations, detailed explanations, and contemporary applications in public administration.
Part 1: The Concept of Bounded Rationality
Understanding Bounded Rationality
Simon’s concept of bounded rationality represents a fundamental departure from classical economic theory. While traditional models assumed decision-makers had perfect information and unlimited cognitive capacity, Simon observed that in reality, organizational decision-makers face significant constraints. These include limited information, time pressures, cognitive processing limitations, and the complexity of real-world problems.
Bounded rationality acknowledges that decision-makers are “intendedly rational, but only limitedly so”—they aim to make rational decisions but are constrained by practical limitations. This framework explains why organizations develop routines, procedures, and structures: to help individuals make decisions within these cognitive bounds.
Breaking with Classical Rationality
Simon challenged the classical economic assumption of “economic man” (homo economicus)—a perfectly rational actor who maximizes utility with complete information. Instead, he proposed “administrative man”—a decision-maker with bounded rationality who seeks satisfactory rather than optimal solutions.
Classical Rationality
The “Economic Man” Assumptions:
- Complete information about all alternatives
- Unlimited cognitive processing capacity
- Clear, consistent preferences
- Ability to calculate optimal solutions
- Maximization of utility/benefits
Bounded Rationality
The “Administrative Man” Reality:
- Limited information about alternatives
- Restricted cognitive processing capacity
- Satisficing rather than optimizing
- Use of heuristics and rules of thumb
- Acceptance of “good enough” solutions
Limits to Rational Decision-Making
Practical Constraints in Organizations
In organizational settings, decision-makers face multiple constraints that prevent perfectly rational decision-making. These limitations aren’t just theoretical—they’re observable in everyday administrative behavior. Organizations respond to these constraints by developing specialized roles, standard operating procedures, and hierarchical structures that help distribute cognitive load and manage complexity.
Information Limitations
Decision-makers cannot access all relevant information due to search costs, time constraints, and organizational barriers.
Cognitive Constraints
Human brains have limited processing capacity—we can only consider a small number of alternatives simultaneously.
Time Constraints
Decisions must often be made under time pressure, preventing exhaustive analysis of all options.
Computational Limits
Even with complete information, many decision problems are computationally complex beyond human capacity.
Part 2: Satisficing vs. Optimizing
The Satisficing Principle
Perhaps Simon’s most famous contribution is the concept of satisficing—a portmanteau of “satisfy” and “suffice.” Rather than searching for the optimal solution (which requires exhaustive evaluation of all alternatives), decision-makers using satisficing establish an aspiration level—a set of criteria that represents a “good enough” solution. They then search sequentially through alternatives until they find one that meets this threshold.
This approach is not irrational; rather, it’s a rational response to bounded rationality. When search costs (time, effort, resources) are considered, stopping at a satisfactory solution often yields better overall outcomes than continuing to search for an elusive optimum.
Optimizing Model
Find the highest peak
Exhaustive search of entire landscape
Process: Search all alternatives, evaluate completely, select absolute best
Requires: Complete information, unlimited time, perfect rationality
Satisficing Model
Process: Search until satisfactory option found, accept “good enough”
Reflects: Real-world constraints, bounded rationality, practical decision-making
The Satisficing Algorithm in Practice
Simon’s satisficing approach can be understood as a step-by-step process that mirrors how actual decision-makers operate in complex environments:
1. Set Aspiration Level
Establish minimum acceptable criteria for a satisfactory solution based on experience, standards, or organizational requirements.
Example: “We need a solution that reduces costs by at least 15%”
2. Sequential Search
Examine alternatives one by one (or in small batches) rather than evaluating all options simultaneously.
Example: “Let’s evaluate these three vendor proposals first”
3. Evaluate Against Criteria
Test each alternative against the established aspiration level to determine if it meets minimum requirements.
Example: “Does this proposal meet our 15% cost reduction target?”
4. Accept or Continue
If current alternative meets criteria: ACCEPT. If not: continue searching until satisfactory option found or search costs exceed benefits.
Example: “This meets our criteria—we’ll select this vendor”
Part 3: Types of Organizational Decisions
Categorizing Decision Complexity
Simon distinguished between two fundamental types of decisions in organizations, recognizing that different kinds of problems require different decision-making approaches. This categorization helps organizations allocate decision-making authority appropriately and develop suitable procedures for each type.
Programmed Decisions
Characteristics:
- Repetitive and routine
- Clear decision rules and procedures
- Lower-level management responsibility
- Can be automated or delegated
- Based on established organizational routines
Examples: Inventory reordering, routine hiring, standard customer service responses
Non-programmed Decisions
Characteristics:
- Novel and unstructured
- No established decision rules
- Upper-level management responsibility
- Requires judgment and creativity
- High uncertainty and complexity
Examples: Mergers and acquisitions, new product development, crisis management
The Intelligence-Design-Choice (IDC) Cycle
Simon’s Decision-Making Process Model
The Intelligence-Design-Choice (IDC) cycle is Simon’s most influential contribution to understanding how decisions are actually made in organizations. Rather than viewing decision-making as a single event, Simon conceptualized it as a process with distinct phases that often loop back on themselves as new information emerges or conditions change.
This model recognizes that decision-making is iterative, not linear. Decision-makers may move back and forth between phases, and the process often includes feedback loops where outcomes from one decision inform future decisions. The IDC cycle provides a framework for structuring complex decision processes in organizations.
The Intelligence-Design-Choice Cycle: A Sequential Process
Simon’s three-phase model (with a fourth review phase added by later theorists) represents the core of organizational decision processes:
Intelligence Phase
Problem Identification & Information Gathering
- Scanning the environment
- Identifying problems/opportunities
- Gathering relevant information
- Defining decision requirements
Design Phase
Solution Development & Analysis
- Generating possible alternatives
- Developing models of the situation
- Predicting outcomes of alternatives
- Establishing evaluation criteria
Choice Phase
Selection & Implementation
- Evaluating alternatives against criteria
- Making the final selection
- Planning implementation
- Allocating resources
Review Phase
Evaluation & Learning
- Monitoring outcomes and impacts
- Comparing results with expectations
- Identifying lessons learned
- Adjusting future decisions
Feedback Loop: Learning for Future Decisions
The review phase creates a feedback loop that connects back to the intelligence phase. Lessons learned from one decision cycle inform problem identification and analysis in future cycles, creating organizational learning and adaptation.
Part 4: The Rationality Spectrum
Understanding Different Approaches to Rationality
Simon’s bounded rationality exists on a spectrum between two extremes: comprehensive rationality (the classical economic ideal) and incrementalism (a pragmatic political approach). Understanding this spectrum helps contextualize Simon’s contribution and shows how different decision-making approaches are appropriate for different contexts.
Each point on the spectrum represents a different balance between the ideal of perfect rationality and the practical constraints of real-world decision-making. Organizations and decision-makers may shift along this spectrum depending on the nature of the problem, available resources, and time constraints.
Comprehensive Rationality
Classical Economics
Perfect information, optimal solutions, unlimited cognitive capacity
Bounded Rationality
Simon’s Theory
Limited information, satisficing, realistic cognitive constraints
Incrementalism
Lindblom’s Theory
Successive limited comparisons, marginal changes, political bargaining
Information Processing in Organizations
How Organizations Manage Information
Simon recognized that organizations develop structures and processes specifically to manage the information processing demands that exceed individual cognitive capacities. These organizational mechanisms help overcome bounded rationality by distributing information processing across multiple individuals and over time.
Scanning
Information Gathering
Collecting data from internal and external sources
Filtering
Selective Attention
Focusing on relevant information, ignoring noise
Structuring
Problem Framing
Organizing information into decision frameworks
Implementation
Action Taking
Executing decisions and monitoring outcomes
Part 5: Applications in Public Administration
Practical Implications for Governance
Simon’s theories have profound implications for public administration. Recognizing bounded rationality helps explain why government agencies develop bureaucratic structures, standard operating procedures, and specialized roles. It also explains why perfect rationality is unattainable in policy-making and why incremental approaches often prevail.
The table below illustrates how bounded rationality manifests in different areas of public administration and what practical strategies can help manage these constraints.
| Public Administration Context | Bounded Rationality Implications | Practical Applications | Policy Examples |
|---|---|---|---|
| Policy Formulation | Policymakers cannot consider all possible policy alternatives or predict all consequences | Use of policy frameworks, incremental changes, pilot programs | Social welfare reform, environmental regulations |
| Budgetary Decisions | Limited information about program effectiveness, political constraints on rationality | Incremental budgeting, performance-based budgeting, zero-based reviews | Federal budget allocation, state spending priorities |
| Organizational Design | Organizational structures reflect bounded rationality through specialization and hierarchy | Departmental specialization, standard operating procedures, decision support systems | Government agency structures, interdepartmental coordination |
| Implementation Challenges | Street-level bureaucrats exercise discretion within bounded rationality constraints | Clear guidelines, training programs, monitoring systems | Social service delivery, law enforcement decisions |
| Crisis Management | High uncertainty, time pressure, and cognitive overload constrain decision-making | Crisis protocols, simulation training, rapid response teams | Natural disaster response, public health emergencies |
| Regulatory Decisions | Limited information about industry practices, compliance costs, and regulatory impacts | Cost-benefit analysis, stakeholder consultation, regulatory impact assessments | Environmental regulations, financial industry oversight |
Case Study: The Cuban Missile Crisis (1962)
President Kennedy’s decision-making during the Cuban Missile Crisis illustrates bounded rationality in high-stakes policy decisions:
Intelligence Phase
Problem: Discovery of Soviet missiles in Cuba through aerial reconnaissance
Limitations: Incomplete information about Soviet intentions and missile capabilities
Design Phase
Options Considered: Diplomatic pressure, blockade (quarantine), air strike, invasion
Satisficing: Chose blockade as option meeting multiple criteria (firm but not immediately escalatory)
Choice Phase
Implementation: Naval blockade combined with secret diplomatic negotiations
Bounded Rationality: Couldn’t predict all Soviet responses, relied on assumptions and limited information
Simon’s Analysis: Kennedy’s ExComm (Executive Committee) exemplified bounded rationality—working with incomplete information, under extreme time pressure, using satisficing to find an acceptable solution rather than an optimal one.
Part 6: Criticisms and Extensions
Evaluating Simon’s Legacy
While Simon’s theories revolutionized decision theory, they have not been without criticism. Some scholars argue that bounded rationality is too pessimistic, while others suggest it doesn’t go far enough in capturing the social and political dimensions of decision-making. Understanding these criticisms helps appreciate the ongoing evolution of decision theory.
Criticisms of Simon’s Theory
Theoretical Criticisms
- Too Pessimistic: Underestimates human capacity for rational analysis in some contexts
- Vague Boundaries: “Bounded” is conceptually fuzzy—how bounded is bounded?
- Overemphasis on Cognitive Limits: Neglects emotional, social, and political factors in decisions
- Circular Reasoning: Any decision can be explained as satisficing after the fact
Practical Limitations
- Decision Support Systems: Technology may expand boundaries of rationality
- Expert Decision-Making: Experts in domain may approach comprehensive rationality
- Cultural Variations: Different cultures may have different rationality boundaries
- Evolution of Cognition: Human cognitive capacities may be evolving with technology
Theoretical Extensions and Related Models
Garbage Can Model
Cohen, March, and Olsen (1972) extended Simon’s work with a model of organizational decision-making as streams of problems, solutions, participants, and choice opportunities mixing randomly.
Incrementalism
Charles Lindblom’s “science of muddling through” (1959) emphasized small, incremental changes rather than comprehensive rational analysis.
Prospect Theory
Kahneman and Tversky (1979) built on bounded rationality with behavioral economics, showing systematic cognitive biases in decision-making.
Naturalistic Decision Making
Klein’s recognition-primed decision model (1998) examines how experts make rapid decisions in complex, time-pressured situations.
Part 7: Contemporary Relevance and Applications
Simon’s Enduring Influence
Herbert Simon’s ideas have gained renewed relevance in the digital age. As organizations grapple with big data, artificial intelligence, and increasing complexity, Simon’s insights about bounded rationality provide crucial guidance for designing decision systems that work within human cognitive limits.
Simon’s Legacy in the Digital Age
Herbert Simon’s theories have gained renewed relevance in the era of artificial intelligence and big data:
Artificial Intelligence
Simon’s work on human problem-solving directly influenced early AI research. Modern machine learning systems often use satisficing algorithms rather than optimization.
Behavioral Economics
Nobel laureates like Daniel Kahneman and Richard Thaler have extended Simon’s bounded rationality into systematic study of cognitive biases and decision heuristics.
Decision Support Systems
Modern DSS are designed to compensate for bounded rationality by providing better information, analysis tools, and decision frameworks.
Modern Public Administration Applications
Evidence-Based Policy
Using research and data to inform policy decisions while acknowledging bounded rationality constraints:
- Systematic reviews of policy effectiveness
- Randomized controlled trials in social policy
- Cost-benefit analysis with sensitivity testing
- Performance measurement and outcome evaluation
Simon Connection: Attempts to expand rationality boundaries through better information
Nudge Theory & Choice Architecture
Designing decision environments to guide choices while preserving freedom:
- Default options for retirement savings
- Simplified forms and processes
- Strategic information presentation
- Timely reminders and prompts
Simon Connection: Works within bounded rationality by structuring choices effectively
Conclusion: Simon’s Enduring Contribution
Herbert Simon’s revolution in decision theory fundamentally changed how we understand organizational behavior and public administration:
Key Enduring Insights
Realistic Human Decision-Making
People are “intendedly rational” but limited by cognitive constraints, information gaps, and time pressures
The Satisficing Principle
In complex environments, seeking “good enough” solutions is often more rational than seeking optimal ones
Organizations as Decision Systems
Organizational structures and procedures exist to compensate for individual bounded rationality
In an era of increasing complexity, information overload, and rapid change, Simon’s insights about bounded rationality are more relevant than ever. The challenge for contemporary public administration is not to achieve perfect rationality but to design decision processes and organizational structures that work effectively within human cognitive limitations.
Final Reflection: The most effective organizations and governments may not be those that make perfectly rational decisions, but those that design systems and processes that help decision-makers navigate complexity within the bounds of human rationality.
