Structural Analysis of Social Issues — Using Systems Thinking to Visualize Why Problems Persist

Introduction

A nonprofit running a food bank continues to deliver food, yet the number of people needing food assistance keeps growing. Employment support programs help people find jobs, yet case after case, individuals return to unemployment shortly after. Interventions address individual problems, but the overall situation remains unchanged. Anyone who has worked on the front lines of social issues will recognize this feeling.

The cause, in most cases, lies in how we see the problem. As long as we respond only to the events in front of us, issues will keep resurfacing in different forms. The thinking framework that enables us to understand the structures behind events is systems thinking.

---

What Is Systems Thinking?

Systems thinking is a way of understanding complex problems by viewing them as collections of interconnected elements — that is, as systems.

Professor Jay Forrester at MIT laid the groundwork through system dynamics, which uses computer simulation. The field developed from the 1970s onward in policy analysis and corporate strategy. In 1990, Peter Senge's The Fifth Discipline positioned systems thinking as "the core discipline that supports an organization's learning capacity," bringing the concept to widespread attention in Japan and beyond.

The core idea is straightforward: instead of seeing problems as linear chains of cause and effect, we view them as feedback loops in which elements continually influence one another.

In the field of social transformation, Donella Meadows made an especially notable contribution. Her 1972 work The Limits to Growth demonstrated the importance of systems analysis, and her 1999 paper "Leverage Points: Places to Intervene in a System" identified twelve levels of intervention within a system. This framework continues to be widely used in the structural analysis of social issues.

---

The Iceberg Model — What Lies Above and Below the Surface

The first tool for viewing social issues as systems is the Iceberg Model.

Waterline

EventsEvents

Visible phenomena

PatternsPatterns

Recurring trends

StructuresStructures

Systems that produce patterns

Mental ModelsMental Models

Underlying values and beliefs

To deeper levelsDeeper intervention points yield greater impact

Just as an iceberg reveals only a fraction of itself above the waterline, a problem's visible "events" are only the tip; far larger structures lie beneath.

[The Four Layers of the Iceberg Model]

  ───────────────────────────
  Above the    ▲ Events
  waterline
  ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─
               ▼ Patterns
               ▼ Structures
               ▼ Mental Models
  ───────────────────────────

Events are the phenomena we observe day to day. "This month, the number of inquiries to the food bank increased" — concrete facts like this belong to this layer. What we typically call "problems" almost always reside here.

Patterns are the trends that emerge when events are viewed over time. A recurring pattern such as "every year, inquiries spike at the end of the fiscal year" becomes visible only when events are accumulated over a longer timeframe.

Structures are the mechanisms that generate patterns. They include institutional factors, inter-organizational relationships, and resource allocation methods. Behind the pattern of "support needs rising at fiscal year-end" lie structures such as the timing of fixed-term employment contract expirations and application deadlines within the social security system.

Mental Models are the values, assumptions, and beliefs that sustain structures. If a society holds the assumption that "poverty is a matter of personal responsibility," this belief shapes both the design of support systems and the behavior of those who need assistance.

The deeper one goes in the iceberg, the greater the impact of change — but the greater the difficulty as well.

---

Causal Loop Diagrams — Mapping Vicious Cycles

The tool for visualizing the structural layer is the Causal Loop Diagram (CLD).

A causal loop diagram connects elements within a system using arrows that represent causal relationships, thereby illustrating the system's feedback structures. There are two types of causal relationships:

• Positive relationship (+): When the cause increases, the effect also increases (and when the cause decreases, the effect also decreases).
• Negative relationship (-): When the cause increases, the effect decreases (and when the cause decreases, the effect increases).

Feedback loops come in two varieties.

Reinforcing Loops (R Loops)

In a reinforcing loop, change amplifies in the same direction. This can work in both beneficial and detrimental ways.

Consider the cycle of poverty as an example:

Low income -> Difficulty investing in education -> Lack of skills/qualifications -> Limited to low-wage jobs -> Low income (returns to start)

Each element increases the next, which in turn feeds back to the first. This is a reinforcing loop operating as a vicious cycle. Without external intervention to break the loop, it continues to perpetuate itself.

Balancing Loops (B Loops)

A balancing loop works to close the gap between a goal and the current state. It functions as a stabilizing mechanism.

When requests for assistance to a food bank increase, the organization responds by expanding food distribution. As distribution rises, the number of inquiries temporarily stabilizes; when demand grows again, requests increase once more. This is a balancing loop. A balancing loop alone rarely addresses root causes; it needs to be considered in conjunction with interventions targeting the reinforcing loops behind it.

---

Leverage Points — Where to Press for Change

Once a causal loop diagram reveals the structure of a system, the next question is: "Where should we intervene?"

In her 1999 paper, Donella Meadows identified twelve levels of intervention points within a system. Arranged from the least to the most effective, they are as follows:

Most policies and support programs concentrate on the lower intervention points, levels 12 through 9: increasing subsidies, building facilities, extending service periods. These measures may appear to produce immediate results, but they do not alter the fundamental structure of the system.

Higher intervention points, by contrast, affect a much broader scope. For instance, changing information asymmetry — who knows what (level 6) — can reach far more people than individual employment support programs. If the structure that prevents disadvantaged populations from accessing the information they need for job-seeking is reformed, the impact extends broadly.

The highest intervention point is the paradigm — a shift in the very lens through which we view the problem. When society's collective understanding shifts from "poverty is an individual failing" to "poverty is a structural problem," everything downstream — from rule-making to institutional design — moves in a cascading fashion.

---

Application to Nonprofits and Social Enterprises

Question the Structure Before Designing Programs

Before designing a program, ask: "Why does this problem keep recurring?" That is all it takes.

Use the Iceberg Model to determine whether your organization is addressing the "events" layer or the "structures" layer. Responding to events is undeniably necessary. However, events-level responses alone will not prevent recurrence. Simply being aware of which layer you are working on can change how you design your programs.

SIIF's Systems Change Investment

A pioneering organization that has integrated the concept of systems change into social investment in Japan is the Social Innovation and Investment Foundation (SIIF) (一般財団法人社会変革推進財団). SIIF views society as "an organic totality of interactions in which elements influence one another" and promotes "systems change investment" — an approach that aims not at solving individual problems but at transforming the structures that produce them. Their work serves as a valuable reference for practice that targets the upper levels of leverage points.

Dialogue Through Causal Loop Diagrams

A causal loop diagram functions not only as an analytical tool but also as a tool for dialogue. The process is simple: write elements on a sheet of paper and connect them with arrows. When multiple stakeholders participate in drawing the diagram together, the process itself builds shared understanding of the problem.

The diagram does not need to be perfect. The experience of "seeing feedback loops that were previously invisible" is what shifts thinking.

---

Related Guides

• Theory of Change Workshop Guide — How to move from structural analysis to a "theory of change" and share it with your team
• Designing Collective Impact — An approach to addressing structural problems that no single organization can solve, through cross-sector collaboration

---

ISVD's Perspective

Social issues do not fit neatly into simple cause-and-effect relationships. Feedback accumulates over time and problems recur within invisible structures.

To keep asking "why does this remain unsolved"; to deliberately combine immediate responses to events with interventions in the underlying structures — systems thinking is the instrument for doing so.

ISVD's SDI Assessment incorporates a process of reading the structure of social issues while simultaneously reflecting on one's own engagement with them. When the structure of a problem becomes visible, so does where you stand within it.