The Search for Leverage Points: Why Complexity Demands Smarter Effort in the Social Sector

by

Edward Addai

edwardaddai64@gmail.com

 

Something May Be Wrong With How We Work

Imagine a district health officer in northern Ghana. Each year, she receives a budget, trains community health workers, runs immunisation campaigns, distributes bed nets and reports results to her superiors. Vaccination coverage climbs modestly. Child mortality falls a little. But stunting rates remain stubbornly high. Girls still drop out of school to care for sick siblings. Teenage pregnancies continue to undermine decades of investment. Something may be wrong, and it is not her effort; it is where the effort is being directed.

Now consider a child protection coordinator in rural Bangladesh. Every six months, she reviews her case files. The numbers of children referred to services have improved. Her case workers are certified and diligent. Yet the same families cycle through the system repeatedly, and the underlying conditions of poverty, patriarchal norms, seasonal hunger remain untouched. She suspects that she and her colleagues are doing the right things but in the wrong places.

 

These are not isolated failures of skill or commitment. They are symptoms of a deeper problem: the social sector's persistent difficulty in finding and acting on leverage points - those places in a system where a small shift in one thing produces large changes in everything else (Meadows, 2008). In a sector overflowing with good intentions, rigorous evaluations, and hard-working professionals, the inability to locate and act on leverage points may be the single most consequential gap between effort and impact.

 

Understanding Leverage in Complex Systems

 

What Is a Leverage Point?

The concept of leverage points was most powerfully articulated by systems thinker Donella Meadows in her landmark essay, later expanded into a book (Meadows, 2008). A leverage point is a place within a complex system such as a corporation, an economy, a living body, a city, an ecosystem, where a small shift in one thing can produce big changes in everything else. Crucially, Meadows identified a hierarchy of leverage points, from the least powerful (changing numbers, such as budgets and subsidies) to the most powerful (changing the goals of the system, the mindset or paradigm out of which the system arises, and the power to transcend paradigms altogether).

For social sector leaders, this framework is both liberating and unsettling. It is liberating because it suggests that not all problems require equal resources - some small, precisely targeted actions can unlock disproportionate change. It is unsettling because it implies that much of what the social sector currently does may be aimed at low-leverage interventions, however well-executed they are.

 

Complexity Science and Why Social Problems Resist Simple Solutions

The social problems that the sector addresses, including poverty, disease, abuse, illiteracy, are not complicated problems. They are complex ones. This distinction, drawn from complexity science and popularised in management and policy through the Cynefin framework (Snowden and Boone, 2007), is foundational. Complicated problems, like engineering a bridge or manufacturing a vaccine, have knowable solutions that can be replicated. Complex problems, like reducing child marriage in rural Ethiopia or improving learning outcomes in urban slums in Karachi, involve non-linear dynamics, feedback loops, emergent behaviour and adaptive agents who change their behaviour in response to interventions (Glouberman and Zimmerman, 2002).

 

Complex Adaptive Systems (CAS) theory, developed through the Santa Fe Institute and scholars such as Holland (1992) and Axelrod and Cohen (1999), holds that in these systems:

 

•         Cause and effect are separated in time and space, making attribution extraordinarily difficult.

•         Systems exhibit non-linearity: small inputs can produce large outputs and vice versa.

•         Feedback loops, both reinforcing and balancing, shape system behaviour.

•         Agents (individuals, organisations, communities) adapt their behaviour in ways that can reinforce or undermine interventions.

•         Emergent outcomes arise that could not have been predicted from the actions of individual components.

 

It is precisely because of these properties that finding leverage points in the social sector is so difficult, and so necessary.

 

Institutional Economics: The Invisible Architecture of the System

Complexity science explains the dynamics of social systems. Institutional economics, drawing on North (1990), Ostrom (1990), and more recently Acemoglu and Robinson (2012), explains why they are structured the way they are. Institutions, understood as the formal and informal rules of the game, shape who has power, what is incentivised, and who bears the costs of failure. In many developing country contexts, the formal institutions such as ministries, courts, schools, hospitals are layered over deep informal institutions: gender norms, ethnic hierarchies, patron-client networks, and religious authorities.

 

For the social sector manager seeking leverage, this is a critical insight. A purely technical intervention such as a new teaching methodology, a better referral pathway for abuse cases, or a community health insurance scheme, may deliver short-term results but fail to stick because it is working against, rather than with or through, the institutional grain. North (1990, p. 3) put it with characteristic precision: 'Institutions are the rules of the game in a society... they structure incentives in human exchange, whether political, social, or economic.'

 

The implication is profound: the highest leverage points in social systems are often institutional i.e. norms, incentive structures, rules, and power arrangements, not the technical parameters of programmes. Yet these are precisely the places the sector finds hardest to act on.

How to Search for Leverage Points in the Social Sector

 

Step 1: Develop a System Map Before a Solution

The search for leverage points begins with disciplined inquiry, not into what to do, but into how the system works. System mapping, whether through causal loop diagrams, outcome mapping, contribution analysis or problem-driven iterative adaptation (Andrews, Pritchett and Woolcock, 2017), allows leaders to move from linear, programme-logic thinking to a richer understanding of the reinforcing and balancing loops that maintain the problem.

 

Consider the challenge of low primary school enrolment among girls in northern Nigeria. A simple linear logic model might read: build more schools → increase access → raise enrolment. Yet a system map might reveal that the binding constraint is not the number of schools but the perception, held by male household decision-makers, that educating girls reduces their family's social standing and marriageability. Burdened by this insight, a leverage point shifts from construction budgets to the norms and incentives that shape household decision-making and the religious authorities and male peer networks that reinforce them.

 

Step 2: Distinguish Symptoms from Structural Causes

Meadows (2008, p. 145) observed that 'before you disturb the system in any way, watch how it behaves.' This counsel is frequently violated in the social sector, where funding cycles, log frames, and donor demands push practitioners toward rapid action before adequate diagnosis. The PDIA approach (Andrews, Pritchett and Woolcock, 2017) provides a rigorous alternative: identify the problem, understand its causes iteratively, and explore solutions through small authorised experiments, feeding what is learned back into the analysis.

 

In child protection, for instance, rates of child sexual abuse in South African townships persistently remain high despite the proliferation of awareness campaigns and hotlines. System analysis repeatedly points to structural causes: extreme poverty forcing both parents to work long and irregular hours; alcohol abuse linked to a mining and migrant labour history; dense and poorly lit informal settlements; and a deep culture of silence around male violence rooted in hegemonic masculinity (Jewkes et al., 2011). The leverage point is not more hotlines; it is the norms, gender power arrangements, and economic conditions that create vulnerability in the first place.

 

Step 3: Look for Feedback Loops and Delay

One of the most powerful leverage points in Meadows' hierarchy is changing the gain around a feedback loop. In development contexts, feedback loops are often broken or delayed, meaning that the system cannot self-correct. Health systems in many sub-Saharan African countries receive data from facilities months after the fact by which time the outbreak has spread, the stock-out has already caused deaths, and the absentee teacher has taught no lessons for a term. When UNICEF's supply chain initiatives in Sierra Leone invested in real-time data systems for essential medicines, stock-out rates dropped significantly not because more medicines were procured, but because feedback loops were repaired and shortened (UNICEF, 2019).

 

The lesson is generalisable: social sector leaders should systematically ask, in any system they are working in, where feedback is absent, slow, or distorted, and whether restoring or accelerating feedback represents a more powerful lever than adding resources or implementing programmes.

 

Step 4: Attend to the Rules, Goals and Paradigms of the System

The top of Meadows' leverage hierarchy - changing rules, goals, and paradigms - is also where institutional economics most powerfully converges with systems thinking. Rules (North's formal and informal institutions), goals (what the system is optimised for), and paradigms (the shared beliefs from which the system arises) are extraordinarily hard to change. But when they do change, the effects are vast and durable.

 

Consider the transformation of primary education in Kerala, India. For decades, scholars have puzzled over why a relatively poor Indian state achieved near-universal literacy while wealthier states lagged. The explanation does not lie in resources or even in the technical quality of teaching; it lies in a paradigm shift: a broad social and political consensus, shaped by left political movements, women's organisations, and local self-government, that universal education was a non-negotiable social good, not a service for elites (Heller, 1999). This shift in the goal and paradigm of the educational system unlocked everything else.

 

For managers and leaders who feel constrained to work within existing paradigms, the practical implication is to ask: what would need to be true about the goals and beliefs of this system for the outcomes we want to be natural, rather than forced? And then to work, often slowly, through coalition, evidence, and narrative, to shift those beliefs.

 

Step 5: Use Complexity-Aware Methods

The social sector has, in recent years, developed a growing toolkit of complexity-aware methods that support the search for leverage. Developmental evaluation (Patton, 2011) applies evaluative thinking in real time to guide programme adaptation in complex environments, rather than delivering post-hoc verdicts on fixed designs. Outcome Harvesting (Wilson-Grau, 2015) works backwards from observable changes to describe what contributions were made and how, without requiring a pre-specified theory. Problem-driven iterative adaptation (Andrews, Pritchett and Woolcock, 2017) explicitly surfaces the political and institutional binding constraints on reform, rather than assuming that technical solutions will be adopted in a neutral environment.

 

Leaders who invest in these methods, even partially, even imperfectly, are systematically more likely to find where their effort will produce the most change.

 

What Happens When the Search Is Not Done or Not Done Well

The Trap of Low-Leverage Busyness

The most common and costly failure in the social sector is not incompetence. It is the systematic application of competence to low-leverage activities. Meadows (2008) herself noted that complex systems often present obvious points of intervention that are not actually high-leverage: they are simply the places where the system makes itself most visible. Organisations with strong monitoring and evaluation cultures know how many people they have reached; fewer know whether the conditions that generate the problem have changed at all.

 

This 'busyness trap' is compounded by the incentive structures of the sector. Donors and governments frequently reward scale, reach, and activity - the number of children vaccinated, the number of teachers trained, the number of cases referred. These are not unimportant metrics, but they measure outputs rather than system change. When institutions are rewarded for outputs, they rationally invest in output-generating activities, even when those activities leave the system's deeper structure untouched (Andrews, 2013).

 

Policy Reform Without Institutional Understanding

The history of development cooperation is littered with technically elegant reforms that failed because they did not account for the institutional and political economy in which they were introduced. North (1990) warned that simply transplanting formal institutions from high-performing to low-performing contexts without attending to informal norms and enforcement mechanisms is likely to fail. The 'best practice' transfer model - replicating what worked in one context without examining why - exemplifies this error.

 

The school voucher experiments in several Latin American countries in the 1990s and 2000s illustrate the point. In theory, vouchers increase competition, improve quality, and empower parents. In practice, where informal institutions such as ethnic segregation, urban-rural gaps in school quality, low parental literacy, and the social prestige of private schooling, structured the educational marketplace, vouchers frequently intensified inequality without improving average learning outcomes (Patrinos, 2012). The leverage point was not in the financing mechanism but in the institutional conditions under which choice could meaningfully operate.

 

The Cost of Complexity Blindness

Perhaps most dangerously, when leaders and managers operate as though social problems are merely complicated rather than complex, as though more of the right inputs will reliably produce the right outputs, they generate interventions that are actively counterproductive. Snowden and Boone (2007, p. 73) warn that 'the ordered world, whether simple or complicated, is the world of experts, and that applying expert-driven, plan-and-implement approaches to genuinely complex problems consistently fails to produce intended outcomes and often produces unintended negative ones.

 

In health, the experience of vertical disease-control programmes in sub-Saharan Africa provides a cautionary illustration. HIV/AIDS programmes, funded generously in the 2000s, produced remarkable treatment results and antiretroviral therapy at scale genuinely saved millions of lives. But the parallel effects on health system capacity, staff motivation, and the institutional logic of primary health care were devastating in many contexts: specialist HIV cadres were created outside civil service structures, facility budgets were distorted, and community trust in general health services eroded (Pfeiffer and Chapman, 2010). The leverage point had been found for HIV, but in acting on it without a system perspective, other parts of the system were destabilised.

 

A Call to Disciplined Imagination

The search for leverage points is not an invitation to abandon what works, or to retreat into analysis paralysis. It is an invitation to a more disciplined form of imagination, one that holds the question 'where in this system can my organisation's effort produce the most durable and disproportionate change?' with the same rigour that is currently applied to project design and financial compliance.

 

For the district health officer in Ghana, that question might lead her away from another immunisation campaign toward an alliance with women's economic empowerment groups who can shift household decision-making about food and health-seeking behaviour. For the child protection coordinator in Bangladesh, it might mean investing in community-level norm change through men's groups and community theatre, rather than another cycle of caseworker certification.

 

The path is neither obvious nor guaranteed. Complex systems are, by definition, not fully knowable in advance. But leaders who engage in the disciplined search for leverage, who map systems before designing solutions, who distinguish symptoms from causes, who attend to the institutional rules and norms that reproduce problems, and who use complexity-aware methods to learn as they act, are systematically more likely to find those rare and powerful places where small shifts produce large, lasting change.

 

The alternative, the endless, earnest, well-funded reproduction of effort without leverage, is a tragedy the sector can no longer afford.

 

Conclusion

The search for leverage points is, at its core, a search for wisdom, for the kind of systemic understanding that allows finite resources, finite attention, and finite human energy to be directed toward the places in complex social systems where they will produce the most good. It demands that social sector leaders resist the natural pull toward visible, measurable, activity-based accountability and instead cultivate a deeper understanding of the systems they are trying to change.

 

Complexity science, through the work of Meadows, Snowden, Holland, and others, gives us the conceptual architecture for that understanding. Institutional economics, through North, Ostrom, and Acemoglu, reminds us that the most powerful rules of the game are often informal, invisible, and deeply embedded. Practical methodologies such as PDIA, developmental evaluation, system mapping, give us tools for navigating these landscapes in real time.

 

None of this is easy. Institutional change is slow. Paradigms resist challenge. Feedback loops are long and the signals are noisy. But the children in the Bangladeshi villages, the girls in northern Nigerian schools, the communities in South African townships, they are not waiting for the sector to find its footing in complexity theory. They are living, right now, in the consequences of interventions aimed at the wrong parts of the system.

 

Finding leverage is not optional. It is the most important professional obligation that social sector leaders carry.

 

References

Acemoglu, D. and Robinson, J.A. (2012) Why Nations Fail: The Origins of Power, Prosperity, and Poverty. New York: Crown Publishers.

 

Andrews, M. (2013) The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions. Cambridge: Cambridge University Press.

 

Andrews, M., Pritchett, L. and Woolcock, M. (2017) Building State Capability: Evidence, Analysis, Action. Oxford: Oxford University Press.

 

Axelrod, R. and Cohen, M.D. (1999) Harnessing Complexity: Organizational Implications of a Scientific Frontier. New York: Free Press.

 

Glouberman, S. and Zimmerman, B. (2002) Complicated and Complex Systems: What Would Successful Reform of Medicare Look Like? Discussion Paper No. 8. Ottawa: Commission on the Future of Health Care in Canada.

 

Heller, P. (1999) The Labor of Development: Workers and the Transformation of Capitalism in Kerala, India. Ithaca: Cornell University Press.

 

Holland, J.H. (1992) Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. Cambridge, MA: MIT Press.

 

Jewkes, R., Sikweyiya, Y., Morrell, R. and Dunkle, K. (2011) 'Gender Inequitable Masculinity and Sexual Entitlement in Rape Perpetration South Africa: Findings of a Cross-Sectional Study', PLoS ONE, 6(12), e29590.

 

Meadows, D.H. (2008) Thinking in Systems: A Primer. Edited by D. Wright. White River Junction, VT: Chelsea Green Publishing.

 

North, D.C. (1990) Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

 

Ostrom, E. (1990) Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.

 

Patton, M.Q. (2011) Developmental Evaluation: Applying Complexity Concepts to Enhance Innovation and Use. New York: Guilford Press.

 

Patrinos, H.A. (2012) 'School Choice in Developing Countries', in Lleras-Muney, A. and Loeb, S. (eds) The Economics of Education. London: Elsevier, pp. 42–53.

 

Pfeiffer, J. and Chapman, R. (2010) 'Anthropological Perspectives on Structural Adjustment and Public Health', Annual Review of Anthropology, 39, pp. 149–165.

 

Snowden, D.J. and Boone, M.E. (2007) 'A Leader's Framework for Decision Making', Harvard Business Review, 85(11), pp. 68–76.

 

UNICEF (2019) Supply Chain Results: Improving Supply Chains for Children. New York: UNICEF Supply Division.

 

Wilson-Grau, R. (2015) 'Outcome Harvesting', in Bamberger, M., Rugh, J. and Mabry, L. (eds) RealWorld Evaluation: Working Under Budget, Time, Data, and Political Constraints. 2nd edn. Thousand Oaks, CA: SAGE, pp. 287–298.