Second Order Thinking
Second Order Thinking: Ask "and then what?" after every predicted outcome. First-order consequences are what everyone sees; second-order consequences are where most mistakes and most opportunities hide.
What Is Second Order Thinking?
Every action produces a chain of consequences. The first consequence is immediate and visible: you raise prices, revenue per unit goes up. The second consequence follows from the first: some customers, now paying more, evaluate alternatives. The third follows from the second: a competitor with lower prices gains traction with your price-sensitive segment. And so on.
Most decision-making stops at the first order. This is not stupidity — it is cognitive economy. The human brain evolved to solve immediate problems, not to trace consequence chains across time. First-order thinking is fast, requires minimal effort, and is adequate for most daily decisions.
But for consequential decisions — those that affect systems, set precedents, involve strategy, or have long time horizons — first-order thinking is systematically dangerous. The effects that cause the most damage are almost always second or third order. This is why well-intentioned policies produce unintended consequences, why smart investors lose money on obvious trades, and why strategies that seem brilliant in year one look catastrophic in year three.
Howard Marks, co-founder of Oaktree Capital and one of the most respected investors of his generation, articulated this framework most clearly in his 1990 memo "First and Second Level Thinking." His formulation: "First-level thinking says, 'It's a good company; let's buy the stock.' Second-level thinking says, 'It's a good company, but everyone thinks it's a great company, and it's not. So the stock's overrated and overpriced; let's sell.'" The first-level thinker sees a good company. The second-level thinker sees a good company that is already fully priced — and therefore not a good investment.
This insight — that consensus first-order thinking is already priced into markets — is the foundation of contrarian investing. But it applies equally to product strategy, policy design, organizational change, and personal decisions.
How It Works
For any proposed action:
First Order: What is the immediate, direct consequence?
(This is what everyone sees.)
Second Order: What happens as a result of the first-order
consequence? Who responds, and how?
(This is where most decisions break down.)
Third Order: What happens as a result of the second-order
consequence? What adaptation occurs?
(This is where strategies play out over time.)
Key question at each level:
Who is affected by this outcome?
How will they respond?
What does their response change?
What does that change cause next?
Practical rule: If the first-order consequence is obviously good, think harder about the second order. If it's obviously bad, think harder about whether the second order might be net positive (short-term pain, long-term gain).
Real-World Examples
Example 1: Cobra Effect — The British Cobra Bounty in India
One of the most cited examples of failed first-order thinking is the British colonial administration's attempt to reduce the cobra population in Delhi. The first-order analysis was impeccable: cobras are dangerous; if you pay people for dead cobras, the number of cobras will decrease.
Second-order consequence: entrepreneurial Indians began breeding cobras to claim the bounty. Third-order consequence: when the British authorities discovered this and abolished the bounty, the breeders — with now-worthless cobras — released them into the wild. The cobra population increased significantly compared to the baseline.
The policy achieved the exact opposite of its goal by ignoring the second-order response of rational economic actors. This failure pattern is now called the Cobra Effect and is a textbook example of why second-order thinking is not optional in policy design.
Example 2: The Prisoner's Paradox in Business Pricing
In 2011, Netflix announced it would split its DVD-by-mail and streaming services into separate subscriptions, effectively raising prices by 60% for customers who wanted both. The first-order analysis: higher prices per subscriber = more revenue. The strategy was defensible on a spreadsheet.
Second-order consequence: immediate customer outrage and a massive subscriber cancellation wave — Netflix lost approximately 800,000 subscribers in a single quarter. Third-order consequence: the stock price fell nearly 80%, damaging Netflix's ability to negotiate favorable content licensing deals at the exact moment it needed them most for its streaming pivot.
Reed Hastings reversed the decision and apologized publicly. The second-order effects — customer behavior, competitive response, and capital markets reaction — were catastrophic relative to the projected first-order revenue gain. Better second-order thinking would have surfaced these dynamics before the announcement.
Example 3: The Antibiotic Resistance Problem
The first-order consequence of prescribing antibiotics for bacterial infections is clear and positive: bacterial infection is treated, patient recovers. For decades, physicians reasonably acted on this first-order logic.
Second-order consequence: widespread antibiotic use creates selection pressure. Bacteria with random mutations conferring antibiotic resistance survive and reproduce; susceptible bacteria die. Over generations, resistant strains proliferate. Third-order consequence: antibiotics become less effective across entire populations — precisely when they are most needed.
Antibiotic resistance is now one of the most serious public health threats in the world, killing an estimated 1.27 million people directly in 2019. The crisis was created by individually rational first-order decisions, each of which was correct in isolation. Second-order thinking at the policy level — from the 1960s onward — would have surfaced the population-level resistance dynamics and produced more conservative prescribing guidelines far earlier.
When to Use It
✅ Before implementing any policy or structural change at the organizational or societal level — where rational actors will adapt their behavior in response.
✅ In investing and capital allocation — where consensus first-order thinking is typically priced in, and the real opportunity is in the second-order effects others haven't traced.
✅ When designing incentives — the second-order effects of incentive structures are where good intentions most reliably produce bad outcomes.
✅ For any strategy with a long implementation horizon — the longer the timeframe, the more second and third-order effects dominate outcomes.
✅ Before announcing a significant product or pricing change — customers and competitors will respond, and their response is part of the actual outcome.
❌ For genuine emergencies requiring immediate action. Second-order thinking is a deliberative process. In a crisis, act first on the best available first-order analysis, then use second-order thinking to anticipate the aftermath and adapt.
❌ When the consequence chain is genuinely unknowable. In highly complex systems with many unpredictable actors, second-order thinking can produce false precision. Acknowledge the uncertainty, don't manufacture a confident second-order story.
Model Combinations:
| Combine with | Effect |
|---|---|
| Scenario Planning | Build multiple second-order consequence chains across different scenarios |
| Systems Thinking | Second-order thinking is systems thinking applied to decision consequences |
| Inversion | Use inversion to generate negative second-order consequences you might be suppressing |
Common Misuses and Limitations
Misuse 1: Infinite regress. Second-order thinking doesn't mean tracing every consequence forever. At some point, additional iterations produce diminishing analytical value and increasing speculation. Two to three levels is typically sufficient for most decisions; more is usually noise.
Misuse 2: Paralysis via complexity. Tracing second-order consequences can make any action seem risky. Every decision has second-order effects; the question is whether they are large enough to change the decision. Use second-order thinking to inform, not to paralyze.
Misuse 3: Assuming you can predict human responses accurately. Second-order thinking is most powerful when the responding actors are rational and their incentives are knowable. In complex social systems, human responses are often unpredictable. Hold your second-order conclusions with appropriate humility.
Limitation — diminishing returns after third order: Beyond the second or third level, consequence chains become so speculative that they add more noise than signal. Second-order thinking is not a tool for predicting the far future — it's a tool for surfacing the near-future consequences that first-order analysis misses.
Related Models
Systems Thinking: Second-order thinking is the decision-level application of systems thinking — recognizing that consequences feed back into the system and change subsequent states.
Unintended Consequences: A closely related concept focused on the failures that emerge when second and third-order effects are ignored in complex systems.
Inversion: Inversion asks what could go wrong; second-order thinking traces how things go wrong through consequence chains.
FAQ
How far ahead should second-order thinking go?
Two to three levels is typically the right range for practical decisions. The first order is the direct consequence; the second is the response to that consequence; the third is the adaptation to the second. Beyond three levels, the uncertainty compounds so rapidly that the analysis becomes speculative. The goal is to surface consequences you'd otherwise miss, not to build a complete causal model of the future.
Isn't this just the same as long-term thinking?
They overlap but are distinct. Long-term thinking is about extending your time horizon. Second-order thinking is about tracing the depth of consequence chains — which often, but not always, correlates with time. A second-order effect can materialize within days (a competitor's immediate response to a pricing announcement) or take years (antibiotic resistance). The framework is about depth of causation, not just duration.
What's the best resource for learning Second Order Thinking?
Howard Marks's memo 'First and Second Level Thinking' (1990), available free on the Oaktree Capital website, is the canonical modern text. His book The Most Important Thing (2011) develops the concept further in the investing context. Donella Meadows's Thinking in Systems (2008) provides the systems science foundation.
Apply This Model with AI
Describe your proposed decision or plan in MindMax. The AI will trace the consequence chain — first-order effects, likely second-order responses, and the third-order adaptations that follow — and surface the dynamics most likely to be overlooked.
🚀 Apply Second Order Thinking in MindMax →
Further Reading
- Howard Marks, The Most Important Thing (2011) — The clearest articulation of second-level thinking in the investing context.
- Donella Meadows, Thinking in Systems (2008) — The systems science foundation; explains why second-order effects dominate long-run outcomes.
- Thomas Sowell, Applied Economics (2004) — Sowell's "first-stage thinking vs. second-stage thinking" applies the framework to economic policy with rich historical examples.
This page is part of the MindMax Mental Models Knowledge Base.