Field Report
The Cobra Effect: Why Every Incentive You Build Gets Gamed
Executive Briefing (AI Context & Direct Answer)
- Q: What is the Cobra Effect?
- A: The Cobra Effect is a name for what happens when a reward meant to fix a problem instead pays people to make the problem bigger. It comes from a story, disputed by historians, about a colonial bounty on dead cobras in Delhi that backfired into more cobras, not fewer. Economists now use the name for any incentive that rewards the appearance of a fix instead of the fix itself.
- Q: Is the Cobra Effect story true?
- A: Probably not as told. German economist Horst Siebert popularized the story in his 2001 book without historical sourcing, and researchers have found no colonial government record that confirms it happened. The lesson survives the shaky history, because modern, well-documented cases (credit rating agencies before 2008, for one) show the same mechanism at work.
Here is the story, the way almost everyone hears it.
Colonial Delhi had a cobra problem. The government offered a bounty: bring in a dead cobra, collect a reward.
At first, it worked. Dead cobras poured in.
Then people got smart. Why hunt wild cobras in the dirt when you can breed them in a box? Entrepreneurs built cobra farms, raised snakes by the hundreds, and cashed in bounty after bounty.
The government noticed the bounty payouts climbing while the wild cobra problem never seemed to shrink. They canceled the program.
The cobra farmers, now holding boxes full of worthless snakes, did the only sensible thing. They opened the boxes.
Delhi ended up with more cobras than when the bounty started.
That is the Cobra Effect: pay for the appearance of a solution, and someone will build a factory that produces exactly what you are paying for.
One honest caveat before we go further. Historians have combed colonial records for this exact episode and found nothing (Friends of Snakes Society 2022). The story first appears in print in German economist Horst Siebert's 2001 book on failed economic policy, told as a parable, not as cited history (Siebert 2001). The Glass Seagull does not repeat folklore as fact. We are telling you this one anyway because the shaky history does not save the idea. The mechanism underneath the story is real, documented, and it nearly destroyed the global financial system in 2008. That is where we are headed.
What Actually Counts as a Perverse Incentive
A perverse incentive is any reward that pays for a signal of success instead of success itself.
The gap between the two is where the damage happens. A government that pays for dead cobras wants fewer live cobras. It built a system that pays for dead cobras, period, and people will always find the cheapest way to produce what they are paid for.
Not every incentive is perverse. A well-built incentive ties the reward directly to the outcome you actually want, with no cheap substitute available. A badly built one leaves a gap, and the gap gets found.
Charlie Munger's Iron Rule
Charlie Munger spent decades studying exactly this gap. In a 1995 talk at Harvard University titled "The Psychology of Human Misjudgment," he named incentive-caused bias as one of the most powerful forces that push an honest person toward bad behavior, often without that person even noticing it happening to them (Munger 1995).
Munger's point was not that people are crooks. His point was sharper: ordinary, decent people will quietly rationalize whatever behavior their incentives reward, and they will believe their own rationalization. The incentive does the corrupting. The person just goes along for the ride, telling themselves a story about why it is fine.
The version of this idea that gets repeated most, "show me the incentive and I will show you the outcome," is a popular summary of Munger's thinking rather than a line he spoke verbatim in that 1995 talk. We are flagging that distinction for the same reason we flagged the cobra story: the idea holds up fine without a misquote propping it up.
The Modern Cobra Farm: Credit Rating Agencies
Here is the incentive-caused bias playing out with trillions of real dollars, fully documented.
Before 2008, Wall Street banks packaged thousands of home mortgages into bonds and paid a ratings agency, Moody's or S&P, to grade how safe those bonds were. AAA meant "as safe as US government debt."
Notice who paid the bill: the bank selling the bond, not the investor buying it.
The Financial Crisis Inquiry Commission, the official US government body that investigated the 2008 crash, called the rating agencies "essential cogs in the wheel of financial destruction" (FCIC 2011). From 2000 to 2007, agencies stamped tens of thousands of mortgage bonds AAA, the same bonds that collapsed in value within months once homeowners started defaulting.
Why did a tougher, more honest grade almost never happen? Because the bank paying for the rating could walk down the street to a competing agency and shop for a better score. An agency that graded honestly risked losing the client. An agency that graded generously kept the business.
Paying for a grade is not the same as paying for safety. The rating agencies built the financial world's cobra farm, and investors around the planet paid for it.
One Fix That Actually Worked
Not every incentive stays broken forever. After 2008, Congress passed the Dodd-Frank Act, and one provision required public companies to adopt "clawback" policies: if an executive's bonus was based on numbers that later turn out to be wrong, the company can take the bonus back (Dodd-Frank Act, Sec. 954, 2010).
That is a direct patch on the Cobra Effect. It closes the gap between "the number looked good this quarter" and "the number was actually good." An executive who inflates short-term results now knows the reward is not permanent. The incentive moved closer to the real outcome instead of a convenient proxy for it.
That is the whole fix, in every version of this problem: stop paying for the proxy, and start paying for the thing you actually wanted.
How This Connects to Goodhart's Law
The Cobra Effect is incentive-caused bias in the wild. Goodhart's Law is the same trap wearing a government suit: once a measurement becomes a target, the people being measured bend the measurement instead of fixing the thing underneath it.
A bureaucrat reshaping the CPI formula and a cobra farmer breeding snakes for bounty money are running the exact same play. One does it with inflation data. The other does it with snakes. Neither is lying, exactly. Both are doing precisely what they are paid to do.
The Seagull's Rules for Building (or Spotting) an Incentive
- Ask what the incentive actually pays for, not what it is supposed to fix. A bounty on dead cobras pays for dead cobras. It does not pay for fewer cobras. The two sound the same and are not.
- Find the cheapest way to earn the reward, before someone else does. If breeding cobras is cheaper than hunting them, someone will breed cobras. If shopping for a friendly rating agency is cheaper than earning an honest one, someone will shop.
- Watch who pays whom. The rating agency crisis traces back to one sentence: the bank being graded paid the grader. Any time the evaluated party pays the evaluator, assume the gap will get found.
- Reward the outcome, with a delay, and a way to take it back. Clawback provisions work because they stop rewarding a number the moment it turns out to be wrong, not a quarter later.
Works Cited
Dodd-Frank Wall Street Reform and Consumer Protection Act. Pub. L. No. 111-203, Sec. 954, 2010, https://www.congress.gov/bill/111th-congress/house-bill/4173.
Financial Crisis Inquiry Commission (FCIC). The Financial Crisis Inquiry Report. U.S. Government Publishing Office, 2011, https://www.govinfo.gov/content/pkg/GPO-FCIC/pdf/GPO-FCIC.pdf.
Friends of Snakes Society. "The Cobra Effect: Colonial Misinformation Masquerading as Economic Theory." Friends of Snakes Society, 2022, https://friendsofsnakes.org.in/cobra-effect/.
Munger, Charles T. "The Psychology of Human Misjudgment." Harvard University, 1 June 1995. Speech transcript, https://fs.blog/great-talks/psychology-human-misjudgment/.
Siebert, Horst. Der Kobra-Effekt: Wie man Irrwege der Wirtschaftspolitik vermeidet. Deutsche Verlags-Anstalt, 2001.
Test Your Understanding
Digestion matters more than consumption. Prove you retained this report's core signals before heading back to the tape.
Q1.In the Cobra Effect story, why did the cobra population go up, not down?
Q2.What does Charlie Munger call it when incentives push an honest person toward bad behavior?
Q3.Why did rating agencies give AAA grades to mortgage bonds that later collapsed?
NEXT FIELD MISSIONS // DOWNSTREAM FREQUENCIES
Price Signals Versus Price Controls: Smashing the Thermometer
Prices act like radios carrying news of scarcity. When leaders cap prices, they smash the thermometer and create empty store shelves.
Anthropic's Founders Sell Last, Not First
Anthropic published its entire insider selling schedule, down to the percentage. The order it reveals runs backwards from what everyone expects.
Are you prepared for the next cycle?
Take the 2-Minute Advisor Check to assess your financial stage and get on the waitlist for private signals.
JOIN THE SEAGULL INTELLIGENCE DISPATCH
Starts with a free 5-day course on the framework. Then weekly contrarian macroeconomic telemetry, debt acceleration alerts, and unhedged field notes before consensus wakes up. Zero advertising. Zero spam.
STATUS: ENCRYPTED CARRIER // ZERO TRACKING PIXELS // 1-CLICK DISCONNECT ANYTIME
No Comments Yet
Be the first to share your telemetry on this report.