Transcript
SPEAKER_1: Alright, so I've been sitting with this story for a few days now, and I keep coming back to the same thought — the French colonial administration in Hanoi basically invented a rat tail economy. Like, that's the actual thing that happened. SPEAKER_2: It really is. And what makes it so striking is that it started from a completely rational place. There was a genuine plague threat. Officials were scared, and they needed a fast solution. SPEAKER_1: Right, so walk me through the threat first. Why were French officials in Hanoi so alarmed about rats specifically around 1902? SPEAKER_2: So this is the tail end — no pun intended — of what historians call the third global plague pandemic. It had started in the 1890s, centered on Hong Kong and southern China. Bubonic plague. And by the early 1900s it was spreading through colonial port cities across Asia. Hanoi was not isolated from that. SPEAKER_1: And they understood the rat connection by then? SPEAKER_2: They did. The science at that point recognized that plague was transmitted by fleas carried on rats. So rat extermination wasn't just a sanitation preference — it was the core public health strategy. Quarantines, urban cleaning, and kill the rats. SPEAKER_1: Here's what I find almost darkly ironic though. The French had just built a modern sewer system under Hanoi. European-style infrastructure. And that sewer system... SPEAKER_2: ...became a rat paradise. Exactly. The underground network gave rats ideal nesting conditions — shelter, warmth, food access. The very symbol of colonial modernity and urban progress had directly amplified the infestation. SPEAKER_1: [chuckle] So they built the problem and then had to solve the problem. SPEAKER_2: And the colonial framing at the time tried to blame Vietnamese neighborhoods for unsanitary conditions. But the infrastructure the French constructed — the sewers, the boulevards — that's what created the habitat. Historians are pretty clear on that point. SPEAKER_1: So what was the first attempt at a solution? SPEAKER_2: Professional teams. Vietnamese workers, overseen by French officials, paid to go into the sewers and hunt rats. But the rat population was enormous and the sewer network was extensive. The teams couldn't keep up. The campaign stalled. SPEAKER_1: Which is when someone had the idea — pay everyone. Open it up. SPEAKER_2: Right. The administration introduced a public bounty program. Residents could bring rat tails to collection points around the city and exchange them for cash. The tail was treated as proof of a kill. Simple, scalable, seemingly data-driven. SPEAKER_1: And it worked — in the sense that people showed up. SPEAKER_2: Massively. By mid-1902, colonial records described enormous numbers of tails being submitted. French observers were reportedly shocked by the volume. The incentive had clearly landed. SPEAKER_1: Wait — but the rat population wasn't actually going down? SPEAKER_2: That's the key idea. The tails were going up. The rats were not going down. And here's why: the bounty only required the tail. Not the carcass. So the verification mechanism had a gap you could drive a cart through. SPEAKER_1: So not dead rats — just tails. SPEAKER_2: Exactly. Some residents figured out they could catch a rat, clip the tail, and release the animal back into the sewers. The rat lives, breeds, produces more rats, and the hunter comes back for more tails. Think of it as a subscription model for rat tails. SPEAKER_1: And there are reports of people going further than that — actually keeping rats? SPEAKER_2: There are. Colonial records and later analyses mention anecdotal instances of individuals keeping rats — what you might call small-scale rat farming — to harvest tails repeatedly. The evidence is anecdotal rather than systematic, so listeners should hold that claim with some care. But the tail-clipping and release behavior? That's documented. SPEAKER_1: [short pause] So the policy created a market — and the market optimized for the metric, not the goal. SPEAKER_2: That's exactly the framing economists and policy scholars use today. The Hanoi rat bounty is now a canonical example of what's sometimes called a perverse incentive — or the cobra effect, named after a similar story from colonial India where a bounty on cobras led people to breed cobras for the reward. SPEAKER_1: And there's a broader name for this pattern, right? When measuring something turns it into a target and it stops reflecting the real goal? SPEAKER_2: Goodhart's Law. The principle that when a measure becomes a target, it ceases to be a good measure. The tail count was supposed to track rat deaths. Once it became the thing people were rewarded for, it tracked tail production — which is a completely different thing. SPEAKER_1: What would a better-designed policy have needed? Like, what should they have measured instead? SPEAKER_2: That's the hard question. Measuring actual rat population decline is difficult. But for example, they could have required carcasses — harder to fake. Or focused on habitat removal: block the sewer entry points, reduce food sources. The takeaway from historians is that the campaign treated Vietnamese residents as passive tools rather than strategic actors who would respond to incentives rationally. SPEAKER_1: And that connects to the colonial power dynamic underneath all of this. The sanitation infrastructure was concentrated in European quarters. The burden of rat-catching fell on Vietnamese residents. And when the policy failed, the framing blamed the population rather than the design. SPEAKER_2: Right. French colonial public health in Indochina consistently prioritized European residents and commercial interests. Vietnamese people were simultaneously blamed for unsanitary conditions and instrumentalized as low-paid labor in campaigns designed by officials who didn't consult them. That gap — between the designers and the people living inside the policy — is a big part of why it failed. SPEAKER_1: So the administration eventually shut it down? SPEAKER_2: They curtailed it once it became clear the rat population hadn't declined and the scheme was being gamed. But the damage was done — not just to the rat campaign, but as a case study in what happens when top-down technocratic policy ignores local social and economic realities. SPEAKER_1: they were measuring tails, but the goal was fewer living rats spreading plague. The moment those two things came apart, the whole program worked against itself. SPEAKER_2: And that's why this story keeps showing up in public health ethics courses and economics classrooms more than a century later. The Hanoi rat bounty isn't just a quirky colonial footnote. It's a precise illustration of what happens when planners reward a proxy and assume the real outcome will follow. It won't — not if the people inside the system have any agency at all. And they always do. SPEAKER_1: What strikes me about that last point — the agency piece — is that it applies well beyond colonial Hanoi. The moment any population is inside a system with clear incentives, they will find the edges of it. SPEAKER_2: Always. And that's not cynicism — that's just rational behavior. The Vietnamese residents weren't doing anything wrong by the logic of the bounty. The rules said: tail equals payment. They followed the rules. The policy design was the failure, not the people. SPEAKER_1: So let's sit with the colonial dimension a bit more, because I think it's easy to tell this as a funny quirky story and miss what's underneath it. The sanitation infrastructure — who was it actually built for? SPEAKER_2: European residents, primarily. French colonial urban planning in Hanoi segregated the European quarter from indigenous neighborhoods. Sewers, paved streets, modern boulevards — those were concentrated in European areas. The Vietnamese majority lived with far less of that infrastructure. SPEAKER_1: And yet the rat problem that infrastructure created — the sewer network as rat habitat — that spread everywhere. SPEAKER_2: Right. The rats didn't respect the colonial zoning. So you had a situation where French engineering produced the infestation, French officials designed the bounty response, and Vietnamese residents were recruited as the low-paid labor to fix it. And when the scheme was gamed, the framing shifted blame onto the population rather than the design. SPEAKER_1: That's a significant power asymmetry. The people bearing the burden of the campaign had no role in designing it. SPEAKER_2: None. And historians argue that's a core reason it failed. When you impose a top-down technocratic measure without collaboration or trust, you lose the local knowledge that might have caught the flaw before it became a crisis. Someone on the ground could have told them: if you only require the tail, people will only bring you the tail. SPEAKER_1: Mm-hmm. So the information gap and the power gap were the same gap. SPEAKER_2: Exactly. And that's a pattern that shows up across colonial public health broadly. French Indochina's plague response consistently prioritized European residents and commercial interests. The welfare of the Vietnamese majority was secondary — which meant the people who knew the city best were the last ones consulted. SPEAKER_1: Let's talk about the global context for a moment, because this wasn't just a Hanoi problem. The plague threat was worldwide at this point. SPEAKER_2: It was. The Hanoi outbreaks were part of what historians call the third global plague pandemic, which began in the 1890s — centered initially on Hong Kong and southern China. It spread through colonial port cities across Asia and beyond. International sanitary conferences were convened. New surveillance regimes were established. Every colonial administration was under pressure to show it was managing the threat. SPEAKER_1: So French officials in Hanoi weren't just worried about rats. They were worried about looking like they had control. SPEAKER_2: That's a fair reading. The bounty program was also a performance of governance — a visible, countable response to an invisible threat. Enormous numbers of tails coming in looked like progress. It looked like data. The problem is that the data was measuring the wrong thing entirely. SPEAKER_1: Think of it like a hospital measuring success by the number of forms filed rather than patient outcomes. The metric looks clean. The underlying reality is a different story. SPEAKER_2: That's a good analogy. And it maps directly onto what policy literature now calls the cobra effect — named after a similar story from colonial India, where a bounty on cobras led people to breed cobras for the reward. The Hanoi case is frequently compared to that one precisely because the mechanism is identical: reward the proxy, distort the outcome. SPEAKER_1: For everyone listening who works in any kind of measurement-driven environment — and that's most people — this is the part that should land hard. What would a genuinely better policy have looked like? SPEAKER_2: Requiring carcasses instead of tails would have been a start — harder to fake. But the deeper fix would have been habitat reduction: block sewer entry points, reduce food sources, address the conditions that made the rat population explode in the first place. Treat the cause, not just the symptom. And critically — involve the people who actually live in the city in designing the solution. SPEAKER_1: [short pause] So the takeaway isn't just about rats. It's about what happens when you mistake the measure for the goal. SPEAKER_2: That's it. The key idea that historians and economists keep returning to is this: the Hanoi rat bounty failed not because the people were uncooperative, but because the policy rewarded a proxy — tails — while assuming the real outcome — fewer living rats — would automatically follow. It won't. Not when the people inside the system are rational actors with their own economic pressures and their own agency. SPEAKER_1: And more than a century later, this story keeps showing up in public health ethics courses and economics classrooms because the lesson is that portable. SPEAKER_2: It really is. The Hanoi rat bounty is now a canonical case study — cited in policy analysis, in discussions of moral hazard, in governance literature about gaming measurement-based incentives. What started as a colonial administrator's practical solution to a plague crisis became one of the clearest illustrations we have of Goodhart's Law in action. When a measure becomes a target, it stops being a good measure. Hanoi, 1902, proved that with rat tails. SPEAKER_1: And that phrase — Goodhart's Law — is worth sitting with for a moment, because it's doing a lot of work in this story. When a measure becomes a target, it stops being a good measure. That's the whole thing, right there. SPEAKER_2: It really is. And what makes the Hanoi case so instructive is that the measure seemed so sensible on the surface. A tail is physical proof. It's countable. It's verifiable. French officials thought they had designed an airtight system. SPEAKER_1: But airtight for whom? Airtight from the perspective of someone sitting in an office counting tails. Not airtight from the perspective of someone who needs income and has figured out that the tail is the only part that matters. SPEAKER_2: Exactly. And that gap — between the administrator's view and the resident's view — is where the whole policy collapsed. The moment you separate the proof from the outcome, you've handed the system over to whoever is most motivated to exploit the gap. SPEAKER_1: So not requiring the carcass was the single design flaw that unraveled everything. SPEAKER_2: It was the critical one, yes. Requiring carcasses would have been harder to fake — you can't clip a carcass and release it. But even that wouldn't have addressed the deeper problem, which was habitat. The rats had an ideal environment in those sewers. Remove the tails, the rats remain. Remove the sewers as rat habitat, and you start actually reducing the population. SPEAKER_1: Think of it like treating a fever with a cold cloth. The cloth addresses the symptom. The infection is still running underneath. SPEAKER_2: That's a clean analogy. And it maps onto what historians argue was the structural failure here — the campaign treated rat tails as the problem rather than the conditions that produced the rats. The sewers, the food sources, the nesting environments. Those were the actual levers. SPEAKER_1: And those levers were harder to pull. Counting tails is fast. Redesigning sewer infrastructure is expensive and slow. SPEAKER_2: Right — and politically, counting tails looked like action. For French officials under pressure from a global plague pandemic, visible data mattered. The third pandemic had started in the 1890s, Hong Kong and southern China were the epicenter, and every colonial administration was being watched. Enormous numbers of tails coming in read as competence. SPEAKER_1: [short pause] So the bounty was partly a public relations exercise dressed up as a public health campaign. SPEAKER_2: That's a fair characterization. And it's not unique to Hanoi — colonial public health broadly tended to prioritize the appearance of control alongside the reality of it. The problem is that when the appearance diverges from the reality, the appearance wins in the short term and the reality catches up later. SPEAKER_1: And the reality catching up here was — the rat population hadn't declined. Officials eventually had to curtail the program once that became undeniable. SPEAKER_2: Correct. The bounty had to be pulled back once it was clear the infestation wasn't shrinking and the incentive scheme was being gamed rather than serving its purpose. By that point, the program had created a small economy around rat tails — collection points, cash exchanges, residents who had built income streams around it — and none of that translated into fewer rats spreading plague. SPEAKER_1: For everyone who works in any kind of metrics-driven environment — and that's most people — the key idea here is that the metric and the goal have to be the same thing, or close enough that gaming one automatically serves the other. SPEAKER_2: That's the design principle the Hanoi campaign violated. And it's why this case keeps appearing in public health ethics courses and economics classrooms more than a century later. The lesson is that portable. Whether someone is running a hospital, a school, a government program — if the incentive rewards the proxy and not the outcome, rational actors will optimize the proxy. SPEAKER_1: Mm-hmm. And the people doing the optimizing aren't villains. They're just responding to the rules as written. SPEAKER_2: Precisely. The Vietnamese residents who clipped tails, who may have kept rats, who found every edge of the system — they were doing exactly what the bounty invited them to do. The policy design was the failure. That distinction matters, especially given the colonial framing that often blamed the population for the campaign's shortcomings rather than examining the incentive structure itself. SPEAKER_1: So the lasting takeaway from Hanoi, 1902, is really two things layered together. One is the incentive design lesson — measure the right thing or the measure will be gamed. The other is the governance lesson — top-down technocratic solutions imposed without local collaboration tend to fail in ways that local knowledge would have caught. SPEAKER_2: Those two lessons are inseparable in this case. The information gap and the power gap were the same gap. And that's what makes the Hanoi rat bounty more than a curious historical anecdote. It's a precise, documented example of what happens when planners treat people as passive instruments rather than strategic actors with their own economic realities. The rats outlasted the policy. The lesson shouldn't. SPEAKER_1: And that phrase — Goodhart's Law — is worth sitting with for a moment, because it's doing a lot of work in this story. When a measure becomes a target, it stops being a good measure. That's the whole thing, right there. SPEAKER_2: It really is. And what makes the Hanoi case so instructive is that the measure seemed so sensible on the surface. A tail is physical proof. It's countable. It's verifiable. French officials thought they had designed an airtight system. SPEAKER_1: But airtight for whom? Airtight from the perspective of someone sitting in an office counting tails. Not airtight from the perspective of someone who needs income and has figured out that the tail is the only part that matters. SPEAKER_2: Exactly. And that gap — between the administrator's view and the resident's view — is where the whole policy collapsed. The moment you separate the proof from the outcome, you've handed the system over to whoever is most motivated to exploit the gap. SPEAKER_1: So not requiring the carcass was the single design flaw that unraveled everything. SPEAKER_2: It was the critical one, yes. Requiring carcasses would have been harder to fake. But even that wouldn't have addressed the deeper problem — habitat. The rats had an ideal environment in those sewers. Remove the tails, the rats remain. Remove the sewers as rat habitat, and you start actually reducing the population. SPEAKER_1: Think of it like treating a fever with a cold cloth. The cloth addresses the symptom. The infection is still running underneath. SPEAKER_2: That's a clean analogy. And it maps onto what historians argue was the structural failure here — the campaign treated rat tails as the problem rather than the conditions that produced the rats. The sewers, the food sources, the nesting environments. Those were the actual levers. SPEAKER_1: And those levers were harder to pull. Counting tails is fast. Redesigning sewer infrastructure is expensive and slow. SPEAKER_2: Right — and politically, counting tails looked like action. For French officials under pressure from a global plague pandemic, visible data mattered. The third pandemic had started in the 1890s, Hong Kong and southern China were the epicenter, and every colonial administration was being watched. Enormous numbers of tails coming in read as competence. SPEAKER_1: [short pause] So the bounty was partly a public relations exercise dressed up as a public health campaign. SPEAKER_2: That's a fair characterization. And it's not unique to Hanoi — colonial public health broadly tended to prioritize the appearance of control alongside the reality of it. The problem is that when the appearance diverges from the reality, the appearance wins in the short term and the reality catches up later. SPEAKER_1: And the reality catching up here was — the rat population hadn't declined. Officials eventually had to curtail the program once that became undeniable. SPEAKER_2: Correct. The bounty had to be pulled back once it was clear the infestation wasn't shrinking and the incentive scheme was being gamed rather than serving its purpose. By that point, the program had created a small economy around rat tails — collection points, cash exchanges, residents who had built income streams around it — and none of that translated into fewer rats spreading plague. SPEAKER_1: For everyone who works in any kind of metrics-driven environment — and that's most people — the key idea here is that the metric and the goal have to be the same thing, or close enough that gaming one automatically serves the other. SPEAKER_2: That's the design principle the Hanoi campaign violated. And it's why this case keeps appearing in public health ethics courses and economics classrooms more than a century later. The lesson is that portable. Whether someone is running a hospital, a school, a government program — if the incentive rewards the proxy and not the outcome, rational actors will optimize the proxy. SPEAKER_1: Mm-hmm. And the people doing the optimizing aren't villains. They're just responding to the rules as written. SPEAKER_2: Precisely. The Vietnamese residents who clipped tails, who may have kept rats, who found every edge of the system — they were doing exactly what the bounty invited them to do. The policy design was the failure. That distinction matters, especially given the colonial framing that often blamed the population for the campaign's shortcomings rather than examining the incentive structure itself. SPEAKER_1: So the lasting takeaway from Hanoi, 1902, is really two things layered together. One is the incentive design lesson — measure the right thing or the measure will be gamed. The other is the governance lesson — top-down technocratic solutions imposed without local collaboration tend to fail in ways that local knowledge would have caught. SPEAKER_2: Those two lessons are inseparable in this case. The information gap and the power gap were the same gap. And that's what makes the Hanoi rat bounty more than a curious historical anecdote. It's a precise, documented example of what happens when planners treat people as passive instruments rather than strategic actors with their own economic realities. The rats outlasted the policy. The lesson shouldn't. SPEAKER_1: And that phrase — Goodhart's Law — is worth sitting with for a moment, because it's doing a lot of work in this story. When a measure becomes a target, it stops being a good measure. That's the whole thing, right there. SPEAKER_2: It really is. And what makes the Hanoi case so instructive is that the measure seemed so sensible on the surface. A tail is physical proof. It's countable. It's verifiable. French officials thought they had designed an airtight system. SPEAKER_1: But airtight for whom? Airtight from the perspective of someone sitting in an office counting tails. Not airtight from the perspective of someone who needs income and has figured out that the tail is the only part that matters. SPEAKER_2: Exactly. And that gap — between the administrator's view and the resident's view — is where the whole policy collapsed. The moment you separate the proof from the outcome, you've handed the system over to whoever is most motivated to exploit the gap. SPEAKER_1: So not requiring the carcass was the single design flaw that unraveled everything. SPEAKER_2: It was the critical one, yes. Requiring carcasses would have been harder to fake. But even that wouldn't have addressed the deeper problem — habitat. The rats had an ideal environment in those sewers. Remove the tails, the rats remain. Remove the sewers as rat habitat, and you start actually reducing the population. SPEAKER_1: Think of it like treating a fever with a cold cloth. The cloth addresses the symptom. The infection is still running underneath. SPEAKER_2: That's a clean analogy. And it maps onto what historians argue was the structural failure here — the campaign treated rat tails as the problem rather than the conditions that produced the rats. The sewers, the food sources, the nesting environments. Those were the actual levers. SPEAKER_1: And those levers were harder to pull. Counting tails is fast. Redesigning sewer infrastructure is expensive and slow. SPEAKER_2: Right — and politically, counting tails looked like action. For French officials under pressure from a global plague pandemic, visible data mattered. The third pandemic had started in the 1890s, Hong Kong and southern China were the epicenter, and every colonial administration was being watched. Enormous numbers of tails coming in read as competence. SPEAKER_1: [short pause] So the bounty was partly a public relations exercise dressed up as a public health campaign. SPEAKER_2: That's a fair characterization. And it's not unique to Hanoi — colonial public health broadly tended to prioritize the appearance of control alongside the reality of it. The problem is that when the appearance diverges from the reality, the appearance wins in the short term and the reality catches up later. SPEAKER_1: And the reality catching up here was — the rat population hadn't declined. Officials eventually had to curtail the program once that became undeniable. SPEAKER_2: Correct. The bounty had to be pulled back once it was clear the infestation wasn't shrinking and the incentive scheme was being gamed rather than serving its purpose. By that point, the program had created a small economy around rat tails — collection points, cash exchanges, residents who had built income streams around it — and none of that translated into fewer rats spreading plague. SPEAKER_1: For everyone who works in any kind of metrics-driven environment — and that's most people — the key idea here is that the metric and the goal have to be the same thing, or close enough that gaming one automatically serves the other. SPEAKER_2: That's the design principle the Hanoi campaign violated. And it's why this case keeps appearing in public health ethics courses and economics classrooms more than a century later. The lesson is that portable. Whether someone is running a hospital, a school, a government program — if the incentive rewards the proxy and not the outcome, rational actors will optimize the proxy. SPEAKER_1: Mm-hmm. And the people doing the optimizing aren't villains. They're just responding to the rules as written. SPEAKER_2: Precisely. The Vietnamese residents who clipped tails, who may have kept rats, who found every edge of the system — they were doing exactly what the bounty invited them to do. The policy design was the failure. That distinction matters, especially given the colonial framing that often blamed the population for the campaign's shortcomings rather than examining the incentive structure itself. SPEAKER_1: So the lasting takeaway from Hanoi, 1902, is really two things layered together. One is the incentive design lesson — measure the right thing or the measure will be gamed. The other is the governance lesson — top-down technocratic solutions imposed without local collaboration tend to fail in ways that local knowledge would have caught. SPEAKER_2: Those two lessons are inseparable in this case. The information gap and the power gap were the same gap. And that's what makes the Hanoi rat bounty more than a curious historical anecdote. It's a precise, documented example of what happens when planners treat people as passive instruments rather than strategic actors with their own economic realities. The rats outlasted the policy. The lesson shouldn't.