What this covers
Michael Munger of Duke University discusses with Russ Roberts why the ability to innovate without seeking approval stands as the organizing principle of political economy. Munger argues that permissionless innovation—a default presumption favoring experimentation over gatekeeping—determines whether societies prosper or stagnate. The conversation draws repeatedly on emergent order, the unplanned coordination that arises when individuals act within shared constraints. Two metaphors anchor this: a perfectly executed English dance where dancers move without collision while following their own lead, and a jazz band where soloists compete yet merge their improvisations into a coherent whole. These images illuminate how markets and decentralized systems coordinate action more reliably than top-down design.
The core mechanism appears in three connected claims: government experts and incumbent competitors are both structurally biased toward the status quo; the true value of most innovations emerges only after deployment and real-world use; and requiring permission raises the marginal cost of experimentation, shrinking the search process that produces growth. Munger illustrates through cases: Federal Express exploiting a postal monopoly loophole; United Airlines preventing disaster by granting ground staff discretion to offer higher compensation; Certificate of Need laws that force hospital entrants to obtain approval from the very competitors harmed by competition. The discussion also addresses costs: permissionless innovation imposes real harm on losing firms and workers facing churn, though general equilibrium effects favor consumers over time. Tensions surface around whether local knowledge and emergent use can reliably prevent catastrophic failure, and whether limiting non-compete agreements—as California does—truly explains Silicon Valley's success over Boston, or merely tells a convenient story.
Munger argues that permissionless innovation—a default presumption in favor of trying new things without seeking approval from either government experts or incumbent competitors—is the single most important concept in political economy, because the value of innovations is usually unknowable in advance and requiring permission systematically suppresses the experimentation that produces growth.
- Both government experts and incumbent competitors are biased toward the status quo and would reject innovations whose value is not yet apparent
- The true value of an innovation often emerges only after it is deployed and users repurpose it (Twitter, FedEx)
- Requiring permission raises the marginal cost of each experiment, slowing the search process that produces innovation
Permissionless innovation is the key factor explaining differential prosperity and growth across nations and sectors.
- Permissionless innovation—the ability to innovate without having to ask permission—is the most important concept in political economy, the key factor explaining why some nations grow and prosper while others remain near the bottom of possible growth paths.
“in the thing that we use to understand why it is that some nations grow and prosper and others bump along near the bottom of the possible range of growth paths, I think permissionless innovation is the most important concept.”
- Massachusetts permitted non-compete agreements while California restricts them, so in California a worker could leave one employer for another and use their knowledge freely, increasing labor mobility and permissionless innovation, which may help explain why Silicon Valley rather than Boston became the hub of internet and sharing-economy innovation—though Munger flags this as a 'Just So Story.'
“California doesn't allow, or at least restricts the use of non-compete agreements. That means that I can go somewhere else without the state or my employer's permission... it meant that the degree of permissionless innovation in Boston held them back.”
- Federal Express exists because of an 'extremely urgent' courier loophole in the postal monopoly law; FedEx had to effectively obtain permission from its government competitor, the Post Office, with courts granting that the urgency exemption applied—which is why FedEx envelopes are labeled 'Extremely Urgent.'
“What they're doing is invoking that loophole in the law that gives the Post Office a monopoly... that's why Federal Express exists.”
- In many states a hospital that wants to open must obtain a Certificate of Need, literally requiring existing hospitals to confirm there is a need they cannot already provide—forcing an entrant to get permission from the very competitors who would be harmed.
“It's even called a Certificate of Need. You have to get your competitors to say, 'You know, we can't provide this service.'”
- Twitter is a paradigmatic case of permissionless innovation because it had almost no rules: its founders did not know what it was (calling it micro-blogging, then a social network, then realizing it was more an information network), and the real innovation came from the many users who repurposed the open platform for unexpected ends, building micro-communities its founders never anticipated.
“the innovation in Twitter was not Twitter. It was the use that people put Twitter to, once it was available... the reason that Twitter is such an important example of permissionless innovation is that it had almost no rules, and so it allowed all these different embellishments and forms.”
Permission-based systems reduce the frequency and efficiency of innovation by filtering experiments through inherently conservative experts and threatened competitors.
- Innovation is a search process in which most attempts fail; if every experiment requires permission, then at the margin the number of experiments falls, so the search for innovations proceeds much more slowly and less efficiently.
“if every time you want to try an experiment you have to get permission, at the margin it's going to reduce the number of those experiments. Which means that the search for innovations will proceed much more slowly and less efficiently.”
- An innovator may have to seek two distinct and equally damaging kinds of permission: from government experts, who are inherently conservative and protect the status quo, and from competitors, who oppose innovations precisely because those innovations would benefit consumers and harm the incumbents' profits.
“the kind of permission that we have to get is either from experts who work for the state and are inherently conservative--they try to protect the status quo because that's what the conventional wisdom is. And the other is, permission from competitors, who might get to make something that's going to harm your profit position.”
- Experts, including economists, are inherently poor at predicting and producing innovations because they reason within existing infrastructure and conventional wisdom, whereas entrepreneurs try something else; most ideas that seem stupid to experts really are bad, but some turn out to be transformative, and experts cannot reliably tell them apart.
“Experts by their nature are really bad at innovation, because what experts understand... we tend to think in terms of the existing infrastructure. Whereas entrepreneurs say, 'Let's try to do something else.'”
- If you had asked computer experts in 1979 or 1980 whether society should devote resources to personal microcomputers, they all would have said no, because politics works at the median and experts are inherently conservative.
“if you had asked experts in computers in, say, 1979, 1980, should we spend society's resources on microcomputers, on personal computers? They all would have said no. Politics works at the median, and experts are inherently conservative.”
Permissionless innovation's success depends on rebuttable presumptions favoring action, low transaction costs, and rule of law that constrains arbitrary discretion.
- Permissionless innovation is a rebuttable presumption in favor of innovation and autonomy: rather than enumerating what is permitted, the system enumerates a list of what is prohibited and treats everything else as permitted by default—analogous to right-turn-on-red in the US versus Germany.
“it's a presumption in favor of innovation. It's a rebuttable presumption. But, the presumption is in favor of: 'Here is the list of things that is prohibited. Everything else is permitted.'”
- Asking forgiveness is favored over asking permission partly because enforcement is costly and the probability of getting caught is much less than one, so an actor may not be caught in the first place—as teenagers intuitively grasp when they do things knowing the answer would be no.
“enforcement is costly. And the probability of getting caught is much less than 1.”
- Hopper's Law, attributed to Rear Admiral and computer scientist Grace Hopper, holds that it is easier to ask forgiveness than permission, because asking permission in advance means you may never get to act and you alert others to your plans; the tech industry invokes it to argue that experimentation with new technologies should be permitted by default.
“it's easier to ask forgiveness than it is to ask permission. So, go ahead and do it; and then, if it screws up, you can ask for forgiveness later. But, if you ask for permission in advance you'll never find out and probably you'll alert other people to your plans.”
- Following Douglass North, transactions costs are a candidate for the most important concept in political economy, and the level of transactions costs—red tape, bureaucracy for using land, property, or starting a business—is a major determinant of how much permissionless innovation a society allows.
“transactions costs are a big factor in determining the level of permissionless innovation.”
- Rule of law means we are governed by principles rather than individual discretion, applied to everyone without privilege, but it is necessary and not sufficient, because the content of the laws still matters—stupid laws applied equally still block beneficial innovation.
“Rule of Law means that we are governed by principles, not by individual discretion; and that those rules apply to everyone without any kind of privilege. But it would still depend on the content of the laws.”
- Twitter's combative, corrosive dynamics—trolls, loss of civility—threaten to turn it from an English Dance into a mosh pit governed by very different rules, raising the question of what limits should apply to permissionless innovation and whether people will increasingly abandon it.
“Twitter is, because of the kind of corrosive, combative dance, like a bit mosh pit. The rules of a mosh pit are a lot different than the rules of the English Dance.”
Effective systems grant discretionary authority to trusted people with local knowledge rather than requiring escalation and advance permission from superiors.
- Decisions should be made by the person on the ground with the most local knowledge rather than escalated for permission, provided that person is trusted and trained; in the military the unit commander is responsible for all the unit does or fails to do, and if you cannot trust them you relieve them of command rather than micromanage, because superiors do not know enough.
“You want the person on the ground with the most information to make the decision-- if, in fact, you trust that person.”
- United Airlines' violent removal of a passenger resulted from employees lacking discretion under rigid rules capping volunteer compensation; after the incident United changed the rules to allow far higher offers (roughly tripling the cap), demonstrating that entrusting employees with discretion to make decisions a manager would make on the spot prevents catastrophic outcomes.
“they didn't have any discretion... United Airlines then changed the rules in just the way that you have suggested, allowing people to innovate without permission. Now, they can go up to $1500 or $2000.”
- Drawing on Taleb, the mess-ups worth worrying about are not everyday errors—which should be allowed—but catastrophic, fragility-revealing failures; systems should have redundancy and grant people enough discretion to detect and prevent things going really, really wrong, as in the United case where discretion would have prevented disaster.
“You want a system that's not fragile with respect to really, really bad mess-ups. You want redundancy, and you want people to have enough discretion to see that and prevent it.”
Permissionless innovation imposes costs on incumbents and workers but generates greater net consumer benefit and systemic adaptation over time.
- Permissionless innovation imposes real harm on losing companies and forces workers into constant churn and turnover that they might prefer to avoid, but the general equilibrium effect over time is more innovation and greater benefit to consumers.
“the problem with it is, it really does cause a lot of harm to the companies that lose. And that churning means that you are constantly turning over where you work... overall you get more innovation and more benefit to consumers.”
- Whether non-compete agreements are good for workers is unclear: they may reduce a worker's freedom to move, but where they exist firms invest more in training and pay higher wages, so banning them does not straightforwardly benefit workers.
“It's not clear whether it's good for workers. In fact, I'd probably get paid quite a bit more in Boston. And I get better trained. They are more willing to invest.”
- Incumbent restaurants in North Carolina have repeatedly used regulations—often invoking health justifications—to prevent food trucks from operating, and the fact that competitors oppose the food trucks is itself evidence that the food trucks would benefit consumers.
“restaurants have succeeded in preventing that... that's what the competitors are worried about, that you are going to provide something to consumers--that's the evidence for why they are opposing it.”
Well-designed market systems and social coordination achieve grace by reconciling individual plans and enabling smooth cooperation without central direction.
- Schiller's metaphor presents the ideal of a beautiful society as a well-executed English dance: from a balcony one sees infinite criss-crossing motions that change direction without colliding, each dancer vacating a position before another arrives, everyone seeming to follow their own lead without getting in anyone's way—a perfect symbol of individually asserted freedom combined with respect for the freedom of others.
“I know of no better image for the ideal of a beautiful society than a well executed English dance... Such a dance is the perfect symbol of one's own individually asserted freedom as well as of one's respect for the freedom of the other.”
- Social dancing exhibits 'grace'—gracefully not bumping into others without ostentation, so it all works smoothly and unnoticed—which is distinct from the feline or animal grace of a gymnastics routine that displays athleticism for applause; the relevant market virtue is the social kind, accomplishing what you want without stepping on others.
“grace is not just graceful motion, but gracefully, you know, not bumping into other people without seeming ostentatiously to say, 'Oh, no, you go ahead.' It's just that it all happens in a way that no one really notices, because it works so smoothly.”
- In the Austrian view, the market and price system reconciles all the conflicting plans people hold: each person changes their plans in response to new information about opportunity cost, prices, and profits, and these plans are reconciled by the rules so smoothly that the coordination goes unnoticed—like dancers reacting moment-to-moment to those around them.
“What the market system does, what the price system does, is reconcile all of the conflicting plans that people have... all of those plans are reconciled by the rules, but it works so smoothly that you don't even notice it.”