Gary Shiffman on the Economics of Violence
Jan 18 2021

Economics-of-Violence-200x300.jpg Economist and author Gary Shiffman of Georgetown University talks about his book, The Economics of Violence, with EconTalk host Russ Roberts. Shiffman argues that we should view terrorism, insurgency, and crime as being less about ideology and more about personal expression and entrepreneurship. He argues that approaching these problems as economists gives us better tools for fighting them.

David Skarbek on Prison Gangs and the Social Order of the Underworld
David Skarbek of King's College London and author of The Social Order of the Underworld: How Prison Gangs Govern the American Penal System talks with EconTalk host Russ Roberts about the written and unwritten rules in America's prisons for the...
Barry Weingast on the Violence Trap
Barry Weingast, the Ward C. Krebs Family Professor of Political Science at Stanford University and a Senior Fellow at Stanford's Hoover Institution talks with EconTalk host Russ Roberts about the role of violence and the threat of violence in maintaining...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Charles Hickenlooper
Jan 18 2021 at 4:18pm

Wasn’t Robinson Crusoe depicted on a slave ship that sunk with only he surviving? Was he then in the bartering business of humans? Didn’t he save Friday from being eaten by cannibals? Was the relationship between Friday and Crusoe based on equality or was Friday his servant? Didn’t Crusoe claim the island as his own with he as the King? Wasn’t Friday made to assimilate into Crusoe’s culture? Where was the give and take of equals for fair trade to improve life on the island other than the slave benefiting  from his master’s supposed benevolence and the master benefiting from his slave’s labor and loyalty?

Jan 19 2021 at 9:10am

Inequality, however you define it, does not equate to slavery. I thought slavery was a trade that involved coercion to force one person to do the bidding of another. Webster’s dictionary is more general, describing slavery as a condition where people have property rights over other people. Either way it doesn’t sound like it would apply to Friday. Friday does call Crusoe “Master”, but this may be an honorific as in what you would call a teacher. Crusoe taught Friday English and European culture so he was Friday’s teacher in that regard.

Russ Roberts
Jan 19 2021 at 10:49am

Most economists (perhaps Gary is one of them, though I don’t know) invoke Crusoe and Friday as an example of two people alone on a desert island with only their wits and whatever resources are there on the island already. It’s usually not a literal or detailed reference to Defoe’s novel, which, by the way, I have not read in at least 50 years.

Jan 20 2021 at 10:26pm

I would be very interested to hear how the analysis toward the end of the podcast applies to the rivalry between China and the US.

Andrew Wallen
Jan 20 2021 at 11:09pm

Thank you for the interesting discussion. Just a couple of thoughts.

It does help to have an understanding of the motivation of insurgents / terrorists. It can be complementary to your approach and the questions you use. For example, you ask if they have a strong mission. That relates to motivation for an insurgency. Understanding the motivation (grievance) of an insurgency helps a legitimate government to address or undermine the grievance that motivates the population (market share) to support the insurgency.

DoD is, of course, prominent in counterterrorism and counterinsurgency, but it is not only DoD. It involves the whole DIME (Diplomatic/Information/Military/Economic), which includes many agencies beyond the DoD.

Thanks again. The questions you pose at around the last 10 minutes for how to evaluate a criminal enterprise were particularly interesting.

Jan 21 2021 at 1:51pm

The podcast noted that people joined illegal organisations even though the expected payoff was low compared with legal work, attracted by the possibility of high rewards. I note that these people (like gamblers) are making a Rawlsian choice – to join a group receiving very unequal rewards, before they know whether they will be one of the lucky ones. Perhaps – for these people – the legal world does not have enough inequality.

This also suggests that targeting illegal wealth by requiring an account of the sources of conspicuous consumption might reduce recruitment more than would be expected for Homo Economicus recruits. This would reduce the likelihood of recruits enjoying the rare high payoff which motivates them, and also make it less likely that successful criminals would advertise these returns by flaunting their wealth.

Jan 22 2021 at 8:23am

I was hoping for some violent ETF suggestions toward the end. Is there literature on the financialization of the afterlife? There seems to be a market. Perhaps governments should stop supporting tribes that believe in the afterlife since that is its own reward. The belief seems to cause trouble.

I suspect exchanges are never positive sum. The total energy cost of protecting the exchange is too high so priced in kilowatts the average exchange is massively negative. But that doesn’t become visible until the land around the market place turns to desert. Civilization is a suicide machine.

John Alcorn
Jan 22 2021 at 9:46am

Good science explains much with little. (Think f=ma in the natural sciences.) See Gary Schiffman’s trenchant iron rule of autocracy:

“the iron rule of autocracy is very simple. It’s this message: If you challenge me, you will lose and I will win. I think that’s like–that’s just it. That’s the rule. So, you apply that anywhere in history, anywhere in the world, at any time in history, anywhere in the world, and the successful autocrat–and the autocrat can be the leader of a drug gang in your neighborhood, or it could be the leader of a political party, you know, in a democratic country. But, if you’re the leader, what you want to convey is: If you challenge me, you will lose and I will win.”

Wow, this explains a helluva lot in social life — and rings true, from experience, casual observation, and wide reading about institutions, organizations, and history.

John Alcorn
Jan 22 2021 at 4:14pm


“The Cosa Nostra, the Sicilian mafia in Sicily, operated for 150 years just fine. They did great. And then there were mafia wars in which prosecutors and judges started to get killed, and then public opinion turned against them. [… .] So, politicians–there tends to be a threshold at which violence is a nuisance versus violence is a national priority.”

Gary Schiffman identifies a mechanism — backlash against violence by organized crime — that can, and often does, increase the probability that government policy towards organized crime will change. However, in any particular historical context, concatenation of several mechanisms is crucial.

In order to explain the change in government policy towards Cosa Nostra in Sicily in the 1980s and 1990s, one must identify other, additional mechanisms. For example, I would invoke also a change in international affairs and a change in Sicilian culture.

In international affairs, the key change was the unraveling of the Cold War. From 1945 to the 1980s, Italy’s Communist Party was strong in northern and central regions. The Christian Democratic Party needed strong support in southern regions. The Devil’s Pact — a coalition of the Christian Democratic Party and Cosa Nostra — met the need. The implicit, corrupt political exchange — with Cosa Nostra as the intermediary in Sicily — consisted in patronage (subsidies and job programs) and immunity from prosecution in exchange for large blocks of votes. Cosa Nostra controlled votes because it had substantial community support, through patronage and private protection. The Devil’s Pact remained a lynchpin of politics during the Cold War. Then Detente, Eurocommunism, glasnost, and finally the collapse of Soviet communism made the Devil’s Pact an unnecessary embarrassment in international affairs.

The change in Sicilian culture involved growth in new middle classes, shaped by education, prosperity, and cosmopolitan exposure.

And charismatic individuals may play a crucial role in the complex process. Sicilian prosecutor Giovanni Falcone made a difference, by talent, temperament, resourcefulness, and drive. For example, Falcone cultivated close relations with the FBI in international drug cases. This relationship enabled him to leverage the Witness Protection Program in the USA, before Italy’s Cold-War regime collapsed.

My methodological point is that a single mechanism rarely suffices to explain a major policy change. Usually, one must specify several causes, and explain how each cause increased the probability of major change in policy towards organized crime,

Jan 23 2021 at 7:03am

Gary Shiffman @ 35:00 : “Self esteem. Like, I have, for whatever reason–and this could be the introduction of culture, religion, other things–but I’ve got some sort of moral code that I live by; and I find value in not deviating from that code. That’s a very rational thing. Economists should be able to account for that pretty simply.”
Russ Roberts: “Yeah, they struggled with it, though. I think sometimes it’s hard to measure and hard to put in the equations as part of it.”

I thought FA Hayek did a particularly good job grappling with this difficult problem of decision making with general rules [adhering to a code] vs. decision making by specific people in specific circumstances which are then applied to others [read: a specific rule applied generally to others is NOT a general rule.]

“One places it here, another there, because it depends on the material content of the will, which is dependent on particular facts and is therefore incapable of a general rule. The insight that general rules must prevail for spontaneity to flourish as reaped by Hume and Kant has never been refused, merely neglected or forgotten. Although welfare has no principle, and, hence, cannot generate spontaneous order, resistance to those [general] rules of justice that made the extended order possible and the enunciation of them as anti-moral stem from the belief that welfare must have a principle and from refusal—and here is where anthropomorphism reenters the picture—to accept that the extended order arise from a competitive process in which success decides, not approval of a great mind, a committee, or a God, or conformity with some understood principle of individual merit.”

“This conduct [decision making by general rule] ought to be seen for what it is: The skill to fit oneself into or align oneself with a pattern of whose very existence one may barely be aware and whose ramifications one has scarcely any knowledge. Most people can after all recognize and adapt themselves to several different patterns of conduct without being able to explain or describe them. How one responds to perceived events would thus by no means necessarily be determined by knowledge of the effects of one’s own actions. For we often do not and cannot have such knowledge. If we cannot have it, there is hardly anything rational about the demand that we ought to have it. And indeed we should be the poorer if what we did were guided solely by the limited knowledge that we do have of such effects.”

Jan 23 2021 at 1:41pm

More Hayek on the same subject:

“There is the question of how our knowledge really does arise. Most knowledge, and I confess it took me some time to recognize this, is obtained not from immediate experience or observation but in the continuous process of sifting a learned tradition which requires individual recognition and following of moral traditions that are not justifiable in terms of the canons of traditional theories of rationality. The tradition is the product of a process of selection from among irrational or rather unjustified beliefs which without anyone’s knowing or intending it, assisted the proliferation of those who followed them with no necessary relationship to the reasons, as for example religious reason, for which they were followed. The process of selection that shaped customs and morality could take account of more factual circumstances than individuals could perceive. And in consequence, tradition is in some respects superior to or wiser than human reason. This decisive insight is one that only a very critical rationalist could recognize.”

This quote and the two above are all from chapter 5 of “The Fatal Conceit” (chapter 7 on the Audible Version). And they show, if nothing else, that Russ was right. Even a masterful handling of the subject matter is still remarkably complex (ie. “hard to measure and hard to put in the equations”). And Gary Shiffman is also correct. Economists should be able to account for these differences readily, although I don’t think it is possible to put it simply, as he had hoped, since our brains make the substitutions that cause these errors automatically, without asking our permission, and without telling us that it has done so.

John Alcorn
Jan 23 2021 at 1:48pm


“So, we have to define victory properly. And then […] we have to fight like an entrepreneur. We have to think about the ways in which we would go to market. When we think about going to market in drug dealing, we think about the DEA [Drug Enforcement Administration]. When we think about counter -terrorism, we think about the DOD [Department of Defense]. It seems to me that that’s an overly narrow and focused view of how we’re supposed to be a market participant. So, we, the United States, and/or our friends and allies around the world should think of ourselves as firms in markets; and let’s go to war as a firm would in this context. So, how do we diminish the ability to recruit? How do we encourage defections? How do we deplete resources that they might have in banks?”

I take it that the “we” in question are American taxpayers and voters. If the remedy is to act like an entrepreneur, then shouldn’t “we” legalize and tax drugs? Presumably, pharmaceutical firms would replace illicit drug cartels. Or perhaps “to fight like an entrepreneur” means something else; namely, to treat prohibition as a constraint and to use economic analysis to find clever ways to defeat illicit suppliers, i.e., to suppress the supply side of the market?

Jan 23 2021 at 3:55pm

We might also want to consider interpreting “to fight like an entrepreneur” as a novel idea to compete fairly with the criminals under normal free market conditions. As in, make a better product (religion, drug, social click, you name it) with fewer side effects at a lower cost to “steal” (ha, a good pun) the customers and employees away from the criminal’s enterprise. Gary Shiffman really wasn’t going that route in the podcast, but it would likely work better than considering the criminal as a businessperson and then attempting to undermine his business systematically using violence and threats. That approach is really just competition by proxy (who can threaten the other guy more effectively, who can rob the other guys reserves, who can kill the other groups leaders, who can disable the other’s delivery networks, who can kidnap or jail the other guys employees, etc.) which ignores the underlying demand driving the whole issue. It’s attempting to treat the symptoms, while deliberately ignoring the cause. Direct competition is likely to work better.

Jan 23 2021 at 3:37pm

I just had a staggering thought. This difference between “general rules” and specific rules applied generally to everyone that Hayek wrote about and Russ and Gary touch on, is really a good explanation of the difference between “Rule of Law” and “Rule of Legislation”. Where law is the formation of general rules over time compiled in religion, culture, and common law jurisprudence (among others) and legislation is regulation created by some to manage specific problems and then applied to everyone primarily through government coercion. Laws (general rules) are created through trial and error performed by individuals across different settings and over time and vetted through competition. Whereas legislation (specific rules applied generally to a population) are created through thoughtful deliberation and design by one or a few individuals at a specific time and in a specific place before being systematically applied to more people, more places, and more times and vetted primarily through a tournament process (debate, popular vote, one on one combat, war, height, pedigree, you name it). Importantly, the tournament process used to vet legislation is independent and usually unrelated to the rational or efficacy of the legislation in question. Admittedly there is some overlap between the two types of decisions tools–law and legislation, which is confusing, but remarkably little by my figuring. This is a profound realization, at least for me. It’s Capitalism vs. Socialism; classical-liberal vs. modern-liberal; right vs. left; Libertarian vs. Democrat and Republicans; individual supremacy vs. tribal supremacy; freedom vs. slavery. It’s the whole thing!

Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

Watch this podcast episode on YouTube:

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

A few more EconTalk podcast episodes:

* As an Amazon Associate, Econlib earns from qualifying purchases.

TimePodcast Episode Highlights

Intro. [Recording date: December 22nd, 2020.]

Russ Roberts: Today is December 22nd, 2020, and my guest is author Gary Shiffman of Georgetown University. He is CEO [Chief Executive Officer] of two firms, Giant Oak and Consilient. Those firms use machine learning to help people deal with risk. Our topic for today is his new book, The Economics of Violence: How Behavioral Science Can Transform Our View of Crime, Insurgency, and Terrorism.

I want to let parents know there may be some adult or disturbing themes in today's episode.

I want to encourage listeners to go to for our annual survey of your favorite episodes.

Gary, welcome to EconTalk.

Gary Shiffman: Hi, Russ. It's really great to be with you. Thanks very much.


Russ Roberts: First of all, tell us a little bit about your career and how you came to write this book.

Gary Shiffman: Yeah, it's an unusual background for an economist, perhaps. I started my career in the U.S. military, in the Navy. My first job out of college, I like to say was the Gulf War, if you remember that a long time ago. So, I just traveled around the world, saw conflict, saw coercion, saw violence, came back to the United States after being overseas for my first four years out of college. I worked in the Pentagon. I worked in the U.S. Senate as a national security adviser to the Republican leadership in the Senate. I worked on issues such as terrorism and peace talks and topics like that.

And at that point, I started my Ph.D. work at George Mason University in economics. And so, I came into economics with a mission, and the mission was to make sense of the world. I had started my academic career in my master's degree at Georgetown, where I now teach. And I learned a lot of political science and IR theory and [crosstalk 00:02:34].

Russ Roberts: International Relations.

Gary Shiffman: International Relations. And I was incredibly disappointed. I was young and eager to change the world, and I was so excited to be at an elite institution, and IR theory just didn't answer any questions for me. It didn't click; it didn't make sense. It wasn't the world that I had just come--spent eight years traveling around and seeing. It didn't work.

And coincidentally, I met for coffee with a young economist named Tyler Cowen, and told him that Adam Smith made sense to me and Hans Morgenthau did not. Tyler said, 'Welcome. Welcome to our community.'

And so I pursued a Ph.D. in economics with the idea of understanding conflict and violence and coercion in the world.

So, my career since then has been the combination of economics and real world application in the National Security space. So, how do we make us safer and more secure? May be better stated: How do we make the world more difficult for people who want to do bad things? And really a focus on changing the world and doing things with real operational impact in the world, but really leveraging and leaning on the academic disciplines that are available to us as practitioners.

And for me--my bias is in economics, but I'm open to any sort of science that helps us make sense of the world, and my default and go-to is economics. Sometimes I describe myself as a behavioral scientist, sometimes as an economist. But, it's this idea of understanding these complex problems. And the reason I like economics is because it is a science of human decision-making. And if you think about National Security, it is fundamentally a human problem. Right? We don't have wars but for people--at least up until 2020. So, it's really about understanding human behavior to have an impact in the world.


Russ Roberts: Yeah. I think the key question is, as you know, what is the role of rationality? More importantly: How do we think about rationality? Because that word is a bit of a suitcase word, a word that you can stuff a lot of things into.

Your book has three case studies. We may get to them in more detail, but you look at Pablo Escobar, Joseph Kony, and Osama Bin Laden, who on the surface are a criminal--a drug dealer, a criminal, an insurgent, and a terrorist. Yet you make the point, I think throughout the book, that all three of them are all three, at the same time; and that may be the standard ways we label people and the ways that we describe their motivation by their, some purported motivator--religion, money, etc.--may be missing the fuller picture.

So, our goal today is to--I'm going to let you try to convince us that, that's a better way to look at than the standard way. And occasionally, I'm going to agree with you, that there may be times I disagree.

I want to start with a quote. You say that you have "grown skeptical of many popular theories of non-state organized violence." Non-State organized violence. "For example," again, quoting you, "I question policies based on slogan, such as 'radical Islam' and other national or religious identities. Instead, I've come to appreciate science-based approaches to national security, particularly those based on social science and economics."

What's wrong with thinking about, say , terrorism as a religious issue? Sometimes it's a political issue. It's not motivated just by religion: it can be motivated by ideology, obviously. But that's the standard way people talk about it in the press and in the halls of Congress. Why is that? What's wrong with that?

Gary Shiffman: Russ, thanks for leading with that question. I think that the language matters, and I think that our propensity to label groups of people by some identity marker is lazy and inaccurate. So, I'm going to state that very strongly. As an economist, we might say that, I might restate that to say, we're confusing correlation with causation. So, the world--25% of the world is Muslim, and there are not that many Islamic terrorists. So, the fact that we attribute Islam with terrorism is just plain inaccurate. I mean, it's a grossly over-simplification.

And we tend to do that because it allows us to make categorizations that are easy to understand and digestible--but not necessarily correct. And, I think that's the main goal of this book, is to really address this question, is: Do these identity labels help us to understand the phenomena that we care about in the world, such as terrorism, insurgency, and drug trafficking, or do they mislead us?

And, I'm arguing in this book that they're mostly misleading, and that there is a better way. And, the better way is to think about markets, to think about humans and institutions and markets and firms and prices and entrepreneurs.

And if we pivot from, 'That's a drug dealer. That's an insurgent. That's a Muslim. That's a Mexican'--if we get rid of those terms, and we say, 'That's a firm, that's an entrepreneur,' we can simplify the story, we can make sense of it, and I think it's more accurate.


Russ Roberts: Okay, we're going to get to that. I want to take one more step back though, and I'd love this--one of my favorite parts of the book, not directly related to violence, but you talk about Robinson Crusoe, who is alone on an island. You say:

What happens to Crusoe if another person, who we will call Friday, appears on the island? We know that humans seek to barter and trade with other humans to better their situation. If we placed a wild boar on the island, Crusoe would likely hunt the boar and roast it for dinner, but it would be absurd to posit that Crusoe would hunt Friday and roast him for dinner. Why is that?

I've never heard that example. It's a beautiful--I have a different take, a different take--a different version of that, but not as dramatic.

But, it's a fantastic point, that the arrival of Friday doesn't make Crusoe poorer--for starters--and that eating him would be a mistake, a tactical error. Why?

Gary Shiffman: In the fact that he just intuitively knows that to be the case. This is this idea--and I know you're an Adam Smith aficionado as well--but we have this fundamental human nature, which is described or may be summarized by the propensity to truck, barter, and exchange one thing for another. So, no matter where I am in the world today, as I travel around--literally, or just reading the paper in the morning, or if I look back through any point in history--if I can go back to this idea that every character that I'm seeing in this plot that I'm witnessing has the propensity to truck, barter, and exchange, then that's a scientific principle from which I can launch into my analysis, and say, 'Well, why are people killing each other? Why is this person coercing these other people?' And, just go back to this fundamental idea that there's a propensity to truck, barter, and exchange.

I like the Robinson Crusoe story because it's a very simple way of making the point that, as a reader, and you read that passage that you just read out loud, you know it's true. You know that Robinson Crusoe didn't think, 'Hey, there's like two days worth of food.' Robinson Crusoe knew that this is an opportunity to truck, barter, and exchange one thing for another [crosstalk 00:11:41].

Russ Roberts: Or cooperate. It doesn't have to be--obviously in this case--it doesn't have to be literal 'truck, barter, and exchange.' If there were 20 people, it would be maybe, but--

Gary Shiffman: Right. With two people, as you said, there's this assumption that they could be better off. So, there's absolutely just the benefit of two people creating more wealth than just one person through specialization and division of labor. Right?

But, then, take that to the furthest extreme, which economists like to put up goalposts and go to--the furthest extreme is: an individual walks into a crowded market in Jerusalem with a backpack full of explosives and detonates it and kills himself and a hundred innocent people around him.

So, you've got this fundamental human nature and the propensity to truck, barter, and exchange and be better off, and you've got suicide terrorism. It's like: How do we reconcile all of this, if these are extremes of human behavior?

And I think economics gives us the tools to think about and address those questions. I don't think that anything is easy, but I think that's a much better approach to understanding violence than is saying, 'Well, that was a Muslim that blew himself up. We know things about Muslims.' That's just not the right way to address things.

Russ Roberts: But, the standard--I would say the standard way that economists often think about this problem: first, of course, there is an economics approach to religion. Larry Iannaccone--who I know you know, you reference him in the book, who has been a guest on the program--talks about that. We may get to that.

But the more narrow way to think about economics in that setting is: Okay, you've got a group of people who are truck, bartering, and exchanging--shopping--in the market peacefully, respecting each other's property rights enforced by a set of police and government, and also norms of how to interact. Then you have a terrorist, a murderer who disrupts that and kills a bunch of people.

A lot of people would say: Well, the economist's approach to the terrorists could be, 'Oh, well, that person didn't have many prospects in life. They didn't have much to live for. If they had a better labor market where they were--if there had been more opportunity, if Israel could trade, which they are now starting to do more of with their neighbors, there's a better chance that those kinds of things won't happen.' And I think that's true. I think that's a good way to think about it.

But I don't know what it--you can go to one extreme and say, 'The terrorist is--it's just an economics standard-of-living problem: you've just got to give them better choices.' Or you could say, 'Oh, they've got this destructive ideology that they believe in and that harms us.' It seems to me it's both. So, why do you want to focus on one over the other?

Gary Shiffman: I don't see it as one over the other. I think that, as economists, we can acknowledge that religion provides institutional rules and constraints on human behavior. So, being Jewish, being Muslim, being Christian, being whatever is going to constrain behavior. And it's okay to acknowledge that.

I think where we go wrong is when we oversimplify the world to say: all Jews are this, all Muslims are this, all Christians are this, all Mexicans are this, all Colombians are this, all Americans are this. And I think we do that. I think we unintentionally go too far with these identity labels.

And I think it's appropriate that we pull back and say, well, maybe we've gone too far with identity labels and think about the role of religion. As I say in the book, it's not Islam as Islam that matters. It's Islam as an institution which shapes human behavior that does matter. And if we think about it that way, we can bring religion back into this discussion. And I love Larry Iannaccone's work and love to incorporate his ideas into this.

So, I'm not saying religion doesn't matter. I'm saying it doesn't matter in the way people think it matters.


Russ Roberts: The part that I love that you emphasize is the kinship part, and the argument that a lot of what we're calling terrorists, or insurgents, or criminals are providing something else that is masked by the things that are costly to us as the non-participants of the terror, the death, the activities. And that the opportunity to affiliate with these groups gives participants actually a way to truck, barter, and exchange more effectively because you can trust their insiders. There's a role for loyalty there.

So that, I thought, was the most provocative idea in the book. Why don't you expand on that?

Gary Shiffman: Sure. So, I think we have this tendency to divide the world into us versus them. And, we do this regularly, whether we're aware of it or not. And, so, this idea of kinship is actually somewhat pervasive in economic literature. Hirshleifer addresses it. There's kinships and fictive kinships, and--

Russ Roberts: Explain what you mean by that, because that's important. What is--so, kinship is: You're my cousin. But, what's a fictive kinship? What do you mean by that?

Gary Shiffman: A fictive kinship, it's not an actual blood kinship or a blood relationship, but we have the psychological construct of some sort a familial tie.

So, we see this in the crime world. We see this in mafias, for example. So, if you study the literature on mafias, or if you just watch--The Godfather, or another movie, or read a novel--the language is all about family. There's a crime family. There are the five crime families of New York. So, it's all family; it's all brother. Your loyalty to your brother in the crime family actually supersedes your loyalty to your blood brother. This is the Cosa Nostra story.

So, we're somewhat--as humans, we're really susceptible to this, to this idea of fictive kinships. And we do it all of the time.

There's literature about: you know, the military creates these fictive kinship. When you go to war and you go to battle--and again, I started my career in the military in a war--you develop very, very strong bonds with these folks you go into battle with that really rival any other blood relationships that you have. So, I think everybody can think about these kind of kinships. It can be as simple as: This weekend I'm rooting for the Washington football team, Russ, and you might be rooting for the Baltimore Ravens. I don't know--

Russ Roberts: Never. Never. Go ahead--

Gary Shiffman: Sorry. I didn't mean to offend.

But, we develop kinships around things like sports teams that we root for. So, these, I think, are really important traits, human traits, that are probably correlated with the long-term survival of our species, but they're present today. And so: We do this, we can assume others do this. And it allows us to build trust. Trust allows us to engage in trade. Right? It's hard to imagine trade in the absence of trust. I think markets require trust.

Trust comes about through lots of different ways, rules, and enforcement, and contract law in courts as one way, but kinship or fictive kinship is another way in which we build trust.

And if you're in the illicit economy--Russ, if you and I were drug dealers and I wanted you to manage my books, if you stole from me, I couldn't sue you. I can't bring you to court--because we're drug dealers.

So, how do I overcome trust, the trust hurdle? Well, if you were a cousin of mine, I might actually let you manage my books because you're less likely to defect on me.

And if you're not my real cousin, I am going to get to know you and your family, and have you over to my house to meet my family, and I'm going to create a fictive kinship with you because that's a way to build trust and to deter defection.

Russ Roberts: And, if I betray that trust, you're going to kill me and maybe my family members.

Gary Shiffman: That's true.

Russ Roberts: Yeah: 'Sorry. It was a nice dinner. I enjoyed it.'


Russ Roberts: I'm going to step back now and try to--this is the way that I see your story, and I'll let you correct me or add to it. We have certain illegal activities that there's still a big demand for--drugs being the most obvious example. People who provide those, take risks, but they can make a lot of money.

To create the networks that would a firm would have to have in a legal economy is going to require substitutes for the ways that people would normally interact. The threat of violence, as you just mentioned, would be one of those ways.

And so, these organizations that are violent, doing illegal things, and sometimes offering an ideology or a salvation theme alongside all that, it's all the same package. It's just a way that an entrepreneur at the top of this organization can have great success.

And you paint both Pablo Escobar and Osama Bin Laden as entrepreneurs, and Joseph Kony, as entrepreneurs who aimed high. The only problem I have with the story is that they both die terrible deaths of, and fairly young, not totally young. So, the challenge I think of making this a so-called rational enterprise is it usually ends in a hail of gunfire.

You could argue that it's worth it: along the way, they get enough benefits. But, these are people who have chosen a very dark niche to express themselves. Your point, I think, which is profound is that--but they are expressing themselves. It's a form of self-actualization. It might be one that you and I would find unappealing, but for them, it's not much different than--I won't name anybody to compare them to Osama bin Laden, or Pablo Escobar. But, think of your favorite CEO of a legitimate firm who has been successful. These people become rich, they become famous, and they have a lot of love from their adherents. Is that a good summary of what you're selling here?

Gary Shiffman: Yeah, absolutely. To the extent that I'm selling a story, I guess: yes.

Russ Roberts: Of course you are.

Gary Shiffman: That's a summary. It's a very good summary.

The myth of Pablo Escobar is that when he was--

Russ Roberts: Tell us a little bit about him, because many people may not have heard of him, know who he is. So, give a little background.

Gary Shiffman: So, Pablo Escobar was the head of the Medellin cartel and he is largely considered the first narco-terrorist. The word 'narco-terrorists' was really first used in the popular vernacular describing Pablo Escobar, Colombian. There's this story abou t Pablo Escobar getting his photo taken in front of the White House here on Pennsylvania Avenue saying that he's going to negotiate with the President of the United States one day. So, I don't know his heart. I don't know what he really wanted, but at least, that's part of the mythology of Pablo, which he had very grand political ambitions.

He started out as a street-level thug. He stole cars and sold the goods of his proceeds. Then he figured out that he could make more money by protecting other people's cars. So, the classic protection racket, which is: you start out doing some criminal activity, and then you realize you can make more money by protecting people from yourself--from that very same criminal activity. This is Pablo Escobar.

So, he runs a protection racket. And then he stumbles into smuggling things across borders, which is another form of protection: 'Plata o plomo,' which is lead or silver. Basically, if you are a Colombian police officer or an enforcement official of some sort, and you run into Pablo, you have a choice. He's going to give you silver--he's going to give you money--and if you don't take it, he's going to shoot you--

Russ Roberts: The lead.

Gary Shiffman: So, he became popular for silver or lead, when you see him.

And then he just happened to be at the right place at the right time, which is the United States discovers its love for cocaine, and cocaine production moves into Columbia. So, this is a guy who's got well-established smuggling routes. He's got massive numbers of police, military, law enforcement, and judges in Columbia on his payroll anyways. And now he can make orders of magnitude more money by smuggling and by paying off government officials, because he's now smuggling cocaine into Miami. And that's where he really becomes wealthy and famous.

Inside of--and I think it's important--in Medellin, which is his hometown, he actually does become--he's the FBI's most wanted criminal and he walks in broad daylight through the streets of Medellin. He's not in hiding, because the people, the whole town of Medellin loves and supports him. He builds roads, schools, hospitals. He provides that social safety net that the government is unable to do because he's actually got more money than the government.

So, this is Pablo Escobar. Yes, he dies of with a bullet to the head by law enforcement agents, eventually.

Russ Roberts: Took a long time.

Gary Shiffman: But, it took a long time. Look at today, we're still talking about him. I mean, I don't know Russ, if he didn't win the game in a sense, right? Did he really not get what he wanted?

Russ Roberts: Well, my favorite part of your book, one of my favorite parts is where he's in prison living like a king, not much different from some of the stories that David Skarbek tells. You reference David's work, who was on the program, of how drug dealers in American prisons extend their reach back into their home neighborhoods through confederates, friends, loyal lieutenants, and so on, demand things that people when they're released from prison--which are enforced through the threat of violence, which is really an important part of this.


Russ Roberts: I want to hear, though, why he builds these schools, and of course, terrorist organizations. Again, I think one of the things that's useful about your framework is that: it should be blurred, the line between terrorists, social providers, and social service providers. Because, they often do all these--they also all sell illicit things. They raise money through illicit things, illegal things. They provide services to nearby citizens. And, they also carry on some ideological political goals.

It's kind of a--another way to think about is sort of a shadow government, in a certain dimension, in these places where formal governments are weak.

But, what's the gain to an Escobar of those schools? Now, he gets the adulation of the people around him, but I guess he also gets the protection. Is that part of the motive there?

Gary Shiffman: Yeah, sure.

Russ Roberts: The loyalty?

Gary Shiffman: So, again: just think about Escobar and Osama Bin Laden as entrepreneurs. They need a safe haven from which to operate, and so they provide social services and social benefits, which increases loyalty and deters defection. It's that simple.

Eli Berman--whose work I'm sure you know--he writes about this, where he looked at the terrorist groups in committing acts of terrorism in Israel, and the groups--Hamas, at the top of the list--engages in more deadly acts of violence and they also provide more social services.

And so, the point he makes is that, it's by providing social services, you create the club. So, you have a club goods model, is what Eli does with it. And so, you encourage loyalty; you deter defection; and your ability to engage in high-value illicit activities is as a function of your ability to deter defection. If you cannot deter defection, you cannot engage in high-value criminal activity.

And so, Pablo Escobar gets a safe haven. He gets the entire city to operate from, by providing public services and public goods to all of the people.

So, it might feed his ego. Again, I don't know his heart and soul. It might make him feel good to walk down the streets and be hailed as a hero.

But it also allows him the freedom to operate a global criminal enterprise with a nice quality of life, in which he's not really hiding or on the the run. It actually seems like a pretty rational and smart thing to do.


Russ Roberts: So, violence plays an unusual role in these stories that I think goes beyond--it's a special kind of incentive effect that I think maybe is underappreciated. I always talk about how, you know, survey results in places where violence is casual and normal or uninformative, those survey results--because people are afraid to tell the truth.

Gary Shiffman: Right.

Russ Roberts: You tell the story--a horrifying story--where Escobar bound, tied up a servant of his, threw him into a swimming pool, and watched him drown in front of his guests, so he could announce, 'This is what happens to those who steal from Pablo Escobar.'

I'm reminded of the group, I can't remember now who it was, when the British were in India. They, somebody--I can't remember who it was now--they found three men sleeping, they would behead two of them and leave the third one to wake up to spread the word, because that person is more effective than three dead people, a person who's survived that then tells folks.

And, it reminded me--strangely because I'm strange--but it reminded me of Solzhenitsyn and Vasily Grossman, who write about how a person in a situation of violence and corruption and degradation, who chooses to stay undegraded, refuses to advance their own self-interest because of what they understand it will do to their sense of self.

And both men honored the people who, in the Gulag, refused to advance themselves at the expense of others.

And, economics says you should. Economics, often--I think bad economics--but economics would say, 'Well, the rational thing is to be violent. The rational thing is to raise the cost. The rational thing is to, in a zero sum game, make sure you get ahead, not behind.'

And I think there's a question of principles, morals, self--I don't know what you want to call it; economics doesn't have that much to say about it. But it seems to me, once violence is on the table, the normal ways that we think about our self-interest and incentives get ratcheted up so high that they kind of overwhelm everything else.

It's like saying to someone, 'Oh, don't worry about betraying Pablo Escobar. We'll pay you enough money to make it worth your while.' Well, there probably isn't enough money for most people because they'll do some things to you that aren't worth any amount of money that you could imagine. Maybe.

Gary Shiffman: Yeah. So, I think two ideas in what you just said that I want to respond to. The first is: I think the iron rule of autocracy is very simple. It's this message: If you challenge me, you will lose and I will win. I think that's like--that's just it. That's the rule. So, you apply that anywhere in history, anywhere in the world, at any time in history, anywhere in the world, and the successful autocrat--and the autocrat can be the leader of a drug gang in your neighborhood, or it could be the leader of a political party, you know, in a democratic country. But, if you're the leader, what you want to convey is: If you challenge me, you will lose and I will win.

So, how do you go about doing that? Well, if you're Pablo Escobar, if you're any sort of criminal, you need to--back to the idea of, 'If you're running the books for our drug gang, I'm going to have you and your family over for dinner.' But, I need to let you know that if you challenge me, you're going to lose it and I'm going to win.

And I think that's a universal rule. You read any novel of fiction, any real life story, and you're going to see a variation on that theme play out. So, that's my first response.

The second is that I think, when economists talk about value, we have to define the word more broadly than something that can be reduced to a dollar value, a monetary value.

So, in the story that you tell--and I'm not familiar with that literature, I've heard you've talked about it a couple of times now, though--is self-esteem. Like, I have, for whatever reason--and this could be the introduction of culture, religion, other things--but I've got some sort of moral code that I live by; and I find value in not deviating from that code. That's a very rational thing. Economists should be able to account for that pretty simply.

Russ Roberts: Yeah, they struggled with it, though. I think sometimes it's hard to measure and hard to put in the equations as part of it.


Russ Roberts: When we mentioned fiction earlier, The Wire is one of my favorite shows, and--at least I saw the first couple of seasons, two or three, I can't remember now. But, the first season opens, is about the drug war, and the Baltimore police at war with the drug dealers on the streets of Baltimore. And the reason I think it's a powerful film, powerful series, and I think David Simon, who created it, has some deep insights, is that: there's kind of a dance between both sides that we are often ignored or peril. Which is that: the police do want to catch the drug dealers, but they also profit from the drug dealers being out there as a source of reason for them to exist, have big budgets, and so on.

And, I wonder how much people really want to catch Pablo Escobar, especially in Columbia. First of all, he killed tons of judges and people who would normally be against him. Nominally, governments are supposed to stop drug dealers. It's not really my--I wonder whether they should. It doesn't seem to make much difference. All it does is--I know we spend a lot of money chasing them around and it doesn't seem to have a big impact on the amount of drugs. Maybe it reduces it to what it would otherwise be: it's possible.

But, the costs are very, very high.

When I think about a less organized government situation, say, in a poorer country than the United States, I mean, how badly do those folks really want to--I mean, is Pablo Escobar really an enemy for them? I'm not sure he's the enemy in the United States either. But we have these narratives--also a different form of fiction--we have these narratives about, that these people are our enemy. Pablo Escobar is providing a service. It's a destructive service cocaine, in a certain way, but it's what people wanted. And his government maybe pretended to try to stop him. I don't know. It could be part of the--and when they tried to stop him, he killed them.

The combination of ineffective governance there creates a very strange dynamic between the two sides, it seems to me.

Gary Shiffman: So, if you think--we have an agency problem in the sense that some politicians have different goals than some police officers might have, and so we have to disaggregate the decision-making process and the preferences and the optimization functions of the different individuals and players involved.

But, in general, what you see, if you look across crime insurgency, terrorism, across countries and cultures and religions around the world, and throughout history, you see that there tends to be a threshold below which politicians don't care and above which the politicians get involved.

So, when Pablo Escobar was smuggling drugs into the United States, he wasn't really a priority inside of Columbia until a prosecutor decides that he thinks it's starting to have a corrupting influence. And so then Pablo kills the prosecutor. And then Pablo starts to kill other politicians, and bombs start blowing up in town. And so, now it becomes a political priority to actually go after Pablo.

But, before that, when he was just selling drugs into the United States, they kind of left him alone. And back to the point you raised earlier, where eventually, they built a prison for the Medellin cartel. And they said, 'You all move into the prison and stop blowing things up, and then we're going to leave you alone.'

And so they actually ran the Medellin cartel from inside of the prison. The barbed wire faced out. Meaning, they had a safe haven in which they didn't want anybody to come in.

The politicians got what they wanted, which was a decrease in violence. They didn't care about the drugs leaving Columbia going to the United States. They just didn't want Columbian innocent people getting killed.

That's a common theme. The Cosa Nostra, the Sicilian mafia in Sicily, operated for 150 years just fine. They did great.

And then there were mafia wars in which prosecutors and judges started to get killed, and then public opinion turned against them.

So, it didn't matter, at that point, how many hospitals, roads and bridges you built as a drug dealer. Once you're killing judges and politicians, public opinion can turn against you.

So, politicians--there tends to be a threshold at which violence is a nuisance versus violence is a national priority.

Mexican elections have gone back and forth over the last 20 years based upon: If we crack down on the drug dealers, that's good, but violence increases. If we don't crack down on the drug dealers, they make more money, but violence drops. Right? And so--

Russ Roberts: And then they face the external pressure from the United States, say, to do something about it. That also plays a role.

Gary Shiffman: Right. Which is, today's news and headlines is that. In The Wire,--and I love The Wire. I'm on the bandwagon of best TV show ever. I'm in that group.

Russ Roberts: Top five for me.

Gary Shiffman: Yeah. What I love about it is, is it does exactly what I'm trying to do in my book, which is it blurs the lines.

Russ Roberts: Yeah.

Gary Shiffman: The drug dealers are bad guys, and the cops are good guys. Well, it doesn't come out that way in The Wire. It turns out that life is more complicated and humans are more complicated. And if we take the firms, entrepreneurs and markets discussion into The Wire, it actually makes sense. So, it's not: let's not label people as cops are good, drug dealers are bad. Let's look at firms, institutions, markets, and incentives, and now you can really understand why people are behaving the way they're behaving. And that's a much more clear and enlightening way to look at the world, I believe.


Russ Roberts: So, I agree with that. Up to a point. Obviously I agree with it because I think incentives matter, and I think markets produce outcomes that no individual intends, and that are just the reality that we need to think about. You can't just change that. The idea that you could make something illegal and therefore solve it is really important.

Actually, that's so obviously true to an economist, and so obviously not true to most people, that maybe we should spend a little bit of time on it. Because I think most people would think, 'Well, if something's bad, we just make it against the law,' without thinking about what, or make something--subsidize something that we think is good, as if that solves a problem. It often leads to a set of more complicated things.

But when we think about The Wire, I think your approach of thinking, you know, 'The drug dealers, they're running a business.' We don't like to think of it, maybe, that way. You and I do, but most people don't. That's kind of weird: 'What do you mean they run a business? They have employees, they have labor issues, they have price issues, they have competition.'

And of course, all of that, because the courts cannot be used, they have to find alternative ways to deal with defection, disloyalty, unpleasant outcomes that didn't happen.

And so as a result, you get violence as a way to deal with that.

We didn't mention, you didn't mention, but your book does: At one point, Pablo Escobar shoots down, takes down an airliner, a, quote, "terrorist act." He's a drug dealer. He's got political ambition, which we didn't talk about.

So, what is he? Is he an insurgent, a terrorist, or drug dealer? He's all three.

Your point, which is, I think profound is that, it's not obvious that it's useful to describe him as, say, a criminal. He is a criminal, but if we think of him as only a criminal, we don't get the full story.

The challenge I have for you is that, when we think about ways to reduce these things--if that's an ideal--it's not obvious how to leverage this under richer understanding.

So, I agree with you, certainly as an economist, that understanding the incentives, institutions, norms, values that folks have, choices that people face, helps us understand more accurately what's going on.

But, you also make the claim that it helps us fight it--if we want to fight it. And, I wasn't convinced by that. So, I want to let you make the case now. You argue that this identity category idea of--you argue that this is unproductive. Why is it more productive--I agree, it's more accurate to realize that there are all kinds of issues involved: ego, alternatives, ambition, and that often these outcomes are a result of those more interesting forces and just, say, an ideological predilection. How does it help us do something about it, if we want to?

Gary Shiffman: Wow. So, there--you can't take action and expect a desired outcome if you don't understand what's causing the situation to begin with.

So, it seems obvious to me that we have to understand, or at least attempt to understand, causation if we want to try to bring about a different outcome. And I think the tools of economics really help us to do that.

There's a large body of literature on radicalization and counter-radicalization, and some of my friends are involved in that. There are large departments of the U.S. government working on it.

And, I'm not sure it's a good idea. I'm not sure it's a good use of funds. Because--just the whole idea, just think about the word 'radical.' I mean, 'radical' is a subjective word. It's not an objective word. I mean, I've got--you know, who doesn't have a radical uncle or something like that? I mean, it could mean so many different things. So, if we want to do counter-radicalization, it seems--I don't know, it just seems too squishy for me to really get my head around as a scientist, as an economist.

But, things that do make sense are the things we've been talking about, which is: How do I think about the drug gang on the streets of Baltimore? As a business, I can make perfect sense of that.

I often ask my students--to their, you know, great frustration--if we need to care about motivation. If I see somebody running a drug gang in Baltimore or running a drug cartel in Columbia, or running a terrorist organization in the United States, or anywhere in the world, and I look at that leader, I look at that entrepreneur, do I need to know why they're doing it?

It might help to know why, but I don't know that I need to know why. And I think that's a good question for us to explore. When we try to understand why we're getting into things like surveys and survey results and things like that, and people are misleading intentionally often, if I know however that they need to generate revenues and minimize costs, and as long as revenues exceed costs and they can make payroll and they can deter defection, that they're going to stay in business, I don't need to know why they're selling drugs or they're blowing up airplanes. I just need--I know how I can help them to fail as a business. And it seems to me that's what I need to know. If I want to put a drug dealer out of business, I don't need to know what's in his heart, or hers. I need to know how to get them to fail as a business.


Russ Roberts: But the way we've done that historically in the United States, we've tried to--I would argue, we've used economics. We've tried--a really primitive form of it--and I think yours is a richer approach; but I'm going to let you have at it. We've tried to raise the cost. We say, 'Look, you sell drugs on the street, you could go to jail, you could get killed.' We've tried to make it more expensive and less attractive. And I would say that has utterly failed. Right? So, what's an alternative approach?

Another approach would be try to create [?] more jobs in the neighborhood to make it harder for those folks to attract lieutenants and privates and people in their army. You can try that approach. What else you got?

And, by the way, the flip side of that, in the terrorist story is: in places with bad governance, you want more governance if you can find a way to create it there, which is hard to do, because they are an alternative to that. And then, ideally, you want more trade to raise the cost to me of destructive non-trade, non-cooperative activity. Does economics have anything more to say than that?

Gary Shiffman: So, just--I'm going to answer that with the three points that I make in the book. But before I do, in The Wire there's this wonderful character named Stringer Bell. Stringer Bell is played by a young Idris Elba, who has gone on to become one of the most famous, you know, and wonderful actors in the world. But, Idris Elba is Stringer Bell who actually, he's the deputy for the particular drugging, the Barksdale gang in Baltimore in this fictional account. And he goes to the local community college and he takes classes in economics to help him run his drug organization better. Which is just a beautiful kind of cap on this idea that it's a business. They're running a business and Stringer Bell is asking questions about, you know, elasticity of demand, of the professor.

Russ Roberts: I remember that scene. I think it was slightly fictional in the sense I don't think taking economics helps you really that much become a better drug dealer, but I got the idea. Yeah.

Gary Shiffman: Right. Well, he's trying to figure out how to price his product in a competitive market. Which is what entrepreneurs do.

So, I suggest--my idea here is: Let's understand a better way to look at phenomena in the world. Let's understand crime, insurgency, and terrorism better and not label it as crime, insurgency, and terrorism, but let's blur the line as you described that we do with Pablo Escobar. So, but, what can we really do about it?

So, I have three ideas.

First is: Let's analyze like business executives. So, if I'm looking at Al-Qaeda or the Islamic state or the Sinaloa Cartel, I'm going to treat them all the same, first of all. I'm going to look at them--and I want to know, if I were a venture capitalist that was thinking about investing in them--would they be a good investment? Would that firm be worthy of my investment dollars?

It's strange to think about the Islamic state and the Sinaloa Cartel as selling shares and raising money. But, I think we should think about it as investors. Do we believe in the business model? Tell me about the leadership structure. Do I think the CEO and the CFO [Chief Financial Officer] are good, solid people? Do they have a strong mission? Do they have product-market fit? Are they selling a product that has elasticity of demand or not? What are their competitors like? Is it a deep market? Do they have massive market share or only a portion of the market share? What does their balance sheet look like? Do they have lots of money in the bank to get them through hard times? Or, are they just getting by month to month? How do they recruit? How do they retain? What's their defection constraint? What's their turnover rate? What percentage of their employee base turns over every year?

These are really standard business questions that any investor would ask of any potential investment. We should be asking these questions of these firms that we're trying to defeat. These are good, solid questions.

Second is, is we want to define victory in market terms. Over the last 15 years, we've been fighting wars in Iraq and Afghanistan. I've been very close in some years, more close in some years than others to this. And, I think we got victory wrong. Victory is not signing a peace treaty and walking away. I don't know that victory ever was that, but it's definitely not that now.

Killing the leader--that doesn't really bring about victory. In market terms, if you're Coke versus Pepsi, or IBM [International Business Machines] versus Cisco or whatever, you don't think about taking out the leadership. You'd think about taking market share.

So, once we've done the analysis of these firms, we have to think about us as a rival firm to them, or our allies as rival firms to them; and we want to take market share from them. Taking market share could mean kill the leader, but it doesn't always mean kill the leader. Sometimes that is productive, sometimes that's not productive in the context of violence and warfare. We have to think about: How do we win in the marketplace in which this firm operates--in which the Sinaloa Cartel and the Islamic State operate? How do we defeat them as a rival firm?

So, we have to define victory properly.

And then the third thing is: we have to fight like an entrepreneur. We have to think about the ways in which we would go to market. When we think about going to market in drug dealing, we think about the DEA [Drug Enforcement Administration]. When we think about counter -terrorism, we think about the DOD [Department of Defense]. It seems to me that that's an overly narrow and focused view of how we're supposed to be a market participant. So, we, the United States, and/or our friends and allies around the world should think of ourselves as firms in markets; and let's go to war as a firm would in this context. So, how do we diminish the ability to recruit? How do we encourage defections? How do we deplete resources that they might have in banks?

I think we do a lot of these th ings piecemeal and case-by-case, but I think we should think more wholistically as a firm competing against a firm in a competitive marketplace.

And that's really, fundamentally, Russ, what I'm arguing is: if we think about terrorism as a function of Islam, and drug dealing as a function of Mexican and Columbia whatever, I think we're missing out on the fact that markets, firms, entrepreneurs are really the key ideas for making a difference in the world.


Russ Roberts: Let's talk for a second about taking out leadership. Sometimes, in the case of, say, a Mafia war, the person who does the killing wants the other side to know who did it. There's a tribal kinship thing for their own side. They avenge something sometimes.

But sometimes you don't want the people to know who did it, and no one takes credit for some of these deaths. One of the things your book taught me is that the virtue of that is that it widens the range of possible enemies you might have that you're not aware of, which spreads you thin. That's a way that you make a firm less effective.

The part, I think, that is effective, that I wonder if you underplay though, is the--talent is scarce. If you take out a leader, it's true that leader will be replaced, but presumably, that leader will not be as talented. And there's also--the competition for that role is diminished when you know that you can be killed.

So, it's a way of turning the violence against that group that they've been imposing on other people. Do you think that's a legitimate way to think about it?

Gary Shiffman: I think that's a simplistic way to think about it. I think that's true sometimes, but I think it's worthwhile to take a step back and ask if that's the case all of the time.

Two examples. First, Steve Levitt and--so, in Levitt and Venkatesh, they did a paper on the drug wars in Chicago, where they actually got the books and the payroll. And they asked this question about: if gang wars are bad for revenue--decrease revenue, increase costs--why do we see violence across gangs?

And, one of their conclusions of their, I think, very interesting paper is that it's a competition. It's a lottery. It's like, if you're street level you could make more money working at Home Depot than street level in a drug gang; but if you go get the minimum wage job, you'll never make a lot of money. But, if you enter the drug gang, and if enough people above you get killed and if you survive, you might actually make a lot of money one day. Right? So, it's a tournament. It's a tournament, is that idea.

Second example is: We worked for years--we, the United States, worked for years--to help the Colombians get rid of Pablo Escobar. And what happened the day we got rid of Pablo Escobar is the Cali Cartel stepped in. We didn't change the market, Russ. We just killed an entrepreneur. And, the Cali Cartel was probably very supportive of our efforts to defeat the head of the Medellin Cartel, because they just took over the market.

So, I think, anecdotally, you could come up with cases in which getting rid of the leader made a difference, and you could come up with instances in which getting rid of the leader doesn't make a difference. Which is why I argue you have to think like a business person and fight like an entrepreneur. And you have to ask yourself: that rival firm, that CEO is super-valuable, and if we get rid of him, it's likely to harm the operations of the firm. Or, that CEO is surrounded by a talented bench, and that CEO i s really not material to our ability to defeat the firm. And I think we have to ask those kinds of questions.

Russ Roberts: My guest today has been Gary Shiffman. His book is The Economics of Violence. Gary, thanks for being part of EconTalk.

Gary Shiffman: Russ, loved it. Thanks very much.