Worse Than War with Anke Hoeffler

Hosts:
Marius R. Busemeyer – Professor of Comparative Political Economy at the University of Konstanz and Speaker of the Cluster of Excellence “The Politics of Inequality”.
Gabriele Spilker – Professor of International Politics and Global Inequality at the University of Konstanz and Co-Speaker of the Cluster.

Guest:
Anke Hoeffler – Professor of Development Research at the University of Konstanz and member of the Cluster of Excellence “The Politics of Inequality”. She is also a member of the German National Academy of Sciences Leopoldina and was awarded an Alexander von Humboldt Professorship in 2019. Her research focuses on economic development, conflict, and violence.

Episode Overview
What can be worse than war? In this episode, Marius R. Busemeyer and Gabriele Spilker speak with Anke Hoeffler about her new book “Worse Than War”, co-authored with James D. Fearon. Hoeffler argues that the violence most people experience does not take place on battlefields, but often at home: intimate partner violence, violence against children, homicide, assault, and suicide account for a major share of the global burden of violence. The conversation explores how such violence can be measured, why its costs are so unequally distributed, and which policies can help prevent it.

Episode Highlights

The “War at Home”

· While war dominates the news, many women and children experience violence in everyday life and behind closed doors.

· The most common form of violence against women globally is intimate partner violence. For children, it is physical punishment by parents.

· War can increase violence in society, but much interpersonal violence occurs independently of armed conflict.

Measuring the Costs of Violence

· Violence at home is difficult to measure because victims often do not report it to the police.

· Victimisation surveys ask people directly about specific acts of violence.

· Hoeffler distinguishes between tangible costs, such as healthcare and productivity losses, and intangible costs, such as suffering and welfare loss.

· The global cost of measurable violence is estimated at around 34 trillion US dollars per year — about 4,000 dollars per person.

What Can Be Done?

· Hoeffler argues that violence is not simply a matter of culture and can change through law, norms, and policy.

· Reducing access to weapons, especially firearms, can lower harm from homicide and suicide.

· Policies that reduce alcohol abuse can also reduce violence, particularly intimate partner violence and public assaults.

· External actors cannot impose change, but they can support local NGOs, social movements, and reform champions.

Links & Further Reading

More about the Cluster of Excellence “The Politics of Inequality”: www.exc.uni-konstanz.de/inequality

Further readings:

o   Hoeffler, A. and Fearon, J. D. (2026): Worse Than War. The Global Costs of Violence. Princeton University Press.

o   Hoeffler, A. (2018). Security and development: Shifting the focus to interpersonal violence. The Economics of Peace and Security Journal, 13(1). 

o   Hoeffler, A. (2017). What are the costs of violence? Politics, Philosophy & Economics ; pp. 422-445.

Contact: cluster.inequality@uni-konstanz.de

New episodes every first Wednesday of the month – subscribe and stay tuned!

Transcript

In_equality Podcast, July 2026 - Anke Hoeffler

Gabi Spilker

Welcome to a new episode of the In_equality Podcast. My name is Gabi Spilker, and I'm a professor of international politics and global inequality at the University of Konstanz. Next to me sits …

Marius Busemeyer

Marius Busemeyer. I'm a professor of comparative political economy and speaker of the Cluster of Excellence “The Politics of Inequality”.

Gabi Spilker

Our guest today is Anke Hoeffler. Anke is a professor of development research at the University of Konstanz and part of our cluster “The Politics of Inequality”. Anke is also a member of the German National Academy of Sciences Leopoldina and received an Alexander von Humboldt Professorship, Germany's highest award for international research, in 2019. Thanks to this professorship, the University of Konstanz was able to lure Anke away from Oxford and come to Konstanz, which we are really proud of. Anke's research centres on economic development as well as on conflict and violence. Today, we want to talk with her about her new book “Worse Than War”, which she wrote together with her co-author James Fearon. The book has just been released by Princeton University Press. A very warm welcome, Anke.

Anke Hoeffler

Thank you very much, Gabi. Thank you, Marius. Thank you for the invitation.

Gabi Spilker

So, Anke, the title of your book is “Worse Than War”. Let's start with the question: What can be worse than war?

Anke Hoeffler

What could be worse than war? We are constantly bombarded with all sorts of images of war, and it looks apocalyptic. So, what could be worse than this? But what is actually worse is the everyday violence that many women and children experience. The wars at home.

Gabi Spilker

Does that mean this is also violence in the context of war, or is that broader than war or complementary to war?

Anke Hoeffler

In a way, both. But a lot of, or most of the violence, is completely separate from war. I call it “war at home”. You were asking about the title, but this is what I think of it. The most common type of violence that women experience globally is from their intimate partner. By intimate partner, we mean their boyfriends or husbands. And the most common type of violence that children experience is parental violence. It's the physical punishment of children that is the most common thing. And that happens at home. And this has got nothing to do with the war per se. But of course, once you've had a war, there's also more violence in society. We know that men who've experienced war are more violent as intimate partners and also parents who've been at war are more violent towards their children.

Marius Busemeyer

But that's obviously very challenging to measure empirically. I mean, we have all sorts of statistics about war victims, how many people die in violent conflict, both between states but also within states. There are big databases on this, even though there's also uncertainty regarding the estimates of war casualties. But how did you manage to come up with a measure of this violence that happens in homes? A lot of it happens behind closed doors, I would assume.

Anke Hoeffler

Yes, it's hidden in plain sight. And it's very difficult to measure and of course, the data is poor. But we thought it was worthwhile to do this whole exercise. 20 years ago, you wouldn't have been able to do it, but now we've got a lot of victimisation surveys. You can't rely on police statistics because in a lot of countries, victims are unlikely to go to the police because they don't trust the police. You really need a whole range of victimisation surveys.

Gabi Spilker

What are victimisation surveys?

Anke Hoeffler

Where people are being asked whether certain things have happened to them. For example, in the past year, have you been physically assaulted? Has something been stolen from your property? Those types of things. We obviously focus on physical violence here.

Gabi Spilker

And these surveys exist in plenty of countries, so that you can compare them across the globe, I suppose.

Anke Hoeffler

Yes, so comparison is, of course, really tricky because what would be theft in one country or assault in one country and what is rape in one country. This is not necessarily defined in the same way everywhere else. You really need to have something that is comparable. And the demographic and health surveys, for example, ask women about their experiences. The best way is not to ask in a general way, have you experienced violence, but it's act-based. So it's very specific: Have you been pushed? Has your hair been pulled? And who's done this to you? These are the typical questions in the demographic and health surveys that have been carried out in many, many different countries.

Marius Busemeyer

But it sounds, to be honest, a little bit like comparing apples with oranges. There are problems of harmonisation between different countries. That can be solved if you ask the same questions to different people in different countries. On the one hand, war is, of course, a completely exceptional situation for many people. And then, on the other hand, this domestic violence is, unfortunately, a thing that occurs every day. It's very hard to compare these two if you want to get a measure of the total amount of violence that is in a society. How did you manage that? How can you compare these different things?

Anke Hoeffler

You're right that we've got very good global databases on war. Uppsala University in Sweden has collected these types of violence data for a long time, and we know that they are comparable across countries. They basically scrape media reports and…

Marius Busemeyer

Scrape, what does it mean?

Anke Hoeffler

They go through these reports from Reuters and BBC and local media and look for how many people have been killed and how many people have been injured. Yes, we do have quite a good idea about how many people die in war. We do also have quite good statistics of how many people die due to homicide outside of war, because there's a body of evidence. We think that these are the most reliable data. There, you can't do a victimization survey. There, you have to rely on official statistics. Official statistics either come from doctors or from the legal system, from the police. And for most countries, they're not the same. For political reasons, very often the legal statistics are much lower than the health ones from doctors, especially in countries that are not democracies. For the poorest world region, Sub-Saharan Africa, we've got very little reliable data. This has been a problem throughout the project. And you have to think of believable ways, you first have to define what you want to measure. Then you look at what you can measure, what's available, but there are still gaps. And then you have to do an imputation. You've got to do a statistical filling in of the data that you haven't got.

Gabi Spilker

Maybe even more precisely: How did you measure the cost of violence exactly? You have these surveys, and then what do you do with the surveys? How do you get to a number that really tells us the cost of violence in country X is so and so high?

Anke Hoeffler

First of all, you need to define what violence is, and that's already difficult. What about child marriage, is that violence? A child is defined by the UN as under 18. But in many different countries, whether she's 17 and a half or 18 years old probably makes very little difference. So you can make up your mind whether you think that this is something like statutory rape, because she shouldn't be married under the age of 18. Or whether you say, any child married at 15 or younger, that we would consider as violence. You have to have a definition of violence. And we are quite open about what we consider as violence and what we do not consider as violence. Reasonably, you could disagree with our definitions. And then also sometimes you just haven't got the data to measure it terribly well, we've got gaps in there. You have the prevalence rates. In terms of the Uppsala data, the homicide data, and then the assault data, which we have from these victimization surveys. And then maybe that's all many people want to know. They find out, yes, that women and children are much, much more likely to become victims of violence. And that's maybe where you want to stop. If you want to go further and say, what's the burden, what's the cost to society, then you need to think about how we put a price on a life and on an injury. And then you multiply out these cost factors by the prevalence rates.

Marius Busemeyer

I think this is a very important point, especially from an economics perspective. You are an economist, from your background, this is relatively normal to think about the costs of human life. I mean, these things that are in healthcare economics, where you measure how much money you spend on certain medical treatments and how much you can generate in terms of benefits for this. But it's still something that might feel a little bit alien to people to think about the cost of violence in terms of economic costs rather than societal costs or well-being. Maybe I'm interpreting this wrong. What is your definition of cost essentially? Is this an economic cost, or does this also include other types of costs that are more social?

Anke Hoeffler

We have tangible costs, something you can touch, that you have to pay out of pocket for. For example, you've been assaulted and you go to a hospital, and you have to pay for this hospital stay. So that's a tangible direct cost. But of course, then you're not able to work. And that's an indirect tangible cost because you've got a productivity loss. But then there's, of course, more to this. It's the cost of suffering. There's a welfare loss to society. And for these intangible costs of suffering, you might want to consider how much a society is willing to pay to avoid a death or an injury.

Marius Busemeyer

So you do provide a cost estimate also of these not purely economic costs?

Anke Hoeffler

Absolutely.

Gabi Spilker

And very precisely, what is the cost? What is the price tag you put on that?

Anke Hoeffler

I found it very difficult to put a price tag on a person's life. Because every life is precious, but it's not invaluable. And as Marius already said, we're making a lot of policy decisions on whether it’s worth saving this life or not. You mentioned medicine, but for example, a lot of traffic departments do this. Should we build an underpass? It's costing so much money and so many people have been killed on this stretch of road. We do it all the time, we're just not very explicit about it. Let me be explicit about it because I think it's helpful for policy decisions. We use a system called the value of a statistical life. The value of a statistical life is a concept that goes back to Thomas Schelling. And there, economists were looking at labor markets. For example, you have a risky job working on an oil platform, and it's a 1 in 1 million chance that you die in this job. So, you're demanding a higher premium for your job. $10 to make up for this risk. Then a certain death would just be, statistically speaking, $10 million. And that's what economists use. It's not the value of your child's life or the president's life, but it is a statistical life.

Marius Busemeyer

And it's how much people would be willing to pay to avoid this, right?

Gabi Spilker

Is this similar to how insurance companies would go about measuring? Of course, there are different ways you could measure it.

Anke Hoeffler

A life insurance premium would be an alternative way of doing it. But not a lot of countries have this life insurance data. And also, if you think about it, in Germany, a lot of people never take this on or their relatives get it on their death. But it's this type of pension fund for them, and they cash it in when they're older.

Marius Busemeyer

We talked a lot about measurement now. Maybe you can provide us with a little bit more concrete estimates on the actual cost of this. What does it actually cost, these different types of violence that you cover in the book? Can you come up with some numbers? This probably varies a lot between countries. Some countries experience war, interstate conflict, others not at all. But all the countries experience some kind of crime, some kind of domestic violence. That's already probably an important difference. But maybe you can come up with some concrete numbers. How much does this actually cost in terms of percentage of GDP or absolute numbers?

Anke Hoeffler

I'm happy to tell you an absolute number. But maybe it's always important to keep in mind the relative sizes of what we call interpersonal violence. The homicides and the assaults and these war burdens, that's of much greater interest to me. If we think globally, we've got $34 trillion US in terms of cost. It's a huge number and I think nobody can quite understand what it means. It would be about $4,000 per person.

Marius Busemeyer

That's the global average, basically.

Anke Hoeffler

That's the global total for one year of all the violence that's happening that we can measure.

Gabi Spilker

And how is this distributed globally? I assume it's not distributed evenly across different countries, but there might be countries that experience more of these costs or suffer more from these costs than others.

Anke Hoeffler

Absolutely. There's a very unequal distribution. Marius already mentioned war. War happens in very few countries. When you're a conflict researcher like me, you tend to forget that most of the world is actually at peace. So that happens typically in very, very poor countries. It is in the Sudans and the Afghanistans of this world that many of these deadly conflicts take place. And then the other type of violence, the interpersonal violence, the very high homicide rates we see throughout Latin America stand out. But what is the highest share in violence is violence against women, in every single world region. You can look at it in countries, world regions, or by income groups. And if you look by income groups, it's the poorer income groups that have a much higher prevalence of violence, not necessarily the costs, because it depends. I know you wanted to get away from this whole measurement issue, but when you ask me about a number, it depends on what the value of a global citizen is. Should that be the same number for every person? A Burundian and a German having the same value, or not. Maybe we should be looking at the German labor market and the Burundian labor market and get these values of statistical life. I think it depends on what problem you want to solve. We are looking very much at the climate change literature, and there it's a global problem. Clearly, we should value every global person's life the same. But if you want to solve a problem in Burundi, it's maybe much better to say, if we want to go towards a cost-benefit analysis, value the Burundian life at the Burundian level.

Marius Busemeyer

That's kind of controversial, I can imagine. 

Gabi Spilker

Yes, it is. And how did you decide for the book?

Anke Hoeffler

We give you a choice, we do both. And of course, there's much more violence in low-income countries. If you use a global value, then this is a much higher value, the $34 trillion. If you do it by country, you've got $23 trillion, that makes a big difference. 

Marius Busemeyer

I see. In terms of the share of the violence that's caused by interpersonal violence relative to the share that's caused by conflict?

Anke Hoeffler

Most people think that, war is the dominant thing in the news, but we find that about 12% of all of these global costs that I told you about is caused by war and terrorism.

Gabi Spilker

That means 88% is caused by something else. Now I understand the title “Worse than War”. That means that the high-income countries are also not free from the cost of violence. Is that correct?

Anke Hoeffler

Absolutely. What's interesting in the high-income countries is that there's a lot of suicide. You might again think suicide is a self-directed violence. That this is not something that we should be considering when we talk about the costs of violence. But maybe you are, again, we give you the choice. We put it in different tables and in different graphs, we put it separately. You were asking me earlier about who experiences the most violence, and its low-income countries and low-income settings. But with suicide, it's the other way around. It's the higher-income countries that experience higher prevalence rates.

Marius Busemeyer

Since we're at the In_equality Cluster, do you also see some association with prevailing levels of inequality? There's the global dimension of inequality, of course, that we talked about a lot, but you also mentioned the difference between high and low-income people within countries. Do you see some association between the cost of violence and inequality?

Anke Hoeffler

We haven't looked further into how violence is distributed within the countries, I think that would be another interesting research project. We were more interested in global inequality. Global in the sense of world regions and income groupings. But also men, women, and children. And we find that, if you are a woman or a child in a low-income setting, in particular in Sub-Saharan Africa, you are much more likely to suffer from violence.

Gabi Spilker

On a global level, you've previously said, it's women in all country contexts who suffer a lot. Is that correct?

Anke Hoeffler

Yes. Intimate partner violence sadly happens in all countries.

Gabi Spilker

And what about children? Is this confined mainly to the low-income countries, or do you also see that across the globe? And the next question would be, what can we do about this? Now that we have all these horrible numbers, what do we do with them?

Anke Hoeffler

Very often, people say it's their culture and it shouldn't and can't be changed, we shouldn't interfere. It's neocolonialism when we present these numbers and say, something ought to change. I don't think it's a cultural issue. Look back at German history, not the distant history. I was married in 1996 and rape within marriage was not illegal, there was no law against it. That was only in 1997. Sometimes it just takes a long time to change norms, but then things can also rapidly change. Sweden was the first country to outlaw all physical punishment of children in 1979, and a lot of different countries followed. I think the experience that we've already had as children, and our children now have is an education that is mainly free of violence. So, it can be changed.

Marius Busemeyer

Maybe the question is less about war and peace on the global scale, but it's more about these domestic policy interventions like educational measures, social policy measures, health measures, that could be actually more effective in lowering the total cost of violence than what we typically think of when we think about conflict, right? Is that one of the implications of your book? What kind of policies do you have in mind?

Anke Hoeffler

Yes. After coming up with all of these sad numbers, we've got two chapters on violence reduction and prevention, one on war prevention and the other on interpersonal violence prevention. At the moment, the international community is putting a lot of effort into reducing conflict. But when you actually look at the success rate, it's a pretty mixed picture. What we do in terms of diplomacy, sanctions, aid, UN peacekeeping, seems to be pretty solid evidence that we can keep the peace with UN peacekeepers. But at the moment, I don't see the international situation as such that we're going to see many more new peacekeeping operations. So given that we put so much emphasis on it, but it's so mixed, I strongly advocate that we look at what could be done domestically against violence against women and children and also against the high homicide rates. You've already mentioned some things, but maybe you have particular interests in policy prevention packages. You really have to reduce the availability of weapons. There's very good evidence that if you reduce access to weapons, mainly firearms, because they're the most deadly weapons, then you reduce harm.

Gabi Spilker

One interesting example I've learned about is in our neighbouring country, Switzerland. There are a lot of suicides happening because of the army’s weapons, is that right?

Anke Hoeffler

Yes, and there was an interesting, we call it a natural experiment, but it's a policy experiment. The size of the military army was drastically reduced in Switzerland about 20 years ago. Those of you who are not living next to Switzerland might need a little bit of an explanation. People have very short army training and then they go home with their entire kit, including their gun. And Switzerland does have a high homicide rate, not the assisted dying, that's not what we're talking about. We're talking about high suicide rates. When the army was reduced, the suicide rate also reduced significantly. Some people used different methods to kill themselves and some people are just determined to kill themselves, but it has been dramatically reduced. That shows you that the availability of weapons, the spontaneous kind of I am going to hurt myself or others, is highly dependent on the easy availability and access to weapons. In Switzerland, we've got a very low homicide rate, so you can't really see an impact on the homicide rate.

Gabi Spilker

What else can you do?

Anke Hoeffler

You can reduce alcohol consumption because alcohol makes people, on average, much more violent. And a lot of intimate partner violence is accompanied by drinking or other substance abuse. These partner programs that have become very popular to address this type of violence, again, we have got very mixed evidence on this, but very good or much better evidence for the programs that address substance abuse, because typically, both partners are very often young. It's much more prevalent in young couples, because this locus of control isn't as established in teenagers and they are physically more matched. There is the idea that men can also be victims of domestic violence, but because they are typically physically stronger, they cause much more harm to women.

Marius Busemeyer

And what about broader measures like education and social policy?

Anke Hoeffler

Yes, everybody likes the idea of I educate you about alcohol, and then all the lobby groups are happy about this. But there's no evidence that this works. You've got to really reduce the availability of alcohol. A typical violent scene is pub closing times, people spill out into the street and start pushing and shoving, and one thing leads to another. That's a typical dangerous situation. For example, you can have plastic glasses because glass has been used very often as a weapon. There are little things, you know. Reducing alcohol consumption around football matches, for example, reduces violence. And you can make alcohol more expensive. So we've got very good evidence from Scotland where a unit of alcohol has a minimum price. People are then consuming alcohol in a different way. Because it's not the glass of wine that you have with your evening meal that's dangerous, but it's the binge drinking. It's this episodic heavy drinking, what the World Health Organization calls it, with high-percentage alcohol, vodka and schnapps and so on, that makes people more violent.

Marius Busemeyer

I was also referring to education in a broader sense, right? Because you mentioned the value orientation that plays a certain role. Also, this topic of neocolonialism that you mentioned. That people or countries might be quite reluctant to accept suggestions on how to redesign the education programs when it comes to gender relations, for instance, or family relations. When this advice comes from the outside, even if it comes from the United Nations, I guess. What could we do in this regard?

Anke Hoeffler

As a professor of development research, I've long come to the conclusion that you cannot change anything from the outside, it has to come from the inside. But every single country has NGOs, social movements, people who are saying this has got to stop. We can't have all of these children being hit and caned by their parents regularly. Also, all the violence that goes on in schools. You've got to try and support these champions that exist in every single country. And it's interesting that a lot of violence reduction research happens in high-income countries, and it's now slowly being taken to low-income countries. I don't mean forced upon. Like, if you are a champion, I can tell you, we know that this and this works. Would you like to try this and then adapt it to the local circumstances? For example, in the United States, we've got extremely good evidence that dating programs work. This is for teenagers aged 14 to 15, before they typically have their first romantic relationship. It's not about sex education, it's about relationship education. It's about what consent is, how do you argue with your partner? How do you resolve conflicts? We've got very good evidence that when these, this is for girls and boys at this age, when you survey them later, they've got much lower rates of intimate partner violence. Now, that doesn't translate. The concept of dating doesn't exist in every country, but it's been countries like Kenya and Haiti that have had very good success with these types of programmes.

Gabi Spilker

Very interesting. Maybe a final question, now that you have put all the numbers together as well as provided a very good overview of what works and what potentially doesn't work. What do you do with this knowledge? How do you get this knowledge across and to whom? Whom do you target as one of the experts in this field, and how to really make a difference in this very, very important field?

Anke Hoeffler

What can you actually ever do as an academic? And how do you have an impact? I think you can just try to change minds and then see whether this snowballs. The podcast seems to be a good start. When I first started doing this economic research on conflict, this wasn't something that development economists were doing. And you just have to be patient and go with your message to different policy shapers and see whether it gains some traction. And sometimes, despite all the odds … My first article, “Greed and Grievance in Civil War” with my then supervisor, Paul Collier, took 5 years to publish. Nobody wanted to hear anything about this topic. 

Gabi Spilker

And today, it is one of the most frequently cited works in this field and a seminal article in this area.

Anke Hoeffler

Yes, thank you. Sometimes it seems unlikely and certain things then take off. With a reputable publisher and a book, I will hopefully be able to distribute the message widely.

Gabi Spilker

And we hope that this podcast helps to distribute this important message. We thank you, Anke, very much for this super interesting conversation. We learned a lot about the cost of violence and that this is, indeed, potentially worse than war. Not to say that war isn't bad, but other forms of violence also exist, mainly targeted against women and children. And we also learned what can help, now we hope that many people will listen to this podcast. That many people listen to what you have to say, maybe read the book. We keep our fingers crossed that the world will potentially be in a better shape in the future. Thank you.

Marius Busemeyer

Thank you.

Anke Hoeffler

Thank you very much for having me.

Gabi Spilker

And to all our listeners, please tune in every first Wednesday of the month.

  (This transcript was created using AI.)