Do Israeli Policies Deter Violence?
With fighting between Israel and Hamas ongoing, Esteban Klor, the Rosita Herczeg professor of economics at The Hebrew University of Jerusalem, weighs in on whether Israeli defense policies deter violence. Part of an IGCC research project called Deterrence with Proxies, Klor analyzes Israel’s controversial use of house demolitions and retaliatory attacks, and finds that the evidence on their effectiveness is mixed. He also discusses how he balances his views about the normative implications of Israeli policies with his job as an empirical researcher. Says Klor: “I would never let my views affect my research, but sometimes the outcome of my research affects the way I look at the world.”
Wars today are fought most often, not directly between adversaries, but through proxies. Think the U.S. in Afghanistan and Iraq, the French in the Sahel, or Iran in the Middle East—powerful countries pursuing their objectives through local intermediaries on the ground.
Deterrence with Proxies, an IGCC project funded by the U.S. Department of Defense’s Minerva Research Initiative, studies these relationships. Tell me more about the project—what question are you trying to answer?
We present an alternative to the practice of countries putting boots on the ground in other countries and trying to buy hearts and minds to stabilize countries. When we started this project in 2014, we understood that in order to achieve peace and prosperity, in order to deter insurgents, you need to take into account that insurgents and the local government have different incentives. The idea is to align those incentives with the goals of, in this case, the U.S. or Israel or other power, by providing a contract, if you will, to local rulers that, through punishments and rewards, will achieve peace and prosperity.
Why would a powerful country like the United States work through a proxy to achieve a goal, and not just do it directly?
There are several advantages. The first advantage is that if you do the opposite—if you do it directly— that means that you will have to do it forever. The idea is not to stay in a given country for an unlimited time period. You train local people to help you out, then give them the power to rule in the way that they’re supposed to rule. The second advantage is that, usually, local agents have better information. And they receive better cooperation from the local population.
Is it cheaper?
Sometimes it is, sometimes it isn’t. Afghanistan was obviously very expensive.
As part of your work on this project, you’ve done a paper with IGCC research director Eli Berman on whether retaliation by Israel for Palestinian attacks can reduce future violence. What did you find?
The paper tries to understand the relationship between Israel, Hamas, and local militants who shoot rockets and missiles against Israel. We look at what type of incentives or punishments given by Israel to Hamas stop Hamas and other militant groups from attacking Israel. What we found is that no matter how hard Israel tries to discipline Hamas, given Hamas’ grievances, they will never be completely deterred. There is no incentive for Hamas to stop shooting completely given that their requests are not being met by Israel.
So, what we see is what economists call a game, where Israel is setting rules to the interaction, defining what it can tolerate and what it won’t tolerate. And that actually, in a way, constrains Hamas attacks. Hamas knows that Israel will tolerate small attacks. But bigger attacks will lead to retaliation. So, Israel is able to achieve what is called “narrow deterrence.” That is, Israel is able to set some rules to the violent confrontations, but is unable to completely stop violent interactions with Hamas and other Gazan militant groups.
You have also studied Israel’s use of house demolitions—which is when the Israeli Army destroys the homes of Palestinians.
The issue of house demolitions is a hot topic in Israel. The legality of this policy is constantly being challenged in front of the Supreme Court. The logic is: the Israeli government claims that it’s very difficult to deter a suicide terrorist. Let’s forget about the normative implications. But from a positive standpoint, I say, well, there are costs and benefits. If I commit a crime, if I’m successful, that’s the benefit. But if I’m not successful, and I’m caught, I will go to jail. That’s the cost. Now, if I’m willing to commit suicide, these cost and benefit calculations don’t make sense—unless I’m afflicting a cost on the people I care about. So maybe we can deter a suicide terrorist if that person knows his kids are going to be homeless.
House demolition has been used even before the creation of the State of Israel. It was enacted by the British Mandate before 1948. When Israel occupied those territories in 1967, that was the law of the land. And because those territories were never annexed formally to the State of Israel, the Israeli military says that it’s legal to use it, because that’s the law in those places. And they keep going in front of the Supreme Court saying that it’s an effective way to deter potential suicide terrorists.
Palestinian organizations and human rights organizations claim that this policy is illegal and against the Geneva Convention. Because you’re punishing innocent civilians, usually the wife and kids of a suicide bomber, while the perpetrator is already dead. These organizations also say that the policy backfires—it’s not effective. It actually creates a backlash and leads to an increase in suicide terrorism.
Are house demolitions only used in response to suicide attacks?
There are three different kinds of house demolitions. So far, we’ve been talking about punitive house demolitions. Punitive demolitions started as a tool against suicide terrorists, but then it expanded to most terror operatives. There’s also what we call precautionary house demolitions or demolitions for military purposes. The third type is called administrative demolitions, which is the demolition of houses that were built without permits. Those demolitions usually occur in East Jerusalem. It’s very difficult for the Palestinian population of East Jerusalem to obtain permits to build houses or to extend their houses. And so they do it anyhow, and every once in a while, the ruler of the area gets them demolished.
How frequently are Palestinian houses demolished?
Right now, it’s infrequent. It was very frequent during the Second Intifada.
You also mentioned precautionary demolition. Can you say more about that, just to clarify the difference?
The Israeli army calls them demolitions for military purposes, and basically, it’s when there’s no claim of any wrongdoing by the owner of the house. It’s all about the house location. So, for example, before the disengagement of 2005, the Israeli Army felt it was important to clear the Gaza border with Egypt, because there were tunnels being used to bring in military equipment and it would be easier to patrol the border without those houses there. Close to 700 houses were demolished. The owners of the houses do receive some compensation, but it’s still a very difficult experience to go through. Usually, it’s houses by the border or houses that overlook roads that are used by Israeli settlers. Again, there is no claim that the owner of the house is going to the roof and shooting. But the fact that somebody could do that makes the house a problem. And so those houses are demolished.
When a house is demolished, are people ever hurt or killed? What happens to the infrastructure itself?
The house is completely demolished. Sometimes it’s an apartment in a building and they fill the apartment with cement. They give a prior warning—it’s required by law—but there are questions about how many days are sufficient. But no people are injured or killed or hurt in any way. But sometimes tenants will appeal to the Supreme Court. When they rule allowing the demolition, the army will come and say, okay, you have five hours, take whatever you need. It’s usually in the middle of the night, to avoid bystanders.
You were looking at whether the use of this tactic would help reduce the likelihood of future attacks—does it?
In the first study from 2015 (with Efraim Benmelech and Claude Berrebi), we were trying to see whether precautionary and punitive house demolitions are an effective tool to combat terrorism. And we found that punitive house demolitions do bring about a decrease in the probability that a suicide attacker will come from a given locality during the month of the demolition. The problem is that the actual impact, while statistically significant, is not that substantial. The deterrence effect disappears after one month. And the effect doesn’t spread to localities surrounding the affected locality.
So while we say that it is effective, it may not be efficient, because we only look at the benefits. We don’t do a cost-benefit analysis. If there is any cost associated with using this policy, then it’s possible that these costs will outweigh the benefits.
For precautionary house demolitions which is an indiscriminate policy of counterterrorism, we conduct the same exercise, and find that this policy leads to a significant and substantial increase in the willingness of individuals to commit suicide terror attacks. Again, we can’t say whether this is inefficient or ineffective, because the actual outcome that the Israeli Army looks at when deciding to conduct precautionary house demolitions is not the number of suicides terror attacks. But it’s still important to know that, whatever the goal is, this policy has an additional cost in terms of leading to a high number of suicide terror attacks.
In these studies, you look at very specific slices of activity between Israel and Palestinians. But of course, there’s all this other stuff happening that might be driving violence—like the occupation itself, Israeli policies that hurt the Palestinian economy, collective punishment. How do you take into account all of those different variables?
We are very careful to try to isolate the effect of house demolitions from all the other things that are going on. We control for the pre-existing levels of violence, and for other forms of counterterrorism, like curfews, targeted killings, closures, etc.
Second, and this is crucial, we rely on what we call panel data. Panel data has temporal and geographical variation. This allows us to estimate deviations from the mean for every locality and to control also for national trends.
What’s the goal of these research projects?
My first goal is to understand what is going on. In Israel you see rockets falling every day. Once in a while you see a plane or a drone and you think: why have we been doing this for 20 years and nothing seems to change? With this project, I learned that this back and forth between Israel and the Palestinians is an equilibrium. The next question is: does it work? Is this the best policy for Israel? Is this the best contract available for the principal?
I think, given the political constraints, it’s a pretty good contract. Even though at first view it looks crazy: we keep shooting at each other and every five years there’s a major military operation. But maybe there’s no way out. It’s sad to realize but I understand something that I didn’t know before.
Regarding the project on house demolitions, the goal is straightforward: is this a good or bad policy to deter terrorism?
In your 2015 paper, you say that the main factors in bringing about the beginning or the end of terror campaigns belong to the political rather than the military realm. If the determinants are political and not tactical, then what’s the purpose of the research?
The beginning or the end of the Intifada, whether or not we are going to have a ceasefire with the Gaza Strip, whether the U.S. military is going to stay in Afghanistan or not—those are political issues. We are looking at the micro level. So long as we need to continue to manage a conflict, what’s the best way to do it? The big political decisions have a great effect on the evolution of a conflict. However, between those big political decisions there is an army and there is a civilian population. The army is trying to achieve some goal and the civilians and militants are trying to achieve a different goal. What’s the best way to micromanage the daily fluctuation of violence while we wait for the next major political decision that will transform this conflict in some other way? This is the focus of my research projects.
As a researcher, do you separate whatever you might think politically from the stuff that you’re studying?
It’s not a conflict for me at all. I guess that’s the beauty of being an empirical economist. I generally start a research project when I have an interesting question and I don’t know the answer. For example, regardless of my political views, I didn’t know whether house demolitions worked or didn’t.
From a normative standpoint, of course I have my own ideas about whether this is the right policy or not, but from a positive standpoint, I was curious. Do they work? How is it possible that Israel is using this policy for this many years and it’s so controversial and we don’t know the answer?
Give me the data and I’m going to do the best analysis I can, and whatever I learn I’m going to write it down and show it to the public. First and foremost, I’m a social scientist, and then I am a citizen of the State of Israel. I would never let my views affect my research, but sometimes the outcome of my research affects the way I look at the world.
In the best-case scenario, what contribution do you hope this research makes?
I’m going to give you an actual example that happened in front of the Supreme Court. The army argued that punitive house demolitions were an effective way to deter suicide attacks. The Supreme Court said, well, we read a paper that found that it’s actually not so clear cut. For me, that’s the kind of impact I want my work to have. I want policymakers to have better information when they make decisions that affect the lives of a lot of people. I don’t know that I can affect Israel’s decision to use house demolitions or drones to attack the Gaza Strip. But at least they can choose to learn from my research and use a policy in sensible ways. That’s a good outcome for me. I did my job. I contributed something to society that goes beyond publishing a paper.