It’s just 35 hours by plane, train and automobile from my East Village apartment to a small displacement camp outside Kitgum town in northern Uganda. Jeannie and I arrived yesterday night, and spent our first day conducting interviews in a camp of about 500 displaced families.

I’m supposed to be pre-testing some of our new survey sections on local political activity, levels of domestic abuse, and women’s empowerment for an upcoming evaluation of a post-conflict development and empowerment program for women. I’m too interested, however, in all of the changes afoot to stick to the script. Peace talks have been progressing, there has been little or no rebel violence or abduction in almost two years, and so I’m most interested to see if people are planning to head back to their homes and land this dry season, now that the grasses are dry and ready to be cut for new roofs.

Four years ago, as rebel violence and attacks intensified, the government forcibly displaced the entire rural Acholi population—roughly 1.5 million people—into squalid ‘protected villages’, or internal displacement camps, often no more than five or ten kilometers from their homes and land. Fear of rebel attacks and government decree kept people from accessing their land (and only livelihood), and so the population has been living on food aid for at least four years. After the violence abated last year, the government allowed people to access their lands again, and some (perhaps a tenth) moved to smaller ‘satellite’ camps closer to their land. These camps remain under-serviced and crowded (the one I visited today still had several thousand people) and I half expected to hear that a majority of households would begin to build huts on their home lands again and begin a transition back to normal life.

Even a few hours of interviews is revealing. First, to my surprise, there still appear to be small numbers of armed rebels scattered throughout the countryside. In the camp I visited today, sightings of small groups of rebels—usually just one or two—are almost a daily occurrence. Not so long ago, encountering a rebel in the bush was an invitation to be abducted, killed, beaten, or even mutilated. With peace talks on, most rebels have congregated in eastern DRC and southern Sudan, but small bands of guerrillas (no one reported seeing more than four or five at once) still seem to be roaming about. They avoid civilians, however, and disappear into the bush when they encounter a civilian. Whatever could they be doing? Are they cut off from joining the main force? Unaware of the peace talks and situation? Gathering intelligence? Poised for retributive attacks should the peace talks spoil, or the rebel leaders captured and brought to the International Criminal Court?

My second surprise is that I could find no one planning a move back home in the next few months. Most are within an hour or two walk of their fields, and are planning to remain in the camp for the moment. All plan for a return to a their ancestral lands. Just not quite yet. The chief reasons? Fear that the peace talks will break down; fear of the small groups of reclusive rebels roaming about.

Such prudence would be perfectly understandable save for two facts. First, the condition of the camps is such that the increased risk of death from poverty and disease is arguably greater than the risk of violence at home. Second, these same people lived at home through far greater risk and violence, in particular the ten years before the army displaced them into camps.

Yet if violence is such a deterrence to living at home, why was forcible displacement necessary? Why did people prefer home to camps when the violence was much greater? Several thoughts occur, though none seem quite satisfactory.

• Return presents a collective action problem, whereby the risk of being targeted for rebel violence is only low if everyone moves home, but high if you are among a very few. In this instance, a public signal, significant event, or government action might be required to coordinate and provoke mass movement.
• Fear of moving home could be an example of the effectiveness of terror tactics—that a very small probability of random but extreme violence can have disproportionately large effects on behavior, perhaps even greater than that of a more sizable and constant risk.
• There could be significant switching costs, including material ones (the cost of rebuilding and moving–admittedly not all that great, as far as I can tell, however) or perhaps even psychological switching costs.
• People are biased towards the status quo because of a lower tolerance for risk on actions taken rather than inaction.
• People are worried that they will lose access to NGO aid and services if they relocate.

I haven’t any real theory or evidence for these, YET. The baseline survey we are conducting for the program evaluation might let us build a panel of displaced persons as they are making the decision to return home. How I will test the above theories is so far beyond me…