Note: All aid worker casualty statistics in this article drawn from “Humanitarian Outcomes (2014), Aid Worker Security Database, https://aidworkersecurity.org/”
It’s been 5 years since the first official World Humanitarian Day, August 19th 2009. The UN sanctioned the day in memory of the suicide bombing of the Canal Hotel, the UN Headquarters in Baghdad on that day in 2003. 11 years ago today, a suicide bomber drove a flatbed truck into the UN compound, parked beneath the window of the UN Special Representative, Sergio Vieira de Mello, and detonated his payload. 22 people, including de Mello himself, were killed, most of them UN personnel.
World Humanitarian Day commemorates their sacrifice, but also remembers aid workers who are targeted by violence while carrying out their work on behalf of populations in need around the world. Each year, and in increasing numbers, scores of aid workers experience violence in the field while bringing aid to people affected by war and disaster. Many, like de Mello and the 21 others who died August 19th 2003, pay with their lives.
As somebody who was involved in a serious security incident some years back, today is a day that means something to me, too, and a chance for me to reflect on the co-workers who were with me that day, some of whom still carry the wounds they sustained. Many of my friends and colleagues in the aid industry have themselves survived security incidents, and many have lost friends and colleagues to violence.
The Aid Worker Security Database is a project which compiles global information about attacks on aid workers. Specifically it looks at acts of violence (as opposed to accidents and illness, which claim even more lives) in which aid workers are killed, seriously injured or kidnapped. The project began gathering data in 1997, so that now, 17 years on, it’s becoming easier to determine trends in the sector.
Let’s take a quick look at the figures.
In 2000, the AWSD recorded 42 violent incidents involving aid workers, involving 91 aid workers around the world (variously with UN, INGOs, LNGOs, IFRC and Red Cross/Red Crescent societies). Of these, 57 were killed, 23 wounded, and 11 kidnapped. 21 were expatriate staff (23%) while 70 were nationals (77%).
In 2013, thirteen years on, the database recorded 251 violent incidents against aid workers, with 460 aid workers caught up in violence. 155 aid workers were killed, 171 injured, and 134 kidnapped. 59 were expatriates (13%) while 401 (87%) were national staff.
The 11-year mean has seen an average of 112 attacks per year against aid workers, with 217 aid workers involved. 80 aid workers a year have been killed, 80 a year injured and 58 a year kidnapped. On average, 18% of aid workers involved in violence are expatriates, while 82% are nationals.
Since 2000, there has been a dramatic shift in the security landscape for aid workers. While the dataset is still relatively small, what is reflected in the figures is backed up by anecdotal evidence from the field, by individual agencies’ experiences and records (most large agencies and the UN have their own internal security incident tracking database for analysis and risk management purposes), and by trends and behaviours within belligerent groups.
Note as well that the AWSD only captures the top end of security incidents- instances of death, serious injury or kidnapping. It excludes less serious incidents (illegal detentions, muggings, threats, robberies, etc.). The AWSD does not factor in the psychological consequences of these incidents, which for victims of violence, including rape, can include a lifetime of psychological distress or struggles with Post-Traumatic Stress Disorder (a report by the Antares Foundation this year reported up to 30% of aid workers exhibit symptoms of PTSD upon returning from the field). It also doesn’t capture the ‘near misses’- times when good management or pure luck has averted a much more serious incident in the field.
On top of this, remember that since 2000 (when most agencies considered themselves ‘neutral’ and therefore relatively immune to being targeted by hostile actors), the UN and other international aid agencies have seen a massive increase in their risk management strategies- increasing their protective and deterrent measures in particular to make it harder for them to be targeted, and generally trying to reduce their exposure to threats.
Despite all this, there has still been a huge increase in incidents and casualties in the field. Between 2000 and 2013 there has been a six-fold increase in the number of attacks, a five-fold increase in the number of casualties, with deaths up 270%, injuries up 740%, and kidnapping up by a substantial 12-fold increase. The only trend that has remained relatively constant is the proportion of expatriate vs. national staff involved in attacks, which hovers for the most part in the 15-20% range.
Overall, the absolute numbers of aid workers in the field (national and expatriate) has increased- so the rate of attacks on aid workers is not quite as pronounced. The AWSD estimates a modest increase in the attack rate against aid workers over the study period, without divulging the precise figures (they are based on its estimates of numbers of aid workers in the field- a difficult figure to pin down).
During the period 2006-2012, this fluctuated between 40 and 60 victims per 100,000 aid workers. Using the average figure of 50 victims per 100,000 aid workers, this results in a probability of a 1 in 2,000 chance of becoming victim of a serious attack as an aid worker. In perspective, the chance of being hit by lightning is estimated at 1 in 1.6 million. The chance of dying in a plane crash (if flying one of the world’s bottom-rated 25 airlines) is 1 in 800,000. The chance of dying in a plane crash (if flying one of the world’s top-rated 25 airlines) is 1 in 12 million. The chance of being killed in a car crash this year is about 1 in 7,000.
There are some key trends we can draw out of these figures.
- Kidnapping is a rapidly growing threat.
The sheer number of aid worker kidnappings- from 11 in 2000 to 134 in 2013- is not just a statistical spike, but reflective of a well-understood development within the operational mandate of belligerent groups. Since the globalization of al Qaeda, its subsidiaries (AQIM, AQAP) and other similar extremist insurgency groups such as al Shabaab and the Taliban, their own experience has taught them that kidnap for ransom is an extremely lucrative trade. Indeed the rapid growth of AQIM in particular has been substantially attributed to the ransom payouts by European nations for the safe release of kidnapped nationals.
As such, AQ has developed guidelines for how to run a kidnap-for-ransom operation targeting foreigners, which has been widely disseminated, and acted upon. And because in the areas where these groups operate, aid workers make up a large, visible and vulnerable population of foreigners, it’s not surprising that they are frequently targeted. In fact, until this year, aid workers have been the #1 victims of overseas kidnap-for-ransom operations. In 2014, employees of global corporations apparently face a higher kidnap risk than aid workers now. But the threat to aid workers is so pronounced that the AWSD released their key 2012 report with the title “The New Normal: Coping with the Kidnapping Threat”.
If there’s a silver lining to this, it’s that kidnappings are, overall, survivable. While in the earlier years of the Iraq insurgency (2003-5) there were a number of high-profile executions of kidnapped foreigners, AQ and its partner agencies have realised that there is a better business-model, and as such, kidnapped foreigners are quite clearly valuable assets, worth protecting to exchange for prisoners or large sums of money. Since 1997, just 14% of aid worker kidnappings have resulted in the death of the hostage- and most of these deaths have occurred either during the hostage-taking itself (generally agreed as the most dangerous time of any kidnapping situation), or during an attempted rescue or escape (also highly dangerous). In short, if you survive the initial hostage-taking, you don’t attempt to escape, and nobody tries to rescue you by force, then your survival chances jump into well above 90% probability.
Just be prepared for a wait. The average holding period prior to release for an expatriate aid worker is a little under 2 months. So be patient, wait it out, and keep your spirits up.
Also, do a hostage survival course.
- Far more national staff are in the line of fire.
This is has always been a given, and this is why the average trend for expatriate v. national staff victims is relatively flat. There are simply a lot more national staff members on the ground than international ones, and therefore when something happens, they’re much more likely to be involved in the incident.
There’s some important reflection for agencies to do around this one, namely because inadvertently or deliberately, aid agencies manage the risks of national staff differently to expatriates. There is a good reason for this. Expatriates do generally have a higher risk profile than national staff (this may not be the case in some contexts, such as those involving ethnic or local political tensions). They cannot blend in, or disappear into a crowd if things go bad. They do not have the language skills to diffuse a tense situation. They have more economic value as a hostage and more political value as a target than nationals (depending on the context). Therefore there need to be more stringent risk management processes around expatriates in most contexts.
And on the flipside, dead national staff members are less damaging to the reputation of an organization than dead expatriates. The abduction and murder of CARE Iraq’s country director, Margaret Hassan, in 2004 was a high-profile media event and crisis for that organization. Ditto the kidnapping of the 2 MSF-Spain aid workers Montserrat Serra and Blanca Thiebaut from Dadaab in 2011 who were held for nearly 2 years, and in both cases would have required a massive organizational response and extensive damage-control. By contrast, while the murder by death-squad of 17 Sri Lankan ACF staff members in 2006 was a tragic event, and likewise the killing in a militant raid of 7 Pakistani World Vision staff members in 2010, neither event made a comparably big splash in the media, and neither had significant political, fundraising or profile consequences for the organizations in question (though I am confident that both events resulted in extensive internal changes to security management processes within those organizations).
As a result, we see the following:
- National staff are more likely to be given clearance to travel into hostile environments, with the expectation that their risk profile is lower and they are better able to navigate the hazards (both statements often true, but they still end up exposed to risk)
- Relative numbers of expatriate staff in the field are reducing. This is partly as a deliberate risk-management strategy, but also reflective of the increase in capacity of national staff members globally. Where many roles used to require expatriates to fill them, the number of field roles for expatriates in most large organizations is decreasing, as nationals are more and more capable to do the jobs, and do them for less than it costs to put an expatriate in that role. Note that this reduces risks to expatriate staff (who are not in harm’s way as a result), but can also reduce the overall risk to the national organization, as with fewer expatriates, the organization’s risk profile reduces, and its interest as a target goes down.
- There is an increasing use of remote-operations in hostile environments. In places like Syria, Somalia and Iraq, agencies are less likely to put their own staff on the ground, and more likely to build a relationship with a pre-existing grass-roots NGO already operational in the field. They then channel programs, funding and resources into those partner agencies to do the work for them- which partly explains the big jump in both absolute and comparative numbers of LNGO and local RCRCS victims in the AWSD. This is known as ‘risk transfer’.
- There is an increase in absolute terms in the number of local start-up NGOs. Similar to the previous point, with the increase in education, global connectivity and awareness in many less developed nations, locals have increasingly recognized the value in the NGO model to access and deliver assistance to needy local populations. As a result, there are far more spontaneous local NGOs appearing in hot-spots around the world, exposing more and more locals to security risks. Not only are these local personnel often directly implicated in crises (see for example many of the community-based organizations that operate in Syria, who will be identified by belligerents as being on one side or the other and therefore as potential targets), but these LNGOs rarely have the resources or the skills base to invest in significant security risk management for their staff and operations.
- Expatriates are more likely to receive security training. Partly because of the higher individual risk profiles of expatriate staff, and the very high cost per capita of providing security training to personnel, most large agencies have prioritized training up their senior managers and deployable expatriate staff in hostile environment survival and risk management. The result is that many field-level staff in hostile environments have little or no security training, and as such may be engaged in more risky behaviour. This is something that is increasingly acknowledged by aid agencies, and programs are beginning to be put in place to rectify this, but it is a slow process. Training 10,000 national staff globally for a large organization takes a lot of time and a lot of precious donor funding in an environment where agencies are being given less and less. For small agencies, it can be a financial impossibility.
3. Expatriate staff face a disproportionate risk in hostile environments.
For reasons already discussed- their perceived political and economic value, their lack of awareness around local risk contexts, their inability to blend or disappear- expatriates do still face a higher risk profile when compared to their local counterparts. In short- they are more likely to be targeted by a belligerent.
Although on the surface this may not appear to be borne out by the accompanying statistics, it very much is. While 2013, for example, saw just 13% of victims being expatriates, recall that there are far fewer expatriates relative to national staff in the field. Even your typical large NGO program with 100 or so local staff might only have 5 to 8 expatriates. In some cases (e.g. Ethiopia, where expatriate numbers are strictly controlled), the ratio can be far lower- just 2 or 3 expatriates for several hundred staff. Many LNGOs- whose ranks increasingly populate the statistics- have no expatriates in the field at all. The AWSD estimates that expatriates are at least twice as likely to be targeted as national staff.
This does make developing a balanced risk management strategy challenging, and ultimately, agencies need to decide where their thresholds of acceptable risk lie, how critical their programs on the ground are to the populations they are supporting, and therefore the extent to which they are willing to place their staff at risk.
- The majority of security incidents occur in a handful of global hotspots.
This goes without saying, really. There’s a reason nobody books vacations to Syria or Mogadishu- because we well understand that bad things regularly happen in these places, and going there puts us at unnecessary risk. The same is true of aid workers.
For example, in 2012, of 170 incidents, fully 130 took place in just 5 nations- Afghanistan (56), South Sudan (21), Syria (18), Somalia (17) and Pakistan (17). (Note that in relative terms, the rate of attack against aid workers was highest during this period in Somalia, as there are vastly fewer aid workers inside Somalia than inside Afghanistan and other locations). This dynamic is pretty easy to interpret looking at global and humanitarian news headlines. As an aid worker, if you’re in one of these places (this last twelve months, other security hotspots include Nigeria, Iraq and Gaza), your proportional probability of being involved in an incident is dramatically increased.
In addition, you can cross-reference this against the most deadly belligerent groups. For example, in IntelCenter’s latest analysis, Boko Haram has revealed itself as the most deadly terrorist group so far this year, with nearly 3,000 attributable killings since the start of 2014 (not, obviously, restricted to aid workers). The Islamic State comes in second place, while other groups with an honourable mention include the al-Nusrah Front, AQAP, al Shabaab and the Taliban. Again, no real surprises here if you follow security trends. As always, good security management tends to begin and end with common sense. But when you map these groups, global hotspots, and areas where aid workers operate, you immediately get a very skewed focus as to where real investment in aid worker security needs to begin.
- When it comes to expatriates, your veteran EAWs are more likely to be involved in a serious security incident.
This is a very simplified and anecdotal descriptor, but there’s a key learning here. The sort of aid worker most likely to be involved in a major security incident is not your excited, idealistic newbie sent off on her first overseas assignment from head office. Why not? Well, because she’s far more likely to be aware of her own limitations, will probably be taking risk advice (like curfews & no-go areas) seriously, and is less likely to be placed into a highly threatening environment. If she’s just arrived in country, her senses will be up and therefore her situational awareness more likely to be running high. Frankly, she’s probably going to be a little edgy- as is the norm when you first arrive somewhere unfamiliar. It’s a basic human reaction.
By contrast, the primary at-risk descriptor here (which I know goes a long way to capturing a number of readers of this blog, and this author hits quite a few of the criteria too) is probably 10 or more years into their aid worker career, have done three or four long-term assignments overseas, and are several months into their current posting. They tend to have a been-there, done-that attitude, which means they will- deliberately or inadvertently- expose themselves to more risks. They are more likely to be deployed to a hostile environment. After the first few months, they’ve settled into a routine and have lost the initial awareness and heightened focus that comes with being in a new environment. And they’re more likely to settle into a ‘nothing’s happened so far, so I’m not really expecting anything to happen now’ mindset.
In short, it’s not just the inexperienced young expatriate staff that agencies need to concentrate on, but critically, the experienced, somewhat-jaded older hands who are moving into the peak of their humanitarian careers.
The sad reality is, any UN, Red Cross or NGO response necessarily requires placing staff at ever-increasing risk these days, and that much is borne out very clearly in the accompanying statistics. The absolute number of attacks, the number of victims, and particularly, the drastic increase in the kidnapping threat, means that being an aid worker- national or expatriate- in many of the world’s hotspots is dangerous business.
For any agency that operates in hostile environments- UN, INGO, LNGO, IFRC or RCRCS- security has to be a central component of operational budgets (donors, please take note), and putting in place a robust, well-informed security strategy is critical for every country office. Management staff need to be trained up in risk management processes, clear security SOPs need developing, a skilled and experienced cadre of security staff need to be employed to support response capability, and field staff need to be trained in basic survival procedures.
Even with all these things in place, sooner or later, chances are most large agencies will face one or more major security incidents, so they must be drilled in critical risk management and containment procedures. It’s sadly so commonplace as to be almost inevitable. But with good processes in place, the chances can be reduced, the impact minimized, and, hopefully, lives spared.