Joseph O'Neill

 of this series, we looked at the problem of violence in its different guises, and at how much damage different forms of violence cause. In this, we will start from the knowledge of the ways in which violence causes the most damage to consider what we can do to help.

In the “Conflict and Violence Assessment Paper” recently released by the Copenhagen Consensus Centre, researchers estimate that sexual and physical violence against women and children costs society at least 5 times more than the cost of homicides, wars, and terrorism put together. Given the large cost of violence worldwide, it is important to prioritise the most effective responses aimed at solving the problem.

To this end, authors evaluate the literature and estimate Benefit-to-Cost Ratios (BCRs) for promising relevant interventions. The paucity of good data in this field makes it difficult to arrive at accurate numbers. The authors themselves admit their BCR estimates are “extremely conjectural”.[1]

, the report finds no evidence that aid prevents civil war onset, with the one proven post-conflict intervention (UN peace-keeping operations) only being applicable in limited circumstances.

 seem more promising. For example, a program enforcing reduction of alcohol consumption in the UK reported quite a high BCR ($17 for each $1 spent) and programs to improve policing in general are also expected to have large BCRs, as they cost little to implement. Unfortunately though, the authors note that current aid programs pay little attention to rates of violent crime, focussing instead on preventing and dealing with the consequences of civil war.

(female genital mutilation, harsh physical discipline, domestic violence), the report notes that merely changing the law to make such practices illegal is unlikely to be effective. The top interventions for these issues were instead those focussed on improving education around specific issues and improving the overall quality of social services. Programs improving parenting skills in the US (BCR = 8), or increasing early response of social workers in child abuse cases in the US (BCR = 13-14), both look promising for preventing violence against children. A program to help social workers identify women at high-risk for domestic violence in the UK also looks promising due to its low rollout costs.

The authors propose the following 5 targets based on their findings:

Although these figures seem promising, we have some reservations about how well the BCRs reported for these programs (which have mostly occurred in rich countries) will translate to the developing world. Reviewers were also less optimistic about the BCRs given by Fearon and Hoeffler, with one concluding that most of the policies considered “are most likely to have modest impact with little evidence that these policies [would be] implemented in a manner that would be exceptional [in the developing world].”[13] This reviewer goes on to provide their own BCR estimates for the report’s top interventions, which range from 0 - 8 as opposed to the original 5 – 17.

BCRs for mitigating violence are in general lower and involve more uncertainty than those for interventions in other areas.

 Compare these values with those for highly effective interventions: micronutrient fortification (BCR estimates often over 20, sometimes over 200)[14], deworming (BCR = 24.7 to 41.6)[15], and subsidising malaria medication (BCR = 35)[16].

Giving opportunities for individual donors

The large total cost of violence suggests that 

violence prevention may be an area in need of funding

. However, the recommendations listed in this study are mostly aimed at policy makers rather than individual donors, and at the current time we do not know of any charities working in this area that present better opportunities than our current top charities.[6] This also applies to donating to the Copenhagen Consensus Centre itself - while we are supportive of Centre’s approach and research, we have some doubts about the centre’s research priorities, and are uncertain about the centre’s ability to significantly influence global aid policy through their research. See our full report on the CCC 

Even with doubts about the accuracy of the estimates made by Fearson and Hoeffler, the work in this paper serves to 

illuminate important areas for future research

in our efforts to help those in the developing world we need to think about unseen problems as well as very public ones

Giving What We Can Summary of Micronutrient Interventions

Worms at Work: Long-run Impacts of Child Health Gains

Copenhagen Consensus 2012 Challenge Paper on Infectious Disease

The Cost of Violence II: What can we do?

What is more costly for society – domestic violence or war? A report released recently by researchers at Oxford and Stanford University argues that, despite much greater coverage in the media, war and civil violence account for less than 5% of the total cost of violence worldwide. The greatest burden instead comes from physical violence against women and children in the home. This has implications for the funding of future interventions against violence in the developing world.

The “Conflict and Violence Assessment Paper”[1] was commissioned by the Copenhagen Consensus Centre (CCC), a think tank working to produce economic cost-benefit analyses of solutions to the world’s biggest problems. The Centre is currently focussed on evaluating proposals for the UN post-2015 Millennium Development Goals (MDGs), which will replace the current MDGs at the end of next year.

The current set of MDGs appear to have been largely successful in influencing development aid spending since they were unanimously endorsed by all UN member states in 2000. With up to $700bn of aid predicted to be spent between 2015 and 2030, the upcoming MDGs will have a role in directing very large amounts of money.

Hundreds of suggestions for post-2015 goals and specific targets have already been put forward[2] but most, although worthy, have not been subject to serious cost-benefit analysis. The CCC aims to provide Benefits-to-Cost Ratios (BCRs) for the various options on the table to help decision-makers prioritize the most effective goals.

We have been impressed by the Centre’s research in the past[3] and so far this year they’ve evaluated proposed MDGs around issues including improving education[4] and dealing with population growth.[5] If there turns out to be a high variance in the cost-effectiveness of potential MDGs (as is the case with the cost-effectiveness of ) a small investment in this kind of cause prioritization could end up doing a huge amount of good.[6]

The issue of violence has already been a focus in the post-2015 MDG discussion, with a UN panel recently highlighting major conflicts, violent crime, and violence against women and girls as high priority problems.[7] In their assessment paper, James Fearon from Stanford and Anke Hoeffler from Oxford University try to determine what kinds of violence we should try to tackle first.

The first step in their analysis is to assign a different “unit cost” to each kind of violent act. This cost represents the average amount of value destroyed by that act in dollar terms. Scaling the unit cost by GDP, and multiplying by the total number of incidents for each act gives the total cost worldwide, expressed as a monetary value or a percentage of global GDP (Table 1).

For example, the unit cost of a single homicide in the US was calculated at $9.1 million, using a model that relates increased wages to increased risk of death across different occupations.[8] Scaling this unit cost by country GDP:US GDP and multiplying by the annual number of homicides for each country[9] gives an estimated global cost of homicide of $700.5 billion USD - 0.82% of world GDP.

A similar approach was used to calculate the total cost of war and terrorism - the total number of war and terrorism-related deaths each year was multiplied by the unit cost for a homicide, and an additional cost was added based on the much larger negative externalities of war compared to homicide (namely, damage to infrastructure and relocation of displaced refugees).

Assigning a unit cost for these forms of violence is slightly more difficult, as their costs are mostly not accounted for in the wage vs. risk-of-death models used previously. Instead, unit cost was calculated by a “jury-compensation” approach that looked at the amount awarded by juries in personal injury cases.[8] Further complicating the calculations was the fact that many forms of non-fatal violence, such as sexual violence against women and children, are widely underreported in both the developed and the developing world. For example, despite the depressingly high worldwide prevalence of domestic (an estimated 30% of all women experience physical or sexual violence from intimate partners in their lifetime[10]), some studies claim that less than 5 percent of this violence is actually reported to the police.[11] Because the authors rely only on reported cases when costing sexual violence against women and children, their costings of these crimes (Table 1) are likely to be underestimates.

To avoid relying on victim reports for child physical abuse, researchers instead estimated prevalence from anonymous surveys on parenting practices.[12] From these data they estimate that 16 percent of all children regularly receive severe physical punishment (labelled Child Abuse in Table 1) , which is defined as “slapping them on the face, head or ears and beating them repeatedly with an implement such as a stick, cane or belt”.[1] Using a unit cost of $95,023 (the cost of a similarly severe assault on an adult), the global cost of child abuse was estimated at $3.59 trillion USD (Table 1).

It is important to note that the total costs reported here do not represent the actual yearly effect of these kinds of violence on global GDP. Rather, they are a way of valuing the welfare costs to the individual and society in objective terms. Another way of think about it is that if all forms of violence (total cost of ~11% GDP) were eliminated tomorrow, it would have a similar effect on human welfare to that of increasing global GDP by 11%.

Table 1. Summary of Cost Estimates Adapted from Hoeffler, Fearson 2014

Taken together, the authors estimate that the total cost of violence to the world is approximately $9.5 trillion each year, with domestic violence against women and children making up $8 trillion of that cost. The largest cost is imposed by violence perpetrated against women by their intimate partners, closely followed by physical violence against children (which is predominantly carried out by parents).

Given the relatively small cost of collective violence (including interstate and civil wars, and terrorism), this report makes a good case for increasing the attention we pay to domestic violence and child abuse, which are currently neglected in the global aid budget, and underreported in the media. As Fearson and Hoeffler put it, “[d]omestic abuse of women and children should no longer be regarded as a private matter but a public health concern.”[1] In the next post, we will look at what are the most promising interventions to reduce the cost of violence - stay tuned!

The North-South Institute’s Post-2015 Proposal Tracker

GWWC blog post on the Copenhagen Consensus Centre

Copenhagen Consensus Centre on Population

Toby Ord: The Moral Imperative Towards Cost-effectiveness

[EP: The Economic Cost of Violence Containment](I 

http://www.visionofhumanity.org/sites/default/files/The%20Economic%20Cost%20of%20Violence%20Containment.pdf

The cost of crime to society: New crime-specific estimates for policy and program evaluation

UN Office on Drugs and Crime: Statistics on homicide, assault, sexual assault and rape

Joseph O'Neill

About

Articles

The Cost of Violence II: What can we do?

The Cost of Violence I: What should our priorities be?