Quite interestingly, a Human Security Index (HSI)--or conversely Human Insecurity Index (HII)--has not yet been developed and released. David Carment and colleagues at Carleton University have an index of human security risk based on a country's number of resource/territorial disputes, armed forces per 1000 individuals in the population, and military expenditures as a share of GDP (http://www.carleton.ca/~dcarment/presents/cifp/sld036.htm). That is too narrow a measure of human security, however. Steve Lonergan and Kent Gustavson at the University of Victoria have done work on developing an index of human insecurity (IHI), but that work has seemingly stalled since 1999, as most of the emphasis has been on environmental security (http://office.geog.uvic.ca/dept/faculty/lonergan/). The Human Development Report (HDR) contains a section called "Ensuring Human Security" that supplies information on a number of variables, but closer inspection of those variables reveals them to be helpful but inadequate as a measure of human security. A HSI would undoubtedly use some of the variables in the HDR, but it would need to go beyond what is found in the HDR in order to have credibility.

A paper from the Harvard University Program on Humanitarian Policy and Conflict Research and the United Nations University, Peace and Governance Program entitled "Security with a Human Face: A Proposal to Creat a Human Security Report" They want to make a report analogous in design and hopefully impact to the HDR. The Human Security Report (HSR) would contain a Human Insecurity Index (HII) that would be "based on indicators as the number of deaths from armed conflict, the incidence of criminal violence, refugee numbers, etc." The HSR "would not seek to develop new datasets, at least not initially, but would rely on extant data sources." The Harvard people (Andrew Mack, the person I know, is now at the University of British Columbia) intend to focus on a relatively narrow definition of human security with violence, either armed conflict or criminal violence, being the critical factor.

In a pilot phase that entailed making a "beta" issue of HSR, the Harvard people propose the following for their HII:
- deaths and injuries arising from war, genocide, or other gross violation of human rights
- data on criminal violence
- data on other human costs resulting from war or criminal violence, e.g. Deterioration of various health indices
- refugee and IDP (internally displaced persons) flows
- damage to societal infrastructure associated with violent conflict
- data on trafficking in persons

This list is rather vague and probably too narrow, but I assume they have made progress from the paper, which was done last fall. I think the HSI outlined below is more sophisticated than the HII, but it may not be at this point.

For me the key question is how to design and construct a HSI that can contribute to conflict early warning. Let us first examine the design issues. An announcement for International Week 2000 (http://www.international.ualberta.ca/iweek/iweek2k/phs.html) argues that human security has two main aspects:

1. Safety from such chronic threats as hunger, disease and repression.

2. Protection from sudden and hurtful disruptions in the patterns of daily life--whether in homes, in jobs or communities.

Moreover, the announcement asserts that the threats to human security fall into seven main categories:

1. Economic insecurity.
2. Food insecurity.
3. Health insecurity.
4. Personal insecurity.
5. Environmental insecurity.
6. Community and cultural insecurity.
7. Political insecurity.

This is not a bad start, although I think environmental insecurity could be subsumed under health insecurity, especially for the purpose of an index pointing to an emerging conflict situation. Let us go one step further and define what could be the components of a HSI based on each of the (remaining) six items. There are many, many possible components of a HSI. The following have been selected because they are comparatively straightforward and feasible to collect, although the emphasis must be on comparatively:

Economic security
- a measure of the per capita income of the country (ideally in Purchasing Power Parity terms)
- a measure of the income or wealth inequality in the country with a focus on the income or wealth of the poorest 20% of the population
- a (true) measure of unemployment
- a measure of the (consumer) inflation rate
- a measure of the extent of a social safety net or welfare system for the elderly and poor

Food security
- a measure of the number of people suffering malnutrition
- a measure of the rate of change of food prices

Health security
- a measure of the availability of health care for the poorest 20% of the population
- a measure of the number of serious illnesses and deaths caused by environmental degradation as a share of the population

Personal security
- the share of the population killed in violent conflict (including massacres/genocide)
- the share of the population turned into refugees or internally displaced persons
- a measure of the likelihood of being the victim of a violent crime (including robbery)
- a measure of the disparity between men in the dominant social group and members of other social groupings such as women, gays, or members of minorities with respect to being a victim of violence

Political security
- a measure of civil liberties and political rights (Freedom House?)
- a measure of the share of the population incarcerated for political crimes
- a measure of the disparity between men in the dominant social group and members of other social groupings with respect to being incarcerated
- a measure of the disparity between men in the dominant social group and members of other social groupings with respect to being abused or killed by government (or government endorsed) security forces

Community and Cultural security
- a measure of the degree to which a country is shifting away from the existing distribution of what languages are spoken or what religions are practiced
- a measure of the degree by which government expenditures are skewed away from a geographic distribution based on population (perhaps controlling for the wealth of the regions)
- a measure of the degree of the freedom to worship

In my opinion, if one could assemble these measures, one would have a pretty good handle on the security or insecurity felt by individuals and groups in a country. Of course, I am open to any alterations/deletions/additions, but at least these measures are a starting point.

For more developed countries, there are some additional indicators that are possibly more relevant in terms of the security/insecurity that the citizens feel:

1. Measure of stock market volatility
2. Direction of stock market index
3. Percent of population that trusts the leader of the government
4. Transparency of economic transactions and political decision making
5. Measure of the state of "essential" services in terms of reliability as well as of problems being fixed

Two more measures perhaps worth mentioning are:

x. A measure of political polarization
y. Ubiquity of firearms in population

They might be useful, but the first may be especially hard to measure, and the impact of the second one is probably dependent on the values of other measures.

If the preceding items are deemed acceptable as measures of human security or insecurity, the next question is can data regarding them be gathered for a large number of countries and/or without heroic effort. I do not have the answer at this point.