Terrorism in Pakistan: is this a development issue?


Listen to the podcast of this post here:

Or download it here: download (.mp3)

The recent terrorist attacks in Pakistan against the Sri Lankan cricket team bring back awful memories of the Mumbai bombings, in which it now appears beyond doubt, Pakistani operatives were also at work. The key question, however, is are there reasons why Pakistan in particular, and other poor countries in general, seem to either be the source, or target, of terrorist activities?


Is Terrorism a Product of Political Conditions?

A brief survey of the literature on terrorism gives mixed results. On the one hand, there are those who suggest that since terrorists are more likely to be educated, and relatively rich, looking to socio-economic factors is an irrelevant exercise. For these authors, the basis of a terrorist’s action is in their identity — the political, religious and fundamentally values-driven nature of their outlook on life. One such contribution (see refs. below) is that of Alberto Abadie in the American Economic Review (vol 96, No.2, pp.50-56). Instead of using terrorism attacks, he uses a measure of Terrorism Risk, as compiled by the World Markets Research Centre group, and attempts to explain this risk measure for a country based on economic, political and geographic variables.

Interestingly, a simple, first-pass model which explains the risk value for a country purely in terms of GDP per capita finds the predictable result that poorer countries are significantly more prone to terrorism. However, this result washes out (that is, it become insignificant) (though the coefficient is still negative) when a more elaborate explanatory specification is used. Here, instead of GDP per capita being important, other factors such as ‘lack of political rights’, ‘linguistic fractionalisation’ within the country, and geographic factors such as country area, elevation, and tropical area become important.

Abadie writes,

“The empirical results reported here show that terrorist risk is not significantly higher for poorer countries, once the effects of other country-specific characteristics, such as the level of political freedom, are taken into account. In contrast with the results for civil wars in Collier and Hoeffler (2004), lack of political freedom is shown to explain terrorism, and it does so in a nonmonotonic way. Countries with intermediate levels of political freedom are shown to be more prone to terrorism than countries with high levels of political freedom or countries with highly authoritarian regimes.” [p.51, Abadie, 2006]

To put that snippet into less technical words, Abadie is noting that poverty (as measured by GDP/cap) or more broadly under-development (as measured by the UN’s Human Development Index) are not key ‘risk factors’ for terrorism in a country. Instead, lack of political freedoms of a particular kind are. The particular kind he mentions are those that are ‘intermediate’ in the political freedoms measure. This measure appears to relate to terrorism risk ‘non-monotonically’ which means ‘not as an increasing function’. The function that comes out of the analysis is pictured below — a combination of a linear component which gives increasing terrorism risk with declining political freedoms, but then the function turns south, and shows that the risk declines with further deterioation in political freedoms.


Reproduced from Abadie, 2006 (see ref below)

That is, countries in transition from authoritarian regimes to democratic institutions are particularly at risk from terrorism. This seems to match the annecdotal experience (consider Iraq and Northern Ireland).

So where does Pakistan fit on this measure? According to the Freedom House index (used by Abadie in the paper mentioned above), which gives a value for political rights from 1 (most free) to 7 (least free) Pakistan gets a value of 4 and is classed as ‘partially-free’ by Freedom House. To put this number in perspective, North Korea and Zimbabwe (to name two) are at the bottom of the political rights register with a reading of 7 (out of 7) and are both classed as ‘not-free’ by Freedom House. For the record, Australia is at 1 (free), sharing this value with Austria, Belgium and Canada (to name just a few near the top of the alphabet).

Anecdotally, this measure rightly reflects the struggle that Pakistan is presently going through to establish free and fair public institutions. Although in principle, elections in Pakistan are free, in practice, many observers point to vote-buying, branch stacking and intimidation in several provinces.

So we may ask the secondary question — why is it that countries in transition are prone to terrorism? One can think of two reasons immediately. First, a very low level of political rights usually means that an authoritarian regime is at the helm (e.g. Iraq, pre-US invasion), and in this context, terrorism activities (attacks on the state) are likely to be harshly cut off before it can even begin. Second, when the political process is still in a state of flux, terrorist activity has most to gain from new-found, fledgling political freedoms. In some cases, this is even explicit, as a political entity might have a formal, political wing, coupled to an informal, militant wing. This was true of the IRA in Northern Ireland (the military wing has been disbanded) and even Nelson Mandela’s African National Congress (ANC) party started life with a strong militant component.

But why would a party ‘arm-up’ in the midst of a political struggle? When in transition, it is far more likely that corruption, bribery and an over-eager connection between the military and current government are features of the present political land-scape. In which case, alternative political parties might find themselves starved of oxygen to make their views known, and under increasing frustration, will take the desperate means of attacking government posts such as the police and military to make their point. Furthermore, a desperate political movement may also find themselves ‘rushing to the bottom’ of pseudo-political tactics, by engaging in their own pressuring of the population whether explicitly or implicitly through threats and demonstrations of harmful intent.

So this perspective helps to understand the present Pakistani plight — with a democratic government struggling to find its feet in the wake of the essentially dictatorial government of former President Mushareef, the terrorist spectre has arisen with impecable and brutal timing.

It is worth remembering the remarkable work of (now) Prime Minister Morgan Tsvangari in Zimbabwe. He has lead a peaceful, non-confrontational, and startelingly patient political movement for change in Zimbabwe from Mugabe’s selfish, cruel and cold regime. So many times have his party members, his supporters, his family and even himself been attacked by Mugabe’s henchmen, and yet, Tsvangiri has returned with calm words of condemnation and peaceful resolve.

But does Poverty have any Role to Play?

A second line of inquiry focuses instead on the socio-economic context that gives rise to the terrorists themselves. For instance Freytag et al. (see ref below) use a cross-country methodology to explain the country of origin of those involved in actual terrorist activities. The authors use the Terrorist Knowledge Base of the MIPT organisation from 1971-2005 as the source of these incidents. Here, the authors advance an economic argument that it is opportunity cost considerations that lead to a person becoming radicalised and engaging in terrorism. To clinch this argument, the authors need to connect a few points. First, they require us to believe that engaging in terrorism gives some kind of utility itself, that is, that terrorism can be consumed like another good, and thus is a possible substitute for other forms of consumption. In other words, we need to believe that if economic circumstances are limited in a country, then the population will engage in terrorism for purely utliitarian purposes, thus bumping up their summed utility.

At first pass, this seems to me to be a difficult argument to swallow. First, how does one explain (economically) a suicide-bomber’s decision? Certainly, the present value of all their future sources of utility will be severly cut-short with a ’successful’ detonation of their device. Thus, we must put a value on an assumed infinite (eternal) stream of positive utility that will be enjoyed in the afterlife. Whilst I’m not adverse to this kind of thinking, it is quite a contentious argument to put. Second, if the argument were true, then we should see an extremely strong correlation between limited economic opportunity and terrorist behaviour. Furthermore, assuming the principal of universality, we should expect that for thousands of years prior to the industrial revolution around 1800, the poverty of all countries would have lead to many terrorist activities. Was this a feature of pre-industrial era England? Probably not.

In any case, the authors of this study consider many variables that may bear on opportunity cost: openness to trade could be positively correlated with terrorism where trade is seen as a threat, whereas, if it is seen as an economic opportunity, trade should be negatively correlated with terrorism; higher GDP per capita should increase opportunity costs; high tax rates (consumption of GDP by the government) should increase opportunity cost since investments are likely in education, health and infrastructure; education of the population should increase opportunity costs; and good institutions (measured by the Fraser index of Economic freedom should lead to higher opportunity cost of terrorism .. and so on.

So does poverty lead to the production of active terrorists? The GDP per capita data have a similar shape to the Political freedom variable used in the study above. That is, terrorists are most prevelant in coutnries with a middle level of incomes, afterwhich, higher wealth leads to their suppression. One can intrepret this as the supply and demand side of terrorist activities. On the supply side, terrorism is undoubtedly expensive, and so higher incomes will lead to a bigger potential pool of income for terrorists, assuming a constant (small) fraction of the popualtion who would send money their way. On the demand side, however, with very high incomes, economic activity is at such a rate, that to engage one’s time in training for, and carrying out potentially harmful terrorist acts, carries a very high opportunity cost, thus supporting the overall hypothesis.

To add nuance to the findings, the authors split the data into Islamic, European and OECD countries. They find mixed results for their opportunity cost hypothesis. A few highlights include: human capital attainment (education rate) results in less radicalisation in OECD and especially Islamic countries, where the effect is highest; population pressure significantly increases the propensity for terrorism, although a large, but ageing population reduces terrorism activity significantly. Interestingly, the Economic Freedom index is significantly protective from terrorist activity for ‘free’ countries, especially in Europe, but higher economic freedoms are associated with terrorist activity in Islamic countries.

So finally, to Pakistan one more time. In terms of incomes and human capital, Pakistan is ranked 128th by GDP per capita (at PPP US$2,370) and has only 50% adult literacy rate. Based on the 2006 figures for the Economic Freedom index, Pakistan ranked at 104 out of 141 countries. Pakistan’s index value of 6.05 puts it alongside that of Russia, Sri Lanka, Bosnia & Herzegovina and Madagascar. Australia ranks 8th and Zimbabwe ranks last (141st) with index values of 8.04 and 2.67 respectively. Furthermore, Pakistan ranks 139 on the UNDP’s 2006 Human Development Index ranking, just infront of Mauritania and just behind Yemen. They get to this ranking by recoding a life expectancy at birth of 64.9 years. Put into perspective, Australia’s life expectancy at birth took approximately the same value in 1930.

Summing Up

So it would seem, that unfortunately, by either the political transition, or poverty and opportunity cost argument, Pakistan is an unsurprising candidate for terrorist activity. It should be noted that this is both as a ‘victim’ of terrorism (due to its present political instability), something that the present government of Pakistan has been quick to point out, especially to the disgruntled Indian government, and as a supplier of terrorists (due to its low level of development, literacy and Economic Freedoms). The latter point is particularly emphasised by the US government who sees the tribal regions of Pakistan that border onto Afghanistan as a most likely source of mobile, radicalised terrorists. Either way, by the numbers, the ongoing fingerprints of terrorism in Pakistan will continue to be a feature for years to come. And it is worth noting briefly that none of this has so far mentioned the nuclear weapons status of Pakistan. No wonder Obama et al. see Pakistan as a major security priority for the present administration.

Refs used in the preparation of this post

  1. Abadi, A. (2006), “Poverty, Political Freedom, and the Roots of Terrorism”, [UN Working-paper version, PDF]
  2. Freytag et al. (2008), “The Origins of Terrorism” [PDF]
  3. UNDP’s Human Development data sheet for Pakistan
  4. Freedom House Political Freedom Index data access

Other refs

  1. A BBC report on the release of Pakistan’s ‘disgraced nuclear scientist Abdul Qadeer Khan’
  2. A brief (unofficial) history of nuclear weapons in Pakistan
  3. Pakistan nuclear regulatory authority

, ,

Comments are closed.