On Genocide , Economic Reasons vs . Ethnic Passion

The traditional vision of genocide is exogenous. In this framework, ethnies have a real sense. The economic approach of conflicts has expressed slight differences in the relation between ethnies and conflicts. However it does not reject this explanation. Here we propose an alternative approach, an endogenous vision of genocide. Genocide appears in society where social capital plays a major role in solidarities. But social capital is a weak asset in the individual portfolio. Economic and social shocks may have impacts on the assets structure and may produce conflicts such as genocide. In this new framework, policy makers may have to adopt prudential rules.


Introduction
Genocide means killing anyone for genetic reasons and race.But generally it has a social sense, concerning extermination of millions of people for "hating man as a man" (Ballet et al. 2007).In human history, given the growing population, genocide concerns more and more people and can be measured in millions (6 in Europe in the 40's, 2 in Cambodia, 1 in Armenia, 1 in Rwanda/Burundi).Genocide is an absolute estimation; ethnocide considers the death of a large proportion of an ethnic group as in colonisation by Europeans.It is a moral tragedy for civilisation and in all the cases, an economic fact with huge material and human consequences.
The main consideration of this phenomenon has been socio-ethnic.In Africa the main reasons are ethnic and especially ethnic fragmentation since Easterly (2008).Furthermore, ethnic conflicts, in this recent economic literature presuppose an "objectivity" of ethnic groups.This realism is denied by modern ethnology; ethnic group being a construction, made sometimes by the "genociders" themselves, not a reality.This is an argument of the historian Mark Levene (1999) emphazing the difficulties of the genociders to delimit the populations to kill.His explanation of genocides is political, as an act of a desperate regime.
Religious, political, ethnical factors are omnipresent as if in these circumstances, universal behaviour had disappeared.More generally, in Africa as in Europe, genocide is a matter of "anthropology", between "appolinian" and "dionisian" populations (Benedict (2005); Mead (2000)), thesis renewed about the Shoah by Goldhagen (1997).Economic analysis has proposed an exogenous theory of ethnocide and ethny.We oppose endogenous theory of ethnocide and conflicts to an objective delimitation of ethnicity.
In the exogenous version, Africa is poor because it is unstable and its instability derives from its ethnic complexity.Conversely we assume that Africa is unstable because it is poor.Social capital becomes a critical asset in these countries.When it is intense and resources are rare, any economic or social shock may transform benevolence into malevolence, passing through ethnic recreation.
In this context economic development policy is not at fault and new prudential considerations must be advanced about economic policy development.

The exogenous vision of ethnocide and ethny versus the endogenous vision of genocide.
In the exogenous vision of ethnocide and ethny, economic fragmentation is postulated as if ethnic groups had a real sense.For instance, Easterly (1997Easterly ( , 2008) ) has constructed an indicator of ethnic fragmentation on the basis of data published by soviet geographers.
So an indicator may be built with the number of ethnic groups and the population of each ethnic group.Then regressions may be done between the number of ethnic groups and the economic performance.For instance according to the above authors, ethnic diversity explains relatively the poor economic performances of Africa.The economic approach of conflicts (Collier and Hoeffler 2008) has expressed faint differences in relations between ethnies and conflicts.For instance, as quoted by Bates (2008), there is a non-monotonic relation.Why are Rwanda or Burundi with two supposed ethnic groups so poor and why was Côte d'Ivoire (Ivory Coast) with 80 ethnic groups able to be a model for Africa?This can be interpreted in a dynamic way, the performance of Côte d'Ivoire in the 70's without ethnic confrontation, the crisis of the 90's being interpreted by ethnic confrontation.
These results are controversial, first on the reality of ethnic groups.Where are the ethnic groups in France?Populations of the underdeveloped coun-tries are not primitive populations as a field of ethnology.Ethnicity is an interpersonal representation changing with the context.What does ethnicity mean in the mega towns and with a large crossbreeding?An ethnic re-creation may take place with economic difficulties and a xenophobic State policy in such a case that the mentality of representations change.For instance the popular representation of Jews in France may change at any moment: they can be seen first as ordinary French people or first by their community reference.
In this last case, the social preference is "malevolent altruism" to externalise frustrations against others and to prefer the disappearance of others by migration even by death.This allows a better density, diminishes the struggle for life, the social recurrent costs for others, the access to the best professions (See for instance, the 18 th October 1940 law in France).In some cases this is the condition of a new social equilibrium with "new" property rights in each party.Genocide is an important way of wealth reallocations.
Genocide begins with an identity research, "a person's sense of self" (Nyborg, Howarth and Brekke (2006) and Akerlof and Kranton (2000)) into a social context, seeing oneself through the eyes of the others.For instance, characteristics may be developed by created signs (yellow star in Europe, blue scarf in Cambodia), imagined characteristics (height as in Rwanda/Burundi), religious origins, differentiated housing (ghettoisation).There is a "re-creation" of identity, managed by the State to marginalize and exclude.From a political point of view, State may be weakened by external shocks (military defeats, wide liberalisation) and has to be restored by nationalism.
So, there is a strong relation between identity and genocide phenomena, interaction rationalised and managed by the State.The economic precondition is a weak economic tissue and a shock of economic policy.These situations correspond to Germany in the 1930's, Turkey in 1915, Rwanda/Burundi in the 1990's, West Africa in the beginning of the 2000's.The ethnic point of view is not immediate but begins when the economic crisis occurs, all the capacities being used and beginning to be destroyed.The economic pre-situations may be very different: huge unemployment and inflation (Germany), high population density on poor lands (Rwanda, Burundi), low returns on expanding rental production (West Africa).
The external shock may occur from markets and from policies.For instance, policy remedies may be strong as liberalisation of economic institutions (marketing boards, credit), big incentives to produce "tradables", money injections and growing debt.These strong economic and social policies may be unsustainable and destroy social networks and personal capacities.
Social crime as genocide appears as the result of an anomic society and of an external shock, especially unsustainable socio-economic conditionalility imposed by institutions.It specially takes place in societies where social capital plays a major role in solidarities.Narayan and Pritchett (1999) mean by social capital "the quantity and quality of associational life and the related social norms".Their survey asks individuals about their membership in voluntary associations or groups, the quality "inclusive" or "exclusive" of group's membership.Therefore, the index of village associational life is a proxy for social capital.There is a preconception about the beneficial effects of social capital.A high index of sociability implies higher incomes, better economic and educational performances.Grootaert and Van Bastelaer (2002) follows this assertion, arguing that social capital has long-term benefits, especially for poor households, such as better access to credit and a resulting better ability to smoothen out income fluctuations by borrowing and/or accumulating assets.According to Fafchamps and Minten (2002), in Madagascar "individuals with better relationship capital can conduct business with each other in a safer and cheaper manner".

Social capital as weak asset
Social capital is greater or less according to societies and their economies.It is seen strongly in certain poor societies, notably in informal urban production and in rural activity.It manifests itself more visibly by important economic transfers where social protection remains essentially private and founded on intergenerational relations.
Social capital plays a great part in poor societies, where transfers or pluriactivity are the main components of the capability to survive.But this huge part of social capital in the assets is explosive.This social capital is volatile and very fragile.Under certain assumptions we can consider that social capital as a weak asset in the individual portfolio.Every individual has not only physical, financial and human capital, but also social capital.The latter may be predominant in poor societies.
First assumption is an inverse relation between individual revenue and risk of conflicts.In this case, poverty appears as strong determinant of conflicts.Collier and Hoeffler (2008) confirm this assumption.Second assumption is an inverse relation between State's redistribution and risk of conflicts.We may assume that government redistribution toward the population, without negative discrimination, ensures a minimal social stability and reduces risk of conflicts.In both hypotheses risk must be interpreted as a probability that a conflict may be brought about.Third assumption is an inverse relation between private solidarity and State's solidarity.Private solidarity is a part of social capital, and then we have an inverse relation between social capital and State redistribution.If individuals can't have grants from government, they express a demand of solidarity in their private network.
Figure 1 represents all the society feasible configurations at a moment of time.Quadrant I relates feasible structures concerning individual assets.We distinguish between social capital and others assets, represented here by the private revenue Yi.Point A indicates individuals' optimisation on the assets structure, with the curve representing individuals' preferences and the inverse relation the feasible structures in the society.The angle of the straight line, passing through A and starting from the axis origin, corresponds to a proportion, here named a, indicating the relative parts of social capital and others assets in the individual assets portfolio.Quadrant II links risk of conflicts to the level of revenue.When the revenue increase risk of conflicts decrease.It is our second assumption.Quadrant III represents our second assumption which draws the inverse relation between State's redistribution and risk of conflicts.Finally, Quadrant IV represents the relation between government solidarity and private solidarity or social capital.
We have to make a clear difference between all the feasible structure of the society at a moment of time and the situation of this society.Feasible structures depend on economic, social and cultural determinants whereas the situation at a moment of time reflects the state of social tensions, redistribution mechanisms and possibilities of production.On figure 1, the polygon drawn by a heavy line represents all the feasible structures of the society.The dotted line indicates the situation of this society.It depends on individuals' choices on their assets portfolio.This set of society relations or configurations allows us to examine effects of economic policies on risk of conflicts.Figure 2 analyses consequential effects of a shock on revenues.The initial situation is at point A1, with a risk of conflicts r1.An economic shock on revenues, decreasing them, may imply two cases.In the first case, the shock has no effect upon the assets structure.It only induces a decrease of private revenue Yi.Individuals optimisation moves from A1 to A2, with a new risk of conflicts r2 higher than r1.We may interpret this case as the consequence of poverty increase.In the second case, the assets structure changes.The optimisation is now on point A3, with a risk of conflicts r3 higher than r1 and r2.The meaning of this second case may be linked to an ethnic or community re-creation.When the private revenue decreases, individuals search and create private solidarity centralised on a community or an ethnic group even if this ethnic group must be re-created.In consequence, opposition between communities or ethnies may be developed and the risk of conflicts between them is increasing.tation may be the following.As social capital is developing, it ensures the valorization of others forms of capital, and implies an increase of revenues.As poverty is diminished risk of conflicts is weaker.The policy has a beneficial effect.In the second case, policy induces a change in the assets structure.Individual optimisation move from A1 to A3, with a new risk of conflicts r3 higher than r1.
Here, policy has a negative effect.We may understand this effect as the consequence of the policy incentive.As policy valorizes social capital, individuals are encouraged to invest in this form of capital.They may do it by a re-creation of communities and ethnies.Private income decreases whereas social income increases.As the matter of fact, as ethnies are re-created, risk of conflicts is higher.government cuts in its expenditures, it develops more clientelism increasing conflicts between groups, those who benefit from the redistribution and those who are excluded.In the second case, individuals change their assets structure, reacting to the policy by a higher use of private solidarity.Then, the situation is now on point A3 with more social capital.It corresponds to a higher risk of conflicts.Because individuals need private solidarity to compensate the cut in government grants, social capital increases.Meanwhile, as creation or re-creation of communities or ethnies is a means to obtain social income, opposition between ethnies or communities may be developed.

Figure 4: Redistribution policy
In fact, we may assume there is a volatility of human behaviour in distress as there are panics on the markets.This volatility concerns social capital, a predominant factor in the poors' portfolios.Therefore, the structure of this personal portfolio changes by the sign of the social capital.But there is also in most of the cases a structural modification in this structure or a sub-structure as an allocation of time.For instance, in poor rural pluriactivity, allocation of time between embedded activities is very weak and it is the capacity to survive.Any change may cause migration or conflicts (exit/voice).Migrations to avoid conflicts with A perturbation of the personal asset structure is accompanied by changes in social capital.If the disturbance of the capacity to survive is too high, this is a source of malevolence and implies a calculus about social crime, reinforced by expectations on other calculus.In this case, self expectation of other malevolence induces either automigration or murder, according to the forces in action.In another sense, migration depends on the probability of being murdered and the probability of murdering according to a presupposed identity.

Policies implications
Genocide is a case of extreme destructuration following a destabilisation of personal assets.This implies a stability hypothesis on the structure of assets, in this case of social capital/other assets considered by the real income Yi/p.The capital social amount depends on the social effort multiplier as an amount of money.The policy implication is to stabilize this structure and to avoid a social "big push" according to a social precaution principle: the supply of social resources must not be superior to a corresponding demand which is constant in the short term.Supply of social resources appeared with the social dimension of adjustment policies of the World Bank (social funds), high labour intensity programs of ILO, and more recently with Social Capital programs.
Each person holds a composite capital with very different assets: economic, social, human, cultural, etc.In the case of the poors, social capital plays a major role in these assets, money or finance being scarce by definition.Social capital allows informal activities or transfers which are the main capabilities in case of monetary poverty.But social capital is a potential resource which implies a social effort or transmission (for instance repeated visits to many persons, multiplication of social links, spreading information and social signs).This social effort or social multiplier (in other terms, the velocity of socialisation) is called here, s.
The key ratio, as shown in the last figures, is Social Capital (S), related to the other real resources as p.Yi.This ratio, S/pYi, expresses the state of "real" socialisation as 1/s or k.
This is a representation of the personal asset structure, and k = 1, is constant in the short term (an old monetarist assumption).
If s, the social multiplier is equal to 1, this shows a strong social interaction.In another manner, this strong social interaction may be written: ΔS.s = Δ p. Yi. with s = 1 (2) Social capital (absolute) variation is equal to real assets (absolute) variation.Social capital may be considered as a complement of individual goods and a constraint on every utility function (Becker, 2000).
In this case, s is constant in the short term as a structural relation, the capital or asset structure being not modified.1/s is very important because social efforts are expected to return "social" and then individual incomes.This return on social investment, as shown by Becker (1974), may be negative and this may be the consequence of envy or hatred (malevolence or negative altruism).To simplify, we will take malevolence as a situation where the relationship dUi/dUj is negative.This relationship is known as the coefficient of altruism indicated as α.This coefficient may be positive (benevolence), negative (malevolence) or nul (neutrality) and may be analysed for one or more persons, and their environment.
The nature of altruism is the key variable of the society concerned.For instance, a negative value (malevolence) of altruism coefficient α may be associated to a lack or an excessive development of Ss.In the case of genocide, there is an intensive social capital in a society for a group and against the other, a high value of the social multiplier s.A group with intensive positive social capital and high resources may fragment between positive and negative social capital according to the divided groups.
If s is important, policies may develop adjustment policies to improve pYi and then the capital structure changes and α may change also.A socialisation of the poor may be malevolent, directed against a common enemy as in Germany in the 1930's and in most of the cases of genocide: s grows and the altruism coefficient becomes negative.Might this socialisation be benevolent as projected by the development planners?If we refer to the identity above, a policy directed on the components of individual assets (Yi) has important consequences on socialisation.This is the problem of the omission of social effects in the framework of structural adjustment policies and of their potential responsibility about conflicts.
For many poor people 1/s is an optimal ratio, preferred to others, taking in account the monetary constraints.In many situations, the main objective would be a stabilization of ΔS.s. on one hand and of p.Yi on the other, or to elaborate a simulation of the social disagreements linked to the adjustment policies on S/pYi.
When social capital Ss, is intense and other resources are rare, any economic or social shock may transform benevolence into malevolence.This difficulty compels the analyst to be very careful when enlarging the notion of capital.Though social capital is heterogeneous and therefore may not give results in one particular direction.The specificity of social capital for an economist is due to the fact that it may have negative components, in particular, accumulated malevolence, as well as benevolence.
For instance, the multi-altitude societies (see Murra (1981)) of the great Africa lakes region observed a socio-economic equilibrium such as a dense population has a sufficient productivity on pluri-activity to survive without assistance.The determinants of life are very weak allocations of time and so real exchange rates between activities.In a small enclosed economy as Burundi or Rwanda, formal and informal productions are the two sectors of the production possibility surface.There is a positive relation between formal activities and intrarural migration, up to a limit where migration is accelerated according to growing disequilibrium between activities and the peculiar terrace cultivation.Conflicts are unavoidable in such a context and take the well-known ethnic form, for instance between "smalls" (Hutus) and "talls" (Tutsis).The labour intensive programs, as imagined by experts, are more efficient on ethnic conflicts than on economic development.Then underdevelopment lies in a primary overvaluation of tradable and more generally of formal activities.

Conclusion
The development policy is based on "big push", "structural adjustment", a variety of external shocks not only on the nation's assets and activities structure, but also on everyone's economic structure.The social consequences of these plans may be very different according to very complex configurations.In poor economies context relation between social capital, other personal assets and conflicts depend on structural components such as k, but also on conjonctural elements as social transmission, s and altruism.Genocide is a special case of altruism where killing others is the only way to survive in a social context where S, k and s are important, α extremely volatile.
The social uncertainty linked to Ss imply not to destabilize the social structure; stabilization is not only a monetarist imperative against inflation and deficits, it must be enlarged to social disagreements (Kanbur 2008), such as genocide.A social stabilization policy must improve a prudential principle where the incertainty linked to social sustainability is strong: the market cannot compensate human massacres.The dead will never come back…

Figure
Figure 1: Society structures Social Capital

Figure 3 :
Figure4represents a change in the redistributive government policy, precisely a cut in the redistribution.On this figure two cases appear also.In the first case, individuals can't use social capital.The redistribution cut produces a change from A1 to A2. Risk of conflicts is higher, increasing from r1 to r2.Two interpretations may be made in this case.First, as government expenditures decrease, poverty increases, and risk of conflicts linked to poverty is higher.Second, as high population density create new conflicts in new areas.And maybe population changes without any control.