Category Archives: Institutions I (Aid)

AID EFFECTIVENESS

AID EFFECTIVENESS: PUTTING RESULTS AT THE FOREFRONT

Franck Wiebe

Millennium Challenge Corporation

A Summary 

In a Nutshell

This is an example of donor policy that undermines domestic institutions.

The cost benefit analysis model used to evaluate projects that the corporation undertake, takes into account on the benefit side, only “changes in local income”. This means that institutional learning could not be a relevant benefit. In other words promoting domestic capacity for autonomy is not taken into account when evaluating projects.

 This means that in evaluating projects one that bypasses domestic policy makers in order to increase the effectiveness of the programme would be considered preferable to those than engage with domestic institutions at the risk of a smaller increase in local income due to governmental inefficient and lack of experience. This undercuts accountability and fundamentally alters, or rather, prevents, a state-citizen relationship based on accountability forming.

 Additionally, the recipient must report to the MCC every quarter. Given that there could be any number of projects in a given year, and a vast array of donors all making similar demands the implication is that donors may get swamped. This is inefficiency from the fragmentation of donor activities that would be partially solved by the Sachs plan of concentrating all donor activities in the UN.

Advertisements

DONORS AND SERVICE DELIVERY

DONORS AND SERVICE DELIVERY

R. Reinikka

A Summary 

In a Nutshell

Donors supply 40% of public resources in 30 countries. Like the previous summary this paper suggests that aid to influence institutional reform can undercut domestic accountability suggesting that donors should:

  1. Strengthen relationships with policymakers as circumvention undermines service delivery
  2. Integrate support into the recipient’s strategy where possible

 Accountability

  • The geographical and political distance between the taxpayers that fund donors and the recipients of the services donors provide breaks the traditional feedback loop in service delivery [restricting voice with an impossibility of exit]. If there was some client feedback it would not work in the same way as domestic accountability.
  • Donors often go straight to the service provider thus sidestepping the policy maker which reduces incentives to good government and institutional learning. This lack of contact with the government can mean that donors fund projects that governments are not interested in and so the project will not integrate well with whatever development strategy the recipient government may have in place. Additionally, donors may fund a sector which was due to receive government funds, thus increasing the fungibility of funds within the budget and allow for e.g. increased military spending.
  • Some projects run by the World Bank specifically bypass domestic organizations by setting up semiautonomous implementation units.

 Voice

  • Donors imposed conditions in an effort to augment weak domestic voice. But this is a poor replacement, and has rarely been seen to work as intended as well as undermining policy ownership. When policymakers do not have to create and legitimize their own viewpoints it becomes easy for them to deny responsibility for the outcomes.
  • This is dealt with to a degree by the PRS strategy of Sachs as it encourages public participation in the poverty strategy, although there are many challenges with effectively managing this type of participation.

 Implications

  • Where state systems are weak they need to be engaged not bypassed in order that they might learn to meet best practice standards. This is essential for the effectiveness of future aid, which has been shown to be more effectively used in countries with good institutions (Dollar & Burnside).
  • There should be a strengthening of domestic institutions even if projects become riskier in terms of measurable outcomes by using domestic capacity

AN AID-INSTITUTIONS PARADOX?

AN AID-INSTITUTIONS PARADOX? A REVIEW ESSAY ON AID DEPENDENCY AND STATE BUILDING IN SUB-SAHARAN AFRICA

Moss, Pettersson & N. van de Walle

 A Summary

In a Nutshell

Aid was traditionally used, in part, as a means of building good quality domestic institutions through conditionality and targeted spending. This approach underlies the capacity building and the “big push” system of aid for improving public infrastructure and institutions [e.g. Sachs]. The idea behind the big push is a compact between the African states and the West whereby African states agree to devote more resources to ending poverty, and in return the West provides a huge increase in aid. The idea is that the large increases can leverage commitment to improve the institutions important for growth and poverty reduction.

 However, there is now much agreement that the neopatrimonial structure of African public life prevents the development of the type of institutions that are good for growth. Resources are not used to promote development, rather the political elites act in a predatory fashion to maintain themselves in power. Additionally it appears that certain types of aid could in fact reduce institutional development. At high levels over a sustained period aid can have distorting effects on some of the very outcome the donors wish to encourage such as policy ownership, institutional development and autonomous growth. Although it is possible that the lack of development we have seen is also attributable to the political instability and periodic violence that affects the region we still need to question why the large volume of aid has achieved few results in improving the institutions of the region.

 Macro Outcomes

  • Dutch disease – large amounts of aid can discourage domestic industrialization by undermining competitiveness (although this can be avoided by exchange rate management by a competent central bank).
  • Budget – aid volatility can prevent governments from budgeting effectively. Additionally the aid can serve as a soft budget constraint on policy makers which encourages the view that budgets are flexible thus encouraging fiscal indiscipline. Evidence of this comes in the form of unsustainable government consumption.
  • The results of these effects can be slower growth, fewer jobs, macro instability, inflation from deficit spending and deterioration of the quality of public servants, all of which can contribute to an inability to move from patrimonialism to a more developmental path.

 Institutional Change

  • Patrimonialism – donor projects allow for patrimonial behaviour. Scholarships, cars, well paid seminar days etc. all undermine the transition to good public institutions. In states where power means access to state privileges and political systems are supported by complex networks of clients, aid is politicized often in place of paying a regular salary. Thus aid reduces institutional development and increases government consumption rather than public investment.
  • Revenues – the system of taxation reflects all other public institutions and collecting revenue from citizens is one of the primary ways of building accountability. Some even argue that the key determination of whether a state will transfer from aid dependency to economic self-sufficiency is whether the state learns to tax. Yet aid reduces the incentive to tax as it is unearned income. This means the institutions of taxation will not develop. It is suggested that grants are seen as free substitute for tax revenues and so reduce taxation, but loans are less de-incentivizing as they must be repaid.
  • State – aid can affect the evolution of state/society relations by undercutting ownership, accountability and participation. Elites do not need public approval for policy as the public do not fund it, and it is easier to submit to donor demands than build an autonomous tax structure. A reliance on aid as a substitute for local resources means the flow of revenues to a government is not related to government efficiency so there will be a tendency to underinvest in developmental capacity. This moral hazard reduces the incentive to adopt good polices and reform inefficient institutions.
  • Democracy – governments do not have to worry about maintaining legitimacy as they do not collect revenues from the population. Thus more unearned income is a cause of weaker democracy. This has been somewhat moderated by the conditions attached to aid, but they are rarely enforced and are a poor substitute for accountability to the citizens. Evidence for this is that the regional governments are characterized by strong presidents and weak legislatures. 

 [Implications

This has implications for the big push ideas of Sachs etc. Whilst institutions may be important for growth, this article suggests that it is not possible to purchase those institutions. First order evidence for this is the huge amount of money that has been spent in trying to develop Africa. The Berg Report called for doubling aid to Africa, which was then doubled again at Gleneagles under the leadership of Tony Blair. And yet the problems persist. This alone should encourage skepticism about the current proposal of increasing aid.

 Second order evidence is that provided in the article i.e. that aid can in fact reduce institutional development. This analysis would seems to imply that aid should be directed to projects where outcomes can have positive institutional effects such as debt relief, peacekeeping and security, together with the provision of public goods such as agriculture and disease research.]

THE BIG PUSH DÉJÀ VU

THE BIG PUSH DÉJÀ VU: A REVIEW Of JEFFREY SACHS’ THE END Of POVERTY

W. Easterly

Journal of Economic Literature 44, 1 (2006), pp. 96-105

 A Summary

In a nutshell: “Big-Push” development economics such as that advocated by JS is not an effective way of helping the developing world. Rather a step by step approach is needed. Sachs advocates large scale social engineering rather than piecemeal reform, and yet this idea (that 1st emerged in the 60s) has been largely dismissed by academics who realise that development is complicated process involving subtle interplays of markets, and systems that operate in a way that is not always immediately reducible by government planners. In other words – we do not always know what we are doing, so we need to tread cautiously and measure the results of our work rather than try to smash poverty with a giant fist of whose effectiveness we are unsure.

 After the fall of communism it was thought that such “shock therapy” would be beneficial. The reasoning was that piecemeal reform of e.g. privatisation would be useless without complimentary reforms of, say, private property. In other words for the sake of effectiveness we cannot wait to measure the success of certain reforms without implementing others at the same time. Sachs says that such a large scale response was critical in addressing high inflation in Poland and Bolivia, yet such solutions do not fit all countries with such varied politics, institutions and incomes. In fact such reform in post-communist states was a disappointment as it was not possible to implement all the reforms at the same time. Such “structural adjustment” is not feasible. In the rich countries of the world progress was slow and considered, not rushed, and thus the authors propose a gradual approach for developing countries.

 The Theory of Aid and the Poverty Trap

  • Poverty traps caused by lack of savings, demographics and insufficient capital stock can be broken by aid argues Sachs. Aid thus has positive effects on growth in developing countries.
  • Easterly contests this: aid in Africa has been high, but it has the lowest growth. Aid has been rising whilst growth rate has actually fallen. Despite long term aid since the 50s, these countries still find themselves in poverty traps, so the healing effects of aid should not be overstated. 
  • Sachs claims that corruption etc. are overplayed as reasons for poverty. Easterly argues in fact bad institutions and policies are the cause of poverty. Empirical research shows (he claims) that bad government is the causal driver of poverty even when we control for initial poverty.
  • Sachs wants to say that the poor countries are ready to take care of themselves, and that increased aid will in fact inspire governmental reform. Yet the evidence suggests that low income is strongly correlated with corruption. Whilst causality of corruption with respect is still debated it seems foolish to dismiss the evidence.

 Poverty as a Technical Problem

  • Easterly argues that Sachs sees poverty as a technical problem that can be fixed with recognised methods. I.e. poverty is solvable with technology and planning.
  • This ignores the social causes of poverty – bad institutions, policies, trading networks that exclude the poor. Sachs want to intervene without any thought for how public and private goods are to be generated by natural incentive.
  • The view of poverty as a technicality lead to Sachs’ administrative solution (the UN administration point).

 Three major problems with this approach

  1. Plans can be decreed at the top but they must be implemented at the bottom. The bottom is unobservable by the top, and the how do you solve the principal-agent problem to incentive the agents to use all the aid etc. for the purpose of poverty reduction. E.g. Uganda where 30-70% government provided drugs disappear before reaching the patients.
  2. Administrators at the top do not have enough information about the situation on the ground to design the right interventions.
  3. Multiple goals/actors weaken the incentives for agents to deliver. Multiple goals means it is hard to get one agent to focus on attaining her goals without worrying about all the others at the same time, so she becomes demoralised and confused. This could be avoided with a very strong principal (in effect what Sachs actually proposes), but we still meet problems 1 and 2.

 The Alternative

  • Piecemeal approach
  • Aid concentrates on finding interventions that work and focussing on them. E.g. vaccination programme that virtually wiped out measles in Southern Africa. E.g. reducing class size. E.g. education subsidies. E.g. clean water. 
  • “The piecemeal engineer knows, like Socrates, how little he knows. He knows only that we can learn from our mistakes… He will avoid undertaking reforms of a complexity and scope which makes it impossible for him to disentangle causes and effects and to know what he is really doing.” K Popper. 
  • “Small moves cumulate into bigger benefits.

A GLOBAL COMPACT TO END POVERTY

A GLOBAL COMPACT TO END POVERTY

J. Sachs

Chapter 14 The End of Poverty: How We Can Make It Happen In Our Lifetime

 A Summary

In a nutshell: a global deal to end poverty. The two sides of the bargain are that poor countries take ending poverty seriously by devoting more resources to this goal; the rich countries in turn make a serious investment in terms of aid. The basic targets that are to be met are the Millennium Development Goals (http://www.un.org/millenniumgoals/)

 Why? Progress is much slower than it seems – in public IMF says how well e.g. Ethiopia is doing whilst in private they admit there is not enough money for health care etc. (health care seen as luxury of development [Sen disagrees here]).

 How? A big push of aid and technology transfers. In order to obtain extra funds we must start being honest with the real state of development in the poorest nations and donors should then realise what their duty is.

 The details

  • Poor countries have no “right” to reach the MDGs – in order to earn the right they must commit to a programme of good governance. For authoritarian and other such regimes “the consequences for the population are likely to be tragic, but the responsibilities of the rich world are also limited” [compare with Collier interventionalist stance]
  • Rich countries must prove they are willing to help well governed countries by providing sufficient aid and other assistance.
  • The Secretary General of the UN must oversee all development, i.e. top down development planning. Individual countries will create a poverty reduction strategy based on the MDGs (such as Ghana and other countries already have – but they are chronically under-funded http://planipolis.iiep.unesco.org/upload/Ghana/PRSP/Ghana%20PRSP%20June%202006.pdf)
  • Currently there is no consultation with the recipient countries as to how much aid they need. They are presented with a figure and told to work with it. When Ghana wrote its PRS it requested $8bn ($75 per person per year) which would enable to meet the MDGs. They were told this was unrealistic and they had to re-draft and re-draft until the figure was down to the $2bn figure that had in fact already been decided upon. With that plan reaching the MDGs is not possible.

 An MDG based PRS

           Differential Diagnosis – identifies investments and policies to achieve MDGs

  • Investment Plan – shows costs, timing and size of investments. [How exactly is a country supposed to write a plan to address all of the MDGs i.e. addressing health, education, poverty, hunger etc? This idea starts from the premise that we know what is wrong with the systems in a country, we know how to solve them, and we will be able to do so for all such defective systems simultaneously. Is this realistic?]
  • Financing Plan – how it is to be paid for, including the funding gap which must then be filled by donor aid.
  • Donor Plan – gives details of donor commitments.
  • Public Management Plan – outlines mechanisms of governance and administration. 
  • By changing planning from year to year to a long term view we can change the world e.g. in two years we can train community health care workers and doctors can be attracted by raised salaries. With 5 year plan med schools can be enlarged. With 10 year plan new schools can be built etc. and health care has been dealt with. I.e. there is sufficient absorptive capacity (ability to use large amounts of aid) in developing countries, but a long view must be taken. [This does not seem to allow for a plan for eventual non-reliance on aid. How is this to be financed after 10 years?]

 The Donor Commitment

    Magnitude – aid must be sufficient to finance the PRS. At present this is not the case.

  • Timing – unlike current methods aid should be given in long term patterns to allow for a 10 year plan to be enacted.
  • Predictability – aid must be delivered in predictable manner, as sudden cessation of aid in countries where it may make up to 30% GDP per year will have serious macroeconomic consequences.
  • Harmonisation – rather than a score of small “pet” projects all aid should be directed through a single agency that provides all bilateral aid.

 Regional Groupings

  • Currently not enough work done on multi-country programmes. E.g. the road that connects the port of Mombassa with the rest of Kenya/Uganda/Rwanda/Burundi. Maintenance is piecemeal whereas the response should be co-ordinated to incorporate all 4 countries.
  • The motive for such collective response is that it increases efficiency, and such projects increase “peer review” and collective pressure for good governance.

 Global Policies for Poverty Reduction

  1. Debt – grants, not debt. Once it became clear the developing world was saddled with too much debt, the payments should have been cancelled. We must make grants (like the Marshall Plan post WWII) as the developing economies are too fragile for huge debt.
  2. Global Trade Policy – developing countries must increase exports to earn forex for imports of capital goods. Trade + aid, (vs. Trade not aid). Trade alone is insufficient as the benefits tens to accrue largely to the rich economies. Agricultural trade specifically must be liberalized (but this will not solve everything as Africa is a net importer of food).
  3. Science for Development – technology needs to be developed and distributed to stop some of the causes of poverty. E.g. disease prevention, new seed crop development, water-management, climate predicting and warning technologies. $7bn a year is needed for R&D.

 Who Administers the Aid?

  • This should be in the hands of the secretary general of the UN as he is best placed to coordinate all the various other agencies.
  • Each country to have a dedicated UN development team.
  • This is because at present the burden of too many agencies at work in each country (and all the paperwork that entails) is too great on the developing world. The agencies do not coordinate and are not providing a targeted response to poverty.
  • It must be the UN as the IMF and World Bank are controlled by the rich (as $1 contribution gets 1 vote, so USA etc. have majority control and the process is not democratic.

AN AFRICAN SUCCESS STORY: BOTSWANA

AN AFRICAN SUCCESS STORY: BOTSWANA

Acemoglu, Johnson & Robinson

In Rodrik, ed., In Search of Prosperity, pp. 80-122

 A Summary

In a Nutshell

This is a classic insitutionalist argument that attempts to explain the extraordinary growth record of Botswana based on the institutions that were in place. These institutions made the choice of economically orthodox policies and possible despite adverse post-colonial conditions of inequality and poverty. The reason that these institutions were able to exist in Botswana was primarily due to the pre-colonial existence of inclusive institutions, and the light handed British colonialism that enabled many of the indigenous institutions to remain in place even after independence. The protection of private property was in the interest of the elites, and there was significant revenues from diamond mining that were distributed widely meaning that there was little to be gained from seeking to capture those revenues.

 Institutions

  • In this context good institutions means that a broad cross-section of society have effective property rights. These contrast with extractive institutions. There is an implied limit to arbitrary and extractive behviour by the state which must be to a certain degree strong in order to protect those rights.

 Relevant History of Botswana

  • The chief was the central figure but there existed a hierarchy of public forums.
  • Land was farmed collectively but there was private ownership of cattle.
  • The location of Botswana was strategic and this was of value to the British rather than any territory or land that was thought to be attractive itself. This meant colonialism was very light.
  • Post-colonial politics saw the emergence of a party that sought to integrate traditional chiefs and the educated elites. This party has ruled ever since (not a great sign of democracy).
  • The president wanted a strong central state to prevent impediment by the regional rulers. The chiefs were gradually stripped of their powers.
  • Sub-soil rights were vested in the government so the government received 50% profits from diamond mines. The profits were used efficiently to invest in the government budget. Resources were thus invested not squandered.

 Political Economy of Botswana

  • The success of Botswana has been the enacting of good policies. But AJR argue these are not causes but outcomes of good institutions.
  • The economic outcomes of the immediate post-colonial system of development were beneficial to the elites and as such they had an interest in preserving the institutions they inherited. Once the diamond income started flowing in there was no struggle for control, as the degree of political stability that had been established along with the widespread use of the diamond funds to make productive investments meant that the elites were not afraid of losing their dominant position. The challenging of the institutional path in other words had a high opportunity cost meaning that no group wanted to fight – they didn’t want to rock the boat.
  • The interests were aligned as the political elites had an interest in preserving the institutional system they inherited upon independence. They could keep their power by following good policies.

[Implications

Following Przeworski, such analyses suffer from endogeneity problems. Why should we accept that it was the institutions of private property that drove growth in Botswana rather than the conditions that enabled those institutions? Putnam however, makes a case for saying that social capital is the foundation of those institutions and enables them to function.

 It is not clear how to disentangle the effects of the diamond revenues from the larger development picture.

 Nevertheless even if we submit to the logic of the paper and agree that it is the institutions of Botswana that have enabled it to develop, it is quite clear that this development is the product of a unique history, a history not shared by any other state, and thus no replicable. It would not be a valid strategy to export the institutions of Botswana to Somalia where the shared ideals and history, and social capital, is so different.

 This would suggest that interventions in Somalia should be based upon piecemeal solutions rather than any grand structural plans (Easterly).]

INSITUTIONS MATTER?

INSITUTIONS MATTER?

A. Przeworski

Government and Opposition 2004

 A Summary

In a Nutshell

The logic of collective action implies that institutions are needed for the type of cooperation that enables a country to develop beyond the type of society based on family and small clans. Ostrom shows that institutions lower the cost of monitoring and enforcing agreements. However, Putnam then goes on to show that in order to explain the existence of institutions we have to look to the foundational civic traditions and social capital. Thus as North argues there are two key propositions to the theory of new institutionalism:

  1. Institutions Matter – they influence norms of belief and action and shape outcomes
  2. Institutions are endogenous inasmuch as their form and function depend upon the conditions under which they emerge and endue.

 Yet, if conditions shape institutions and institutions transmit the causal effects of these conditions how do we distinguish the effects of institutions separately from the conditions which gave rise to them? In other words, how is the problem of reverse causality overcome? Only by answering this question can we judge the effectiveness of implanting institutions into some particular conditions in the hope that they will function in the same way that they did elsewhere. Is it the institutions or the conditions that matter?

 For example we may observe the growth rates of Zaire under a dictatorial rule and question whether the growth would have been different had it experienced democracy. And yet it is only possible to observe one state of the world, thus counterfactuals are created to re-create the same conditions of Zaire under democracy. Yet there is no way to determine if the counterfactuals are valid

 This implies that projects of institutional reform must take as a point of departure the actual conditions not blueprints based on institutions that have been successful elsewhere. It is also important that institutions reflect the basic power structure of society otherwise they will not be “self-enforcing”.

SOCIAL CAPITAL AND INSITUTIONAL SUCCESS

SOCIAL CAPITAL AND INSITUTIONAL SUCCESS

R.D. Putnam

Cp. 6 of  Making Democracy Work: Civic Traditions in Modern Italy

A Summary 

In a Nutshell

Collective action problems mean that in the absence of a credible mutual commitment each member of a group has an incentive to defect and become a free rider. In such a situation actors must forgo many opportunities for mutual gain “ruefully but rationally”. One potential solution is to allow for third party enforcement. However this is expensive as the neutral third party (potentially the state) must monitor the contracts and enforce compensation such that it is cheaper for the defector to comply with the contract rather than break it. Even more basically impartial enforcement is a public good and so subject to the same problem it is trying to solve i.e. the third-party itself may have incentive to defect from impartial enforcement. Thus, impartial 3rd party enforcement is not generally considered a stable equilibrium where no player has an incentive to defect.

 So why then do we see cooperation? Indefinite repetition of games is important whereby the defector can be punished in subsequent rounds. This requires that there be abundant free information about past cooperation of players, which would seem to suggest that cooperation is less likely than the observed level in society.

 Ostrom argues that formal institutions reduce transaction costs of monitoring and enforcing agreements. The success of these institutions are conditional on the boundaries of the institution being well defined, that affected parties participate in creating the rules that there are graduated sanctions, and that the conflict resolution method is sufficiently low cost. Whilst this may be the case the question arises as to how such institutions come about? What are the conditions that enable them to be created? The participants themselves cannot create them for the same reason that they need it in the first place. The answer proffered is: community trust.

 Thus voluntary cooperation is much easier is communities that have a large stock of social capital in the form of norms of reciprocity and networks of civic engagement. When individuals have lived in such a situation with social trust and capital, and shared norms of reciprocity exist, there exists the foundations for building the institutional arrangements for resolving collective action problems.

 Trust is an essential part of this social capital and every commercial transaction involves a degree of trust. “Trust lubricates cooperation”. An individual expects the other actor to choose cooperation outside of the ability of a formal institution to punish defection.

 [Implications

The implication is that institutions cannot be imported into situations in which they have not developed organically, as the social capital foundations are not strong enough. Thus the key to cooperative behviour is not just getting the rules right. For even in the presence of strong contract law for example, commercial undertakings would be next to impossible if the level of mistrust in society meant that full contracts would need to be drawn up for any commercial interaction, as well as the need to include the expected costs of prosecuting upon defection which would appear highly probable. This would seem to indicate that the protection of private property for example cannot be the whole story as to the successful development of certain states {e.g. Botswana – see AJR}, as there must be some initial conditions of trust that enabled the institution of private property to be built.

 The argument also seems to move from conditions to institutional development and not the other way around, which answers Przeworski up to a point.]

Group Size and Group Behaviour

Group Size and Group Behaviour

M. Olson

Chapter 2 of The Logic of Collective Action: Public Goods and the Theory of Groups

A Summary 

In a Nutshell

It is not in the interests of rational members of large groups to contribute toward achieving the common interest so long as the group members are free to pursue their own individual interests. This is because when the number of members of a group is sufficiently large such that the typical participant knows that his own efforts will not make any difference to the outcome and he will be affected by the outcome in much the same way whether he makes any contributions or not. Put another way, as the cost of one individual’s defection is shared by the whole group, the individual himself pays very little, but the benefits from defection are the same as though he had contributed. Therefore it is rational for the individual to defect, or shirk, from the agreed plan of action of the group.

Being in a small or “privileged” group is much more advantageous. Collective action is possible because the incentives affecting each member dictate that each individual has an interest in acting such that the group does not fail. The incentives that face the group are the same that face individuals, whereas for large groups this is not the case.

This conclusion holds even if the supposed outcomes of large group action would provide large benefits to the members. Additionally, the argument is formulated on the assumption that there is perfect consensus, thus in the real world where no such consensus exists, collective action problems of free-riding are exacerbated by argument over initial goals.

Social incentives can incentivize group members to act in the interests of the group. This can be social position pressure, threat of exclusion. These are called “selective incentives” that can be mobilized to animate a latent group. The nature of the incentives is that they can distinguish between individuals. However, generally such social sanctions can only function in smaller groups. This is because in larger groups it is not rational to exclude someone whose actions by definition had no effect on the outcome of the group, and secondly because the large group will not be a friendship group meaning the defector is unlikely to be socially affected if he fails to make the sacrifices on the group’s behalf.

The argument is not affected by claims that people act selflessly. This is because even a perfectly selfless individual will not think it rational to undertake action where his own contribution will be utterly imperceptible. “A man who tried to hold back a flood with a pail would probably be considered more of a crank than a saint, even by those he was trying to help”.

[Implications

People are rational, but individually rational choices are not necessarily in line with the socially optimal choice. This has implications for voting, for aid giving, for mass organization. Partiuclarly with regard to mass organization, even if Marx could convice workers that they were being exploited and needed to throw off the chains of oppression, the very size of the group in question will place a heavy presumption against the successful organization of a revolution. There will not be a revolution precisely because so many people want it.

This implies that insitutions that are able to overcome these collective action problems are necessary if society is progress beyond the tribal/family unit. For example the state can selectively incentivize individuals to co-operate for the greater good (by using tax funds to pay workers to build a road for example). Also the state can enforce property rights which are an essential component of the institutional hypothesis.]