|
|
|
Conceptualizing RFI’s versus GFI’s - Ravi Kanbur
|
|
|
2. GFI’s for GPG’s, RFI’s for RPG’s
|
| |
I take as given the urge in rich countries to transfer resources and know how to
poor countries, based on a desire to help them overcome their poverty, and will base the arguments on this paper on this assumption. Whether this latter desire stems from true
altruism or from a feeling that the poverty of others will eventually have negative
consequences for the rich is an interesting question but one that is not relevant to my task
here. Rather, I will ask—given this urge, what is the best way to execute it, and does the
current configuration of RFI’s and GFI’s come close to this best configuration? In taking
this analytical route I put to one side the obvious fact that poverty reduction is not all that
drives the urge to transfer resources, and development assistance is not the only reason
why rich countries support RFI’s and GFI’s. Narrow self-interest within rich countries is
oftentimes present, as in the case of tied aid, and the foreign policy interests of rich
countries are never fully divorced from the concerns they express in RFI’s and GFI’s.
Despite these caveats, it is nevertheless a useful organizing framework to lay out
the advantages and disadvantages of alternative institutional arrangements for delivering
official development assistance. The most obvious starting point is bilateral assistance.
The advantage of this form of assistance is that the taxpayers of the donor countries can
have, in principle, close scrutiny of this assistance. The disadvantages also flow from this
very same feature—domestic political processes, and domestic distributional struggles,
can begin to influence the nature and composition of development assistance. This is seen
most clearly in the various pressures that come on a government to tie its aid to purchases
from its own suppliers.
But an equally powerful advantage of a donor country pooling its resources with
other donor countries is that it can benefit from economies of scale and scope in development assistance activities. At the simplest level, capacity to evaluate assistance
projects and programs can be centralized and the costs shared. Hence, in principle, a
pooling of resources in a rich donors’ club like the EU should have considerable cost
advantages. However, pooling of countries in turn introduces the problem of differences
in preferences of the different members on how the resources should be used. Whatever
consensus is reached, it will always be unsatisfactory to some countries’ preferences as
donors. The cost advantages therefore have to be balanced against this disadvantage.
Faced with the costs and benefits of the two alternatives—going it alone or joining a rich
countries’ club—therefore, a rational response for each donor country will be to
diversify, to have some of its development assistance flow through bilateral channels and
some through channels that group together rich country donors.
The argument above does not as yet provided a rationale for RFI’s, since it
discusses groupings of donors, not groupings of recipients. What are the costs and
benefits of grouping recipients into groups that are defined by geographical region? The
cleanest argument comes from a consideration of cross-border externalities and multicountry
public goods. Cross-border externalities have received increasing attention in
recent years, and do not need detailed elaboration here. Suffice it to say that when
development in or actions by one country have an impact on other countries, an impact
that is not mediated by classical competitive markets, the presence of such an externality
can lead to sub optimal policy outcomes for the group of countries encompassed by that
externality. There are gains to be had for all countries by developing a mechanism that
coordinates policies and actions, with compensation as necessary. Such mechanisms satisfy the classical economic definition of a public good between the relevant countries,
and it is well established that such public goods will be undersupplied.
The link between cross-border externalities, multi-country public goods, and
development assistance becomes clear if all or some of the countries encompassed by the
externality are poor countries, and the proposed mechanism will benefit them. If these
conditions are met, then since the objective of development assistance is to improve the
lot of poor countries, it follows that mechanisms that address such cross-border
externalities have a natural claim on aid resources. These mechanisms are not costless—
far from it. A use of donor resources to finance them is a legitimate use of aid resources.
However, by definition, the use of resources in this way requires groups of
countries encompassed by externalities to come together. Let us consider a progression of
groupings, starting with near neighbors at one extreme, and gradually expanding to
include all countries in the world for externalities that are truly global in nature. If setting
up of these country groupings were entirely costless, then a separate one could be set up
for each externality and dealt with separately. But it is very likely that there will be
economies of scale and of scope. If the fixed cost of setting up institutions is large, there
is an argument for grouping together the groups themselves. One criterion for such
grouping is the likelihood of the occurrence of cross-country externalities across the
different groups. Given the high likelihood that geographical proximity is a key
determinant of the likelihood of certain types of physical externalities (water table, forest cover, migration, civil war, transportation, smuggling, etc.), it does seem that a regional
grouping is a good initial cut.
Of course, the regional cut is not appropriate in some cases. For truly global
phenomena like global warming, global financial contagion, global spread of infectious
diseases, or global conflict, a global grouping is needed. To the extent that these
phenomena affect the well being of poor countries, and global coordination could
improve their lot, global mechanisms for addressing these problems, and global
institutions that do so, also have a claim on aid resources. There may be yet other types of
externalities where neither regional nor global groupings are appropriate—the producers
of a single commodity like coffee come to mind—but, while recognizing their
importance, I will not pursue these further here.
The above argument suggests a division of labor—regional externalities to be
dealt with by regional institutions, global externalities by global ones. Put another way,
RFI’s to supply Regional Public Goods (RPG’s), and GFI’s to supply Global Public
Goods (GPG’s). Of course the division is not clear-cut, since sometimes there are some
spillovers from a regional phenomenon to the global setting. But even a cursory
examination of the operations of the World Bank and the Regional Development Banks
(RDB’s) does not suggest a division of labor in addressing cross-border externalities.
Granted that such operations are a small part of the operations of these institutions, but
the World Bank is more often than not involved in the supply of what are regional
specific multi-country public goods, with either no involvement or overlapping involvement of the relevant RDB. But our argument suggests strongly that the World
Bank should reduce its role in and the RDB’s should increase their role in such activities.
To the extent that there is lack of capacity in the RDB’s to undertake these activities, this
is an argument for the donor community to build up RDB capacities, perhaps even using
some of the resources currently devoted to World Bank activities. Such specialization
would also be helpful because it would allow the World Bank to focus on its comparative
advantage—global externalities.
|
|
Related documents