Journal Issue


International Monetary Fund. External Relations Dept.
Published Date:
December 1982
  • ShareShare
Show Summary Details

Triggering graduation

In the Bank activity item, “World Bank’s graduation policy reaffirmed” (June 1982) there is a feature that ought to be clarified in light of recent developments in the literature and policy of evaluating countries. The third point stated:

Graduation will normally occur within five years after a country reaches a per capita gross national product (GNP) benchmark, but this period might be longer if the situation deteriorates during the phase-out period.

In this day and age, given the volume of material on the inherent inadequacy of the per capita GNP concept/index as a meaningful measure of the standard of living in a country, I find it unbelievable that per capita GNP is still being used as a measure to determine the phase-out of countries from the World Bank borrowing program.

I am sure that the technicians and the technocrats of the Bank are fully aware of the developments in the Physical Quality of Life Index (PQLI). Morris David Morris’ Measuring the Condition of the World’s Poor offers very useful insights into the nature and usefulness of the PQLI. I suggest that it is time for the Bank to begin using PQLI as the basis for its graduation program from Bank borrowing.

The point is made half-heartedly in the feature that “The graduation issue, including any problem that may arise in the application of the per capita GNP level that triggers graduation, will be reviewed annually.” This, to me, is only a partial solution. There needs to be a complete review of the graduation issue in an attempt to bring it in line with current political economic reality. PQLI offers this political economic reality.

S.B. Jones-Hendrickson, PhD

Associate Professor of Economics

College of the Virgin Islands

Bevan Waide, Director of the Bank’s Country Policy Department, replies

Determination of eligibility for graduation based on per capita GNP is simply a starting point for a careful review of each country’s situation and the development of a flexible program to phase down and ultimately end Bank lending. This review examines the country’s overall economic condition and its capacity to sustain a long-term development program. Particular attention is paid to two important factors that influence the pace of graduation—access to external capital markets on reasonable terms and the extent of progress in establishing key institutions for economic and social development. In reviewing the latter, a broad range of social indicators—including the three that make up the PQLI—are taken into consideration, because countries that have similar levels of per capita income may show substantial differences in their progress in these areas.

The Bank is aware of the inadequacies of per capita income measures and supports international efforts to refine such measures, that is, the United Nations International Comparison Project and the Living Standards Measurement Study. The PQLI itself is a controversial index and has its own substantial data problems. Per capita income, suitably adjusted where necessary, seems to remain the best available index for use in establishing an initial graduation benchmark, because it provides the most widely accepted basis for cross-country comparisons.

Oil and water?

I am taking the highly unusual step of writing to you with regard to the “Review Article” in Finance & Development (March 1982) by Salah El Serafy of my book (with Stuart Sinclair), Oil, Debt and Development: OPEC in the Third World.

(1) El Serafy’s main theme is that “user cost” rather than cartel theory is a better explanation of oil price behavior in the 1970s. But El Serafy makes no attempt whatsoever to lend empirical support to this assertion. In fact, user cost pricing is so highly sensitive to discount rate, expected future prices, and time scale as to make it very difficult to choose between the competing theories.

(2) El Serafy also asserts that there are “relatively low demand elasticities facing individual sellers” (p. 40). This is incorrect since cross elasticities of demand for oil exported by different sellers can be quite high (see R. Kuenne in Kyklos, Vol. 32, Fasc. 4, 1979, pp. 695–717).

(3) With regard to the OPEC aid-oil balance, El Serafy says that our analysis is “inadequate” (p. 41) since we leave out of account boosted workers’ remittances and trade flows. However, we deliberately excluded these because their opportunity costs are positive. For example, (a) domestic consumers lose the services of labor which temporarily migrates to OPEC members; and (b) exports to OPEC are not produced at zero cost and might be diverted from other markets.

(4) El Serafy clearly misses the subtleties of Marshallian offer curve analysis: in fact it is not an “obvious proposition that higher prices benefit the sellers at the expense of the buyers” (p. 39). The whole point is to show optimum monopoly pricing. Cartel prices could be so high as to reduce the members’ welfare.

(5) El Serafy’s own view of OPEC changes when it is convenient for his argument. Thus, when discussing oil prices he finds it convenient, in effect, to view OPEC as a loose association of oil producers. But when discussing OPEC aid and per capita income, it becomes convenient for him to “reaggregate” the members so as to be able to claim that “per capita ‘income’ for OPEC as a group is, in fact, little more than half the world average” (p. 41).

C. Paul Hallwood

University of Aberdeen

United Kingdom

Salah El Serafy, Senior Evaluation Officer in the Bank’s Operations Evaluation Department, replies

(1) I take my cue not from the Keynesian concept of user cost but from Hotelling and the theory of exhaustible resources.

(2) As a group, OPEC possesses a sizable portion of the world petroleum resources and a significant market-share of world trade in petroleum. The same can be said of certain members of OPEC, taken separately. Even without collusion, therefore, oligopolistic elements are present in the petroleum market. The higher the market-share, the lower the (absolute) price elasticity of demand facing any one seller. Large market-shares inevitably lead to finite elasticities of fairly low magnitudes. Kuenne’s values (whether for own or cross elasticities) corroborate this. Add depletability and dubious financial and other investments and you get backward rising supply curves and the desire for conservation. Lower quantities (and higher prices) would thus seem to follow both from market structure (without collusion) and from the fact that oil is not reproducible.

(3) Increased exports and labor migration may have positive costs, but since this ‘trade’ exists one must assume it is profitable for the exporters. In the world of economics nothing is without cost, least of all petroleum itself.

(4) I see no subtlety in depicting two sets of indifference curves and two offer curves and asserting that the petroleum market outcome is at the extreme edge of what Bowley (Mathematical Groundwork) called the bargaining locus. At the bottom of page 53 you yourself realize that it is doubtful that equilibrium is where you show it in the graph.

(5) There is no inconsistency in treating OPEC as a group in certain contexts, while considering the individual interests of its members in others. For market behavior both approaches are necessary. As a subgroup of developing countries, confronted with common problems, aggregating them can be useful. But it is not necessary.

Other Resources Citing This Publication