the first place through inflationary financing. In subsequent rounds of the operation of the multiplier, however, the effects would be similar on the whole to those obtaining from domestically financed investment. Perhaps a brief mention should be made of the probable operation of a negative multiplier as well. The decline in the Handicrafts & Small Scale Industrial sector can partly be attributed to the uncertainties and restrictions imposed upon the private sector by the Revolution, but it could certainly also be attributed in part to the decline in demand that the inflationary process would have caused, as higher proportions of family budgets were being diverted to those producing and/or trading in foodstuffs. Finally, our analysis based on n and y runs in terms of national aggregates, although clearly the model underlying it is one involving inter-sectoral transfers of marketed surplus from agriculture to non-agriculture. This aggregation conceals some important aspects of the process. First, it ignores the possibility of non-inflationary productive investments based on rural labour accumu-lation, and thereby exaggerates the strategic necessity of extracting and transferring rural surplus product. Second, it ignores the fact that the rural sector is largely insulated from the urban food market, since it has the first claim to foodgrain consumption. The implication follows that the urban areas are thus rendered even more vulnerable to variations in agricultural output and marketed product. This fact was starkly highlighted by the recent Ethiopian experience, and the extreme consequences for the urban population form the subject matter of the following section which documents the im-poverishment of an increasing proportion of urban households.

ABSTRACT

The other vital event for the poorer sections was the increase in the relative importance of the grain purchasable from the kebele shops at controlled prices. During 1981/82 these shops supplied 44.7 kgs of foodgrains per household per month in Addis Ababa as against the nominal ration level of 65 kgs per month. For the remaining one-third, the household would have had to rely on the open market where, prices, on average, were 50-60 per cent higher in 1981/82. While the AMC's supplies clearly perform a vital role, a few qualifications need to be made. First, the coverage of urban areas other than Addis Ababa is very restricted but, since price data are not available for cities other than Addis, it is difficult to form any idea of the relative difference. However, it is probable that open market prices are lower in the smaller towns, and the net effect might be less than would be implied by the Addis differentials indicated in Table 9. The second point concerns the movement in the relative prices of the 'poor' and the 'rich' cereals baskets at the kebele shop. Kebele prices of the superior cereals dropped in 1979/80-1981/82 by about four per cent, while the inferior ones have risen by nine per cent. The open market prices also show a similar disparity with the former group registering an increase of 11.6 per cent as against one of 26.6 per cent for the latter group. While in times of inflation and real income decline, the open market movements can be explained by a part of the population switching to the consumption of inferior grains, the kebele price movements could also possibly be an indication of misguided policy. Third, the decline in the prices of teff at the kebele is matched by a rise of ten-fold in the quantity of teff handled by the AMC. The evidence certainly argues that while the increases in the coverage of the AMC will have helped the poorer sections, the change in relative prices has moderated, although clearly not obliterated, that gain. In general, the relative price movement went against the 'poor' cereals basket by 13.4 per cent in two

be noted that while agricultural GDP per capita fell by 5.1 per cent over the reference period, and GDP per capita by one per cent, the real GDP generated by 'Public Administration and Defence' rose by 22.5 per cent in per capita terms. This sector improved its share in GDP from 6.97 per cent in 1974/75 to 8.61 per cent in 1979/80. Some of the reasons for this are readily understandable, yet it does tell us something about how some of the surplus generated was utilised.