42 SOCIAL SCIENTIST excess demand in the two tradable sectors is not of a high order. Therefore, the inflationary pressures on the economy critically depend on our ability to raise the rate of growth of output of the non tradable sectors, particularly Railway transport and Electricity. Deficit spending beyond 10 per cent generates excess demand in Agriculture because of which the foreign exchange requirement shoots up substantially. More important, the growth rates required to meet the excess demand in non-tradable sectors seem too high to be achieved. The final question is what type of inference can we draw for the present day situation on the basis of the above results ? In the period subsequent to our period of reference, the Union Government expenditure net of tranters. has increased roughly 1.4 times in real terms. Any percentage increase at this higher level of government expenditure implies a much larger real deficit than we have postulated in our excercise. Similarly, the economy must have been better integrated in the intervening period and therefore sectoral multipliers (Our K[ in Section I) for a given composition of government expenditure will be higher which in turn mean that the additional expenditure required (Our G[* in Section I) to reach the full capacity level in different sectors will be correspondingly lower than what we have estimated in our exercise. This should be viewed in the following context. Although the capacity of different sectors must have been raised in the intervening period, the level and the structure of excess capacity as worked out for four broad sectors viz. Basic Goods, Capital goods, Intermediate goods and Consumer goods have remained remarkably stable between 1971-809. The composition of government expenditure, at least the part relating to commodity purchases, might have been changed. However, one does not expect drastic changes in the composition of government expenditure. To the exetent the present pattern of sectoral demand conforms to the one implicit in this excercise, a much lower percentage of deficit spending than what we have indicated is likely to generate inflationary pressure. 1 The sectoral composition of the expenditure on commodity purchases has been generated by scrutinising and shifting the data scattered over a large number of sources like (i) Detailed Demands for Grants, (iij Directory of Government Purchases, (iii) Supplement to the Report by the Railway Board on Indian Railway etc. But in the case of salary disbursals, there is no way of working out the commodity vector directly. This has been done indirectly by establishing a relationship between the salary distribution of the government employees and the consumer behaviour with the help of three parameter lognormal distribution. For further details see : Atui Sarma and V.B. Tulasidhar, Economic Impact of Government Expenditure, Concept Publishing Co: New Delhi, 1984. Chapters 3 & 4. 2 The Fifth Plan Input matrix consists of 66 sectors. The coefficients are for 1973-74 at '1971-72 prices. The domestic coefficient matrix has been generated by deducting import coefficients matrix from the total coefficient matrix. These matrices are given in Annexurc III of