Population_growth_in_indiaa_reappraisal.pdf

THE ECONOMIC WEEKLY

September 16, 1961

Population Growth in India A Reappraisal T C Chellaswami The 1961 Census count exceeded even the upper limit of expectation of projections worked out earlier. An attempt is made here to critically examine the difference between the enumerated and the estimated population totals and to determine whether the assumptions' underlying the earlier projections stand in need of revision and, if so, in what direction and to what extent. An attempt is also made to estimate the growth of population in the period 1961*76 on the basis of modified assumptions in regard to future mortality and fertility trends in the light of the 1961 Census count and recent N S S estimates. T H E announcement o f the p r o v i sional p o p u l a t i o n count of the 1961 Census created an understandable stir. It has already p r o m p t e d the planners to re-examine the v a r i ous p o p u l a t i o n forecasts. W h i l e the newspapers flashed in head-lines that p o p u l a t i o n g r o w t h has belied all forecasts, the Registrar General in his publication '1961 Census P r o v i s i o n a l P o p u l a t i o n Totals' stated that "the count has exceeded even the upper l i m i t of expectation o f projections w o r k e d out i n past years". A l l these seem to i m p l y the need for revision in the assumptions u n d e r l y i n g the various p r o j e c t i o n s Indeed some demographers seem to t h i n k so. T h i s paper tries to examine whether there is any case f o r revising the assumptions at the present stage and, if so. to what extent and in what d i r e c t i o n . Table I below gives three different estimates f o r the p o p u l a t i o n of I n d i a ( e x c l u d i n g J a m m u and Kashm i r and the T r i b a l Areas of Assam) for the p e r i o d 1951-76 at 5 year intervals assuming zero net m i g r a t i o n based on certain assumptions regarding m o r t a l i t v and fertility trends d u r i n g 1951-76. The base 1951 p o p u l a t i o n used f o r the pro j e c t i o n was not corrected for net under-enumeration at the 1951 Census.

Adjustments Necessary Before a t t e m p t i n g the c o m p a r i son between the enumerated and estimated ' p o p u l a t i o n f i g u r e s for 1961. adjustments for the f o l l o w i n g factors are necessary so as to make them comparable :— (i) geographical the t w o figures;

coverage

of

( i i ) m i g r a t i o n factor in the two sets of figures; and. ( i i i ) the net under-count of the

p o p u l a t i o n at the Censuses of 1951 and 1 9 6 1 . The p o p u l a t i o n according to the 1961 Census after adjusting the same f o r ( i ) geographical coverage ( i i ) net m i g r a t i o n into the I n d i a n U n i o n f r o m abroad d u r i n g 1951-61, and ( i i i ) net under-count at the 1961 Census, is obtained as explained below: (Figures in millions) The population of the Indian Union according to 1961 Census 438 Minus population of Jammu Kashmir and the Tribal areas of Assam according to 1961 Census 4 Minus estimated population of migrants into India from abroad and their natural increase during the decade 1951-61* 3 Plus the' estimated net under-count at the 1961 Census of about one per cent t 44 Population of Indian Union in 1961 after adjusting for the three factors 435 Comparable projected population in 1961 except for adjustment for net under-count of 3951 Census population according to author's projections 424 Difference

11

* In an article in The Economic Weekly, April IS, 1961, migration from East Pakistan during the decade 1951 61 was analysed and was estimated to be about 2,1 million (during the decade 19511961)'. The total number of migrants into India and their natural increase during the decade 1951-61 from East and West Pakistan, Tibet and Nepal is assumed to be about 3 millions. + No estimate of net under-count for 1961 Census is available as yet. Hence, it is assumed that this would be of the order of about one percent

As adjustments f o r differences in the geographical coverage and m i g r a t i o n factor have already been made, the short-fall by 11 m i l l i o n in the estimated p o p u l a t i o n could o n l y be explained by ( i ) net under-count at the 1951 Census a n d / o r ( i i ) the nature of the assumptions. U s i n g the same method and assumptions followed in the author's forecasts, a p o p u l a t i o n o f 135 m i l l i o n i n 1961 could have been obtained if the base p o p u l a t i o n of 1951 was of the o r d e r of ,364 m i l l i o n . AS already e x p l a i n ed the 1951 base p o p u l a t i o n used for the projection was 357 m i l l i o n arid this does not allow f o r the net under-count of about 1.1 per cent revealed by the post-1951 Census sample verification check, T h i s u l i der-count alone w o u l d raise the base 1951 population to 361 m i l l i o n . This leaves a gap of 3 m i l l i o n .

Under-count in 1951 It may be noted that the 1961 Census p o p u l a t i o n has been corrected for an under-count of about 1 per cent. There is reason to believe that the actual net under-count at the 1961 Census m i g h t have been lower than that at the 1951 Censu. due to improvements in coverage, methods and organisation d u r i n g the 1961 Census and this w o u l d reduce further the above stated gap of 3 m i l l i o n . M o r e o v e r , the estimated net undercount of 4 m i l l i o n during 1951 Census is l i k e l y to he rather on the low side because the design adopted for the sample verification of 1951 Census count was not satisfactory

Table 1 : Population of the I n d i a n U n i o n (Excluding Jammu & Kashmir and Tribal Areas of Assam) Projection 1951 1956 1961 1966 Coalr and Hoover 357 384 424 473 T Chellaswami 357 386 424 460 Planning Commission Expert Commirtre headed by Registrar Genera! 357 386 425 473

1471

1971 532 521 521

1976 601 581

THE ECONOMIC WEEKLY

September 16, 1961 The verification officers were required to check the enumeration of per sous from a random sample of households selected f r o m the National Register of Citizens which was itself prepared d u r i n g the census The verification whether any occupied house had escaped enumeration was done by ascertaining whether three occupied houses nearest to the sample house had been included in the National Register. This procedure was not satisfactory because of the fact that only such houses or households which are inaccessible or located in isolated places are l i k e l y to be missed 1 .

Nature of Assumptions The assumptions u n d e r l y i n g the a u t h o r s earlier p r o j e c t i o n " are rather on the h i g h side and are. therefore, expected to yield a higher population for 1961. That the expectation of life at b i r t h w o u l d increase at an average rate of 9 years every decade is too optimistic, j u d g e d by the available international experience. A U N R Report, discussing the assumptions of future m o r t a l i t y declines, proposed as a model, at annual gain of 0.5 years in expectat i o n of life at b i r t h wherever the expectation of l i f e at b i r t h is less 1 Coale and Hoover, while examining the size of the 1951 population, observe: "Our analysis of the age and sex distribution in 1951 indicates that there was a substantial undercount of children under 5 in the Indian Census and also indicates less persuasively, an undercount of females A rough estimate of the two deficiencies in the Census Count is 20 to 25 million

than 55 y e a r s . As regards f e r t i l i t y , it has been assumed that the general f e r t i l i t y rate (G F R) of .177 corresponding to a crude b i r t h rule (C B R) of 43 per 1000 population assumed for the quinquennium 1951-56, would remain unchanged upto the m i d d l e of the quinquenn i u m 1961-66. T h i s is by no means an assumption on the low side. In Fact the projection of 1961 popula tiou (442 m i l l i o n , w i t h correction for under-count) on the earlier assumptions w o u l d give a population of about 495 m i l l i o n in 1966. F r o m the discussions in the preceding paragraphs it w i l l be clear that the gap of three m i l l i o n stated above could only be explained by the higher net under-count at the 1951 Census, as the assumptionu n d e r l y i n g the projection are considered to be on the h i g h side.

Intimates of Population Growth It is not the i n t e n t i o n of this paper to argue for the projections. The m a i n objective here is to emphasise that w i t h our present know "Methods for Population Projection Sex and Age ( S T / S O A/Series Population Studies, No 25)

For details regarding the original assumptions, their justification and the method of projecting the population. reference may be made to: T Chellaswarm. "Mortality and Fertility Trends and Population Growth in India 1956 76 )" in 'India's Population', Asia Publishing House, 1960.

1472

by A,

ledge and in the absence of detailed age d i s t r i b u t i o n of the 1961 Census Count and the 1961 Official L i f e Table, it is very difficult t o assess the v a l i d i t y of the assumptions on m o r t a l i t y and f e r t i l i t y trends d u r i n g the period 1951-61. On the other hand, the above analysis shows that the assumptions of the p r o j e c t i o n for the decade 1951-61 m i g h t be such as to y i e l d a figure in 1961 at least equal to that revealed by the actual Count of 1961 if not higher. Hence any drastic or major revision of the assumptions, especially towards the higher side, is not advisable at the present stage. However, a* future estimates of p o p u l a t i o n for the period 1961-76 are required for perspective plan ning and for other policy purposes estimates of population f o r the period 1966-76 at 5 intervals on the basis of slightly m o d i f i e d assumptions are furnished in the foll o w i n g paragraphs. The modified assumptions together w i t h reasons for the modifications are explained below. Table 2 gives the summary result, and tables 3 and 4 give the detailed

THE ECONOMIC WEEKLY age d i s t r i b u t i o n of p o p u l a t i o n estimates for the p e r i o d 1966-76 at five year intervals on the basis of slightly modified assumptions in the l i g h t of the data revealed by the 1961 Census and the recent N S S estimates of crude b i r t h and death rates. The base p o p u l a t i o n assumed and the modified assumptions together w i t h reasons f o r the modifications are explained b r i e f l y below. The base p o p u l a t i o n for the year 1961 is taken as 442 m i l l i o n (438 m i l l i o n of actual Count + 4 m i l l i o n of net under-count d u r i n g the Census). Sex d i s t r i b u t i o n of this total p o p u l a t i o n is the same as revealed by the 1961 Census, (The number of females for every 1000 males according to 1961 Census was 9 4 0 ) . The age d i s t r i b u t i o n of this populat i o n is assumed to be the same as that of the author's projected population for 1961. As assumed in the earlier proj e c t i o n , the expectation of life at b i r t h increases at the rate of 9 year?: per decade upto the middle of the quinquennium 1961-66. On this assumption, the crude death-rate for the q u i n q e n n i u m 1961-66 works out to 16.2 and hence it is assumed that the rate of further decline in morta lity w o u l d be lower than that as sumed for earlier q u i n q u e n n i u m s . It has therefore, been assumed that,

September 16, 1961 Table 4 :

Population by Broad (in millions)

for the p e r i o d after the m i d d l e of the quinquennium 1961-66, the expectation of life at b i r t h would increase at an average rate of 0.7 years per annum as against 0.9 years per annum assumed p r e v i ously. The hypothetical L i f e Tablebased on U N M o d e l L i f e Tables adjusted for the I n d i a n mortality conditions have been used f o r projecting the 1961 p o p u l a t i o n . The expectation of life at b i r t h increases f r o m about 35 w a r s for the q u i n q u e n n i u m 1916-51 to about 49 years for the q u i n q u e n n i u m 1961-66 and finally reaches the level of about 56 w a r s by 1971-70. (These estimates are for the q u i n q u e n n i u m ending in the given year and are assumed to apply to the mid-year of the qui quermium ). N S S estimate of crude birth rate is s l i g h t l y higher than that obtained under the projection for the q u i n q u e n n i u m 1956-61. Hence

1473

Age

Groups

it is assumed that f e r t i l i t y decline would start at the end, instead of f r o m the middle, of the quinquenn i u m 1961-66. f u r t h e r the rate of decline after 1966 w o u l d be slightly lower than that assumed earlier. A c c o r d i n g l y it has been assumed that the General F e r t i l i t y Rate (defined as the ratio of total births d u r i n g a year to the mean female population in child-bearing ages 15-49 of .177 corresponding to a Crude B i r t h Rate of 43 per 1000 population assumed for the q u i n quennium 1951-56 would remain unchanged upto the end of the q u i n q u e n n i u m 1961-66 and there after decline to .170 and .160 for the quinquenniums 1966-71 and 1971-76 respectively. T h r o u g h o u t , the sex ratio at b i r t h has been assumed as 105 ( i e 105 male births for every 100 female births). Zero net m i g r a t i o n i n t o I n d i a f r o m abroad d u r i n g 1960-70 has also been assumed.