Malawi - Population sector study (Vol. 2 of 2) : Annexes

Report No. 8513-MAI Malawi Population Sector Study (In Two Volumes) Volume II: Annexes May 29, 1992 Population and Human Resources Division Southern Africa Department African Regional Office FOR OFFICIAL USE ONLY Docunmet Ottbe Wo,ld hun Ths 40cumenthas, a restncted dfistributon arid may be used by recipients on)hi, the perfomniance-of their officlaUduties Uts contents may not otherwise lbe dfisdlos without World. Bank authorization.  FOR OFFICIAL USE OnY Annex I Demographic Note 1 This document has a rusted disrbution and may be used by recipients only in the perfonnance of their official duties. Its contents may not otherwise be disclosed without World Bank authod2ation.  D eorV-hic Note A.1. Much demographic analysis has already been done for Malawi, in hotih 11'Iawi and the World Bank, on the data base extending up to and including the 1977 cens'ns . llte kesults of the analyses can be found in Volume IV of the CSO's 1977 Census Report, in the Bank's PHN Technical Note 86-20 ("The Demography of Malawi') and in the Bank's 1985 Popuilation Sector Review. Together these results formed the basis for the Bank's demographic estimates and projections as set out in the Population Sector Review. A.2. Since then, fresh data on population dynamics in Malawi have become available. In addition to the preliminary returns from the 1987 census, the full results of the 1982 Demographic Survey and the 1984 Family Formation Survey (the latter funded under the first Family Health Project) have now been published. Furthermore, the mortality and fertility data from the 1977 census have been tabulated by district, allowing detailed exploration of mortality and fertility differentials across the country for the first time; a partial analysis of these new tabulations was presented in the report of the 1984 Seminar on the Analysis of the 1977 Malawi Population Census and its Utilization in Development Planning. Thus the Bank's estimates and projections required review and updating for this Study -- though of course a further more comprehensive review and updating will again be necessary when the full 1987 census results come out. A.3. This Note gives details of two particular areas of attention: namely the district results form the 1977 census, and the new mortality data from the 1982 and 1984 surveys. Since analysis of the new ferdlity data from 1982 and 1984 did not reveal any need to change current estmates of levels and trends in fertility, this aspect of population dynamics is not treated in detail here. 1977 District Data on Mortality and Fertility A.4. 1977 district-level tables of children ever born, children surviving and births in the past year are reproduced in Annex II, Section A. The data were analyzed by the Brass P/F and child suJrival (Trussell variant) methods, using the microcomputer AFEMO PC program which is based on the techniques recommended in the UN's Manual X ("Indirect Techniques for Demographic Estimation", UN, New York, 1983); print-outs are also reproduced in Annex II, Secion B. A.5. Results of the AFEMO analysis are presented in Tables A.1 and A.2, and shown grically in summary in Figures A1-A4. Figures A.1 and A.2 map indicators of recent Ilevel of childhod morlWity and ferdlit by district; the fertility indicator is the total fertility rate (TFR) estmatd from the average of the P/P-adjusted values for the three age-groups 20-24, 25- 29 and 30-34, and the childhood mortality indicator is the value of the proportion dying before age S (q5) calcuated as the average of q5 values for the three age-groups 20-24, 25-29 and 30-34, using the Coale-Demeny South model life tables. The fertility indicators refer to the yeai before the census, or 1976-77, and the mortality indicators to a period roughly 5 years before the census, or the latter half of 1972.  t t * l. U reA.1 - 2- MA~LAVV4 _ A77 PERCE OFCH SWE e a 5, B t b tS~~~~~~M.f .~~~~~~~~~~~~~~~~ ,:K.)4 I M V ZA iEte @ MOZAMBIQES .. , l L LUN- e  ~~~~~~~~19 am A h^ FIg4a ..ZM 3 E w~~~~~~~~~~~~~~~mo TO foO fta D i7s w ZAMAMA ~ ~ ~ ~ ~ ~ ~ ALW \;.\~~~ ~~~~ lp 4, 0 nc ws -~~~~-- W L} I~~~~~~IIr ;AMA -D .N -z e MOAMIQUE=  4 - A.6. These results show a pattern of strong and consistent differentials in both mortality and fertility across Malawi. Fertility is, extremely high everywhere in Malawi, even in the two major cities of Lilongwe and Blantyre ('FRs of 7.8 and 7.7 respectively). However there is a clear cluster of spectacularly high fertility levels in the core districts of Central Region, where TFRs range from 8.7 to 9.9; the only di.strict in this range outside this area is Chitipa in the extrme north (at 8.7). Elsewhere fertility becomes in general lower moving north and south, though there also appears to be a pattern of higher, fertility along the eastern side of the country in Northern and Southern Regions (TFRs generally in the 8.1-8.4 range) and lower fertility along the lakeshore (TFRs generally 7.3-8.2) and in the core Shire highlands (TFRs of 7.2-8.1). A.7. Childhood momlity is also extremely high throughout the country, even in the two major cities (at q5s of 24 and 20 present respectively in Lilongwe and Blantyre cities). Again, however, there is a clear cluster of spectacularly high mortality levels in the core districts of Central Region, with q5s between 39 and 41 percent in Ntchisi, Dawa, Lilongwe (Rural) and Salima. These levels are unique in Malawi. From this cluster, mortality levels in general fall progressively to north and south. The outer districts in Central Region are only slightly lower, at q-s of 34-37 percent; similar levels occur also in the adjacent northernmost districts of Souther Region, Mangochi and Machinga (34-35 percent) and in the southernmost districts Chikwawa and Nsanje (34-36 percent). A cluster of lower-level mortality is found in the Shire highlands (27-32 percent) and the lowest levels of all occur in Northern Region (24-30 percent except for Mzimba, which is adjacent to Central Region, at 32 percent). A.8. Figures A.3, A.4 and A.5 display graphically by region the trends in childhood mortality in each district between the early 1960s and mid-1970s, again using q5 as an indicator; these trends are also summarized in Table A.2 into two values for two dme points, one around 1965 and the other around 1972, and percentage changes are mapped in this summary form in Figure A.6. A.9. There is clearly substantial variation across the country in childhood mortality trends, as well as levels. During the period of approximately the mid-1960s to the early 1970s, changes in mortality varied from a 10 percent decline in qs to an actual increase of 2 percent, and in much of Malawi there was little real change at all. Generally speaking, mortality levels stagnated in most of Central Region, in the southern highlands of Northern Region, and in the northern half of Southern Region; however there were marked declines in the southern half of Southern Region, in the norhern highlands of Northern Region, and (to a lesser extent) along most of the lakeshore districts. The two major cities, Lilongwe and Blantyre, also reported substantial declines in their already relatively low mortality, but the effect of migration are difficult to disentangle from any real change in mortality conditions in the cities. It will be important to repeat this analysis for the 1987 census data in order to monitor trends over more recent periods.  _ X"D =_IW"7 B SMCI IN NORNES REG 4. 42 4, 37 32 32 3 - ai). FIN~~~ -. - - ~~~~~~~~~~~~~b ~ ~ ~ ~ ~ ~ ~ ~ ~ tf X ., -- 4%V 3 43 som _A_EMO _ _ a __4 _U__ __; 444~ ~ ~ ~ ~~~sp 3a 4' 21" - i - 3 449F S ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~S  Fbo A* iiui.H OuMwwMU 1imunj MSv rCr INCENTRAL. pAwow I Um M iA3 7 CUSS OMA O14 CHftD Stfi I V 46 44 - - ppp} JS-~~ - ao ' \s a0Da 4- ~~~~~~~% ___ - -- - 'my 4% -p KPX X ., 20'. S8D 24~~~~~~~~~~~ 92~~~~~~~~~~~~~ 4~~~~~v ci 's  -7- ~ As.- 1SO neoco uonmuw iMuuArB OISWNCTINSOUT E64 REWN F LE 1W t W M AA N4 CHID SURVIVAL 4, 3 - e-_--- -. 4"~~~~~~~~%4 a9 - $- - - 4% x% -_ _ _-_ 4t~~~~~~~~4 4U~~~~~~~~~4 _=37< .x ....u9 as~~~~~~~~~~~~~~o c_ - - --__ 3 4 a b* a2 - .% - - a - - - - - - - Z 3 4a 2? #: 25~~~~~~~~~~~~~~~~~4 22 21