The Population of Turkey

Home , Hatay Province, Marmara Region, Turkish population

1974

World Population .Year

THE POPULATION OF TURKEY

CI.CR.E.D. Series

Printed by Dogu? Ltd. Çlrketl - ANKARA 19 7 4 World Population Year

THE POPULATION OF TURKEY by

Figen Karadayi, Serim Timur, Miroslav Macura, Sanaira Yener, Halûk Cillov, Handan Tezmen, Sunday Üner, Hüsnü A. Ki§ni§çi

The Institute of Population Studies Hacettepe University Ankara

PREFACE Until recent years, the importance of population growth as a determinant in social and economic changes has been underrated and little attention was paid to the characteristics of a population. Dur- ing the short period of time following the Second World War, the accelerated population growth rate in many countries has drawn our attention to the population problems. Especially in the less developed countries numerous attempts were made to solve the serious problems related to rapid population growth, and several research works were initiated. Today, population growth and the composition of the population has been recognized everywhere both as the cause and determinant of social and economic development. During the last decade, similar attemps and studies have been made in Turkey, even if not as numerous as elsewhere. Turkey as a newcomer in this field has information and data on the subject of population, but only very limited and somewhat widely scattered. The Institute of Population Studies, being the only one of its kind in Turkey, has established a documentation center which com- piles available information and collects data for the use of scientists, planners and politicians. At the same time, in preparink a monographp for the 1974 Pop- ulation Year, we have attempted to follow the outline as suggested by UNFPA and CICRED ; past, present and future progress has been explained accordingly. The monograph is divided into seven chapters as listed in the Table of Contents. Whereas some of the facts used in the preparation of this . monograph are not quite up to date, others are the direct products of research projects conducted in very recent years. The works of several authors in this field are contained in this monograph. Thus, we sincerely hope that the presentation of this monograph will be of value to policitians, leaders, and administrators, as well as create a public interest in this field. It will also be an important source for social researchers. Gratitude is herewith expressed to the authorities of UNFPA and CICRED for making the financial arrangements, and the authors of the chapters of this monograph - Dr. Halûk Cillov, Figen Kara- dayi, Dr. Miroslav Macura, Handan Tezmen, Dr. Serim Timur, Dr. Sunday Üner, and Samira Yener for their contributions. I would also like to thank Hasan Serinken as the coordinator and Alev Bosuter, Ismail Igsever, Ergun Törüner, Mehmet Ögec and §ükrü Ögec for their services in typing and designing; Güliz Kunt and Türkiz Gökgöl for their help in editing and Feyzullah Ertugrul for his services in the printing of the monograph.

Dr. Hiisnti Ki§ni§çi Director Institute of Population Studies Hacettepe University CONTENTS

Page Preface 5 Chapter I Population Growth in Turkey, by Figen Karadayi, M. A. 11 Chapter II Components of Growth Section A.... Fertility, by Serim Timur, Ph. D 27 Section B .... Mortality, by Miroslav Macura, Ph. D 39 Section C .... International Migration, by Samira Yener, 46 M. A. Chapter in The Structure of The Turkish Population, by Halûk Cillov, Ph. D 57 Chapter IV International Migration and Population Redistribution in Turkey, by Handan Tezmen, M. A 87 Chapter V The Labour Force, by Sunday Uner, Ph. D Ill Chapter VI Population Projections, by Samira Yener, M. A 127 Chapter VII Economic and Social Implications of Population Growth and Population Policy, by Hüsnü A. Ki§ni§çi, M. D 139

Chapter I

POPULATION GROWTH IN TURKEY by Figen KARADAYI, M. A.

POPULATION GROWTH IN TURKEY

Figen Karadayi, M. A. Institute of Population Studies Hacettepe University

During the period 1650 -1800, the devolpment of new techniques in agriculture, in manufacturing, in transportation and communication had increasingly important effects on the population growth in the west. Statistics of European countries show that in most of them the increasing rate of population growth was due to the decline in the death rate rather than to any change in the birth rate (Thompson, 1965). During the first half of the nineteenth century, in the west as a whole, the death rate fell rapidly to the lower 20ies per thousand, whereas the aggregate birth rate probably remained almost stationary and the rate of natural increase averaged 11-12 per thousand per year (Thompson, 1965). In the nineteenth century, while the western countries were characterized with a rapid population growth, in the Otoman Empire there was no significant increase in the population. However, the Turkish statistical information during this period was insufficient and not reliable for making comparisons with European countries. In the twentieth century, most of the western countries had reached a low level in death rates and moreover, the birth rates had also declined to such a low level that the gap between deaths and and births was diminishing at a fairly rapid speed, leading to a slower rate of population growth. The historical experience of the western countries diffused to developing countries like Turkey after a century. However, it produced even higher rates of population growth. After the establishment of the Republic in 1923, Turkish govern- ments followed a strong pronatalistic policy for many years, in order to compensate for the great loss in population during the pre- ceeding years of war. II THE POPULATION OF TURKEY

Trends in the Actual Rate of Population Growth

During this century, more reliable statistical information on the characteristics of population and its growth was collected. Al- though there were several population registrations during the Otto- man Empire, the actual population census on a scientific bases was conducted for the first time in 1927. This was followed by the 1935 census after which the censuses were repeated at five year intervals. In the 1927 census, the Turkish population was recorded as 13,648,000 (Table 1). As it can be observed from the table, the population increase rate has been rather fast during this century. According to the results of the 1935 census, there was an increase of 2.5 million in the population as compared with the previous census. During the 1927-1935 period, the actual rate of increase in population was 21 per thousand. However, this high rate of inci'ease was partly due to the increase in the number of immigration (around 200,000) to the mainland from the lands lost in the war. Secondly, it was partly due to the under-enumeration of population (around 400,000) in the previous census of, 1927 (Cillov, 1980). Thus considering two factors, the actual population increase between the two dates becomes 1.9 million instead of 2.5 million, consequently the natural rate of increase becomes 16 per thousand, somewhat lower than the actual rate of increase. However, this rate was still one of the highest rates compared to other countries. Between 1935 -1940 the Province of Hatay with its 208,000 population was added to the lands of Turkey, and at the same time 120,000 immigrants entered the country. Thus, if these factors are considered, a natural rate of increase is expected to be somewhat lower than the actual. In the period 1940 -1945, which was during the Second World War, there was a decline in the annual rate of increase, dropping from 19.6 per thousand to 10.6 per thousand. Firstly, there was no observable migration movement in this period. There were also other reasons which had depressing effects on the rate of natural increase.

12 POPULATION GROWTH IN TURKEY

The main reason was the fact that the majority of the male population was enlisted. Also, due to limited economic means the number of marriages, pregnancies, and births were decreased voluntarily. Decrease in the births was also attributable to the fact that the proportion of the fertile population in the 20 - 35 age group was low because of the First World War. On the other hand, the death rates during the period were also high.

Table 1

Population, and Intercensal Growth Rates According to the Censuses Actual Rate of Population Population Growth Census Year (thousands) (per thousand) 1927 13,648 21.1 1935 16,158 19.6 1940 17,821 10.6 1945 18,790 21.7 1950 20,947 27.7 1955 24,065 28.5 1960 27,755 24.6 1965 31,391 25.6 1970 35,661 ~" ""

SOUTCG : Census of Population. Social and Economic Characteristics of Population. 24 Oct. 1965, State Institute of Statistics. Printine Division Publication No. 568, Ankara 1969.

13 THE POPULATION OF TURKEY

During the period 1945 -1950 following the war, there was again a sharp rise in the rate of increase to 21.7 per thousand, which was due to several factors. Mainly it was affected by the increase in marriages and births after the demobilization of the male population. It is also clear that, after the wars, even the number of births in the old marriages increased. Secondly, it was due to a sharp fall in the death rate as a consequence of the improvement of health and medical services, and the increased use of antibiotics.

In the period 1950 -1955, the actual rate of increase was 27.7 per thousand. This was extremely high compared to other European countries. However, if we disregard the influence of 125,000 immigrants, the natural rate of increase becomes a little lower, but is still a high rate. This was again due to the rise in the number of births as the result of an increase in the marriage rate and in the> proportion of women in reproductive ages. Table 2. shows that the proportion of reproductive women has risen to 20 percent in 1950. Among these the proportion of married women was 54.6 percent in 1950 and 61.9 percent in 1955, whereas it went down to 59.3 percent in 1960. Table 2

The Proportion of Total and Married Women in the 15 - 35 Age Group by Census Years Proportion of Proportion of Women in 15 - 24 Married Women in Age group to Total 15 - 24 Age Group Census Year Female Population (%) (%) 1935 13.8 55.4 1940 13.6 — 1945 17.4 48.8 1950 20.1 54.6 1955 18.6 61.9 1960 16.1 59.3

Source : Sarç Ömer Celftl, "Turkish Population Increase" (Turkish), Tipta Yenilikler, Eczaci- basi Has Fabrikasi, No. 9, Nov. 1967, p. 22-24.

14 POPULATION GROWTH IN TURKEY

Between 1960 -1965 a decline in the rate of increase became fairly apparent. However, this was mainly due to the international migration, where 200,000 workers at reproductive ages left Turkey to work in foreign countries. Secondly, the proportion of population in reproductive ages was somewhat lower (Table 2). This was due to the fact that, the cohort born during the second world War came to reproductive ages during this period. So a decline in the natural increase was not attributable to the change in the fertility trend, but to the change in the age structure of the population. In 1965 -1970 period, the actual rate of increase became slightly higher with a rate of 25.6 per thousand, compared to the rate in the preceding period, which was 24.6. Increase in the rate was again due to the increase in the proportion of the population in reproductive ages. In this period, larger cohort born after the war in the period 1945 -1950, came to reproductive ages. However, the negative influence of emmigration of workers to foreign countries is still present. In summary, from the first census of 1927 to the last one in 1970, the population of Turkey had increased from 13,648,000 to 35,661,000. The annual rate of increase has always been high for each census period except for the war years. With 25 per thousand rate of population increase, Turkey's population doubles itself every 28 years. This rate is the highest of all the European countries where the rate is below 10 per thousand (Table 3). On the other hand, Turkey has a slightly lower growth rate than some of the Middle Eastern and Latin American countries, where it goes up to 3.4 per thousand .

15 THE POPULATION OF TURKEY

Table 3 Population Growth Rate of Several Selected Countries. Annual Rate Annual Rate of Population of Population Country Growth (%) Country Growth (%) France 0.7 Japan 1.1 Germany (Republic) 0.4 Syria 2.7 Norway 0.9 Arab Republic 2.8 Sweden 0.5 Iraq 3.4 Bulgaria 0.7 Israel 2.4 Rumania 1.3 Pakistan 3.3 Netherlands 1.1 India 2.6 U.S.S.R. 1.0 Mexico 3.4 Greece 0.8 Turkey 2.7 U.S.A. 1.1

Source : 19Y1 World Population Data Sheet. Population Information ior 145 countries. The high rate of population growth in Turkey was basically influenced by a high rate of fertility and by the gap between the births and deaths. The role of immigration to the country is rather negligible. Thus it is necessary to look over the changes in the different components of population growth. Changes in the Components of Population Growth Population growth in a country changes under the combined influence of several factors. These are basically fertility, mortality and migration. Where the gap between birth rates (B.R.) and death rates (D.R.) gives the natural increase (NRI) of a country, this, combined with the influence of migration (M.R.) constitutes the actual rate of population growth (R). B.R. — D.R. + MR = R

In order to observe the trends in natural rate of increase, the contribution of fertility and mortality should be studied. 16 POPULATION GROWTH IN TURKEY

Changes in Fertility by Census Years Depending on the age distribution of quinquennial Turkish censuses from 1935 to 1970, changes and trends in birth, death, migration rates and population growth have been studied by Shorter (Shorter, 1973). According to the results of this study, it is observed that over 35 years CBR has dropped steadily from 51.0 per thousand to 34.4 per thousand, which was about a 24 percent decrease in the birth rate (Table 4). The current rate of 38.4 per thousand is still very high compared to other European countries which have rates of 13-17 per thousand. However, Turkey has a lower rate when compared to some Middle Eastern countries such as 50 per thousand in Pakistan, 48 per thousand in Iran, 47 per thousand in Syria (Population In- formation, 1971). The decline in CBR in 1960 -1965 may be basically attributable to the changes in the age structure of the population. During this period, the proportion of the reproductive population is low due to the arrival of war cohort at child bearing ages. The depressing effect of this, WiH continue until 1970 when the normal size-cohorts come to child bearing ages. Table 4 Crude Birth and Death Rates and Natural Rate of Increase of Population by Census Periods Natural Rate Crude Birth Crude Death of Population Rate Rate Increase Census Periods (per thousand) (per thousand) (per thousand) 1935 -1940 51.0 34.6 16.4 1940 -1945 50.2 39.6 10.4 1945 -1950 47.8 28.1 19.7 1950 -1955 46.7 20.7 26.0 1955 -1960 45.0 17.8 27.2 1960 -1965 41.2 14.7 26.5 1965 -1970 38.4 12.7 25.7 Source : Shorter, C. Frederic and B. Tekçe. "Demographic Determinants of Urbanization in Turl&ey, 1935-1970. Forthcoming Pub- lication : Benedict, P., E. Tümertekin and F. Mansur (eds.) Turkey : Geographical and Social Perspectives. Leiden : E. J. Brill, 1973. 17 THE POPULATION OB1 TURKEY

Changes in. Mortality by Census Years Over the past 35 years the death rate has also fallen steadily from 34.6 per thousand to 12.7 per thousand Which is about a 47 percent drop in the original rate (Table 4). The death rate, compared to the birth rate has experienced a more rapid decline in the same period. It rises up to 39.6 per thousand during the war. years, and drops rapidly thereafter, especially in the two following census periods of 1945 -1950 and 1950 -1955, because of the improvements in the health conditions. Then the decline becomes more gradual. The rate in Turkey is slightly higher than that of the European countries where it is around 7 -11 per thousand, but it is lower than that of India, Pakistan, or Iran. Bole of Natural Increase in Population Growth This shows the gap between the CBR and CDR which is the most important contributing factor to population growth. Changes in the natural rate of increase are given in Table 4. The biggest decline in NRI is between the periods 1935 -1940 and 1940 -1945. This is mainly due to the increase of deaths during the war. Then it rises rapidly in the post war period of 1945 -1950. This is mainly due to the application of better medical technology with the result of a sharp decline in the death rates. Because of this decline in the period 1950 -1955 is also ¡big, the natural rate of increase goes up again to 26.0 per thousand. Thereafter the death rates experience only a gradual decline. Thus the rate of natural increase drops gradually parallel to the gradual decline in the gap between birth and death rates. In summary, it is clear that the excess of birth rate over death rate, increased sharply after the period of 1935 -1940 up to 1950 and declined sharply thereafter. Bole of Migration in Population Growth The joint contribution of the natural rate of increase and international migration gives the total actual rate of population growth of the country. The influence of migration may be positive as the form of immigration, or negative as emmigration. The influence of migration has been small but positive until 1960 (Table 5).

18 POPULATION GROWTH IN TURKEY

Table 5 Changes in the Natural Rate of Increase and in International Migration as Components of Actual Population Growth in Turkey, by Census Period 1935 -1970.

Natural Rate Rate of Inter- Actual Rate of of Population national Migra- Population Increase tion Increase Period (per löiousand) (per thousand) (per thousand) 1935 -1940 16.4 3.2 19.6 1940 -1945 10.6 — 10.6 1945 -1950 19.7 2.0 21.7 1950 -1955 26.0 1.7 27.7 ; 1955 -1960 27.2 1.3 28.5 ' 1960 -1965 26.5 —1.9 24.6 1965 -1970 25.7 —O.I 25.6

Note : The rate of international migration was calculated by subtracting the natural rate of increase from the actual rate of population increase. (Crude Birth Rate - Crude Death Bate) + International = Actual Bal« of population Natural Bate of Increase Migration Increase

However, after the 1960 flow of Turkish workers to foreign countries reversed the influence of migration to be negative the rate of actual population growth decreased to 24.6 per thousand (Table 5). The decrease in the actual rate of population growth after 1955 -1960 Was basically due to the role of emigration, the rate of natural increase is not very different from the previous period.

In Turkey, the role of natural increase has been dominant compared to the role of migration, in determining actual population growth. Turkey's population growth has been basically influenced with the rapid decline in the death rates which increased the gap between the births and the deaths. International migration played rather a minor role. Its role increased in importance especially after the 1950 's. The relationship of the three components of growth can be clearly seen in figure 1. The larger proportion of population growth is by the natural increase of population, that is, the gap between the crude birth and death rates. The gap increases after 19 Figure 1. Crude Birth and Death Rates, and Rates of Increase of Population, 1935 -1970

1935 19¿0 19/.5 1950 1955 1960 1965 1970 55-i— i i i i

50- Crude Birth Rate Z.S-

Z.0-

35- •z. o 30- o0. a.

20- < in o 15- Crude Death Rate

a. 30-

Net immigration % 25H

20- \ Net Emmigration

15-

10-

0 1935 19¿0 1950 1955 1960 1965 1970

CENSUS YEARS

Source :. Shorter, C. Frederic and B. Tekçe. "Demographic Determinants of Urbanization in Turkey, 1935-1970. Forthcoming Publication: Benedict, P, E. Tumertekin. and F. Mansur (eds.) Turkey: Geographical and Social Perspectives. Leiden: E. J. Brill, 1973. •

20 POPULATION GROWTH IN TURKEY the 1940 - 45 period with the sharp decline in death rates. In the lower part of the chart, it is observable that international migration adds to population growth until the 1950's while it shows negative effects after the 1960 - 65 period. When the growth of separate settlements in Turkey is considered, the role of migration in terms of rural to urban migration becomes more important. Population Growth in Rural and Urban Places Together with the problem of a high rate of population growth, the rapid growth of urban population is another problem facing Turkey. During the 30 - year period of 1935 -1965, as the population of Turkey doubled, the population of the urban areas (places over 10,000 population) increased three and one half times (Seetharam, 1973). In fact, the growth of the urban population in time is due to the influence of several factors. These are : the natural increase due to the gap between fertility and mortality, international migration, birth of new urban places (with population over 10,000), and mainly rural to urban migration (Shorter, 1973). Among the components determining urban growth in the 1965 -1970 period, rural to urban migration rate is 52.4 per thousand which is the highest among the other components (Shorter, 1973). Then comes the birth of new cities with 20.5 per thousand. International migration and natural increase play a depressing role in the population growth in urban areas. Differential natural increase is minus 7.7 per thousand and differential international migration is minus 3.5 per thousand. Altogether the components make up an urbanization rate of 61.7 per thousand (Shorter, 1973). It is clear here that the role of rural to urban migration as one component of urban population growth is highest compared to other factors. The influence of natural increase in determining rapid growth of urban population is minor. The gap in death rates in urban and rural areas is smaller than the gap in fertility. Urban ¡birth rate is lower than the rural birth rate whereas the population growth of urban places is higher than the growth in rural areas. The natural rate of growth is 27.2 per thousand in rural areas whereas it is 20.7

21 THE POPULATION OF TURKEY per thousand in urban areas (Turkish Demographic Survey, 1970). The increase in the number of urban areas has been two and one half times from 81 in 1935 to 199 in 1965. It is ciar from Table 6. that the rate of urban population has always been greater than the rate in rural areas. Since the rate of natural growth, on the contrary, is higher in rural areas, the higher rate of growth in urban areas is basically provided by the rural to urban migration. The whole period from 1935 -1965 can be divided into two stages where the first shows almost no increase in percent urban and the second is a period of rapid change (Table 6). The larger proportion of increase in percent urban, that is 12 percent increase, occured from 1950 to 1965. Although the urban growth rates were always higher than the rural rates, the difference has been greater since 1950. Also, the rates of urbanization show rapid increase after 1950. The process of urbanization in Turkey becomes faster as the population of the city becomes larger. So, the rate is higher in the metropolitan areas. Table 6 Rate of Urban and Rural Growth and Rate of Urbanization from 1935 to 1970 Census Percent Urban Growth Rural Growth Rate of Urbani- Year Urban Rate (%) Rate (%) zation (%) 1935 16.5 4.0 1.3 1.9 1940 18.1 1.5 1.0 0.4 1945 18.5 2.6 2.2 0.3 1950 18.8 7.6 1.9 4.6 1955 22.5 6.5 2.1 4.3 1960 25.9 6.0 1.4 5.2 1965 29.8 Source : Snetharam, 1C. S. Urbanization in Turkey, 1935-1965, a contributed paper submitted to Section 2.3 on Bis Cities of the General Conference of the IUSSP, Liege, August- September, 1973. The total growth rate, contribution of natural increase, and the time for doubling the population are given in Table 7. The difference between the total growth and natural increase may be roughly attributable to the influence of rural to urban migration which seems to play the greatest role in the population growth of the three metropolitan areas.

22 POPULATION GROWTH IN TURKEY

Table 7

Natural and Total Growth Rates of Three Metropolitan Areas in Turkey- Actual Doub- Natural Growth Total Growth ling Time City Rate (%) Rate (%) (years) Ankara 2.2 7.8 9 Istanbul 1.2 4.1 17 Izmir 1.5 4.7 15 Source : Anderson, Lewis S., Socio - economic Factors in. the People Boom. Presented at the Ankara Chapter of the Society for International Development, June, 1970.

REFERENCES — Anderson, Lewis S,, "Socio-economic Factors in the People-boom". Presented at the Ankara Chapter of the Society for International Development, June 22, 1970. — Cillov, Halûk, Nüfus ístatistilderi ve Demografinin Genial Esaslan (Population Statistics and the Principles of Demography), Ser- met Matbaasi, Istanbul 1960, p. 48-55. — Cilov, Halûk, "Demographic Factors on the Economic Develop- ment of Turkey", International Union for the Scientific Study of Population General Conference; London, September 1969. — Census of Population, Social and Economic Characteristics of Po- pulation, 24 October 1965. State Institute of Statistics, Printing Division Publication, No. 568, Ankara, 1989. — Population Census of Turkey, 25 October 1970. — Temporary Results. State Institute of Statistics, Publication no: 616, An- kara, 1970. — Samara Yener, 1960 -1965 Dönemiude Köyden §>ehire Gögler (Rural to Urban Migration in the Period of 1960-1965). State Planning Organization, Publication No : 212, Ankara, July, 1970. — Sarc, Orner Celâl, "Türkiye'de Nüfus Artisi" (Population In- crease in Turkey). Tlpta Yenilikler — Eczacibasi îlâç Fabrikas.i No. 9, November 1967, p. 22-24.

23 THE POPULATION OF TURKEY

— Seetharam, K. S., "Urbanization in Turkey, 1935-1965", a Con- tributed paper submitted to Section 2.3 on Big Cities of the Ge- neral Conference of the IUSSP, Liege, August - September 1973. — Shorter, F. C, "Information on Fertility Mortalit and Population Growth in Turkey" Turkish Demography, Proceedings of a Con- ference, Feb. 21124i 1968. Hacettepe University, Institute of Pop- ulation Studies, Publication No. 7, 1969. — Shorter, F. C. and Belgin Tekçe, "Demographic Determinants of Urbanization in Turkey, 1935-1970". Fortcoming Publication in Benedict, P., E. Tiimertekin and F. Mansur (eds). Turkey : Geographical and Social Perspectives. S. Leiden; E. J. Brill, 1973. — Statistical Yearbook of Turkey, State Institute of Statistics, Printing Division, Puglication No. 580, Ankara, 1969. — Thompson, W. S. and David Lewis, Population Problems, Mc- Graw-Hill Book Company, U. S., 1965. — Turkish Demographic Survey : Vital Statistics 1966-67, School of Public Health, Ankara, 1970. — Türldye Nüfusu Üzerine fotatistiksel Çahsmalar, (Statistical Studies on the Turkish Population) Devlet Planlama Teskilâù, Yayin No. 1012, Ankara, March, 1971. — 1971 World Population Data Sheet, Population Information for 145 countries.

24 Chapter II

COMPONENTS OF GROWTH

Section A-Fertility

by Serim TtMUR, P.h. D.

Section B - Mortality by Miroslav MACURA, Ph.D.

Section C - International Migration

by Samira YENER, M. A.

COMPONENTS OF GROWTH

Section A — Fertility

Serim Timur, Ph.D. Institute of Population Studies Hacettepe University

Taken as a whole the population of Turkey may be placed in the early stage of the demographic transition. The crude birth rate for the country as a whole is as high as 40 per thousand, or using more refined measures, the total fertility rate is 5.3 and the gross repro- duction rate is 2.61. The aggragate national fertility measures, however, do not reveal the striking rural-urban and regional differences. Fertility is lowest in the metropolitan centers of Istanbul and Izmir which are at the western end of the country, where economic and social development is highest. Fertility in rural areas of the more developed west is also low as compared to the less developed central1 and eastern parts of the country. The level of socio-economic development indicates that metropolitan cities may be in the second phase of fertility transition while the more backward rural areas may be in the initial phase. Hence, as would be expected by the historical pattern of fertility differentials in the world, we find a strong inverse relation between fertility and socio-economic characteristics of' couples in metropolitan cities and a weak inverse or even a direct relation in rural areas. We shall succesively discuss fertility differences : i) between urban and rural areas and geographic regions, and ii) according to some socio-economic characteristics such as educational attain- ment, income, and husband's occupation2.

1 Vital Statistics from the Turkish Demographic Survey 1366-67 Ministry of Health and Racial Welfare, School of Public Health Publication, Ankara: 1370, pp. 121-126. 2 Additional Information and a more detailed discussion on fertility differentials and socio- economic characteristics in Turkey can be found in Timur Serim, "Socio-Economic Deter- minants of Differential Fertility in Turkey", paper submitted to the Second European Population Conference, Strasbourg, 1-7 Sept., 1971. 31 pages, and in Karadayi Figen, Soclo-Economlc Correlates of Fertility Behavior and Attitudes In Turkey", Master's Thesis. Hacettepe University, Ankara, 1971. • • •

27 THE POPULATION OF TURKEY

Regional Differences The basic fertility estimates of the main five geographic regions and urban-rural strata which are derived from the Turkish Demo- graphic Survey of 1965 -1967 are shown in Table l.The crude birth rate in rural areas with less than 10,000 population is 44 per thousand whersas it is 31 per thousand in the urban areas. The corresponding total fertility rates are 6.1 and 3.9 respectively. Fertility is lowest in the metropolitan areas of Ankara, Istanbul and Izmir. Crude birth rates are 31, 25, and, 24 per thousand respectively.

TABLE 1 Urban - Rural and Regional Fertility Rates in Turkey Crude Birth Total Rate Fertility Region/Area (Per 1000) Rate Total 39.6 5,303.5 TURKEY Rural 43.9 6,124.0 Urban 31.4 3,880.5 Ankara 31.2 3,299.5 METROPOLITAN AREAS Istanbul 24.2 2,654.0 Izmir 24.8 2,916.0 REGION I Rural 52.7 7,095.5 (Central Anatolia) Urban 34.7 4,293.5 REGION H Rural 38.1 5,153.0 (Black Sea Coast Urban 34.7 4,343.5 REGION DI Rural 35.6 4,932.5 (Western Turkey) Urban 26.5 3,140.5 REGION IV Rural 40.3 6,072.0 (Mediterranean) Urban 31.9 4,278.0 REGION V Rural 49.7 7,436.0 (Eastern Turkey) Urban 43.9 6,095.0

Source : Vital Statistics from the Turkish Démographie Survey 1966-67, Ministry of Hea'th and Social Welfare, School of Public Health Publications, Ankara, 19T0. p. 37 - 39.

28 COMPONENTS OF GROWTH

Regional differences are nearly as sharp, with Eastern Turkey (Region IV) yielding the highest fertility followed by Central Ana- tolia, (I) the Black Sea Coast (II), the Mediterranean Region (IV), and Western Turkey (HI).

1968 Haoettepe Population Survey The data for the following sections of this chapter are taken from a nationwide multi-stage probability sample survey on "Family Structure and Population Problems in Turkey" conducted by the staff of the Hacettepe Institute of Population Studies in 1968. The estimates presented here are based on data relating to 3,200 currently married women under 45 years of age and their husbands.

Fertility Measure Used The basic measure of fertility used in this chapter is the average number of live-born children of currently married women in reproductive ages (44 years old and younger). In order to minimize the difficulties that may arise in comparing groups with differing compositions of marital duration, data on children ever^born have been standardized for duration of marriage, using as the standard the marriage duration distribution of all women in the sample. In comparing the standardized means of live^born children, in general, a difference of 0.5 may be taken to ibe statistically significant at the 5 percent level.

Urban-Rural Residence The earliest and most rapid declines in the birth rates in the West took place in the largest cities, and the latest and slowest were seen in rural areas, while the smaller towns stayed between the two extremes. The same pattern is observed in Turkey when the mean number of live-born children are compared by community size. The average number of live-born children for all Turkish women in reproductive ages is 3.9. This figures is higher (4.2) in rural areas with less than 2,000 population and considerably lower in metropolitan areas (2.7). Small towns and other urban areas are

29 THE POPULATION OF TURKEY in between, with an average live birth of 3.8 and 3.4 respectively (Table 2).

TABLE 2 Number of Live-born Children to Currently Married Women in Reproductive Ages, by Community Size Currently Married Standardized Women Aged 15-44 Mean No. of Number Community Size Live born Per cent of Sample (Population) Chilren Distribution Women Less than 2,000 4.2 60.8 1423 2,000 -14,999 3.8 13.2 . 337 15,000 and over 3.4 15.7 1189 Metropolitan (Ankara, Istanbul, Izmir) 2.7 10.4 318

TURKEY 3.9 100.0 3267

Education of Women It is usually believed that one of the reasons for fertility decline in developed countries is the educational advancement of the general population. As the level of education rises, fertility usually declines in both economically developed and underdeveloped countries. Furthermore, numerous studies have shown that the education of the wife is more strongly associated with fertility than the education of the husband. In Turkey, where only about one third of the women are literate, educated women form a highly select group, and declines in fertility are expected even at very moderate levels of education. Although illiterate women have almost 3 times as many births as women university graduates, a marked reduction in fertility is also observed

30 COMPONENTS OF GROWTH with as low as 5 years of education. As shown in Table 3, the average number of live births is 4.2 for illiterate women, 3.2 for non-educated literate women, 2.8 for those with primary school education, 2 for those who have completed secondary school or lycee, and 1.4 for university graduates. In other Words, mere literacy reduces fertility by 24 percent; 5 years of education readuces it by 33 percent; both 8 and 12 years of education, 'by 50 per cent; and 16 years of education by 67 percent. However, rather than a. regular accelerating effect of education on fertility, we may distinguish three turning points where fertility, is reduced by one fourth from the preceding educational category. These three points of marked break are mere literacy, secondary school and university education. As in all other socio-economic characteristics, there is a marked relation between urban-rural residence and female education. Although 80 percent of the metropolitan women are literate, this proportion is 43 percent in other urban areas, 39 percent in small towns, and only 17 percent in the villages. Does, however, a specific educational level affect fertility at the same rate in both urban and rural areas? Since there are few women of higher educational levels in the rural areas and small towns, the effect of educational level on fertility by urban - rural residence can be seen by combining middle school, lycée, and university levels (Table 4). The data by community type reveal that illiterate women have virtually the same high level of fertility in metpopolitan, urban, and rural areas. When we look at the degree of change in fertility associated With educational level, we see that mere literacy reduces fertility by almost 40 percent in metropolitan centers, whereas this proportion is 20 percent in other urban areas, 30 percent in towns, and only 12 percent in the villages.

31 THE POPULATION OF TURKEY

TABLE 3

Mean Number of Lave Born Children by Woman's Education and Reductions in Fertility by Educational Level (Standardized by Duration of Marriage, Women 14-44) Percent Percent Percent Reduction in Reduction in Distribution Fertility Fertility of. Mean from the from Women Educational No. of Category of Preceding in each Level LBC Eliterates Category Category

Illiterate 4.2 — — 68.2 Did not complete primary school 3.2 24 24 12.0 Completed prim. Sch. (5 years) 2.8 33 13 15.9 Completed Sec. Sch. (8 years) 2.1 50 25 2.1 Completed Lycée (12 years) 2.0 52 5 1.3 University grad. (16 years) 1.4 67 30 0.5 TOTAL 3.9 100.0 NUMBER 3267 3267 3267 3267

Primary school education, likewise, reduces fertility by 40 percent in metropolitan centers but by only 25 percent in rural areas. Among literate and primary school educated peasant ¡women, fertility is much higher than among women of the same educational level in urban areas. Thus, both urban-rural residence and education interact to affect" fertility. Although education produces a reduction in fertility irrespective of community type, a specific level of edu-

32 COMPONENTS OF GROWTH cation does not produce the same level of fertility in both urban and rural areas. In other words, educational level and fertility are highly interrelated, but most of the effects of education operate through residence, metropolitan residence being especially influential.

TABLE 4 ; Mean Number of Live-born Children by Education of Woman and by Community Type (Standardized by duration of marriage, women 14-44)

Community Type Metro- politan City Town Village Turkey Education of Women LBC LBC LBC LBC LBC

Illiterate 4.1 3.8 4.3 4.3 4.2 (Percent distribution) (20) (51) (61) (83) (68) Did not complete prim. 2.5' 3.1 3.1 3.8 3.2 Sch. (Percent distribution) (23) (18) (15) (8) (12) Completed primary Sch. 2.6 2.7 2.6 3.2 2.8 (Percent distribution) (33) (26) (21) (9) (16) Secondary and over 1.9 2.9 — 2.1 (Percent distribution) (24) (5) — (4)

TOTAL LBC 2.7 3.2 3.8 4.2 3.9 TOTAL PER CENT (100) (100) (100) (100) (100) NUMBER 318 1189 337 1423 3267

Ratio of highest to lowest (Lowest = 1) 2.16 1.41 1.65 1.34 2.00

Husband's Occupation As expected by classical trends, the differentiation of occupational groups with respect to fertility is most pronounced among

33 THE POPULATION OF TURKEY the professionals and farmers (Table 5). While the former have an average of 2.3 live-born children, the latter have an average of 4.4. This comparison, however, must be tempered with the fact that farmers constitute the largest occupational category (51 percent), whereas professionals constitute the smalest (2 percent), Commerce and businessmen occupy the second lowest fertility category followed by clerks and! junior civil servants. After farmers, workers, and artisans are in the next two high fertility categories. The same general pattern is observed in metropolitan centers but fertility is lower than the national average in every category, except for workers. In other urban areas, once more professionals and civil servants have the lowest fertility whereas commerce and businessmen as well as farmers (absentees) have the highest fertility. In all existing categories, towns have a higher fertiEty than metropolitan and urban areas. In rural areas where businessmen and professionals are non-existent, fertility is highest for all the remaining categories and no differentials are observed.

Income A classification of families by per capita income shows a regular decline in fertility with rising levels, of income (Table 6). In order to maximize the accuracy of income reporting, questions were asked about the income of each member of the household, as well as each possible source of income in cash or in kind. In this context, per capita income is a more refined variable than total household income. It controls for the total number of family mem- bers and earners. Otherwise, some of the extended families may be included in the highest income bracket in spite of the fact that their standard of living may not be high.

34 COMPONENTS OF GROWTH

HABLE 5 Mean Number of live-Born Children by Husband's Occupation and by Community Type (Standardized by Marriage Duration, Women 14-44)

Community Type Metro- Husband's politan City Town Village Turkey Occupation LBC LBC LBC LBC LBC

Commerce and Businessmen 2.4 3.7 X — 2.8 (Percent distribution) (6) (3) (2) — (2) Professionals 1.9 3.0 X — 2.3 (Percent distribution) (8) (2) (1) — (2) Clerks and junior civil servants 2.9 3.0 3.8 — 3.1 (Percent distribution) (31) (22) (13) (1) (10) Artisans and small scale retailers 2.4 3.4 4.0 3.9 3.5 (Percent distribution) (32) (39) (33) (8) (20) Workers 4.0 3.3 3.3 4.4 4.4 (Percent distribution) (23) (29) (25) (6) (15) Farmers and farm workers — 3.5 4.8 4.4 4.4 (Percent distribution) — (5) (25) (85) (51)

TOTAL LBC 2.8 3.3 3.8 4.3 3.9 TOTAL PERCENT (100) (100) (100) (100) (100) NUMBER 238 250 937 251 949 Ratio of highest to lowest (lowest = 1) 2.11 1.23 1.26 1.13 1.91

x Lass than 20 cases 35 THE POPULATION OF TURKEY

TABLE 6 Mean Number of Live-Born Children by per Capita Income and by Community Type (Standardized by Marriage Duration, Women 14-44)

Annual per Metro- capita family politan City Town Village Turkey income in TL. (x) LBC LBC LBC LBC LBC

1-499 — 4.4 5.3 4.6 4.6 (Percent distribution) (6) (16) (38) (26) 500 - 999 4.7 4.2 4.2 4.3 4.3 (Percent distribution) (3) (18) (22) (29) (24) 1,000 -1,999 4.1 3.5 3.7 4.0 3.8 (Percent distribution) (22) (36) (30) (16) (22) 2,000 - 3,999 2.7 2.8 2.8 3.4 3,0 (Percent distribution) (34) (28) (20) (ID (17) 4,000 + 2.2 2.3 3.1 3.7 2.7 (Percent distrbution) (40) (12) (13) (6) (12)

TOTAL LBC 2.8 3.4 3.8 4.2 3.8 TOTAL PERCENT (100) (100) (100) (100) (100) NUMBER 249 919 248 1005 2421 Ratio of highest to lowest (lowest = 1) 2.14 1.91 1.89 1.35 1.70

x 1 U. S. dollar = 14 Turkish liras The number of children ever-born is 4.6 in the lowest income bracket whereas it is 2.7 in the highest income group (Table 6). Although regular but slight decreases in fertility are observed among the three lowest categories within each community type, a break in fertility occurs only in the fourth income group. It should be noted that among the three lowest categories comprising 72 per-

•36 COMPONENTS OF GROWTH cent of the total population, there are virtually no differentials by community type. In other words, 25 percent of the metropolitan population, 60 percent of the urbanités, 68 percent of the town population and 83 percent of the villagers earning less than 2,000 TL. per capita have the same high fertility (about 4 live births.) The most interesting finding in Table 6, however, is that fertility exhibits a U-shaped curve in towns, and villages; the highest fertility being found in the lowest income bracket, and gradually decreasing to the highest group, where it increases slightly. Thus, rural fertility seems to be positively related to income in the highest group. This pattern perfectly fits the classical fertility cycle theory. There is a well-defined transition from inverse association of economic status to fertility in urban areas to direct relation among the upper strata in rural areas. Besides income, the amount of land owned, another indicator of rural wealth, also reveals the same positive association of fertility with size of large land holdings.

Summary

In the presentation of the socio-economic determinants of Turkish fertility differentials, an attempt was made to combine the aggragate and individual levels of analyses. In a nationwide sample of Turkish couples, community structural implications for differentials at the individual level were investigated in addition to the impact of individual socio-economic characteristics. Although socio-economic variables were negatively associated with fertility at the individual level, the degree of this association and sometimes even the direction was found to be different between urban and rural communities. Thus, our findings so far clearly reveal that fertility varies with specific or different conditions of socio-economic structure. The birth rate in Turkey is affected essentially by the same causes acting towards its decline in developed countries. Contrary

37 THE POPULATION OF TURKEY to some recent studies in developing countries, almost all the tra- ditional (or historic) fertility differentials in industrially advanced countries have been observed in Turkey. In the three metropolitan centers (Ankara, Istanbul, and Izmir) that have gone farthest in fertility decline, there were extremely wide fertility differentials by social class. The differentials were less pronounced in other urban areas where fertility levels were also high. In rural areas where fertility was highest, there were no differentials among socio -eco- nomis groups. In fact, high levels of income and large landowner- ship were positively associated with fertility.

38 COMPONENTS OF GROWTH Section B — Mortality Miroslav Macura, Ph.D. Institute of Population. Studies Hacettepe University; This section presents information on mortality in various segments of the Turkish population. The information is derived from three data sources : population censuses, periodic Demographic sample surveys and the Turkish Demographic Survey. Mortality indices based on. the first two sources were derived,using some of the estimation techniques recently developed for extracting demographic measures from incomplete and. defective data (5). Indeces based on the TDS materials were calculated using convertional procedures as well as some of the techniques used with the other, two sources. The census and survey materials were used to reconstruct national trends in adult mortality for the past four decades and to ascertain levels of infant and early child mortality at a few dates in the past decade and a half. The same materials were used to learn about regional and urban-rural mortality differentials in the late 1960s. Trends in adult mortality by sex since the mid 1930s were re- constructed from the census age-sex distributions and the TDS data (4). Estimates of adult mortality, expressed in terms of expectation of life at age five, are shown in Table 1. Some of the estimates, it is almost certain, are biased in one Way or another, due to changing completeness of census enumeration from one census to another and possibly due to other data defects. Thus, it is lively that the pick in e°5 attained during the 1955-60 period is spurious and that'it was produced by a coverage of the 1960 census that was more complete than.that either of the 1955 or the 1965 census. In spite of '• the errors in the estimates, the time pattern of mortality change in the population aged five and over is clear. The mortality increased during the years of the World War II and underwent a rapid decline in a decade and a half folowing the War. The decline continued through the 1960s, at a pace much slower than before.

39 THE POPULATION OF TURKEY

Table 1 Expectation of life at Age 5 Estimated from Intercensal Survivorship Rates (1935-65) and the TDS Data (1966-67) (years)

e°5 ,fe°5 Period Males Females 1935 - 40 51.5 49.6 1940 - 45 46.5 46.2 1945 - 50/a 52.4 53.4 1950 - 55/a 57.6 59.2 1955-60 62.2 63.5 1960-65 59.2 61.9 1966-67 60.6 63.9 a The estimates for 1945 - 50 and 1950 - 55 were obtained by using ten year cumulated survivorship rates for 1945 - 55 and interpolating. This was necessary to avoid the effects of certain irregulativßs in the 1950 census. Source: The whole table, except the 1966-67 figures, is reproduced from (4). The 1966-67 figures are derived from both-sexes e°5 published in (2).

The estimates of e°5 also illustrate change in sex differences in adult mortality over time. They show that adult females experienced a mortality heavier than that of adult 'males in the early years of the period. Males were subject to comparatively heavier mortality later on. The change in the differential resulted from a comparatively faster decline in female mortality. The decline in mortality among adults was accompanied by a decline in infant and early child mortality. It is certain that some decline took place prior to the mid 1950s, but it is not possible to determine when did it start and how it progressed. For the years after the mid 1950s, the estimates of infant mortality for both sexes combined do exist. They were obtained from data of children ever born and children surviving collected in the surveys of 1963 and 1968, in the TDS and in the 1970 census and from the TDS current death data. The estimates, derived using the Brass mortality pro- cedure and the üfe table technique, are shown together with their reference dates in Table 2.

40 COMPONENTS OF GROWTH

The table shows a decline of some fourty points during a decade between the late 1950s and the late 1960s. To what extent the estimated levels and the implied tempo of decline are accurate is somewhat difficult to say. It is probably true that the infant death rate of the late 1960s was in the upper 140s or perhaps lower 150s. Indeed, this is what two independent estimates imply. The estimate for the late 1950s is not probaply equally strong. It is lively that, to an unknown extent, it understates infant mortality of the period, primarily due to a tendency to understate numbers of death children who were born to women passed age 30, or even 25 at the time of survey. If this were true ,the estimated tempo of decline would be undurstated, too. Different communities of Turkey participated in various deg- rees in the described mortality decline. This is revealed by regional and community-size differentials in mortality recently uncovered using the TDS and 1970 census data. Table 2 Estimates of Infant Mortality, Both Sexes Combined

Infant Deaths Reference Date Source of Data per Thousand of Births Estimate 187/a 1955-60 Retrospective data from the 1963 and 1968 Surveys and the TDS 152 1966-67 Current Data frqm the TDS 145 1968 Retrospective Data from the 1970 Census a) Th¡3 rate of 187 for 1955-60 was reached as an average of three infant death rates. Each of the three rates reflects mortality experienced by children ever born to women age 20-34 in one of the three surveys (1963-1968 and 1966-67 TOS). Source : Infant death rates f jr 1955 - 60 and for 1966 - 67, on the one hand, for 1968, on the other, come from (3) and (1), respectively. 41 THE POPULATION OF TURKEY

Reliable information on regional differentials is available for two years in the late 1960s : 1967 and 1968. Mortality indeces of 19*67, calculated from the TDS information and shown in the lower part of Table 3, refer to two large regions of the country. Western and Coastal provinces, on the one hand, and Central and Eastern provinces, on the other. The indeces show that general, adult and child mortalitp were ail higher in the former than in thlei latter region. As far as child mortality is concerned, Map 1 of 1968 infant death rates (derived from the 1970 census data) supports this finding to a large extent. It also makes it possible to drow somewhat more specific conclusion about regional differentials. Namely, that there is a considerable mortality' variation within each of the two large regions, especially within the second one. And also, that it is not necessarily true that Coastal or Western provinces enjoy lower mortality than that of Central and Eastern provinces. Information on mortality differentials by size of community presented here refers to 1967, comes from the TDS and is shown in the upper part of Table 3. When three types of mortality indices presented for populations of metropolitan centers, and all-urlban and rural areas are examined it is possible to generalize that mortality is lowest in the metropolitan areas, highest in rural areas and intermediate in all-urban places. It is evident that differences in mortality between all-urban and metropolitan are smaller (almost nonexistent among adults) than differences between rural and all- urban areas.

42 o

•I

ils HI- o THE POPULATION OF TURKEY

Table 3

Relatively Acceptable Indices of Mortality by Size and Location of Community, Centered on 1967

(ex° in years; radix •= 1000 for I5)

FEMALES MALES General Adult Child General Adult Child 0 o Population eu° e5 • 15 ^ e5 15

By Size of Community : AH Urban (2,000 + ) 60.7 64.8 868 57.0 62.1 849 Rural (under 2,000) 52.4 61.6 784 49.2 59.3 763 By Location of Community Metropolitan Centers 61.2 64.8 875 57.8 61.4 868 Western and Coastal Provinces3) 56.1 63.2 820 53.5 60.8 808 Central and Eastern Provinces3) 49.4 60.3 753 45.6 58.2 718 a) The population oí the metropolitan centers (Ankara. Istanbul and Izmir) are excluded. Note : The estimates by size of communnlty were prepared by using one year of data from all TDS subdivisions. Source : (3) • I •' . Î ' • I • •'

44 COMPONENTS OF GROWTH

REFERENCES

(1) Macura, Miroslav, Unpublished materials on Turkish mortality. (2) School of Public Health, Vital Statistics from the Turkish Demographic Survey, 1966-67, Ankara, 1970. (3) Shorter, Frederic C, Demography of Turkey (forthcoming). (4) Shorter, Frederic C, "Information on Fertility, Mortality and Population Growth in Turkey" Population Index, Vol. 34, No. 1, January-March, 1968. (5) United Nations, Methods of Estimating Basic Demographic Measures from Incomplete Data, New York, 1967.

45 COMPONENTS OF GROWTH Section C — International Migration

Samira Yener, M. A. State Planning Organization

Although the growth rate of the Turkish population has been influenced marginally in the past by changes in territory and the flow of international migration, until I960, the effect of these movements on the growth rate was negligible. However, after 1960, the net flow of migration switched from positive to negative, i.e. from an inflow of Balkan Turks in the previous period, to a net outflow of Turkish families to Western Europe. Therefore, this chapter will mostly concentrate on the emigration of Turkish workers. • • .... As a component of growth, the effect of emigrants is estimated not only by the actual numbers of emigrants, but also by their demographic impact on each census as a result of their own growth characteristics. A recent analysis of census data and intercensal emigration movements reveals the fact that the demographic impact of this outward movement results in the reduction of the birth rate from 42.01 to 40.8.1 I. Causes of Labour Emigration The main causes of labour movement from Turkey to other countries are : a) A labour shortage due to the rapid economic growth and low birth rates in a number of European nations. b) A high rate of unemployment prevailing in Turkey. c) The current earning differentials between internal and external labour markets.

1 Yener Samira : A paper presented to HO, Seminar on Population and Population problems In Turkey, 11 December 1972.

46 1 COMPONENTS OF GROWTH

The pull effect of wage differentials and demand for labour attracts the skilled and employed workers of urban and industrial centers,2 wehereas the unemployed and unskilled workers with limited job opportunities in the agricultural sector bring about an outward movement indicating the push effect of unemployment.

II. Trend Over Time

The statistics related with international migration are incomplete due to the fact that they do not cover all of the workers abroad. Informal entries to other countries are made even though an official channel, the Employment Bureau, attached to the Ministry of Labour, exists. According to statistics obtained from the aforementioned office, emigrations through official channels reached 569 thousand by 1971 (Table 1). When the foreign labour statistics are studied (e.g. Germán) it is observed that the annual increase¡in the number of workers exceed the number sent by official channels. Table 2. shows the number of workers working in Germany and the annual net increase in these figures. According to this table, the net increase between 1970 anid'71 is 96 thousand, whereas the number of workers sent officially to the European countries add up to a total of 88 thousand in the same year. This reveals an out inter-country movement of workers in Europe.as well as the existence of informal migration.

2 Saver, Ertan Ziver "A Socio-Economic Appraisal of Turkish Emigration (1965 - 72)", a paper presented to SOPEMI. 47 THE POPULATION OF TURKEY

Table 1 The Number of Workers Sent Abroad Through the Ministry of Labour Years Male Female Total

1960-63 39,834 3,155 42,989 1964 62,000 4,176 66,176 1965 40,341 11,179 51,520 1966 24,647 9,763 34,410 1967 5,414 3,533 8,947 1968 31,863 11,341 43,204 1969 83,210 20,765 103,975 1970 108,799 20,776 129,575 1971 74,242 14,200 88,442

TOTAL 470,350 98,888 569,238

Source : Employment Office, Th£ Ministry of Labour.

Table 2 The Net Increase of Turkish Workers in Germany (June 1960-71)! Years Total Number of Workers Annual Increase

1961 5,193b 1963 27,144 21,951 1964 69,211 42,067 1965 121,121 51,910 1966 157,978 36,857 1967 137,081 20,897 1968 139,336 2,255 1969 212,951 73,615 1970 327,985 115,034 1971 424,374 96,389

Source : Amthliche Nachrichten der Bundesanstalt Fuer . Arbeit. Arbeitsstatistik 1971 Jahreszahlen J 1157 B — 'August 1972. 48 COMPONENTS OF CROWTB

The emigration of labour from Turkey reached its peak in the 1970's, after which a decline was observed. This fluctuation is due to the consequence of external factors effecting the aggregate demand for labour, rather than the difficulties of meeting this demand in Turkey. When the distribution of workers, according to their country of destination is analyzed, it is clearly seen that more than 80 % went to Germany, and the percentage of those going to France increased, while the percentage for Holland and Austria seems to have decreased. As the statistics on return migrants are inadequate both in coverage and quality, the net increase of Turkish workers in other countries is estimated through the stock statistics of these countries. It has been found that during the 1965-70 period, Turkish workers abroad showed a net increase of around 297 thousand:1 in. Length of Stay According to a survey conducted by the German Research Institute2 the length of stay of the Turkish workers employed in 1962 is as follows : Length of stay Turkish Workers

(percentage)

Males Females

Less than 2 years 20 35 Between 2-4 years 37 42 Between 4-7 years 39 21 More than 7 years 4 2

100 100 As is seen from the table above, the greater percentage of the workers shows a tendency to stay for more than two years. 1 Yener Samira "An Estimate of a Net Increase of Turkish Workers Abroad" State Planning Office, P.D. 5. 2 Bundesanstalt Survey, Autumn 1968. 49 T.HE, POPULATION OF TfJRKEY

IV: Social and Economic Characteristics of Turkish Workers Com- position by Sex and Age The great proportion of Turkish emigrants in Germany are male (78 % in 1972). The age distribution is. as follows : . . '. Age - Male (•%) Female (%)

Under 25 7 32 25-35 60 41 35-45 29 25 Over 45 10 2

TOTAL 100 100 According to Bohning's study1 on the marital status1 and distribution of married couples and whether or not the wife was living with her husband in Germany, allows us to conclude that there is. an increased tendency of the married ¡workers to go abroad bring, their families with them. The proportion of married workers has risen from 78- % (in 1966) to 82 % (in 1968). The proportion of ' married workers living with their 'wives has gone up from 16 % to 28%. ,,......

V. Geographical Origin of Emigration

The provinces of the Marmara Region furnish the highest number of emigrants. Thds is the most developed area of Turkey where the rate of emigration is nearly twice that of the national average for the year 1971. Inf act the Marmara region is not actually the sending region, external migration being an extension of the rural-urban internal migration 'biases the results.

1 W. R. Bohning, Foreign Workers in Post War Germany, University of Kent; 1970 Table 2. p. ;6.- COMPONENTS OP QROWTR

Table 3

Emigration Rates According to the Regions

Emigration Rate Regions Emigrants (thousands)

Marmara Region 27,344 8.92 Central Anatolia 21,614 6.11 Agean Region 11,087 6.23 Antalya 2,407 4.80, Çukurova 3,867 4.53 Western Black Sea 4,014 2.29 ~ Eastern Black Sea 8,502 4.17 Eastern Anatolia 9,607 3.07

National Total 88,442 5.65

Bourco : State Planning Organization.

VI. Occupational Status

According to the skill definitions that are included in the official statistics of the Turkish Employment Office Bulletin, more than two thirds (64.6 % in 1972) of the emigrants are unskilled workers with very limited work experience. The breakdown of the skilled category in terms of occupations is shown in the following table. THE POPULATION OF TURKEY

Table 4 Occupations of Emigrants Occupations .. 1967 1972 Unskilled Workers 56,005 56,486 Miners 5,173 1,572 Textile Workers 1,562 1,527- Tailors 2,338 2,472 Leather Operators 41 — Blacksmiths (hot) 11 75 Moulders 223 102 Fitters, turners 2,845 1,778 Mechanics and repairmen 498 311 Plumbers 144 79 Assemblyman 57 104 Welders 1,803 903 Blacksmiths (cold) 246 157 Electricians 417 254 Carpenters 5,005 4,969 Bricklayers 4,017 5,172 Other 7,157 9,268 Total skilled workers 31,537 28,743 Total 88,442 85,229 As far as the sector of employment abroad is concerned, the change in the industrial1 structure of Turkish workers can be traced in Table 5. . . Table 5 ... Industrial Structure of Turkish Workers in Germany (toy the end of September 1971) Trans ., Agri- Construc- Com- r>or- Serv- culture Mining Industry tion merce tation ices 1964 0.7 14.5 54.5 21.8 1.5 3.3 3.7 1968 0.7 6.5 67.5 15.6 2.2 2.2 5.4 1971 0.9 7.3 65.5 15.9 2.6 2.3 5.4 Sourca : Bundesanstalt für Arbeit, J 1157. B.

52 COMPONENTS OF GROWTH

As compared to 1964, the proportion of workers occupied with mining and construction decreased, whereas those in industry increased to a considerable extend.

VII. Immigration.

As far as the immigration is concerned, the Turkish population can be considered a closed population, i.e. unaffected by migration or by losses or gains due to territorial changes during the last three decades, or until the Í960's, with the exception of the annexation of Hatay Province in 1939, and of the influx of a considerable number of Turks from Bulgaria in the 1950's. During the 1965-70 period, the influx from Bulgaria was approximately 30,000.

53 THE POPULATION OF TURKEY

REFERENCES

Bacci M. Iivi, M. Hagmann, "Report an the Demographic and Social Pattern of Migrants in Europe, Especially With Regard to Inernational Migrations," paper presented to Council of Europe, 2nd European Population Cenference - Strassbourg 1971. Hume, Ian M'., "Migrant Workers in Western Europe," Working Paper prepared for International Bank for Reconstruction and Development, October 1970. Miller, Duncan R., Emigrant Turkish Workers : A Socio-Eco- nomic Analysis in Essays on Labor Force and Employment in Turkey, Miller Duncan, ed. AID publication. Saver, Eftan Ziver, "A Socio - Economic Appraisal of Turkish Emigration (1965-1972)," paper presented to SOPEMI Meeting Paris, 4-5 July 1973. Yener, Samira, "An Estimate of Net Increase of Turkish Workers Abroad," S.P.O. Mimeogr. 10 pages, September 1972. Chapter III

THE STRUCTURE

THE TURKISH POPULATION

by Haluk CfiLLOV, Ph.D.

THE STRUCTURE OF THE TURKISH POPULATION

Haluk Cillov, Ph.D. Faculty of Economics Istanbul University

I. Sex Distribution of the Population 1. Importance of the male-female ratio

In the first part of the monograph the developments in Turkey's population Were discussed. In this part we will discuss the structure of the population by it's various characteristics. The sex distribution of a population is one of it's important characteristics. In a population, the male-female ratio may be an important element affecting the strength and development of that population. In addition, the functions of. a population varies according to it's sex distribution. For this reason, the knowledge of the sex distribution of a country's population enables one to evaluate the social and economic structure, and even political strentgh of that population. When the populations of the countries are examined, one observes a natural balance in male-female ratio. However, in many countries the female population is slightly more than the male population, or vice versa. In 1970 the number of males per 1000 females in Great Britain was 947, in West Gremany 914, in Greece 951, in Italy 958, in Yugoslavia 962, in the United States 948, in Soviet Union 854. On the other hand, there were 1064 males per 1000 females in India, 1110 in Pakistan, 1013 in Peru and 1012 in Egypt (UAR).1 Generally a female surplus is observed in many countries. Although more male births are recorded than female births, the death rates are higher in males, and this tips the scale in favour of the female .population. 57 THE POPULATION OF TURKEY

In observing the populations, certain factors are noted which disturb the balance between the numbers of males and females :

a. Age structure is the most important factor disturbing the balance. In general, more births occur in places Where the population is young, and this raises the number of males. The imbalance in favour of males in Turkey may partly be attributed to this factor.

b. International migration is an artificial force affecting the male-female ratio. The males take the lead in migration, and this results in la malte surplus in countries! whiere in migration take place, and a female surplus in countries where there is outmigration. The 1023 to 1000 male-female ratio in Australia and the slight increase in the female proportion in Turkey in 1965-70 period are related to international migration.

c. The wars are another factor which disturb the sex balance. In war years there is generally a decrease in the male population. In France in 1910 there were 1035 females per 1000 males. An increase in that number to 1103 in 192Ö is related to the First World War. . ; • i \ i n ,-•'-! ;••; r !

d. Finally, difficult living conditions for females in some countries disturb the balance. In People's Republic of China the fewer number of females than males may partly be attributed to this factor. . »

2. Sex Distribution of the Turkish Population

In Turkey the first general census was taken in 1927. Beginning in 1935 a census was taken every five years.

1. According to the censuses taken between 1927-1970, the number of males per 1000 females are shown in Table 1.

58 THE STRUCTURE OF THE TURKISH POPULATION

Table 1

Number of Males per 1000 Females : Years Years

1927 927 males 1955 1031 males 1935 965 " 1960 1042 1940 966 " 1965 1032 1945 1012 " 1970 1026 1950 1018 "

As we see, there were more women than men in Turkey until 1945 and the reverse is true after that year. There are several reasons for this : a. The 1927 census was taken after two major Wars, the First World War and Independence War. The findings of this survey reflects the losses in male population in these wars. b. As Will be seen later, the number of births in Turkey its high. Furthermore, the high number of male births is more in Turkey than most other countries. Although general birth statistics are not prepared in our country, it is estimated that there are 108 male births per 100 female births.2 The surplus in male births contributed to the near balance established by 1945. c. During the 1927-45 period Turkey received immigrants, especially from the Balkan countries. Since the male population is more numerous among immigrants, it can be understood that the change in male-female ratio in males' favour has been affected by these immigrations. i 2. Several factors affected the change in male-female ratio ID men's favour until 1960. But, beginning in 1963, a large number of Turks went abroad as labourers. Since most of these labourers were men, a reverse trend began taking place-in the population structure. Although the male population is still1 larger, in 1965 and 1970 censuses we see a decrease in the proportion of the male population. 59 THE POPULATION OF TURKEY

3. The distribution of Turkish population shows its characteristics especially when different age groups are taken into consideration. While there is a 1 to 3 percent male surplus in the 0-15 age group in most other countries, this figure is as high as ten percent in Turkey. This phenomeno is related to births, and results in a more than average male surplus in young ages. The balance in the male-female ratio which normally takes place in the 15-24 age group shifts to higher age groups (35-45 age groups) in Turkey in rounded numbers in the censuses. In fact, in 1935 census the male surplus continues to be present up to 25 years of age while the balance is established in the 35-44 age group in 1950-70 period even though the circumstances were similar. In Table 2, the numbers of men per 1000 women in different age groups are shown (1965 and 1970 censuses)3. Table 2 Number of Menper 1000 women Age Groups 1965 1970 Age Groups 1965 1970

0 — 4 1041 1027 35 — 39 1067 1028 5 — 9 1073 1054 40 — 44 1040 1014 10 — 14 1137 1123 45 — 49 1037 1095 15 — 19 1133 1048 50 — 54 1036 914 20 — 24 1068 1124 55 — 59 1083 1144 25 — 29 908 856 60 — 64 902 907 30 — 34 1003 873 65 + 744 864 3. Geographic Distribution of Population by Sex The male surplus observed in the Turkish population shows variations according to the distribution of the population in the country. The variance is especially pronounced between the villages and towns. In 1965 there were 1032 men per 1000 women in Turkey-while this ratio was 3157 to 1000 in towns (settlements with populations over 2000) and 983 to 1000 in villages. In other words, there is a concentration of males in towns and females in villages. 60 THE STRUCTURE OF THE TURKISH POPULATION

This situation is not exactly like the other countries. Although it is generally true that we see a male surplus in towns, the female surplus in villages is greater in Turkey than most other places.. In fact in some countries the male surplus in villages is greater than it is in towns. In Great Britain there are 936 males per 1000 females in towns while this ratio is 989 to 1000 in villages4. In the United States these ratios are 940 ton 1000 in towns and 1042 to 1000 in villages according to 1960 census. In other words the situation is quite the reverse in the United States where there is a male surplus in villages. The male surplus we observe in Turkish towns becomes more apparent in the metropolitan cities. According to 1970 census there were 1153 males per 1000 females in Istanbul and 1173 males per 1000 females in Ankara. We relate this situation in Turkey to rural to unban migration which is particularly greater towards big cities5. We should also point out that the nature of this migration in Turkey is different from Western countries. Whereas in Western countries peopis migrate to unban centres to find better living conditions, in Turkey it is generally done for a temporary period to supplement the farm income. The concentration of big male populations in Istanbul and An- kara is explained by people's belief that the possibility of earning a/better income in Istanbul is greater, and Ankara is a bureaucrats' city. Also there is a concentration, of military bases around big. cities which may contribute to the malö surplus in such places. In conclusion, we may say that although there is a male surplus in Turkey, it is not a dangerous one. In fact it may be accepted' that this situation increases the military and economic strength of the country. Furthermore, the balance we see between the numbers of men and women in marriage age (25-44) eliminates the danger of not being able to find marriage partners. II. Age Distribution of the Population 1. Importance of the Age Distribution ' . The age distribution of the. population is important in. deter- 61 THE POPULATION OF TURKEY mining the needs and attitudes of human beings as well as their social functions. In this respect studies on age distribution is an important part of population studies. Hence, to find out the level of fertility, or age specific death rates or causes of death we need to know the distribution of the population by ages. Thus, to study the events associated with the age composition we have to know the age distribution of that population. The differences in the definition of age by various countries result in difficulties When determining the age composition. Generally, we accept number of days, weeks, months and years since birth of a person as his age. However, it is not easy to determine age with this definition, it is especially difficult in countries where the literacy rate is low. Age misreporting is generally due to ignorance or indifference as well as preference of the digits ending in (0) and (5). We see the effect of these factors in Turkey, where the literacy level is relatively low. 1970 Census6 has shown that in Turkey 38,367 people (1.0.7 per thousand) did not know their ages. The preference for the ages ending in 5 or 0 is still observed in the 1970 Census,7 although at a lower level, Hence, although the number of males aged 30 is 374,346 those aged 29 and 31 are 150,260 and 130,130 respectively. It is observed that females round their ages too, and at older ages this tendency is more pronounced/ Hence, in 1970 Census, the females aged 40 werfe 503,654 whereas those aged 39 were 101,515 and 41 were 77,309. This strong tendency is observed in males at very old ages. For example, in the same census the number of men aged 80 were 29,135 as opposed to 5,621 and 4,383 at ages 79 and 81.

2. Characteristics of the Age Distribution in Turkey A. The Turkish population is young. Those aged 15 or younger in 1970 were 14,888,793 or 41.8 % of the total. If we define the youth as those under 20, then this proportion is as high as 52.1 %. If we compare with France and England where those aged 15 or 62 THE STRUCTURE OF THE TURKISH POPULATION

younger are 32.2 % and 29.8% respectively, it is easily seen how large the population at younger ages is in Turkey. The proportion under 15 has been steadily increasing in Turkey, due to the increase in births, as seen in the following table. Some other countries are also shown for comparison7. Proportion aged. 15 or Under Country Year %

Turkey 1955 39.4 1965 41.9 » 1970 41.8 France 1968 23.7 Greece 1969 25.1 Yugoslavia 1968 28.3 U.S.A. 1970 28.5 The table below shows the distribution of the young population by age and sex : Age % of Males % of Females ,% of Total

0 — 4 14.3 14.3 14.3 5 — 9 14.8 14.4 14.6 ,10 — 14 13.4 12.4 12.9 15 — 19 • 10.5 10.3 10.4 It is seen above that, .

a¡ the first two age groups are especially large : ; b. in addition to the large number of births, the facts that males understate their ages to postpone military duty and that the females understate their ages due to their fear not being able to get married, causes a relative increase in the younger ages in both sexes* . - . • B. In contrast to the number of younger people, we observe a relatively small number of those in the middle ages, especially the active-population, 15-64; in 1970 those in this age group numbered 63 THE POPULATION OF TURKEY

19,196,141. The following table gives their percentages in 1955, 1965 and 1970; other countries are given for comparison : Proportion Countries Years Aged 15-64% Turkey 1955 57.1 1965 54.1 » 1970 53.9 France 1968 62.8 Greece 1969 65.0 Yugoslavia 1968 64.3 U.S.A. 1970 61.6 The above figures show that the proportion of those aged 15-64 are relatively small and have a decreasing trend. However, these figures have been influenced by the preference for younger ages, and thus some of those in the 15-64 age group are seen in the 0-15 age group9. The following table shows that the population aged 15-34 is larger than that aged 35-64 in the 1955, 1965 and 1970 Censuses. Proportion Proposition Years Aged 15-34 Aged 35-64% 1955 33.7 % 23.4 1965 31.1 % 23.0 1970 30.9 •% 23£ The above findings also support the fact that the Turkish population is a young one. In fact when those aged 15-34 are compared, it is 30.9 % in Turkey, whereas in France it is 28.1 •%, and in Greece and USA 29.7 %. Those aged 35-64 are 23.0 % in Turkey but 34.7 % in France, 35 % in Greece and 31.9 % in USA. However, this small percentage in the ages 35-64 is due to other factors. Those aged 35-64 in the 1960 and 1970 Censuses were born between 1910 and 1925 and these years correspond to the First World War and the Independence War, and it is known that the number of births in these years were significantly small. Hence the small proportion aged 35-64 is temporary and will change when the effect of these two wars reduce.

64 THE STRUCTURE OF THE TURKISH POPULATION

C. Finally, those 64 years old and above were 1,543,248 in the 1970 Census. This is 4.3 % of the total population and is relatively small when compared with other countries. Proportion Aged Country Year 64 or more

Turkey 1955 3.4 % 1965 4.0 % 1970 4.3 % France 1968 13.4 % Greece 1969 9.9 % Yugoslavia 1968 7.5 % U.S.A. 1970 9.9 %• The proportion above 64 is small because the mean length of life in Turkey is relatively smaller than that in the Western Countries. However, this proportion has been rising1 since 1935, which shows that the mean length of life has also been increasing. D. The median age for males was 19 and for females 20, according to the 1960 Census figures. The median for the total population was also 20. It is observed in 1970 that the median has decreased to 18 in each sex, hence implying again a trend towards a younger population. ; The median age in 1960 was 28 in USA, 31 in Greece, 33 in France and 35 in England and that it was 20 in Turkey shows1 how young the Turkish population is. Briefly, the above findings show that the proportion of the people at younger ages is large and that at older ages is small. E. Finally, we can observe the main feature of the age distribution from the tables shown above. The large proportion in the age group 5 - 9 is due to the high birth rates observed in 1955-1965, especially in 1955-1960. The shortages in some of age groups between 20 and 59 may create some socio-economic problems. Especially, the shortage in the most productive ages, 15-34, reflects lower production capacity of a country. 65 THE POPULATION OF TURKEY

We must note, however, that these irregularities in the age distribution are not important enough to have effects on the national economy in Turkey. We may even add that it has some good effects, as will be shown below. i) The large proportion under 15 shows that the population of the country is growing and it will increase the potential labour force in the future. ii) The small proportion at older ages has an economic advantage, in the sense that their burden on the working population is relatively low. We note however that the dependency ratio 10 is large and has a tendency to get larger. This ratio was 75 % in 1955 n and has reached to 84.8 in 1965 and 85.6 in 1970. We must also keep in mind that this tendency is due to the increase in the proportion at younger ages, and would have further increased, had there been an increase in the proportion at older ages.

III. Distribution of the Disabled Population

Aibove we discussed the physical characteristics of the population. Although not quite as important as the characteristics we previously discussed, it is necessary to take a look at the disability situation among the physical make up of the population of the country. We should point out that the data concerning the physical structure is not always reliable. For this reason questions about disability are not included in every census. For instance, in 1970 census such questions were not asked. On the other hand, in 1965 census people were asked whether they had any visible physical disability as whether they had it all their Üves. In Turkey in 1955 there were 316,943 people with visible physical disability. This number went up to 380,985 in 1965. However, the proportion of physically disable people has decreased from 1.31 percent in 1955 to 1.21 percent in 1965. In 1965, 62.« percent of the disabled were males and 37.4 percent were females. 66 THE STRUCTURE OF THE TURKISH POPULATION

Table 3 Disability Situation of the Population in 1955 and 1965 1955 1965 %

Arm disorders 35,543 10.3 27,121 7.1 Hunchbacks 7,204 2.3 5,743 1.5 Blinds 101,182 31.9 112,956 29.6 Leg paralyses 6.293 2.0 19,932 5.2 Deaf 34,037 10.7 52,859 13.9 Lame 107,882 34.0 142,890 37.6 Various disability 9,098 2.9 5,998 1.6 Other 3,063 1.0 Unknown 15,641 4.9 13,486 3.5

Total number of disabled 316,943 100.0 380,986 100.0 23,747,820 — 31,010,436 — Total population 24,064,763 — 31,391,421 — In Table 3 we observe that : a. The majority of the disabled in Turkey are the lame ones. b. The blinds take the second place. 74,778 of the blinds were blind in the one eye and the remaining 38,Í78 were blind in both eyes. c. The third group of the disabled are the deaf and mutes. 25,028 of these people were deaf and 27,830 were both deaf and mute. I ;'•. ,• . d. In published statistics we are able to find the sex and age distribution of the disabled and whether they were disabled all their lives13. However we are unable to establish any trends here. The reasons for disability varies according to the nature of disability, and there are no definite relations between age and disability. However we observe two important points when we look at the distribution of the disabled by age : i. There is a considerable decrease in the number of the disabled under five years of age in 1965. This is a healthy indication for the physical structure of the future generations. 67 THE POPULATION OF TURKEY

ii. We also observe that the disabled and especially the lame people concentrate in the 40-44 and 50-64 age groups. The reasons for this may be found in the effects of the First World War, the Independence War and the Balkan Wars. IV. Distribution of the Population by Marital Status To study the age and sex structures of a population is not sufficient when one is investigating the composition of that population. The social and economic characteristics of the population need to be studied as well. The marital status of the people is the first characteristic to be looked at, for the social and economic trends of a population is strongly related to this characteristic. We find out the marital status of the people through questions included in the censuses. Beginning with the first general census of 1927 in all censuses questions of this nature are included and the findings are published. In Table 4 the marital status and the sex distribution of the people according to 1970 figures are shown14. Table 4 Marital Status and Sex Distribution of the Turkish Population in 1970 (in thousands) % M 4312 36.6 Single F 3015 25.9 T 7327 31.3 M 7122 60.5 Married F 7384 63.6 T 14506 62.0 M 230 2.0 Widowed F 1090 9.4 T 1320 5.7 M 54 0.5 Divorced F 79 0.6 ;. T 133 0.5 M 47 0.4 Unknown F 46 0.4 T 93 0.4

68 THE STRUCTURE OF TUE TURKISH POPULATION

1. In this table it is observed that : a. A great majority of the population is married (62 %). b. The single population constitute one third of the total population. c. The number of the divorced and especially the widowed are small in the Turkish population. The presence of a greater number of widows is largely due to their smaller chance of getting married again. The same tiling is true for the divorced. 2. It is also necessary to study the developments in the marital status of thé population. There is a difficulty here. Until the 1970 census, the marital status of the people were recorded after 15 years of age, but in 1970, this was done for people over 12. We believe this was a mistake, for it limits the possibility of comparison between different censuses. To avoid this difficulty, we compared the 1955 and 1965 figures.15 Table 5 The Distribution of Population Over 15 years of Age by Marital Status and Sex (in percentage) 1955 1965

M 26.8 26.5 Single F 13.6 14.5 T 20.2 20.5 M 69.8 70.4 . . Married F 71.0 72.7 T 70.4 71.5 M 2.2 2.3 Widowed F 13.4 11.7 T 7.8 7.0

• M 0.6 0.7 Divorced F 1.0 0.9 T 0.8 0.9 M 0.6 0.1 Unknown F 1.0 0.2 T 0.8 0.1

69 THE POPULATION OF TURRET

In the above table we observe that : i) There has been a slight increase in the proportion married. ii) This increase is especially due to the reduction in the number of widows (i.e. widows getting married again). iii) The proportion of single and divorced has been relatively stable. V. Family Structure When studying the characteristics of a population, we need to know the family structure in that population. Data on family structure: and life-styles of the families is being collected since 1955. It is hard to state that the data was reliable in 1955. However, in addition to the censuses of 1960, 1965 and 1970, some national surveys I6 have collected data on life-styles and have shown the characteristics of the family structure in Turkey. In the above mentioned censuses in Turkey, the household has been defined as : One person or a group of persons with or without a family relationship, who live in the same house or in the same part of the house, who share their meals, earnings and expenditures and who take part in the management of the household and who render services to the household. The household head is the person who is responsible for the earnings and expenditures of the household and actually manages the house. According to the 1970 classifications established by sampling methods, we observe that the number of families is 5,492,538. The distribution of families by residence could be calculated only in 1955 and 1960. Thus, we see that in 1955, 32 % of families were living in rural districts and the rest in urban areas. The rate of families living in cities increased to 36 % in 1960. Distribution of the Families by Size The size of the families differs according to each household. In fact, it is not possible to derive the composition of families from the census. Although some valuable data were obtained on this subject in research conducted on 4,800 families in 1968 17, we find it useful to review the increase of the household sizes between 1955- 70 as seen in the censuses, thus providing a comparison.

70 THE STRUCTURE OF THE TURKISH POPULATION

The sizes of the households in 1955-70 are shown in table 6.

Table 6

The Evaluation of the Households by Size

1955 1965 1970

1 person families 167,116 140,950 167,524 2 " " 407,534 473,391 497,693 3 " 544,359 591,069 549,935 4 " " 674,594 820,756 754,263 5 678,494 890,502 824,765 6 " " 571,809 794,359 729,088 7 + " 1,168,994 1,824,970 1,845,508 Unknown 24,276 119 123,762

Total 4,237,176 5,536,116 5,492,538

It can be seen that in Turkey the households are crowded. Tak- ing the year 1970 as a basis, we can calculate that 33.6 % of the families consist of 7 or more persons, 15 % of 5 persons, and 14 % of 4 persons.

Characteristics of the Households

After determining the families forming the society, it is possible to see the characteristics of the households. Examination of these characteristics is of great importance as they show the economic and social lives of the families. At present, 34 % of the households are living in two-room houses, 22 % in three-room houses, and 17 % in one-room houses.15 If we compare this situation to 1955, we see that the sizes of the houses occupied by the families have not changed, much. The table below confirms this point.

71 THE POPULATION OF TURKEY

Table 7

Households by Number of Rooms Occupied

1955 1970

1 room, houses 1,192,150 908,950 2 room houses 1,639,488 1,887,640 3 room houses 732,757 1,221,521 4 room houses 382,634 682,960 5 + room houses 214,688 777,068 Unknown 75,459 14,399

Total 4,237,176 5,492,538

It is important from the points of view of economic and social characteristics of the familes whether tihte¡y live in their own or rented house and how much rent they pay. ; In fact, in 1970, 4,473,076 out of 5,492,538 families (81 %) were not paying rents. In other words, it is very probable that they lived in their own houses. When it is considered that 86 % of the households were living in their houses in 1955, it can be concluded that a greater number of the families have moved to rented houses within the last fifteen years. Of the 4,473,000 rent paying families in 1970, 13.2 % paid up to 199 TL. per month, and therefore we observe that the ones living in the low rent houses were the majority. We see that many of the families were paying 200-299 TL. per month and the number of families paying higher rents, that is to say, those living in big and luxurious houses were few. Monthly rents paid by households in the years 1955 and 1970 are given below.

72 THE STRUCTURE OF THE TURKISH POPULATION

Table 8

Households by Monthly Rent Paid

1955 1970 Amount of Rent Paid No. of Families No. of Families

LESS than 200 TL. 507.158 591,763 200 — 299 TL. 12,933 163,822 300— 399 TL. 4,629 105,151 400 — 499 TL. 2,010 57,830 500 — 599 TL. 1,019 33,733 600 — 699 TL. 522 17,900 700 + TL. 888 46,169 Unknown 38,091 3,083

The table above shows that the move towards houses with high rentals increased between 1955-70. This is true, but should, be considered in relation, to increased rents. We have seen above that families living in rented houses pay low rents. Houses with few rooms are widespread. This situation gives the idea that generally speaking the families are poor and residences are not of high quality because they are cheap. In order to investigate this matter, it is necessary to know the physical conditions of the houses. In the 1965 and 70 censuses, data related to the facilities that houses have, were obtained. Taking these data as a basis, one can discover the technical specifications of the houses and the social lives of the families. a. First, it ¡must be indicated that kitchen facilities of houses used by families are not very satisfactory. As can be seen in the table below, in 1970, only 49 % of the 5,492538 families had separate kitchens and 42 % had none. 73 THE POPULATION OF TURKEY

Table 9 Kitchen Distribution in Houses Number Families with a separate kitchen 2,698,721 49.1 Families with a common kitchen 477,303 8.7 Families with no kitchen 2,312,127 42.1 Unknown 4,387 0.1 TOTAL 5,492,538 100.0 b. The situation with regard to toilet facilities is even worse. In 1970, only 60 % of the 5,403,000 households had the exclusive use of toilets; 27 % had no toilet at all. Table 10 Toilet Distribution in Houses Number Families with separate toilet 3,324,279 60.5 Families with a common toilet 680,484 12.4 Families with no toilet 1,483,977 27.0 Unknown 3,798 0.1 TOTAL 5,492,538 100.0 c. The position regarding the distribution of the bath facilities is not any better. One learns that 56 % of the 5,493,000 families have no baths, only 39 % have exclusive use of one and 5 % are using common ones. ' Table 11 Bath Distribution in Houses Number % Families with a separate bathroom 2,125,150 38.7 Families with a common bathroom 278,364 5.0 Families with no bathroom 3,085,562 56.2 Unknown 3,462 0;l TOTAL 5,492,538 100.0

74 THE STRUCTURE OF THE TURKISH POPULATION

d. The statistics show that electricity, water and heating facilities are also in bad condition. i. First, one sees that only 39 % of houses used by families have electricity. ii. 34 % of these houses have drinking water and the rest do without it. iii.' Finally one learns that 81 % of families are using stoves, 14 % are using fireplaces and only 3 % have central heating. These findings are interesting as they help to illustrate the social lives of families and the unsatisfactory state of housing in Turkey. It is possible to observe the families' major occupations as well as distribution of families working in various professions according to the rentals they pay. However, because of some amendments made to occupational categories in 1970, it is not possible to compare the development of the heads of household according to their occupations between the years 1955-1970. Therefore the occupational categories of the family heads in 1970 will be examined. Table 12 Occupational Categories for the Household Heads Number Scientist and technicians, self employed 209,952 3.8 Private entrepreneurs and directors 45,470 0.9 Administrative personnel 161,212 2.9 Commercial and sales personnel 287,333 5.2 Personal services 248,427 4.5 Agricultural employees 2,827,830 51.5 Other than those agriculturally employed 953,700 17.4 Undetermined professions 758,614 13.8 TOTAL 5,492,538 100.0 a. It is seen that in Turkey 52 % of household heads' occupations are related to agriculture. 17 % of them are working as labourers or artisans in fields other than agriculture, and 4.5 % are in personal services. This shows the importance of agriculture in the economy of Turkey.

75 THE POPULATION OF TURKEY

b. It must be added that 96 % the household heads working in agriculture live in their own houses. Most of the rented houses are occupied by scientists and technicians together with the self-employed. c. Finally, another interesting point arises which illustrâtes the economic structure of families : i. Among the ones living in houses of low rental1 (1-199) liras), labouers come first while entrepreneurs have the lowest proportion. ii. Houses with high rental are occupied mostly by technicians together with entrepreneurs and directors. On the other hand the number of family heads working in agriculture living in houses with high rental is veryfew. VI. Distribution of Population by Nationality \ In the censuses, the question of nationality was asked. The establishing of this question has importance particularly in countries with many foreigners, or those where the number of foreigners / may change by large amounts19. Nationality is described as "the allegiance to which persons are legally attached". The great majority of Turkish society are "Turkish citizens", Vhîe division of Turkey's population by nationality according to the tour censuses is shown below20. Table 13 Distribution of Turkey's Population by Nationality Turkish Foreigner Years Male Female Male Female 1935 7,999 8,195 28 26 1945 9,427 9,326 20 18 1955 12,208 11,812 24 18 1960 14,129 13,568 34 21 As seen in the table, the percentage of foreigners is very low in Turkey. In addition to that, it is observed that this percentage is descreasing year by year. In fact, in 1935, the rate of the foreigners (including both sexes) was 0.3 % of the total population, and this rate was reduced to 0.2 % in 1960.

76 THE STRUCTURE OF THE TURKISH POPULATION

As can be derived from the table above : It is possible to discern the nationalities of foreigners in the statistics. In 1960 U. S. citizens predominated with a total number of 16,102, Yugoslavs with 3,996, Italians with 3,598, West Germans with 2,840 and English with 2,616. VII. Distribution of Population by Religion In the censuses in Turkey, the religious affiliations of the people were looked at. The major religion of Turkish society is Islam and the relative distribution of people of other religions is very low. However even the stability seen in the distribution of population by religion is interesting. . This is shown in Table 14 below, according to the 1955 and 1965 censuses. Table 14 Distribution of Population by Religion 1955 1965 Religion No. % No. % Moslem 23,804,048 98.9 31,129,854 99.1 Christian 208,867 0.8 206,825 0.7 Jewish 45,995 0.2 38,267 0.1 Other religions 2,746 13,661 Nonreligious 613 0.1 1,212 0.1 Uknown 2,494 602 Total 24,064,763 100.00 31,391,421 100.00 1. As can be seen in the table, 99 % of the Turkish society is Moslem. This rate has retained its stability over the last ten years. Christians are relatively diminishing in number. The same situation is observed for the Jewish population; On the other hand, the number of people, of other religions and ;of the nonreligious has increased slightly. 2. In addition, it is possible to observe the distribution of the different Christian sects. This is shown in the table below according to the 1955 and 1965 censuses. 77 THE POPULATION OF TURKEY

Christians 1955 1965 Catholic 21,784 25,833 Orthodox 86,655 73,725 Protestant 8,952 22,983 Gregorian 60,071 69,526 Others 31,405 14,758 Total 208,867 206,825 It is observed that i) The Orthodox represent the majority in Christian communities but their number is relatively decreasing. ii) Gregorians come second. Mi) The number of Catholics is increasing, however, there are only 25,000 of them. iv) The highest increase is seen among the Protestants. VIH. The Cultural Level of the Population 1. Literacy. The social and economic power of a country depends on its literacy. The economic development potential of a country with a high rate of literacy is high. Therefore, it is necessary to know the literacy and even the cultural level of the population. In Turkish censuses literacy is a specific category. In some years (1935 and 1945) even the number of people who could read but not write, or could write but not read, was recorded separately. In addition, literates and illiterates are classified by sex and age. Literacy in Turkey according to the 1927-1970 censuses is shown below22. : i Table 15 Literacy in Turkey (thousands) Illiterates Literates Unknown 0-5 Age Year Male Female Male Female Male Female Group 1927 4,036 5,354 852 260 7 7 3,133 1950 4,516 6,819 4,055 1,725 36 44 3,753 1960 5,324 8,301 6,158 2,743 9 7 5,213 1965 4,689 8,450 8,372 4,133 13 8 5,727 1970 4,620 8,726 10,302 5,825 11 11 6,172

78 THE STRUCTURE OF THE TURKISH POPULATION

a) The literate population is increasing year by year. In 1927 89.4 % of the population over 6 years of age was illiterate. This rate was reduced to 51.2 % in 1965 and to 45.3 % in 1970. b) There are more males who are literate than females. At present 69 % of the male population is literate. This rate was only 17.4 % in 1927, which shows the significance of the development. c) The percentage of female literacy is low. But the relative progress of the rate of literacy among females is higher than that of the males. In 1927 only 4.7 % of the females were literate whereas the rate increased to 40 % in 1970. d) Great progress is recorded in literacy after 1960. This is considered a positive development. Thé relative progress of literacy in Turkey is shown below.

Table 16 Literacy Percentage ( % ) The Rate of Total Population of 6 Years Year Male Female of Age and Over 1927 17.4 4.7 10.6 1955 55.8 25.5 40.9 1960 53.6 24.8 39.5 . . 1965 64.0 32.8 48.8 1970 69.0 40.0 54.7 2. The decreasing rate of illeteracy in Turkey is quite promising when compared with that of other underdeveloped countries. Ac- cording to estimates, this rate was 60-65 % for the population over 15 years of age in 1968. It was 85-95 % for Saudi Arabia and Afghanistan, 75-85 % for Pakistan, and 90-85 % for Ethiopia23. The rate of illiteracy for the same age group is 15-20 % in Greece, 10-20 % in Spain; 5 % in Hungary and Poland, 3 % in France; 2 % in the Soviet Union, 1 % in West Germany, Sweden and Austria. These numbers show that Turkey has the lowest rate of literacy in Europe. ; 3. Statistics showing the distribution of literate population by province are also available. 79 THE POPULATION OF TURKEY

a. If the conditions of literacy according to age are examined, some interesting results can be achieved. In 1970 the great majority of literates was among the young people. For example 71 % of the population aged 10 were literate in comparison to 32 % for the age group 50-54, and 16 % for the age group over. 65. b. It must be indicated that the majority of literates live in cities. : Results of the 1965 census show that 66.9 % of the population living in cities are literate. This percentage is 38.53 % for the rural areas. We must add that both in urban and rural aras, the male, population predominates among the literates. The distribution.of the literate population over 6 years of age living in urban and rural areas is shown below :

Table 17 Percentages of Literates (1965) Male Female Total Urban areas 79.4 52.3 66.9 Rural areas 54.6 22.9 38.5

c. In comparison to 1935, a vast progress in the education level of the urban and particularly rural population has been made. As a matter of fact, the urban literacy rate was 47 %, and the rural 14 % in. 1935; these have increased to 67 % and 39 % respectively in 1965. 2. The Mother Tongue of the Population. In the censuses, the language of the population also forms a specific category. a. There are several methodological points of view as to how to describe the "mother tongue"24. In Turkey, it is assumed that "the language spoken within the house among the family" is the mother tongue. •'.••:

80 THE STRUCTURE OF THE TURKISH POPULATION

In fact, although the description is quite clear, soma mistakes are made when establishing what the mother tongue is by considering it as the "mother's tongue". If the coincidental errors are ignored, it can be said that in Turkish society, Turkish is the language predominantly spoken. In the 1955 and 1960 censuses it is seen that Turkish is the mother tongue of 90 % of the population. Also in the 1965 census, 28,289,680 people (90,1 % of the total population) declared their mother tongue to be Turkish. b. The remaining 10 % of the population speak other languages. The most popular foreign languages are "Islamic minority" languages" and Kurdish is the predominant one among these. In 1965 it was ascertained that 2.2 million persons (7 % of the total population) spoke Kurdish, followed by Arabic which was wpoken by 365 thousand persons (1 %). Among the other minority languages, 48 thousand people speak Greek and 33 thousand Armenian. In addition, 28 thousand speak English, 5 thousand speak Ger- man while 7 thousand declared Serbian to be their mother tongue. c. It is also possible to observe the distribution of the population according to sex and province. 3. The Cultural Level of the Population. It is necessary to know a country's rate of literacy as well as its cultural level. Questions related to this matter were asked in the census. a. Among the 10,478,234 literate people in the 1965 population census; 28.3 % had no education 58.2 % had no primary school education 6.6 % had no middle school education 2.6 % had a lycee education 4.3 % had some higher education (vocational schools). The 1970 census made it possible to learn the cultural level of the population which is classified according to sexes below25.

81 THE POPULATION OF TURKEY

Table 18 The Cultural Level of Literates, 1970 (Thousands)

Male Female Total Number % Number• % Number %

No education 1,841 21.0 952 20.6 2,793 20.9 Primary School 5,429 62.1 2,978 64.6 8,407 62.9 Middle School 734 8.4 367 8.0 1,101 8.2 Lycee 282 3.2 123 2.7 405 3.0 Vocational School 257 3.0 137 3.0 394 3.0 Higher Education 195 2.2 / 42 0.9 237 1.8 Unknown 10 0.1 11 0.2 21 0.2 Total 8,748 100.0 4,610 100.0 13,358 100.0

Examining the above data : i. 62.9 % of total literates had a primary school education. This percentage is almost parallel for both males and females. ii. Approximately 8 % of the literates had a middle school education and 3 % had lycee education. Here the male percentage is higher than the female as the latter do not have the ambition to continue. iii. The percentage of literates having had higher education is 1.77 % which is very low. This percentage is even lower for females. b. The literate population of Turkey is concentrated in the younger age groups : 40.6 %" of literates are in the 11-19 age group. This is related to the fact that younger generations have a greater opportunity to be educated. It is derived from the statistics that the younger generation has more opportunity for higher education than the older generation and this is considered a promising progress. .

82 THE STRUCTURE OF THE TURKISH POPULATION

NOTES

1. Calculated from Demographic Yearbook 1970, United Na- tions, 1971. 2. Haluik Cillov, Turkey's Economy, Istanbul, 1972, p. 62 (In Turkish). 3. Calculated from the Turkish State Institute of Statistics publications : 1965 Census of Population, Ankara, 1968, p. 39, 1970 Census of Population, Ankara, 1971, p. 2. 4. Demographic Yearbook 1970, p. 370. 5. Erol Tiimertekin, Distribution of the Male-Female Ratio in Turkey. Journal of the Institute of Geography, Volume 5, Number 9, Istanbul, 1958 (in Turkish). 6. 1970 Census of Population Results Based on 1 % Sample: 7. De H. Furgac, "Incorrect Age Reporting in Turkish censuses of Population", Journal of Faculty of Economics, Volume 10, Number 1-4, Istanbul, 1960, p. 64 (in Turkish). 8. In this section, the Turkish data is calculated from Statistical Yearbook 1968 and 1970 Census of Population, and. the foreign data from U. N. Demographic Yearbook 1970. 9. Orner Celâl Sarc, Age and Sex Distribution of Population, Diyarbakir Lectures, Istanbul, 1942 (in Turkish). 10. Ratio of those under 15 and above 65 to 15-65 age group. 11. Second Five Year Development Plan, State Planning Organi- zation, Ankara, 1967, p. 48. 12. Statistical Yearbook 1965, p. 64, and 1965 Census of Pro- pulation, p. 242. 13. "1965 Census of Population; Social and Economic Charac- terists of the Population", State Institute of Statistics, Ankara,

. p. 242. . • . ' . • • . . , ••';• ;;";

83 THE POPULATION OF TURKEY

14. 1970 figures are calculated from 1970 Census of Population Sample Results. 15. Haiuk Cillbv, Turkey's Economy, Istanbul, 1972 (in Turkish). 16. Serim Timur, Turkish Family Structure, Haeetitepe Uni- versity, Ankara, 1972. 17. Ibid. 18. Results of 1970 Census of Population, 1964-65 Statistical Yearbook and 1968 Statistical Yearbook. 19. Methods de Secensement de la Population N. U., New York, 1950, p. 70. 20. 1955 Census of Population, Istanbul, 1961, p. 122, Statistical Yearbook 1960-62, Ankara, 1964, p. 79. We should add that the distribution of the population by nationality is not yet calculated for the 1965 and 1970 Censuses of Population. 21. 1963 Statistical Yearbook, p. 63 and 1965 Census of Popula- tion : Social and Economic Characteristics of the Population, Ankara, 1969, p. 227. The distribution of the population by religious affiliation is not done for 1970 Census yet. 22. "Statistical Yearbook 1959," Ankartai, 1961, p. 79, "Statistical Yearbook 1960", Ankara, 1964, p. 76, "Statistical Yearbook 1968", Ankara, 1969, p. 40, "1970 Census of Population", An- kara, 1972, p. 3. 23. Population. Information, Population Reference Bureau, Washington D. C, March 1968. 24. Methods de Recensement de la Population, p. 83. 25. 1970 Census of Population Sample Results, p. 4. 26. Haluk Cillbv, Turkey's Economy, Istanbul, 1972, p. 96, (in Turkish). 27. Demographic Yearbook 1963, a U. N. publication, New York, 1964, pp. 417-481. 28. Table is prepared from the data in 1970 Census of Population Results. 29. Haluk Cillov, Turkey's Economy, p. 106. 30. There are around 2 million disguised unemployed in Turkey.

84 Chapter IV

INTERNAL MIGRATION AND POPULATION REDISTRIBUTION

by Handan TEZMEN, M. A.

INTERNAL MIGRATION AND POPULATION REDISTRIBUTION IN TURKEY

Handan Tezmen, M. A. Ph.D. Student at University of Pennsylvania

I. Introduction The main objective of this paper is to analize the trends and volume of internal migration in Turkey for the period 1955-1965. Turkey can be considered as an illustrative example for the discussion of high and constant fertility, high but fast declining mortality and rapid urbanization in today's underdeveloped countries. Therefore, from this point of view, to analize the pattern of internal migration can be also helpful in understanding, in general, the various effects of economic variables on internal migration. H. Spatial Unit of Analysis and Definitions For this paper we adopted the five regions (see map I) of the Turkish Demographic Survey as the spatial units to analize the internal migration among these regions. These regional boundaries ivhich are arbitrarily drawn to coincide with the provincial boundaries are approximately composed of 13 provinces (there are 67 provinces in Turkey) with similar social, economic and demographic characteristics. Before going into the analysis of internal migration, let us first give its definition : Internal migration can be- defined as "change of residence from one clearly defined geographical unit to another within the national boundaries"1. On the basis of this general definition ¡we can also define the lifetime migrant as a person whose area of residence at the census

1 D. S. Thomas, "Research Memorandum on Migration Differentials", Social Sciences Research Council Bulletin, (43. N. Y. 1938), P. 4.

87 / Í i

j :• / .^ /. \ -' ,3 {, \ ;/' i • i i '_ Ö' ""•' \ • 0. i : c • • Ss ;• : • i¿ ..• \

" 1 * • V. \ • •: 1 N \ • Z J s • U4 i S*- Q '•-: (' }}^ \ \ / \ \ • : s I/ - ! t >' 1 • -ZÍ--/ ^\'\I -. — V ^ ,;-•• ,¡ v ; c \ \c i -i.s )•••• ri 'r\^ «

KA R • lea l j A| 17* 2 : ,_! f ? '*••- 'm v^ )• / W- Of /f á '•••.•' 9 A ÍG E *». "iiT^ • Al.T i ••i/ f! = "•i• .. 0 • 1 :TWS \; • y.. 1! *V ¿£ i 1 M i c~> THE POPULATION OF TURKEY or survey date differs from his area of birth2. Thug, it is possible to classify the population enumerated into two groups : (1) Non-migrant : A person whose place of residence, at the time of enumeration, is the same place where he was born. (2) Migrant : A person whose place of residence is different from his place of birth, at the time of enumeration, within the national boundaries. in. Data and Methodology In order to give the possible maximum knowledge about the internal migration volume and direction, we made the study in both lifetime migration method and census survival1 ratio method. Be- cause lif etime migration provides us 'both the volume and direction of migration while the latter method gives us migration volume by ages. A) Lifetime Migration Censuses are the only source of data to estimate migration in Turkey. The iformation directly related with migration which censuses have is the cross classification of the province of birth and the province of residence by sex. These data are represented as a matrix of 67 by 67, that is by province of birth and province of resindence, exluding the foreign-born. We reduced this 67 by 67 matrix to 5 by 5 (Tables IA, IB, IC and ID). These tables show the distribution of population by regions and sex, where the volumes show the places of birth and the rows show the places of residence, and the diagonal cells give the lif e time migrants for each region. Thus the migrant category is subvidided into migration streams on the basis of five regions of birth places and five regions of places of residence. The sum of lifetime immigrants or lifetime outmigrants gives the number of persons who were enumerated away from their birth places; that is the number of lifetime migrants for the country3. This

2 United Nations, Methods of Measuring Internal Migration, ST/SOA/Serles A/4T, New York, 1970, P. 2. ' 3 United Nations, op. dt. p. 5. 89 CO CO 00 cq CwO ai o CO CO o •**Oí •* o 00 tr- CO CO in O3 i 00 03 rH T-í rH co tr- o o Ir- rH 00 g IN IN CO rH IN l

tí IN in tr- 00 cq CO CD in o o o CO t- 00 CN 4- 00 00 t- O3 i-< in a Ü in o CO CO So •* 1 T + 1 I

,aJ S? 29 1 89 7 59 9 4- CO 23 3 1 O -44 O3 S m o t> n cq O i rH

O3 o in

cs 03 1 ra-••3 CO O3 CD o •* t-

c 20 7 55 0 T-i rH O Migr a

> oo CO CD rH t- tí CO a co t- in rH o m CO r-^ r-i tr- m 'oí cq CO 00 CM t~ O3 a) •* r-i O CO t- H a> CO tí r-t t- rH

IN tr- CO 00 M CO 03 00 in I oo O in 03 00 O CM co IN in 00 CO To CO O3 in

iD 16 0 tí ï •?

IO Birt h 00 CO C3 in V B T-í IO I— a CO H IM rH IN tí o T-l CO r^ •*

CO 6 18 3 'S ' S ;(Tot a CO IN IN CO 00 o 1—' tí '§? in H O tí co w o IN IN . . J3 O 'So) o o tí w cq N rH IN ri H 2 M CO rH O3 in CO > rH tí in CO IN Ir- o in 03 CO

C O Ä iH th e 66 7 co m O3 nu s 1 .2 oo O0 «S:

¡nu s 's & 3 83 6 tíO 8 I «M

S5 g in O. rH CO- co 03 in CM m

1Í5 O Oí CO Oi OÍ' in iH IN. T-l O CO o co «0 o CX) IN o CD

CM rH Oí t> Oi CO CD co O rH CgO •tf CO in IN O O CO CO CD 1O CO O fi J? «o

tí O IN o m Oi in 'S) Oí Oí CO Ci o tí i rH o CM > CD in in oo H

in o 00 a co t- CO o O) H •* r-t (3 o m in I m H CD Cl CD tí CD • N rH CM rH in Oî CD l-l 00 o H CO CM CO O To co 3 CO CM,

CO Oí ' in a l- 01 00 o Oí CD 1m- l> cq cO 11*~ tí M T-l

rj tl) H H S t > g g a • a a a .2 T O O Q O '¡10 so 'oo $1 a>

ffia •s* S?I a> SS co e» 00

S g S S co N t- oo 1(»5 CO tO 1-1 co O TH CO co T-l r-l IN

01 eo r-l CO CD •* co o o S S | S S eq iß H

r-l Oi « r-l C0 i-l •* ü 00 CN o ce w o> to w w TH to i-i t- Oí IN es

1A en eu IN N S M •* IN oo CO 05 ^ ri o

IN r-l CO S i i CO 00 CN S 00

CO s oo rj 1-1 en r-l JH cp co o

•g

r-l •* Ol co 00 O> •* CM mm« t- m co to to ce co iH to rH t- co 2 2

•a - i

92 CO CO 74 5 53 2 25 3 72 5 18 1 CO IM rH t- ClMO itn CO CN CO CN eq

CO OO CO oCO i* 4-1 *J 8 S rH CN i-l 0> cj 1-1 g i-l CO

CN « en CO CN CO &* CO "i CO © co tN (N 2 8 i-H 1 § 00

t- O 0* O „s i ÏH ?5 S CO i Co in IM t- O1 g in CO 00 ^

ft oo 03 CO oo Q M "* * oo 00 1O iH 1O tN I S t

co g co

CO CO rH CO

tí S 00 00 CN CC CD 35 CO IN w (N CO i-i CO 8

U6 o t- CO t- 1-H S (N oo IM 1Í5 CO

fe tN O 00 5005 S© C¡5O O § CO Cp © «5 1O CO O) tN CN rH a>

CO P S •I PU)

93 FIGURE IA - Ratios of lifetime in-migrants and lifetime out-migrants, to the non migrants, by regions, and sex; 1955

• Males 30- R%3 Females i .20-

.10-

.00- « j m P mTG m^ .10-

.20-

.30-

I II HT • E -, Y Regien FIGURE IB- Ratios of lifetime in-migramts and lifetime outmigrants, to the non migrants, by regions and sex; 1965

Region FIGURE HA- Ratios of net lifetime! migrants, to the non migrants, by regions, and sex; 1955

Y. Region FIGURE IIB - Ratios of net lifetime migrants, to the non migrants, by regions, and sex; 1965

IS. Y Region INTERNAL MIGRATION AND POPULATION REDISTRIBUTION total is obtained from the aibove mentioned tables by subtracting the numbers in the diagonal cells from the corner grand total (see notes below the above mentioned tables). The sum of net life time gains (or the total of net losses) is a measure of redistribution due to lifetime migration for the country as a whole (Tablell). This table is obtained from the tables mentioned formerly by horizontal totals and summing the differences with like signs. The sum total of lifetime immigrants for all the regions in a country is equal to the sum total of Mf etime outmigrants, since each immigrant to an area is an outmigrant from some other area4. There- fore the sum of net balances for all areas is necessarily zero. After obtaining in and out and net lifetime migration as explained above, we have computed the percentage table (table HI) for the in and out lifetime migrants for 1955 and 1965 by regions and sex. There are several theoretical problems in the measurement of migration particularly in lifetime migration. It is impossible to com- pute a rate in terms of annual number of migrants. However, for the purpose of comparison among regions, intensity and as a relative measure of the migration phenomenon, migrant to non-migrant ratios are calculated using migrants as numerator and non-migrants as denominator (Table IV). Findings : In order to understand the findings better it is necessary to point out very briefely some socio-economic characteristics of the regions. Among the regions, there is an uneven distribution of natural richness. For instance in region IV the agricultural land is abundant and rich and new openings of land for cultivation is very usual through irrigation whereas no such possibility exists in Region II which is very mountaneous even though the main economic activity is still agriculture.

4 ibkL P. 5 96 s g s s s es —i eo" rH T-i W S, rt 1-1 Ci

•41 00 t- OS H rH IN O) "Ï t- CO 00 e * 6 «

Iß 00 o CO

OS OS in a o 00 t- 3 00 r-i rH es 3) o CO t~ O al es in en

»G ao, gin gm I «.C.O .". OS o S3 a a a 8

Cl os os e} t- os eo io os in •*. m oo"

00 CO rH m CO CO CI es N OÍ OÍ o 03 CÓ 00 C0 t-l ri rt

Ol H * qO es o

S g g g g ï tí"S § I 'f 97 r-( £- 1 +• r + r

t -* O3 »o•* n5 O O O o

0> H rH to ¡Ó O M CO u5

s s i 1 +

O) m - N I

t- o co w te O O iH o

N O 2 O S S S 8 O H O O I". + I f « rl » $ It t- M M S

g t- •* o S t- •t • * •* S3 »-1 O O iH P O

03 2 P SñS 8 I í + I

M O

S cq t- 00 M o O ©

g g g § § •a •& •& •& •&

98 TABLE IV VOLUME OF NET MIGRATION BY SEX, AGE AND REGION

Males Region I Region II Region III Region IV Region V

10—14 — 5 975 — 2 077 + 54 299 — 310 — 27 523 15—19 —12 035 — 38925 + 109 036 — 8 368 — 48 721 20—24 —44 683 — 72 754 + 115 677 —22157 + 8 397 25—34 — 7960 — 9 630 — 4 776 + 16 952 — 88 022 35—44 — 5 300 — 1732 — 16 747 — 4 778 — 1625 45-54 — 85 — 3 691 + 15 834 — 2 956 — 9 646 55—64 — 1287 + 2 832 — 3 410 — 1220 + 5 492 65 + i+ 1406 + 4641 + 15 834 + 1580 — 2 746

TOTAL —75 919 —121336 —267 800 —21257 -164 394

Females 10—14 —18 976 — 21686 + 64 048 + 1377 — 56 660 15—19 — 8 586 — 10 502 + 62 942 — 6360 — 41108 20—24 — 2 406 — 22 269 :+ 8 391 — 1143 + 17270 25—34 1+19 862 — 37 712 :+ 401 —12 240 ,+ 9 682 35—44 —20 987 — 13 330 ,+ 44 087 + 1268 — 26 541 45—54 — 3 505 — 2072 ;+ 19 591 — 3 569 — 11 942 55—64 — 1575 — 4960 + 10135 + 211 — 6481 65+ — 9 695 — 2119 ;+ 15 926 — 425 — 8 355

TOTAL —45 868 —114 650 +225 521 —20 881 —124135

99 \ INTERNAL MIGRATION AND POPULATION REDISTRIBUTION

It is not only the agricultural land, but also the urban settlement is very uneven due to the uneven concentrtion of industrial centes. In this respect Region III is the most industrialized and urbanized rer gion within which two metropolitan areas, Izmir and Istanbul, are located. The least urban and industrialized and with relatively poor. incuLtivaJble land is Region V, Region I can be called as the transitory region in every respect, especially for the migration flows. It is the transition area between Region V to III and Region II to IV. If we. exclude Ankara, the nation's capital, it is mainly agricultural and rural in settlement. Throughout the period under study, Region in has been the predominant center of attraction for migrants. It is the only region which gains population through migration in both census year (figures IA, IB, IIA end IIB). All the others are losing population due to internal migration. Both Regions II and V are the main regions of loss of population. While Region II loses population to Region I and III, Region V loses population mostly to IV and I. There we see the important role of distance in migration. The general view given by these data is that a large proportion of the migrants to any region was 'born in the adjacent regions. The highest loss of population is from Region II where agriculture is predominant and cultivable land is scarce. In 1965 this loss was reduced for Region II while the second most losing region, Region V, began to lose more. As for Region I, if we had controlled for Ankara, it would have shown a much higher rate of loss of population. Region IV is the most stable region from the migration point of view due to its rich and abundant agricultural land. ~ Furthermore, when we look at the findings from the sex selectivity point of view we see that males are much more migratory than females. In 1955, without exception, females migrated less than males. In 1965 the sex difference was reduced even though the main pattern remained the same.

100 THE POPULATION OF TURKEY

So far, this method of analysis provided us with the estimates1 of the volume and direction of migration by sex among regions. However, from our definition of lifetime migration, we can conclude that the estimation of migration by this method gives us the mini- mum of the total amount, as the definition assumes a single movement directly from the region of birth to the region of residence. • Therefore the final result is balance of surviving migrants which excludes return migrants, multiple movement and dead migrants. For this reason, now let us look at the net migration rates by age and sex to complete the analysis of internal migration. B) Net Migration Estimates by Census Survival Ratio Method In order to estimate the net migration from the census age data an adjusted census survival ratio method is used. A census survival ratio is simply the ratio of the population age x at a given census to the population (x-t) at the census (t) years earlier. Computed for a nation as a whole, this ratio is then multiplied by the population aged (x-t) of the spatial units at the first census, and the "expected surviors" subtracted from the corresponding population at the second census to yield estimates of net migration5. ; There are three basic assumptions involved in this method for the utilization of the survival ratios : i. Closed population : This assumption cannot be valid for Tur- gey due to the migration of Turkish workers to West Germany during the 1960-65 period. Therefore international migration adjustments were made to obtain the closed population figures6. Briefly, the method of adjustment involves the estimation of the survivors of the external migrants in 1960 and the application of this figure to the 1965 census population to get the closed population figure.

5 K. C. Zachariah, A Historical Study of Internal Migration in the Indian Subcontinent, 1S01-1931, New York, Asia Publishing House, 1964, Ch. 4 P. 136. 6 During this adjustment process, the immigration of 29.300 people from the Balkan, countries and the emigration of 220,000 Turkish workers, 20,000 families and 4,800 nstumes were taken into account. These figures are taken from Samira Berksan's unpubliahed paper- "Interregional Migration in Turkey, 1955-65", Universiyt of Pennsylvania, May 1969. 101 • INTERNAL MIGRATION AND POPULATION REDISTRIBUTION

ii. Equality of survival ratio among1 regions : Also this assumption is not valid for Turkey where mortality rates, especially the infant and child mortality rates, vary considerably among regions. Therefore an adjustment is made by using the data of Turkish Demographic Survey. We obtained e°o and e°5 values first from the Turkish Demographic Survey and then by choosing the corresponding model life tables. After choosing these life tables we adjusted our national census survival ratios for mortality differences among regions7. iii. The third assumption is that the ratio of the enumerated population in any age-sex group to the actual population is the same at each census, for each region as for the nation. The fact that the percent error in the national census is the weighted average of the regions, percent errors can be taken for a built-in correction factor for under or over enumeration for the utilization of the census survival ratios. Therefore having no other source for the adjustment of enumeration errors, we can assume the variation range for the regions is small enough to proceed With the method of census survival ratio. / We computed the rates of net migration by taking the average population as our base; that is, by taking the arithmetic mean of the cohort population at the beginning of the decade and the cohort population at the end of the decade. Findings : In general the findings through this method are in line with the findings of the former analysis. Again, the two most losing regions are the second and the fifth. Region IV is the most stable. Region I, due to the inclusion of An- kara, does not show heavy losses. Economically the most developed region, Region III, is the only gaining one. Region IV is also the only region where female migration is equal to the male migration in volume and rate, while in all the other regions males are much more migratory compared to females.

7 This method has been explained and utilized in detail in the above mentioned atujdy made by S. Berksan. 102 THE POPULATION OF TURKEY

The labor force movement from the east to the west is again one of the observations clear through our findings. As we said before, the main advantage of this method is that, it gives us the possibility of analizing age selectivity in migration. This age selectivity of migration is mainly clue to : i) Induced migration in the case of the males who are required to go to the military service, ii) Migration of the labor force from rural to urban or from less developed to more developed agricultural regions. This labor force movement starts at early ages, (Table IV and ' Table V). The peak ages for migration are 20-24 for males and 10- 19 for females in general. This general pattern for males does not show itself only in Region V where the 20-24 age group shows a heavy loss. The gain can be attributed to the military establishments in this region which draw young males from all over the country. For the loss in the 25-34 age group, the reason can be due to the fact that this region ia economically the least developed one where there is lack of job opportunities. The main reason for the very young women to migrate more than the others is again an economic one. In Turkey the service sector is growing faster than the rest of the economy, and this growth is partially due to the abundance of job opportunities in domestic services. Another reason behind these high rates of migration for younger females is.the family migration. In Turkey, as in most of the other countries, there is an age difference between the husbands and Wives. Therefore when a male of 20-24 migrates with his family, the same movement affects the female migration rate of younger ages. To conclude our findings let us see the general internal mobility in Turkey by referring to another study made by Irene Taeuber.8 According to this study the level of internal mobility was fairly low arid did not show any significant increase over the decades. In the

8 Irene Taeuber, "Population and Modernization In, Turkey", Population Index, Vol. 24 April 1958.

103 TABLE V RATE OF NET MIGRATION BY SEX, AGE AND REGION

Males Region I Region II Region HI Region IV Region V

10—14 — 11.82 — 5.07 + 117.89 — .90 — 69.03 15—19 — 31.16 —140.09 + 259.00 — 33.94 —169.00 20—24 —157.23 —374.82 :+313.99 —124.16 + 36.01 25—34 — 15.01 — 25.99 — 5.98 + 52.00 —220.90 35—44 — 13.78 — 6.01 — 30.09 — 20.00 — 6.07 45—54 — .31 — 19.00 + 45.01 — 18.81 — 60.97 55—64 — 5.24 + 15.84 — 10.99 — 9.23 • + 44.99 65 + + 9.92 + 39.29 — 9.03 + 16.01 — 22.99

TOTAL — 27.96 — 62.07 + 77.98 — 12.99 — 83.99

Females 10—14 — 42.01 — 61.16 :+148.96 + 4.00 —169.95 15—19 — 24.00 — 40.05 + 167.97 — 28.14 —164.02 20—24 — 8.70 —113.91 + 28.01 — 6.99 + 91.98 25—34 + 37.00 — 89.98 + .59 — 39.02 + 24.99 35—44 — 52.01 — 44.12 + 92.04 + 5.11 — 96.03 45—54 — 14.81 — 9.90 ;+ 62.06 — 23.93 — 73.90 55—64 — 7.02 — 25.95 + 33.00 + 1.02 — 51.05 65 + — 50.03 — 10.84 + 49.03 — 3.04 — 69.92

TOTAL — 16.96 — 55.01 + 72.00 — 13.00 — 67.99 lOi THE POPULATION OF TURKEY same study it is found that only 8 % of the Turkish population lived outside their province of birth in 1935, and by 1950 the figure was still only 9 % \ However, between 1950-60. this proportion rose to 13 % 10. Although these figures underestimate the total internal migration, they are still sufficient to indicate that since 1950 there has been a significant increase in internal mobility due mostly to the growing of the industrial sector and the advancement of technology in agriculture and transportation which altogether facilititated and gave rise to the need for the laJbor force redistribution among regions. IV. Interregional Migration Mobility, among provinces within the regional boundaries reveals a trend similar to that of intraregional, migration. In all the regions the percentage of people who were born and living in the same province at the time of the census decreased between 1955 and 1965. However this decrease is greater than the decrease of those who were born and living in the same region. Therefore, the ratio of those who were born and living in the same province to those who were born and living in the same region is decreased from 98.2 % in 1955 to about 96.3 '% in 1965. This means that in the ten year period, the number of intraregional migrants has increased by approximately two percent. Another noteworthy finding is that every region has shown a gain of approximately 1 to 3 percent in intraregional migration. This fact stands in contradiction to the large differences we have observed among regions with regard to gains (or losses) from intraregional migration. Thus, we can conclude that intraregional migration is increasing at approximately similar rates in all the regions whereas the interregional migration grows fast but unproportionately with regard to different regions which creates problems from the population distribution and balanced economic growth point of view.

9 Ibid, p. 109. 10 Turkey, Institute of Statistics, Census of Population, 23 Oct. 1960, Ankara, 1962.-

105 INTERNAL MIGRATION AND POPULATION REDISTRIBUTION1

V. Population Redistribution and Urbanization To have a complete picture of the mobility of the Turkish population, let us very briefly look at the population redistribution. The index of redistribution that we utilized for this purpose can be defined as the percent of population in 1960 or 1965 that would have to be redistributed to obtain the 1955 distribution.

The results we obtained are as follows : Index of Redistribution

Males Females 1960 3.15 2.85 1965 3.89 3.61

Depending on these results we can state that the population redistribution in Turkey during the 1955-65 period has been moderate. It is also evident that even if the redistribution of females is at lower levels, the rate of increase over the 10 years is very similar to that of males. When we look at the situation from the regional point of view the region that has the highest gain due to immigration between 1955 and 1965 is again Region III which is relatively the most industrialized one. Therefore we can generalize this fact as "the population redistribution is from the eastern part to the western part of Turkey where most of the large cities and the relatively advanced industrial establishments are located". This increasing concentration of population in the west is also the characteristic of the Turkish urbanization process. Due to migration from the rural to urban areas the growth rate of the cities has always exceeded that of the country as a whole. Another fact is that the larger cities are growing at a faster rate than the smaller cities, and thus they form a larger part of the total population than they did before11.

11 Irene Taeut^r, op. cit p. 104 106 THE POPULATION OF TURKEY

Kemal özok, also, notes that 61 % of those who have ever migrated within Turkey were living in cities of 100,000 or over at the time of the 1960 census, and six of the nine of these cities are in the west12. Thus we can conclude that the process of urbanization in Tur- key has contributed significantly to the unequal distribution of population over the country through internal migration.

12 Kemal özok, "Urbanization and Internal Migration In Turkey", United Nations, World Population Conference, 1965, (unpublished paper), p. 6

107

Chapter V

THE LABOUR FORCE

by Sunday ÜNER, Ph.D.

THE LABOUR FORCE Sunday Uner, Ph.D. Institute of Population Studies Hacettepe University Introduction The labour force is generally defined as that part of the population which participates in economic activities. In other words that part of the population which contributes to the supply of labour available for the production of economic goods and services during a specified period. Therefore, the labour force of a country covers the individuals who work for profits and gains or those who are not at work but are eager and looking for work and profit That means that the labour force covers all those who are employed and unemployed. In Turkey, data on economic activities of individuals are collected by means of quinguennial censuses. Data on occupational distribution of the labour force are hardly reliable for the period of 1927 to 1945. From 1950 onwards, censuses provide reliable data on the distribution of the labour force by individual occupations, by branch of economic activity and employment status collected according to international standards. Size of the Labour Force and Participation Rates , Turkey's population reflects the characteristics of less developed countries. The size of the population and the high rate of population growth are the dominant factors in determining the size of the labour force. Labour force participation rates determine what portion of the population joins the labour force. If the effects of participation rates are not considered, the population with a high rate of growth means, also a high rate of growth in the size of the labour force. It is obvious that this does not mean the full employment of the labour force. On the contrary, the increasing part may affect the size of visible and disguised unemployed or underemployed. In fact, a fast growth of the labour force due to

11* THE POPULATION OF TURKEY a high population growth rate gives rise to an increase in the rate of visible and disguised unemployment as in the case of most of the less developed countries and Turkey. Turkey's labour force was 5 million in 1927. In 1965, it reached 13.5 million, an increase of more than 2.5 times in 42 years. Accord- ing to the 1970 census, it is now 14.5 million. In addition to the effects of a high population growth rate, increasing rates of female participation has generated the countinuous increase of the labour force. As it can be observed in other less developed countries, 70 .% of Turkey's population lives in rural areas. This phenomenon does not only affect the size of the labour force through high fertility rates, but it also increases the rates of female participation through the economic activities of small agricultural establishments employing unpaid family workers. The size of the population and its age and sex structure is not sufficient for determining the size of the labour force. Although the 15-64 age group of the population is a potential source for labour, it is hardly possible that the bulk of this group is involved in economic activities. A number of factors of different origins affect the labour force participation rates which determine the size of the labour force. Apart from employment opportunities, many factors of non-economic origins affect the rates of female participation. Changes in the labour force participation rates in Turkey are given in Table 1. Table 1 Changes in the Labour Force Participation Rates (1950-1970) Total Female Male ;Labour ]Force Labour Labour Labour Participation Rates Years Force Force Force Total Female Male 1950 10,724 4,766 5,956 51 46 57 1955 12,040 5,282 6,758 50 45 55 1960 12,993 5,295 7,697 47 39 54 1965 13,591 5,187 8,404 43 33 52 1970 14,533 na na 41 na na Source : H. Cilov, Turkish Economy, p. 93, Sennet Matbaasi, Istanbul (in Turkish) and State Planning Organization, Third Five Year Plan. 112 TBE LABOUR FORCE

Inspection of Table 1. provides the following information : 1. Male participation rates are higher than female participation rates as it is observed in other developed and less developed countries. 2. Labour force participation rate has started to decline from 1955 onwards. This decline has been rather sharp in recent years. 3. In the same period, female participation rates have shown a sharper decline than mate rates. These changes in female participation are largely due to the census methods and different definitions. Especially the treatment of unpaid family workers affects to a great extent female participation rates. 4. In Turkey, the labour force participation rates have to do with the age structure of the population. Meanwhile it partly reflects the level of economic development of the country. In fact, the high participation rates for females due to the effect of unpaid family workers inflate the genera! participation rates. If this effect could be eliminated, the general participation rates would (be a much lower level. 5. Finally, with the present data available in Turkey, it is very difficult to set up the real trends in the participation rates. Different techniques employed at each census with different definitions make it more complicated. Therefore, one should act cautiously when comparing these rates.

Labour Force Participation and Unemployment

Turkey has a young population with high rate of growth. Hence, a considerable portion of youngsters join the labour force every year. Labour demand can not match this dynamic growth of labour supply. The chronic imbalance between the labour supply and demand forms the structure of employment and limits its size. Therefore, the solution of the problem largely depends on the possibility of an increase in labour demand to a degree that Will absorb visible an disguised unemployed and/or underemployed. 113 THE POPULATION OF TURKEY

Although the strategic importance of population growth in solving the employment problem has been recognized earlier with the introduction of the Firts Five Year Plan in Turkey, the approaches were different. The first plan sees the solution of the employment problem as an independent target. The second plan tackles it as a dependent variable of the income-increasing target. Finally, the third plan poses the employment as a dependent variable of economic growth. Accepting that industrialization alone will not create enough jobs to solve the employment problem in the short run, the third plan foresees special provisions in this context. Table 2 indicates the trends in employment for the period of 1962-1972.

114 III its

O CO r-j Ui »O T)Î CO \ñ CO <£> i-î I? OOrlHNMiíinfflNllO

riHHHrIHHriHHH ¡ ..i OOOOOOOOTHT-I

î T™I i'* î î î î TH ^H ^H TH £i ^5 î î T^ co ^î *rTi c? ^^ co i^ O Cv( CO IQ €O 00 *O TH C^í ^Î CO 'I O ^D ^5 c5 w> ^^ TH TH TH î r4 -3 I

o> p ö ci ^î cô a w io oö T-Î -^î t^

C^CO^lOCDt— OOGi O ri W CO CO CD CO ^ CO CO CO £*• t** b- HHT-ÍHT-ÍHTWH T-ÍTH-H 1

115 THE POPULATION OF TURKEY

The first point to consider is the emergence of the "labour surplus" which was sharply increased during the indicated period. This clearly shows that employment targets have not been realized. According to the figures of the Third Plan, the "labour surplus" which indicated the difference between the labour supply and employment has risen from 235 thousand in 1962 to 750-800 thousand in 1972. With the addition of surplus labour in agriculture, the total labour surplus has reached 1.6 million- at the end of the second plan. The second point is related with the structure of employment. The third plan identifies the labour surplus as the main problem to be tackled. It also claims that the nature of the labour surplus can be recognized as underemployment rather than visible unemployment. But, due to a lack of employment statistics, the portion of visible unemployment within the total labour surplus can not be figured out definitely. Therefore, the third plan admits that the plan figure is only an estimate which covers both unemployed and underemployed. Composition of Labour Force by Major Economic Sectors The sectoral distribution of Turkish labour force reflects the characteristics of a less developed economy. The majority of labour force works in agriculture. Table 3 shows the distribution between 1955-1970. - Table 3 Distribution of Labour Force by Major Economic Sectors

1955 1960 1965 1970

Agriculture 77.4 74.0 71.4 67.0 Industry 8.1 9.6 10.2 12.1 Services 4.0 . 5.4 11.1 11.4 Others 10.5 11.0 7.3 9.5

100.0 100.0 100.0 100.0 Source : State Institute of Statistics. 116 THE LABOUR FORCE

Although there is a considerable decrease in its coverage, still 67 % of the labour force works in agriculture in 1970. The industrial sector covers 12.1 % of the labour force, 8.4 % of this ratio covers the manufacturing industry. Same ratios for 1965 was 10.2 % and 7.2 % respectively. That implies a transfer of labour from agricultural sector. For the third sector a sharp increase can be observed after 1965.

Composition of Labour Force by Branch of Economic Activity- Distribution of labour force by branches of economic activities shows similar trends as in the distribution by economic sectors. Table 4. gives the distribution of labour force by branch of economic activity between 1955-1970.

Table 4 Distribution of Labour Force by Branches of Economic Activities

1955 1960 1965 1970 Agriculture, Hunting Forestry and Fishing 77.4 74.0 71.4 67.0 Mines and Extraction Industries 0.5 0.7 0.6 0.8 Manufacturing Industries 6.0 7.2 7.2 8.4 Construction 1.6 1.7 2.4 2.8 Electricity, Gass and Water Production 0.1 0.2 0.2 0.1 Commerce, Banking, and Insurance 2.8 3.2 2.9 6.1 Transportation and Communications Services 0.5 1.6 2.1 2.6 Services 4.0 5.5 6.4 11.4 Not well defined 6.0 5.5 6.5 0.9 Source : State Institute of Statistics. 117 THE POPULATION OF TURKEY

The ratio of those working in agriculture, hunting, forestry and fishing is as high as it can be expected in less developed countries. The ratios of those working in the branch of extraction industries and mines together with electricity, gas production are low as they aie in most of the developed or less developed countries. The ratio for the construction industries is also low. This is a common feature of less developed countries. On the other hand, the ratio for commerce, banking and insurance is only 3 %. The same ratio for developed economies reaches up to 15-20 '%. Similar observations can be made in the branches or transportation, communication and activities related to services. When again compared with the ratios of developed countries, the ratios are found to be much lower. Another distinction of developed economics can be seen in the ratio of those engaged in manufacturing industries. The ratio here is 25-35'% for developed countries. In Turkey, the same ratio is only 8.4 %. One of the features of the Turkish economy is the slow rate of growth of industrialization. In spite of the efforts of recent years, the rate of industrialization has not yet reached the level that can inject an industrial character to the structure of the labour force by eliminating the effects of a high population growth.

Occupation Composition of the Labour Force The occupation composition of the labour force in Turkey again reflects its less developed and agricultural character. The following table gives the distribution of labour force by occupational groups according to 1965 and 1970 census.

118 THE LABOUR FORCE

Table 5

Distribution of Labour Force by Occupation and Sex in 1965 ( % )

1965 1970 Male Female Total Total Professional and Technical Personnel and Related Occupations 2.7 1.1 2.1 3.4 Administrators and Enterpreneurs 3.8 . 0.8 2.6 2.9 Salesmen and Related Occupations 4.4 — 2.8 3.1 Farmers, hunters, forestrymen, and related occupations 57.0 94.0 71.7 66.8 Miners and related occupations for extraction 0.7 — 0.5 0.7 Occupations related to Trans- portation and Communications 2.7 — 2.0 2.5 Craftsmen and construction workers 12.0 2.8 8.7 12.6 Occupations related to services 4.8 — 3.5 3.9 Unknown occupations 10.0 _ 3.7 4.6

Sources : State Institute of Statistics, 1965 Census and 1970 Census.

Inspection of Table 5. provides us with the fact that a great majority of the population engages in occupations of agricultural nature. This is especially certain for females (94 % in 1965). Even for males the relative importance of other occupations of non-agricultural nature is quite low.

Composition of Labour Force by Employment Status

Censuses in Turkey provide also information about employment or occupational status of individuals who are involved in economic activities. Table 6. below gives the breakdown of the labour force by employment status acording to the 1965 census. 119 THE POPULATION OF TURKEY

Table 6 Distribution of Labour Force by Employment Status in 1965-1970 (%)

1965 1970 Number Ratio Number Ratio (thousand) (%) (thousand) (%) Employers 133 1.0 101 0.7 Self employed 3,886 28.7 4,164 28.6 Unpaid Family Workers 6,419 47.3 6,438 44.3 Employee 3,638 22.4 3,757 25.9 Unknown 82 0.6 73 0.5 Total 13,558 100.0 14,533 100.0

Sources : State Institute of Statistics. 1965 and 1970 Census. Table 6. reveals that a great portion of the labour force com- prises unpaid family workers. (47 % in 1965 and 44 % in 1970). This group is largely employed in agricultural activities. We should give atention to the fact that the ratio of this group in developed countries is only 1-5 %. Parallel to this group, own-account workers are the second largest group (28.7 % of the labour force). We should note that unpaid family workers are employed "by this group rather than the employers group. These two groups together cover 73 % of the labour force in Turkey. Meanwhile, the ratio of these two groups in developed countries only reach 15-20 %. This feature also denotes the size and organizational shape of economic activities in Turkey as it'does in other less developed countries. The sex differentiation in each statuts group is also interesting. According to the 1965 census, 96 % of the employers and 90 % of own-account workers are males. On the other hand, in the unpaid family workers group, the ratio is 26 % for males and 76 % for females. We should add that males in this group are belonging to rather young age groups.

120 THE LABOUR FORCE

Labour Force and Education Educational institutions are regarded as the. major means of training the various types of the labour required in less developed countries. In this content, knowledge ajbout the educational attainments of the labour force and the degree of literacy are useful. Taible 7. gives the distribution of literates by branch of economic activity.

Table 7 Distribution of Literates Iby Branch of Economic Activity

Branch of Activity 1960 1965 1970 Agriculture 16.4 33.4 37.9 Extraction Industries 44.7 71.1 74.9 Manufacturing Industries 56.7 83.7 81.0 Electricity, gas, water production 48.8 70.4 93.0 Construction 43.1 8218 66.0 Commerce 62.5 91.0 84.1 Financial Associations 91.2 98.0 98.0 Services 81.5 76.7 91.4 Not well defined 61.0 86.0 96.0

Source : State Institute of Statistics.

Table 7 indicates that the literacy ratio is high for most of the economic branches. Among these branches of economic activity, financial activities, electricity, gas and water production have the highest literacy ratios. As can be expected, agriculture, mining and construction have lower ratios. The latter activities require less skilled labourers in the less developed countries. Taible 8 gives us the educational attainments of the labour force. When the 1960 and 1970 ratios are compared, a slight improvement can be observed in 1970. But ratios above the primary school level are very low and reflect the poor level of the education on the whole for the labour force. 121 CO

I ffiMO« CO Oi

I CI} O) Iß t> 01 00 O

03 Ö O •"§ o •g H

122 THE LABOUR FORCE

REFERENCES

1. Cillov, Haluk, Turkey's Population (in Turkish), Istan- bul; 1962. 2. Ekin, Nusret, Labour Force and Economic Growth (in Turkish), Istanbul, 1968. 3. Ekin, Nusret, Unemployment in Developing Countries and in Turkey (in Turkish), Istanbul, 1972. 4. Gürtan, Kenan, Methods of Demographic Analysis (in Turkish), Istanbul, 1968. 5. State Institute of Statistics : Population Censuses (1950- 1970). 6. State Planning Organization : First Five Year Plan (1963- 1967). State Planning Organization : Second Five Year Plan (1968- 1972). State Planning Organization : New Strategy and Develop- ment Plan, Third Five Year (1973-1977), (in Turkish). 7. Türkay, Orhan, Disguised Unemployment (in Turkish), Ankara, 1968. 8. Tuncer, Baran, The Impact of Population Growth on the Turkish Economy, Hacettepe University Publication No. 3, Ankara, 1969.

123

Chapter VI

POPULATION PROJECTIONS

by Samiia YENER, M. A.

Chapter VI POPULATION PROJECTIONS

Samira Yemer, M. A. State Planning Organization

Introduction The role of population projections within the economic and social planning context is two-fold. In view of the fact that the population, by determining the labour supply, is an essential factor in the production of all goods and services, and by defining the number of consumers, it describes the needs of the nation. However, one should bear in mind that a population projection is not a prediction of future trends but an illustration of the growth patterns Which a population would experience under sets of assumptions about the demographic situation. How close the predic- tions to the actual population comes, depends on several factors such as : (a) the accuracy of the base data and the vital rates, (b) the accuracy of the actual population estimated through the future census, (c) the accuracy in estimating the patterns of migration, mortality and fertility. Thus, the main aim of the projections is to provide a range of assumptions concerning the future and indicating the most possible one among them. In this section, consideration will be given primarily to population projections for Turkey on the whole, as covering the period from 1970 to 1980 (the medium range) and to 2000 (the long range), by five year intervals.

127 THE POPULATION OF TURKEY

For the preparation of the Third Five Year Plan, six different sets were developed by the State Planning Organization to study the range of possibilities for the future trend of the population, four of which will be presented below. The brief description of the model that is used for the projections is as follows : (a) The component, method of projection was utilized with a given starting age distribution and the possible future improvements in the selected survivorship rates among the East Model levels \ as well as the possible change in the shape as well as the level of fertility schedule. . (b) The census results, were corrected for misreporting of ages. But, as it is not possible to detect the omission errors in enumeration due to the lack of a post enumeration survey, the 1 per cent sample is taken as face value. The previous census show that the sampling errors were about 01 %. Therefore no correction was made for sampling errors. (c) For all the assumptions the improvements in mortality have the same trend which enable an increase in life expectation at birth to 73 years for females and 67 years for males in the year 2000. (d) With regard to the fertility assumptions, three different sets were prepared : (1) High fertility (both shape and the level of the fertility schedule is assumed to be constant). (2) Low fertility (starting with 1970, both shape and level of the fertility will change according to the target set for 1995). The prevailing fertility level (2.95) and the shape of the fertility schedule in Istanbul at present is taken as a target to be reached by the whole population for 1995. According to the recent studies, during the 1965-70 period, the total fertility level is estimated to be

1 Coale Ansley, Paul Demeny "Regional Model Life Tables and Stable Populations", Princeton, Princeton University Press, 1966. 128 ; POPULATION

5.78| for the total population K (3) Medium fertility. The decline will not be as sharp as in (2) starting in 1970, but with a gradual decline of 5 per cent in level and shape up to the year 1985, it will then reach the target in 1995. This is what to expect. A drastic decline can not be started in 1970, because the family planning programs are not yet fully implemented and the socio-economic change which forms the basis of the acceptance of such programs Will be bearing their fruits ten years from now. In addition, the change in age structure after 15 years will strengthen the decline in fertility by slowing down the growth of child bearing ages. In two of the three alternatives the emmigration assumption was not introduced. With the fourth alternative the possible emmigration of one million people till 1995 was given consideration. The results of four sets of projections are presented on Table 1 a, b, c, d, together with the changes in the perspective vital rates and expectation of Life and GRRS. I. Labour Force Projections The labour force is generally defined as that part of the population which contributes to the supply of labour available for the production of economic goods and services during a specified, period including employers, self-employed persons, and those who assist without pay in a family economic enterprise, as well as employees. It also includes unemployed persons, who are not at work and aré seeking work and profit. Lalbour force projections can be categorized into two main groups as-estimation of labour supply, and labour demand. Labour supply projections are in fact the product of two separate projections, a projection of total population by sex and age and a projection of the activity rates for each group considered. Non-agricultural labour supply estimates : As the readability of the statistics on inactive population is somewhat inaccurate, it has been difficult to estimate the trend of this population for the future. Therefore, another method was utilized for projecting the supply of non-agricultural labour force. 1 Yener, Samlra "III. Bes Yillik Niifus Projeksionlan lçln kullamlan Yöntem", DPT: 1300-SPD : 254, Ankara 1973. ': 129 THE POPULATION OF TURKEY.

The brief description of this method is as follows : the open unemployment figures obtained on the basis of household surveys ', together with the employment figures derived from the population censuses, yielded the current labour supply figures. These in turn lead to more reliable estimates of the activity rates. By taking into account the past trend and the projected changes in the economy's manpower needs and the growth of the urban population, the future activity rates of the non-agricultural population are estimated. (Table 2.) As for estimating the agricultural labour supply, it has been assumed that until the year of 1995, the current activity rates will be maintained.

Table 2 Trends in Future Activity Rates

1972 1977 1987 1995

Rural 38 38 38 38 Urban 38 38 39 41 , , Total population 38 38 39 40 • . -i

Source : Third Five Year Development Plan.

By using the total urban and rural population estimates an

1 Household Surveys on Labour Force art being conducted by th» Stat« InitituU of

130 POPULATION PROJECTIONS

b) Agricultural employment estimates. For the short run, the employment estimates are derived with the method which takes into account the number of workers required to attain given levels of output and productivity, through the usage of average hour iworked per day at the peak season. For the long term projections, e.g. for the year 1995, the target of eliminating disguised unemployment till 1995 was emphasized in the Third Five Year Plan. Accordingly, the level of employment for that year and the years preceding it will be exactly that appropriate to production requirement and to the maximum labour productivity which production techniques will permit. Table 3 Projections of Labour Supply and Labour Demand (millions) Civilian Labour Force 1972 1977 1987 1995

(15-64 age group)1 Civilian Labour Supply 14.3 16.0 21.0 26.0 Agricultural 8.8 8.6 8.0 6.2 . Non-agricultural 5.5 7.4 13.0 19.8 Labour Demand 13.5 14.9 19.0 25.2 Agricultural Employment 8.8 8.6 8.0 6.2 Labour Demand in most active season 7.9 7.9 7.5 6.2 Disguised Unemployment 0.9 0.7 0.5 — Non-Agricultural Employment 4.7 6.3 11.0 19.0 Industry 1.5 2.1 3.4 5.6 Services 3.2 4.2 7.6 13.4 Non-Agricultural unemployment 0.8 1.1 2.0 0.8 1 Hounding «TO» should b» consldared. H. Regional Population Projections Provincial estimates ibased on the crude extrapolation techniques were made. However, the anaJLysia of 1970 censuses are required for refined provincial1 estimates, the sex and age distribution and migration estimates which will be obtained after the results are completed.

131 s

•s 2 t Cl

•a o o •a

a o 'S) S

132 POPULATION PROJECTIONS in. Urban and Rural Population Projections

According to the targets set within the Third Five Year Plan, a functional and gradual pattern of urbanization comparising 75 percent of the population will be realized in 1995. The rural^urban population projections of both medium and long range were made on the basis of the. above mentioned target, and the following distribution of the population among these two main groups were maintained :

; Tableo Rural and Urban Distribution of the Population

1972 1977 1987 1995

Urban as 5% of total 38 47 62 75

Rural as Jio Of total 62 53 38 25

Source : HI. Uve Year Plan.

IV. Household Projections

Projections of numbers of household are useful in economic and social planning in view of the fact that there' exists a variety of services for which the households are the units of consumption. The method of household projections which are included in the Third Five Year Plan, consists of utilizing the ratio of the number of households to the total population. It is estimated that with the urbanization and industrialization, there will be a further decline in this ratio which follows a reduction of one percent per year. This reduction is assumed to be the same for both rural and urban areas. Table 6 shows how household projections extend till the year 1995 on the basis of the above mentioned methodology.

133 THE POPULATION OF TUBKBY

Table 6 Household Projections !

Urban Rural No. of Average No. of Average Size Households Size of Households Year of Household (thousands) Household '(thousands)" 1972 5.09 2,806 5.83 3,988 1973 5.08 2,990 5.82 3,966 1974 5.07 3,192 5.81 3,987 1975 5.06 3,405 5.80 3,997 1976 5.05 3,636 5.79 3,992 1977 5.04 3,880 5.78 3,893 1978 5.01 4,630 5.75 3,997 1985 4.96 6,330 5.70 3,775 1995 4.86 10,000 5.60 2,920

Source : III. Five Year Development Plan, pp 836.

V. Projections of School Population

These types of projections can be categorized into two components. The first is estimation of the future school age population: which is a section of total population projections. The second is the estimation of enrollment or scholarization ratios. The future enrollment ratios are presented on Table 7. While setting the targets for the secondary education, the enrollment ratios in future years are assumed to be compatible with the future economic and social deveJiopment taken aa targets to be reached for the years 1977 and 1995. In other words, the scholarization rates of 1995 will be in accordance with tüie social and economic conditions of 1995. For the higher education, the scholarization target» of technical education were developed on the basis of manpower projections, whereas for the other sections of higher education, the current trend is assumed to prevail, : POPULATION PROJECTIONS

Table 7 Targets for Scholarization

1972 1977 1995 Primary education 1st cycle 84 100 100 ; 2nd cycle 30.4 50 75 Secondary education 18 25 45. Higher education 6.8 9 15

Source- State Planning Organization

Table 8 Fiiturfe School Population (Primary and Secondary Education) 1973 -1996 (in thousand population)

Primary "Education" Lycee (Secondary Ed.)! First Cycle Second Cycle School Scho. School Scho. School Scho. Years Pop. Rate Pop. Rate Pop. Rate1

1973—74 5,193 90 1,170 44.3 326 13.2 1974—75 5,408 92 1,208 44.9 338 13.4 1975—76 5,689 95 1,261 46.0 345 13.4 1976—77 5,928 97 1,343 48.1 352 13.4 1977—78 6,242 100 1,441 50.7 359 13.4 1982—83 7,214 100 1,789 57.0 444 15.1 1987—88 8,166 100 2,259 46.0 516 15.8 1992—93 9,520 100 2,927 71.0 595 15.9 1995—96 9,859 100 3,454 75.2 645 15.7

Source : Third Five Y0«r Development Plan, PMe 743. TBE POPULATION OF TURKEY

REFERENCES

Alpay, Aysel, "Age Heaping Problem in the Turkish Censuses 1935 to 1965", Istatistik, Turk îstatistik Dernegi Dergisi Cilt 1, Sayi 3. Brass, William, et al, The Demography of Tropical Africa, New Jersey, Princeton University Press, 1968. Brass, WiHimi, "Uses of Census and Survey Data for Estimation of Vital Rates", U. N., African Seminar on Vital Statistics, Addis Ababa, 1964. Coate, Ansley, Paul Demeny, Regional Model Life Tables and Stable Populations, Princeton, Princeton University Press, 1966. Hume, Ian, "Migrant Workers in Western Europe", International Development Bank Report, May 1971. Republic of Turkey Prime Ministry, State Institute of Statistics, 25 October 1970 Census of Population 1 percent Results ' (mimeographed 28 pages). Republic of Turkey, Prime Ministry, State Institute of Statistics, 1965 Census of Population, Social and Economic Characteristics of the Population, S. I. S. Printing Division, Ankara, 1969. Republic of Turkey, Prime Ministrp, State Planning Organization, Third Five Year Development Plan 1973-1977, DPT 1272, PRIME MINISTRY Printing Office, Ankara 1973. Republic of Turkey, Ministry of Health and Social Welfare, School of Public Health, Vital Statistics from the Turkish Demographic Survey 1966-67. Shorter, Frederic and Demieny, Paul, Türkiye'de ölüm Se- viyesi, Dogurganhk ve Ya§ Yapisi Tahmini, Istanbul1 University Yayim No. 1306, 1968. U. N. Manual IV, Method of Estimating Basic Demographic Measures from Incomplete Data, New York, 1967. Yener, Samara, Method of Population Projections Utilized in the Third Five Year Plan, DPT : 1300, SPD : 254, Ankara, 1973. Yener, Samara, Population Projections. SPO (Thesis) October, 1969 (Mimeographed 25 pages). Table 1. a High Fertility Population Projection (Thousand)

FEMALES

AGE 1970 1975 1980 1985 1990 1995 2000

0 2740.3 3198.8 3902.4 4674.21 5513.6 6492.0 7729.6 5 2366.3 2641.8 3111.3 3827.1 4616.8 5475.4 6468.7 10 2144.4 2352.1 2629.0 3099.7 3815.9 4606.7 5466.7 15 1924.1 2131.4 2340.1 2618.1 3089.5 ' 3806.4 : 4598.0 20 1341.5 1907.6 • 2115.8 2325.9 2605.6 3078.1 3795.7 25 1090.6 1327.1 1890.1 2099.7 2311.8 2953.2 ' 3067.0 30 1960.7 1076.3 1312.9 1873.2 2084.4 2298.4 2581.6 35 1070.1 1044.8 1062.3 1298.3 1855.8 2068.6 ' 2284.7 40 862.4 1050.4 1027.8 1047.2 1282.4 1836.6 2051.1 45 685.6 841.8 1027.6 1007.7 1029.0 1262.9 1813.0 50 487.7 662.7 815.8 998.5 981.7 1005.4 1237.8 55 559.7 463.9 632.6 781.4 959.9 947.7 974.7 60 449.9 517.6 431.2 591.0 734.0 907.1 901.3 65 392.2 395.6 458.5 384.7 531.6 666.3 831.5 70 231.3 315.8 321,8 376.7 319.8 447.7 ' 569.1 75 + 199.4 250.8 337.8 388.1 :459.1 461.5 569.8 IOT 7605.6 20178.4 23417.1 27391.4 32191.1 37954.2 14940.2

MALES

0 2766.7 3282.7 4011.8 4814.2 5686.2 6703.8 7999.6 5 2441.2 2663.5 3185.3 3921.8 4739.6 5630.3 6664.0 10 2215.6 2426.8 • 2649.7 3170.8 . 3906.1 . 4723.1 5613.4 15 1896.8 2200.0 2411.1 2634.0 3153.8 3887.4 4703.0 20 1504.8 1875.7 •2177.3 2388.4 2611.5 3129.7 3860.9 25 1177.1 1484.9 • 1852.8 2153.0 2364.2 2587.8 3104.3 30 1002.3 1160.7 • 1465.8 1830.9 2129.8 2341.3 , 2565.4 35 ' 1124.7 985.9 1143.1 1445.5 . • . 1807.8 2105.5 2317.4 40 926.0 1100.5 966.2 . 1122.0 1420.9 . 1779.S 2076.1 45 709.5 897.4 1068.5 939.9 1093.5 1387.4 1741.4 50 485.4 676.2 857.1 1022.8 901.7 1051.6 . 1337.8 55 603.0 450.8 629.6 . 800.1 957.4 846.5 . 990.6 60 465.5 •539.2 • 404.4 566.6 722.5 • 868.0 .. 770.9 65 377.6 392.9 457.0 344.2 484.6 621.3 750.9 70 222.0 290.0 303.4 355.0 269.2 . 381.8 , 493.7 75 + ' 142.8 207.0 281.5 325.2 382.8 354.1 , 423.8 TOT 18061.0 20634.3 23864.5 27834.2 . 32631.4 38399.2 ..45413.3

TOTAL 35666.6 . 40812.7 47281.6 55225.6 , 64822.5 76353.4 90353.4

BIRTH RATE - 39.3 40.6 40.8 40.3 39.5 39.2 DEATH RATE 12.3 11.1 9.8 8.2 6.8 5.6 NATURAL INC 27.0 2914 31.1 32.0 32.7 : 33.7

MIGRATION 0.0 . .0.0 0.0 . 0.0 0.0 0.0 POP INCREASE 27.0 29.4 31.1 32.0 32.7 33.7

LIFE EXPECTANCY E (0) FEMALES 59.4 62.3 65.0 67.8 70.4 72.9 É (0) MALES 55.8 58.2 60.5 62.8 65.0 67.0 FERTILITY GRR. •2.82 2.82 2.82 2.82 2.82. 2.82

I Table 1. b Medium Fertility Population Projection (Thousand)

FEMALES

AGE 1970 1975 1980 1985 1990 1995 2000

0 2740.3 3038.3 3524.3 4062.5 4081.6 3903.2 5 2366.3 2641.8 2955.2 3456.3 4012.7 4053.3 10 2144.4 2352.1 2629.0 2944.1 3445.2 4003.8 15 1924.1 2131.4 2340.1 2618.1 2934.5 3437.6 20 1341.5 1907.6 2115.8 2325.9 2605.6 2923.7 25 1090.0 1327.1 1890.1 2099.7 2311.8 2593.2 30 1060.7 1076.3 1312.9 1873.2 2084.4 2298.4 35 1070.1 1044.8 1062.3 1298.3 1855.8 2068.6 40 862.4 1050.4 1027.8 1047.2 1262.4 1836.6 45 685.6 841.8 1027.6 1007.7 1029.0 1262.9 50 487.7 662.7 815.8 998.5 981.7 1005.4 55 559.7 463.9 632.6 781.4 959.9 947.7 60 449.9 517.6 431.2 591.0 734.0 907.1 65 392.2 395.S 458.5 384.7 531.6 666.3 70 231.3 315.8 321.8 376.7 319.8 447.7 75 + 199.4 250.8 337.8 388.1 459.1 461.5 TOT 17605.6 20017.9 22882.9 26253.4 29630.2 32817.3

MALES

0 2766.7 3118.0 3623.1 4184.2 4209.4 4030.5 5 2441.2 2663.5 3025.4 3541.9 4119.4 4168.0 10 2215.6 2426.8 2649.7 3011.7 3527.6 4105.0 15 1896.8 2200.0 2411.1 2634.0 2995.6 3510.8 20 1504.8 1875.7 2177.3 2388.4 2611.5 2972.7 25 1177.1 1484.9 1852.8 2153.0 2364.2 2587.8 30 1002.3 1160.7 1465.8 1830.9 2129.8 2341.3 35 1124.7 985.9 1143.1 1445.5 % 1807.8 2105.5 40 926.0 1100.5 966.2 1122.0 1420.9 1779.8 45 709.5 897.4 1068.5 939.9 1093.5 1387.4 50 485.4 676.2 857.1 1022.8 901.7 1051.6 55 603 0 450.8 629.6 800.1 957.4 846.5 60 465.5 539.2 404.4 566.6 722.5 868.0 65 377.6 392.9 457.0 344.2 484.6 621.3 70 222 0 290.0 303.4 355.0 269.2 381.8 75 + 142.8 207.0 2S1.5 325.2 382.8 354.1 TOT 18061.0 20469.6 23316.0 26665.1 29997.7 33112.0

TOTAL 35666.6 40487.5 . 46198.9 52918.6 59628.0 65929.3 1

BIRTH RATE 37.5 37.2 36.7 31.8 26.7 21.7 DEATH RATE 12.1 10.8 9.5 7.9 6.6 5.7 NATURAL INC 25.4 26.4 27.2 23.9 20.1 16.1

LIFE EXPECTANCY E (0) FEMALES 59.4 62.3 65.0 67.8 70.4 72.9 E (0) MALES 55.8 58.2 60.5 62.8 65.0 67.0

FERTILITY GRR. 2.S8 2.55 2.42 2.04 1.67 1.29 n Table 1. c Low Fertility Population Projection (Thousand)

FEMALES

AGE 1970 1975 1980 1985 1990 1995 2000

0 2740.3 2905.5 3234.6 3040.2 3586.5 3565.4 3378.2 5 2366.3 2641.8 2826.1 3172.2 3405.9 3561.6 3552.6 10 2144.4 2352.1 2629.0 2815.4 3162.9 3398.4 3555.9 15 1924.1 2131.4 2340.1 2618.1 2806.2 3155.0 3392.0 20 1341.5 1907.6 2115.8 2325.9 2605.6 2795.9 3146.1 25 1090.0 1327.1 1890.1 2099.7 2311.8 2593.2 2765.7 30 1060.7 1076.3 1312.9 1873.2 2084.4 2298.4 2581.6 35 1071.1 1044.8 1062.3 1298.3 1855.8 2068.6 2284.7 40 862.4 1050.4 1027.8 1047.2 1282.4 1836.6 2051.1 45 685.6 841.8 1027.6 1007.7 1029.0 1262.9 1813.0 50 487.7 662.7 815.8 998.5 981.7 1005.4 1237.8 55 559.7 463.9 632.6 781.4 959.9 947.7 974.7 60 449.9 517.6 431.2 591.0 734.0 907.1 901.3 65 392.2 395.6 458.5 384.7 531.6 666.3 831.5 70 231.3 315.8 321.8 376.7 319.8 447.7 569.1 75 + 199.4 250.8 337.8 388.1 459.1 461.5 569.8 TOT 17605.6 19885.1 22464.0 25226.3 28116.7 30971.9 33625.2

MALES

a 2766.7 2981.7 3325.3 3551.5 3698.7 3681.7 3496.2 5 2441.2 2663.5 2893.2 3250.7 3496.5 3662.3 3659.9 10 2215.6 2426.8 2649.7 2880.0 3237.6 3484.3 3651.4 . 15 1896.8 2200.0 2411.1 2634.0 2864.6 3222.2 3469.5 20 1504.8 1875.7 2177.3 2388.4 2611.5 2842.7 3200.2 25 1177.1 1484.9 1852.8 2153.0 2364.2 2587.8 2819.7 30 1002.3 1160.7 1465.8 1830.9 2129.8 2341.2 2565.4 35 1124.7 985.9 1143.1 1445.5 1807.8 2105.5 2317.4 40 926.0 1100.5 966.2 1122.0 1420.9 1779.8 2076.1 45 709.5 897.4 1068.5 939.9 1093.5 1387.4 1741.4 50 485.4 676.2 857.1 1022.8 901.7 1051.6 1337.8 . 55 603.0 450.8 629.6 800.1 957.4 846.5 990.6 60 465.5 539.2 404.4 566.6 722.5 868.0 770.9 65 377.6 392.9 457.0 344.2 484.6 621.3 750.9 70 222.0 290.0 303.4 355.0 269.2 381.8 493.7 75 + 142.8 207.0 281.5 325.2 382.8 354.1 423.8 TOT 18061.0 20333.3 22885.9 25609.6 28443.2 31218.2 33764.8

TOTAL 35666.6 40218.4 45349.8 50835.9 56559.9 62190.1 67390.0

BIRTH HATE 36.0 34.6 32.1 29.3 25.8 22.1 DEATH RATE 11.9 10.6 9.3 7.9 6.8 6.0 NATURAL INC 24.0 24.0 22.8 21.3 19.0 16.1

MIGRATION 0.0 0.0 0.0 0.0 0.0 0.0 POP INCREASE 24.0 24.0 22.8 21.3 19.0 16.1

LIFE EXPECTANCY E (0) FEMALBS 59.4 62.3 65.0 67.8 70.4 72.9 E (0) MALES . 55.8 58.2 60.5 62.8 65.0 67.0 FERTILITY GRR. 2.56 2.31 2.05 1.80 1.55 1.29 ra Table l.d Medium Fertility + Migration Population Projection (Thousand)

FEMALES

AGE 1970 1975 1980 1985 1990 1995 2000

• 0 27-40.3 3007.1 3494.9 3983.2 4021.8 3854.2 .5 2366.3 2627.6 2910.5 3427.5 3934.4 3993.9 10 2144.4 2339.2 2601.9 2899.6 3417.5 3925.7 15 1924.1 2123.4 2319.0 2591.1 2890.1 3408.9 20 1341.5 1889.4 2089.4 2305.0 2578.7 2879.4 25 1090.0 1298.0 1843.0 . 2073.6 2291.0 2566.5 30 1060.7 1052.5 1264.2 1826.5 2058.4 2277.7 35 1070.1 1026.4 1020.4 1246.3 1809.6 2042.9 40 862.4 1036.0 995.3 1005.9 1230.9 1790.8 45 . 685.6 834.4 1006.1 975.8 988.4 1212.3 50 487.7 661.7 807.7 977.6 950.7 965.8 55 559.7 463.9 631.7 773.6 939.9 917.8 60 • 449.9 517.6 431.2 590.1 726.7 888.2 65 392.2 395.6 458.5 384.7 530.8 659.6 70 231.3 315.8 321.8 376.7 319.8 447.0 75 + 199.4 250.8 337.8 388.1 459.1 461.5 TOT 17605.6 49839.0 22529.4 25825.1 29147.7 32292.2

MALES

0 2766.7 3085.9 3592.9 4102.5 4147.7 3979.9 5 2441.2 2668.5 2979.2 3512.3 4039.0 4106.9 10 2215.6 2472.6 2620.6 2965.7 3498.2 4024.9 ,15 1896.8 2192.7 2389.7 2605.1 2949.9 3481.5 20 1504.8 1867.0 2161.5 2367.2 2582.9 2927.3 25 1Í77.1 1432.5 1790.8 2137.3 2343.2 2559.3 30 1002.3 1069.8 1322.2 1769.6 2114.3 2320.6 35 1124.7 913.5 981.2 1303.8 1747.3 2090.2 40 926.0 1057.3 852.0 963.1 1281.7 1720.2 45 709.5 875.0 1004.1 820.7 938.6 1251.5 50 435.4 870.1 • 829.6 961.1 795.0 • 902.6 55 ' ' 603.0 450.8 623.9 774.4 • 899.6 746.4 60 465.5 539.2 404.4 56L4 699.4 815.7 65 377.6 ' 392.9 ' 457.0 344.2 480.2 601.4 70 222.0 290.0 303.4 355.0 269.2 378.4 75 + 242.8 207.0 281.5 225.2 382.8 354.1

: 22593.8 25876.7 29168.8 32260.7 ' TOT 18061.0 20104.0 • 45123.3 51701.9 58316.5 64552.9 TOTAL 35666.6 39942.9 21.9 BIRTH RATE 37.'4 37.7 36.8 32.0 27.0 5.7- DEATH RATE 12.1 11.0 9.6 8:0 6.3 16.2 NATURAL INC 25.3 26.7 27.2 24:1 20.3 0.0 MIGRATION —2.6 —2.4 0.0 0.0 0.0 16.2 POP' INCREASE' 22.6 24.3 ' 27.2 24.1 20.3

LIPS EXPECTANCY 72.9 E' (0) FEMALES 59.4 62.3 65.0 67.3 70.4 67.0 E (0) MALES ; 55.8 58.2 60.5 62.8 65.0 1.29 FERTILITY GRR. 2.68 2.55 2.42 2.04 1.67 rv Chapter VU

ECONOMIC AND SOCIAL IMPLICATIONS OF POPULATION GROWTH AND POPULATION POLICY

by Hüsnü A. KÎSNÎSÇÎ, M. D.

ECONOMIC AND SOCIAL IMPLICATIONS OF POPULATION GROWTH AND POPULATION POLICY

Hüsnü A. Kisn%i, M. D. Institute of Population Studies Hacettepe University

Especially after World War U, many countries came face to face with the problem] of a high rate of population growth, such as had never previously been experienced. The problem is being felt more seriously in less-developed countries in which the annual per capita income is already very low. In the post-war years, due to the improvement of health «HT ditíona, there was a large decrease in mortatiitp rates. On the other hand, birth rates did not change considerably when compared to those of the pre-war years. This situation is the main cause for the high' rate of population growth Which the world is facing at the present. Lately, the high' rate of population growth' has also drawn the attention of the Turkish public, like it has in many other less developed countries. According to economists, this high rate in Tur- key may, in time, present a serious threat to this country. Economist have been studying the economic consequences of population growth for a long time and were primarily interested in the size of the population. Lately, this approach' has been replaced by stressing the importance of the composition of the population. Today, it is stressed that the age distribution of the population is one of the most important factors which affect the economy of a country. In countries where the population growth rate is higher than in others, the ratio of the age group below 15 is considerably higher in proportion to the total population. Therefore, a large part of economic resources are necessarily spent for consum.ptk>n purposes, which in turn causes a decrease in the amount to be spent for investments. ,;,.!• ¡ , ; • . ; 139 THE POPULATION OF TURKEY

In this article, after studying the social and economical consequences of the high population growth rate in Turkey, we will, consider the relationship between economic, social and population policies. Economic Development in Turkey Even though income per capita in Turkey seems to be somewhat higher compared to other less developed countries, when the economic structure of the country is examined closely it presents itself as an underdeveloped economy. IT IS ACCCEPTED IN GENERAL THAT THERE ARE MANY indicators which show a country's economic development. But for comparative purposes between countries and for showing changes through time in the same country, per capita income seems to be most widely used indicator, despite the difficulties involved in using it. High rates of population growth In Turkey seem to have negative effects on per capita income. This shows itself in a study, taking 1961 prices constant and taking the base year of 1950 GNP as 100.0. In 1955, GNP appears as 136.0 while per capita GNP is 118.2. In 1965 these indices were found to be 215.5 and 141.0 respectively. This negative effect will appear more clearly when per capita income projections are carried out, as can be seen from Table 1. .

Table 1 Projections of GNP and Per Capita GNP Per Capita GNP (TL.) 2 % Population 2 .6 % Population Years GNP* Growth Growth 1965 73,127.1 2,329 2,329 1972 119,363.1 3,310 3,177 1992 461,897.2 8,622 7,359 2000 793,625.3 12,643 10,297 * GNP is assumed to increase by 7 % throughout the period Source: B. Tuncer; The Impact of Population. Growth, op. thç Turkish Ifconomy, H. U. Publi- cations No. 3. 1968. p. 57 HO ECONOMIC AND SOCIAL IMPLICATIONS

If the present level of population growth continues to increase at the same rate, in the year 2000 per capita income will be 2,346 TL. less than what it would have been if this rate had decreased to 2 %. As can also be seen from Table 1, this figures is slightly more than per capita income in 1965. In other words, if population growth could be slowed down from the existing rate of 2.6 % per year to 2 % per year, capita income diferencia in the year 2000 will be

23 %•'•• •• i '•/'• . • . -.....::-. i

Income Distribution

As there have been few studies carried out on income distribution in Turkey, it is difficult to say much on this topic. Nevertheless the available studies show that there is an inverse relation between the income level and the birth rate. If the birth rate increases while the income level is decreasing, it will firstly increase the proportion of low income groups in the total population having an adverse effect on the growth of a country's economy. Again when the birth rate stays high in this group their share of the income, in proportion to the total income, will decrease in time. Of course, a high rate is not the only cause for the changes in unequal income distribution, but its effect on the low income groups cannot entirely be ignored. In a study carried out by Bulutay et al, it has ibeen pointed out that in Turkey the average household income after taxes is 11,077 TL. The same study shows that the annual income of 42 % of the families is below 5,000 TL., 48.6 % earned between 5,000 - 25,000, and only 9.2 % earned more than 25,000 TL. The situation becomes more interesting when we consider individual income distribution after tax. In 1968 average individual income after tax is 1,805 TL. Of these, 93.5 % earned less than 5,000 TL., 5.7 % earned 2,500 - 5,000 TL., and only 0.8 % had an average income of over 25,000 TL. Table 2 shows this distribution more clearly.

141 THE POPULATION OF TURKEY.

Table 2 Percentage of Families and Their Share of Income; From Low to High Income Groups

20 4.5 3.0 20 4.5 3.0 20 8.5 7.0 40 13.0 10.0 20 ' 11.5 10.0 60 24.5 20.0 20 18.5 20.0 80 43.0 40.0 20 57.0 60.0 100 100.0 100.0

Source Tuncer Bulutay et al.,: Income Distribution in Turkey 1968. A. U. 8BF Publication«, No. 325. 6evinc Matbaasi, Ankara 1971. P. 16.

Inspite of all the socio-economic measurements taken in Turkey, the income distribution between the years 1963-1968 showed * negative trend. As we can also see from Table 2. the share of the low income groups decreased considerably iwhile the share of the high income groups increased.

Demographic Investment

As a consequence of a high rate of population growth, there alao appears another important economic problem! in less developed countries. ; While thé rate of population growth in Turkey between 1960- 1965 was 2.46 ,%, it increased to 2.56 % in the period 1965-1970. If we also consider the population working abroad, the rate of population growth reached 2.73% in the later period. ¡When iwe consider that this increase is closely related to the children born in poBt-War

142 ECONOMIC AND SOCIAL IMPLICATIONS

years, who are now reaching the fertile stage, the population will grow at least at this rate within a very short period. On the other hand, in the (state) plans, the rate of economic growth has been put forward as being 7 %. Also, fixed capital-income ratio has been planned as 3 % while the total investments consist of only the 21.% of the GNP during the period 1962-1972. That is to say, a growth rate of 2.6 % in the population causea the 7.7 !% of the share laid aside for investments from the GNP to be absorbed by the increasing population as demographic investments. Therefore, only 13.3 % are left for the economic investment» which are necessary for economic development.

Capital Requirements foe Employment Another socio-economic problem to be cosawdered in countries where there is a high population growth: rute, is a high supply of the labour force. This situation continuously produces a need for finding additional employment possibilities. The provision of additional employment possibilities is closely related to the opening of new Work areas and this, in turn, requires investment la less developed countries, per capita income is already; low, therefore saving possibilities are limited, which makes investments difficult. The above mentioned problem is more clearly determined when we project the effects of certain rates of population growth' on the amount of capital which wil absorb the labour force untUi the year 2000. Here, two basic assumptions have 'been considered. Firstly, the mcreasing rate of labour force supply will1 move to sectors other than agriculture (services and industries), until the year 20ÓO. Secondly, until1 this year, capital requirements per labourer for employment, on the average, will be 50,000 TL. The results of this estimation can be seen in Table 3. which also shows that the high rate of population increase has adverse effects on employment.

143 THE POPULATION OF TURKEY

Table 3 Capital Requirements for Employment (in million TLi.) Capital Requirements for Employment* Labour Supply 2 % Pop. .2.6 % Pop. 2 % Pop. 2.6 % Pop. Years Growth Growth Growth Growth 1967 7,625 9,975 305 399 1972 16,500 22,100 330 442 1992 26,500 39,950 520 799 2000 31,900 52,100 638 1,042

• Labour supply in given years is 79 % of the 15-64 age group. • The labour supply figures are multiplied by 25,000 TL. in the year 1967, and by ,50,000 up to 2,000.

Population Growth, Food Supply and Consumption At present, some of the less; developed countries experiencing a rapid population growth are faced with a food shortage, and in other countries, even though there is a certain amount of increase in food production, food shortage will inevitably present a problem to them. The procurement of a sufficient food supply for the people living in every country is an important problem and of deep concern to policy makers. Here we will study the pattern of food supply in Turkey in the past, present and future. No doubt, in order to study the food problem seriously, we need data on production and consumption levels of various goods, and food consumption patterns amongs various socio-economic groups in different regions of the country. Since at this state the carrying put of such a study does not seem possible, in this article only the production and the consumption of the basic food supplies vital to human life, such as wheat, meat and milk, have been selected as the three basic staples of food, The problem of food-shortage which comes together with a rapid population growth and which has become a subject of great 144 ECONOMIC AND SOCIAL IMPLICATIONS controversy recently, if studied carefully, does not seem to present a problem for Turkey at the present, or in the near future. The number one food item in terms of consmption is wheat; for example, in 1972, wheat consumption per person Was 193.4 kg. This figure was 141.1 kg. in 1950. Although, from 1950 onwards, the application of new techniques in agriculture resulted, im a considerable increase in the total production of goods, this did not reflect itself in the per person wheat consumption figures. One of the main reasons for this phenomenon is of course a rapid rate of population growth.

Table 4 Per Capita Wheat Production and Consumption Total Production Total Consumption Per Capita Years (tons) (tons) Consumption (kg.) 1950 3,871,900 3,058,810 146.1 1967 7,935,000 6,269,000 189.7 1972 9498,000 7,266,000 193.4 Note : It is assumed that 79 % of the total production is used for consumption purposes.

Table 5 Per Capita Wheat Consumption (Projections) Per Capita Consumption (kg.) 2 % Pop. 2.6 % Pop. Production Consumption Years Increase Incrase (tons) (tons) 1992 244.9 209.1 16,620,000 13,130,000 2000 264.8 215.7 21,041,000 16,622,000 Reta : It Is assumed that 79 % of total production is used for consumption purposes. Source : B. Tencer, the Impact of Population Growth on the Turkish Economy p. 58.

The other staple chosen as representative of food consumption is meat. Turkey is considered to have a great potential for animal husbandry but the level of meat production is relatively low and, consequently, consumption is also below the world standards. 145 POPULATION OF TURKEY

An exact figure for the total meat consumption in Turkey is difficult to estimate since only the quantity coming out of the slaughterhouses is known. Also, a large number of slaughterhouses are situated in areas where the population is above 2,000. Therefore it is assumed that the consumption pattern of areas with 2,000 people is representative of the areas where the population is below 2,000. Actually, in this study we are more concerned with the relationship between per capita meat consumption and the high rate of population growth. Therefore, the non-availability of exact meat consumption figures does not seem to create an important problem. Estimated meat production in. Turkey was 177,520 tons in 1950. It has been assumed that 95 % of production goes in to consumption. This shows us that for 1950, per capita consumption is 8.051 kg. Although total meat consumption increased 174 % between the period 1950-1967, per capita meat consumption increased only 74 %, Table 6. shows meat production, consumption, and per capita consumption figures for different years.

Table 6 Supply of Meat and Per Capita Meat Consumption Per Capita Production Consumption Consumption Years (in tons) (in tons) (kg.) 1950 177,520 168,644 8.051 1967 486,000 462,000 13.980 1972 629,000 597,000 15.890 Table 7. shows the amount! of meat that would be avaliable for consumption up to 2,000 according to two rates of population growth. There would be a 7,500 kg. difference in per capita meat production between the growth rates of 2 % and 2.6 %. This difference seems low, but when it is considered that in 1950 the per capita consumption estimation was only 8,051 kg. its importance becomes quite clear.

146 ECONOMIC AND SOCIAL IMPLICATIONS

Table 7 Supply of Meat and Per Capita Meat Consumption (Projections) Per Capita Consumption (kg.) Production Consumption 2 % Pop. 2.6 % Pop. Years (in tons) (in tons) Growth Growth 1992 1,765,000 1,677,000 31.300 26.720 2000 2,666,000 2,533,000 40.350 32.860 Nota : The production Is assumed to increase by 5.3 % annually. Source : B. Tuncer, the Impact oí Population Growth on the Turkish Economy p. 59. The third item chosen for examination in this study is milk. Again, when we have production, consumption, and per cepita production estimates of milk for different years and for different population increase rates, we can see the negative effect of a rapid population growth on this item. Supply of milk and per capita wilk consumption figures are given in Table 8. As can be seen from these figures, although supply of milk increased 25 % in 1972 compared to 1967, the figures for per capita milk consumption increase only reached 10 % for the same years. This situation can be related to a high rate of population growth. • Table 8 Supply of Milk and Per Capita Milk Consumption Production Per Capita Consumption Years (tons) (kg.) 1950 3,173,000 151.427 1967 3,234,000 97.900 1972 4,050,000 107.800 Note : It Is assumed that total production m used for consumption. In Table 9. the effects of a high rate of population growth can be observed more clearly. If we assume a 4.6 % average annual increase in milk production and a 2 % population increase, per capita milk consumption in 2.000 will be 163.80 kg. If the population increases by 2.6 % annually, per capita milk consumption will be only 144.000 kg. for the same year.

147 THE POPULATION OP TURKEY

Table 9 Supply of Milk and Per Capita Milk Consumption Per Capita Consumption (kg.) Production 2 % Population 2.6 % Population Years (in tons) Growth Growth 1992 9,952,000 185.800 158.500 2000 14,262,000 227.200 185.000 Nnte : It is assumed that production -will increase 4.6 % annually. Source : B. Tuncer, the Impact of Population Growth on the Turkish Economy p. 60. Education The manpower factor which is of 'great importance from the point of economic development, presents a problem in Turkey like it does in many other less developed countries. Although important developments have been accomplished in manpower in the planned period, there are still problems involved in the desired quality of manpower. In 1971, only 48 % of those who graduated from primary schools found possibilities of attending secondary schools, and the rest took their places in the production process without getting any kind of vocational training. Since public education for training manpower in Turkey is on a very low level, attendance in formal educational institutions has gained in importance. Therefore, studying the qualitative changes in the educational institutions seems to be the most important point for consideration in Turkey's educational level. In general, the following trends have (been observed in the 7-12 age group between the period 1950-1970: While in 1950 there were 3,183,200 people in this age group, this figure rose to 6,024,594 in 1970. That is to say, from 1950 to 1970, there has been a 89 % increase in this age group. But there was an increase of 210 % in the number of primary school students for the same time period. In the secondary education age group, 13-18, the observed increase is 1,957,348 persons which makes a total of 4,843,848, for the above mentioned period. This gives us a 68 % increase in the number of

148 ECONOMIC AND SOCIAL IMPLICATIONS secondary school children for the last 20 years. Consequently it can be said, that there its a great discrepancy between the level of increase in the school age group and the schooling ratio. In 1950, the schooling ratio in the primary school level was 50.8 %. This figure reached only 83.2% in 1970 while the rate of increase in the primary school students is 89 %. Also, the schooling ratio for the secondary schools which was 7.7 % reached only 27.1 % in 1970, while the increase in the number of students has been 491 % between 1950 and 1970. Qn the other hand, the number of primary school teachers in 1950 was 35,870, and the number of students per teacher was 45. In 1970, the number of teachers reached 132,721, but the teacher student ratio could be ony decreased to 38. The situation in the secondary education level also presents problems. There werte 15,458 teachers at this level in 1950. This figure reached 47,620 in 1970. Therefore, whereas there were 14 students per teacher in the previous year, the ratio reached 28 in the later period. We can also consider education from the point of expenditures involved in this area. In particular, the relation between different rates of population increase and per student expenditures until the year 2000 seems to give meaningful results. For this estimation, a 100 % schooling ratio for primary schools from 1972 to 2000 has been assumed. There were 578,000 students in the 7-12 age group in 1972, and per student expenditure was estimated as 338.2 TL. Accordingy, the total expenditure is calculated to ¡be 1,955.8 million TL. On the basis of a % 2 annual population growth, this age group will reach 7,836 thousand in 1992. During the year per student expenditure will be 610.2 TL., and the total expenditures will reach 4,781.5 million TL. A 2.6 % population growth rate will give us 8,778 thousand people who will require 5,356.8 million TL. in total. With a 2 % annual population growth rate for the year 2000, there will be 9,053 thousand students in the age group 7-12, which 149 THE POPULATION OF TURKEY will require 773.0 million TL. per student, and a total of 6,998.0 million TL. will have to be spent. If the annual population growth is 2.6 % this age group population will be 10,547 thousand with a requirement of 8,158.2 million TL. in total. In a country where rapid industrialization is a target, the need for a high level of educational investment creates obstacles for other economic investments. This situation is created because of a high rate of population growth. On the other hand, as mentioned above, a high rate of population growth also creates difficult social problems in adidition to creating economic problems.

Health The main reason for a slower population growth made prior to 1950, was a high death rate. However, since 1950, there havei been many improvements in Turkey's health conditions. Between 1953 - 63 per capita health expenditures rose 7 % annually, and the rate of increase since this period has been around 9.5 % annually. This unusual rate of increase is assumed to fall to 7 % until 2,000 and will constitute 7.5 % of the total health expenditures. Therefore, per capita health expenditures which have been 27.5 TL. in 1965, will reach 504.3 TL. It is also estimated that a 2.6 % population growth will require 38,866.9 million TL., and a 2 % growth will require 31,654.9 million TL. This shows us that only a 0.6 % difference in population growth causes a 7.212 million TL. difference in expenditures, and this, of course, is an obstacle in Turkey's economic development. We can also consider the health problem from another angle : In 1950 there was a total of 3,020 doctors in Turkey. This figure reached 13,843 which constitutes an increase of 358 %. On the otheU hand, while there was one doctor to 6,933 persons, this figure decreased to 2,577 in 1970. Also, the number of beds which was 18,837 in 1950, reached 72,034, but the number of beds per 100,000 persons has only increased from 89.9 to 103.3. 150 ECONOMIC AND SOCIAL IMPLICATIONS

Urbanization One of the most important issues Turkey is facing at prer sent is the high rate of rural to urban migration which has speeded up especially during the recent years. Various social and economic reasons have been stated for this event. According to one view, the main reason for migration is the "pull" of the newly developing industrial centers in the urban areas. Howover, in many studies carried out in Turkey, it is stated that as a result of a high birth rate the villages become too crowded in relation to their relatively fixed resources. Consequently, this view stresses the "push" of the rural areas as the main reason for internal migration. The urban population grew considerably faster after 1950. That year only 18.7 % of the total population lived in urban areas, but this ratio reached 26.3 % by 1960. The rate of increase in the city population was about 5.0 % per annum between 1960-65, 6.0 % in 1970¿ and it reached 35.9 % of the total population in 1970. According to estimated figures, the rate of urbanization will decrease in the near1 future and will fall to 3.0 % until1 the year 2000. This, will give us a ratio of 60 % living in urban areas. Another important point which1 should be stressed, is the quality of urbanization in Turkey. Urbanization in Turkey does not occur because of industrialization or the development of real urban functions. The process is taking place in the form of accumulation of population (demographic urbanization) in certain areas. The movement from' rural to urban areas is directed especially to cities of more than 100,000. In the planned period, the population movement has been mostly to centers above ;50,000. At present, more than one third of the total urban population lives in areas over 500,000. Although urbanization can be considered to be a natural event during transitional periods, and is also considered as giving an im- petus to development, an increase in the urban population which reaches one million per annum brings with it important social problems, • 15» THE POPULATION OF TURKEY

Aliso, in countries where per capita income is already low and where the pace of economic growth is slower than desired, accumulation of population in cities prevents the implementation of social and,economic measurements necessary for the development of the urban areas.

Housing

Meeting the housing requirements of people in every society is an important social and economic issue. Thus housing becomes one of the most serious problems involved in Turkey's urbanization process. As there are no dwelling censuses carried out in Turkey, we cannot give a dynamic picture in this area. Also, for penetrating studies we need the socio-economic levels of the groups who have housing requirements. Therefore, when dealing with the housing requirements, we will consider areas over 10,000 population. This does not mean that the problem is non-existent for areas with a smaller population, but our assumption is that the problem is more serious in the larger settlements. When it is estimated that the rate of urbanization will be 6.0 to 6.5 % in the following years anuually the housing problem gains in importance. In 1963 approximately 23.8 % and in 1971 about 21.7 % of the total investment was spent for housing. In order to determine the number of dwellings necessary for the needs of an increasing population, we must know the number of those who will live in cities in the future, as well as the average size of the families. It is estimated that there will be one tenth of 1 % decline in the family size by 1980, and 2 tenths of 1 % by 2000. This would make the family size about 4.5 persons in 2000 as against 5.0 in 1965 and 4.9 in 1972. If there is a 2 % population increase by 2000, the number of housing requirements for the same year will be 8,495 thousand in areas with more than 10,000 population. If the annual population increase is 2.6 %, this figure reaches 10,432. Another problem in a high rate of rural to urban migration ia the development of "gecekondus" or squatter houses. Since the 15? ECONOMIC AND SOCIAL IMPLICATIONS relatively poorer people who move into cities are not able to provide themselves with decent dwellings the suburb areas of the cities have become the site of "squatter" houses, lacking sewage, water, electricity and gas. According to calculations, there were 450,000 squatter houses in 1967 which constitutes 21 % of the housing units. The "gece- kondu" problem will continue to exist for some time, unless, standard housing units are built at the required speed. This requirement has, ofcourse, social1 as well as economic consequences.

Population Policy as an Element in Social and Economic Policy During certain periods in Turkey pronatalist or antinatalist policies have been preveiant depending on the socio-economic factors of the country. There is no doubt that a population policy will favour an increased 'growth rate when the population of the country is growing at a low rate, and there is the need for manpower and spe- cialization for utilizing the natural resources and for the military. However, the problem to be kept in mind is the fulfillment of the health, education, food, housing and employment requirements of the growing population. According to 1927 census results Turkey's population was 13 million 648 thousand. In the next 7 years from 1927 to 1935 the average rate of population growth stayed at 2.1 % per annum. During 1935-45 this rate decreased to 1 %, increasing to 2.2 % between 1945 -1950, and finally reaching 2.9 % in 1960. After 1960, there was a minor decrease in the population growth rate. Accor- dingly, Turkey's population increased by 22,048 million in 43 years. This short review of population trend in Turkey should be considered together with the population policies implemented in the country. Regarding the population policies put into effect, the history of the Turkish Republic can be divided into two periods. The first starts from the beginning of the Republic (1923) and extends to the mid-sixties during which a high birth rate was officially reco- mended and supported. During the second period which starts from the 1960's an antinatalist policy was adopted.

153 THE POPULATION OF TURKEY

The fundemental reasons for the population policy during the first period, were demographic/military and economic in nature. Prior to the establishment of the Republic, involment in various wars such as the Italian War, the Balkan Wars World War I and the War of Independence caused a considerable decrease in the male population which, in turn reflected a temporary decrease in the birth rate. Apart from that during the first period, the total death rate (30-40 per thousand.) and the infant mortality rate were alt an increasingly high level. Moreover, during this period the power of á country was judged by the size of its army, and rtihe size of the Turkish army was considerably small compared to the armies of the more powerfull countries. The main economic consideration for the acceptance of the population policy in this period was the shortage of manpower in the agricultural sector. Also the fact that increased production necessia- ted large scale investments and that children started working at early ages made possible the bringing up of children at low costs. Due to these roughly outlined reasons the policy makers of the young Republic had pass a law declaring any kind of family planning method to be illegal, in order to put the previously mentioned population policy into effect. According to this law, any kind of publication and advertisement on the subject of population planning methods, and manufacture, sale and import of the population planning devices were prohibited. Also, some tax exemption benefits were given to those families who had more than certain number of children. From the demographic point of view, there appeared a significant decrease in the death rates (15 per thousand) after 1950's. At the same time, birth rate reached 40 per thousand as a result of the pronatalist policy adopted. Both these factors have been the main causes of present rate of population increase. With an increase in the average expectation of life at birth and birth rate a balance between the males and females at fertile ages was established. Consequently, the increase in the fertility level caused a distortion of the balance between the economically active

154 ECONOMIC AND SOCIAL IMPLICATIONS and the dependent segments of the population. For example, in 1955 every 1.000 economically active population had to support 753 persons whereas this number increased to 836 per thousand in 1970. Another factor causing a change in the Population Planning Law in Turkey was the mother and child health conditions. Although we have no reliable information, it is calculated that the number of women who die because of abortion is about 10.000 per annum. Also, infant death rate seems to be very high (155 per thousand). Due to the aJbove mentioned factors, it was felt that the time had come for a change from the pronatalist approach, which was implemented for a long time to the Population Control Act which was consequently put into action. However prior to discussing the extend of the benefits of Popu. latiori Control Act it will be more penetrating to consider the socio- economic policies in Turkey. In the Third Development Plan (1973-1977) the long term goals are stated as; improvement in the standards of living; industrialization, less dependency on external sources, solving of the unemployment problem and improvement of the income distribution. There- fore, the socio-economic implications of a high rate of population increase so far discussed should be considered in connection with the social and economic policies adopted. A high population growth rate will have an adverse effect on the accumulation of capital, which in turn will hinder investments. The main reason for this adverse effect is the high proportion of the consuming age group in the total population. The dependency of a large group makes savings difficult, which thus limits investments. As previously mentioned, there is yet another relationship between population growth and investments. This is the prdblem between invenstments (demographic) which aim to prevent a decrease in percapita income and economic investments. Since any improvement in the standard of living is related to an increase in per capita income, the higher the rate of increase in the latter, the faster an improvement in the former will be registered. 155 THE POPULATION OF TURKEY

The investments required for the next 10 years in order to reach an increase of 4 % in per capita income, show a difference of 15 % with a 2.6 % population increase, andl with a 2 % population increase. The composition of the investments is also influenced by the population growth. Mainly, as the proportion of adult population is low, investments such as health, education and housing reach a high level. Consequently, the increase of the income level through total investments, will stay at a low state. As in other less developed countries, unemployment presents a problem for Turkey. The high population growth rate increases the supply of labour, but job opportunities are still limited. It is difficult to give the correct unemployment and over-employment figures in Turkey for the time being. However, according to the Third Five Year Plan, it is stated that in 1972 ^ there were 1.6 million unemployed and/or unproductive labour force in Turkey. At present, it seems difficult to solve the employment problem through creating new job opportunities, for, even a decrease at the present level of population growth will not influence the people Who will join the labour force for the next 15 years. It should also be stated that the lowering of the dependency to the external sources, which is an economic target, is related to the proportion of the investments that can be taken from the GNP. This becomes difficult to the extent that the rate of population growth affects the level of demographic investments. Available data shows that income distribution trend in Turkey is working against the low income groups. In a country with a big unemployment problem it is inevitable that the majority of the families will receive a very small share of the total income. It is quite difficult to have a just income distribution in such a country. This situation is felt seriously in Turkey. Previously its was stated that the birth rate in low income groups is quite high. This means that more people will share an income.

156 ECONOMIC AND SOCIAL IMPLICATIONS

Consequently, it can be said that social and economic policies in Turkey do not seem to take into consideration the high rate of population growth. We can therefore pose the following question : What has been achieved so far? — On the 1st of April 1965 the Population Control Act was put into action; the prohibition of the manufacturing, im- porting and selling of birth control devices was abolished. A Po- pulation Planning Department was established within the administra- tion of the Ministry of Health, its first target t>eing the achievement of a decrease in the present level of the population to 2 %. However, due to the organizational factors involved and the programs not fitting the socio-economic structure of Turkey, not much has been achieved in the stated program. According to many surveys, it is stated that the women's attitudes and practices in the use of birth control devices differ according to their socio-economic levels. Therefore, implementation of a population policy where social and economic policies are not taken into consideration, will not be successful in reducing the birth rates.

157 THE POPULATION OF TURKEY

REFERENCES

Bulutay, Tuncay; Timur, Serim; and Ersil, Hasan, Income Distri- bution in Turkey, 1968, Sevinç Matbaasi, Ankara, 1971. Gökgöl, TürMz, A., Socio-Economie Determinants of KAP in Po- pulation Planning in Turkey, M. A. Thesis. Hacettepe University, Ankara, 1972. Gökgöl, Türkiz A., "An Essay on the Relationship of Population and Education", Unpublished paper. Hacettepe University, Ankara, 1971. Keles, Rugen and Can, T., The Problem of "Squatters Huts", Re- public of Turkey, Prime Ministry, State Planning Organization, An- kara, 1966. Republic of Turkey, Prime Ministry, State Planning Organization, Third Five Year Development Plan 1973-1977. Publication No. DPT 1272. Prime Ministry Printing Office, Ankara, 1973. Republic of Turkey, Prime Ministry State Planning Organization, Second Five Year Development Plan 1968-1972. Prime Ministry Printing Office, Ankara, 1967. Republic of Turkey, Prime Ministry State Institute of Statics, Census of Population 1950, Publication No. 359. Ankara, 1950. Republic of Turkey, Prime Ministry State Institute of Statistics, Census of Population 1960, Publication No. 452. Ankara, 1964. Republic of Turkey, Prime Ministry, State Planning Organization, Problems of Urbanization. Report of Special Commission, Ankara, 1972. Republic of Turkey, Prime Ministry, State Institut© of Statistics, Turkish Statistical Manual 1971. Publications No. 670. Ankara, 1973. Tuncer, Baran, The Impact of Population Growth on the Turkish Economy, Hacettepe University Publication No. 3, Dogus Matbaasi, Ankara, 1968.

158