Demography And Population Registers Research Paper

1. Contents

Population registers, also referred to as household registers, are longitudinal, nominative records of the households and individuals that make up a community. They follow households through time, recording changes in their composition and other characteristics. Registers also follow individuals, recording when and by what means they enter and exit the household or community. They usually record their relationship to the head of the household. In most cases, they measure socioeconomic status at either the household or individual level. European registers, for example, often record the occupation of the household head. In a rural setting they may pro-vide details on landholding, including type of tenancy.

Authorities originally established population registers for diverse reasons. In China, where systematic population registration has perhaps the longest his- tory, successive governments used registers for centuries as tools for administration. In parts of Europe during the nineteenth century, civil authorities also introduced them as a tool for administration. In Japan, the state originally established a nationwide system of local population registers in the late seventeenth century to prevent the spread of Christianity, though of course it quickly found the registers useful for more routine tasks (Cornell and Hayami 1986). In some places, religious authorities maintained population registers. For example, parish priests in Sweden during the nineteenth century compiled catechetical registers every year that listed households and their members to identify candidates for literacy exams. In Italy, churches compiled status animarum every year that listed all of the households in the parish along with their members (Manfredini 1996).

Even though registers had diverse origins and purposes, they are broadly similar in terms of format. Almost all of them are organized in one of two ways. Many of the systems established in continental Europe during the nineteenth century were continuous. A register was created that contained a separate entry for each household in a community. Once created, often from the results of a census, it was updated on a continual basis. As events such as births, marriages, occupation changes, migrations, and deaths occurred in a household, annotations were made to its entry. Once annotations accumulated to the point where further updates were impractical, the register was retired and a new one was created containing only current information.

In the remaining locations, mainly in Asia but also in some parts of Europe, households in a community were recorded in census-like fashion at fixed intervals, most commonly annually, but in some cases less frequently. Copying information from the previous register and updating it produced a new register. Entries commonly included additional annotations indicating whether or not demographic events such as birth, migration, marriage, or death had occurred since the last update. Since households and their members were usually listed in roughly the same order in successive updates, entries for individuals and households are easily linked from one register to the next. Demographic events and other changes in status not specifically annotated can be inferred by comparison of entries in successive registers.

Historical studies that use population registers often supplement them with data from other sources. In most European settings, it is possible to use names, ages, and other identifying information to link to baptism, marriage, and burial records in parish registers and obtain additional information about the timing of demographic events. In some locations, links can be made to vital registration data, including birth, death, and marriage certificates. Such linkage can turn up vital events that the population registers missed. In some cases, it is also possible to link to entries in land and tax registers, yielding richer detail on socioeconomic status.

Population registers combine features of the sources most commonly used to study population and society in the past. Like both genealogies and family reconstitutions from parish registers, population registers are inherently longitudinal. They follow individuals through time, allowing for the reconstruction of life histories. Like parish registers, population registers usually provide a nearly complete record of the vital events occurring in a community. Like censuses, registers describe entire communities, identifying who was present and at risk of experiencing a demographic event or other transition. They also provide details on household characteristics, including composition, socioeconomic status, and members’ relationships to each other.

Registers accordingly overcome many of these sources’ limitations and allow for new avenues of inquiry. In contrast with genealogies, they cover entire communities, not selected families. Whereas genealogies are usually retrospective, compiled based on the recollections of family members, registers are prospective. In contrast with family reconstitutions and genealogies, registers describe residential context, not just kinship. Finally, unlike most censuses, registers allow for the direct calculation of demographic and other rates because they are longitudinal, and include information on demographic events and other transitions.

2. Applications

Registers may be used to examine the determinants of marriage, childbearing, remarriage, and death because like panel studies they allow use of event-history techniques that measure the effects of time-varying community, household, and individual characteristics on the chances of these and other events (Alter 1988). Registers can also be used to illuminate the links between economic conditions and demographic behavior. Whereas previous studies have been limited to correlating prices and aggregate vital rates at the regional or national level, application of combined time-series and event-history techniques to register data allows comparison of the price responses of different subgroups of the population, yielding new insights underlying causal mechanisms (Bengtsson 1993).

Register data lend themselves to the study of household dynamics. Classic studies of historical households were limited to examinations of size and structure by their reliance on cross-sectional census data. Registers, by contrast, follow changes in the structure and composition of households over time. Indeed, one of the first applications of register data was in a study of household dynamics in mid-nineteenth century Belgium (Van de Walle 1976). From register data it is possible to compute the rates at which households with different characteristics dis-solve, disappear, or change structure.

Topics such as migration, adoption, and occupational mobility that are difficult or impossible to study with traditional sources for historical demography and family history are ideal for analysis with registers. Studies of adoption have been carried out in several Asian locations, including Japan and Taiwan (Wolf and Huang 1980). As for migration, studies have been carried out in both Europe and Asia. They usually focus on the determinants of the timing of family migration and, especially in Europe, individual home-leaving. Examination of occupational mobility is also possible because through record linkage, it is straight-forward to examine the relationship of children’s statuses as adults to the statuses of their parents, grandparents, and other kin.

3. Limitations

Like all other historical sources, registers suffer from limitations. One of the most common is that they may fail to record short-term movements into and out of a community. While such problems are most common in systems that record the population at fixed intervals because they miss those who enter and exit again between updates, they also exist in continuous registration systems. In-migrants may never be recorded if they are only in the community for a few weeks or months because they leave or die soon after arriving. Similarly, those who reside in the area but who migrate out for a very short time may never be identified as having been away. Infants who die soon after they are born may leave no record of their existence, leading to underestimates of the level of mortality in the first months of life.

Where service migration was part of the life cycle, older adolescents and young adults are often poorly covered from the time they leave their parent’s household to the time they marry and form one of their own. If they remain in the area, they are difficult to link to their natal households. Therefore, it is hard to tell whether young persons who leave their natal households are still in the area but living and working in other households, or have left the area completely. For this reason, studies in settings where service migration was common often exclude from consideration older teenagers and young adults who have never married.

Demographic and other events that occur to residents when they are outside the community are unlikely to be recorded. This is not a serious problem, however, because population registers also allow individuals who have left the community to be excluded from consideration starting from the time of their departure. The caveat remains that if the individuals who left the community were selected on some characteristic that affected the chances of the event of interest, rates computed for those who remained might differ from the ones that would be observed if no one left.

4. Availability

In Asia, historical registers for China, Korea, and Japan are available. Within China, there are at least three types of registers. In Taiwan province, Japanese colonial authorities established a registration system that generated data of extremely high quality from 1895 to 1945 (Wolf and Huang 1980). In northeast China from the mid-eighteenth century to the be-ginning of the twentieth, triennial registers recorded the residents of state farm systems in great detail (Lee and Campbell 1997). Registers from what is known as the baojia system exist for other parts of the country in the eighteenth and nineteenth centuries, but they appear to be of very poor quality (Skinner 1987). In Japan, there are annually updated population registers for many villages from the mid-seventeenth century to the mid-nineteenth century (Cornell and Hayami 1986). In Korea, triennial registers remarkably similar in format to the northeast Chinese ones exist.

In Europe, registers exist for parts of Belgium, Italy, The Netherlands, Sweden, and elsewhere during the eighteenth and nineteenth centuries. Belgian registers are continuous (Alter 1997), as are those from The Netherlands, where there is now an effort underway to use them to compile a nationally representative sample of individuals (Mandemakers 1994). Italian registers, consisting of annual status animarum, exist for a number of parishes (Manfredini 1996). In at least one Italian City, Venice, detailed data from a continuous registration system that operated during the middle of the nineteenth century also exist (Derosas 1989). In Sweden, catechetical registers exist for the nineteenth century. Some of these data have already been transcribed into databases at the Demographic Database at Umea University.

Bibliography:

1. Alter G 1988 Family and the Female Life Course: The Women of Verviers, Belgium, 1849–1880. University of Wisconsin Press, Madison, WI
2. Bengtsson T 1993 Combined time-series and life event analysis. The impact of economic fluctuations and air temperature on adult mortality by sex and occupation in a Swedish mining parish, 1757–1850. In: Reher D, Schofield S (eds.) Old and New Methods in Historical Demography. Clarendon Press, Oxford, UK
3. Cornell L L, Hayami A 1986 The shumon-aratame-cho: Japan’s population registers. Journal of Family History 11: 311–28
4. Derosas R 1989 A database for the study of the Italian population registers. Historical Social Research Historische Sozialforschung 52: 59–65
5. Lee J Z, Campbell C D 1997 Fate and Fortune in Rural China: Social Organization and Population Behavior in Liaoning, 1774–1873. Cambridge University Press, Cambridge, UK
6. Mandemakers K 1994 Historical sample of the population of the Netherlands (HSN). Backgrounds, objectives and international context. In: Marker H J, Pagh H (eds.) Yesterday. Proceedings from the 6th international conference Association of History and Computing, Odense 1991. Odense University Press, Odense, Denmark
7. Manfredini M 1996 L’utilizzo degli Status Animarum nellerico-struzioni nominative: miglioramenti informativi qualitativive quantitativi. Il caso di Madregolo (1629–1914) [The use of Status Animarum in nominative reconstruction: improvements from qualitative and quantitative information]. Bollettino di Demografia Storica. 24–25: 113–29
8. Skinner G W 1987 Sichuan’s population in the nineteenth century: lessons from disaggregated data. Late Imperial China 8: 1–79
9. Van de Walle E 1976 Household dynamics in a Belgian village, 1847–1866. Journal of Family History 1: 80–94
10. Wolf A P, Huang C H 1980 Marriage and Adoption in China, 1845–1945. Stanford University Press, Stanford, CA