North Korea Census

Household Registration (Hojuk)
Research Use: Household registration is an excellent source of vital information and family relationships for all social classes, including commoners and slaves. These records are particularly valuable because they include female names.

Record Type: Registration of the population by household, similar to a census, maintained by the civil government for taxation purposes. Background: The original intent was to update the registers every three years, but the actual frequency of updating has not been determined. Household registration appears to be similar to the population registration systems used in China, such as the bao jia. Apparently two copies were made, 1) a single-sheet family copy and 2) a government copy listing separate households in a bound volume. Format can vary considerably.

Time Period: These records are known to exist from the 1600s until the Japanese took over Korea in 1910. (In the present FamilySearch Library collection, the earliest entry is 1615, the latest, 1901.) Earlier census-like records existed, but those were apparently destroyed in the Japanese invasions of the 1590s.

Contents: Typical entries include the name of the master or head of each household and often give the names of four ancestors in his direct line. They also provide titles and positions held, clan origin, place of residence, the names of the spouse and children, their ages and/or date of birth, and the status of individuals. The names of slaves are often listed. Records of commoners may list occupational skills or trades. The actual format can vary considerably.

Location: Seoul National University Library has the largest unfilmed accumulation of household registration which is part of the Kyujanggak collection, probably about 100 rolls. Other small collections can be found in local archives, in university libraries, and in private ownership.

Percentage in FamilySearch Library: About 10%, the Library acquired several rolls from the Central National Library in Seoul.

Population Coverage: While the census was supposed to cover all households, the actual coverage has yet to be fully ascertained. This is one of the few Korean records that records the names of commoners and slaves and also lists females. Originally the record likely covered a large portion of the population; however, because of record destruction the actual coverage is probably less than 10%.

Reliability: The household registration was compiled as a primary source for census information and is generally reliable.

Family Census Register (Hojuk Dungbon)
Research Use: Family census registration records are a primary source of birth, marriage, and death information in Korea. They identify names of parents, prove other relationships, and are very useful for linking generations.

Record Type: This is the modern civil registration system for Korea.

Background: The family census registration system was introduced during the Japanese occupation beginning in 1910. Because of Japanese influence these registers are quite similar to the population register [koseki] or the resident registration [juminhyo] records used in Japan. There have been modifications in this registration system since Korea became independent of Japan. The system likely varies from the North to the South.

Time Period: 1910 to present.

Contents: Specific content is unknown. Based on similarities with Japanese records it is likely that they give place of residence, name and birth date of the head of the household, the name and birth date of the wife, marriage date and place; names of others in the household including: children, grandchildren, brothers and sisters of the household head, servants, etc., with birth dates and places. Death dates are also recorded. These registers may also include occupations, criminal activity, place of origin, date of move-in or move-out. May also give names and birth dates of the parents and grandparents of the head of household and of his wife.

Location: In District Registration offices.

Population Coverage: Coverage should now be nearly 100%, but coverage would have been less when the system started and full coverage would not likely have been achieved until the 1960s.

Reliability: These records are very reliable.