Estimating Vietnamese population over the 20th century is a difficult challenge for more than one reason.
First, there is the lack of data. Like many developing countries, Vietnam has set up a vital registration
and census administration, step by step, in a process that has taken all of the 20th century. Started at the
beginning of the century, data collection reached some kind of satisfying level in the 1930s, when WWII
and the liberation wars interrupted its development until the late 1970s. At that time vital registration
and, particularly, the census organisation were remodelled by international standards. As a result,
Vietnam’s demographic data diverge considerably, both in its availability and in its inner structure.
Other difficulties have political backgrounds. Before WWII demographic data collection was organised
by the French administration It relied on local authorities, that were more than suspicious about it.
There is no doubt data quality suffered from this resistance. After the departure of the French
administration, in 1955, and until the end of the American war in 1975, data collection, as far as it
continued, was organised separately in the North and in the South, and demographic indicators were
used as arguments in the ideological confrontation between the two parts of the country. Even in the late
1970s data may be suspected to be ideologically biased. It is not before the 1980s, when the Renovation
movement sweeps away many socialist ideas and practices in Vietnam's society, economy and
administration, that demographic data is collected and published using scientific methods and
international standards.
A last difficulty comes from the demographic evolution itself. In 1989, Vietnam has accomplished part
of the mortality transition and started its fertility transition. So far, it follows a well-known pattern of
many developing countries. But Vietnam knew many exceptional events that disturbed this pattern: two
world wars and two liberation wars, a major famine in 1944-1945 and an important out-migration flow
after 1975. These events have influenced Vietnam's demographic history at levels that are not always
well known.
Facing such difficulties, we cannot aim to reconstitute exactly Vietnam's demographic 20th century
history. What we may try is to figure out a model that combines all available information and that fits as
well as possible to all of this at the time. If, moreover, this model is demographically consistent, then we
may accept it as a stylised version of real Vietnamese 20th century demographic history.