Internal migration

Internal migration is the movement of people within a country involving the establishment of a new temporary or permanent residence. These movements can be voluntary, as in the case of a person who decides to move from a rural to urban area, or they can be forced, in which case they refer to movements of Internally Displaced Persons (IDPs), further described in the sub-section of Part II, Chapter 3 on forced displacement. It is important to note that the term covers foreigners as well as nationals, provided the movement involves a change of residence (IOM, 2019a).

Table 1: Core migration indicators and topics for measuring international migration & temporary mobility (UNSD, 2022b) 

EMDP2C1T1

What are the challenges of measuring internal migration?

The complexity of defining the limits of internal migration is underscored by questions about the size of geographic data-gathering units and the duration of stay in a new place of residence. For example, internal migration from one province to another in a very big country like China is very different from internal migration between provinces in a small country like The Netherlands. Furthermore, the dynamics of small-scale internal migration are likely to differ from the dynamics of large-scale internal migration. For example, in India small-area internal migration is characterized by movements of women for marriage while large-area internal migration is characterized by movements of men for educational or economic opportunities (Skeldon, 2018)

There are also challenges with finding appropriate measures for the duration of stay in the new place of residence. Many countries do not have registration systems that record changes in the usual place of residence, which means that most data are generated retrospectively through survey instruments with questions about birthplace, place of residence one and/or five years prior and last place of residence. The table below shows how the number of geographic data-gathering units and the intervals used to measure a change in place of residence vary considerably across population censuses (Bell and) Charles-Edward, 2014). Birthplace is the least satisfactory measure as persons may have migrated internally and returned to their birthplace or persons may have moved from their birthplace to their new place of residence many years ago. On the other hand, as shown in the table below, using birthplace offers greater international coverage. Questions about previous place of residence provide better estimates, especially when they are measured using both one- and five-year intervals. 

Table 2: Smallest spatial units recorded in country, by migration data type (Bell and Charles-Edward, 2014)

Number of zones One year Five years Undefined interval Other defined interval Duration Birth place Any migration data

2-24

 3 8 8 2 5 25 13
25-99 7 6 17 4 9 27 20
100-249 1 5 2 3 2 10 11
250-999 3 11 5 3 5 17 17
1000+ 9 12 5 8 16 17 34
Not specified 6 11 17 12 35 27 47
Total 29 53 54 32 72 123 142

(Source: IMAGE Inventory of Internal Migration data collections. * 2000 Census round)

Case study: Using the aggregate crude migration intensity to measure internal migration

One solution to the problem of varying geographic data-gathering units is to measure internal migration based on all changes of address including local residential mobility, which is sometimes referred to as the overall or aggregate crude migration intensity (ACMI). The downside of this approach is that persons who changed address but moved down the street are also counted as internal migrants. Nonetheless, the ACMI can still be useful for comparing internal migration rates across countries and investigating the factors associated with an increase or decrease in average migration, as demonstrated in a study by Bell and Charles-Edwards (2014), who compiled data on internal migration from 82 countries. Comparing the ACMI for lifetime, one-year, and five-year intervals across countries, Bell and Charles-Edwards show that migration intensities are negatively correlated with economic downturns, population ageing and the equilibrium of regional economies but are positively related to national development. Internal migration intensities are also connected to international migration rates, as countries with net international migration gains show high levels of internal migration (suggesting potential substitution or displacement) while those with net international migration losses show low levels of internal migration.

How to improve internal migration data?

Bell and Charles-Edwards (2014) outline five key recommendations for the international statistical community to improve cross-national comparisons of internal migration:

  1. Internal migration should be measured over a fixed interval, ideally over one or five years, with a lower priority accorded to place of birth within a country;
  2. Data on place of residence should be coded to the smallest geographical zone feasible;
  3. Priority should be given to collecting data on all changes of usual address to facilitate comparisons of overall migration intensity;
  4. Data on duration of residence, if collected, should be recorded as length of residence rather than year of arrival, and indicate explicitly the spatial unit to which it refers;
  5. Statistical agencies should disseminate a range of standard outputs on internal migration including detailed origin-destination flow matrices.
< Table of contents
Checklist
References
Resources