The measurement of remittances has been no straightforward affair because states, international organizations, banks, and researchers use varying concepts, definitions, and methods to measure and report remittances (see Kapur 2003 and World Bank 2007). Although there has been a general agreement that migrant remittances have grown and reached substantial levels relative to other international transfers of funds, inconsistencies in definitions and data have contributed to some uncertainty about the magnitude and trends in remittances and their impact on development.

Concepts and methods used to compile data are not the same in all countries.  Some countries report incomplete or no official data while others include data from non-governmental sources such as commercial transfer companies.  Distinctions between investments and remittances can be rather ambiguous.  These inconsistencies pose problems in compiling and comparing national remittance statistics (World Bank 2007).

Another common problem in estimating remittances is that official records capture formal transfers but not informal remittances sent through family, friends, or “black market” operators.  Though the amount of informal transfers is unreported, it has been estimated by the World Bank as roughly equal to formal remittances

The most frequently cited official data on remittances is that provided by the International Monetary Fund based on balance of payments data.  In response to a request from the G7 meeting of 2004 for improved remittance data, the Statistics Department of the IMF, the Development Data Group of the World Bank, and the Statistics Division of the United Nations formed the International Working Group on Improving Data on Remittances, which proposed more unified concepts and definitions for measuring remittances that have since been adopted by the IMF for compiling balance of payments statistics.  According to the new definitions, total remittances include all personal transfers (cash or in kind goods transferred between resident and non-resident households), compensation of employees (wages of workers abroad net of taxes and travel expenses), and social benefits (social security and pension fund payments).  For a history, definitions, and discussions of these recommendations and adoption by the IMF, see World Bank 2007 and IMF 2008.

Household surveys are a useful source for enhancing official data on remittances.  For social scientists, representative household surveys have been extremely valuable in improving the quality of data on remittances for particular populations, particularly of informal transfers, and providing insights into the determinants of remittance flows. Smaller, often non-representative surveys and ethnographic studies have been useful in revealing the nature of various informal remittance transfer systems, such as the hawala and hundi systems, (Kapur 2003; Salomone 2006), and the importance of in-kind remittances.  (For a January, 2008 discussion of the role of household surveys in measuring remittances, which was organized by the United Nations Economic Commission on Europe, World Bank, and US Census Bureau, see: http://www.unece.org/stats/documents/2008.01.migration.htm).

Surveys also shed light on the importance of internal remittances, which are probably more frequent within rural households but smaller in value than international remittances (see topic 5 on internal remittances).

Notwithstanding definitional, data, and measurement problems, it is certain that global remittance flows have increased substantially over the past two decades. According to IMF Balance of Payments Statistics data, recorded remittances to developing countries have increased from $31.1 billion in 1990 to $76.8 billion in 2000 to reach $240 billion in 2007 (Ratha et al. 2007). The real increase in remittances is probably lower than these figures suggest because part of the increase in officially recorded remittances reflects improvements in defining and recording remittances, a shift in the sending of remittances from informal to formal channels (Kapur 2003), and a depreciation of the US dollar during that time period.

Registered remittances now amount to well over two times the amount of official development assistance and are widely considered to be a more stable, counter cyclical, and sustainable source of foreign currency than foreign direct investments (Salomone 2006).  Not all remittances originate from wealthy countries. According to recent estimates by Ratha and Shaw (2007), South-South remittances represent 10 to 29 percent of total remittances received in developing countries.