Lawrence H. Officer - A New Balance of Payments for the United States, 1790–1919: International Movement of Free and Enslaved People, Funds, Goods and Services

Immigrant remittances via money transfers are also termed immigrant remittances via financial intermediaries or simply immigrant remittances. I begin with benchmark years, then develop inter/ extrapolation series to fill the entire period. There are four benchmark figures, the most important of which is for the year 1907 . This is the year on which the other benchmarks depend and the only year with an existing country-distribution of remittances. It is also a year with wide divergences among estimates and disagreement as to the correct figure. Charles F. Speare (, p. 231) the range 150200. The adopted figure is $ 175 million .

Rationale: Magee and Thompson (, pp. 549553) present a model in which a multiplicative factor reduces total reported remittances via financial intermediaries and money orders. The parameters of the factor are the share of immigrant remittances in reported (immigrant plus commercial) remittances and the share of unreported remittances in total (reported plus unreported) remittances. Applying their two alternative sets of parameters to the Immigration Commission figure, the results are 153 and 194, averaging 173.5close to 175, the midpoint of the BWT range.

For 1913 : $225 million , the midpoint of the range 200250 ( Williams

I follow Dickens ( The years 19081912 are interpolated between 1907 and 1913, 19141918 between 1913 and 1919.

For 18601907, I develop an interpolator index with country components, and country weights computed as proportions of total remittances in 1907., pp. 449451).

For the number of remitters , consider Europe first. I begin with official immigration for each European component country (or country group) 18201908, convert to proportion of total Europe (the residual constituting proportion for other Europe ), and transform to calendar years 18201907..

A similar procedure is followed for China and Japan, with immigration proportion relative to oceanic rest-of-world. (Sects. 56). Then other-countries net immigration admittances are the sum of other Europe , other oceanic rest-of-world, and Canada .

Finally, following Magee and Thompson (, pp. 559560), the number of remitters for a given country is the moving sum of current-year and past-four-years net immigration admittances .

Immigrant earning capacity is a prospective concept, so pertains to the current year. It is measured as the product of the U.S. money wage and employment proportion (one minus 1/100th the unemployment rate).

I move to individual-country indexes . In the case of two countriesthe United Kingdom and Austria-Hungarythe indexes are direct remittance series, therefore clearly dominating the country default index. For the United Kingdom, the basic series 18481875 is immigrant remittances via banks and mercantile houses from the United States and BNA . Per necessity, the series is extended to 1860 via the 1880 overlap with the Austria-Hungary default index.

International money orders issued in the United States to the specific country are judged superior to the default index, but only after there has been sufficient experience with the facility so that its use becomes routine and providing that there are at least ten usable calendar years. Germany (18851907), Italy (18851907), and Scandinavia (18901907) fall into that category. For each country, linking to the default index carries the series back to 1860. For Russia-Finland, the Balkans, Greece, China, Japan, and other-countries, the default index must be used for the entire 18601907 period.

Each countrys (or country-groups) component series is converted to an index number 1907=1; the series are weighted per note 3; and the interpolator index , thereby obtained for 18601907, interpolates total remittances 18611906 between 1860 and 1907.

I adopt Norths (, we have U.K. remittances 18481860. As extrapolator series 17901848: take the U.K. proportion of Europe immigration times European total immigration admittances , both lagged three years. This series extends U.K. remittances to 1790 via the 1848 overlap. For benchmark 1860, U.K. remittances are $2.6 million and therefore other-Europe $2.0 (4.6 - 2.6) million. The extrapolator series for other-Europe is one minus the three-year-lagged U.K. proportion times three-year-lagged European total immigration admittances; and, with the 1860 overlap, the extrapolator carries other-Europe back to 1790. The sum of U.K. and other-Europe remittances 17901859 completes the series.