Gender Ratios in Global Migrations, 1850-2000 Data collection funded by the National Science Foundation and the National Institutes of Health Trent Alexander.

Gender Ratios in Global Migrations,

1850-2000

Data collection funded by the National Science Foundation and the National Institutes of Health

Trent [email protected]

Donna [email protected]

Katharine [email protected]

Johanna [email protected]

Gender ratios in mass migrations

Ravenstein’s “Laws” assert that men vastly outnumber women among long-distance movers

World history and demography textbooks present this as conventional wisdom about the past

Recently, researchers at US-Dept of Labor and UN have suggested a “remarkable” shift in migrant gender balance has implications for policies re integration (social programs, labor

market impact, public assistance, family structure issues)

But nobody has any decent data...until now

Key questions

Migrant gender balance: has there been change over time?

Do patterns in migrant gender balance differ for internal migrants as compared to international migrants?

Do patterns in migrant gender balance differ for short-distance migrants as compared to long-distance migrants?

Methods

Use individual-level census data from Minnesota Pop Center

IPUMS-USA: 15 censuses from 1850-2000

IPUMS-International 29 countries 94 censuses from 1960-2005

North Atlantic Population Project: 4 countries 8 censuses from 1851-1901

Methods

Use information in the datasets to identify...

internal migrants international migrants non-migrants

Bring in additional geographic information to determine...

how far migrants moved whether or not they moved from an adjacent country or region

Methods: identifying migrants

Coding issue—which types of borders count?

international migrant: any national border-crossing counts

internal migrant: only crossing of first-level administrative units counts

state, province, department, district, etc. huge numbers of people moved within the boundaries of these areas we do not count them as migrants

Methods: adding geographic information

How far did migrants move, in kilometers?

determined geographic coordinates to all countries and first-level units used the geographic mid-point of each area

attached geographic coordinates to our data for place of residence place of birth

use geometry to determine distance between place of residence and place of birth

Key question number 1

Migrant gender balance: has there been change over time?

YES, clearly.

Source: Integrated Public Use Microdata Series files (IPUMS-USA)

Figure 1. Women's representation among U.S. foreign-born, 18+ years

30%

35%

40%

45%

50%

55%

60%

1850

1860

1870

1880

1900

1910

1920

1930

1940

1950

1960

1970

1980

1990

2000

2005

/200

6

Per

cen

tag

e fe

mal

e

Source: Integrated Public Use Microdata Series files (IPUMS-USA)

Figure 2. Women's representation among U.S. foreign-born, 18+ years

30%

35%

40%

45%

50%

55%

60%

Pe

rce

nta

ge

fe

ma

le

Age-standardized

Unstandardized

Figure 3.Age-standardized estimates of women's representation among U.S. foreign-born, by region of

birth, 18+ years

0%

10%

20%

30%

40%

50%

60%

Percen

tag

e f

em

ale

NW Europe

SE Europe

Mexico

Other Latin America

Asia

Africa

Source: Integrated Public Use Microdata Series files (IPUMS-USA)

Figure 4 Age-standardized estimates of women's representation among the foreign-born in

18 countries, 18+ years

30%

35%

40%

45%

50%

55%

60%

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Pe

rce

nta

ge

fe

ma

le

Argentina

Brazil

Canada

Chile

Colombia

Costa Rica

Ecuador

France

Kenya

Malaysia

Mexico

Panama

Philippines

Romania

South Africa

Spain

Uganda

Venezuela

Model 4

Source: Integrated Public Use Microdata Series files (IPUMS-International)

Key question number 1, cont’d

Migrant gender balance: has there been change over time?

YES, but it seems like we’ll need to focus on individual countries to make sense of all this

Some countries consistently DRAW more women example of Argentina

Some countries consistently SEND more women example of Peru

We have no idea WHY. But we could dig much deeper in the data.

Figure 6Women's representation among persons moving to Argentina and Brazil,

between 1995-2000, by country of origin

Source: Integrated Public Use Microdata Series files (IPUMS-International)

0

10

20

30

40

50

60

70

Peru Paraguay Uruguay Bolivia Chile United States

Country of origin

Perc

en

tag

e f

em

ale

Argentina

Brazil

Figure 7Women's representation among persons moving from the U.S. or Peru

between 1995-2000, by country of destination

Source: Integrated Public Use Microdata Series files (IPUMS-International)

0

10

20

30

40

50

60

70

Argentina Chile Spain Colombia Brazil

Country of destination

Pe

rce

nta

ge

fe

ma

le

Peru

United States

Key question number 2

Do patterns differ for internal migrants as compared to international migrants?

YES....

Figure 8Women’s representation among internal and international

migrants in the United States

Source: Integrated Public Use Microdata Series files (IPUMS-USA)

30%

35%

40%

45%

50%

55%

60%

Perc

en

tag

e f

em

ale

Internal

International

30

35

40

45

50

55

60

1950 1960 1970 1980 1990 2000

Per

cen

tag

e fe

mal

e

Argentina

Brazil

Chile

Colombia

Costa Rica

Kenya

Mexico

Philippines

Romania

South Africa

Spain

Uganda

United States

Venezuela

Decade Average

Figure 9Women’s representation among internal migrants in 14 countries

Source: Integrated Public Use Microdata Series files (IPUMS-International)

Key question number 3

Do patterns differ for short-distance migrants as compared to long-distance migrants?

Yes, a little bit Women have typically moved slightly shorter distances Magnitude of difference is minor

Long-term trends in distance moved are interesting but don’t seem to vary by gender

Figure 10 Distance women moved as a proportion of distance men moved in 12 countries,

international migrants

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1960s 1970s 1980s 1990s 2000s

Wo

me

n's

dis

tan

ce

div

ide

d b

y m

en

's d

ista

nc

e

Brazil

Argentina

Chile

Colombia

Costa Rica

Kenya

Mexico

Romania

Spain

Uganda

USA

Venezuela

Decade average

Source: Integrated Public Use Microdata Series files (IPUMS-International)

Figure 11Distance moved (in km) for internal and international migrants in the United States

0

2,000

4,000

6,000

8,000

10,000

12,000

1850 1860 1870 1880 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2005

Km

mo

ved International: men

International: women

Internal: men

Internal: women

Source: Integrated Public Use Microdata Series files (IPUMS-USA)

Figure 12Distance moved (in km) for international migrants, women only,

by country of destination

0

2,000

4,000

6,000

8,000

10,000

1960s 1970s 1980s 1990s 2000s

Km

m

ov

ed

brazil

chile

colombia

costa rica

kenya

mexico

romania

spain

uganda

USA

venezuela

argentina

Source: Integrated Public Use Microdata Series files (IPUMS-International)

What else could we be doing?

Focus on changing dynamics within particular countries Focus on atypical countries? Or prototypical ones?

Look at migrant gender balance along other dimensions education, marital status, employment status, others?

Probably many other things we haven’t thought of yet

Additional information about the data at http://ipums.org

Trent [email protected]

Donna [email protected]

Thank you.

“Use it for good, never for evil.”

Katharine [email protected]

Johanna [email protected]