Skip to content
Contents
South Asians in America

Chapter 6 · Second generation

The second generation of South Asian Americans

In 2023–25, about 2.2 million US-born children and adults had a parent born in South Asia, 61% of them under 18; in 2015–25, 82% of those aged 25 to 44 held a bachelor's degree, against 71% of South Asian immigrants of their parents' generation, aged 55 to 74, and 41% of Americans aged 25 to 44 whose parents were born in the US.

Affiliation
NYU Stern
Published

Second generation here means people born in the US with at least one parent born in India, Pakistan, Bangladesh, Nepal, Sri Lanka or Bhutan [1]. The Current Population Survey asks everyone where their mother and father were born [2], so it identifies the second generation whether or not parents live in the same household. It cannot identify South Asian Americans of the third generation, whose parents were born in the US.The March Current Population Survey (CPS) of 2023, 2024 and 2025 estimates an average of 2.24 million people born in the US to a parent born in South Asia, 61% of them under 18 [3]. Counting only parents born in India, Pakistan or Bangladesh, the countries the CPS codes in every year since 1995, the number rose from 0.40 million in 1996–98 to 2.15 million in 2023–25. In the pooled 2015–2025 samples the second generation's median age was 12, 66% were under 18 and 17% were 25 to 44. 85% had two parents born in South Asia, and 73% had a parent born in India.

In 2023–25, adults of the second generation made up 28% of adult citizens who were born in India, Pakistan or Bangladesh or born in the US to a parent born there, against 20% in 1996–98. Their number rose from about 84,000 to 862,000. The 1996–98 figures rest on 38 or fewer adult second-generation records a year.

Figure 1

The second generation with a parent born in India, Pakistan or Bangladesh grew from 0.40 million in 1996–98 to 2.15 million in 2023–25

People in the US by generation, parents or own birthplace in India, Pakistan or Bangladesh, March CPS, single years.

Show data
India, Pakistan and Bangladesh are the South Asian birthplaces the CPS codes in every year from 1995; Sri Lanka and Nepal are coded from 2007 and Bhutan from 2013, and including them adds 4% to the 2023–25 second generation. Earlier years are left out because the 1994 file codes only India and the 1994–95 weights are not comparable with later years. Single-year estimates rest on 150 to 861 second-generation records (35 to 349 adults). The 2014 estimate uses the redesigned three-eighths file [4].Source: CPS Annual Social and Economic Supplement 1996–2025 via IPUMS CPS, weighted.

At 25 to 44, the second generation held bachelor's degrees more often than immigrants of their parents' generation and twice as often as the third generation

In the pooled 2015–2025 samples, 81.8% of second-generation South Asian Americans aged 25 to 44 had a bachelor's degree (90% interval 79.6 to 84.0). Among South Asian immigrants of their parents' generation, people aged 55 to 74 who arrived at 18 or older in 1995 or earlier, 71.4% did; the difference is 10.4 points (7.1 to 13.8). Graduate degrees were about as common in the two groups (42.0% and 39.2%). The third generation or later aged 25 to 44 had 40.7% with a bachelor's degree and 13.2% with a graduate degree, and the second generation of other Asian origins 62.9% and 20.0%.

South Asian immigrants aged 25 to 44 who arrived at 13 or older are the second generation's contemporaries (median ages 35 and 32). 73% of them were not citizens and 60% had arrived within the previous ten years. 85.1% held a bachelor's degree, 3.3 points more than the second generation (90% interval 0.9 to 5.7), and 49.7% a graduate degree.

Employment differed by sex. Standardized to the second generation's ages, 55% of immigrant women aged 25 to 44 who arrived at 13 or older were employed, against 80% of second-generation women and 76% of third-generation women; for men the rates were 91%, 85% and 83%. Among full-time, year-round wage earners, the second generation's median earnings were $94,500 in 2024 dollars ($89,900 to $99,100). With the other groups reweighted to the second generation's ages, the median was $105,400 for immigrants who arrived at 13 or older, $75,100 for the second generation of other Asian origins and $61,100 for the third generation. 4.2% of the second generation lived below the poverty line, against 9.8% of the third generation. In the 2019–2025 samples, which carry the coverage question used here (1,115 second-generation records), 4.2% of the second generation had no health insurance in the previous year, against 9.3% of the third generation.

43% of the second generation aged 25 to 44 were married, against age-standardized rates of 75% for immigrants who arrived at 13 or older and 46% for the third generation. Immigrants who arrived before age 13 fall between the second generation and those who arrived at 13 or older on employment and marriage, but fewer of them held a bachelor's degree than in either group (76.0%, against 81.8% and 85.1%); their age-standardized median earnings, $94,000, were about the same as the second generation's.

Studies that compare the two generations over the same age range also find bachelor's degree rates within a few points of each other for second-generation and immigrant Indian Americans. Pooling the CPS for 2008 to 2016, Tran, Lee and Huang found that 80% of second-generation Indian Americans aged 25 to 40 held a bachelor's degree, against 79% of the first generation [5]. In the ACS for 2021–23, which records nativity but not parents' birthplace, 76% of US-born Indian Americans aged 25 and over held a bachelor's degree, against 77% of Indian immigrants [6]. Lee and Zhou call immigrant groups that are more educated than both the people who stayed behind and the host population hyper-selected [7]. At ages 25 to 44, both South Asian immigrants who arrived at 13 or older (85.1%) and the second generation (81.8%) held bachelor's degrees at about twice the rate of the third generation (40.7%).

Figure 2

Second-generation South Asian Americans aged 25 to 44 held bachelor's degrees at 82%, above immigrants of their parents' generation (71%) and twice the third generation's rate (41%)

Adults aged 25 to 44 (the parents' generation row: 55 to 74), pooled March CPS 2015–2025, with 90% intervals. Employment, earnings and marriage are standardized to the second generation's age distribution; earnings are for full-time, year-round wage earners, in 2024 dollars.

  • South Asian immigrants
  • South Asian second generation
  • Other Asian American second generation
  • Third generation or later
  • Parents' generation; third or later, non-Hispanic white

Show data
Generation is from the person's and parents' birthplaces. Arrival age is approximate because the CPS records arrival in ranges of years [8]. Poverty is the official measure. Intervals are not computed for the third generation, whose samples are large.Source: CPS Annual Social and Economic Supplement 2015–2025 via IPUMS CPS, weighted.

Nearly half of US-born children of South Asian immigrants speak a South Asian language at home

The American Community Survey (ACS) of 2022–24 counts an annual average of about 1.14 million US-born children under 18 who live with at least one parent born in South Asia, and 276,000 immigrant children who live with such a parent [9, 10]. The ACS sees only parents in the same household; the CPS for the same years estimates 1.35 million US-born children under 18 with a South Asian-born parent. Among the US-born aged 5 to 17, 46.5% spoke a South Asian language at home (90% interval 45.7 to 47.3) and 50.1% spoke only English. The share speaking a South Asian language was 54.5% when both parents were born in South Asia and 6.4% when the other parent was of another race. 93% of US-born children with two South Asian-born parents spoke only English or spoke it very well.

Hindi (7.3%), Telugu (6.7%), Urdu (6.3%) and Bengali (5.4%) were the languages most often spoken. Telugu, a language of southern India, was spoken by about as many of these children as Hindi.

Figure 3

54.5% of US-born children with two South Asian-born parents spoke a South Asian language at home; with one South Asian-born parent and one parent of another race, 6.4%

Children under 18 living with at least one parent born in South Asia, pooled ACS 2022–24, with 90% intervals. Language measures are for ages 5 to 17.

  • US-born, both parents born in South Asia
  • US-born, other family types
  • Immigrant children

Show data
Parents are those in the same household. "Other parent born elsewhere" means two parents present, one born in South Asia and one born elsewhere. "South Asian by race" means that parent was recorded as single-race South Asian, with the same codes as for the children; "another race" is any other answer, including other Asian groups and two or more races. Parents include a second mother or father in same-sex couples. South Asian languages are Indo-Aryan and Dravidian languages of the subcontinent, including Hindi, Urdu, Bengali, Punjabi, Gujarati, Marathi, Nepali, Sinhala, Telugu, Tamil, Malayalam and Kannada.Source: ACS 1-year samples 2022–2024 via IPUMS USA, weighted.

Figure 4

Hindi, Telugu, Urdu and Bengali were the languages most often spoken at home by US-born children of South Asian immigrants

Share of US-born children aged 5 to 17 living with a South Asian-born parent who speak each language at home, pooled ACS 2022–24, with 90% intervals. Languages spoken by at least 1% of these children.

Show data
The ACS records one language other than English per person.Source: ACS 1-year samples 2022–2024 via IPUMS USA, weighted.

Children of a South Asian-born parent and a parent of another race were usually recorded as multiracial

Of US-born children living with two parents, at least one of them born in South Asia, 76% had two South Asian-born parents. Those children were almost all recorded as single-race South Asian (94.7%), as were children whose other parent was born elsewhere but is South Asian by race (98.0%). When the other parent was of another race, 9.9% were recorded as single-race South Asian and 56.3% as two or more races. Across all US-born children living with a South Asian-born parent, 83.8% were recorded as single-race South Asian, so the single-race definition used in chapters 1, 2, 3, 4, 5 and 7 would leave out about 16% of them. Chapter 8 identifies South Asian Americans by their own or a parent's birthplace; it counts any citizen not born in South Asia with a South Asian-born parent, a broader rule than this chapter's requirement that the person be born in the US. Using the CPS for 2003 to 2013, Duncan and Trejo estimate that 24% of second-generation Indian Americans aged 25 to 59 do not report an Asian race, and that among Asian Americans those who do not report an Asian race have less schooling [11], which implies that counts based on race overstate the second generation's education. The CPS figures in this chapter use parents' birthplace instead; restricting them to people who report Asian race alone changes the second generation's bachelor's share only from 81.8% to 82.4%.

What these data cannot show

The adult comparisons rest on 1,638 second-generation records from 1,276 people in 1,097 households over eleven years. Under the CPS 4-8-4 rotation a household can be interviewed in March of two consecutive years [12], so some people appear twice; the intervals resample households across years. The CPS records race only as Asian, so it cannot identify South Asian Americans of the third generation, and it records arrival year in ranges, so the line between immigrants who arrived before and after age 13 is approximate.

The ACS links children only to parents in the same household, and its language questions record which language is spoken at home and how well English is spoken. They do not measure how well a child speaks the South Asian language, and the person completing the survey answers for the children. Neither survey follows people as they age, and this chapter does not link adults to their own parents, so each comparison between generations is between different people observed in the same years.

Data sources

Methods

Generations (CPS). Born in South Asia: BPL India, Pakistan, Bangladesh, Nepal, Sri Lanka or Bhutan. The trend uses India, Pakistan and Bangladesh only, coded in every year from 1995; Sri Lanka and Nepal enter the codes in 2007 and Bhutan in 2013, and the 1994 file codes only India. Arrived before 13: the midpoint of the arrival range (YRIMMIG) minus the birth year (survey year minus age) is under 13. Second generation: born in the US or an outlying area with a mother or father (MBPL, FBPL) born in South Asia. Other Asian second generation: US-born with a parent born in East or Southeast Asia, Afghanistan or Burma and neither parent in South Asia. Third generation or later: US-born with both parents US-born (NATIVITY = 1). Parents' generation: born in South Asia, aged 55 to 74, arrived at 18 or older, with an arrival range whose midpoint is before 1995; every range kept ends in 1995 or earlier, and the 2015–2025 files code 1994 and 1995 as one range. The 2014 sample keeps only the redesigned three-eighths file (HFLAG = 1), whose weights alone sum to the population. Electorate shares count citizens born in India, Pakistan or Bangladesh (CITIZEN 3 or 4) and second-generation adults.

Outcomes (CPS, ages 25–44, pooled 2015–2025). Bachelor's degree or more (EDUC ≥ 111), graduate degree (≥ 123), employed (EMPSTAT 10 or 12), married (MARST 1 or 2), official poverty (POVERTY = 10). No coverage in the previous year (ANYCOVLY = 1) is available in the 2019–2025 samples only. Earnings are INCWAGE for people with positive wage income who worked 50 or more weeks, usually 35 or more hours a week, in the previous calendar year, moved to 2024 dollars with the CPI-U for that year [13]. Age-standardized figures reweight each group to the second generation's distribution across ages 25–29, 30–34, 35–39 and 40–44. Medians are interpolated within $2,500 bins.

Children (ACS 2022–24). Children under 18 whose co-resident mother or father (BPLD_MOM, BPLD_POP, and in same-sex couples BPLD_MOM2, BPLD_POP2) was born in South Asia. US-born means born in the US (CITIZEN = 0); immigrant children are those born abroad who are not US citizens at birth (CITIZEN 2 to 5), which leaves out 24,000 children born abroad to American parents. Counts are annual averages over the three samples. Language at home is LANGUAGED; English ability is SPEAKENG. Race is RACED: single-race South Asian uses the codes of chapters 1, 2, 3, 4, 5 and 7; two or more races is RACE 8 or 9; the other parent's race is RACED_MOM, RACED_POP, RACED_MOM2 or RACED_POP2.

Uncertainty. CPS intervals come from a Poisson bootstrap (2000 replicates) that gives each household one draw shared across all years it appears. ACS intervals use the 80 replicate weights, pairing replicate r across years. Differences between groups are computed within each replicate, so a household with members in both groups is resampled once; 180 of the 1,097 second-generation households also contain a parents'-generation immigrant. Intervals are 90%. In the pooled estimates, cells with fewer than 50 records are suppressed and those with fewer than 200 are marked †; the single-year trend counts are shown with their record counts instead.

Reproduction. south-asian-voters/chapters/06-second-generation/build_data.py computes every number and writes data.json; build_page.py and charts.js assemble this page; refs.json holds the references and quoted outside facts.

References

  1. US Census Bureau (2026). About Foreign Born Data Tables Based on the Annual Social and Economic (ASEC) Supplement to the Current Population Survey (CPS). census.gov, definitions of first, second and third-and-higher generations (page revised August 27, 2026). www.census.gov/topics/population/foreign-born/guidance/cps-guidance/about-fb-cps-asec.html
  2. IPUMS CPS (2026). MBPL: Mother's birthplace, and FBPL: Father's birthplace, variable descriptions. IPUMS CPS. cps.ipums.org/cps-action/variables/MBPL
  3. Flood, S.; King, M.; Rodgers, R.; Ruggles, S.; Warren, J. R.; Backman, D.; Breton, E.; Cooper, G.; Rivera Drew, J. A.; Richards, S.; Van Riper, D.; Williams, K. C. W. (2025). IPUMS CPS: Version 13.0 [dataset]. Minneapolis, MN: IPUMS. doi:10.18128/D030.V13.0
  4. IPUMS CPS (2026). The 2014 ASEC: the 3/8 and 5/8 files, and HFLAG: Flag for the 3/8 file 2014. IPUMS CPS. cps.ipums.org/cps/three_eighths.shtml
  5. Tran, V. C.; Lee, J.; Huang, T. J. (2019). Revisiting the Asian second-generation advantage. Ethnic and Racial Studies 42(13): 2248–2269. doi:10.1080/01419870.2019.1579920
  6. Im, C. (2025). Facts about Indians in the U.S.. Pew Research Center, May 1, 2025. www.pewresearch.org/race-and-ethnicity/fact-sheet/asian-americans-indians-in-the-u-s/
  7. Lee, J.; Zhou, M. (2015). The Asian American Achievement Paradox. New York: Russell Sage Foundation.
  8. IPUMS CPS (2026). YRIMMIG: Year of immigration, variable description and comparability notes. IPUMS CPS. cps.ipums.org/cps-action/variables/YRIMMIG
  9. Ruggles, S.; Flood, S.; Sobek, M.; Backman, D.; Cooper, G.; Rivera Drew, J. A.; Richards, S.; Rodgers, R.; Schroeder, J.; Williams, K. C. W. (2025). IPUMS USA: Version 16.0 [dataset]. Minneapolis, MN: IPUMS. doi:10.18128/D010.V16.0
  10. IPUMS USA (2026). Frequently asked questions: Extract option, attach characteristics. IPUMS USA. usa.ipums.org/usa-action/faq#ques38
  11. Duncan, B.; Trejo, S. J. (2016). The Complexity of Immigrant Generations: Implications for Assessing the Socioeconomic Integration of Hispanics and Asians. NBER Working Paper 21982; published in ILR Review 70(5): 1146–1175 (2017). www.nber.org/papers/w21982
  12. US Census Bureau (2022). Basic CPS Household Nonresponse. CPS technical documentation (4-8-4 rotation pattern; page revised July 26, 2022). www.census.gov/programs-surveys/cps/technical-documentation/methodology/basic-cps-household-nonresponse.html
  13. US Bureau of Labor Statistics (2026). Consumer Price Index for All Urban Consumers (CPI-U), US city average, all items, series CUUR0000SA0, annual averages. US Bureau of Labor Statistics. data.bls.gov/timeseries/CUUR0000SA0

How to cite

@misc{singh2026secondgeneration,
  author = {Singh, Vishal},
  title = {The second generation of South Asian Americans},
  year = {2026},
  url = {https://vishalsingh.org/south-asians/second-generation},
  note = {South Asians in America, chapter 6. Published 7 November 2026.}
}

Acknowledgments

CPS and ACS microdata are from IPUMS CPS and IPUMS USA, University of Minnesota. IPUMS is not responsible for the analysis here.