States and Metro Areas
The vast majority of data that appears for states and metro areas calculated by the New American Economy research team using various publicly available datasets. Primary among these were microdata from the 2014 American Community Survey (ACS), downloaded from the Integrated Public Use Microdata Series (IPUMS) database.1 We define immigrants as individuals born to non-U.S. citizens outside of the United States and U.S. territories.
For metro areas, we use the Office of Management and Budget’s (OMB) 2013 definitions of metropolitan statistical areas (MSA). The OMB defines a metropolitan statistical area as an area that has at least one city of 50,000 or more residents with surrounding areas that have a “high degree of social and economic integration” with the core city or cities such as commuting ties. MSA are made up of counties or county equivalents (e.g. parishes in Louisiana) and one county cannot be divided between metro areas or metro/non-metro areas.
A full list of the MSAs and their component counties can be found here.
Unless specifically stated, the method for calculating data points for states and metro areas is the same, using the same data source.
Data points on the number, share, and age breakdown of both U.S.- and foreign-born residents of each state and metro area were calculated using microdata from the 2014 ACS.
The data on the number of self-employed immigrants and the business income generated by immigrant entrepreneurs come from the 2014 ACS 1-year sample.
The number of employees at immigrant owned firms is estimated by using the 2007 Survey of Business Owners (SBO) Public Use Microdata Sample, which is the most recent microdata on business owners currently available.2 The estimates are weighted using the tabulation weights provided in the dataset. We define immigrant-owned businesses as firms with at least one foreign-born owner. For confidentiality, the data exclude businesses classified as publicly owned firms because they can be easily identified in many states. Based on our own analysis, we believe that many of the publicly owned firms excluded from this data are companies with 500 employees or more. As a result, the final number of employees at immigrant owned companies in this report is a conservative estimate, and is likely lower than the true value.
Fortune Magazine ranks U.S. companies by revenue and publishes a list of top 500 companies and their annual revenue as well as their employment level each year. To produce our estimates, we use the 2015 Fortune 500 list.3 Our estimates in this section build on past work done by PNAE examining each of the Fortune 500 firms in the country in 2011, and determining who founded them.4 We then use publicly available data, including historical U.S. Census records and information obtained directly by the firms, to determine the background of each founder. In the rare cases where we could not determine a founder’s background, we assumed that the individual was U.S.-born to be conservative in our estimates. Some firms created through the merger of a large number of smaller companies or public entities were also excluded from our analysis. These included all companies in the utilities sector and several in insurance.
To produce the Fortune 500 estimates for each state, we allocate firms to the states where their current headquarters are located. We then aggregate and report the annual revenue and employment of the firms in each state that we identify as “New American” Fortune 500 companies. These are firms with at least one founder who was an immigrant or the child of immigrants.
Income and Tax Contributions
As in past NAE briefs, we use the term “spending power.”5 Here and elsewhere we define spending power as the disposable income leftover after subtracting federal, state, and local taxes from household income.
Using the 2014 ACS 1-year microdata sample, we estimate the aggregate household income, tax contributions, and spending power of foreign-born households. We estimate state and local taxes using the tax rates estimates produced by Institute on Taxation and Economic Policy by state income quintiles.6 For federal tax rate estimates, we use data released by the Congressional Budget Office in 2014 and calculate the federal tax based on the household income federal tax bracket.7
Educational attainment figures for both U.S.- and foreign-born individuals ages 25 and older were calculated using microdata from the 2014 ACS.
For states, we use the North American Industry Classification System (NAICS), to classify and estimate the industries with the largest share of foreign-born workers. All individuals 16 years old and above are included in these calculations. The total number of workers for certain industries in some states is extremely small, thus giving unreliable results. In order to avoid this, we calculate the percentile distribution of the total number of workers per industry per state and drop the industries in each that fall below a certain threshold. Estimated occupations with the largest share of foreign-born workers per state also follow the same restrictions—the universe is restricted to workers age 16 and above, and the occupations per state that fall under minimum threshold excluded.
For metro areas, we use broad categories of industries—sometimes called the “2-digit” NAICS codes, to determine which industry groups have the highest share of immigrant workers. Again, similar to states, we drop particularly small industry groups as to avoid giving estimates for industries with small sample sizes and inaccurate results.
We use the STEM occupation list released by U.S. Census Bureau to determine the number and share of foreign-born STEM workers as well as the number of unemployed STEM workers from 2014 ACS 1-year data.8 Per U.S. Census classification, healthcare workers such as physicians and dentists are not counted as working in the STEM occupations. All unemployed workers who list their previous job as a STEM occupation are counted as unemployed STEM workers. The data on STEM graduates are from the 2014 Integrated Postsecondary Education Data System (IPEDS) completion survey.9
We estimate the number of unemployed healthcare workers using the 2014 ACS 1-year sample. Healthcare workers are healthcare practitioners and technical occupations, or healthcare support occupations as defined by U.S. Census Bureau.10 Unemployed healthcare workers are individuals who report their previous job as a healthcare occupation, and their employment status as currently not working but looking for work. We took the number of job postings for healthcare workers from the Burning Glass Labor Insight tool, a database that scours online sources and identifies the number and types of job postings. We describe this resource in detail in the section on STEM methodology.
We then delve into specific occupations within the broader healthcare industry. To produce the figures on the total number of physicians and psychiatrists and the share educated abroad, we use American Medical Association (AMA) Physician Masterfile data.
For the share of foreign-born nurses and home health aides, we use the 2014 ACS 5-year sample data because data from the 1-year sample is too small to make reliable estimates.
We access the agriculture GDP by state from Bureau of Economic Analysis, which includes GDP contributions from the agriculture, forestry, fishing, and hunting industry.11 The share of foreign-born agricultural workers is estimated using 2014 ACS 1-year sample. The Quarterly Census of Employment and Wage (QCEW) was used to estimate the percentage of crop farms producing fresh fruits and vegetables.
The data in the housing section comes from the 2014 ACS 1-year sample. Immigrant homeowners are defined as foreign-born householders who reported living in their own home. We estimate the amount of housing wealth held by immigrant households by aggregating the total housing value of homes owned by immigrant–led households. We also estimate the amount of rent paid by immigrant-led households by aggregating the rent paid by such families. We then calculate the share of housing wealth and rent that immigrant households held or paid compared to the total population. For characteristics of homeowners, a foreign-born new homebuyer is defined as a household with a foreign-born household head who owned and moved to the current residence within the last five years.
We obtain the size and share of postsecondary students who are international in each state from the 2014 Integrated Postsecondary Education Data System (IPEDS) fall enrollment data. Those figures are then applied to preexisting work previously done by NAFSA, an organization representing professionals employed in the international offices of colleges and universities across the United States. NAFSA has developed an economic value tool and methodology that estimates the total economic benefit and jobs created or supported by international students and their dependents in each state.12 The economic contributions include the costs of higher education along with living expenses minus U.S.-based financial support that international students receive.
Because the enrollment data from IPEDS that we use is different from the underlying data used by NAFSA, our figures differ slightly from the NAFSA estimates of the economic contributions made by international students in the 2014-2015 school year.
The estimates for registration rates and active voters for native-born and foreign-born population are calculated from the Voter Supplement in the Current Population Survey (CPS) for the years 2008, 2010, 2012, and 2014 using the IPUMS database. The sample in CPS includes only civilian non-institutional persons. Foreign-born individuals who stated having voted between 2008 and 2014 are termed active voters.
Using data from the 2014 ACS 1-year sample, we estimate the number and share of foreign-born eligible voters. We define them as naturalized citizens aged 18 or older who live in housing units. Persons living in institutional group quarters such as correctional facilities and non- institutional group quarters such as residential treatment facilities for adults are excluded from the estimation. We also estimate the number of new foreign-born voters who will become eligible to vote in 2016 and 2020, either by turning 18 or through naturalization, as well as the total number of foreign-born voters in these years. The estimates of newly eligible voters for 2016 include naturalized citizen ages 16 and 17 as of 2014 (thereby becoming of voting age by 2016). Those eligible to vote in 2020 include all naturalized citizens ages 12-17 in 2014. Applicable mortality rates are also applied.13 In addition, we estimate newly naturalized citizens using data from the Department of Homeland Security, which show the two-year average of new naturalized citizens by state.14 We discount from these numbers the percentage of children below 18 in households with a naturalized householder by state. Estimates of total foreign-born voters include naturalized citizens aged 18 or older in 2014, discounted by average U.S. mortality rates by age brackets, summed to the pool of newly eligible foreign-born voters.
Margin of victory in 2012 refers to President Barack Obama’s margin of victory over Republican candidate Mitt Romney in terms of popular vote. The margins are negative in states that Romney won in 2012.15
Using data from the 2014 ACS, we applied the methodological approach outlined by Harvard University economist George Borjas to arrive at an estimate of the undocumented immigrant population in the overall United States and individual states.16 The foreign-born population is adjusted for misreporting in two ways. Foreign-born individuals who reported naturalization are reclassified as non-naturalized if the individual had resided in the United States for less than six years (as of 2014) or, if married to a U.S. citizen, for less than three years. We use the following criteria to code foreign-born individuals as legal U.S. residents:
- Arrived in the U.S. before 1980
- Citizens and children less than 18 years old reporting that at least one U.S.-born parent
- Recipients of Social Security benefits, SSI, Medicaid, Medicare, military insurance, or public assistance
- Households with at least one citizen that received SNAP benefits
- People in the Armed Forces and veterans
- People attending college or graduate school
- Working in occupations requiring a license
- Government employees, and people working in the public administration sector
- Any of the above conditions applies to the householder’s spouse
The remainder of the foreign-born population that do not meet these criteria are reclassified as undocumented. Estimates regarding the economic contribution of undocumented immigrants and the role they play in various industries are made using the same methods used to capture this information for the broader immigrant population in the broader brief. When estimating the aggregate household income, spending power, and tax contributions, we are not able to make reliable estimates for undocumented-led households in Alaska, Maine, Montana, North Dakota, South Dakota, Vermont, and West Virginia due to the small sample size of undocumented-led households in ACS.
Due to the irregular nature of U.S. congressional districts, analysis of each district was not possible using microdata from the 2014 5-Year American Community Survey. Instead, we use estimates for hundreds of demographic and socioeconomic indicators published by the U.S. Census Bureau on the American FactFinder data portal. Using these estimates, we calculate all of the demographic and socioeconomic characteristics of each congressional district’s U.S.-born and immigrant population as listed on each district page. Income, tax, and spending power figures were calculated using average household income estimates from American FactFinder and then weighted by congressional district to equal the total income, tax, and spending power figures calculated from microdata for each state. For Districts-At-Large, state data on household income, taxes, and spending power are used instead since the geographies are coterminous.
1 Steven Ruggles, Katie Genadek, Ronald Goeken, Josiah Grover, and Matthew Sobek. Integrated Public Use Microdata Series: Version 6.0 [Machine-readable database]. Minneapolis: University of Minnesota, 2015.
2 U.S. Census Bureau, Survey of Business Owners and Self-Employed Persons Datasets. Available online.
3 “Fortune 500,” Fortune, 2015. Available online.
4 “The ‘New American’ Fortune 500,” Partnership for a New American Economy, 2011. Available online.
5 “The Power of the Purse: The Contribution of Hispanics to America’s Spending Power and Tax Revenues in 2013,” Partnership for a New American Economy, 2014. Available online.
6 “Who Pays? A Distributional Analysis of the Tax Systems in All 50 States (5th edition),” Institute on Taxation and Economic Policy, 2014. Available online.
7 “The Distribution of Household Income and Federal Taxes, 2011”, Congressional Budget Office, Washington, D.C., 2014. Available online.
8 U.S. Census Bureau, “STEM, STEM-related, and Non-STEM Occupation Code List 2010,”
2010. Available online.
9 National Center for Education Statistics, “Integrated Postsecondary Education Data System.” Available online.
10 U.S. Census Bureau. “2010 Occupation Code List.” Available online.
11 U.S. Department of Commerce, “Gross Domestic Product (GDP) by State,” accessed July 2, 2016. Available online.
12 NAFSA, “International Student Economic Value Tool.” Available online.
13 U.S. Department of Health and Human Services, “National Vital Statistics Reports, Deaths: Final Data for 2013”, 2016. Available online.
14 Department of Homeland Security, “Yearbook of Immigration Statistics: 2014 Naturalizations, Table 22 – Persons Naturalized by State or Territory of Residence: FY 2005 to 2014.” Available online.
15 Federal Election Commission. “Federal Elections 2012: Elections for the President, the U.S. Senate and the U.S. Representatives”, 2013. Available online.
16 George J. Borjas, “The Labor Supply of Undocumented Immigrants,” NBER Working Paper (National Bureau of Economic Research, Inc, 2016). Available online.