Middle Eastern or North African in U.S. Government Surveys: A Preview of MENA Demographics

This brief approximates responses to the proposed MENA question, as well as the creation of a new Hispanic or Latino race category, using ancestry and country of origin responses from the American Community Survey (ACS). I find a total MENA population of 3.8 million in 2021—about 1 percent of the total U.S. population. The MENA population is most demographically similar to the Non-Hispanic White and Asian populations in terms of age, earnings, education, and occupation. Specifically, the MENA population is more highly educated and has higher earnings than Non-Hispanic Whites, but lower levels of education and earnings than the Asian population. (This brief uppercases races and ethnicities to be consistent with U.S. government stylistic guidelines.)

Statistics on race and ethnicity in the United States represent the self-identified background of respondents to censuses and surveys according to standard categories determined by the OMB with assistance from the U.S. Census Bureau. Current race and ethnicity definitions were set in a two-question format in 1997 by the OMB through its Statistical Policy Directive No. 15: Standards for Maintaining, Collecting, and Presenting Federal Data on Race and Ethnicity (SPD 15).² The OMB recently proposed substantial revisions to the 1997 standards for collecting and presenting federal data related to race and ethnicity. The proposed revisions consist of three primary changes to data collection:³

1. Collapse race and ethnicity into a single query that adds Hispanic or Latino origin as a separate race category.
2. Introduce a separate race category for those of Middle Eastern or North African (MENA) descent.
3. Obtain by default more detailed racial breakouts within the updated racial categories.

While the current standards allow a wide range of race and ethnicity combinations, including multiple racial identities, there is a debate among experts as to whether the current standards require updating. The OMB asserts that the existing standards need revision due to significant demographic shifts in the American population since 1997. These changes include the growing population of individuals who identify as multiracial, shifts in the overall demographic landscape, and the addition of new national-origin groups through immigration over the past several decades.⁴ Similarly, civil rights groups argue for expanding the scope of data collected on race and ethnicity to ensure compliance with civil rights laws and to prevent discriminatory practices. They believe more comprehensive data collection is necessary to address disparities and provide equitable treatment.⁵

Some critics contend that updating the racial standards in government surveys is unnecessary from a methodological standpoint and may decrease data quality. They argue that broadening the categories could lead to confusion and inaccurate reporting, particularly for the White and Hispanic or Latino groups.⁶ Opponents also express concerns that the changes may have cultural and legal implications, potentially creating divisions and conflicts.⁷ There are apprehensions that the new standards may extend beyond statistical purposes and be utilized to allocate federal program eligibility based on the interests of specific groups rather than objective criteria.⁸ Although the OMB states that the standards “are not to be used for determining the eligibility of population groups for participation in Federal programs,” there is mixed historical evidence that the data are used in practice for rewarding government grants and other benefits.⁹

An uncontested demographic change over the past two decades is that the population identifying as multiracial increased dramatically since 2010, according to the 2020 enumeration. In particular, the share of people identifying as “Some Other Race” in combination with another race increased by 733 percent since 2010.¹⁰ Considering the breadth of the White category, which includes those with MENA ancestry, the population identifying as both White and Some Other Race increased by over 1,000 percent over the same time frame. In comparison, the share of the Hispanic or Latino population identifying as White alone decreased by 53 percent over the decade.¹¹ An issue with these comparisons is that it’s unclear whether this shift results from changes in demographics or data processing. For the 2020 census, the Census Bureau implemented various improvements to the design and coding of the race and ethnicity questions in the census questionnaire.¹² Therefore, whether the new statistical standards will reflect these demographic changes is uncertain without further data from the 2020 enumeration.

There has been no attempt to model how the proposed changes will manifest in the data. Effectively weighing the benefits and costs of adopting the new statistical standards for race and ethnicity requires a counterfactual to quantify how the new standard might affect the demographic landscape of the U.S. population. This brief will jointly evaluate the addition of Hispanic or Latino as a separate race category and the creation of a separate race category for those of MENA descent by applying these new proposed standards to existing public-use data from the ACS.¹³ Since the MENA population is not officially recorded in the ACS, information on respondents’ ancestry and birthplace provides a useful proxy for the likely MENA population present in the ACS. The resulting data provide a useful counterfactual demographic profile of the United States if the OMB’s revisions to race and ethnicity were implemented. I was not able to implement the third proposed change—more detailed racial breakouts within the updated racial categories—but the elevation of Hispanic or Latino to a new race category and the creation of the MENA race category are the most consequential changes that could also affect each other, while more detailed racial breakdowns in each racial category shouldn’t affect the topline results.

Since the 1970s, federal statistical standards in the United States have included race and ethnicity information. In 1977, the OMB introduced Statistical Policy Directive No. 15, which established the initial standardized approach to categorizing race and ethnicity.¹⁴ The aim was to create a minimum set of designations that would facilitate data reporting for civil rights compliance, program administration, grant-making, and tabulating statistical data.¹⁵

An important issue with any categorization of race and ethnicity is that the definitions themselves “are sociopolitical constructs and are not an attempt to define race and ethnicity biologically or genetically” and are inherently “dynamic and fluid.”¹⁶ As such, the original OMB classifications aimed to develop a minimum set of categories that could be consistently used across different data sources such as the decennial census, surveys, and federal forms. These categories needed to be broad enough to allow respondents to accurately report their racial and ethnic identity in a way that could be reaggregated into the minimum set of standardized groups.

Box 1 shows the original 1977 race and ethnicity category designations.¹⁷ The OMB subsequently revised the designations to form the current standards in the 1997 SPD 15, which follows an expanded two-question format.¹⁸ The updated proposed minimum categories combine two ethnic and five racial categories.¹⁹

In addition to the minimum categories, respondents to the census and many other federal surveys such as the ACS are also asked about their country of origin and ancestry. For example, the 2020 census questionnaire allowed respondents to write in their self-identified origin. In the context of MENA, one may mark “White” and report their origin and ancestry as Egyptian. Since the Census Bureau has not yet released decennial data for these self-reported ancestries, evaluating the introduction of a MENA category requires additional information to develop a counterfactual MENA population to compare with the existing standards.

Existing research focuses primarily on the foreign-born MENA population in the United States. A 2015 report by the Migration Policy Institute highlights key socioeconomic statistics of the foreign-born MENA population using data from the 2013 ACS.²⁰ The report shows that the MENA population exhibits higher education levels than the foreign-born population as a whole and the native-born population. Many individuals within the MENA population have obtained at least a bachelor’s degree or higher educational qualifications. Regarding occupational patterns, the report indicates that the foreign-born MENA population is disproportionately represented in management, business, the sciences, the arts, and sales occupations. This finding suggests a tendency toward employment in professional and managerial roles within various sectors. A more recent 2019 report by the Partnership for a New American Economy using data from the 2015 ACS expands this research and corroborates the high educational attainment and occupational tendencies of the MENA population.²¹

Overall, existing research highlights the strong educational profile of the MENA population in the United States and their concentration in professional occupations. Such findings contribute to our understanding of the socioeconomic characteristics and contributions of the MENA population within the larger context of immigration and the U.S. labor market. However, these studies consider only the foreign-born MENA population. The subsequent empirical work in this brief expands the analysis to include a broader measure of the entire MENA population in the United States.

This brief uses data from the most recent 2021 ACS public-use microdata sample (PUMS) produced by the Census Bureau and disseminated by the Integrated Public Use Microdata Series (IPUMS) database.²² The ACS, a large-scale survey designed to replace the long-form decennial census, collects a wide range of information on respondents’ demographic, economic, and housing characteristics, sampling around 3 million respondents on a rolling basis each year. The ACS also includes detailed demographic information on race, ethnicity, and ancestry that can be used to measure the potential MENA population.

While the ACS already has detailed information on race and respondents of Hispanic or Latino origin, additional information is needed to delineate respondents with a MENA background. Combining the responses to ancestry and birthplace questions provides a proxy for respondents likely to self-identify as MENA in the proposed race and ethnicity question revision. Respondents in the ACS microdata are classified as MENA if they either report ancestry from the MENA countries or were born in a MENA country. Research by the Migration Policy Institute uses a similar method to describe the MENA diaspora, defined as those who “self-identify as having ancestry ties to a specific country of origin.”²³ This classification is best suited to describe the likely MENA population universe, as the respondents share a connection to the region either by ancestral origin or birthplace. Table A1 contains the detailed ancestry and birthplace codes used to construct the MENA category.²⁴

Using the wealth of socioeconomic information in the ACS, the resulting data provide age, sex, employment status, income, and educational attainment, among other things. Applying the new MENA category and integrating the Hispanic or Latino origin information creates a useful counterfactual demographic composition, as suggested by the OMB proposal. Each tabulation is weighted using either person or household weights, as noted in each table header, and represents selected statistics from the official Data Profile tables published by the Census Bureau. The Data Profile tables summarize commonly requested socioeconomic indicators from the ACS.²⁵ Margins of error (MOEs) for each estimate represent sampling variability and reliability for a 90 percent confidence interval. These are computed using the successive difference replication (SDR) method. A larger margin of error relative to the estimate indicates a lower reliability for that estimate.²⁶

The 2021 ACS shows a total of 3.8 million people with either a Middle Eastern or North African ancestry or birthplace—around 1 percent of the U.S. population.²⁷ Table 1 shows general social and demographic data for the MENA population, and for the other race and ethnicity categories adjusted by subtracting respondents classified as MENA in 2021. These data include age, educational attainment, nativity and citizenship status, and English-language ability, among others. Economic characteristics such as employment status and income are shown in Table 2. Each table shows the estimate and its margin of error.

The MENA population leans slightly toward men, with around 150,000 more men than women. The sex ratio between men and women is 1.08 men per woman, compared to a ratio of 0.98 men per woman nationally.

Individuals with a MENA background are relatively young compared to the general population. The working-age (ages 20–64) MENA population represents 62.7 percent of the total MENA population, compared to 57.8 percent of the White population and 58.3 percent nationwide. Compared to other race and ethnicity categories, the working-age MENA population aligns more closely with Black and Asian populations, with respective working-age populations of 60.3 and 64.5 percent. The median age for the MENA population is 37 years, aligning closely with the Asian population at 38 years. Comparatively, the White population is older, with a median age of 44 years, and the Hispanic or Latino population is much younger, with a median age of 30.

The MENA resident population has a very high level of educational attainment compared to the U.S. population as a whole. About 53.2 percent of the MENA population ages 25 and over holds a bachelor’s degree or higher, compared to 35 percent nationwide. Compared to other race categories, the MENA population has higher educational attainment than the White population at 38.6 percent and just below the Asian population at 56.8 percent.

MENA immigrants were 4.9 percent of the total foreign-born population in 2021, with a total of 2.2 million MENA immigrants. About 58.1 percent of the MENA population residing in the United States were foreign-born. MENA immigrants are more likely to be naturalized citizens than the overall foreign-born population. Nearly 70 percent of MENA immigrants were naturalized citizens, compared to 53 percent of the total foreign-born population.

In 2021, around 2 million MENA individuals were engaged in the civilian labor force (ages 16 and over). Their labor-force participation rate is 65 percent, slightly above the national rate of 64 percent, while their unemployment rate was relatively high (8.8 percent) compared to the nationwide unemployment rate of 6.7 percent.

MENA employment was concentrated in management, business, science, and arts occupations, with around 54 percent of employed workers ages 16 and over. In comparison, the share of Whites and Asians in the same occupations was 47 percent and 58 percent, respectively.

MENA households had relatively high incomes in 2021. The median income of MENA households is $74,000 and the average is $115,000. The MENA household income is higher than that of White households, which have a median income of $73,000 and a mean income of about $102,000. Similarly, median and mean household income for Asians was $99,000 and about $134,000, respectively.

The poverty rate for the MENA population was around 17 percent, 4 percent higher than the nationwide poverty rate of 12.9 percent.

Using ancestry and birthplace to estimate the likely MENA population shows a total of around 3.8 million MENA residents in the United States in 2021. Over half of the MENA population is foreign-born; most are naturalized citizens. MENA labor-force participation, educational attainment, and earnings often exceed the nationwide population average. In other words, the MENA population boasts strong economic and educational standing, higher than that of White Americans and below that of Asian Americans.

Data presented in this brief provide a first empirical estimation of the MENA population as proposed by the OMB and are based on a broad, diasporic interpretation of the potential MENA population. The data are especially valuable for policymakers and researchers to determine how the OMB proposal might appear in the data and whether they accurately reflect the diversity of the American public.

The data used in this brief are sourced from the 2021 ACS 1‑year estimates published by the Integrated Public Use Microdata Series (IPUMS) database. The microdata were tabulated for each population universe using the person weights for person-level statistics and the household weights for household-level statistics.

Margins of error are calculated using the successive difference replication (SDR) method per Census Bureau guidance.²⁸ The variance estimates from the SDR method are computed by taking the sum of squared differences between a replicate estimate and the full-sample-weighted estimate.

Letting x₀ denote the full-sample estimate and x_r (r = 1, …, 80) denote the replicate estimate for each replication, the variance of the full-sample estimate is as follows:

Each replicate estimate x_r creates a replicate data set by applying randomization techniques to the original ACS public-use sample that account for the survey’s complex design. Each replicate data set represents a plausible alternative version of the original or subset while preserving its design features and sampling characteristics. The variance of the full-sample statistic can then be estimated by taking the difference between each replicate subsample and the full-sample estimate for each replication.

The margin of error for each estimate, assuming a 90 percent confidence level, is therefore