Population projections for U.S. counties by age, sex, and race controlled to shared socioeconomic pathway

Abstract

Small area and subnational population projections are important for understanding long-term demographic changes. I provide county-level population projections by age, sex, and race in five-year intervals for the period 2020-2100 for all U.S. counties. Using historic U.S. census data in temporally rectified county boundaries and race groups for the period 1990-2015, I calculate cohort-change ratios (CCRs) and cohort-change differences (CCDs) for eighteen five-year age groups (0-85+ ), two sex groups (Male and Female), and four race groups (White NH, Black NH, Other NH, Hispanic) for all U.S counties. I then project these CCRs/CCDs using ARIMA models as inputs into Leslie matrix population projection models and control the projections to the Shared Socioeconomic Pathways. I validate the methods using ex-post facto evaluations using data from 1969-2000 to project 2000-2015. My results are reasonably accurate for this period. These data have numerous potential uses and can serve as inputs for addressing questions involving sub-national demographic change in the United States.

Figure 1. Lexis Diagrams for CCRs and CCDs.

(a) Demonstrates the general framework for Cohort-change ratios and (b) the general framework for cohort-change differences using a toy example. The “observed” populations are in bold while the projected populations are italicized.

Figure 2. The five Shared Socioeconomic Pathways (SSPs).

Adapted from. (a) Shows the relationship between mitigation and adaptation and the five SSPs while (b) shows the projected populations under the five SSPs.

Figure 3. Map of county errors of the total population in 2015 using the CCD/CCR model.

This figure shows the geographic distribution of absolute percent errors. Most states and counties have low error rates of the total population with isolated pockets of large errors. The missing counties in Colorado are due to geographic boundary changes associated with the creation of Broomfield County in 2001.

Figure 4. Age structures of various county types.

This figure compares the projected age structures to the observed age structures in twelve counties across four county types using the CCD/CCR model. (a) Demonstrates counties with major universities, (b) demonstrates sample suburban counties, (c) demonstrates sample retirement counties, and (d) demonstrates sample counties with large cities. All four county types have age structures largely preserved despite widely different age structures.

Figure 5. Errors by age group.

This figure plots the Median Algebraic Percent Error (ALPE) by age group (a) and the Mean Absolute Percent Error by age group (b).

Figure 6. Race group errors.

(a) Shows the Algebraic Percent Errors for all three methods and (b) shows the APE distribution of errors.

Figure 7. Projected numeric and percentage population changes for the five SSPs between 2020 and 2100 for counties in the continental United States.

AK and HI are available in the final projections but are excluded from these maps due to space considerations and to improve interpretability. (a) Shows the projected numeric change and (b) shows the projected percentage change.

Figure 8. Comparisons to various State-level Population Projections.

Several states produce timely population projections. This figure compares six states’ independent population projections to mine produced here. All state-level projections are the black dotted lines. Texas, Alaska, and Arizona include projections of uncertainty and their uncertainty (high, medium, low scenarios) is displayed as the gray shaded area on the respective panels.