Migration and Income Flows in South Dakota
Migration is a key component of population change. It is typically more volatile than natural increase and can be a significant indicator of the relative economic strength and attractiveness of places. Income changes due to migration flows can also be a substantial component of overall changes in the state.
Prior columns have detailed trends in South Dakota’s migration, both domestic and international, as they reflect movement among states and among counties in South Dakota. This study has a somewhat different focus, asking about the characteristics of households and individuals moving into and out of South Dakota. The initial view utilizes Internal Revenue Service (IRS) migration data based on matched returns from consecutive years to examine a couple of measures of differences among the inflow and outflow groups. Data from the American Community Survey (ACS) are also examined to provide a more detailed profile of the groups moving to and from South Dakota.
IRS Migration Findings
The IRS data provide extensive coverage by including all tax returns, and its numbers are essential in the larger Census Bureau annual population estimation program. While little detail is included in the data, the information on returns, exemptions, and aggregate adjusted gross income (AGI) allow for creating measures of household size and income exchanges. The 2021–22 South Dakota data are the most recent available and are summarized in Table 1.
2021–22 was not only a strong year for migration; it was also an exceptional year in terms of income differences. The net flow of income was more than $583 million. This is about 1% of total personal income in the 2021–22 period and is equal to 18% of the change in personal income between 2021 and 2022.
Exemptions per return (Table 2) varied among different migration flows, with lower values for domestic interstate and within-state migration and higher values for foreign migration and non-migrants. The patterns held for both inmigrants and outmigrants, with little difference between the inmigrant and outmigrant values for any particular group. The value for all interstate migrants in the U.S. was 1.72—very close to South Dakota’s values.
Income per return, on the other hand, showed great variation both among flow types and between inmigrants and outmigrants (Table 2). Of greatest interest is the large difference of about $20,000 between the inmigrant and outmigrant flows. A similar gap exists between the interstate and within-state flows. Non-migrants averaged about $10,000 higher than the inmigrant group. The average AGI per return was still higher by about $2,000 than the non-migrants.
Comparisons Over Time
Trends over time in migration and aggregate income are shown in Figures 1 and 2. The positive net inmigration period of 2004–05 to 2013–14 was followed by largely offsetting in- and outflows until 2019, when even larger positive net migration returned. Income differences, in nominal dollars, largely follow the migration trend until 2014, when roughly zero net migration still led to net income inflow. Growing net migration has led to even larger net income flows, somewhat exaggerated on the graphs due to the use of nominal dollars.
Exemptions per return have stayed roughly between 1.75 and 1.95 over time (Figure 3). There have been numerous movements up and down over the years with no perceptible trend upward or downward. It has been slightly higher for inmigrants in almost every year, tending toward about 0.05 higher among inmigrants in recent years.
Adjusted Gross Income per return, measured in nominal terms, shows the expected upward trend consistent with price increases (Figure 4). With the exception of 2011–12 and 2012–13, AGI per return has been higher for inmigrants. A growing gap between inmigrant and outmigrant AGI per return is clear in recent years.
Migrant Characteristics from the American Community Survey
The American Community Survey (ACS) is conducted annually. As a survey, it collects data on only a small sample from the population. The Census Bureau reports that coverage is about one percent of the population, so it provides information with significant margins of error. In fact, the 2023 ACS estimates of migration flows between states show numerous examples of margins of error that are nearly the size of the estimated flow. For example, the estimated flow from Montana to South Dakota in 2023 is 680 with a margin of error of 539.
While the individual between-state flow sizes leave significant uncertainty, the overall estimates in and out of South Dakota are similar to IRS estimates, and it is still reasonable to examine the characteristics of individuals reflected in the Public Use Microdata Sample (PUMS) provided by the Census Bureau. The numbers reported below use the weights assigned by the Census Bureau to capture how many individuals each of the observations represents. Since outmigrants to other countries are not included in the survey, only domestic inmigrants were examined to assure comparability among groups.
The 2023 ACS estimated 30,055 inmigrants to South Dakota from other states, 29,464 outmigrants from South Dakota to other states, and 874,969 people who remained in South Dakota. The PUMS data provided 335 outmigrant records, 235 inmigrant records, and 9,214 nonmovers. Although there were many more outmigrants than inmigrants, the weights identified for the observations indicate that approximately equal numbers were represented by the sample.
The three groups had some notable differences in racial composition (Table 4). The percentages White were fairly similar although several percentage points higher among inmigrants than outmigrants. More striking are the differences in the Black and Native American percentages. More than eleven percent of the outmigrants were Black, while the nonmover and inmigrant percentages were both around three percent. The Native American percentages were both over ten percent among the nonmovers and inmigrants but only about six percent among the outmigrants. The Hispanic percentage, while never exceeding eight percent, is notably higher among outmigrants than inmigrants.
Educational attainment (Table 5) was fairly similar among the three groups, with 5–10% of each group over the age of 24 having less than a high school education, 50–60% being high school graduates (possibly with some college or an Associate’s degree), 21–26% having a Bachelor’s degree, 5–8% with a Master’s degree, and 3–8% with more advanced degrees.
Comparisons of the places of birth and citizenship (Table 6) provide additional insight into migration patterns. More than 95% of the nonmovers and inmigrants were born in the United States, while only 91% of the outmigrants were U.S. natives. Virtually none of the inmigrants and only 2% of nonmovers were noncitizens, while 6.1% of the outmigrants were noncitizens.
The ACS provides a number of economic measures (Table 7). Labor force participation rates are similar among the three groups, all with unemployment under 2%. Income and earnings levels are slightly higher among inmigrants than outmigrants, while nonmovers have the highest total income and earnings.
Comparison to 2021 and 2019
To assess change over time, the single-year PUMS files for 2018–19 and 2020–21 were examined. Table 8 summarizes comparable metrics for 2019 and 2021. Patterns indicate COVID-related shifts, rising native-born inmigrants, and declining noncitizen migration.
Conclusion
The limited IRS data indicate that South Dakota is benefiting economically from migration exchanges when measured by adjusted gross income. There is no clear pattern of significant differences in family sizes as proxied by exemptions per return.
The richer ACS data reveal intriguing demographic contrasts among migrant groups. While not all differences have immediate economic implications, they could influence long-term trends. The findings support the IRS evidence of migration’s beneficial effect but should be interpreted with caution due to small sample sizes. The exchange of college students remains a particularly important factor driving net inmigration.
