defining the “rural” in rural...

28

AM

BE

R W

AV

ES

ECONOMIC RESEARCH SERVICE/USDA

F E A T U R E

Defining the “Rural”in Rural America

John Cromartie [email protected]

Shawn Bucholtz

VO

LU

ME

6 �

ISS

UE

3

Ken Hammond, USDA

29

AM

BE

R W

AV

ES

F E A T U R E

JU

NE

20

08

The term “rural” conjures widely shared images of farms,ranches, villages, small towns, and open spaces. Yet, when itcomes to distinguishing rural from urban places, researchers andpolicymakers employ a dizzying array of definitions. The use ofmultiple definitions reflects the reality that rural and urban aremultidimensional concepts, making clear-cut distinctionsbetween the two difficult. Is population density the definingconcern, or is it geographic isolation? Is it small population sizethat makes it necessary to distinguish rural from urban? If so,how small is rural? Because the U.S. is a nation in which so manypeople live in areas that are not clearly rural or urban, seeminglysmall changes in the way rural areas are defined can have largeimpacts on who and what are considered rural.

Researchers and policymakers share the task of choosingappropriately from among the more than two dozen rural defi-nitions currently used by Federal agencies. For example,research on suburban development and its effect on rural realestate prices would probably define rural differently than a

study designed to track and explain economic and socialchanges affecting rural people and places. Programs developedto address the unique problems that small rural governmentsface will not necessarily target the same rural areas as will pro-grams that are developed to help rural businesses operating incredit-constrained markets. The key is to use a rural-urban def-inition that best fits the needs of a specific activity, recognizingthat any simple dichotomy hides a complex rural-urban contin-uum, with very gentle gradations from one level to the next.

Delineating a precise line between rural and urbanAmerica that best serves the purpose, given the complexity oftoday’s settlement patterns, involves answering two questions:

1. Is a given urban entity defined in terms of its administra-tive boundaries, its land-use patterns, or its economicinfluence?

2. What is the minimum population size for an entity to beconsidered urban?

� The share of the U.S. population considered rural rangesfrom 17 to 49 percent depending on the definition used..

� Rural definitions can be based on administrative, land-use, or economic concepts, exhibiting considerablevariation in socioeconomic characteristics and well-beingof the measured population.

� For research projects and economic development pro-grams alike, the appropriate definition of rural will be thatwhich meets the goals of the endeavor.

WWW.ERS.USDA.GOV/AMBERWAVES

By identifying urban areas first, ruralis defined as the territory that is notincluded. Good decisionmaking in choos-ing an appropriate rural definitionrequires an understanding of the key char-acteristics of urban entities and how they,in turn, determine the characteristics ofrural definitions derived from them.

Challenge Number One:Choosing an Appropriate UrbanBoundary

There are three different concepts of“urban” that lead to very different bound-ary definitions, and thus to very differentrural definitions:

• The administrative concept, used bymany USDA rural development pro-grams, defines urban along municipalor other jurisdictional boundaries.

• The land-use concept, used by theCensus Bureau, identifies urban areasbased on how densely settled the areais—the picture of settlement you getfrom an airplane.

• The economic concept, used in mostrural research applications, recog-

nizes the influence of cities on labor,trade, and media markets that extendwell beyond densely settled cores toinclude broader “commuting areas.”

These three concepts represent pro-gressively expansive urban boundariesthat differ considerably from one another.

For instance, in 2000, Peoria, IL, as definedby its municipal boundaries, encompassedan area with about 113,000 people but asan economic entity it included nearly367,000 people. Applying a land-use con-cept resulted in an area with a populationbetween these two alternatives.

30

AM

BE

R W

AV

ES


VO

LU

ME

6 �

ISS

UE

3

F E A T U R E

Source: USDA, Economic Research Service, using data from the U.S. Census Bureau.

Three ways to define Peoria

Administrative concept: Peoria City, population 112,936

Land-use concept: Peoria Urban Area, population 247,172

Economic concept: Peoria Metro Area, population 366,899

Question 1:For any given urban entity, where is its boundary?

Peoria Metro Area

Illinois

Tim McCabe, USDA

For rural development programs thatprovide assistance to or through local gov-ernments, an administrative definition ofrural is often a starting point for determin-ing program eligibility. On the other hand,infrastructure programs meant to over-come the disadvantages sparsely populatedareas face in providing water and sewerservices may find rural definitions derivedfrom the land-use concept helpful in target-ing assistance. For programs requiring thecoordination of efforts within broader mar-ket areas, such as area-wide transportationplanning assistance, a definition based oneconomic concepts may be more appropri-ate. So ubiquitous are county-level data thatresearchers often divide urban and ruralareas along county lines, making “non-metro” the de facto economic definition ofrural for most research purposes (see box,“How Are the Boundaries Between Ruraland Urban Developed?”).

Challenge Number Two:Choosing a Population SizeThreshold

In addition to being defined as thearea outside urban boundaries—deter-

mined in different ways depending on theconcept—rural includes some set oftowns and villages below a chosen popula-tion threshold. For the 1910 Census, ruralmeant open countryside and any placewith fewer than 2,500 people. Though theCensus Bureau modified its definitionover the decades to keep up with subur-ban expansion, it did not change the 2,500population threshold as the minimumsize for urban places. Over the sameperiod, thresholds for some USDA ruraldevelopment programs were adjustedupward, arguably an appropriate responseto rapid urbanization. For example, theRural Housing Program began in 1949,serving communities with fewer than2,500 people, but it now sets eligibility atless than 20,000 people.

Proponents of a higher thresholdpoint out that towns of 2,500 people typi-cally have not maintained the levels anddiversity of employment, goods, and serv-ices that existed in 1910. The tremendoustransportation and communicationadvances of the past 100 years helpedreorganize economic and social activitiesaround larger towns and cities. The debate

over an appropriate population sizethreshold between rural and urban placesis ongoing. Definitions used by Federalagencies use population-size thresholdsranging from 2,500 to 50,000 people. Forinstance, the definition of rural used forUSDA’s Community Facilities programsconsists of territory outside Census placesof 20,000 or more. In contrast, the defini-tion of nonmetro areas used by mostresearchers applies a 50,000 populationthreshold.

Different Definitions Mean BigDifferences in Rural Populations

Depending on the boundary choiceand the population threshold, the share ofthe U.S. population defined as rural andits socioeconomic characteristics vary sub-stantially. In 2000, 21 percent of the U.S.population was designated rural using theCensus Bureau’s land-use definition (out-side urban areas of 2,500 or more people),compared with 17 percent for economi-cally based nonmetro areas (outside metroareas of 50,000 or more).

However, alternative definitionsincrease that range from 7 to 49 percent.

31

AM

BE

R W

AV

ES


JU

NE

20

08

F E A T U R E


Depending on the definition, population size thresholds range from 2,500 to 50,000 people

Dublin, GA Richmond, IN

Saginaw, MI

2,195 2,728 15,857 39,124 61,799

Buena Vista, CO

Bridgeport, AL

Always rural Sometimes urban, sometimes rural Always urban

Question 2:What is the minimum population size for an entity to be considered urban?

Raising the population size threshold forthe land-use definition from 2,500 to50,000 increases the rural population from21 to 32 percent. Lowering the thresholdfor the economic definition from 50,000 to10,000 decreases the rural populationfrom 17 to 7 percent.

Holding the population threshold con-stant at the minimum level of 2,500 people

but moving from an administrative to aland-use definition drops the U.S. ruralpopulation by a third, from 31 to 21 per-cent. This change represents a shift in thedesignation of people who live in areas typ-ically described as suburban, who arecounted as rural under the narroweradministrative concept but as urban underthe land-use version. A similar shift in sub-

urban population occurs at the upper pop-ulation threshold of 50,000, where ruralpopulation decreases from 32 percentbased on the Census Bureau’s land-use con-cept to 17 percent under the economicdefinition. (For descriptions of each defini-tion, see box, “How Are the BoundariesBetween Rural and Urban Developed?”)

Alternating the definition of ruralalso varies the socioeconomic characteris-tics of designated areas. Rural populationsconsistently show lower education andincome levels than the overall U.S. popula-tion, however they are defined. Given thatrural definitions based on administrativeboundaries include larger shares of whatcould be classed as suburban areas, theeducation and income levels of their pop-ulations are closer to those of the generalU.S. population. The suburban populationcounted as rural is much smaller for theeconomic definition represented by non-metro areas. Thus, the share of the ruralpopulation with a college degree dropsfrom 28 to 18 percent across this rangeand the average household income dropsfrom $56,000 to $40,000.

Multiple Measures of RuralServe Multiple Purposes

Rural definitions are not limited to theoptions discussed here. For instance, ERSprovides an alternative to OMB’s metro andmicro definitions that uses census tractsinstead of counties (see box: “What Are theRUCA Codes?”). In the 2002 farm b-ill,administrative and land-use concepts wereessentially combined for a rural definitionadopted by several USDA funding pro-grams. Eligible territory includes areasoutside Census places of 50,000 or moreand their adjacent urban areas.

With so many options, which defini-tion is best? The choice of a rural definitionshould be based on the purpose of the activ-ity. For instance, analyzing the effect ofpopulation loss on per capita fiscal costs forrural communities is best approached using

F E A T U R E

VO

LU

ME

6 �

ISS

UE

3A

MB

ER

WA

VE

S

32


Rural population size and characteristics vary by definition

Rural defined as territory outside

Census Bureau Census Bureau OMB metro andplaces urban areas micro areas

(administrative) (land-use) (economic)

with populations less than:

U.S. total 2,500 20,000 2,500 50,000 10,000 50,000

Population, 2000 (millions) 281.0 87.7 138.5 59.1 89.5 19.9 48.8

Percent of populationdefined as rural na 31.1 49.2 21.0 31.8 7.1 17.4

Percent with acollege degree 30.7 26.8 28.3 22.5 22.9 18.5 20.8

Average householdincome ($1,000) 57.0 56.0 56.0 51.0 49.0 40.0 43.0

na = not applicable.

Source: USDA, Economic Research Service using data from the U.S. Census Bureau’s 2000decennial census.

What Are the RUCA Codes?

Counties are often too large, especially in Western States, to accurately represent labormarket areas in all cases. Thus, metro and micro areas often include territory that islegitimately rural from both a land-use and economic perspective. ERS Rural-UrbanCommuting Area (RUCA) codes provide an alternative, economic classification usingcensus tracts rather than counties. Although relatively new, these codes have beenwidely adopted for both research and policy, especially in rural health applications.

RUCA codes follow (as closely as possible) the same concepts and criteria used todefine metro and micro areas. By using the more detailed census tracts, they provide adifferent geographic pattern of settlement classification. While counties are generallytoo large to delineate labor market areas below the 10,000 population threshold,RUCA codes identify such areas for towns with populations as small as 2,500.Additional information and files containing the codes are available in the ERS MeasuringRurality Briefing Room: www.ers.usda.gov/briefing/rurality/ruralurban commutingareas/

F E A T U R E

JU

NE

20

08


AM

BE

R W

AV

ES

33

How Are the Boundaries Between Rural and Urban Developed?

Rural definitions based on the administrative concept start with

the Census Bureau’s list of “places.” Most places listed in the

2000 Census are incorporated entities with legally prescribed

boundaries (e.g., Peoria City), but some are locally recognized,

unincorporated communities. Rural is defined as territory out-

side these place boundaries, together with places smaller than

a selected population threshold. For example, USDA’s Telecom

Hardship Loan Program defines rural as any area outside

Census places of 5,000 or more people.

Rural definitions based on the land-use concept most often

start with the Census Bureau’s set of urban areas, consisting of

densely settled territory. Rural as defined by the Census Bureau

includes open countryside and settlements with fewer than

2,500 residents. Urban areas are specifically designed to cap-

ture densely settled territory regardless of where municipal

boundaries are drawn. They include adjacent suburbs that are

outside place boundaries and exclude any territory within

places that does not meet the density criteria.

The most widely used rural definition based on the economic

concept consists of the 2,050 nonmetropolitan (nonmetro)

counties lying outside metro boundaries. Metropolitan (metro)

areas are county-based entities that account for the economic

influence of cities. The Office of Management and Budget

(OMB) defines them as:

• Core counties with one or more urban areas of 50,000

people or more, and;

• Outlying counties economically tied to the core counties,

as measured by the share of the employed population

that commutes to and from core counties.

Using these criteria, urban entities are defined as countywide or

multicounty labor market areas extending well beyond their

built-up cores.

Prior to 2000, the land-use concept (Census urban areas) and

the economic concept (OMB metro areas) were not applied to

urban entities below 50,000 people. In 2000, the Census

Bureau added urban areas ranging in size from 2,500 to 49,999

(labeling them urban clusters to distinguish them from the

larger urbanized areas that had been defined since 1950). OMB

added a new micropolitan (micro) area classification, using the

same criteria as used for metro areas but lowering the thresh-

old to 10,000 people. These modifications greatly increase the

flexibility of researchers and administrators to tailor rural def-

initions to different target populations.

Ken Hammond, USDA

administrative boundaries because taxationand service provision often follow theselines. Tracking urbanization and its influ-ence on farmland prices is best approachedfrom a land-use definition that can distin-guish built-up territory from surrounding,less developed land and the degree to whichthis boundary shifts over time. Mappingdiscontinuities in the supply and demandfor medical services and analyzing theireffect on rural well-being would likely focuson distance to labor markets as a key deter-minant of health care accessibility.

In any application involving measure-ment, data availability will play a majorrole. Studies of the effects of unemploy-ment, poverty, retirement, industrialrestructuring, and other trends on ruralareas cannot easily employ administrativeor land-use definitions because data arenot available to support the m. County-level, economic definitions (nonmetro

areas) dominate rural research for this rea-son. However, researchers need tocarefully analyze and report the implica-tions of any definitional choice: Who isincluded in the study and who is left out?What information is being masked byusing large geographical building blocks,such as counties? How does this ruralgeography vary by State?

Policymakers face the same questionswhen crafting eligibility rules that best fitparticular rural programs but are not aslimited by data considerations.Considerable flexibility exists in tailoringdefinitions to suit a given application,and the appropriate choice may varydepending on program goals. A programproviding housing assistance may bedesigned to target more isolated or eco-nomically distressed rural settings thanwould programs designed to stimulatebusiness starts and job creation. Rural

communities lacking access to healthservices may not be the same areas miss-ing broadband support. Carefulconsideration of alternative definitions ofrural and their socioeconomic characteris-tics has the potential to improve theoverall efficiency of economic develop-ment programs by enabling them to bettertarget the intended beneficiaries.

34

AM

BE

R W

AV

ES


F E A T U R E

ERS Rural Definitions Data Product,www.ers.usda.gov/data/ruraldefini-tions/

The ERS Briefing Room on MeasuringRurality, www.ers.usda.gov/briefing/rurality/

The ERS Briefing Room on Land Use,Value, and Management,www.ers.usda.gov/briefing/landuse/

You may also be interested in . . .

This article is drawn from . . .

VO

LU

ME

6 �

ISS

UE

3

Ken Hammond, USDA

Springfield, MO

St. Louis, MO

Little Rock, AR

Memphis, TN

Dallas-Fort Worth, TX

OklahomaCity, OK

Tulsa, OK

Wichita, KS

Kansas City, MO

Urbanized areas

Metro areas

Note: Map shows only selected metro areas.


County-based metro areas extend far beyond their densely settled cores

www.ers.usda.gov/emphases/rural/ataglance.htm

Concise Summaries of Selected IssuesBy USDA’s Economic Research Service

Rural America At A Glance . . .

Rural Poverty At A GlanceInformation onpoverty trends anddemographic charac-teristics of the ruralpoor. While metro andnonmetro areas haveshared similar pat-terns of reductionsand increases inpoverty rates overtime, the nonmetropoverty rate consis-tently remains higherthan the metropoverty rate. Largemetro-nonmetro gapsalso exist whenpoverty is analyzedby race, ethnicity,age, and family structure.July 2004

Rural America At A GlanceThis annual reportcovers current socialand economic indica-tors for rural America,reporting on trends inemployment andearnings, populationand migration, pover-ty and income, andFederal programfunding. Key indica-tors are provided, foruse by public and pri-vate decisionmakersand others, in effortsto enhance the eco-nomic opportunitiesand quality of life forrural people and theircommunities.September 2007

Rural TransportationAt A GlanceThe effects of deregu-lation, devolution ofFederal transportationresponsibilities to theStates, increasedFederal funding, andheightened securityconcerns are dis-cussed in the contextof each mode of transportation. While93 percent of ruralhouseholds haveaccess to a vehicle,high proportions ofcarless rural house-holds are clustered in the South, Appala-chia, the Southwest,and Alaska.January 2005

Rural Hispanics At A GlanceHispanic populationgrowth has helped tostem decades of pop-ulation decline inmany rural areas. Thepamphlet draws onthe latest informationfrom the 2000 Censusand other Federaldata sources to high-light the growth of theHispanic population inthe U.S. and its geo-graphic dispersion tothe Midwest andSoutheast. The pam-phlet also summarizesdemographic charac-teristics and the mostrecent indicators ofsocial and economicconditions forHispanics. December 2005

Rural Children At A GlanceDemographic, social,and economic charac-teristics of ruralchildren in families.Although rural childpoverty rates declinedin the 1990s, theyremain higher thanthe rates for urbanchildren (21 percentvs. 18 percent). In2003, 2.7 million ruralchildren were poor,representing 36 per-cent of the rural poor.Child poverty is heav-ily concentrated in the South.April 2005

Rural Education At A GlanceInformation from the2000 Census andother Federal sourceson the educationcharacteristics of ruralworkers and counties.The report finds thatracial educational dif-ferences remain largeand that adult educa-tion levels remain farbelow the nationalaverage in many ruralcounties, particularlyin the South. Countieswith more educatedpopulations appear tohave performed better economically in the1990s and have lowerpoverty rates.January 2004

defining the “rural” in rural...

Documents