Dyess Colony and Experiments in Rural Relief During the Great Depression

Before Johnny Cash shot a man in Reno to watch him die or met his true love in the North Country or got the blues in Folsom Prison, he spent his boyhood in Dyess Colony. I might know much more about Cash’s upbringing, but Walk the Line opens with an ominous shot of a table saw and I turned it off right then (it turns out that the table saw leads to the maiming and eventual death of Cash’s older brother Jack).

My first thought when learning about Dyess—while binge-watching Ken Burns’ Country Music… oh pre-pandemic and pre-parenthood life—was “there’s a paper here, I’m just not sure what.” Consider this post as the first draft of some of those ideas.

So what was Dyess Colony? The name is a misnomer; Dyess wasn’t a colony so much as an experiment in rural relief during the New Deal. The Dyess Colony was located in northwest Arkansas, in a county that sits on the banks of the Mississippi River (the aptly named Mississippi county). Established in 1934 by William Reynolds Dyess, a local cotton baron and WPA official, struggling farm families were given a farmstead, a house, a barn, and, eventually a path to ownership.

With farmland and houses for 500 families, Dyess (also known as Colonization Project No. 1) was the largest New Deal era resettlement project. It was one of a number of audacious efforts at “social reform in a time of national crisis” (Newkirk 2014). Crises call for innovative (and at times crazy) policies and there are a lot of novel plans in the American Rescue Plan but unless we all missed it, I don’t think any new towns or communities are going to be created whole-cloth out of the Arkansas swamps.

Who “settled” Dyess? Administrators looked for only “experienced farmers who, through no fault of their own, had lost their farms” in the Great Depression (Pittman 1970). Farmers had to be healthy, young (under 50), and an Arkansas resident.

There was one other requirement of Dyess colonists: they had to be white. Like many New Deal policies—a story traced in detail in Ira Katznelson’s When Affirmative Action Was White—Dyess and other experimental communities created by the federal government was segregated. This policy likely suited Ray Cash, the father of Dyess’ most famous son, just fine. The elder Cash was a racist who “boasted of attending lynchings” (Tunnell and Hamm 2009). Holley (1971) describes some of the African-American-only resettlement programs in the Jim Crow South. But it also suggests that one long-term effect of Dyess and programs like it won’t be on racial harmony.

So what were the effects of this audacious project? Contemporaries (like today’s social scientists) knew a good experiment when they saw one. One 1939 report began: “The recently established rural resettlement colonies furnish excellent laboratories for sociological investigation.” (Loomis and Davidson 1939: Sociometrics and the Study of New Rural Communities). These contemporary sociologists drew some phenomenally complicated network diagrams by hand to predict which colonists moved out or remained and also measured no real change in religiosity among residents (the hypothesis being that “social agencies” like churches might “develop in response to the needs of the people” when “heterogeneous groups of people were established in a new environment” (Loomis and Davidson 1939: Social Agencies in the Planned Rural Communities). But Dyess and the other programs haven’t drawn a lot of social science research since then. 

Following the Dyess residents ahead in the census (to 1940 and beyond) or in other administrative records could be fascinating.  Apparently,  thousands filled out the six-page application forms but only 500 families were selected and settled in the colony. If anyone can find the forms buried in an Arkansas archive, let me know. I don’t think there’s a clean RCT or RD in there, but that would let us answer the question of how being resettled in such a program affected the “colonists”. 

Did the government create a long-lasting and durable settlement out of swampland? The short answer is no. Today, the census estimates the population of Dyess, AR to be just 363 people, much smaller than the 500 families who lived in Dyess during the Great Depression. The failure of Dyess to really take root could be a bit of a surprise as other natural experiments in American economic history have had very long-run consequences on settlement patterns. After all, the returns to scale in density reinforce agglomeration. In a paper I love teaching to my undergrads, Hoyt Bleakley and Jeff Linn (QJE 2012) study portage. First, a bit of a lesson in geography (hydrology?). Changes in elevation along rivers lead to rapids and falls. Boats (and the people and goods they are carrying) have to come out of the water. Overland hauling or portage sites became natural locations for economic activity (repair shops, hotels, taverns, and much more) to spring up. Before you know it, you have a town and then a city on your hands. As Bleakley and Linn show, at the “fall line, a geomorphological feature in the southeastern United States marking the final rapids on rivers before the ocean” there are clusters of density, visible today from satellite images.

In the western US, the US army played a similar role to navigable rivers. As my former student Chelsea Carter showed in her job market paper, counties with historical forts are denser today. The most obvious examples of army-led development still have “fort” in their names, from Fort Wayne, IN, to Fort Worth, TX, to Fort Collins, CO. The forts solved a coordination problem: which location of many in a vast countryside should people settle in?

The creation of Dyess had many of these ingredients. The Rural Electrification Administration built a powerhouse and 43 miles of lines; a bank (out of the cafe) provided loans and deposits. A connection to the main railroad was laid down. Pittman (1970) lists the amenities of Dyess Center: a hospital, a barbershop, a service station, a movie theater, a poultry plant (maybe that one is a disamenity?), a shoe shop, a garage, a machine shop, a feed mill, and a post office. All of these fixed investments in a coordinated location could have led to a virtuous cycle of agglomeration. But it didn’t. Why?

Part of the explanation could be the political uncertainty and infighting over the project. During the New Deal, Dyess was managed by several different federal agencies and made enemies of at least one Arkansas governor. Also, in an era when urban and suburban America grew and powered the war effort and post-war economy, Dyess was very much rural. The planned houses in Dyess were not built together in the town center but on individual farm plots (twenty or forty acres farms).

By 1970, Dyess Colony was just “a typical small-farm town of the Mississippi Delta” (Pittman 1970). But at least it gave us Johnny Cash.


  • James Feigenbaum

    I am an Assistant Professor in the Boston University Department of Economics. I am also a Faculty Research Fellow at the NBER in the Development of the American Economy program. My primary research interests are in labor economics and economic history.

Leave a Reply