1960s RIOTS: THE UNITED STATESThe United States possesses a long, terrible history of racial rioting. Until World War II, the vast majority of urban riots involved white-on-black violence. Perhaps the most infamous example during the nineteenth century was the 1863 riot in New York City, during which immigrants angry over the Civil War draft in the North attacked and killed scores of African Americans. As African Americans left the rural South and moved to Northern cities they increasingly competed with whites for housing and jobs. Racial tension sometimes escalated into white-on-black riots such as those in St. Louis and Chicago in 1919. In 1943 riots broke out in Newark and Detroit that in character bear a resemblance to those that occurred in the 1960s, including clashes between African Americans and police, as well as looting and arson of businesses in African American neighborhoods. Even when viewed against the backdrop of the 1940s riots, those occurring in the 1960s were without historical precedent. From the early 1960s through the early 1970s, hundreds of riots originating in African American neighborhoods broke out in American cities. Since the 1960s the United States has not been completely immune to racial riots, as examples in Miami (1980), Los Angeles (1992), Cincinnati (2001), and Benton Harbor, Michigan (2003), attest.
The standard academic definition of a 1960s race-related riot, as originally put forth by Spilerman (1970, 1971), was a “spontaneous” occurrence with a minimum of thirty participants, some black, that resulted in violent outcomes, such as arson, looting, or death. With this definition in mind, social scientists have collected information on the location, timing, and severity of the 1960s riots. Currently, the most comprehensive data set is that collected by Carter (1986), which covers the period 1964 to 1971 and includes the dates and locations of more than 700 civil disturbances, as well as numbers of deaths, injuries, arrests, and occurrences of arson. The peak years of riot activity were 1967 to 1969, especially 1968 in the aftermath of the assassination of Martin Luther King Jr. Most riots were not severe, in the sense of widespread loss of life or property destruction, but a small number were extraordinarily violent. By far the deadliest riots were those in Detroit in July 1967 (43 deaths), Los Angeles in August 1965 (34 deaths), and Newark in July 1967 (24 deaths).
Social scientists have long tried to identify city-level factors associated with the incidence and severity of the 1960s riots. In general, after accounting for each city’s black population size in 1960 and for region, little or no variation in incidence and severity can be accounted for by pre-riot city-level measures of African Americans’ absolute or relative (black-to-white) economic status. The point is not that the 1960s riots had no underlying causes but rather, if the black population in a given city was of a sufficient size, a riot could happen at almost any time in the mid-to-late 1960s if there was an appropriate spark. Most sparks were local--for example, in the Watts, Los Angeles, riot in 1965, the arrest of an intoxicated black motorist led to a wider altercation with neighborhood residents and eventually a huge riot. By contrast, the King assassination was a national spark that had the potential to incite rioting across the country.
The King riots figure importantly in two recent studies of the economic impact of the 1960s riots in urban areas (Collins and Margo 2004a, 2004b; King 2003). In the several weeks following that assassination, more than 100 riots erupted. However, riots did not occur everywhere, and a key factor in determining the likelihood and severity of a post-King riot was the level of rainfall. A high level of rainfall in April 1968 significantly reduced rioting and is a source of exogenous variation across cities. Using predicted severity based on rainfall, Collins and Margo show that the various economic outcomes for African Americans--for example, median household income, employment rates, and the value of owner-occupied housing--declined sharply between 1960 and 1970 in cities that experienced severe riots. Moreover, these declines persisted and in some cases worsened in the 1970s. Collins and Smith (2007) extend the cross-city analysis to a case study of Cleveland, showing that black neighborhoods that bore the brunt of the rioting experienced economically significant and persistent declines in property values and population that were not continuations of trends in place prior to the violence.
The standard academic definition of a 1960s race-related riot, as originally put forth by Spilerman (1970, 1971), was a “spontaneous” occurrence with a minimum of thirty participants, some black, that resulted in violent outcomes, such as arson, looting, or death. With this definition in mind, social scientists have collected information on the location, timing, and severity of the 1960s riots. Currently, the most comprehensive data set is that collected by Carter (1986), which covers the period 1964 to 1971 and includes the dates and locations of more than 700 civil disturbances, as well as numbers of deaths, injuries, arrests, and occurrences of arson. The peak years of riot activity were 1967 to 1969, especially 1968 in the aftermath of the assassination of Martin Luther King Jr. Most riots were not severe, in the sense of widespread loss of life or property destruction, but a small number were extraordinarily violent. By far the deadliest riots were those in Detroit in July 1967 (43 deaths), Los Angeles in August 1965 (34 deaths), and Newark in July 1967 (24 deaths).
Social scientists have long tried to identify city-level factors associated with the incidence and severity of the 1960s riots. In general, after accounting for each city’s black population size in 1960 and for region, little or no variation in incidence and severity can be accounted for by pre-riot city-level measures of African Americans’ absolute or relative (black-to-white) economic status. The point is not that the 1960s riots had no underlying causes but rather, if the black population in a given city was of a sufficient size, a riot could happen at almost any time in the mid-to-late 1960s if there was an appropriate spark. Most sparks were local--for example, in the Watts, Los Angeles, riot in 1965, the arrest of an intoxicated black motorist led to a wider altercation with neighborhood residents and eventually a huge riot. By contrast, the King assassination was a national spark that had the potential to incite rioting across the country.
The King riots figure importantly in two recent studies of the economic impact of the 1960s riots in urban areas (Collins and Margo 2004a, 2004b; King 2003). In the several weeks following that assassination, more than 100 riots erupted. However, riots did not occur everywhere, and a key factor in determining the likelihood and severity of a post-King riot was the level of rainfall. A high level of rainfall in April 1968 significantly reduced rioting and is a source of exogenous variation across cities. Using predicted severity based on rainfall, Collins and Margo show that the various economic outcomes for African Americans--for example, median household income, employment rates, and the value of owner-occupied housing--declined sharply between 1960 and 1970 in cities that experienced severe riots. Moreover, these declines persisted and in some cases worsened in the 1970s. Collins and Smith (2007) extend the cross-city analysis to a case study of Cleveland, showing that black neighborhoods that bore the brunt of the rioting experienced economically significant and persistent declines in property values and population that were not continuations of trends in place prior to the violence.