Smart. Sharp. Funny. Fearless.
Saturday, December 3, 2016

Do Presidents Cause Increased Homicides (And Suicides)? The Data Says Yes

As a noted psychiatrist and public health expert, Dr. James Gilligan has spent years at Harvard and New York University investigating the reasons (and possible remedies) for the deadly violence that has long marred American society and differentiated the United States from other advanced countries. Over the past four decades, his clinical work has brought him into contact with prisoners convicted of homicide, and with both prisoners and ordinary citizens who were close to suicide, the two major categories of lethal violence, which are generally classified as entirely different problems arising from criminal evil and mental illness, respectively.

But as Gilligan explains in his path-breaking new book Why Some Politicians Are More Dangerous Than Others, many years of studying homicide and suicide statistics in the United States led him to notice that the rates of both would rise and fall in tandem. During certain periods over the past hundred years or so, both homicides and suicides increased dramatically, and fell very quickly at other times; when laid out on a graph from 1900 to 2007 (the last year for which the government has assembled complete data), the combined violence data created a set of distinct peaks and valleys.

What Gilligan realized, as he charted levels of murder and self-murder, was that he had actually created a map showing three huge “epidemics of lethal violence” over the past century, interrupted by periods when the rates fell to lower, more “normal” levels.

And then, after trying to figure out what was causing those epidemics of violence, he noticed something else about them.

All three outbreaks of broad, sustained violence tracked very closely with the presidential election cycle: “Specifically, rates of suicide and homicide began rising to epidemic levels only after a Republican was elected president, and remained within that range throughout the time Republicans occupied the White House. The increase began during their first year or years in office, and peaked in their last year or years. They did not reverse direction and fall below epidemic levels until after Democrats took office, with the fall occurring within the first year or two of the new Democratic administration, and the rates [of lethal violence] usually reaching their lowest point during the last year or years in which a Democratic president occupied the White House.”

The correlation between politics and violence, however strong, did not necessarily indicate a causal relationship; proving that would require further and deeper research into what Gilligan calls a “murder mystery.”

But his first responsibility as a scientist was to make certain that the startling statistical relationship he had discovered was real. The U.S. government has been collecting annual data on all causes of death since 1900, which are available in databases maintained by the National Center for Health Statistics.

Gilligan took those figures and adjusted them for age, to ensure that changes in the percentage of young or old people in the population over time had not distorted the picture (since young people commit more murders and old people are more likely to commit suicide). He also had to account for the fact that not every state had sent its mortality data to Washington from the earliest days of statistical collection in 1900.

Yet no matter how he sliced and diced the numbers, those peaks and valleys of death still correlated neatly with Republican and Democratic presidencies.

Gilligan writes that he was surprised by his initial findings – and even more surprised to discover, upon further analysis, that the association between the political party of the president and the consistent, pronounced swings in suicide, homicide, and total lethal violence (combining suicide and homicide rates) were statistically significant. In other words, they were not the product of mere coincidence or chance. When he calculated the mean and median rates of violent death in America over the past century, he found that they were 19.4 and 20 per 100,000 population – and that periods when rates rose above those levels matched almost perfectly with Republican administrations, while periods when rates fell below those levels matched almost perfectly with Democratic administrations.

On page 12, Why Some Politicians Are More Dangerous Than Others includes a graph showing a sharp rise in the rates of violence between 1900 and 1916 (corresponding to the presidencies of William McKinley, Theodore Roosevelt, and William Howard Taft), which then fall from 1916 to 1920 (Woodrow Wilson), when they again start rising until they reach a high peak in 1932 (after Warren Harding, Calvin Coolidge and Herbert Hoover).

Then total violence declines steadily and steeply from 1932 to 1944 (under Franklin Delano Roosevelt), with a single year of post-war increase that then disappears. Rates of violence remain low for the next two decades (from Harry Truman through Dwight Eisenhower, John F. Kennedy and Lyndon B. Johnson). The next substantial increase doesn’t begin until 1969 (under Richard Nixon), and the rates of violence then continue on an epidemic plateau for more than two decades (from Nixon through Gerald Ford, Jimmy Carter, Ronald Reagan, and George H.W. Bush). Finally another precipitous decline commences in 1993 (following the inauguration of Bill Clinton), followed by another rise that begins in 2001 (when George W. Bush succeeds Clinton).

Not only have the rates of violence risen under Republicans and declined under Democrats, but it is possible to calculate, as Gilligan did, the cumulative number of additional violent deaths that occurred when Republicans were in power. Over the past 108 years, that cumulative increase in deaths per 100,000 was 38.2; when applied to the current U.S. population of roughly 300 million, that would come to about 114,000 additional deaths per year.

So based on the historical data, Republican administrations appear to be very dangerous, in the starkest sense, for public health. But is there a direct causal relationship between Republican presidents and rates of violent death?

Tomorrow we will see what Gilligan found when he looked behind the reasons that people kill each other and themselves – and why he believes those fatal decisions are so strongly influenced by presidential policy.

Click here for reuse options!
Copyright 2011 The National Memo