Violent Crime Rates Under New York City Mayor Giuliani

My previous post detailed the drop in the violent crime and murder rates in New York City since their peak twenty-five years ago and showed that it led New York State in improving crime statistics.

The differences were dramatic. In 1990, there were 2245 murders in New York City and 360 in the rest of New York State. (In 1990, New York City had approximately 40% of the state's population.) In 2014, there were 333 murders in New York City and 284 in the rest of the state. (In 2014, New York City had approximately 42% of the state population.)


Mayor Giuliani

In this post I'm going to look at the question of how much of a real decrease took place during Rudy Giuliani's tenure as mayor of New York City.

In my opinion, Rudy Giuliani is a shrill blowhard. However, it is possible that he is a shrill blowhard who was competent at something. I am going to look at statistics as to whether the drop in violent crime rate was real, substantive and represented a better performance than the United States as a whole over the same period of time. I am not going to address the methods to achieve these changes: I'm taking on a delicate enough issue as it is. While numerical changes succumb to statistics, policy matters are a forever-long debate.

Background

Rudy Giuliani, the 107th mayor of New York City, served from January 1, 1994 through December 31, 2001. This is convenient, there are no half-year or even half-month statistics to look at. Giuliani succeeded Mayor David Dinkins, in office from 1989 through 1993.

Giuliani had three police commissioners. The first was William Bratton from January 1, 1994 to April 15, 1996. The second was Howard Safir from April 15, 1996 to August 18, 2000. The third was Bernard Kalik from August 21, 2000 to December 31, 2001. (Bratton would return as police commissioner under current Mayor de Blasio.)

According to the FBI Uniform Crime Reporting, in 1993, the year before Giuliani took office, the violent crime rate in New York City was 2089.8 per 100,000 population. This was down by 12.3% from its peak in 1990. The murder rate was 26.5 per 100,000 population down by 13.6% from its peak in 1990.

In 2001, the violent crime rate in New York City was 927.5 per 100,000 population (official FBI statistics, more on this below) and the murder rate was 8.9. These represented decreases of 55.6% and 66.4%, respectively.

First question: violent crime was decreasing from 1990 to 1993, so, did that decrease accelerate? If we project the rate of decrease under the last three years of Dinkins' administration throughout the Giuliani years we would have had a 29.4% decrease in violent crimes and a 32.3% decrease in murders. The actual numbers (as mentioned above) came out to be 55.6% and 66.4%.

So far, things are looking good for Giuliani. Let's look at the thorny problem of the census figures. The FBI Uniform Crime Report Statistics are based in part on actual census figures every 10 years, in this case, 1990 and 2000. For the years in between, the numbers are based on official census projections. The problem is that the census projections for New York City in the 1990s were wrong. From 1990 to 1999, the census projection showed a total of 1.46% total growth. The actual growth over the ten years was 9.36%. This is reflected in the following figures.

Year   Population, NYC
1990  7,322,564
1993  7,347,257
1997  7,320,477 (down from 1990!)
1999  7,429,263
2000  8,008,278
2001  8,023,018

This artifact provided the violent crime rates to show an artificial dip between 1999 and 2000. While the number of violent crimes in New York City fell from 78,495 to 75,692 (-3.6%), the violent crime rate dropped by -11.0%. Similarly, while murders rose from 664 to 673, the murder rate dropped by -6.0%.

That may seem like an unfair boost to Giuliani's statistics, but the 2000 census did represent a real count. What is unfair is that throughout the 1990s, with the population being undercounted, the crime rates were artificially high, to a greater degree as the decade went on. The undercount overestimated the 1993 statistics, thereby favoring Giuliani's statistics. Indeed, the total number of murders bottomed out in 1998 at 633, while there were 714 in 2001, his final year. Without the census anomaly, the murder rate would have been increasing during those final years.

What was the actual population in 1993 (or 1999)? There is no way to be sure, but if we model a linear growth from 1990 to 2000, the 1993 population would be 2.5% higher (and subsequent crime rates, 2.4% lower) than the estimated figure. This works out to be more than it might seem. The projected changes in crime rates act like compound interest, with each year's number affecting the next.

So let's redo the 1990 to 1993 (pre-Giuliani) change and acquire the extrapolated figures that look at what would happen if the numbers continued to fall at the corrected 1990 to 1993 rate, and 1993 to 2001 (Giuliani numbers) using a linear model of population growth.

Giuliani's numbers change only slightly. Now violent crime was down -54.5% and murder -65.6%. The extrapolated changes using the 1990 to 1993 rate now bring a decrease of -33.9% in violent crimes and -36.5% in murders, if extrapolated to 2001.

Still Giuliani's stats are performing well. Furthermore, the New York City statistics outperformed those of the United States as a whole which had a drop of -32.5% in violent crime rates and -41.1% in murder rates between 1993 and 2001.

Let's look at when the changes occurred.

Violent Crimes, New York City
Year  Year-to-Year % Change
1991  -3.28
1992  -7.21
1993  -4.66
1994  -11.9
1995  -17.2
1996  -14.2
1997  -6.71
1998  -8.30
1999  -8.88
2000  -4.94
2001  -1.87

You may note that the most substantial changes took place in the years 1994 to 1996.

William Bratton served as police commissioner from January 1, 1994 to April 15, 1996, two full years and 105 days. I don't have the data to divide up partial years, but in his first two years as police commissioner, the violent crime rate dropped -27.1% and the murder rate dropped -40.6%. Counting the first full two years (1997 and 1998) of his successor, Howard Safir, the violent crime rate dropped -14.5% and the murder rate -36.7%. In the next three years of Safir and Kalik (1999 through 2001) the violent crime rate dropped another -15.0% while the murder rate rose by 10.7%. Overall, from 1997 through 2001, the violent crime rate dropped -27.3% while the murder rate dropped 30.0%. The five full years after Bratton the drop in crime rate performed equally or less well than the two full years under Bratton.

Year-to-Year Changes in the Violent Crime and Murder Rates, from 1990 to 2001. The figures from 1991 through 1999 were adjusted to account for a linear population growth from 1990 to 2000. Note the murder rate increased over the final three years.

The Cumulative Change in Violent Crime and Murder Rates. As above, adjusted for a linear population growth. The cumulative changes are calculated as 1991 through 1993 in comparison to 1990, and as 1994 through 2001 in comparison to 1993 (Dinkins' final year).

My personal verdict so far. Did Giuliani do something right? Yes: he hired Bill Bratton. Did he do something wrong? Yes: he let Bill Bratton go, forcing him out of office.

Giuliani's Final Year.

In David Dinkins final year as mayor, terrorists attacked the World Trade Center, leading to several deaths and thousands of injuries. In Giuliani's final year as mayor, terrorists attacked the World Trade Center leading to thousands of deaths and injuries.

The FBI Uniform Crime Statistics don't include the deaths and injuries from the 2001 attacks. They do include the deaths and injuries from the 1993 World Trade Center attack, the 1995 Oklahoma City attack, in fact, all other terrorist attacks on American soil.

Let's be clear: Giuliani was not behind the 9/11 attacks. Giuliani was not behind any of the 7,175 other murders that took place in New York City when he was mayor. And, of course, Dinkins was not behind the 1993 attack.

Still, the erasure of murders from the counts, just because the number is horrific, is wrong. Giuliani is not responsible for the 9/11 attacks, but is he responsible for any of the deaths or injuries of the 9/11 attacks? This is a thorny issue. A mayor does not only reduce crime through law enforcement, but, among other things, also through making sure street lamps work, and that emergency crews have timely access and coordinated communication. The last of these was a real issue when it came to firefighters hearing the evacuation orders after the first World Trade Center tower fell.

Should the communication have been better? Yes. You can argue that we know that because hindsight is 20/20, but: the buck stops at the mayor's office. Everything he did right to make New York less dangerous has to be weighed with what he didn't do.

As a result of the 9/11 attacks in New York City, 2,753 victims died and an additional 6,294 were treated for injuries at local hospitals. If these were included with Giuliani's statistics, as the World Trade Center attacks were for Dinkins, the murder rate would have reached its peak in 2001 with 43.2 murders per 100,000, an increase of 67.2% over the course of the Giuliani administration. The violent crime would be adjusted to 1040.3 per 100,000, still a healthy decrease of -49.0%.


Martin Hill Ortiz is the author of Never Kill A Friend, Ransom Note Press.

Never Kill A Friend, Ransom Note Press

Never Kill A Friend is available for purchase in hard cover format and as an ebook.
The story follows Shelley Krieg, an African-American detective for the Washington DC Metro PD as she tries to undo a wrong which sent an innocent teenager to prison.

Hard cover: Amazon US
Kindle: Amazon US
Hard cover: Amazon UK
Kindle: Amazon UK
Barnes and Noble