Schneier Unfair to CCTV: Exploring CCTV AdvantagesBy John Honovich, Published on Jul 01, 2008
While Bruce Schneier is a brilliant security thinker, I find his recent arguments against CCTV to be unfair. This is a critical issue because Schneider's opinions are singularly critical to shaping both security practitioner's and the general public's views on security. I hope to offer a different perspective and additional information to better help the community understand this important issue.
Here are three main points from Schneier's CCTV Camera article:
- “Criminals can easily adapt by moving their crimes to someplace not watched by a camera and there will always be such places.”
- “By their very nature, cameras result in underused and misallocated police resources.”
- “The funds spent on CCTV cameras would be far better spent on hiring experienced police officers.”
In this report, I will offer 4 main claims:
- CCTV cameras increase the costs of crimes
- CCTV cameras are far less expensive than police
- CCTV cameras are force multipliers
- Cost of CCTV systems has dropped substantially in the last 5 years
Not Just Shifting Around Crime
A common Schneier critique is that CCTV cameras just shift crime around. The supermarket adds cameras and the criminals just move to the mini-mart, etc., etc. As such, there is little or no social value being generated, just private individuals pushing off crime to their neighbors.
This is a tricky argument. You could make the same case for using a lock on your door. If you use a lock, the criminal simply shifts next door and robs them. Clearly, not everything can be locked and the criminals will simply move to those places without locks. QED, we should not buy locks anymore and simply spend more money on hiring experienced police officers.
CCTV cameras decrease the value of conducting crimes. Assets have different worth. Locations have different costs. Criminals are trying to maximize the total value of their thefts by choosing high value targets with sufficiently low risks of being caught. By placing cameras on assets with the highest worth and in areas where the cost is low to access, cameras can decrease the value of crimes.
Let's look at fraud in banks. Fraud is a huge problem for banks. A competent fraudster can walk into a bank and in 10 minutes steal $5,000 while getting a smile and a thank you from a teller. Almost every bank I have ever seen has cameras. Are the banks simply shifting the crime somewhere else? The crime may shift but the value to the criminal is much lower and the potential loss to society is lower because there are very few targets as attractive to a criminal as a bank.
When crime is harder and less profitable, criminals take less from society and are are less motivated to conduct crimes. While solving murders and terrorist events get all the sizzle, cameras are valuable in solving 'mundane' and routine day to day issues like theft, assault, vandalism, etc. that generate real economic returns to society.
How much actual reduction is a genuine discussion for debate and evaluation. My point here is not to claim that cameras eliminate all crime but to discredit this unqualified notion that cameras simply shift crime around and create no social value.
How much Better is a Police Officer than a Camera?
Schneier recommends spending money on police officers rather than cameras. Certainly, police officers are better than cameras much like having Bruce Schneier run your Information Security program is better than just having free ware anti-virus on your PC. The substantive question is that given the relative costs, how much more valuable is police officers than cameras.
A police officer is about 200-300 times more expensive than a security camera. A 24/7/365 police officer rotation likely costs $150-$200K USD. Today, a security camera generally costs $2,000 to $4,000 USD (installed, networked and with a video management system). Given a 5 year lifespan, the annual cost of a security camera is $500 - $1000 USD (factoring in service costs and the time value of money).
Security cameras are a very inexpensive substitute for policing. You can literally have hundreds of cameras for the cost of 1 more police officer. That being said, if security cameras do not work, then it does not matter if they are free. As a first step, I do think it's important that everyone understands just how incredibly inexpensive cameras are compared to people. In my previous points, I noted how security cameras can decrease the value of crime to criminals and how they can help in solving and reducing mundane crimes. Next, let's look at how they can help the police.
Schneier's position that cameras, by their very nature, “result in underused and misallocated police resources” is in stark opposition to the experience and understanding of most every working physical security manager. Moreover, it opposes the recommendation of Sandia Lab's Mary Lynn Garcia in her seminal work, “The Design and Evaluation of Physical Protection Systems.”
Cameras are force multipliers – tools that make a given group of people more effective without adding more people. Force multiplication is an old and critical concept in security. You will hear military people talk about it routinely. I have personally seen cameras successfully used as force multipliers in city wide surveillance, malls, stadiums and major retailers, just to name a few. They allow limited operators to cover more areas through centralized direction.
Improper use of video surveillance will absolutely result in problems but I believe it is absolutely false to conclude that it is “by their very nature” as Schneider does. Certainly if you have 500 cameras over 10 square miles and 5 people, you will have serious problems. But if you have 30 cameras across 1 square mile and 3 people, you can very effectively coordinate resources.
Properly used, cameras allow security responders to cover a greater area without increasing headcount. This is where the cost differential between cameras and officers becomes most critical. Rather than adding more officers costing hundreds of thousands of dollars per year, a fraction of that amount can be used make existing officers more effective. Even if crime stays the same relative to hiring an additional officer, it decreases the cost of security which is a societal good.
Camera System Costs Dropping
Schneier focuses on historical results of CCTV cameras. I totally agree with him on this as it is always dangerous and risky to make rosy projections about how things will improve in the future.
One thing that has already occurred and is a matter of fact is how far the cost of camera systems has fallen in the past five years. In 2003, the typical public surveillance solution consists of proprietary analog fiber systems which routinely ran $4,000 USD or more per camera simply for the infrastructure. Today, public surveillance solutions use IP networks. Even in the worst case scenario where you have build your own IP network, the overall cost reduction for the system is 50% or greater.
Similar decreases in the cost of megapixel cameras and video analytics have also occurred, providing greater evidence quality and better response to high threat activities.
All of this is actual pricing and production deployments today. This will inevitably lead to greater success in the field. How much success is certainly an issue to be measured and tracked but significant improvements are reality, not speculation.
While I share many of Schneier's concerns, his consistent and uncritical criticism of video surveillance needs questioning. I hope the points I have raised spurs further discussion and reflection of this matter. As this is a very popular topic, I am looking forward to your thoughts in the comments.