Article in ssrn electronic Journal · July 015 doi: 10. 2139/ssrn. 2634590 citations 32 reads 1,108 author: Some of the authors of this publication are also working on these related projects

Figure 7: Botnets
0 
1000000 
2000000 
3000000 
4000000 
5000000 
6000000 
7000000 
8000000 
9000000 
10000000 
0 
5000 
10000 
15000 
20000 
25000 
2008 
2009 
2010 
2011 
2012 
2013 
2014 
BotNets Per 1,000,000 Units 
Bot Zombies 
Bots per 1,000,000 Internet Users 
Bots per 1,000,000 Active Mobile Broadband Subscriptions 
Bots per 1,000,000 Email Users 
As is clear from Figure 7, while both the normalized and 
the absolute numbers point to a decline in the number of 
10 This conceptualization focuses on the risk of having a computer 
become a botnet and not the other side of the issue of whether a botnet 
will be used to launch a DDoS attack on a website. Looking from this 
angle, the normalization of botnets around the number of Internet, active 
mobile broadband subscriptions or email users expresses how large the 
criminal element is as a proportion of all users. 

---

GLOBAL CyBERSPACE IS SAFER THAN yOu THINk: REAL TRENdS IN CyBERCRIME 
ERIC JARdINE • 13
botnet computers between 2008 and 2014, the normalized 
numbers show a far steeper drop.
11
The absolute number of 
botnet zombies, which is a count of the number of infected 
computers worldwide, fell from 9,437,536 in 2008 to only 
1,900,000 in 2014, which is a drop of 79.9 percentage points. 
In contrast, the number of botnets normalized around 
the number of Internet users fell from 6,041.69 botnets 
per 1,000,000 Internet users to 650 botnet computers per 
1,000,000 users during this same period, amounting to a 
decrease of 89.24 percent. Similar magnitude declines are 
found for both active mobile subscriptions (–96.3) and email 
users (–89.5). This data suggests that the absolute figures 
overrepresent the insecurity of cyberspace compared to 
the normalized numbers by exaggerating the problem of 
botnets as a potential vector of cybercrime.