In the Case Studies section
eTransparency Case Study No.12
Text 2920/117: Reporting Police Wrongdoing via SMS in the Philippines
Case Study Author
Erwin A. Alampay (erwinalampay@bigfoot.com)
Application
The Philippines National Police - the country's unified police force - introduced a text messaging system in 2002 enabling the public to report wrongdoings by police officers (as well as by criminals).
Application Description
Given the ubiquity of cell phones in the Philippines (by 2003, there were well over 16 million cell phone subscribers out of a total population of some 80 million), there is a logic in making use of the technology for e-transparency purposes. The Philippines National Police therefore set up a text messaging service that allows citizens to seek emergency assistance, to report crime, but also to report on wrongdoing by police officers. The system can be used for any type of wrongdoing: abuse of members of the public, ineffective policing, police officers demanding bribes in relation to real or bogus crimes, protection payments demanded from businesses, failure to arrest criminals, or even direct involvement in criminal activity.
The process is initiated by a mobile phone user sending a text message (originally to text 2920, then to be subsumed by text 117). The message passes through the telecom provider's server and automatically goes into a computer at the Complaints Referral Action Centre (CRAC) which prints out the message on a text information tracking sheet. After proper evaluation, the message is passed on to the concerned PNP unit. For emergencies, the message would be relayed by telephone. For other short-term messages, fax might be used. For more sensitive issues - such as accusations of police wrongdoing - the message might be delivered by hand. When the concerned PNP unit receives the message, it takes action as necessary. In the case of a wrongdoing complaint, the message would be sent to the head of the unit involved, who would initiate some kind of investigation.
With the mobile phone number recorded, the monitoring team of the PNP text system should inform the sender on the status of the case reported, and on any action taken. CRAC should maintain a record of each complaint until that complaint is finally dealt with.
The text messaging system can also be used to report corruption in other public agencies, which the police themselves could then investigate.
Role of ICT
Cellular (mobile) phones are the primary ICT involved, using their short-messaging system (SMS/text) facility. The text messages are automatically captured into CRAC's database. Of course, ICTs only play a role here at the immediate front-end of the system, and for recording purposes. All other aspects of the procedure are undertaken in the conventional manner following print-out of the message from the CRAC server.
Application Drivers/Purpose
The overall intention behind text 2920 was to provide a low-cost method by which members of the public could be encouraged to report crimes, including crime and wrongdoing by the police. It was specifically marketed as a means for the public to report abusive, bad, and corrupt police officers (known in the Philippines as " abakada cops": rascals in uniform).
Stakeholders
There are two key stakeholders for the text messaging system: all serving police officers, and members of the public. Victims of crime are the other main group who have consciously been included in the 'stakeholder map' for the system. Though not consciously included, criminals represent the final major stakeholders. Minor stakeholders are the other players in the criminal justice system such as the judiciary and prison systems.
Transparency and the Poor
There is no overt part in the initiative that addresses the needs of the poor, although the assumption is that cell phones are pervasive enough to have penetrated at least some poor households, albeit not as much as in non-poor households.
Impact: Costs and Benefits
Unfortunately, no cost data are available on the text messaging system, but some performance indicators are known. In its first six months of operation in 2003, the text 2920 system received just over 29,000 messages. Of these, 33% were requests for assistance; 25% were reports of illegal drug use or gambling; and 13% - more than twenty messages per day - were to report on abakada cops.
Using the text messaging facility brings some potential benefits to the police complaints procedure. Messages are easy and cheap to send, and are captured onto the complaints database without human intervention. Because it is automated, the system can handle many messages simultaneously, and it avoids clogging up police phone lines. The system does not increase the number of potential informants, but it has reduced the barriers to complaints about police wrongdoing. It may therefore act as some kind of deterrent to police wrongdoing, but there is no evidence as yet to support this supposition.
The texting system certainly works in the sense that messages can be sent in, will be recorded for action, and then processed by the police complaints procedure. It seems to have fallen down somewhat in progress reporting: some callers at least report that they are not given case updates or outcomes unless they make subsequent follow-up requests. Nor is there any evidence available about the outcomes of messages sent in: no statistics are kept on how complaints are resolved, or on charges brought against police officers.
While text messaging may help increase front-end reporting, it may have little back-end impact on police behaviour. For example, in one reported case, a text message complaint was passed on to a police station commander about a particular serving officer. In course of the enquiry, that police officer swore an official affidavit stating his innocence of the charges. Without any corroborating evidence, it comes down to the word of the text messager vs. the word of the police officer. Since the text messager remains anonymous, unless they take specific action to come forward, such cases are likely to result in no action other than an initial inquiry. As described below, anonymity has also brought other problems for the text 2920/117 service.
Evaluation: Failure or Success?
There has been no formal evaluation of the project to date. As a way of seeking police assistance and reporting crimes, it does appear successful: one assumes that all crimes in progress were attended by the police. As a means of reporting police wrongdoing it is also successful. However, there is no evidence yet that it has had any impact on the actual level of wrongdoing or, indeed, whether it has led to any disciplinary or legal actions to be taken against police officers.
Enablers/Critical Success Factors
- Low usage barriers . There is a similar Web-based system for reporting crimes and corruption to the police: CRAIVIS (Complaints, Referral, Action, Investigation and Verification Information System). However, usage rates for this system are far lower than for the SMS system. Texting's success comes from its basis on a technology that is pervasive, familiar, and cheap. Probably the majority of citizens own a mobile phone, or have access to one (e.g. through a family member, friend or neighbour). They need no special training in order to use the system; just basic information about where to text. Costs are also low - text messages cost roughly the equivalent of one US cent each to send.
- Cost sharing . The time and financial costs of writing and sending the initial report are borne by the citizen, not by the state, helping to make the project more financially sustainable.
Constraints/Challenges
- Anonymity . 90% of cell phone subscribers in the Philippines use pre-paid lines, thus offering them some degree of anonymity when they use the text service (because only their number rather than their actual identity can be disclosed). This encourages reporting. However, as seen above, it hampers the successful resolution of police complaints. It has also encouraged prank text messages, with nearly 10% of all messages received on the old text 2920 system being classified as pranks, many of which have led to a (costly) police response. If this number rises - and particularly if the system is used to level false accusations against officers - this may undermine both the credibility and financial viability of the service. In mid-2003, text 2920 was merged into a centralised emergency reporting system, text 117. It has been reported that over 98% of messages to that service turn out to be pranks.
- Lack of public trust . Reducing the barriers to reporting helps one part of a police transparency/accountability system. However, people will ultimately only bother to report wrongdoing by police officers if they have confidence that their reports will be acted upon. The text 2920/117 system has not yet been able to demonstrate this. Further, there are frequent media reports about police being themselves involved in kidnapping, drug pushing, and abuse of power. This discourages the public from making any report, since they do not trust the police to do any law enforcement job properly.
Recommendations
- Address anonymity in reporting systems . In this type of e-transparency system, it seems that the costs of anonymity are starting to outweigh the benefits. Therefore, measures to address anonymity may need to be taken. The text 2920 system introduced 'Club 2920': a means for pre-registration of identified phones/callers, who would then be prioritised in dealing with 2920 messages. The new system (text 117) has planned the introduction of a mechanism of caller identification, linked to cell subscriber records. Legislation also needs to catch up, with penalties for prank texters. On the Web-based version of the system, those lodging a complaint are required to give their name and contact details and, in some cases, to come in person to present their complaint.
- Provide feedback . For the public to have confidence in e-transparency systems, they must perceive that their input has some impact. Therefore, back-end processes need to be put in place to set performance standards for post-message response times and actions, and to provide status reports to those who have lodged a complaint.
Further Information
Case Details
Case Editor : Richard Heeks.
Author Data Sources/Role : Interviews, Observation and Documents; No Direct Role.
Centrality of Transparency : Mixed. Type : Transaction. Audience : Mixed. Content : Registration. Sector : Security Services. Outcome : No Independent Evaluation.
Region : South-East Asia. Start Date : 2002. Submission Date : November 2003.