ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

India’s Contentious Crime Statistics

What should we know to make our crime statistics reliable?


The credibility of the National Crime Records Bureau’s (NCRB) statistics is perennially doubted in this country, especially since the late 1990s, when the numbers began revealing declining trends in the incidence of crime per lakh population. With the recent release of the NCRB’s Crime in India 2017, this mistrust has only exacerbated. First, due to the timing of the release of the report, which, following the NCRB’s conventional practice of annual release for almost six decades now, should have been released in 2017. Second, during this hiatus in publicising the annual crime statistics of 2017, the bureau did not come up with convincing responses to the right to information applications seeking the reasons for this undue delay. Third, and specifically in light of the first two reasons, it is not unusual that a staggering 33% decline in the crime rate within half a decade—from 576.99 in 2010 to 388.6 in 2017—is appearing unfounded to many. An intuitive argument is that neither the current demographic transition nor the economic transformation seem to provide conducive conditions for such a decline.

Crime declines in India, however, are in tandem with the global trend of “crime drop”—observed mostly in the industrialised countries since the 1990s—that has not only been well-documented, but also much debated. While concurrence (or the lack of it) regarding the veracity of such crime-drop statistics is usually facilitated by corroborative (or divergent) estimates from other data sources (such as the United Nations Interregional Crime and Justice Research Institute’s International Crime ­Victims Surveys [ICVS]), in most of these countries, the process of ratification is taken a step ahead by the academic discourses that explore the drivers underlying the drop. In the Indian case, the very basic verification of official crime statistics is impaired by the dearth of alternative benchmarks. Whatever little was available, such as the only ICVS conducted in 1992, had generated victimisation estimates that were way higher than comparable official estimates of crime rates for that year.

However, the downward trend in crime rates in India may find some traction if viewed in a causality framework, parti­cularly in view of the “security hypothesis”—that is, change in the quality and quantity of security provisions potentially drives down crimes—given India’s burgeoning electronic security market, which is expected to grow to $2.4 billion by 2020 at a projected compound annual growth rate of 32% between 2017 and 2023. But, the consideration of this hypothetical framework to interpret the NCRB’s crime statistics may involve certain pitfalls. First, while technology-led security may reduce (traditional) acquisitive crimes like burglary, car theft, street crimes, etc, it also opens up the opportunity for new types of crimes, such as cyber thefts that use the very same technology. Then, is the relatively low number of cyber crimes in the NCRB’s report a matter of some procedural glitch? Second, if technology-based surveillance can reduce the opportunity for violent crimes in general, then why does it not apply for the marginalised categories of the population, namely women, Scheduled Castes (SCs), and Scheduled Tribes (STs)? Or, is the year-on-year increase in the crime rate for women, SCs, and STs a fallout of the increased visibility, and hence awareness and reporting of such crimes, effected by technology-led security? If that is the case, then, logically, data on mob lynching should have also been a part of this report.

While one may find it difficult to refute the government’s claim regarding the unreliability of lynching data received from the police stations on the grounds of the absence of any official definition of mob lynching, it is equally hard to dismiss the dwindling credibility of some of the NCRB’s reclassifications of old crimes, and/or classifications of new crimes by the same logic. For instance, the new category of “crimes by anti-national elements.” Without any contour for defining “anti-national,” this classification suffers from the lacuna of being undefined in legislative discourse. Given such selective and unprincipled inclusion of data, it may not be incorrect to assume that the official crime data in India is politically motivated.

In this context, the apprehension about the credibility of a crime drop in India becomes real, for one cannot ignore the fact that these statistics form the performance parameter of the criminal justice system, particularly the police force. Thus, there is ample incentive for subsequent under-reporting along the line ministries. Even a commoner’s expectation and perception of crime data is no different. We tend to overlook the fact that crime is as much a function of socio-economic factors, as it is of the political and criminal justice systems. For instance, changing demographic profiles can bring about substantial changes in levels, rates, and even the nature of crimes. In this context, official policies should be evaluated more holistically, that is, in terms of their impact on the capacitation of individuals, and in turn, intercepting the incidence of crime. In such a scenario of flux, if we continue viewing crime statistics as merely a tool for blaming and shaming the executive branch of the criminal justice system, we stand at the risk of losing sight of some overarching issues of governance related to the entitlement and empowerment of the people.

Moreover, by reducing crime statistics to a yardstick for assessing official performance, we are actually ratifying the ­official mindset that these numbers are of the most relevance to their producers. Hence, the government can conveniently deny, procrastinate, or control activities for generating alter­native databases—such as crime victimisation surveys—for cross-checking its crime data. One must remember two things at this juncture. First, that it is the government’s imperative to govern, which provides the rationale for producing indicators/data to form the basis of its decision-making. Second, it is the quality of this data that determines how decisions are made by fundamentally shaping what we know about the contingent issue. Thus, while pressing for alternative databases, we should be mindful that the purpose of these is not delimited to merely identify the existing crime estimates as poor, but to understand that if poor, why and how poor the current numbers are.


Updated On : 22nd Nov, 2019


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top