[This note was submitted to the Election Commission of India.]
The National Statistical Commission under the chairmanship of Dr. C. Rangarajan (2001) recommended [please see Reference (1)] for conferring statutory status to the Commission on the lines of what has been done in the United Kingdom (UK). In the UK, under the terms of their Statistics and Registration Service Act, 2007, the UK Statistics Authority headed by the National Statistician is responsible for promoting and safeguarding the integrity of official statistics. Statistical activities are conducted in the UK in accordance with the Code of Practice for Official Statistics notified by the Authority. During election time, specific guidance [please see Reference (2)] on statistical activities is provided to political parties and Government offices as a part of General election guidance. In addition to this, there is a practice of the Authority’s Chair addressing letters to leaders of UK’s political parties reminding them of their duties to use statistics responsibly during the election period. A letter written by Sir David Norgrove, the Authority’s Chair [please see Reference (3)] in November 2019 is extracted below:
“To: Leaders of UK political parties
I write to ask for your support and leadership to ensure that the use of statistics during this general election campaign serves the public good.
The UK Statistics Authority has the statutory objective to promote and safeguard the production and publication of official statistics. My predecessors and I have in the past been obliged to write publicly about the misuse of statistics in both pre-election and pre-referendum periods. Statistics can be a powerful support for an argument but misuse damages their integrity, causes confusion and undermines trust. It can also lead debate to focus too much on the statistics themselves, distracting from the issues at hand. This is particularly important during the intense public scrutiny of an election campaign, where misinformation can spread quickly.
I would ask that statistical sources should be clear and accessible to all; any caveats or limitations in the statistics should be respected; and campaigns should not pick out single numbers that differ from the picture painted by the statistics as a whole.
I am sending identical letters to the leaders of the main political parties, with a copy to the Cabinet Secretary.”
In India, a National Statistical Commission (NSC) was constituted in July 2006 through an executive order as an interim arrangement which would evolve a statutory Commission based on its experience and it has been functioning since then as an advisory body on statistical matters. The NSC has not been bestowed statutory status yet, although it has made detailed recommendations on the subject in its annual report for the year 2011-12[please see Reference (4)]. A Code of Statistical Practice as a part of National Policy on Official Statistics was recommended by the NSC in its annual report 2010-11 [please see Reference (5)]. The Code is like the one notified in the UK. In its annual report 2013-14 [please see Reference (6)], the NSC slightly modified the National Policy on Official Statistics and recommended adoption of the United Nations Fundamental Principles of Official Statistics (FPOS). Nothing was heard about the Policy after what was said by the Hon’ble Finance Minister in the Budget speech 2020-21. Relevant extract of the speech is given below:
“86. There is a growing need for the Indian Statistical system to meet the challenges of real time monitoring of our increasingly complex economy. Data must have strong credibility. The proposed new National Policy on Official Statistics would use latest technology including AI. It would lay down a road-map towards modernised data collection, integrated information portal and timely dissemination of information.”
Be that as it may, the NSC in India as of now neither has the status nor the locus standi to act like the UK Statistics Authority in protecting the integrity of official statistics and in seeking cooperation of political parties during elections for the purpose. Hence, the Election Commissions (at the Centre and in the States) may have to take the initiative in this direction, as a part of their regular exercise of notifying a Model Code of Conduct (MCC) during each election process.
The Election Commission of India is vested with the responsibility of conducting elections to Lok Sabha and State Legislative Assemblies whereas the State Election Commissions are entrusted with the conduct of elections to Rural and Urban Local Bodies in the States. During an election process, the political party or persons in power have an obvious advantage over others. An Election Commission ensures a level playing field for all the political parties and persons in election fray, through strict observance by them of a MCC evolved with the consensus of political parties. A set of Dos and Don’ts are prescribed under MCC for the purpose.
During general elections to the Lok Sabha, the MCC is applicable throughout the country. During elections to any Legislative Assembly, the MCC would be applicable in the concerned State. During bye-elections, the MCC would be applicable in the concerned constituency, in case it is comprised in State Capital / Metropolitan City / Municipal Corporation, and in the entire district(s) covering the Constituency otherwise. When elections to Rural and Urban Local Bodies are conducted, the MCC applies to the respective areas. MCC is enforced from the date of announcement of election schedule by the concerned Commission till the process of elections are completed. The duration for which the MCC would be effective, hereinafter referred to as MCC period, would vary normally from a month to three months.
Nobody would expect the Government to come to a halt during MCC period. Statistics, being central to heated debates during elections, should not be halted. However, there could be temptations to misuse or misrepresent or misquote statistics during elections and somebody should check such acts to protect the public from being misled.
As of now, there is no direct mention in MCC in respect of any embargo on statistical activities or statistical releases during election process. The portion in MCC which may perhaps be remotely connected with the issue is extracted below:
“The party in power whether at the Centre or in the State or States concerned, shall ensure that no cause is given for any complaint that it has used its official position for the purposes of its election campaign and in particular –
…………………………………………….
(iv) Issue of advertisement at the cost of public exchequer in the newspapers and other media and the misuse of official mass media during the election period for partisan coverage of political news and publicity regarding achievements with a view to furthering the prospects of the party in power shall be scrupulously avoided.”
It is a universally accepted fact that official statistical releases either directly or indirectly reflect the performance of the Government or the political party in power. Citizens are entitled to know the statistics produced with public funds, whether the statistics are in favour of or against the government of the day. Although statistical releases are neither in the form of an advertisement nor are intended to give publicity on the achievements of the party in power, they are likely to be misused or overstated during an election process.
Who approves statistical releases in each of the government offices is a moot question. In case of the Ministry of Statistics and Programme Implementation (MOSPI), which is the nodal Ministry on statistical matters at the Centre, all the reports are released either with the approval of the Chief Statistician of India (CSI) or with the approval of the National Statistical Commission (NSC) and not of the concerned Minister. The statistical methodology leading to the releases, evolved from time to time in consultation with Experts and the NSC, is placed in the public domain. Pre-release access to the statistical reports is not given to any politician, bureaucrat, or Minister, except GDP estimates in case of which only the Chief Economic Adviser (Ministry of Finance) is provided with a copy of the release only half an hour before the scheduled release and this fact is also placed in the public domain. Some of the releases of the MOSPI indicate results at all India/ industry level etc., (not State-wise), whereas others, in particular, the Consumer Price Indices include State-wise position. There are a few pre-announced releases of MOSPI. The National Accounts Statistics (GDP data) are disseminated as per the standards laid down for the real sector by the International Monetary Fund (IMF) under Special Data Dissemination Standards (SDDS). The press releases on GDP are issued from time to time according to an advance release calendar, which is available at the MOSPI’s website. The MOSPI releases monthly data on inflation and Consumer Price Indices on the 12th of next month (in case 12th happens to be holiday, on the next working day). The MOSPI also releases monthly data on Index of Industrial Production on 12th of every month (in case 12th happens to be holiday, on the previous working day) with about six weeks lag.
There are other statistical releases of the Central Government and of the State Governments and the schedule of release of most of them are not pre-announced. Moreover, most of these releases are made from time to time either with the approval by the concerned Minister or with that of a senior officer in the hierarchy. Often, such Ministers or senior officers do not involve themselves in statistical operations and they approve releases on being briefed about the key results and their implications. One may expect some bias in such releases, even during non-election times. Another intriguing fact is that some of the releases of the MOSPI, such as the National Accounts, are based on data of other Ministries at the Centre and in the States. The statistical releases made in the country from time to time depict various socio-economic indicators at national level, or State level, or District level, or sub-district level and it would not be desirable to stop the releases even when the results of one or more constituencies going for elections are directly available or could be culled out from the releases.
Statistical releases may be based on censuses or sample surveys or administrative statistics. The releases could be regular and / or periodical and at times based on international commitments. There could be other releases based on a census or a sample survey conducted occasionally. Thus, some of the statistics disseminated are regular/ periodical products, whereas the remaining are ad hoc or otherwise.
At the international level, the ten Fundamental Principles of Official Statistics (FPOS) adopted by the United Nations General Assembly, through Resolution No. 68/261 in its 73rd plenary meeting held on 29th January 2014, guided many countries to evolve statistical laws, protocols, and procedures to enhance the independence and integrity of their statistical systems. The Government of India formally notified adoption of these principles on 15th June 2016 [please see Reference (7)]. Official Statistics are collected with public funds and hence, are to be disseminated to the public on a timely and impartial basis. The first two principles of the FPOS given below emphasise this point:
“Principle 1. Official statistics provide an indispensable element in the information system of a democratic society, serving the Government, the economy, and the public with data about the economic, demographic, social and environmental situation. To this end, official statistics that meet the test of practical utility are to be compiled and made available on an impartial basis by official statistical agencies to honour citizens’ entitlement to public information.
Principle 2. To retain trust in official statistics, the statistical agencies need to decide according to strictly professional considerations, including scientific principles and professional ethics, on the methods and procedures for the collection, processing, storage and presentation of statistical data.”
The Government at different levels produces and disseminates statistics from time to time on various subjects falling in their domain. Hence, the Indian Official Statistical System is a decentralsied system. Different political parties may be running the Governments at different levels. If the respective political bosses approve dissemination of statistics at these levels, it may be presumed by the public that such releases could be influenced by their political interests. The presumption persists even when officers approve the releases, for the simple reason that they may not do anything to antagonise their political bosses. Such presumptions have been impacting the credibility of the statistical system adversely and jeopardising public trust in statistics. If there is an established mechanism to check misuse of statistics, then disseminating statistics will not be tantamount to violation of MCC. Best mechanism for the purpose is to have a set of protocols laid down through the usual consultation process by the Election Commission of India to guide all the producers of official statistics and political parties during election times. The same protocols may be prescribed by the State Election Commissions as well. The set of protocols suggested for the purpose is given at Annex. I.
References:
(1) Rangarajan Commission’s report (chapter-14) is available at: https://www.mospi.gov.in/14-indian-statistical-system
(2) General election guidance 2019: guidance for civil servants is available at: https://www.gov.uk/government/publications/election-guidance-for-civil-servants/general-election-guidance-2019-guidance-for-civil-servants
(3) The role of the Statistics Authority in the 2019 General Election is available at: https://osr.statisticsauthority.gov.uk/blog/the-role-of-the-statistics-authority-in-the-2019-general-election/
(4) Annual report for the year 2011-12 of the NSC is available at: https://www.mospi.gov.in/sites/default/files/annual_report/nsc_AR_2011-12.pdf
(5) Annual report of the NSC for the year 2010-11 is available at: https://www.mospi.gov.in/sites/default/files/annual_report/nsc_annual_report_2010-11_eng_6sep12.pdf
(6) Annual report of the NSC for the year 2013-14 is available at: https://www.mospi.gov.in/sites/default/files/annual_report/nsc_AR_2013-14.pdf
(7) The Gazette Notification on adoption of Fundamental Principles of Official Statistics is available at: https://www.mospi.gov.in/sites/default/files/main_menu/gazette_notification/UNFPOS_11july16.pdf
Annex.I
Protocols for statistical activities during MCC period
Statistics are very important for informal and formal democratic debates. Government (at different levels) produces and publishes a wide range of statistics on different facets of the economy and society. It is neither possible nor desirable to put a stop to activities leading to statistical releases, particularly during elections when intense democratic debates are happening around. However, it has been considered necessary to lay down certain protocols consistent with the Fundamental Principles of Official Statistics to provide guidance to producers of official statistics and political parties and persons in the election fray to be followed by them during the MCC period. The protocols are given below:
(1) The utmost care should be taken as usual to ensure that statistical results and findings are presented impartially in an objective manner.
(2) Regular statistical products or releases being brought out in the form of press notes or bulletins or publications or electronic releases or through uploads in dedicated internet sites in accordance with a pre-announced calendar or any international commitment shall be continued unabated during MCC period.
(3) Any ad hoc statistical release (other than regular statistical products or releases), even when a pre-announcement was made specifying that the release would be made during a certain period, say, a week or a month or a quarter, but not on a particular day, should not be made during the MCC period. In case of a need to release such products in public interest during the MCC period, approval of the National Statistical Commission shall be taken for such release including the content of release.
(4) Publication of scheduled statistical releases shall continue during the MCC period. However, if the date of any statistical release falls on polling day itself, then the release shall be made only after 5 p.m. on that day or it shall be postponed to the following day.
(5) Any request made by any citizen or a candidate contesting election for furnishing unpublished statistics shall be honoured and met in a manner that identity of any person to whom the statistics or part of it relate would not be possible to be culled out even by the process of elimination.
(6) Descriptive analysis which is often given along with statistical results requires special attention, because it is a potent source for misuse or misinterpretation or adverse criticism. Hence, such analyses issued during the MCC period shall be highly impartial and restricted to only basic facts. In case of doubt, the advice of the National Statistical Commission shall be taken.
(7) Existing practices, if any, provided for pre-release access of statistical products for the purpose of briefing to any Minister or a senior officer in the Government who is responsible for policy making on the subject or a related subject or who may interact with media immediately after the release, may be continued. However, no practice of this nature shall be introduced for the first time during the MCC period.
(8) Officials involved in the production of official statistics shall not give any oral briefing to any Minister during the MCC period. However, if there is a practice to brief a Minister in respect of market-sensitive statistics or any other statistics which have an implication for any public service, such practice should be continued.
(9) Requests, if any, received from any person, including a candidate contesting in an election, for guidance on methodology followed in any existing statistical product shall be honoured and fulfilled. However, any request received from any person for advice on the interpretation or analysis of statistics released should be handled with great care, as the advice furnished is likely to be misused. In such cases, analysis or interpretation shall be restricted to only basic facts. In case of doubt, the advice of the National Statistical Commission shall be taken.
(10) Arrangements for free or regular mailings, if any, existing prior to the MCC period in respect of statistical releases to organisations and important persons outside the Government setup shall be continued during the MCC period.
(11) Any request received during the MCC period for supply of a few copies of leaflets, background material or free publications relating to existing statistical products shall be dealt with in the same manner as if such request was received prior to the MCC period. However, bulk supply of such material should not be made to any person or organisation during the MCC period, as there is a possibility of using the material for propaganda purposes in elections.
(12) Launching or continuing to conduct any regular or periodical census or a survey of individuals, households, businesses, or other organisations may be continued during the MCC period. Ad hoc surveys in respect of purely socio-economic facets may also be either launched or continued. However, ad hoc surveys that are likely to lead to any controversy or are related to an election issue should be either postponed or abandoned. Similarly, new Censuses or surveys of a continuing nature, unless pre-announced, shall not be launched during the MCC period, because new government to be formed after election may take a different view about them. In case of doubt, the advice of the National Statistical Commission shall be taken.
(13) On-going technical consultations on existing statistical products or on statistical products likely to be evolved shall be continued as usual, but without any publicity during the MCC period. But consultations with the public, including user meetings on statistical products, shall not be undertaken during the MCC period.
(14) On receipt of any complaint from any person or a specific reference from an Election Commission during the MCC period on misinterpretations or misrepresentations of statistics, the National Statistical Commission shall consider the complaints / references and publish facts, under intimation to the Election Commission, relating to underlying issues without making any reference to the complaints or to the source or cause of the complaints. (15) These protocols may not cover every eventuality, but the principles thereof should be applied to specific circumstances during the MCC period. In case of any doubt on any issue relating to statistical activities not covered in these protocols, advice of the concerned Election Commission shall be sought. The Election Commission, in consultation with the National Statistical Commission where necessary, will give appropriate directions.
Ranganadham 