Torture the data, and it will confess to anything.

—Ronald Coase

The mystery behind the unprecedented and foolish decision to demonetise 86% of the currency in circulation may never be fully unravelled. However, by now it is more than clear that it was taken by none other than der Führer. In the characteristic style of his mother organisation, Rashtriya Swayamsevak Sangh (RSS), Narendra Modi went on shifting the goalposts when the foolishness of the move began to unravel. When the system failed to provide cash to people, he swiftly asked them to go digital: to make India a cashless economy, which he later moderated to less-cash economy.

While launching an Aadhaar-based e-transaction biometric app abbreviated as BHIM (Bharat Interface for Money) on 30 December 2016, he did not forget to score a political point by saying it was named after Babasaheb Ambedkar. This failed to cut ice with people, scores of whom had already faced untold hardships by then.

The expectation that a significant amount of old currency in black money would never come back to the Reserve Bank of India also did not come true. Embarrassingly for the Modi government, nearly all the demonetised currency found its way back to the banking system. Modi swiftly shifted the goalpost saying that the data generated by the process of demonetisation would be mined with data analytics tools to catch those with black money. Scores of “Modiphiles,” eager to uphold his wise words, rushed to pontificate on how big data analytics can help eliminate black money, not realising that the data lacked a crucial piece of information.

Alchemy of Analytics

Big data is typically defined by three “Vs,” volume (hugeness of data), variety (multifariousness of data: audio, video, text, signals), and velocity (rapidity with which data swells). Analytics is the combination of statistical modelling and machine learning. Big data analytics examines large data sets to uncover patterns, correlations, trends and myriad other useful information. While this is true, big data analytics is not a magic wand that could produce something out of nothing.

True, currency notes have a country identifier, denomination, unique serial number, and a mechanism for counterfeit prevention. These data can be captured easily and in real time by cash counting machines provided they are equipped with sensors to detect and store serial numbers of currency notes that are run through them. With these data, notes can then be traced to the end user or account holder. The algorithms can be built to indicate the rough location of hoarded money once the collected data is run through them.

However, as for the demonetisation process data, the fact remains that the data of serial numbers of currency notes deposited was not collected. Serial numbers of notes deposited are not linked to depositors. In the absence of this crucial information, there is no way to construe deposits of money as illegitimate, let alone, trace them to a particular person. The ways in which the old currency was exchanged with new ones are: (i) ordinary people exchanged their hard-earned money standing in queues; (ii) exchanges done through agents paying commissions ranging from 20% to 40%; and (iii) deposits after paying penal tax at 50%. Out of these three ways only the middle one is illegitimate, as it converted black money into white. This happened in two ways: one, through connivance with bank officials, and two, through engaging a battery of poor people to exchange old currency at a commission of around 10%. Crores of rupees have been exchanged with these methods. What could analytics make out of such data? Expectedly, there will obviously be a surge in bank deposits but can it be viewed as illegitimate money? The only thing demonetisation has done is convert the black money of criminals into white.

Fooling the People

As explained in my earlier column (“Demonetisation—Modi Digs a Ditch for the BJP,” EPW, 3 December 2016), black money in cash (that includes jewellery) is just about 5% of all such illegally held wealth. Therefore, if the intention was to trace black money, then currency was an unlikely candidate. The fountainhead of black money is the corporate sector, with its patronage network of politicians and bureaucrats. Tax exemption to donations to political parties, and anonymity of donors to the extent of₹20,000 per donor are still allowed. The political parties thus became a conduit to make black money into white with anonymity for criminals. None other than the Chief Election Commissioner Nasim Zaidi has termed these political parties, numbering today over 1,900, “as conduits for siphoning off black money.” Of course, the main beneficiary is the Bharatiya Janata Party (BJP).

Pretending to catch suspect cases by sifting through big data of currency deposit is like trying to catch fish after letting bulk of them escape through a big hole. Is Modi going to flag the large unexplained deposits before the declaration of demonetisation? After all, his claim of secrecy of the decision is exploded by the media reports that there were huge transactions just weeks prior to the announcement. Who were these depositors? Did they belong to the inner circle of the BJP?

Do you really require big data analytics tools to identify pickpockets in the crowd when there are robber gangs roaming around in broad daylight? According to a report compiled by the Association for Democratic Reforms (ADR) based on the election affidavits of the candidates, the assets of 165 Members of Parliament (MPs) re-elected to the 16th Lok Sabha (of the total 168, the affidavits of three MPs being not clearly available on the Election Commission of India website as per ADR) had on an average risen by a whopping 137% between 2009 and 2014. Modi’s own party topped the list in both assets as well as criminal cases. In Uttar Pradesh, where the BJP won 71 out of the 80 seats, Varun Gandhi saw his assets grow by 625%. In 2009, according to Varun’s affidavit, his assets stood at₹4.93 crore. In 2014 it shot up to₹35.73 crore, an increase of₹30.81 crore. His mother Maneka Gandhi’s assets saw a rise of 105%. If the rise in assets were a proxy for corruption, the BJP clearly scored over the Congress.

The assets of BJP’s re-elected MPs jumped from₹5.11 crore to₹12.6 crore in 2014, an increase of 146%, while the Congress saw an increase of 104%, rising from₹5.66 crore in 2009 to₹5.90 crore in 2014. Modi needs to answer how these politicians claiming to do public service are transformed into financial wizards. Similar wizardry is observable in the bureaucrats, without whom politicians’ wizardry may not be possible. It is an open secret that bureaucrats, particularly those controlling administration, police, and in regulatory posts among others all have huge assets disproportionate to their sources of income. How many of them are ever investigated, let alone, convicted? The vulgar inequality that brings India a dubious distinction of being the most unequal country in the world, with its 57 billionaires owning up 58% of its wealth (Oxfam 2016) is after all not produced by honest money.

Problems with Analytics

Big data analytics, a new paradigm of data-driven decisions, has enormous implications, both positive as well as negative. Philosophically, it spells the “end of theory” (Anderson 2008). Big data looks for the correlation rather than the causation: the “what” rather than the “why.” To those who are enamoured by this new paradigm, a recent White House report “Big Data: A Report on Algorithmic Systems, Opportunity, and Civil Rights” may serve as caution about its risks. It states, “[t]he algorithmic systems that turn data into information are not infallible—they rely on the imperfect inputs, logic, probability, and people who design them.” An earlier White House report had warned of the potential of encoding discrimination in automated and secretive decisions that analytics entail through its complex algorithms. The benefits of big data are seriously tempered by concerns over privacy and data protection. Advances of the data ecosystem turns the power relationships between government, business, and individuals on its head, and can lead to racial or other profiling, discrimination, over-criminalisation, and other restricted freedoms.

While the entire world is concerned with these issues, the Indian government is pushing its digital juggernaut oblivious of its dangers. It is upbeat about Aadhaar data, which it wants to leverage for digitising every transaction with biometric identification. Despite demonstrations by experts that biometrics are unreliable for financial transactions, Modi flaunted BHIM as “your thumb as your bank.” Contrary to its stated objective to create a unique identity, soon after Aadhaar was launched in 2009, an Aadhaar-authentication application programming interface (API) was created, making it available for businesses. As Nandan Nilekani, its architect recently averred, just an “Aadhaar-enabled biometric smartphone” is estimated to create a $600 billion opportunity (Credit Suisse 2016).

There is of course not an iota of consideration to what happens to the privacy of Indians or security of their crucial data. When this issue came up in the Supreme Court in August 2015, the Attorney General had settled it saying that the people of this country do not have a right to privacy. Interestingly, in the case to strike down defamation as a crime, around the same time, the government pleaded exactly the opposite that they had to protect the privacy rights of people. The Supreme Court had rightly restricted the use of Aadhaar card to only six areas—rations in the public distribution system, liquefied petroleum gas, the Jan Dhan Yojana, the Mahatma Gandhi National Rural Employment Guarantee Act, and pensions—and that too, voluntarily. But in utter contempt of it, the government has been bulldozing it to make it mandatory all over. Obviously, it wants to take complete control of our lives in contravention of the constitutional guarantees, and reduce us to be guinea pigs, the subservient automatons, with the application of big data analytics.

When people could quietly endure the disaster of demonetisation, this malfeasance perhaps may only be a minor irritant!

References

Anderson, C (2008): “The End of Theory: The Data Deluge Makes the Scientific Method Obsolete,” Wired, 23 June, www.wired.com/2008/06/pb-theory.

Credit Suisse (2016): “India Financial Sector,” Equity Research Asia Pacific Region, Diversified Financials, Credit Suisse, 29 June, https://www.credit-suisse.com/media/cc/docs/cn/india-digital-banking.pdf.

Oxfam (2016): “An Economy for the 99%,” Oxfam Briefing Paper, January.

Anand Teltumbde ([email protected]) is a writer and civil rights activist with the Committee for the Protection of Democratic Rights, Mumbai.