Digital transformation for FBR transparency
The Federal Bureau of Revenue is in the eye of the storm once again
The writer is the former chairman of NADRA
The Federal Bureau of Revenue (FBR) is in the eye of the storm once again. First, it has reportedly missed the revised revenue collection target for the first half of the current fiscal year by a wide margin despite several measures and double-digit consumer inflation. Second, the FBR’s reputation to combat corruption and fraud is again marred with charges of bad governance, given its own employees were found involved in corruption.
Tax evasion, fraudulent refund claims and issues related to mis-declaration of imported goods are not just personal crimes. With the current economic crisis when demands for government services increase but those willing to pay for it are disproportionately less, the result is also a crime against society and democracy. This leads to a broader economic crisis that limits the government’s ability to provide essential services to the people.
Good governance is impossible without adequate state capacity to collect revenues. Digital technologies can enhance the FBR’s capacity if its leadership focuses on broader accountability systems around technological deployment and structural changes. Big data and analytics are key considerations in this ongoing challenge, as they offer smart but quick reforms with the ability to identify and thwart efforts to commit tax fraud and abuse in terms of recording value of imported goods. Through analytical models, this data-driven insight can trigger proactive investigations, allowing the FBR to prevent improper payments, losses, and improve overall service levels.
Pakistan Revenue Automation Ltd (PRAL) was supposed to increase efficiency of tax and revenue regimes using Information and Communication Technologies (ICT). As a member of the Board of PRAL (2012-14), I challenged the laissez-faire approach and the executive management with innovative ideas of digital solutions. I pressed hard to make my point that digital systems were required to transform the institution. But alas, the rich rent seekers and intermediaries in the bureaucratic labyrinth were and still are very well protected!
The good news is that big data and analytics offer key opportunities to gain an advantage in the fight against tax evasion and fraud. Combined with evidence of identity (EOI) data — e.g. the CNIC and biometrics — it allows tax authorities to identify patterns of fraud and take proactive steps to mitigate losses. But the first step in this regard is the digitisation of records and integrating the databases of the government’s revenue streams.
In 2012, we used the analytics model and found that only a fraction of Pakistanis paid income taxes. By integrating data across various government databases, reconciling it with the citizen database with NADRA, big data analytics helped identify 3.5 million tax evaders. It was estimated that if a basic minimum tax rate was applied, Pakistan would have $3.5 billion right away — an amount outweighing Pakistan’s annual request for loans to the IMF. I hope the FBR made use of this data!
The integration of data across multiple sources provides greater visibility into various analyses, enhancing the connection between government agencies and tax authorities. The resulting predictive and prescriptive models can be deployed on high performance big data platforms which rely on large parallelism and scalable distributed storage systems to quickly scan, filter and analyse different data streams.
Embedding these models within applications provides operational intelligence to help users anticipate tax evasion, fraud and abuse while improving services to citizens. More importantly, it can allow tax collectors to take advantage of these opportunities within a real-time window. These models are currently being used in many developing and developed countries.
Smart governments from developing countries are catching on: the solution for solving complex service delivery problems lies in properly mined data and analytics. But deploying ICT technologies needs the will of the government as the blowback from rent seekers within the system will test its resolve!
Published in The Express Tribune, February 22nd, 2020.
Tax evasion, fraudulent refund claims and issues related to mis-declaration of imported goods are not just personal crimes. With the current economic crisis when demands for government services increase but those willing to pay for it are disproportionately less, the result is also a crime against society and democracy. This leads to a broader economic crisis that limits the government’s ability to provide essential services to the people.
Good governance is impossible without adequate state capacity to collect revenues. Digital technologies can enhance the FBR’s capacity if its leadership focuses on broader accountability systems around technological deployment and structural changes. Big data and analytics are key considerations in this ongoing challenge, as they offer smart but quick reforms with the ability to identify and thwart efforts to commit tax fraud and abuse in terms of recording value of imported goods. Through analytical models, this data-driven insight can trigger proactive investigations, allowing the FBR to prevent improper payments, losses, and improve overall service levels.
Pakistan Revenue Automation Ltd (PRAL) was supposed to increase efficiency of tax and revenue regimes using Information and Communication Technologies (ICT). As a member of the Board of PRAL (2012-14), I challenged the laissez-faire approach and the executive management with innovative ideas of digital solutions. I pressed hard to make my point that digital systems were required to transform the institution. But alas, the rich rent seekers and intermediaries in the bureaucratic labyrinth were and still are very well protected!
The good news is that big data and analytics offer key opportunities to gain an advantage in the fight against tax evasion and fraud. Combined with evidence of identity (EOI) data — e.g. the CNIC and biometrics — it allows tax authorities to identify patterns of fraud and take proactive steps to mitigate losses. But the first step in this regard is the digitisation of records and integrating the databases of the government’s revenue streams.
In 2012, we used the analytics model and found that only a fraction of Pakistanis paid income taxes. By integrating data across various government databases, reconciling it with the citizen database with NADRA, big data analytics helped identify 3.5 million tax evaders. It was estimated that if a basic minimum tax rate was applied, Pakistan would have $3.5 billion right away — an amount outweighing Pakistan’s annual request for loans to the IMF. I hope the FBR made use of this data!
The integration of data across multiple sources provides greater visibility into various analyses, enhancing the connection between government agencies and tax authorities. The resulting predictive and prescriptive models can be deployed on high performance big data platforms which rely on large parallelism and scalable distributed storage systems to quickly scan, filter and analyse different data streams.
Embedding these models within applications provides operational intelligence to help users anticipate tax evasion, fraud and abuse while improving services to citizens. More importantly, it can allow tax collectors to take advantage of these opportunities within a real-time window. These models are currently being used in many developing and developed countries.
Smart governments from developing countries are catching on: the solution for solving complex service delivery problems lies in properly mined data and analytics. But deploying ICT technologies needs the will of the government as the blowback from rent seekers within the system will test its resolve!
Published in The Express Tribune, February 22nd, 2020.