Combating e-invoice fraud with big data technology

Combating e-invoice fraud with big data technology

The enormous volume of data generated also creates major challenges in fraud control.

An e-invoice generated from a cash register at Hồng Hà Stationary Store. — VNA/VNS Photo

The rise of invoice trading and tax fraud in the digital environment is posing major challenges to tax administration and revenue protection efforts.

Against this backdrop, tax authorities are increasingly turning to big data technologies to harness the vast amount of information generated by the e-invoicing system, enabling them to manage risks more effectively and detect violations at an early stage.

According to tax authority statistics, 100 per cent of eligible businesses, organisations, household businesses and individual business operators have adopted e-invoices.

The system has received and processed approximately 23.3 billion e-invoices since implementation began.

In particular, 463,855 business establishments have registered to use e-invoices generated from cash registers, an increase of 474 per cent compared to the same period.

Nguyễn Thế Bình, chief accountant at Quang Ngai Sugar Joint Stock Company, told the press that the rollout of e-invoices has been one of the tax administration's most significant achievements in advancing the digital transformation.

Beyond changing tax declaration methods, e-invoices help businesses optimise accounting operations, data management and system administration, while also enabling tax authorities to improve management efficiency, he said.

However, the enormous volume of data generated also creates major challenges in fraud control.

While violations previously occurred mainly through paper invoices, the electronic environment now allows invoices to be issued and circulated almost instantly. Within a short period, fraudsters can conduct thousands of transactions, overwhelming traditional inspection methods.

In recent years, law enforcement agencies and tax authorities have uncovered a number of large value-added tax invoice trading networks.

According to the General Department of Taxation (GDT), the use of e-invoices has created a vast data source for modern tax administration. However, the enormous number of daily transactions and increasingly sophisticated fraud schemes require authorities to accelerate the use of digital technologies, AI and big data analytics to identify risks at an early stage.

The Fourth Industrial Revolution drove a strong shift from traditional tax administration to data-driven governance, said Nguyễn Thu Trà, head of GDT's Compliance Management and Taxpayer Service Division.

Instead of relying primarily on scheduled audits and inspections, the tax authority is now gradually transitioning toward a risk management model based on data analysis. 

Technology experts have said that big data’s greatest advantage lies in its ability to collect, store and process enormous volumes of information from multiple sources.

Beyond invoice data, the system can also integrate information from business registration databases, banking systems, customs authorities, social insurance records, legal representative information and tax compliance histories.

This enables authorities to build comprehensive business profiles and identify unusual patterns.

Another highly effective application is business network analysis. Rather than examining each company individually, big data makes it possible to map relationships among thousands of businesses through invoice transactions and financial flows.

This allows tax authorities to detect groups of companies that frequently trade invoices with one another, share the same address or legal representatives, or exhibit unusual financial connections.

Experts consider this a powerful tool for dismantling organised invoice trading networks, which are often concealed through multiple layers of transactions and are difficult to detect using conventional audit methods.

The application of big data also helps the tax sector shift from a post-audit approach to a proactive monitoring and pre-audit model.

In the past, many fraud cases were only discovered months or even years later. Today, suspicious transactions can be identified almost immediately after an invoice is issued.

This significantly reduces the risk of budget revenue losses while improving the efficiency of Government resources.

Instead of conducting broad inspections, tax authorities can focus on high-risk entities, thus reducing unnecessary burdens on legitimate businesses.

Nevertheless, applying big data in tax administration still faces numerous challenges. For one, data remains fragmented across various ministries and agencies and has yet to be fully standardised.

Data integration and sharing mechanisms among relevant authorities also require further improvement.

In addition, substantial investment is needed in technological infrastructure, data security and the development of a highly skilled workforce.

GDT Director General Mai Xuân Thành said the tax sector has made important progress in digitising tax administration through the successful implementation of e-invoices and electronic platforms serving taxpayers.

However, if it stops at its current achievements, the sector risks falling behind the rapid evolution of global tax governance models, he said.

It is thus necessary to confront existing realities, accurately assess current gaps and recognise that digital transformation is not merely a technological task, but a comprehensive restructuring of tax administration methods, he added. 

Bizhub