Electronic invoice monitoring systems to improve tax compliance: Evidence from Pakistan

Improving revenue collection and tax capacity is one of the toughest challenges for most low- and middle-income countries. Technology can be a useful tool to combat weak documentation and a lack of transparency in such economies. However, the impact of such interventions across literature has varied per the overall compliance regime.

In our research, we aim to quantify the effectiveness of an invoice monitoring system launched in Punjab, Pakistan, and to empower the government in making an informed decision on the scale and importance of the system in the future. We collaborate with the Punjab Revenue Authority (PRA) in Punjab, Pakistan. PRA is responsible for collecting the Sales Tax on Services in Punjab (a value-added tax), which accounts for more than 60% of the provincial tax revenue.

Electronic Invoice Monitoring System

PRA introduced an Electronic Invoice Monitoring System (EIMS) to improve tax yields and prevent tampering of reported sales/transaction data. The system collects invoice-level data from service providers in real-time, generating a unique digital invoice for each transaction, which is recorded on the server of the tax authorities. The system was rolled out across Punjab in various phases, focusing on a handful of services. The law has also defined an eligibility criterion for the adoption of this technology. All registered taxpayers in designated service sectors having an annual turnover of PKR 10 million and above are liable to be monitored through EIMS.

We investigate -

1. How did the introduction of the EIMS impact the reported sales and tax liability of treated firms?
2. What are the major considerations and constraints behind a firm’s decision to adopt the EIMS?

Data and methodology

The study aimed to evaluate the impact of the EIMS intervention on the restaurant sector, which contains the largest number of EIMS adopters across service categories. The study used anonymised sales tax returns provided by the PRA from 2012-2021 and included data from about 2,500 total restaurants. The data is on a monthly level for each firm beginning from the first month the firm gets registered, and the rollout of these machines started in February 2020. To study the impacts of adopting EIMS, we analyse how a range of tax-relevant outcomes change around the time that a taxpayer begins to use EIMS. For this, we need a control group of taxpayers that did not adopt the EIMS system to estimate the extent to which the sales reported by taxpayers that do adopt EIMS would have grown even if they had not adopted EIMS. We use the taxpayers who are eligible to adopt EIMS but had not, as of December 2021, adopted EIMS.

Results

Our findings suggest that within three months of adopting EIMS;

1. Reported sales increase by ~43%
2. Tax liability goes up by ~27%
3. Value added per unit of output increases by ~0.4

The results suggest that the introduction of the EIMS system had a significant positive impact on the reported sales and tax liability of treated firms. Adopters of the EIMS system had higher reported sales, input costs, and tax liabilities compared to both notified and not notified non-adopters. The results suggest that the EIMS intervention increased tax compliance among adopters, resulting in higher reported sales and tax liability.

Barriers to adoption

The large increase in reported revenue makes it imperative to understand the key barriers to adoption faced by firms. For this, we conducted a series of focus groups with taxpayers that had adopted the machines and with those that remained non-compliant.

• Taxpayers cited a lack of incentives as the biggest barrier to adoption of EIMS. Taxpayers mandated to adopt did not receive any tax breaks or other financial incentives under the current policy.
• The behaviour of non-compliant firms also negatively affected adoption of these machines. Taxpayers complying with EIMS might be at a price disadvantage compared to those that do not adopt the machines, nudging competitors towards non-compliance.
• There are no consumer incentive schemes; both to deter taxpayers from non-compliant behaviour and to increase consumer participation in the enforcement process.
• Many taxpayers do not comply with the machines due to IT-related compatibility issues as EIMS requires integration with existing sales software.
• Due to the legal requirement of maintaining manual tax records, the effect on reducing compliance costs through EIMS is also minimal.

Policy implications

Our analysis shows a large increase in reported sales (~43%)  which could be offset by taxpayers simply increasing the input costs to minimise the increase in tax liability. However, our positive estimate of tax liability (~27%) clearly shows that the government is gaining revenue through this intervention. This evidence can be used to increase adoption of EIMS across eligible taxpayers and it may even lead to the expansion of the machine to other service sectors.

Incentive structures can be used to increase acceptance of these machines. These incentives can work at the level of the taxpayer and the consumer to ensure that the accountability chain is not broken at any level. It could also ease the enforcement burden of the tax authority. Removing the need for manual records could act as a financial incentive for compliant firms since it can be costly to maintain and store manual invoices over a few years.

There needs to be greater focus on re-evaluating enforcement strategies to maximise uptake of the machine. Experimentation and research in this area would be useful from a policy standpoint.