Author ORCID Identifier

Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Graduate School


Public Policy and Administration

First Advisor

Dr. David R. Agrawal

Second Advisor

Dr. William H. Hoyt


Value added tax (VAT) based on credit invoice system is the most common consumption tax in the world. Despite its self-regulating nature, VAT faces challenges in developing countries who have limited state capacity to check evasion and enforce tax on informal sectors of the economy. The tax authorities introduce policy interventions that can target the evasive behavior of firms interacting with informal sectors. My dissertation seeks to provide insight into three such policy reforms in Pakistan’s VAT regime. Therefore, this dissertation is composed of three essays.

In first essay of my dissertation, titled “Using Computerization to enforce VAT: Evidence from Pakistan”, I study a policy intervention which empowered a computerized system to check invoices and reject input tax claims based on risk-based criteria. I use administrative tax data for the universe of VAT returns filed in Pakistan from tax year 2009 to 2016 to estimate the impact of this reform on the firms operating domestically. Using the exporters not subject to the reform as a control group, I find that the input tax claims fell by 2.36 million Pak Rs. per treated firm, representing a decline in input tax claims to the tune of Pak Rs. 86 billion. Firm heterogeneity analysis by business activity and firm structure shows a decline ranging from 30% to 90%. Surprisingly, the corporations and partnerships also show significant reduction in input tax claims from 50-70%. Contrary to the expectations, the huge volume of evasion shows that VAT implementation in limited tax capacity regimes may not yield the expected revenue efficiency gains.

Second essay of my dissertation titled, “Is Minimum the Maximum? Tax Burden on Informal Sector in VAT: Evidence from Pakistan”, analyzes another policy reform. In developing countries, a substantial amount of revenue at import stage is now collected from VAT instead of traditional import tariffs. This modern approach assumes negligible VAT evasion at post-importation stage. I test this assumption through universe of monthly VAT returns filed in Pakistan for tax years 2009 to 2016 to estimate evasion by firms exclusively engaged in imports. I utilize kinks produced by minimum value addition thresholds to estimate evasion of VAT post-importation. I estimate an average evasion rate of nearly 78%. Using changes in thresholds over years, I provide evidence that this minimum tax collection is the best-case scenario for revenue efficiency. The firms show strong bunching at or below threshold with about 40-60% of the firms showing bunching behavior. My results support the view that, absent deviations from standard, replacing high import tariffs with VAT would decrease welfare.

Third essay of my dissertation titled, “The Deterrence Value of Tax Audits: Estimates from a Randomized Audit Program”, analyzes a randomized audit program. It is a joint project with Michael Best and Mazhar Waseem. In modern tax systems audit is the sole instrument through which the tax authority can detect noncompliance and create deterrence. We exploit a national program of randomized audits covering the entire population of VAT filers from Pakistan to study how much evasion audit uncovers and how much evasion it prevents by changing behavior. While audit uncovers a substantial amount of evasion (the evasion rate among firms in the bottom three size quartiles is more than 100%), it does not deter future cheating. Examining more than ten intensive and extensive margin outcomes, we detect no effect of audit on proximate or distant firm behavior. Our results suggest audits are sub optimally utilized in checking mechanical violations of law instead of creating deterrence against evasion.

Digital Object Identifier (DOI)