Effective Personal Income Tax Schedules inBangladesh: Evidence from Administrative Returns

Taxes are the invisible architecture of a modern economy, shaping how money moves from individuals to the state and back again into public services. At the heart of this system lies a simple question: how much of a person's earnings actually goes to the government? In theory, tax laws are written as clear ladders, where income steps up into higher brackets and the rate of tax climbs with each rung. This structure is designed to be progressive, meaning those who earn more pay a larger share. However, the reality of a tax bill is rarely just a matter of applying a single rate to a total income. The final amount a person owes is often shaped by a complex web of deductions, special rules for certain types of income, minimum payments, and credits that are not visible in the headline numbers. Understanding the true burden requires looking past the official rules to see what actually happens on the tax returns people file.

A recent study by Khademul I. Chowdhury, a researcher at the Australian National University, turns a microscope onto this reality in Bangladesh. The work relies on a massive, anonymized collection of more than two million individual tax returns from the 2022–23 tax year. This dataset covers roughly 57 percent of all returns submitted that year, offering a rare, large-scale glimpse into how the tax system functions in practice for a developing economy. The researcher's goal was not to judge whether people are paying the right amount or to measure how much tax is being hidden, but simply to map the relationship between the income people report and the tax they actually pay. By analyzing these records, the study reveals a tax landscape that looks quite different from the official schedule printed in law books.

The official tax schedule in Bangladesh operates on a graduated scale. For most taxpayers, the first 300,000 Bangladeshi taka of income is tax-free. Above that threshold, the rate climbs in steps: 5 percent, then 10, 15, 20, and finally 25 percent for the highest earners. This is the "slab schedule," a mechanical benchmark that assumes a taxpayer's liability is calculated solely by applying these rates to their total income. The study compares this theoretical benchmark against the actual numbers found on the tax returns. The results show that the real-world tax burden is far more concentrated than the official rates might suggest. While the average effective tax rate remains low for the majority of filers, it rises sharply only at the very top of the income distribution.

The data paints a picture of extreme concentration. The top 20 percent of taxpayers in the database are responsible for 92.4 percent of the total tax liability recorded. Even more striking, the top 1 percent alone accounts for nearly half of all the taxes paid. For the bottom 60 percent of filers, who collectively earn about 27 percent of the reported taxable income, the share of taxes paid is a mere 2.2 percent. This disparity highlights that the tax system, as it is actually experienced by those who file returns, places a heavy weight on a small group of high earners while the vast majority pay very little.

When the researcher looked at the average tax rate across the income distribution, a surprising pattern emerged. For a large portion of the lower and middle-income filers, the actual tax rate they paid was higher than what the official slab schedule would predict. This seems counterintuitive, given that the official rules offer a zero rate on the first 300,000 taka. The explanation lies in the complexity of the tax code. Many taxpayers have income subject to withholding taxes, final tax provisions, or minimum tax rules that apply regardless of their total income. These mechanisms can create a tax bill even when the general schedule suggests there should be none. As a result, the effective tax rate for many people near the bottom of the filing population is positive, whereas the simple benchmark would show zero.

However, as income rises toward the top, the relationship flips. At the highest income levels, the actual average tax rate paid is lower than the marginal rate suggested by the official slab schedule. The study found that while the official schedule might call for a 25 percent rate on the highest earnings, the effective rate observed in the data is lower. This gap is likely due to the various deductions, rebates, and preferential treatments available to high-income earners, which reduce their final liability. The study also calculated how tax liability changes as one moves up the income ladder, finding that the slope of this increase is generally flatter than the steep steps of the official rate schedule. This indicates that the official rates do not fully capture the incentives or the actual financial impact on the wealthiest individuals.

To make sense of these patterns, the researcher fitted mathematical curves to the data, creating smooth descriptions of how tax rates change with income. One of the key findings was that the shape of this curve depends heavily on who is included in the calculation. When the analysis included all filers, the curve appeared relatively flat, suggesting a gentle increase in tax rates as income grows. But when the researcher restricted the data to only those earning at least 300,000 taka—the general exemption threshold—the curve became much steeper. This shift demonstrates that the lower end of the income distribution, where small fixed taxes and special rules create irregularities, significantly distorts the overall picture. The study concludes that there is no single, universal description of the tax schedule; the shape changes depending on which group of people is being examined.

The research also clarifies what these numbers do and do not tell us. The data comes from filed returns, meaning it reflects the behavior of those who participate in the formal tax system. It does not capture the income of those who do not file, nor does it reveal unreported income. Therefore, the findings describe the tax burden of the observed population, not the entire country. The study explicitly avoids interpreting the differences between the official rules and the actual payments as evidence of tax evasion or enforcement gaps. Instead, it attributes the variation to the complex interplay of legal provisions, reporting choices, and the specific composition of income sources. The gap between the simple slab schedule and the real-world outcome is a feature of the system's design, not necessarily a failure of compliance.

Ultimately, this study provides a crucial set of tools for economists and policymakers who build models to understand how taxes affect the economy. By offering a precise map of the effective tax rates in Bangladesh, the research allows for better simulations of how changes in tax policy might influence behavior and revenue. It shows that relying solely on the official rate schedule is insufficient for understanding the true economic landscape. The real tax system is a mosaic of rules and exceptions that creates a unique burden for every income level. For the vast majority of filers, the burden is light and irregular; for the wealthy, it is heavy but softened by the very mechanisms designed to make the system progressive. The study leaves us with a clear understanding that in Bangladesh, as in many places, the tax you pay is rarely just the rate you see on paper.