Pakistan’s tax challenge: four conversations, one problem

A few days after speaking at a tax workshop in Karachi, I found myself having four very different conversations about Pakistan’s tax system.

The first was with a salaried taxpayer whose income is taxed before it reaches his bank account.

The second was with a documented businessperson who felt that being in the system meant being permanently available for notices, reconciliations, audits and explanations. The third was with a smaller operator. His question was blunt: why should he fully document himself if documentation only increases his exposure?

The fourth conversation was quieter, but in many ways just as important. It was with a low-wage worker earning roughly the minimum wage prescribed by law. Between rent, food, utilities and transport, very little remained at the end of the month. Yet his concern was not immediate survival alone. He spoke about his two children and his determination to educate them.

He wanted to send them to a private school in his neighbourhood, not out of luxury but out of a widely held belief: that access to better schooling is one of the few ways to move up in society. The government school available to him, in his view, could not offer the same opportunity.

Even then, he was uncertain. Would a low-cost private school, operating within a parallel education system, really allow his children to compete in the long run? Or, would it simply create another layer of inequality?

His question was not framed as tax policy. It was framed as a struggle for mobility. But it pointed to something deeper. If taxation is meant to redistribute opportunity, then its success must ultimately be visible in services that narrow these gaps.

When access to quality education, health and basic public services remains uneven, the burden of aspiration shifts back to the individual regardless of his income level.

This also raises a more specific question of responsibility. After the 18th Amendment, core services such as education and health lie primarily with the provinces. If taxation is meant to redistribute opportunity, then that redistribution must become visible through provincial delivery.

When quality education and healthcare remain uneven, the question is not only how much is collected, but whether it is translating into outcomes where it matters most.

These four conversations capture Pakistan’s tax problem better than many tables and statistics. The issue is not merely revenue collection. It is visibility, fairness, documentation, trust and the relationship between the state and the citizen.

Pakistan’s tax crisis, in that sense, is not simply a fiscal problem. It is also a governance problem, a documentation problem and a state-capacity problem. Sustainable reform cannot come merely from higher rates, more withholding taxes or more enforcement notices. It requires rebuilding credibility between the state and the citizen.

The salaried taxpayer

A salaried person earning Rs 300,000 per month has annual income of Rs 3.6 million. Under the relevant salary slab, his annual tax comes to approximately Rs 466,000, or Rs 38,833 per month. His monthly take-home income is reduced to around Rs 261,000.

In Karachi, Lahore or Islamabad, after rent, school fees, utilities, groceries, transport, medical expenses and family support, the difference between gross income and real economic capacity becomes very clear.

The salaried taxpayer is also taxed again when he spends. Sales tax on goods, taxes in utility bills, petroleum-related levies and taxes embedded in prices all reduce the value of his already-taxed income. This is why many citizens feel that they are taxed everywhere, even when they cannot always identify the tax separately.

This is where policy discussions sometimes lose touch with lived reality. Gross income is not the same thing as economic comfort. A person may look reasonably comfortable on paper, but after-tax deduction and ordinary household expenses, his real capacity to save, invest or absorb emergencies may be far narrower than the number suggests.

The documented business

The documented businessperson faces a different form of pressure. He is not merely a taxpayer. He is also a collection agent for the state. He deducts withholding taxes, deposits them, issues invoices, reconciles statements, responds to notices, manages audits, follows refunds and deals with multiple tax authorities in case he is a service provider operating in more than one province.

In theory, this is part of formal economic life. In practice, when the burden becomes excessive, documentation begins to feel like a disadvantage. This disadvantage is visible in the marketplace.

A documented supplier issuing a proper invoice may have to charge sales tax, receive payment through documented channels and accept future reporting exposure. An informal supplier may offer a lower cash price, with no invoice, no sales tax trail and no reconciliation burden.

The customer may not think in terms of tax policy. He simply compares prices. If the informal price is lower, the compliant business loses.

When informality becomes a market advantage

This is how informality becomes self-reinforcing. The compliant business loses price competitiveness. The informal business gains a market advantage. The customer becomes accustomed to lower undocumented prices.

The state loses revenue. The burden shifts back to those already visible. The comparison becomes sharper when we add a third person with similar economic capacity but outside effective tax visibility.

A small business operator with taxable income of Rs 300,000 per month may pay around Rs 810,000 annually under the applicable non-salaried rate structure, leaving him with about Rs 232,500 per month. The salaried taxpayer takes home around Rs 261,000. But a person who successfully remains undocumented, understates income and transacts largely in cash may retain far closer to the full Rs 300,000.

This is where the problem becomes more than a tax calculation. The person contributing least may enjoy the highest immediate cash advantage. The system tells the salaried person that visibility is costly, the documented business that compliance reduces competitiveness, and the informal operator that staying outside the system may be financially rewarding. No voluntary tax culture can grow in such an environment.

The point is not simply that one category is taxed more than another. The deeper point is that three persons with similar monthly economic capacity may face entirely different outcomes depending on whether their income is salary, documented business income, or income that remains outside effective visibility.

A further complexity can be added to the above examples if a person is operating in a sector where the receipts are subject to withholding taxes which are otherwise treated as ‘minimum tax’.

The economy is not entirely invisible anymore

But Pakistan’s economy is no longer entirely invisible. The Pakistan Bureau of Statistics’ (PBS’) Economic Census 2023 (formally released in 2025) provides an important new starting point. It records around 7 million economic establishments and captures information such as business activity, location, employment size and classification of economic activity. It also points towards Pakistan’s first comprehensive Business Register.

This should be treated as a 2023 baseline, not as a perfectly current picture of today’s economy. Economic activity changes quickly. Businesses open, close, shift locations, expand or contract. Therefore, the figures need to be updated, extrapolated and cross-checked with current administrative data. But even as a baseline, the Economic Census tells us something important: Pakistan’s economy is not unknowable.

Geo-tagging, digital mapping, establishment classification and business-register development can help the state understand where economic activity exists, where formalisation is needed, and where tax records do not match commercial reality. Combined with digital invoicing, Point of Sale (POS) integration, banking data, provincial sales tax records, customs data, property information and utility footprints, this can become a powerful tool for broadening the tax base.

What household data adds to the picture?

The Economic Census, however, does not directly tell us the income or profitability of each sector. Its value lies in mapping economic activity and establishments. For household income, consumption and living standards, the more relevant source is the Household Integrated Economic Survey. PBS describes HIES 2024–25 as the first fully digital HIES round, conducted from September 2024 to June 2025, covering 32,814 households across Pakistan, including AJK and GB.

This matters for tax policy because taxation should not be designed only by looking at visible transactions or registered businesses. It should also be informed by household capacity, income sources, consumption patterns and vulnerability.

HIES helps explain how households earn and spend. It can show the relative importance of wages and salaries, non-agricultural activities, remittances, crop income, livestock income and other sources. It can also show differences between urban and rural households, and between lower-income and higher-income groups.

This does not mean HIES can be used to assess the taxable income of a particular person. It cannot. But it can help policymakers understand where economic capacity may exist, where households remain vulnerable, and where policy assumptions may be unrealistic.

Read together, the Economic Census and HIES provide two different but complementary pictures. The Economic Census shows where economic activity exists. HIES shows how income and consumption are experienced at the household level.

A sensible tax policy should use both.

Agricultural income tax

This becomes especially relevant in the case of agricultural income tax. Agricultural income tax is constitutionally a provincial subject. Yet, for decades, it has remained one of the most sensitive and uneven areas of Pakistan’s fiscal system. The debate is often reduced to two extremes. One side argues that agriculture remains under-taxed. The other responds that many rural households are small, vulnerable and dependent on uncertain agricultural returns. Both concerns have some force. The mistake lies in treating agriculture as one uniform category.

A small subsistence farmer, a household keeping livestock for survival, a medium cultivator with seasonal marketable surplus, and a large commercial agricultural operator are not the same economic reality. Yet public debate often places them under one broad label: agriculture. This is where data can help.

HIES can show how important crop and livestock income is within household income. The Agriculture Census can help map farms, land utilisation, tenure, cropping patterns, livestock and agricultural machinery. For provincial agricultural income tax authorities, this kind of data should not become a blunt enforcement weapon. Its better use is segmentation.

It can help separate subsistence households from medium farmers. It can distinguish ordinary rural survival activity from large commercial agricultural capacity. It can help identify cases where substantial agricultural income is plausible and cases where agricultural income may merely be used as an explanation for unexplained wealth.

In other words, the objective should not be to tax every rural household. The objective should be to stop agricultural income from remaining either politically untouchable at the top or administratively misunderstood at the bottom.

Data classification before penalisation

The same principle applies across the economy. A neighbourhood tailor, a home-based tutor, a small beauty parlour, a street food vendor, a livestock household, a medium retailer and a large undocumented distributor are not the same economic reality. They may all appear in data, but they should not receive the same policy response.

The first use of data should therefore be classification, not coercion. Technology should help the state ask better questions. It should not merely generate more notices. For agricultural income tax, this means provinces should combine PBS data with land revenue records, crop reporting data, irrigation records, livestock records, market committee data, sugar mill procurement data, cotton ginning data, rice and flour mill procurement data, agricultural lending data, tube-well electricity consumption, machinery ownership and FBR return information where agricultural income is claimed as exempt.

Such data can help test whether declared agricultural income is plausible. But it should not be used to create automatic tax demands. A credible data system must allow explanation, correction and appeal.

Asset-based taxation

The coming budget may also become important for another reason. Pakistan appears to be moving through a moment of reconsideration in relation to asset-based taxation.

(To be continued tomorrow and on Monday)

Copyright Business Recorder, 2026