How do you pay for a programme like PTI's Ehsas?

Just raising more revenue won't turn Pakistan into a welfare state.
Published April 24, 2019

Some eight months into his premiership, Imran Khan has finally given a tangible indication of what his vow to turn the country of 220 million into a welfare state means.

It turns out it means a lot: creating jobs, addressing elite capture, establishing safety nets and human capital development, all packed into Ehsas, a broad social welfare plan.

From where it stands now, Ehsas seems to be more than a single-government programme. It indicates a fundamental shift in the extent to which the government is willing to intervene in the market and the society.

Any analyses of the programme itself can wait until more information on specific policy actions is available; for now, there is an essential question to ask: how can Ehsas be funded?

Why a welfare state?

Let’s start by asking what justifies an intervention of this nature by the government.

The most apparent reason is the need for redistribution — taking the gains from those who are economically stronger and redistributing them to those who are economically weaker. This redistribution happens in most countries, but to significantly varying degree.

Even the most robust economic growth leaves some people behind — a rising tide does not lift all boats — but more so in Pakistan, where growth has at best been unsustainable and unequal. A large proportion of the population experiences various dimensions of poverty (according to one estimate, one in four Pakistanis lives in severe poverty).

What makes poverty complicated is that it is more than just a lack of money; healthcare, education and access to credit services all influence poverty — and expansive welfare states focus on providing most of them.

Ehsas: Show me the money

Apart from poverty, another way of looking at the role of modern welfare states is their ability to provide an opportunity for people to share risk systematically. Even if you aren’t poor, you can benefit from sharing this risk when going through shocks, such as during unemployment.

Consider this: when you’re healthy and have a job, you pay the state, which allows someone else who might be going through an illness to benefit from some form of assistance. Tomorrow, you can benefit from the welfare state if you’re in a similar situation.

Funding a welfare state

The key determinant to a state’s ability to redistribute economic gains and provide a mechanism to share risk between citizens lies in its ability to raise revenue through taxes. Data backs this: welfare states collect a large proportion of their GDPs in taxes.

Norway collects 38 per cent, Denmark about 45pc, the United Kingdom about 33pc and Sweden about 44pc. Pakistan, in comparison, only collects about 12pc of its GDP in taxes. The chart below illustrates this.

This means that while the government might want to turn Pakistan into a welfare state, it doesn’t have much financial capacity to undertake any substantial investment to that end, especially when this lack of revenue is combined with rising debt repayment and defence spending. It is not possible to be a debt-ridden security state and a welfare state at the same time.

So, what can be done? One way to change this is how the government spends the revenue it already collects. Call it expenditure-side reforms. Let’s look at two heads of the budget closely: defence and development.

The 2018-19 budget allocated 21pc of the total budget outlay for defence (or about 3.2pc of the GDP). This allocation doesn’t include a large chunk which is spent on pensions for soldiers and military hardware acquisitions, nor the actual defence spending which seems to be rising faster than projected.

Now, look at development. The 2018-19 budget allocated about 19pc of the budget for development, such as infrastructure, hospitals and so on. This head has since been drastically cut by the new government and will form an even smaller proportion of actual spending when the fiscal year ends in a few months.

To become a welfare state, this trend would have to reverse, with the government cutting defence expenditure and allocating more money for welfare projects — while also reforming what is meant by development spending. This would mean building and incorporating institutions which allow risk-sharing to occur as part of this or a different head altogether.

A decrease in the defence budget is likely to have political costs associated with it and I will not go into them. A more feasible solution, then, might be to capture a larger piece of the economy through taxes. Call them revenue-side reforms.

What is holding us back from raising more taxes?

Let’s split them into three broad reasons.

First, the government just can’t. This is the answer we might be thinking about, and it is a large part of the answer. We have an ineffective enforcement regime: the Federal Board of Revenue and provincial taxation departments have limited capacity to enforce tax obligations.

Some of this is due to the absence of people with the right skills, the technology or the right set of incentives to encourage tax officials to work effectively. This can be solved through investments and creating a better set of incentives for tax officials.

We have evidence from an experiment in Punjab that performance-based pay for tax collectors led to a significant increase in tax revenue. So, tweaking incentives could have a positive impact, but there is another dimension to these constraints.

Evidence from elsewhere shows the importance of information in determining the state’s ability to tax — in other words, what if, due to the structure of our economy, the government can’t enforce tax obligations properly?

In most modern welfare states, the state can enforce tax obligations because the economy is well-documented; the state knows how much income you’re making or assets you own.

In Pakistan, however, a large chunk of the economy is informal. This is the reason why we rely so much on indirect taxes, which don’t require much information, but end up harming the poorest the most.

Also read: Why Pakistan will go to the IMF again, and again and again

Second, the government doesn’t want to. In theory, if we want to raise more taxes, we should be able to make the necessary investments — after all, we did make such investments in building a large modern military.

But we haven’t. The reason for this might be political: it is not in the interests of those who control political power to invest in tax-raising capacity.

There is some evidence here. Tim Besley and Torsten Persson show that countries where the executive (who runs the country) is exposed to strong institutional constraints — independent judiciary, free press and political competition — tend to collect higher tax revenues than countries where constraints on executive power are weaker.

Countries with strong executive constraints are also more likely to rely on income taxes than other, more indirect forms of taxes. Why is this?

It could be that in such countries, political leaders don’t want to raise significant (specifically, direct) taxes so they cannot be held accountable by the people. Evidence shows that people respond to taxation by lobbying the state for better governance, but this is speculation on my part.

Third, people don’t want to give taxes to the government — after all, only about a million people actively file for annual tax returns in Pakistan. Some degree of compliance can be enforced through tax authorities, but some would have to be a result of voluntary compliance arising from factors such as trust in the state.

Experiences from around the world allude to factors like intrinsic motivation (“we really want to pay taxes because we want to be honest or patriotic”), social norms (other people guilting, shaming or just nudging us to pay taxes), reciprocity (the relationship between us and the state) and culture (long-run social norms).

But how much remains debated. For example, an experiment from Denmark shows that the tax evasion rate is high for self-reported income (that is, when people have the opportunity to evade taxes) but since over 90pc of people are exposed to some form of third-party reporting, they cannot evade even if they want to. Voluntary compliance might be overrated.

Easier said than done

The answer is very likely to be, in part, all three, and any move towards a large-scale welfare programme must combine policy actions to increase revenue through better enforcement, political commitment in investing in tax-raising capacity and building stronger compliance. Just raising more revenue won't turn Pakistan into a welfare state.

We will also need to think about how we spend. If the increase in revenue is coupled with an increase in non-welfare spending, whether it is significantly higher operating expenditures or more spending for defence, there is little use in increasing revenue.

This requires political commitment, and alas in Pakistan’s case, that has to be made not just by politicians.

Illustration by Soonhal Khan

Are you examining Pakistan's social safety nets? Share your insights with us at