Debating political economy of the budget

by Asad Sayeed and Kabeer Dawani 

Two rupee coin by Rehman Chughtai

In the last week of October, an event was hosted by a multilateral organization in collaboration with a local think tank to discuss ‘The Political Economy of the Budget.’ The invite was a pleasant surprise for two reasons.  First, it is unusual in Pakistan to come across a seminar with the ‘political economy’ prefix as part of it. Budgets are especially amenable to political economy analysis given the explicitly political nature of resource allocations and revenue collection. Second, seminars and discussions on the budget generally begin in the second quarter of the year. Debating budget related issues so early in the fiscal year is also a welcome initiative.  

In hindsight, we expected too much from a public event in Pakistan. As the general drift of discussion goes, ‘politics’ is used as an unwelcome addendum to allocation of resources based on some form of pristine economic logic (usually based on neo-classical criteria). This is how the discussion carried on during the panel discussion and by and large that is how the accompanying document also treated the subject. 

The discussion was driven by issues highlighted through exhibits. It was unfortunate though that without proper contextualization, the numbers put up on the exhibits tended to mislead the audience. There was one exhibit which stated that 92% of the Federal Government budget goes to debt servicing, defense, transfers to provinces and subsidies. The general drift of the argument was the lament that these locked in expenditures eat up such a large chunk of the budget that they leave little for expenditures on public welfare.  This line of argument masks the fact that after the 18^th Constitutional Amendment and the 7^th NFC Award, the responsibility for social sector expenditures and the finances for undertaking this responsibility have been shifted to provincial governments in their entirety. It was also not clearly pointed out that the exhibit refers to only current expenditure of the Federal Government and not its development component.  The biggest welfare program of the Federal Government – the Benazir Income Support Program – is for some inexplicable reason hidden in the development budget. 

But more importantly in the post 18^th Amendment and 7^th NFC environment, 92% of spending on the above ‘non-welfare’ heads is expected and also part of the bifurcation of jurisdiction across tiers of government envisaged in the 18^th Amendment. In fact, as Figure 1 demonstrates, if we add law enforcement related expenditures of the federal government (Rangers, Frontier Constabulary etc.) and expenditure on ‘organs of the state’ – the parliament, Supreme Court, President and Prime Minister’s house and secretariats – this cost escalates to 96% of the current expenditure on the part of the Federal Government. 

Figure 1: Federal Current Expenditures (Source: Annual Budget Statement 2014-15 and Budget-in-Brief 2014-15)

Now, say, for whatever reason, the structure of the NFC is such that the federal government is able to reduce any of this ‘locked in’ expenditure, the principal beneficiary will be provincial governments and by implication expenditure on social services. 

The relevant political economy issues about the opaqueness and lack of transparency of the defense budget were also left out of the discussion.  The most relevant issue in this realm, however, is the large expenditure on energy subsidies. As we see in Figure 2, more than three fourths of Federal Government subsidies have gone to the energy sector.  A political economy debate would delve into determining the beneficiaries of this subsidy: Is it the rich, the poor, the middle classes, an ethnic group or interest group(s) that corners these subsidies? 

Figure 2: What is subsidized? (Source: Budget-in-Brief 2014-15)

Another exhibit at the seminar stated that agriculture contributes 0.5% of taxes as a percentage of its sectoral contribution to GDP. Consequently, most of the discussion tended to highlight the low collection of agriculture tax as the most important reason for Pakistan’s abysmally low tax-GDP ratio. That collection of the agriculture income tax is extremely low is a fact; however, the issue was neither contextualized appropriately, nor was its political economy implications brought to the fore.  

First of all, the tax on agricultural income is administered provincially whereas all other taxes on income are administered by the Federal Government.  Agricultural tax has been consigned to the provincial domain since before partition and all three Pakistani Constitutions have treated it as such. India also treats the agricultural tax as a provincial subject.  It is a relevant question then as to why this anomaly on agricultural tax has existed for so long, not only across Constitutions but also the number of occasions where the Constitution has been amended in Pakistan’s history by both civil and military governments? 

The other relevant question is about clarifying whether the agricultural tax is primarily a question of equity or of revenue yield.  Recent research by Dr. Anjum Nasim estimates that the revenue potential of the agricultural income tax is not as much as it is often made out to be; the revenue generated across all provinces would have been between Rs. 80 billion to Rs. 112 billion for the year 2009-10. Taking the upper limit, this would have amounted to less than 8% of total tax revenues for that year. Although not insignificant when compared to existing agriculture tax collection, it surely does not appear to be the most buoyant tax. 

The case for the imposition of an appropriate agricultural tax for reasons of equity is, however, compelling.  This in turn opens up a broader array of issues regarding equity in taxation also. It is not only the agriculturists who avoid paying taxes on their incomes and assets but urban property dwellers, the services sector and the corporate sector also systemically evade taxes.  In the urban property market, it is routine that property transactions are registered at one third the value of the actual transaction.  Similarly, while the services sector has a share of 50%+ in GDP, its share in income taxes is a mere 5.7%.  A World Bank study estimated that the potential corporate tax for the year 2007-08 was Rs. 450 Billion less than what was paid. 

Matters pertaining to government revenues and expenditure priorities are perhaps the most publicly visible and relevant, and the importance of debating them is self explanatory. The above are some illustrative examples of questions that need to be asked and answers sought through informed debate and discussion.