I often come across the notion that our defence budget is the second-largest expenditure after debt servicing, far exceeding civilian government budget and it has been increasing abnormally over the years. Recently, I decided to take an objective look at it and see whether this impression is valid. And this is what I found.
The civilian government’s share in the federal budget 2018-19, excluding debt servicing, is 74% higher than our defence budget (including military pensions).
A true comparison of civilian versus military budgets, however, should look at the overall public-sector resource pie, including provincial budgets. This way, it is also easier to look past changes in federal versus provincial split of resources and functions, before and after the 18thamendment.
When looked at the overall public-sector expenditure pie (excluding debt servicing), military budget claims 1/6th of it. For every rupee spent by the military, the civilian government spends Rs4.6. Even more interestingly, this proportion was exactly the same 10 years ago, meaning that both the civilian and military budgets have grown at exactly the same rates.
In fact, the ‘running of civilian government’ head at the federal level, mostly covering salaries and employees’ expenditure, grew at a much faster pace of 16% a year, even after taking into account significant thinning post 18th amendment. The ‘defence affairs and services’ on the other hand grew by 13% a year.
Moreover, the World Bank data shows that in the last 30 years, Pakistan’s military expenditure as percentage of GDP has come down by 50%. This rate of reduction was in fact faster than that for South Asia.
If this is true, then what prevents people from seeing this? A big part of the confusion comes from how budget heads are categorised. Undoubtedly, the ‘defence affairs and services’ head is double than that of ‘running of the civilian government’. But this comparison is not fair.
The allocation for defence includes both recurrent (salaries, etc) and development (new equipment, facilities, etc) budget. The civilian side, on the other hand, has separate heads for recurrent and development budgets. Even within the recurrent budget, the head for ‘running of the civilian government’ does not include civilian pensions, grants or even subsidies. The grants include transfers to railways to meet its losses and those to AJK and Gilgit-Baltistan, whereas the subsidies include allocations for Wapda, PEPCO, Utility Stores, etc. These heads definitely belong to the civilian government budget.
On the other hand, some of the heads that are presently booked outside ‘defence affairs and services’, such as military pensions and development allocations for defence and defence production divisions should be clubbed with defence budget. The estimates given above were calculated after making these adjustments.
One wonders then what is really behind the country’s economic woes. Firstly, both the defence and civilian budget allocations during the last decade have grown by 14% per annum, far exceeding the average GDP growth of 4%. Secondly, the persistent fiscal and current deficits have contributed to piling up of domestic and foreign debt. Thirdly, some parts of the public sector expenditure have outpaced others, creating an imbalance. The federal pensions, for instance, have grown by 18% per annum since 2009, whereas pensions in Punjab grew by 23% a year in the last five years. Lastly, while the tax revenues managed to grow at 13% barely keeping pace with expenditure, the non-tax revenues increased only by 3.5%, slowing down overall yearly revenue growth to 11%.
Unfortunately the civil-military debate has kept our focus away from these real underlying reasons of fiscal mismanagement. The defence budget is a much larger debate, which must be settled after due discussion, but it should not be used to hide public sector inefficiencies, slow revenue growth or poor service delivery outcomes.