Bridging data gaps in India: Case of subsidy spending

Accurate measurement of subsidy spending by the central and state governments faces considerable challenges due to the complexity, diversity, and opacity of the subsidy system. The paper reviews the current state of subsidy accounting in India and the quality of data reported in budgets and finance accounts in this regard. It highlights the need for a clear definition and reporting by both the central and state governments, while attempting to calculate a more accurate figure for the extent of subsidies. The paper aims to make the case for transparent and disaggregated data on subsidies as being crucial for understanding the size, design, and impact of the State’s principal instrument to achieve equity and welfare.

Measurement of India’s expenditure on subsidies faces three key challenges:

· There is no single and clear definition of subsidy in government documents. For example, it is unclear whether cash or in-kind transfers, implicit subsidies, and tax rebates are to be classified as subsidy. Their accounting by the central and state governments, therefore, also differs.

· Governments often leverage other sources of financing towards subsidisation which may appear in statements other than those of explicit subsidies. For example, some subsidies are provided off-budget, through public sector enterprises or government-backed entities, rather than being directly recorded in the government’s budget.

· Budgets also often do not show the full fiscal impact of current subsidy programmes, as some of their expenses may have been deferred. For example, governments often run arrears or accumulate carry-forward liabilities to be paid in subsequent budget outlays, when actual subsidy payments are mis-accounted for during the budget-setting. While these may be reported in future budgets, it brings significant mis-reporting on an annual basis.

The reported subsidy expenditure of the Centre and states, therefore, does not fully reflect the actual or true spending on subsidies at any given point in time. Thus, this paper attempts to include (1) other related subsidy spending; and (2) other means of financing subsidies to understand the true fiscal burden of subsidisation. We perform this exercise to find improved actual estimates of the subsidy burden of the Centre, as well as its breakdown of food, fertiliser, and fuel subsidies. Our estimation reveals that the actual subsidy spending of the Centre in FY25 is at least 18% more than what is budgeted. The Centre has also significantly mis-reported its spending on fuel subsidy, as payments deferred in the past are being repaid presently. Similar issues exist in estimating states’ subsidy spending. This renders data on subsidy expenditure incomplete and inconsistent:

· Incompleteness: Many forms of subsidies are left unaccounted for, either because of the ambiguity in classifying them or because their financing is through means other than budgetary allocations. These are difficult to quantify and may not always be recorded as part of official subsidy figures, despite being significant in size.

· Inconsistency: What constitutes subsidy spending can change across the years within a particular department, ministry, or State, leading to inconsistencies in reporting. Without the change in the accounting practice clearly delineated, a cursory look at the numbers can paint an incorrect picture.

In the case of states’ subsidy spending, a bigger and more preliminary concern is the incomparability of their subsidy statements:

· Incomparability: As a result of the above two data gaps, central and state government subsidy statements are incomparable. There is no standardised framework that aggregates all types of subsidies across the economy, leading to underreporting or overreporting in some cases.

As a result, making cross-state comparisons over a period of time is prone to significant inaccuracies. The paper attempts to address the comparability of reported food, power, and fuel subsidy spending by states by estimating the true or actual spending on these items by each state. The paper finds that accounting for food subsidy, for instance, shows large variations in underreporting across state governments—actual food subsidy spending by states in FY23 was at least three times more than what was reported.

Finally, because of the large extent of data gaps, it is important to caution the readers that the subsidy spending in this paper is itself underestimated, as many other subsidy-related and off-budget financed subsidies have been left out. Because strict identification of each government spending item would be complicated, we have limited ourselves in this paper to the most widely discussed subsidies, subsidy-equivalent programmes, as well as financing means.

It is, therefore, imperative to understand the true extent of subsidy spending. This can significantly assist in eliminating inefficient subsidies and improve targeting of the existing ones. This requires:

· A comprehensive definition of subsidies could be statutorily established by way of an overarching PFM law that applies uniformly to the Centre as well as the states.

· Additionally, the scope of the statement of subsidies must be expanded to include other subsidy-related spending and other means of financing subsidies, for improved transparency.

· Most importantly, India’s shift to accrual-based accounting will allow subsidy estimates to reflect the total burden of subsidies in any given year, irrespective of whether cash payments to subsidy-disbursing ministries or organisations have been made or deferred to the future.

Such an expansive scope of the problem requires constructive and proactive overhaul of the entire PFM framework of India, by the governments, the CAG, and the Finance Commission.

This paper can be accessed here.

This paper is authored by Shruti Gupta, research associate, Radha Malani, research analyst and Anoop Singh, distinguished fellow, CSEP.