Is Sri Lanka addicted to missing revenue targets?

Track how many of the recommendations in these diagnostics are put into action, and you will have the answer

It is all too familiar that Sri Lanka has a problem in meeting its budgeted revenue targets.
Recently, a government member of parliament (MP) who voted in favour of the budget last year, said that the tax increases that were required to achieve the revenue targeted for 2023 were never practical. This was to explain why government revenue was hugely off target this year. The multiple flaws in his understanding and argument were explained in a fact check published in the Daily Mirror of Thursday, October 26.

The Factcheck shows why the MP’s understanding of what is practically feasible is not correct. But the fact that Sri Lanka would miss the revenue targets by a mile was easy to see for those who scrutinised the budget projections.
In December last year (2022), Verité Research hosted its bi-annual Public Finance Lounge, to which high-level stakeholders are invited. One of the presentations was an analytical demonstration of why revenue targets set out in the budget were likely to fall short by about 11%. 

A recent press release by the International Monetary Fund (IMF) says that revenue is now expected to fall short by about 15%. The point is that this shortfall in revenue in 2023 is not mysterious. Much of it was foreseeable even as the budget was being presented in November 2022.
But delving into this issue raises not just the question of poor understanding among members of parliament on why targets are not achieved, but another deeper and darker question: Is Sri Lanka structurally addicted to missing the revenue targets in its budgets?

Signs of a structural addiction to missing revenue targets?

A look at decades of budget performance suggests that Sri Lanka might indeed have a compulsive proclivity, to its serious detriment (aka addition), to missing revenue targets that are approved in its budgets.

This addiction might also be structurally entrenched. Since as far back as the year 2000, no government has achieved a revenue target that was put down in its budget (we don’t have access to budget speeches prior). The addiction seems so severe, that even when put inside the straight-jacket of an IMF programme, Sri Lanka mostly fails to achieve even the minimum revenue targets in the programme.

Since the turn of the century, Sri Lanka has had three IMF programmes before the current one. The revenue targets agreed on the 14th IMF programme with a duration of three years, starting in 2003, were never achieved, not even in the first year. On the 15th programme which ran for 20 months starting in 2009, it was only achieved in the first year. On the 16th programme with an initial duration of 3 years, starting in 2016, the revenue target was achieved only in the first year, though the second year was a near miss, instead of being wildly off the mark. In this 17th programme, the revenue target, by all indications, is going to be missed by a mile in the first year itself.

This begs the question; why does Parliament keep approving budgets with revenue targets that the government compulsively fails to achieve? 
There are three structural factors that could be enabling this addiction-like behaviour: (1) Sri Lanka’s bureaucracy does not produce proper analysis to estimate revenue (2) Decision-makers disregard analysis and put forward impossibly rosy revenue expectation (3) There is systemic bad faith in implementing revenue related policies in the budget.

The 2023 Budget saw all three enablers of this structural addiction at work

Verité Research has been supporting Parliament Committees (when engaged) and the top decision-makers of the country (through its public finance lounge) with budget analysis since 2016. Every year, the organisation provides an alternative assessment of the practicality of the revenue targets.
There are past reports of the Committee on Public Finance (COPF) in parliament, which show how the revenue outcome projections provided to the committee by Verité Research during budget debates were much closer to the actual outcomes seen a year later than what was presented in the budget.

In December 2022, Verité Research showed that revenue targets for the 2023 budget will fail by at least 11%. The basis for that analysis and ensuing events suggest that all three enablers of the addiction, noted above, are operative in Sri Lanka. These are just three examples to show how each of these enablers has contributed to the revenue shortfall in 2023.

Addiction Enabler 1: Improper analysis. The new social security contribution levy (SSCL) is similar to the previous Nation Building Tax (NBT) that existed in 2019. In estimating its collection, there are three considerations: (a) the increase in collection due to nominal GDP growth since 2019 (b) the SSCL rate increasing to 2.5% from the 2% NBT; and (c) the threshold of the SSCL increasing 10-fold to LKR 120 million, whereas the NBT threshold was LKR 12 million per annum. The projections for SSCL in the budget adjusted the 2019 NBT revenue for changes “a” and “b”, but failed to adjust for the change “c”. This analytical mistake could result in the actual SSCL collection falling short by over 20%

Addiction Enabler 2: Disregarding analysis. Using available data, and stated policy directions, Verité Research was able to project that taxes on international trade would fall short by around 25% of what was in the budget for 2023. But trade taxes and quantities are estimated every year, without such a large deviation. This could indicate a higher decision-making nexus that is disregarding internal analysis and inflating the revenue expectations in the budget. 

Addiction Enabler 3: Bad Faith Implementation. There was a Casino Entrance Levy of USD 100 per person which had never been collected, since being introduced in 2015, until 2023. There were multiple promises even in mid-2022 that it would finally be collected. But implementation went awry. Collection was started only around 20th February 2023, and then legislation was brought in to — retrospectively from April 1st — remove this tax entirely for all foreign clients, and reduce it by 50% for local clients. These were revenue-reducing policies that were not part of the expectations of the IMF programme commitments, nor the approved budget.

Two governance diagnostics contain the solutions

The structural addiction to missing revenue targets falls within a larger malaise in Sri Lanka. Its budgets have long lacked both credibility and accountability.
Solving the credibility problem can begin with this: the government publishes, along with the budget, a supporting analysis of every revenue projection and revenue-impacting budget proposal. This will impede the second addiction enabler of disregarding proper analysis, and also impede the first addiction enabler, by encouraging proper analysis, knowing it will be subject to public scrutiny.

Solving the accountability problem can begin with this: The government publishes along with the budget all variances in revenue from what was projected in the previous budget, along with explanations and actions being taken to rectify the significant variances. This will impede the third addiction enabler by discouraging bad faith implementation since it will now be made visible to the public.

The above solutions and much more are provided in two convergent and complementary governance diagnostics for Sri Lanka that were published in September. First is the Civil Society Governance Diagnostic. Second is the IMF Governance Diagnostic, which is the only technical assistance analysis by the IMF that has been published under the current programme. The recommendations in these diagnostics might be pivotal in Sri Lanka breaking its addiction to failing on economic targets.
Is Sri Lanka serious about getting on a path of sustained economic recovery? Track how many of the recommendations in these diagnostics are put into action, and you will have the answer.
Dr. Nishan de Mel is an economist and the Executive Director of Verité Research.