Predicting revenue: Getting it right2017-02-06 16:54:49
Forecasting revenue is very difficult – analysts are asked to predict economic shifts, taxpayer behaviour, and the outcomes of complex tax laws, often with limited data. But striving for the best forecasts is vital as the foundation of the government budget and delivering good public policy.
In Indonesia, revenue at the national government level is sourced mainly from income taxes, both company and individual, along with the Value Added Tax (VAT), excises (on goods such as cigarettes and alcohol), royalties from resources, and dividends from State Owned Enterprises.
Sound revenue forecasts underpin good budget planning and lead to efficient spending outcomes. By contrast, overestimating revenue can lead to difficult and disruptive spending cuts, often in high priority but easy to cut areas like infrastructure. Underestimates can also be problematic, leading to short-term spending to ‘use up’ additional revenue.
At a three-day workshop held over 18-20 January 2017, around 30 government officials from the Ministry of Finance developed skills in better revenue forecasting. Interactive exercises using the examples of income tax and VAT collections helped participants hone their skills, with some working well into the night!
Participants also discussed the tax capacity of Indonesia. A higher level of development goes together with a higher capacity to pay and collect taxes. Indonesia’s tax to GDP ratio is low compared to its peers, at around 11 percent of GDP. Supporting Indonesia’s efforts to better estimate and understand tax capacity is key to improving revenue performance.
“Countries such as Australia also find forecasting revenue difficult. But a good process and investing the time to develop the right tools and data can help” said Matthew Quillinan, AIPEG Budget Policy Adviser. AIPEG will conduct a follow-up workshop on evaluating the revenue impact of tax policy proposals and continue to mentor government officials with on-the-job support for better revenue forecasting.