Erroneous macroeconomic forecasting and how to correct it

19
Gede Sandra People Sovereignty Movement (Pergerakan Kedaulatan Rakyat) Analyst

IO – Macroeconomic forecasting is a core function of the Ministry of Finance (the State Treasury). It deals with any instructions from the Ministry of Finance in various policy spectrums, as a complementary activity adetermination of costs (The Australian Government’s Review of Treasury Macroeconomic and Revenue Forecasting, 2013).

Reliable macroeconomic forecasting can inform us whether our economy is nearing the ending of an expansive phase, or at the start of a recession. Its purpose is to ensure that economic actors can anticipate these latest developments and avoid accompanying reactions (Bernstein and Silbert, 1984).

Amid the current economic recession, blamed on the COVID-19 pandemic, all economic actors strongly need reliable macroeconomic forecasting from their country’s Ministry of Finance. Unfortunately, the Indonesian Ministry of Finance is incapable of forecasting our macroeconomic condition, thus hindering our business actors from anticipating the latest developments and making it harder for our economy to recover from a persistent recession. With such huge damage, I find it quite odd that our Minister of Finance is not part of the Cabinet reshuffle.

Three Incorrect Macroeconomic Predictions

Throughout 2020, our Ministry of Finance has made three erroneous macroeconomic forecasts, each involving economic growth, the GDP, and inflation rates. First error: In April 2020, Minister of Finance Sri Mulyani Indrawati forecasted that Indonesia’s economy will grow at a 4.5-4.7% level in Quarter I of 2020. This forecast sorely missed the mark: Statistics Indonesia (Badan Pusat Statistik – “BPS”) recorded our economic growth for the period at a mere 2.97%.

Second error: In July 2020, the Minister of Finance forecasted our economic growth in Semester I of 2020 to be in a -1.1% to -0.4% range. On the contrary, BPS recorded a growth of -5.32% for the period. Another major miscalculation. 

Third error: In October 2020,  Sri Mulyani predicted that our economic growth in Quarter III of 2020 would be in a -2.9% to -1.1% range. BPS recorded a growth rate of -3.49%.

The error percentages of the Minister’s economic growth forecast for Quarter I, II, and III of 2020 (calculated from the median point) were 1.63%, 4.57%, and 1.49% respectively. Because the forecasting error is always higher than the reality achieved, it is a positive bias. If the error is lower than the reality, it’s a negative bias. A positive bias shows that Government forecasts tend to be over-optimistic.

Other than when predicting economic growth, the Ministry of Finance has also forecasted inflation erroneously. At the start of the year, the Ministry set the 2020 inflation target at 3.1%, as expressed in the 2020 State Budget. In August 2020, the Minister revised the inflation target for 2020 to 2.5%. Finally, on 1 December 2020, Sri Mulyani revised again her forecast of 2020 inflation to a mere 1.5%. At least her error is not that far off the mark now, with current reports saying that our inflation rate in 2020 was 1.68%.

What to do now?

In order to reduce errors in upcoming macroeconomic forecasts, the Ministry of Finance should review past forecasts and errors and officially publish this document as a means of public accountability. This is the accountability practice common among the Ministries of Finance (Treasuries) of other countries, most specifically among the Commonwealth countries of Australia, New Zealand, and Canada. 

The Review of Treasury Macroeconomic  and Revenue Forecasting (2013) issued by the Australian Ministry of Finance suggests that reliable forecasting includes not only in-depth analyses of major economic sectors, but it also utilizes other information that fine-tunes the econometric models being used for the forecasting.

Such additional information may include: 1) Consumer and business surveys to identify possible directions for household expenses, prieasevate investments, and employment recruitment; 2) information gathered from regular dialogues with the business community about new trends identified; 3) data on housing construction or other types of construction that describes shortterm construction; 4) the opinion of experts in other Government agencies, as well as data on the growth of productive age population and non-rural commodity production estimates; and 5) assessments made based on corporate experience and corporate track record.

And finally, the macroeconomicforecast must undergo internalreview (senior Government economists)and external peer review (experteconomists from outside of the Government).