DIW Study: Use Of AI Increases Accuracy Of ECB Policy Predictions By Up To 80%

Illustration of the German Institute for Economic Research (DIW Berlin) (photo: x @DIW_Berlin)

JAKARTA A recent study from the German Institute for Economic Research (DIW Berlin) revealed that the use of artificial intelligence (AI) significantly increases its accuracy in predicting the monetary policy of the European Central Bank (ECB).

The study analyzed ECB communications from January 2019 to March 2025, using a specially designed AI-based text analysis model to understand policy signals from any official ECB statement.

According to Kerstin Bernoth, DIW expert and study author, the AI used is able to examine every sentence in the ECB statement individually and categorize it as a signal of restrictive, expansive, or neutral policies.

"The central bank uses language as a monetary policy tool," said Bernoth. "The choice of words in speech, press releases, or interviews has never been done carelessly. All are carefully selected and can be a guide for future policies."

Prediction Accuracy Rises From 70% To 80%

When combined with other variables such as inflation rates, economic policy uncertainty, and previous interest rate trends, this AI-enhanced forecast model was able to increase the accuracy of the forecast for interest rate changes from about 70% to 80%.

This shows that a new approach to natural language analysis technology can be an important tool in responding to and anticipating the direction of central bank policy, which has a major impact on financial markets, exchange rates, and the investment climate.

Ahead of the ECB meeting on Thursday 17 April, the prediction model developed by DIW showed there was likely to be a cut in interest rates, although the communication tone of the ECB recently sounded more neutral.

Analysts surveyed by Reuters also expect the ECB benchmark interest rate to drop from 2.5% to 2.25%, as global trade weakens due to new rates and increased uncertainty that suppresses consumption and investment.

This finding further confirms that verbal communication is an important instrument in modern monetary policy. With AI, economists now have a sharper tool to interpret the language used by the central bank.

Meanwhile, the use of AI in economic policy also opens up new opportunities for more accurate modeling and prediction, which can ultimately help investors, policymakers, and market players make more precise and fast decisions.

With the increasing reliability of AI-based predictions, ECB measures and other central banks could be anticipated more quickly by the market, reduce uncertainty, and strengthen macroeconomic stability in the European region.