This study investigates the impact of inaccurate assumptions on economic forecast precision. We construct a new dataset comprising an unbalanced panel of annual German GDP forecasts from various institutions, taking into account their underlying assumptions. We explicitly control for different forecast horizons to reflect the information available at the time of release. Our analysis reveals that approximately 75% of the variation in squared forecast errors can be attributed to the variation in squared errors of the initial assumptions. This finding emphasizes the importance of accurate assumptions in economic forecasting and suggests that forecasters should transparently disclose their assumptions to enhance the usefulness of their forecasts in shaping effective policy recommendations.

Die Projektgruppe Gemeinschaftsdiagnose prognostiziert für das Jahr 2025 einen leichten Zuwachs des Bruttoinlandsprodukts in Deutschland um 0,2 %. Im weiteren Prognosezeitraum stimuliert die expansive Finanzpolitik die Konjunktur. Die Institute prognostizieren für die kommenden beiden Jahre Expansionsraten von 1,3 % und 1,4 %. Strukturelle Probleme wie abnehmende Wettbewerbsfähigkeit und der demografische Wandel bleiben bestehen. Um Wachstumsperspektiven für die deutsche Wirtschaft zu schaffen, bedarf es einer umfangreichen Reformpolitik. Zur Orientierung präsentieren die Institute einen Zwölf-Punkte-Kompass für den Herbst der Reformen.

Accurate macroeconomic forecasts are essential for effective policy decisions, yet their precision depends on the accuracy of the underlying assumptions. This paper examines the extent to which assumption errors affect forecast accuracy, introducing the average squared assumption error (ASAE) as a valid instrument to address endogeneity. Using double/debiased machine learning (DML) techniques and partial linear instrumental variable (PLIV) models, we analyze GDP growth forecasts for Germany, conditioning on key exogenous variables such as oil price, exchange rate, and world trade. We find that traditional ordinary least squares (OLS) techniques systematically underestimate the influence of assumption errors, particularly with respect to world trade, while DML effectively mitigates endogeneity, reduces multicollinearity, and captures nonlinearities in the data. However, the effect of oil price assumption errors on GDP forecast errors remains ambiguous. These results underscore the importance of advanced econometric tools to improve the evaluation of macroeconomic forecasts.