A good economic forecast is one of the most important inputs for policy makers. It is an art and science that is constantly evolving. At the core are the basic time series methods. Those range from judgmental methods that use expert judgement to adjust forecasts produced by a suite of models to dynamic stochastic general equilibrium (DSGE) models that use modern economic theory.
At the most basic level, these methodologies predict the behavior of a set of economic variables—for example GDP, consumption, investment, interest rates, and industry employment—viewed as statistical time series. This approach requires a large amount of knowledge of the behavioral patterns of these variables, and not a little faith that the estimated behavioral patterns will persist.
As a practical matter, economic forecasts are typically reported on a quarterly basis. However, for a number of reasons it is often more useful to have information on an annual basis—particularly if the information is being used to inform policies that will impact the economy over several years.
Real GDP growth is expected to slow this year and next in response to a rise in trade barriers and heightened policy uncertainty. Global economic activity is projected to pick up over the longer term, reflecting an expected recovery in oil-producing countries and a stabilization of armed conflicts. However, downside risks include the possibility that global trade tensions reescalate or that inflationary pressures resurge. In addition, higher delinquency rates on credit cards and auto loans could restrain consumer spending.