Economic Forecasting Approach

Moody's Analytics methodological approach to forecasting.

With extensive experience in national and subnational data management and forecasting, Moody's Analytics is a trusted, leading provider of economic and financial forecasts for major countries/jurisdictions; global metropolitan areas; the U.S.; and all its states, metropolitan areas and counties.

Global Economies

The Moody's Analytics global model is a system of simultaneous equations that produce between 50 and 300 forecast variables for 100+ countries/jurisdictions. The equations are estimated using a standard template of simultaneous equation model, encompassing the entire macro economy of a nation. The standardization makes comparisons across geographies possible and allows us to bring in nuances relevant to each country/jurisdiction.

The model's core equations are a small set of series that drive the rest of the model. The core includes final demand, monetary and fiscal policy, trade, and labor market conditions. Auxiliary variables are modeled as a function of the core set of equations. This approach makes it easy to re-estimate and control the forecast system.

The forecasts are then put through numerous accuracy checks and the models are shocked to simulate the effect of macroeconomic events such as currency fluctuations, government stimulus, and global recessions. The analysts for those countries/jurisdictions review each variable to ensure that the forecast is internally consistent and in line with expectations for that nation or region.

Global Subnational Areas

As the importance of understanding metropolitan areas as centers of economic activity has grown over the past two decades, Moody's Analytics has developed subnational analysis and forecasts for more than 500 metro areas worldwide, and is committed to further expanding geographical coverage.

Each month Moody's Analytics updates the national forecasts and the metro area historical data are extended with any newly available observations. The updated national forecasts and metro area history are then input into the metro area models, and the forecasts are updated.

In preparing the monthly update of the metro area forecasts, weight is given to the initial model-based forecasts and to the previous month's forecast using a lineup procedure. This process eliminates erratic changes to the forecast that do not have a firm economic basis.

The new metro area forecast is then reviewed by the analyst who is the expert for the area. The metro analyst makes adjustments based on his/her knowledge of factors of current conditions or pending shifts in industry conditions or government policy that the model will not be able to pick up.

Extensive diagnostics are used to ensure that the forecasts are stable from month to month and consistent with the current business cycle outlook of each nation and the overall health of the global economy.

U.S. National and Regional Areas

Our approach to forecasting the U.S. and its regional economies is informed by both national and regional analysis. The process starts each month with the U.S. forecast. The Moody's Analytics U.S. macro model is a large, 1,600-equation simultaneous-equilibrium model that allows for interrelationships among all of the sectors of the U.S. economy, including production, income, financial markets, consumer spending, and labor markets. Our chief economist, Mark Zandi, and his staff discuss the economic outlook and refine the forecast so that it accurately reflects our view of where the U.S. economy is headed, in both the near and the long term.

After the U.S. forecast is finalized, those results are used in the regional forecasts. The national drivers are run through models for each state and metro area. In addition to the national forecasts, regional variables such as costs of doing business are included in the regression equations whenever possible. Demographics and industrial structure also play key roles in defining the regional outlooks within the context of the U.S. macro forecast. Once the model-based forecast is derived, analysts review each state and metro area forecast to make sure the results are consistent with both the national forecast and regional-specific conditions.

At the end of the state and metro forecasting process, the sum of the states and metro area forecasts are checked against the U.S. forecast for variables such as employment and housing permits. We do not require the sums of the regional forecasts to add up to the national total—the sums of states for most concepts do not equal the U.S. in history—but we do check them against one another to ensure that the shape of the national forecast is consistent with the regional forecasts.