Configuring baseline projections

Table of contents

• Configuring baseline projections
  • Specifying baseline projection adjustments
  • Standard productivity, population, and autonomous energy efficiency adjustments to baseline projections
    • The labor_augmenting_technical_change.csv file
    • The baseline_exogenous_projections.csv file
  • Potential output projections in the baseline

Users have very precise control over baseline projections in G-Cubed models. This control is provided through various files in the data folder.

Specifying baseline projection adjustments

The baseline projections are modified using adjustments to exogenous variables in the model. The adjustments that are applied when generating the baseline projections must be specified in a design file. This design file has the same syntax and interpretation as the design files used in simulation experiments.

The model configuration file specifies the name of the design file that is used to generate the baseline projections as the value of the BaselineDesignFile setting. If that setting is included in the model configuration then the named design file must be in the data folder of the model. The standard name used in G-Cubed models is baseline_design.csv but other names can be used.

If the BaselineDesignFile setting is missing from the model configuration file then there will be no adjustments to exogenous variables when generating the baseline projections. In other words, productivity and population growth rates will be zero in the baseline projections.

The baseline design file determines how the baseline projections are tailored by exogenous variable adjustments by indicating the ‘simulation layer adjustment files to be applied when generating the baseline projections. In contrast to the simulation layers used in experiments all of these baseline simulations layers must have the same event year and that event year must be the first projection year.

Typically the baseline design file references just one simulation layer adjustment file, the product.csv file described below. However, it is possible to reference multiple simulation layer adjustment files in the baseline design file. In this case, the simulation layer adjustment files are applied in the order they are listed in the baseline design file. Working with multiple simulation layer adjustment files can be useful as a way of dividing up types of adjustments being made to the baseline projections.

Standard productivity, population, and autonomous energy efficiency adjustments to baseline projections

Exogenous adjustments to labor-agumenting productivity, population, and autonomous energy efficiency improvements are made to baseline projections in the G-Cubed model. These adjustments are based on the information provided in various files in the model’s data folder. Details of the baseline adjustments for the catchup growth model are available. Likewise, details of the baseline adjustments for the autonomous energy efficiency improvements are also available.

Those exogenous adjustments are compiled into a set of shocks to exogenous variables when the baseline projections are created. That compilation process generates two additional files in the data folder, labor_augmenting_technical_change.csv and baseline_exogenous_projections.csv.

In some situations, it can be useful to directly override those adjustments directly rather than by making changes to the source data files. For this reason, there is some complexity in the way that the model handles these adjustments. This complexity is addressed in the following descriptions of these files.

The labor_augmenting_technical_change.csv file

In model builds before 179, this was typically named the modprod.csv file.

The labor_augmenting_technical_change.csv file contains projections of labor-augmenting technical change adjustments for all sectors in all regions. The projections are from the first projection year to the last projection year. these adjustments are calculated from the information in the prodmat.csv file. When G-Cubed is run, this file is only regenerated if it cannot be found in the data folder..

The baseline_exogenous_projections.csv file

In model builds before 179, this was typically named the product.csv file.

The baseline_exogenous_projections.csv file contains all of the standard adjustments to exogenous variables that are typically applied when creating baseline projections for a G-Cubed model. It includes the effective labor productivity adjustments based on labor-productivity growth rate projections for all sectors in all regions (sourced from the labor_augmenting_technical_change.csv file) and population growth rates (sourced from the labor_force_growth_rates.csv file). It includes the the autonomous energy efficiency improvements from the autonomous_energy_efficiency_improvements.csv file. It includes the implied potential output growth rates for all sectors in all regions. When G-Cubed is run, This file is only regenerated if it cannot be found in the data folder..

The name of this file can be altered from its default by changing the following configuration settings in the model configuration file: ExogenousGrowthAndEfficiencyAdjustmentsFile.

If the labor_augmenting_technical_change.csv file was deleted and regenerated but the baseline_exogenous_projections.csv file was not, the baseline_exogenous_projections.csv file will be out of date.

In summary:

To fully regenerate the baseline_exogenous_projections.csv file, you must delete both the labor_augmenting_technical_change.csv and baseline_exogenous_projections.csv files and then recreate the baseline projections for the model.

When the model is run, the population growth rates and the labor-augmenting productivity growth rates are combined to form exogenous effective labor productivity growth rate projections. These are also stored in the model’s data folder in two additional files:

• labor_augmenting_technical_change.csv - labor-augmenting technical change for all sectors in all regions.
• baseline_exogenous_projections.csv - effective labor productivity growth rates for all sectors in all regions, Autonomous energy efficiency improvements, and potential output growth rates.

If you make changes to the other data files, these two generated data files will need to be deleted before the model is run again. They will be regenerated with the new data.

You can make direct alterations to the information in these two generated data files. Those direct alterations will not be overwritten until you delete the files and run the model again.

Potential output projections in the baseline

Particular care needs to be taken when generating potential output growth rate projections. These are exogenous projections and they need to be consistent with the other growth rates (productivity and population growth rates for example) in the model. If that is not the case, then monetary policy will tend to distort experiment results because the monetary policy rules typically include a responese to deviations from potential output.