External parameter files

External parameters that are specified for all products, all sectors, or over time, are supplied in three comma-separated variable (CSV) files in the inputs folder. The files provided with the Freedonia sample model have the names product_parameters.csv, sector_parameters.csv, and time_series.csv. However, the names are set in the configuration file's general settings. That means that different external parameter files can be used for different scenarios.

The other files in the Freedonia sample model input folder include the supply-use table, which is described on a separate page, and some optional input files, which are described below.

Product parameters

The product parameters file has the structure shown below. There are as many rows in the table as there are products.

The first two columns are the product codes and product names. These should be in the same order as the products in the supply-use tables.

The next two columns provide the initial income elasticities of demand^[1] for exports (with respect to global GDP) and domestic final consumption (with respect to total wages). These elasticities change over time, as determined by additional parameters in the configuration file.

The final two columns are bracketed because they are optional. They specify how imports and exports respond to relative changes between domestic and world prices. If the columns are not present, these elasticities are set to zero by default, meaning no price sensitivity.

For more information on how AMES uses the elasticities, see the detailed explanation of export demand, final domestic demand and imports.

Elasticities can be estimated from historical data, although that requires knowledge of statistical methods. Otherwise, they can be drawn from published studies or from already existing economic models.

The corresponding variables are:

• export_elast_demand0 : the initial value for $\underline{\eta}^\text{exp}_k$ for product $k$
• wage_elast_demand0 : the initial value for $\underline{\eta}^\text{wage}_k$ for product $k$
• import_price_elast : $\underline{\phi}^\text{imp}_k$ for product $k$
• export_price_elast : $\underline{\phi}^\text{export}_k$ for product $k$

Here is the example from the Freedonia sample model:

Sector parameters

The structure of the sector parameters file is shown below. There are as many rows in the table as there are sectors.

The first two columns are the sector codes and sector names. These should be in the same order as the sectors in the supply-use tables.

In addition to the sector codes and names, this file contains a single required parameter for each sector, the depreciation rate. Where available, this can be calculated from national statistics. Otherwise, the delta parameter reported for the whole economy in the Penn World Table can be applied to each sector.

The file also includes several optional parameters, shown in brackets, which determine labor productivity growth and changes in intermediate demand coefficients at sector level. The productivity parameters might be assumed to be the same across sectors, estimated from historical data, or drawn from published studies. If values are the same across sectors, then they can be specified as default values in the configuration file and the corresponding columns can be omitted from this file.

The corresponding variables for sector $i$ are:

• depr_rate : $\underline{\delta}_i$
• KV_coeff : $\underline{\alpha}^\text{KV}_i$
• KV_intercept : $\underline{\beta}^\text{KV}_i$
• empl0 : $\underline{L}_{i0}$
• labor_prod_gr : $\underline{\beta}^\text{KV}_i$ (with $\underline{\alpha}^\text{KV}_i = 0$)
• rate_const : $\underline{a}_i$ (an optional parameter)
• exponent : $\underline{\vartheta}_i$ (an optional parameter)

Here is the example from the Freedonia sample model:

Time series

The structure of the time series file is shown below. It contains several parameters: the world growth rate, the world inflation rate, the growth rate of the working-age population, the exchange rate, and parameters for the labor productivity calculation.

Only the working_age_gr column is required – each of the others is optional, so they are bracketed.

• If either the world_gr or world_infl_rate columns are missing, then the corresponding default value from the configuration file is applied for all years;
• If the exchange_rate column is missing, then it defaults to a constant exchange rate;
• If any of the KV_coeff, KV_intercept, or labor_prod_gr columns is missing, then the corresponding default value from the configuration file is applied for all years.

Scenarios for the world economic growth rate can be drawn from other studies, such as the Shared Socioeconomic Pathways (SSP) database. The working age growth rate can be calculated from national projections or the UN Population Prospects database.

Other parameters have less well-established sources of estimates. The labor productivity parameters (the Kaldor-Verdoorn parameters KV_coeff and KV_intercept or the labor productivity growth rate labor_prod_gr) might be assumed constant, estimated from historical data, or drawn from published studies. If they are constant, then they can be omitted from this table and specified as default values in the configuration file. Assumptions regarding the world inflation rate and the exchange rate can be based on historical patterns, other modeling studies, or consultation with experts.

The corresponding variables are:

• world_gr : $\underline{\gamma}^\text{world}$
• world_infl_rate : $\underline{\pi}_{w,k}$ (applied uniformly to all products $k$, but modified by an optional file specifying world real price trends for selected tradeables)
• working_age_gr : $\underline{\hat{N}}$
• exchange_rate : $\underline{e}$
• KV_coeff : $\underline{\alpha}^\text{KV}$
• KV_intercept : $\underline{\beta}^\text{KV}$
• labor_prod_gr : $\underline{\beta}^\text{KV}$ (with $\underline{\alpha}^\text{KV} = 0$)

Here is the example from the Freedonia sample model:

Optional input files

External inputs for investment, potential output, and prices can be supplied from LEAP. Investment expenditure is collected automatically, while potential output and prices can be specified in the configuration file. Additionally, for non-energy sectors, it is possible to specify any or all of four optional input files for: investment demand; potential output; maximum capacity utilization; and real prices for tradeables.

Investment demand

The AMES model calculates investment for non-energy sectors based on expected demand and profitability: see the explanation of potential output in the Technical Details. However, for public infrastructure investment – which is driven by policy goals, rather than private profitability, and where the capital stock is not associated with a particular sector – investment must be specified exogenously. (When externally specified investment is associated with a particular sector, it is better to specify potential output: see below.)

The investment demand parameter file has the following structure:

Note that not all years need to be included. For example, if there is investment expenditure in 2025, but not in 2026, then the year 2026 does not have to be included in the file. The values are in "real" monetary terms, and are used directly by AMES, so the units should be the same as those for the supply and use tables.

Potential output

In the AMES model, potential output is determined by investment: see potential output in the Technical Details. In some cases, it is best to override this behavior. For example, output from agriculture might be determined by an external crop model, or the output from the mining sector might be constrained by the availability of the ore. In other cases, the production level might be set as a policy target or through a sector-specific planning document. In these cases, potential output can be specified for specific sectors, and AMES will calculate investment.

The potential output parameter file has the following structure:

The sequence of values is converted internally into an index. For this reason, values must be specified for all years. However, values should be specified only for sectors with exogenous potential output. Other sectors, where AMES simulates the change in potential output, should not appear in this file.

Maximum capacity utilization

Capacity utilization in the AMES model is determined in each time step by solving a linear goal program. By default, maximum capacity utilization is equal to 1.0. However, in some cases, capacity utilization might be constrained. For example, during a disease outbreak, some service sector activities may be limited, and during a drought, manufacturing plants that rely on cooling or process water might have to curtail production. In these cases, a maximum level of capacity utilization less than one can be specified exogenously.

The maximum capacity utilization file has the following structure:

In the file, all years must be listed. However, values do not have to specified for all years. If capacity utilization is unconstrained in some year, the maximum level can be omitted, and AMES will set it equal to 1.0. An example of this is shown in cell (y_2,sec_2) in the table above.

World real price trends for selected tradeables

In the AMES model, "world" prices for goods and services are specified exogenously, while domestic prices are calculated as a markup on costs. By default, world prices for all tradeables grow at a uniform, user-specified world inflation rate. However, optionally, real world price indices for all or some tradeables can be specified. The real price trend is then adjusted for inflation at the world inflation rate.

The real price file has the following structure:

The sequence of values is converted internally into an index. For this reason, values must be specified for all years. However, values should be specified only for products for which a real price index is specified. For other products, the real world price is assumed to be constant, so the nominal price rises at the world inflation rate. Price indices for non-tradeables are ignored, because the world price is irrelevant; together with other domestic prices, AMES calculates prices of non-tradeables endogenously based on a markup.