Situation & Outlook Macroeconomic Modeling

Situation & Outlook Macroeconomic Modeling

Decision-Making Information

ProximityOne
information resources & solutions
(888) DMI-SOLN
(888) 364-7656

The Situation & Outlook model is a mathematical simplification of a national and regional economy. It consists of a system of equations designed to capture the structural relationships between broad economic aggregates - such as gross domestic product (GDP), personal income, employment, and inflation.

The Situation & Outlook model operates much like a large-scale system of simultaneous equations where historical data is used to estimate parameters, allowing the system to project future trends or simulate the impact of specific shocks.

The Situtation & Outlook model relies heavily on historical time-series data and estimated behavioral equations. These are highly practical for generating highly granular, real-world forecasts (such as county-level personal income or output projections).

Here is a breakdown of how the Situation & Outlook model is structured and operates:

1. The Building Blocks: Variables

Macroeconomic models classify economic indicators into two primary categories:

Endogenous Variables: These are the variables the model is designed to explain and forecast. Their values are determined within the system. Common examples include total consumption, regional GDP, or unemployment rates.
Exogenous Variables: These are inputs determined outside the model. The model takes them as given. They often include policy instruments (like the federal funds rate or tax rates), demographic shifts, or external shocks (like a sudden change in global oil prices).

2. The Structural Framework: Equations

The mechanics of the model rely on a set of equations that link these variables together. These generally fall into three types:

Accounting Identities: These are definitional truths that must always hold and require no statistical estimation. The most famous is the national income identity:
Y = C + I + G + NX
Where GDP (Y) equals Consumption (C) + Investment (I) + Government Spending (G) + Net Exports (NX).
Behavioral Equations: These describe how different economic agents (households, firms) behave based on historical data. For example, a consumption function might link current consumer spending to personal income and interest rates. The parameters (coefficients) of these equations are typically estimated using econometric techniques on historical time-series data from agencies like the Bureau of Economic Analysis (BEA), the Census Bureau, or the Bureau of Labor Statistics (BLS).
Equilibrium Conditions: These equations ensure that markets clear within the model, such as setting aggregate labor supply equal to aggregate labor demand.

3. Estimation and Calibration

Before the model can generate forecasts, it must be estimated.

Estimation: Historical datasets are ingested into statistical software to estimate the coefficients of the behavioral equations using methods like Ordinary Least Squares (OLS) or Maximum Likelihood.
Calibration: In highly theoretical models, rather than estimating every parameter, some parameters (like the discount factor or capital share of output) are "calibrated" using established microeconomic evidence or long-term historical averages to ensure the model behaves realistically in a steady state.

4. Simulation and Forecasting

Once the system of equations is specified and the parameters are set, the model is put to work:

Baseline Forecasting: By plugging in projected paths for the exogenous variables (e.g., expected population growth or planned government spending), the model solves the system of equations across future time periods (e.g., quarterly through 2030) to generate a baseline forecast of GDP, income, and other endogenous variables.
Scenario Analysis (Shocking the Model): The true power of the model lies in counterfactuals. An analyst can introduce a "shock" to the system - such as a 2% increase in interest rates or a sudden drop in regional employment - and observe the dynamic path the endogenous variables take as the system adjusts back to equilibrium over time.

Example of a behanioval equation

A classic example of a behavioral equation in macroeconomics is the Aggregate Consumption Function. Unlike an accounting identity that is true by definition, this equation attempts to mathematically describe how households behave when deciding how much to spend versus how much to save.

Here is how a standard, dynamically specified consumption function might look in an econometric model:

C_t = β₀ + β₁Y_t + β₂r_t + β₃C_t-1 + ε_t

Here is the breakdown of the components:

C_t (Endogenous Variable): Total real consumer spending in the current quarter (t).
.. This is the behavior the equation is trying to predict.

Y_t (Explanatory Variable): Real disposable personal income.

r_t (Explanatory Variable): The real interest rate, capturing the cost of borrowing or the reward for saving.

C_t-1(Lagged Variable): Consumer spending in the previous quarter, included to capture "habit persistence" (the idea that households adjust their consumption habits slowly).

β₀,β₁,β₂ ,β₃ (Parameters/Coefficients): These are the structural values estimated using historical statistical data. For example, β₁ represents the marginal propensity to consume-how much out of every additional dollar of income a household will spend.

ε_t (Stochastic Error Term): This captures the random, unobservable shocks to consumption that the model cannot explain (e.g., a sudden change in consumer confidence).

A Regional Level Example

When building forecasting models to project quarterly economic trends at the sub-national level—such as a county-level personal income or gross domestic product model-behavioral equations are adapted to capture regional dynamics.

A behavioral equation for County-Level Personal Income (PI) might look like this:

PI_i,t = α₀ + α₁EMP_i,t + α₂W_i,t + α₃PI_i,t-1 + ε_i,t

In this regional model:

PI_i,t is the personal income for county i in quarter t.

EMP_i,t is the local employment level (derived from Bureau of Labor Statistics Quarterly Census of Employment and Wages data).

W_i,t is the national average wage rate (an exogenous macro-level driver).

ε_i,t is the county-specific error term.

To make this operational, historical time-series data are added for each county, runs regressions to estimate the α coefficients, and then programs the resulting equations into the software to calculate the projected values through future quarters.

ProximityOne User Group .. goto top
Join the ProximityOne User Group to keep up-to-date with new developments relating to metros and component geography decision-making information resources. Receive updates and access to tools and resources available only to members. Use this form to join the User Group.

Support Using these Resources
Learn more about accessing and using demographic-economic data and related analytical tools. Join us in a Data Analytics Lab session. There is no fee for these Web sessions. Each informal session is focused on a specific topic. The open structure also provides for Q&A and discussion of application issues of interest to participants.

Additional Information
ProximityOne develops geodemographic-economic data and analytical tools and helps organizations knit together and use diverse data in a decision-making and analytical framework. We develop custom demographic/economic estimates and projections, develop geographic and geocoded address files, and assist with impact and geospatial analyses. Wide-ranging organizations use our tools (software, data, methodologies) to analyze their own data integrated with other data. Contact us (888-364-7656) with questions about data covered in this section or to discuss custom estimates, projections or analyses for your areas of interest.