State Space Models

All state space models are written and estimated in the R programming language. The models are available here with instructions and R procedures for manipulating the models here here.

Friday, June 6, 2025

World-System (1980-2080) Alternative Forecasts for US CG Debt


The US Congress is currently debating a Spending Bill being advanced by the Trump II Administration that will add $2.4 Trillion to the Deficit. In a prior post (here), my model for the US Economy forecasts that Central Government (CG) Debt will decline (the forecast was made after the Trump I Administration). I have also suggested that the outcomes of the Trump Administrations cannot be forecast because they are essentially a random walk (here).

In the graphic above I give four forecasts for US CG Debt (data taken from the World Development Indicators, see the Boiler Plate). The Random Walk (RW) line is meant to show that Debt Shocks (see the Trump I Administration) can happen at anytime in the future and are unpredictable. The prediction driven by the USL20 Model (US, dashed green line) shows the decreasing path of US CB Debt. The prediction driven by the WL20 Model (W, dashed blue line) shows debt peaking around 2050. The Business as Usual Line (BAU) shows debt growing linearly (it actually levels off in 2500).

One point to make is that all these models are stable (largest eigenvalue < 1.0)**. This means that none of my models predict that US CG DEBT is out of control or will grow forever (uncontrolled exponential growth). However, other Econometric models (for example, the Penn-Wharton Budget Model, PWBM) forecast that the US has about 20 years before Debt levels become unsustainable.

My best guess? Given that my forecasting models are stable and that the Trump Shocks will eventually be history, it seems reasonable to project that US CG Debt will keep pace with growth in the US Economy and will eventually stabilize when economic growth is over (see my forecasts for US  Economic growth here).

NOTES

** All my models are estimated from historical data and are not assumed, prior to estimation, as either stable or unstable (as are most Economic Models). For more information about the models I use and data sources, see the Boiler Plate.

I analyze the causes of the US CG Debt Bubble in another post.

No comments:

Post a Comment