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.

Wednesday, October 15, 2025

World-System (1980-2100) Will Political Instability Hurt the French Economy?

 


Recently, the NY Times reported (here) that "...after three government upheavals in France since last summer, the damage to the French economy has already been done." In this post, I will take a look at the data. Recent data on French GDP comes from the World Bank and ends in 2024. The graphic above shows French Real GDP (GDP in constant 2015 US$) with a forecast and 98% confidence intervals out to 2100. 

French GDP is forecast to grow until about 2045 and then collapse after that. 

In terms of "damage to the French economy," the 2020 COVID-19 Pandemic was a major shock (outside the lower 98% confidence interval). In 2024, French GDP was a bit above the forecast attractor path (dashed red line). Political turmoil might reasonably be expected to bring the economy back to the attractor path but not create another COVID-style shock.

Possibly the most concerning aspect of the GDP forecast is the collapse starting around 2045. Why is the French Economy forecast to be in Growth-and-Collapse mode?

If you have been following my prior postings on the French Economy (here) you will see that France is approaching a Steady State Economy (maybe).*** French GDP, on the other hand, is best predicted by the World System (see the Notes below), which is in growth-and-collapse mode (see the WL203 Model). To avoid this future, some geopolitical re-orientation** will be needed.

Looking at the AIC statistics (below). One approach would be to reorient the French Economy to Business as Usual (BAU, [16.16 <  AIC =  21.38 < 25.79]) with no external geopolitical alignments, which produces the steady-state forecast above.  Compared to the World System model (W, [-97.78 < AIC =   -46.24 < -9.145]) which has the best AIC, this will require some geopolitical work. One path would be through the Random Walk (RW, [ -67.1 <   AIC =  -56.52 < -47.09]) which has a very good  AIC but is not predictive of the future.

One charitable analysis of current French Political Instability is that it is part of a Random Walk walk process to find a different Geopolitical Alignment. There is no guarantee that the process will end in a Steady State Economy.

You can experiment with the FRL20 BAU model here and explore alternative futures.

Notes

** The New York Times also reports (here) that France's ability as a geopolitical actor has been compromised by the current Political Instability.

*** The issue surrounding unending (exponential, unstable) Economic Growth, Limits to Growth, Steady State Economies,  and Growth-and-Collapse Modes have been uncovered (again--the debate has been ongoing since the 1970s) in the 2025 Nobel Prize for Economics. Joel Mokyr, a historian and one of the recipients, argues (here) that a Steady State Economy where growth has ended is unlikely. If history is any guide, new innovations will always break the Steady State and create more opportunities for growth. However, see my discussion of French Technology Cycles here and here. Also see Joseph Schumpeter's Creative Destruction model (here). I will review all these models in a future post, but as an obvious summary: no one knows the future. I can only report on forecasts from models and wait to see what happens to the actual French Economy.

AIC Statistics



Technology Models

The Nobel Price in Economics was recently award to Phillipe Aghion, Peter Howitt and Joel Mokyr for their work on How Technology Drives Economic Growth. Reviewers have claimed that their work has relevance for current Growth Policy. For this post, I can just point out that the two technology input models, TECHE (TECH Efficiency) and TECHP (Tech Productivity) are reasonable competitors but not the best models.

An interesting point is that the Aghion-Howitt model with Creative Destruction defaults to a Random Walk (RW) with drift.

I will investigate La French TECH in a future post (here).





Thursday, October 9, 2025

World-System (1970-2150) Six Futures for Colombian Growth

 


The Trump II Administration's Gunboat Diplomacy in the Caribbean  has brought Colombia into the supposed War on Narco-trafficers. The BBC Reports (here) that

Colombian President Gustavo Petro has said that a boat recently bombed by the US was "Colombian with Colombian citizens inside", an allegation the White House called "baseless".

To provide some background to Trump's legally questionable military actions in International Waters, it would seem useful to understand the Economy of Colombia and how it might be affected by possible US military action. Google AI provides the following overview:


Colombia also has an historical reputation as a Cocaine Trafficker, but the Cocaine trade has recently been disrupted as a result of domestic and global forces that are reshaping the drug industry.

My first question, in this post, is what Geopolitical Alliances would be best for Colombia and how might Trump's Drug War affect the alliances. The graphic at the beginning of this post shows six "reasonable" future alignments for Colombia (some "unreasonable" futures are discussed below). Alignment with the US is cyclical and unstable, as is alignment with Russia (RU) and just continuing Business as Usual (BAU). Focusing on Globalization (KOF) provides the most stable future and is the best model from the standpoint of the AIC (see the Notes below, [23.47 < AIC = 51.86 < 81.82). Linkage with the World System (W) generates collapse starting around 2025.


There are a number of the collapse scenarios for Colombian Geopolitical Alignment (graphic above): Emphasis on Technical Efficiency (TECHE) and linkage with China (CN), the differences being when the collapse is forecast to start.






Notes




Google AI on the Economy of Columbia:


ChatGPT on Columbia Globalization:

 AIC Statistics:




Tuesday, September 30, 2025

World-System (1960-2100) Eight Forecasts for Venezuelan Growth

 


November 14, 2025. The New York Times is reporting (here) that:

The Trump administration is rapidly escalating its pressure campaign against Venezuela, with America’s largest aircraft carrier, the Ford, about to take up a position within striking distance of the country, even as President Trump’s aides provide conflicting accounts of what, exactly, they are seeking to achieve.

The urgent question in the media right now seems to be whether Trump will invade Venezuela or just continue marching around in the Caribbean trying to insert his Manifest Destiny. My question is more general: what is best for the Economy of Venezuela, particularly in terms of Geopolitical Alignments?

The graphic above shows eight different growth forecasts for VE1 (the overall growth component of the the VEL20 model, see the Notes below). From the AIC statistics alone, the Random Walk (RW, not surprising since "the Venezuelan economy has been in a state of total collapse since 2013" here) is the best short-term model while Geopolitical Alignment with the Latin American Region (LAC) provides the best stable Attractor Path. But, these Geopolitical Alignments are not the only options.

The "best" options for a stable growth path, just looking at the time series plot, are for Venezuela to pursue Technical Efficiency (TECHE), another version of the Business-as-Usual (BAU) model with emphasis on Technology. Another way of saying that the Venezuelan Economy should be left alone to pursue it's own business.

Linkages with the US, World System (W) and (worst of all, China, CN) are all worse than a Random Walk. Emphasizing Neoliberal Technical Productivity (TECHP) is also worse than a RW.

You can run the VE20 BAU model here and explore alternative futures.


Notes

Background


VEL20 Measurement Model






A More Detailed Look at the Economy of Portugal

In a prior post (here), I reviewed an journalistic article from the Atlantic (here) that tried to understand the puzzle of Portugal's economy using a sampling of conventional wisdom from macroeconomic analysis. One of the articles mentioned in the piece was published by the prestigious Brookings Institution and written by Prof. Ricardo Reis of Columbia University (here). It provides a more in depth, academic, macro-economic analysis that is worth some careful reading. The graph above is taken from earlier post and annotated with various historical fictions about the economy of Portugal from A-E. I will refer to each fiction in the discussion below.





Notes


World-System (1960-2100) Venezuela Growth Forecast

Venezuela is in the news right now for a number of reasons, mainly that the Trump Administration appears to be preparing an invasion. For the future, the VE20 model produces a rather wide range of overall growth forecasts (98% Bootstrap Prediction Interval on VE1) from exponential growth to collapse. 

The VE20 BAU Model is stable and will reach Steady State Economy sometime after 2100. For exponential growth or collapse to occur, the model would have to be destabilized, which might happen as a result of invasions or other external interference. An argument can be made that the approach Steady State Economy is part of the explanation for the ongoing  crises.

You can run the VE20 BAU model here and explore alternative futures.


Notes

Data are taken from the World Development Indicators (WDI). All variables are in standard scores. The methodology used to create forecasts is similar to the one used by the Atlanta Federal Reserves GDPNow app. Prediction intervals are generated using a Bootstrap algorithm in the R programming language (see the Boiler Plate).

You can run the VE20 Model here.

Measurement Model

The VE1 state variable is an overall weighting of all the indicators, VE2 is an Unemployment-GDP Controller and VE3 is an Environmental-Human Development vs. Emissions-Energy controller. The time plots are presented below.

VE20 BAU System Matrix


For the BAU model [147.6 < AIC = 167.2 < 182.8].

VE20 BAU Model Step-Ahead Predictions


The VE20 BAU model does a fairly good job of tracking the historical state variables except for turning points and during crisis periods, of which there are many.





 

World System (1940-2040) Inflation in Venezuela


NPR recently ran a story (here) titled "As Inflation Soars, Venezuela's Leader Opts for Drastic Steps." Venezuela has had past historical episodes with inflation, particularly in the 1990's and the new outbreak caught my attention. Why the new outbreak? To provide some context for my explanation, here's some background.

The conventional explanation for problems in Venezuela's economy (here) invokes the petroleum and manufacturing sectors.


However, results from using the VEL20 model indicate that the best driver for controlling Inflation in Venezuela is Globalization (the KOF Index) and (from the graphic above) Globalization peaked around 2000 (right when inflation was at it's lowest). The peak in KOF was essentially a "Globalization Bubble" during the Chavez government (1998-2013).


Notes


The Systems Model for Inflation (F=GDP Deflator) with input from Globalization (G=KOF). The model provides both the best short-term and attractor path predictor for Inflation [426.1 < AIC =   453.1 < 477.1].