
.
Founder
In the AI age, almost any topic can be consulted in an instant at near zero marginal cost. This means that the market for off-the-shelf reports on products and markets is undergoing a significant contraction. A company in the building sector, for example, can use AI Deep Research to interrogate the conditions in Chile or Morocco in its niche; it doesn’t need to buy a customized report.
However, when it comes to understanding the economies of different countries this logic breaks down. Of course, AI Deep Research can generate multi-page summaries, often of quality, but it is in the interpretation of this output where the AI model falls short. Countries, alas, are not like the market for given products (e.g. window frames) where various metrics can create a reasonably reliable picture of the business opportunity.
A country is the functioning of the whole and to date, or at any time in the future, there is no way of modelling it mathematically except for short term forecasting. Countries are inherently complex and economic trajectories are determined by a myriad of factors which comprise political, social, external, geopolitical and other variables. Every country is on its own path, which is a function of its history and the decisions taken by Government and other actors over many decades. Its future will also be determined by its path and the capacity society and the political system has for reform.
Basic macroeconomic summaries, historical data aggregations, and standardized sector overviews are now routinely synthesized by artificial intelligence engines at near-zero marginal cost. However, this commoditization of descriptive reporting has not eliminated the necessity of economic intelligence; rather, it has shifted corporate demand toward specialized Economics Country Reports grounded explicitly in expert human judgment.
The displacement of human analysts by automated systems encounters immediate structural barriers when applied to macro-level country assessments. Machine learning models excel at processing real-time sentiment metrics across vast, multi-lingual datasets. Yet, when tasked with evaluating country-level tail risks, structural regime changes, or complex state behaviour, algorithmic outputs exhibit systemic vulnerabilities.
There are various competing providers of Economics Country Reports who focus on current analysis and short term forecasting. We, on the other hand, take a long term view. Looking back over several decades we let the data tell the story and refer to political events to explain the data. When the past trajectory of a country is laid out it is possible to look into the future and speculate on its likely future course.
Any company operating internationally needs to understand the markets in which it does business. This does, of course, include sectoral information but also encompasses the overall political and economic environment. If a company is making a long term commitment it need a long term view.
Banks, insurance companies, asset managers and other financial entities are continuously reevaluating the risks in various markets. Whereas this is predominantly based on short-term forecasting there is also a role for a long term view since the short term eventually merges with the long term.
Our analysis is to a significant extent derived from the databases provided by multilateral institutions including the IMF, World Bank, UN etc. but the users of such data will also benefit from our analysis.