Today's Letter IEDI presents data recently released by the International Monetary Fund (IMF) on the set of industrial policies adopted by countries in 2023, based on the database of the "New Industrial Policy Observatory" (NIPO).

The IMF Observatory records and monitors different industrial policy actions (taken in a broad concept), implemented around the world and tries to fill an important information gap. 

Unlike other policies, such as monetary and fiscal policy, there is no systematic and quality data on a global scale for industrial policy. As a result, the Fund warns that empirical evaluations on the subject tend to be partial and limited, usually restricted to sectors or national cases.

The data from the new Observatory were compiled and analyzed in the IMF document “The Return of Industrial Policy in Data,” prepared by Simon Evenett, Adam Jakubik, Fernando Martín and Michele Ruta, released in January 2024. It is an effort to identify patterns of this new phase of industrial policy in advanced and emerging countries.

The database compiles information on 75 jurisdictions (countries or economic blocs), representing 94% of world GDP, that adopted some type of industrial policy measure last year, highlighting the form of interventions and the motivations indicated by governments. In 2023, an additional 2,580 industrial policy measures were identified worldwide, 71% of which may impact global trade dynamics, according to the IMF.

With these data, the Fund's researchers not only call attention to some recent trends, but also carry out preliminary econometric exercises to verify the correlation between the use of industrial policy and elements of political economy, dynamics of retaliation by other governments, or domestic macroeconomic (cyclical) conditions.

Some of the main findings of the study are summarized below:

 •  The recent wave of industrial policy is mostly driven by advanced economies, with subsidies being the main instrument used;

 • Industrial policies are concentrated in China, the European Union and the United States, which together account for 48% of the measures;

 •  As for instruments, domestic subsidies, export incentives and localization measures (local content, for example) are, in order, the three most common instruments in advanced economies, which have more fiscal space;

 • Trade restrictions on imports are the second main instrument used by emerging and developing economies, after domestic subsidies, mirroring advanced economies' export subsidies;

 •  Instruments vary greatly on a case-by-case basis: export incentives are most employed by Canada, Germany, Japan, and South Korea; localization measures by the United States and India; restrictions on exports by China, India, and Russia; import barriers by low-income developing countries;

 •  In the current phase, the use of industrial policy is no longer restricted to the existence and correction of market failures;

 • Strategic competitiveness is the main reason why governments use industrial policy measures (more than 1/3 of the cases), associated with other rising objectives, such as combating climate change (28%), supply chain resilience (15%) and national security/geopolitical tensions (20%);

 •  Countries with high export concentration tend to adopt industrial policy measures more often, consistent with the idea of promoting economic diversification through industrial policy;

 •  The actions implemented are positively correlated with the previous use of measures by other countries in the same sectors, which indicates a “ripple effect” dynamics in the adoption of industrial policy;

 •  Data for Jan'21 to Dec'23 for China, the European Union and the United States show that, on average, there is a 73.8% probability that a subsidy for a given product by one of these large economies will lead to the adoption by another of them of a subsidy for the same product within one year;

 •  Sectorally, in early 2023 the emphasis was on medical goods, but they were quickly overtaken by military/civilian dual use products, advanced technologies products, including low-carbon ones, as well as semiconductors and critical minerals;

 •  There is a positive correlation between the number of industrial policy measures and the occurrence of elections in the year of their implementation or in the following year;

 •  Right-leaning governments on the political spectrum tend to adopt more industrial policy measures than left-leaning governments; and better-rated governments (in terms of effectiveness) are also more likely to undertake industrial policy measures;

 • Domestic macroeconomic (i.e., cyclical) conditions may also be correlated with the use of industrial policy measures, such as in exchange rate appreciation episodes, which erode local product competitiveness.

The full text is available in Portuguese.