Industrial policies have experienced a resurgence in recent years, including in emerging markets, as governments seek to reshape their economies and assert greater control over industrial sectors. This trend is likely driven by political economy factors, such as a desire for a more active state, and in response to foreign governments supporting their own domestic industries. This first in a series of three columns expands the available cross-country data on industrial policies and documents key patterns in how these policies are implemented, particularly in economies with lower administrative and fiscal capacity.
Industrial policies – policies to change the sector composition of production in an economy – have seen a resurgence in recent years. While their track record has been mixed (Hasanov and Cherif 2019, Kalouptsidi et al. 2019, Juhász et al. 2023, Barwicket al. 2024, Millot and Rawdanowicz 2024), their growing popularity may reflect domestic political economy considerations: a desire for the state to play a larger role in the economy (Kóczán and Plekhanov 2024) and voters’ preference for subsidies over taxes as the costs of the former are less salient (EBRD 2024a). Industrial policies may also appear more attractive when other countries are supporting their own industries, especially in the presence of increasing geopolitical fragmentation.
In recent research (EBRD 2024b, Kóczán et al. forthcoming), we build on the database of Juhász et al. (2023) using large language model (LLM) processing to document recent trends in industrial policies, in particular in emerging markets. We document four key stylised facts.
1. The use of industrial policies is on the rise, including in economies with less administrative and fiscal capacity
The use of industrial policies is on the rise in advanced economies and in other emerging markets, with increases in both the number of new policies announced in a given year and the number of policies in place at any given point in time (see Figure 1). The proportion of exports and imports covered by industrial policies has also been rising. Our results suggest that while industrial policies are more commonly seen in higher-income economies, they have been increasingly deployed in economies with lower levels of administrative and fiscal capacity.
Figure 1 The number of industrial policies in place has increased rapidly since 2019
Source: GTA, Kóczán et al. (2024), Juhász et al. (2023) and authors’ calculations.
Note: Consistent data on China are not available for the period 2021-22 owing to lags in reporting.
2. Economies with lower administrative and fiscal capacity are more likely to rely on distortive instruments
The instruments deployed to pursue industrial policies differ vastly. To a large extent, the choice of policy instrument is dictated by the sector, the objective of the policy and the structure of the market. Globally, grants (supporting innovation or IT start-ups, for example), export finance, import tariffs, and loans and loan guarantees provided by the state (often on concessional terms) are the most common instruments, accounting for 67% of industrial policies. Other commonly used instruments include public procurement requirements favouring certain producers, incentives for localising value added in production chains, financial assistance abroad and production subsidies.
We further find that while distortive instruments, such as import/export bans, quotas and licensing requirements are, on average, used relatively infrequently, they are more common in economies with lower levels of administrative and fiscal capacity (see Figure 2, which plots instruments against the average administrative capacity of the economies that implement them on the horizontal axis and the average capacity to raise fiscal revenue on the vertical axis).
In contrast, instruments such as trade finance, incentives to localise value added, and localisation requirements in public procurement are associated with relatively high levels of administrative capacity. Accordingly, they are more common in richer economies.
Figure 2 Less distortive instruments require higher bureaucratic quality and greater revenue raising capacity
Source: Juhász et al. (2023), Kóczán et al. (2024), V-Dem and authors’ calculations.
Note: The size of each bubble is proportionate to the number of industrial policies that use the relevant instrument globally. Based on 147 economies. “Bureaucratic quality” refers to the V-Dem indicator assessing the rigour and impartiality of a public administration, while “revenue-raising capacity” refers to the V-Dem indicator assessing sources of fiscal revenue. Figures for bureaucratic quality and revenue-raising capacity are averages over the period 2010-21 for economies that implement industrial policies using the relevant instrument. The figure only shows instruments that are used to implement at least 75 policies globally, with selected instruments labelled. Instruments that are considered highly distortive in IMF (2024) labelled in red.
3. Most industrial policies discriminate against foreign interests and firm-specific policies are common
While competitive elements can help minimise distortions associated with ‘picking winners’, recent industrial policies, both in advanced economies and in emerging markets, have predominantly discriminated against foreign interests (for instance, by establishing import barriers or subsidising domestic producers; see also Evenett et al. 2024).
Industrial policies that are firm-specific are also common (see Figure 3). Over 80% of industrial policies in Canada and the US target specific firms (through export-import loans, for instance). China stands out as having a high percentage of firm-specific policies for its level of development, with its policies typically targeting large (often state owned) enterprises in the manufacturing sector. In poorer countries, the share of firm-specific policies tends to be lower.
Figure 3 In lower-income economies, industrial policies are less likely to target specific firms
Source: GTA, Juhász et al. (2023), Kóczán, Marino and Plekhanov (2024) and authors’ calculations.
Note: The size of each bubble is proportionate to the total number of industrial policies announced in the relevant economy over the period 2010-22. Figures for firm-specific policies are based on the GTA classification. Data on firm-specific policies are averages covering the period 2010-22. Only economies with at least 10 industrial policies are shown. The line is fitted to all economies shown in the figure, with selected economies labelled. The horizontal axis shows GDP per capita in US dollars at market exchange rates.
4. Sunset clauses have become more widespread
Subsidies given to specific firms or narrowly defined industries can result in ‘addiction’ and calls for policies to be extended indefinitely, regardless of their benefits. As a result, industrial policies tend to be easier to introduce than abandon, with infant industry policies often continued well beyond those industries’ childhood years. Our findings suggest that the incorporation of sunset clauses (automatic end dates) has become more common in recent years. This development, which has coincided with the rise in more addictive instruments such as subsidies, is welcome and may, to some extent, indicate that countries are learning from decades of past experience with industrial policies. Nonetheless, policies with sunset clauses remain a minority.
Policy implications
Past experience with industrial policies offers several lessons. First, policies should incorporate competitive pressures and market-based tests, such as outward orientation and incentives for knowledge transfer. Evaluation mechanisms must be seen as an ongoing learning process. Including sunset clauses from the outset can help prevent policy ‘addiction’ and make it easier to phase out support measures when they are no longer needed. Finally, given the critical importance of effective implementation, industrial policies should be paired with continuous investment in administrative capacity and bureaucratic quality.
Source : VOXeu
