## 2.1 Industrial Policies; 2.2 Patents; 4.1 Protectionist industrial policies; Appendix extensive/intensive margins (excerpt)

## Source details

**Canonical URL:** [2.1 Industrial Policies; 2.2 Patents; 4.1 Protectionist industrial policies; Appendix extensive/intensive margins (excerpt)](https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026047-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2026/english/wpiea2026047-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2026/english/wpiea2026047-source-pdf.pdf.json)

---

### Definition, scope, and measurement of Industrial Policies (IPs)
- Industrial policies (IPs) are defined as goal-oriented state interventions intended to shift the composition of economic activity (Juhász et al., 2025).
- Dataset construction:
  - Source: Global Trade Alert (GTA) database (Evenett and Fritz, 2020) via a machine-learning classifier applied to policy descriptions recorded in GTA between 2009 and 2022.
  - Coverage: 177 countries, 31 manufacturing sectors (ISIC 2-digit), period 2009–2019.
  - An IP count is built for each country-sector-year; non-reported counts treated as zeros if the country has never reported an IP in that sector since 2009.
  - Reporting-lag adjustment: retains only policies recorded in the same calendar year as their announcement.
  - Multi-sector/country policies are double counted to capture spread.
- Concordance and mapping:
  - Targeted products coded with HS1992 product codes and CPC industry codes.
  - WITS concordance links products targeted by IPs to ISIC Rev. 2 sectors (WITS, 2025).

### Classification by discriminatory intent and instrument
- GTA red-amber-green classification used:
  - Protectionist IPs: “red” measures that “almost certainly discriminate against foreign commercial interests.”
  - Liberalizing IPs: “green” measures that “liberalize towards foreign commercial interests.”
  - Ambiguous (“amber”) measures excluded; they represent 5 percent of IPs in GTA, compared to 63 percent for protectionist IPs and 32 percent for liberalizing ones.
- Instrument groups (aligned with UN MAST):
  - (i) trade barriers (export/import restrictions)
  - (ii) domestic subsidies
  - (iii) export incentives
  - (iv) local content requirements
  - (v) other instruments (e.g., public procurement or FDI measures)
- Instrumental composition (2009–2019):
  - Protectionist IPs: subsidies 55% and export incentives 35%.
  - Liberalizing IPs: import liberalization measures 84%.
  - Other instruments less implemented, limiting variation and precision.

### Identification of climate-related and low-carbon technology (LCT) IPs
- Three-step classification:
  1. Map HS codes corresponding to LCTs from literature; code related products as LCTs.
  2. Aggregate to sectors; define climate-related IPs as those targeting sectors where at least 70 percent of products are LCTs.
  3. Use a large language model (LLM) to detect climate mitigation or environmental motives in GTA policy descriptions; manual validation resolves HS-LLM disagreements.
- Methodology references include Huang et al. (2025) for LLM methods and several studies for product identification (Pigato et al., 2020; OECD/Eurostat, 1999; Rosenow and Mealy, 2024; Kowalski and Legendre, 2023; Goldschlag et al., 2020; Hasna et al., 2023; Mealy and Teytelboym, 2022).

### Use and trends of IPs
- GTA compiles governments’ policy measures that may disadvantage foreign commercial actors; use of IPs has increased sharply since 2017 across AEs and EMDEs.
- Analysis focuses on changes in the number of active IPs rather than stock levels and includes country and year fixed effects.

### Limitations and caveats of the IP data
- Reporting heterogeneity may bias IP counts.
- Data start date 2009 undercounts pre-existing IP stocks; mitigation: focus on changes and include fixed effects.
- GTA captures only policies affecting foreign commercial interests; purely domestic or subnational interventions omitted. Robustness check excludes countries with potentially large subnational IPs (see Section 4.5 in the source).
- Dataset records presence, not intensity, of policies; cannot discuss policy magnitudes.
  - Note: New Industrial Policy Observatory analysis shows a positive correlation between IP count and the total value of subsidies in 2023 (Evenett et al., 2024).

---

### Measurement and data sources for Patents (INPACT-S)
- Innovation measured using patent applications via INPACT-S (LaBelle et al., 2024) leveraging PATSTAT.
- Coverage: domestic patent applications and flows across manufacturing sectors (ISIC Rev. 2) from 1980 to 2019.
- Attribution:
  - Fractional counting assigns patents by residence of applicant and inventors.
  - For regional patent authorities (e.g., European Patent Office), a weighted-dispersion method distributes applications across member states.
- INPACT-S includes all family filings; dataset spans 1980–2019, 31 manufacturing industries, 212 countries.
- Constructed country-sector-year measures:
  - domestic inventions – same country of invention and application,
  - patents received from abroad,
  - patents submitted abroad.
- Merged IP–patent dataset covers 180 countries, 31 sectors, period 2009–2019.
- Empirical patterns:
  - Highest yearly patent applicants: China, the USA, Japan, Korea, and Germany.
  - Top three sectors by patent applications: chemistry; medical and precision equipment; computing machinery.
- Dependent variable in local projection regressions:
  - ∆^h log(PatentStock_{c,s,t+h} + 10^{-6})
- Regressors include ∆IPStock_{c,s,t}, lagged log patent stock log(PatentStock_{c,s,t−1} + 10^{-6}), lagged IPStock_{c,s,t−1}, controls X_{c,s,t}, fixed effects λ_{c,s}, λ_{c,t}, λ_{s,t}.
- Limitations:
  - Cross-country and cross-industry patenting behavior heterogeneity; patent counts may not capture patent quality and can reflect strategic behavior.
  - Inclusion of all family filings traces innovation trajectories but complicates isolating original inventive steps.

---

### Average effects of protectionist industrial policies on patenting
- On average, protectionist IPs do not have a significant effect on patent applications by domestic inventors over the four-year horizon.
- Foreign inventors:
  - Protectionist IP introduction is, on average, associated with a 1.4 percent increase in the number of foreign patent applications in the targeted sector.
  - Effect is short-lived and becomes statistically insignificant after the second year.
- Interpretation:
  - Domestic inventors likely have innovations already in the pipeline and cannot accelerate filings to benefit from the intervention.
  - Foreign inventors appear to front-load patent applications for existing innovations in response to policy implementation—securing market access or strengthening intellectual property protection—rather than creating new innovations.

### Heterogeneity by instrument
- Subsidies and export incentives drive the short-term increase in foreign patent applications:
  - An additional protectionist subsidy is associated with a 2 percent increase in received foreign patent applications in the first year.
  - Export incentives have a slightly higher and more sustained effect than subsidies, but:
    - Export incentives are less frequently implemented, yielding less variation.
    - Estimates for export incentives are more imprecise and statistically insignificant; first stage weaker than with subsidies.
- Robustness: Appendix Figure A.16 shows stronger statistical significance and instrument relevance when standard errors are clustered at the country-sector level.

### Comparison with literature and mechanisms
- Findings align with evidence on export-oriented IPs supporting East Asia’s export-led growth model.
- Mechanism: export-oriented IPs help firms overcome domestic market limitations, facilitating technological capability accumulation and scale economies.

### Country-income heterogeneity
- Advanced Economies (AEs):
  - Protectionist IPs have no significant impact on patenting by domestic and foreign inventors over the observed horizon.
  - Note: in AEs most protectionist IPs were implemented after 2017, limiting analysis beyond two years.
  - AEs were the destination of 74 percent of all patents submitted by foreign inventors between 1990 and 2019.
  - Between 1990 and 2021, AEs accounted for 96% percent of patent filed abroad.
- Emerging Market and Developing Economies (EMDEs):
  - Protectionist IPs associated with an increase in received foreign patent applications, consistent with acceleration of transfers of existing technologies.
- Liberalizing IPs:
  - Promote cross-border patenting, particularly from AEs to other countries.
  - In EMDEs, liberalizing policies not significantly associated with increased patenting (possibly due to low variation).
  - In AEs, liberalizing IPs significantly increase outward patenting by inventors based in AEs.
  - Lifting one additional liberalizing policy is associated with a 4 percent increase in received foreign patent applications after four years.

### Targeting: sectoral variation and policy implications
- Average protectionist IPs do not increase domestic patenting within four years, but targeted IPs can:
  - Infant industries:
    - IPs targeting infant industries are followed by a statistically significant increase in patenting by domestic inventors after two years, reaching about a one percent increase after two years.
    - These IPs do not increase foreign patent applications.
    - Interpretation: targeted support for young, financially constrained, and high potential-for-learning-by-doing firms can generate positive innovation responses; effects stronger for younger and smaller firms by alleviating financial frictions.
    - Methodological note: OLS estimates used because infant-industry IPs are strategic rather than retaliatory; absence of upward pre-trends supports plausibility of causal interpretation or a lower bound estimate.
  - Low-carbon technologies:
    - Protectionist IPs with climate objectives are associated with a gradual and statistically significant increase in patenting by domestic inventors, reaching over 0.7 percent three years after implementation.
  - Innovation-central industries:
    - One additional protectionist IP targeting an innovation-central industry is associated with a 3.7 percent increase in received foreign patent applications.
    - One additional liberalizing IP applied to an innovation-central industry is associated with a 10.6 percent significant increase in received foreign patent applications after 4 years.
    - Industries defined as innovation central if eigenvalue centrality in the country’s innovation network is above the 80th percentile; results robust to using median cutoff and PageRank centrality.

### Synthesis on timing and design
- Protectionist IPs tend to accelerate patent filings for existing innovations—especially by foreign inventors—rather than stimulate new innovation within four years.
- Liberalizing IPs (mainly lifting import barriers) generate larger and more persistent increases in received foreign patent applications over a four-year horizon.
- Effectiveness depends on design, timing of implementation/removal, and targeting; infant-industry and climate-targeted IPs can foster domestic innovation within considered horizons.

---

### Appendix: Extensive vs. intensive margin findings and robustness
- Extensive margin:
  - Coefficients on the extensive margin are significant, indicating climate-related IPs foster development of new innovation ecosystems in previously un-patented sectors.
  - Dependent variable for extensive-margin regressions equals 1 if the sector-year-country has more than one patent, zero otherwise.
  - Decomposition shows an increasing pre-trend on the extensive margin (entry of new patenting sectors) and a decreasing pre-trend on the intensive margin.
- Climate-related vs non-climate-related IPs:
  - Climate-related IPs show stronger effects on developing new patenting sectors (extensive margin) compared to non-climate-related IPs.
  - Difference between climate-related and non-climate-related IPs is insignificant for foreign patenting; both facilitate foreign patenting to the same extent.
- Innovation-central industries:
  - Targeting innovation-central industries yields substantially larger foreign patenting effects; liberalizing policies on central industries yield particularly large effects.
  - Non-central industries: no statistically significant impacts; liberalizing IPs may temporarily decrease received patent applications in non-central sectors.
- Heterogeneity by country income and policy type summarized:
  - Protectionist IPs disproportionately facilitate receipt of patents in EMDEs.
  - Liberalizing IPs encourage cross-country patenting from AEs.
  - Protectionist export-oriented policies outperform subsidies in temporarily increasing received foreign patents; liberalizing IPs produce more persistent cross-border patenting.
  - Both protectionist and liberalizing IPs have no significant effect on domestic patenting on average within four years.
- Robustness checks:
  - Hansen J-test: over-identifying restrictions not rejected across horizons for liberalizing IP IVs.
  - Excluding China, the United States, and Germany leaves results robust.
  - Clustering at the country-sector level improves first-stage F-statistics; IV results unchanged.
  - Including broader trade policies does not change results materially.
  - Policy-instrument granularity: subsidies and export incentives results robust under alternative clustering; instrumenting by same tool in other countries/sectors yields similar shapes but smaller magnitudes for export incentives.
  - Centrality measures and thresholds: results robust to PageRank vs eigenvector centrality and different cutoffs.
  - OLS vs IV: OLS indicates associations but subject to selection bias and pre-trends; IV specifications show absence of pre-trends where instrument validity holds.

### Policy implications and conclusions
- Industrial policies can foster patent applications when well designed.
- Targeting is essential:
  - Prioritize infant industries (e.g., LCTs) to boost domestic patent applications in the medium term by relaxing financing constraints and accelerating innovation in early-stage, high-potential sectors.
  - Prioritize industries central in the country’s innovation network to yield larger gains in received foreign patent applications.
- Effects vary by country fundamentals: firms benefit more from IPs in countries with better governance or financial market development.
- IPs tend to increase technology diffusion to EMDEs; further research required to determine whether this translates into technological development and productivity increases.
- Broader outcomes—expected benefits, spillovers, policy alternatives, general equilibrium dynamics such as trade adjustments or cross-country spillovers—should be carefully evaluated prior to implementation.

*Source: IMF Working Paper — excerpts from sections 2.1, 2.2, 4.1, Appendix, and associated figures/notes as provided in the source PDF.*

### 2.1    Industrial Policies .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  

### 2.1    Industrial Policies

### Definition and scope
- Industrial policies (IPs) are defined as goal-oriented state interventions intended to shift the composition of economic activity (Juhász et al., 2025).
- The analysis uses IPs identified within the Global Trade Alert (GTA) database (Evenett and Fritz, 2020) via a machine-learning classifier applied to policy descriptions recorded in GTA between 2009 and 2022.
- The constructed dataset covers 177 countries, 31 manufacturing sectors (ISIC 2-digit), and the period 2009–2019.

### Construction of the IP measure
- IP counts:
  - A count is built of IPs in place in each country, sector, and year.
  - Non-reported policy counts are treated as zeros if the country has never reported an IP in that sector since 2009.
  - To ensure consistency over time, a reporting-lag adjustment retains only policies recorded in the same calendar year as their announcement.
  - An IP applying to multiple sectors or countries is double counted in the analysis to capture spread.
- Concordance and mapping:
  - Targeted products are coded with HS1992 product codes and CPC industry codes.
  - WITS concordance links products targeted by IPs to ISIC Rev. 2 sectors (WITS, 2025).

### Classification by discriminatory intent and instrument
- GTA red-amber-green classification is used:
  - Protectionist IPs: “red” measures that “almost certainly discriminate against foreign commercial interests.”
  - Liberalizing IPs: “green” measures that “liberalize towards foreign commercial interests.”
  - Ambiguous (“amber”) measures are excluded from the analysis; they represent only 5 percent of IPs in GTA, compared to 63 percent for protectionist IPs and 32 percent for liberalizing ones.
- Instrument groups (aggregated into five groups aligned with the UN MAST classification):
  - (i) trade barriers (export/import restrictions)
  - (ii) domestic subsidies
  - (iii) export incentives
  - (iv) local content requirements
  - (v) other instruments (e.g., public procurement or FDI measures)
- Instrumental composition over 2009–2019:
  - Protectionist IPs were predominantly composed of subsidies (55%) and export incentives (35%).
  - Liberalizing IPs are primarily import liberalization measures (84%).
  - Other IP instruments have been less implemented historically, limiting variation and result precision for those instruments.

### Identification of climate-related and low-carbon technology (LCT) IPs
- Three-step classification to identify IPs supporting LCTs:
  1. Rely on literature to map HS codes corresponding to LCTs, coding related products as LCTs in the product-country-year dataset.
  2. Aggregate to sectors and define climate-related IPs as those targeting sectors where at least 70 percent of products are LCTs.
  3. Use a large language model (LLM) to detect climate mitigation or environmental motives in GTA policy descriptions; manual validation resolves HS-LLM disagreements.
- The methodology references Huang et al. (2025) for LLM methods and Pigato et al. (2020); OECD/Eurostat (1999); Rosenow and Mealy (2024); Kowalski and Legendre (2023); Goldschlag et al. (2020); Hasna et al. (2023); Mealy and Teytelboym (2022) for product identification.

### Use and trends
- The GTA, established in 2008, systematically compiles governments’ policy measures and announcements that may disadvantage foreign commercial actors.
- The use of IPs has increased sharply since 2017 across both advanced economies (AEs) and emerging markets and developing economies (EMDEs).
- Panel A of Figure 1 (described) shows year-to-year change in active country-sector IPs in AEs and EMDEs; Panel B shows composition of change in active IPs between 2009 and 2019.

### Limitations and caveats of the IP data
- Reporting heterogeneity: Differences in countries’ reporting standards may bias IP counts.
- Data start date: The data begins in 2009, undercounting IP stocks in countries with pre-existing IP frameworks.
  - Mitigation: Analysis focuses on changes in the number of active IPs rather than stock levels and includes country and year fixed effects.
- Coverage: GTA captures only policies affecting foreign commercial interests, omitting purely domestic or subnational interventions.
  - Mitigation: Robustness check excludes countries with potentially large numbers of subnational IPs (see Section 4.5 in the source).
- Intensity vs. presence: The dataset captures the presence, not the intensity, of policies, preventing discussion of policy magnitudes.
  - Note: New Industrial Policy Observatory analysis shows a positive correlation between IP count and the total value of subsidies in 2023 (Evenett et al., 2024), which provides some reassurance about representativeness.

*Italic: Source: IMF Working Paper — 2.1 Industrial Policies (excerpt).*

### 2.2    Patents

### 2.2    Patents

### Measurement and data sources
- Innovation activity is measured using patent applications across countries and sectors, building on the INPACT-S dataset developed by LaBelle et al. (2024).
- INPACT-S leverages the PATSTAT database to quantify domestic patent applications and flows of patent applications from one country to another in manufacturing sectors (ISIC Rev. 2) from 1980 to 2019.
- Attribution method:
  - Fractional counting assigns patent applications to origin countries based on both the residence of the applicant and of the inventors.
  - For patents submitted to a regional patent authority (such as the European Patent Office), a weighted-dispersion method distributes applications across individual member states to improve spatial precision.
- Coverage:
  - INPACT-S includes all patent filings, regardless of family size, allowing tracing of the complete trajectory of an innovation (including subsequent filings that often reflect incremental refinements).
  - Resulting dataset spans 1980–2019, covering 31 manufacturing industries and 212 countries.

### Aggregation and constructed measures
- Three country-sector-year measures are constructed:
  - domestic inventions – same country of invention and application,
  - patents received from abroad,
  - patents submitted abroad.
- These indicators are merged with the IP dataset at the country-year-sector level, producing data that covers 180 countries, 31 manufacturing sectors, and the period from 2009 to 2019.
- Empirical representation and notable patterns:
  - China, the USA, Japan, Korea, and Germany have the highest yearly number of patent applications over the considered period (as shown in Figure A.2).
  - The three sectors receiving the most patent applications are chemistry, medical and precision equipment, and computing machinery (Figure A.1).

### Equation detail (as used in empirical analysis)
- Dependent variable specification in local projection regressions includes a small constant to retain zero observations:
  - ∆^h log(PatentStock_{c,s,t+h} + 10^{-6})
- Regressors include changes in IP stock (∆IPStock_{c,s,t}), lagged log patent stock log(PatentStock_{c,s,t−1} + 10^{-6}), lagged IPStock_{c,s,t−1}, controls X_{c,s,t}, and fixed effects λ_{c,s}, λ_{c,t}, λ_{s,t}.

### Limitations and cautions in interpreting patent counts
- Cross-country and cross-industry differences in patenting behaviors can affect comparability (some sectors/countries are systematically patent-intensive; others rarely patent). The analysis partially addresses this via sector and country fixed effects and robustness checks excluding main patenting economies.
- Patent application counts may not capture patent quality (Hall et al., 2001).
- Patent applications can reflect strategic behaviors rather than genuine inventions (Blind et al., 2006).
- The inclusion of all family filings helps trace innovation trajectories but makes isolating the original inventive step more difficult.

*Source: IMF Working Paper — section 2.2 Patents (from the provided chapter PDF).*

### 4.1    Protectionist industrial policies

### 4.1 Protectionist industrial policies

### Average effects on patenting (domestic and foreign)
- On average, protectionist industrial policies do not have a significant effect on patent applications by domestic inventors over the considered four-year horizon.
- Figure A.4 in the Annex shows that patent applications submitted abroad by domestic inventors does not change, consistent with the null effect on domestic innovation.
- Protectionist industrial policies are associated with a short-lived increase in patent applications from foreign inventors:
  - The introduction of an additional protectionist industrial policy is, on average, associated with a 1.4 percent increase in the number of foreign patent applications in the targeted sector.
  - This effect dissipates rapidly and becomes statistically insignificant after the second year.
- Interpretation:
  - Domestic inventors likely have innovations already in the pipeline in their home country prior to policy implementation and therefore cannot accelerate filings to benefit from the intervention.
  - Foreign inventors appear to front-load patent applications for existing innovations in response to policy implementation—securing market access or strengthening intellectual property protection—rather than creating new innovations.

### Heterogeneity by instrument
- Subsidies and export incentives drive the short-term average increase in foreign patent applications following protectionist IP implementation:
  - An additional protectionist subsidy is associated with a 2 percent increase in received foreign patent applications in the first year.
  - Export incentives are associated with a slightly higher and more sustained effect than subsidies, but:
    - Export incentives are less frequently implemented than subsidies, resulting in less variation in the IP variable.
    - Estimates for export incentives are more imprecise (and statistically insignificant) and the first stage is weaker than with subsidies.
- Robustness:
  - Appendix Figure A.16 confirms robustness under alternative clustering assumptions, with stronger statistical significance and instrument relevance when standard errors are clustered at the country-sector level.

### Comparison with broader literature and mechanisms
- Findings align with literature on export-oriented industrial policies supporting East Asia’s export-led growth model.
- Mechanism: export-oriented IPs help firms overcome domestic market limitations, facilitating accumulation of technological capabilities and scale economies that domestic markets alone may not deliver.
- Recent empirical evidence points to more sustained effects of export-oriented IPs on firm performance and trade competitiveness relative to subsidies.

### Country-income heterogeneity
- Protectionist IPs have no significant impact on patent applications by domestic and foreign inventors in AEs over the observed horizon.
  - This may partly be because, in AEs, most protectionist IPs were implemented after 2017, preventing analysis beyond two years.
- Protectionist IPs are associated with an increase in received foreign patent applications in EMDEs, consistent with acceleration of transfers of existing technologies rather than fostering novel innovation.
- Cross-country patenting patterns:
  - AEs have been the destination of 74 percent of all patents submitted by foreign inventors between 1990 and 2019.
  - Between 1990 and 2021, AEs accounted for 96% percent of patent filed abroad.
- Liberalizing IPs promote cross-border patenting, particularly from AEs to other countries:
  - In EMDEs, liberalizing policies are not significantly associated with increased patenting activity (possibly due to low variation in liberalizing IPs in EMDEs).
  - In AEs, liberalizing IPs significantly increase outward patenting by inventors based in AEs.

### Targeting: sectoral variation and policy implications
- While average IPs do not increase domestic patenting within four years, appropriately targeted IPs can yield different outcomes. The chapter focuses on three sectoral targets: infant industries, low-carbon technologies, and innovation-central industries.

- Infant industries:
  - IPs targeting infant industries are followed by a statistically significant increase in patenting by domestic inventors after two years, reaching about a one percent increase after two years.
  - IPs targeting infant industries do not increase foreign patent applications.
  - Interpretation:
    - Targeted support for young, financially constrained, and high potential-for-learning-by-doing firms can generate positive innovation responses.
    - IPs have stronger effects on younger and smaller firms, likely by alleviating financial frictions.
  - Methodological note:
    - OLS estimates are used to study infant-industry IPs because these policies are strategic rather than retaliatory, limiting the strength of the instrument; the absence of upward pre-trends supports the plausibility of a causal interpretation or a lower bound estimation.

- Low-carbon technologies:
  - Protectionist IPs with climate objectives are associated with a gradual and statistically significant increase in patenting by domestic inventors, reaching over 0.7 percent three years after implementation.

### Synthesis and temporal considerations
- Protectionist IPs tend to accelerate patent filings for existing (non-new) innovations—especially by foreign inventors—rather than stimulate new innovation within the first four years.
- Liberalizing IPs (mainly lifting import barriers) appear to generate larger and more persistent increases in received foreign patent applications over a four-year horizon:
  - Lifting one additional liberalizing policy is associated with a 4 percent increase in received foreign patent applications after four years.
- The effectiveness of industrial policies depends on design, timing of implementation and removal, and targeting; some IPs (infant-industry and climate-targeted IPs) can foster domestic innovation within the considered horizons, while average protectionist measures do not.

*IMF WORKING PAPERS — Shaping Innovation: Can Industrial Policies Boost Patent Applications?*

### Appendix indicates that the coefficients on the extensive margin

### Appendix indicates that the coefficients on the extensive margin

### Extensive vs. intensive margin findings
- The coefficients on the extensive margin are significant, suggesting that climate-related IPs foster the development of new innovation ecosystems in sectors previously un-patented.
- The extensive margin effects are estimated by using a dependent variable equal to 1 if the sector-year-country has more than one patent and zero otherwise. As such, the corresponding regression estimates the effect of IPs on developing unpatented sectors.
- By contrast, the average IP has a muted effect on domestic innovation as shown by Figure 3A’s IV estimates that tackle the selection bias of non-climate-related IPs targeting sectors already experiencing innovation gains.
- A decomposition into extensive and intensive margin shows that the pre-trend results from an increasing pre-trend on the extensive margin (i.e., entry of new patenting sectors) and a decreasing pre-trend on the intensive margin.

### Climate-related vs non-climate-related IPs and domestic/foreign patenting
- Climate-related IPs show stronger effects on developing new patenting sectors (extensive margin) compared to non-climate-related IPs.
- The difference between climate-related and non-climate-related IPs is insignificant when it comes to foreign patenting; both policies facilitate foreign patenting to the same extent (shown by overlap of curves in Figure 8B and Figure A.8 of the Appendix).
- Figure 8 summary (OLS estimates):
  - Panel A (Domestic inventors): y-axis measures percentage change in patent applications submitted by domestic inventors; x-axis is horizon in years; green line = effect of climate-related IPs; brown line = effect of non-climate-related IPs. Percentage changes estimated with 100×(exp(βh)−1). Regression controls: non-protectionist IPs, non-IP trade policies, sector-year, country-year and country-sector fixed effects. Standard errors clustered at the country level. Dashed lines represent 90 percent confidence intervals.
  - Panel B (Foreign inventors): analogous setup for patent applications submitted by foreign inventors.

### Innovation-central industries
- Targeting innovation-central industries yields substantially larger foreign patenting effects:
  - One additional protectionist IP targeting an innovation-central industry is associated with a 3.7 percent increase in received foreign patent applications, nearly double the average effect.
  - One additional liberalizing industrial policy applied to an innovation-central industry is associated with a 10.6 percent significant increase in received foreign patent applications after 4 years.
- IPs applied to non-central industries show no statistically significant impact; liberalizing IPs may temporarily decrease received patent applications in non-central sectors.
- Industries are defined as innovation central if their eigenvalue centrality in their country’s innovation network is above the 80th percentile of the country’s innovation centrality distribution (Section 2.3). Results are robust to:
  - Changing cutoff from the 80th percentile to the median (Figure A.18).
  - Replacing eigenvector centrality with PageRank centrality (Figure A.19).

### Heterogeneity by country income group and policy type
- Protectionist IPs disproportionately facilitate the receipt of patent applications in EMDEs, as foreign inventors may try to protect their invention or secure access to markets impacted by these policies.
- Liberalizing IPs encourage cross-country patenting from AEs, potentially due to access to lower-cost or higher-quality imported inputs or increased product market competition.
- Protectionist IPs temporarily increase received foreign patents, with export-oriented policies outperforming subsidies.
- Liberalizing IPs (e.g., lifting trade barriers) produce more persistent and broader cross-border patenting.
- Both protectionist and liberalizing IPs have no significant effect on domestic patenting on average, suggesting that new innovation likely takes more than four years to materialize.

### Robustness checks and inference
- Hansen J-test: For liberalizing IPs instrumented with two IVs, Figures A.11 and A.12 show the null hypothesis that over-identifying restrictions are valid cannot be rejected across all horizons.
- Excluding major trading/patenting economies: Excluding China, the United States, and Germany from estimations leaves results robust (Figure A.13).
- Clustering standard errors at the country-sector level: Results remain consistent; first-stage F-statistics improve and overall IV results are unchanged (Figure A.14).
- Broadening trade policy responses: Including a broader set of trade policies does not change results; magnitude and shape of effects largely unchanged, possibly smaller estimates (Figure A.15).
- Policy instrument variations:
  - Results for subsidies and export incentives are unchanged when clustering standard errors at the country-sector level; first-stage F-statistics are above 10 under this less conservative assumption (Figure A.16).
  - Using the same policy tool in other sectors and countries as an instrument yields similar shapes but slightly smaller magnitudes and insignificance for export incentives (Figure A.17).
- Centrality robustness: Results on innovation-central sectors are not sensitive to centrality metric or threshold (Figures A.18 and A.19).
- OLS vs IV:
  - OLS provides a benchmark and indicates associations but is subject to selection bias and pre-trends.
  - Example OLS finding: An additional protectionist IP correlates with a 0.18 percentage point increase in domestic patent applications after one year, but an upward pre-trend suggests selection bias.
  - IV specification shows absence of pre-trends, supporting causal interpretation where instrument validity holds.

### Policy implications and conclusions
- Industrial policies can foster patent applications when well designed.
- Targeting is essential:
  - Prioritizing infant industries (e.g., LCTs) can effectively boost domestic patent applications in the medium term by relaxing financing constraints and accelerating innovation in early-stage, high-potential sectors.
  - Prioritizing industries that are central in the country’s innovation network can yield larger gains in received foreign patent applications.
- Effects vary by country fundamentals: firms benefit more from IPs in countries with better governance or financial market development.
- IPs tend to increase technology diffusion to EMDEs; additional research needed to determine whether this translates into technological development and productivity increases.
- Broader outcomes (expected benefits, spillovers, policy alternatives, general equilibrium dynamics such as trade adjustments or cross-country spillovers) should be carefully evaluated prior to implementation.

*Source: IMF Working Paper — Appendix material as provided in the content unit.*

### References

### References

### Key themes covered by cited literature
- Carbon taxes, path dependency, and directed technical change in the auto industry.
- Industrial policy design, competition, and political economy.
- Transition to green technology and green innovation diffusion.
- Trade liberalization, intermediate inputs, and effects on productivity and innovation.
- Patent behavior, patent citation data, and the technological composition of industries.
- Misallocation, manufacturing TFP, and capital allocation implications.
- Empirical and theoretical assessments of industrial policy effectiveness and measurement.
- Raw materials critical for the green transition, production, trade, and export restrictions.
- Global value chains, state aid effects, and firm-level responses to industrial interventions.

### Selected notable references (as listed in the source)
- Aghion, Philippe, Antoine Dechezleprêtre, David Hemous, Ralf Martin, and John Van Reenen, “Carbon Taxes, Path Dependency, and Directed Technical Change: Evidence from the Auto Industry,” Journal of Political Economy, 2016,124(1), 1–51. doi:10.1086/684581. Publisher: University of Chicago Press.
- , Jing Cai, Mathias Dewatripont, Luosha Du, Ann Harrison, and Patrick Legros, “Industrial Policy and Competition,” American Economic Journal: Macroeconomics, October 2015,7(4), 1–32.
- , Lint Barrage, Eric Donald, David Hémous, and Ernest Liu, “Transition to Green Technology along the Supply Chain,” Working Paper 33934, National Bureau of Economic Research June 2025.
- Akcigit, Ufuk and Nathan Goldschlag, “Where Have All the “Creative Talents” Gone? Employment Dynamics of US Inventors,” Working Paper 31085, National Bureau of Economic Research March 2023.
- Alston, Julian M., Philip G. Pardey, Devin Serfas, and Shanchao Wang, “Slow Magic: Agricultural Versus Industrial R&D Lag Models,” Annual Review of Resource Economics, October 2023,15 (Volume 15, 2023), 471–493. Publisher: Annual Reviews.
- Amiti, Mary and Jozef Konings, “Trade liberalization, intermediate inputs, and productivity: Evidence from indonesia,” American Economic Review, December 2007,97(5), 1611–1638.
- Bailey, Michael A., Anton Strezhnev, and Erik Voeten, “Estimating dynamic state preferences from united nations voting data,” The Journal of Conflict Resolution, 2017,61(2), 430–456. Publisher: Sage Publications, Inc.
- Baquié, Sandra, Yueling Huang, Florence Jaumotte, Jaden Kim, Rafael Machado Parente, and Samuel Pienknagura, “Industrial Policies: Handle with care,” IMF Staff Discussion Notes, March 2025,002, 1. Publisher: International Monetary Fund (IMF).
- Bartelme, Dominick G., Arnaud Costinot, Dave Donaldson, and Andrés Rodríguez-Clare, “The Textbook Case for Industrial Policy: Theory Meets Data,” Working Paper 26193, National Bureau of Economic Research August 2019.
- Barwick, Panle Jia, Hyuk soo Kwon, Shanjun Li, and Nahim Zahur, “Drive Down the Cost: Learning by Doing and Government Policies in the Global EV Battery Industry *,” Working Paper 33378, National Bureau of Economic Research March 2025.
- , Hyuk-Soo Kwon, Shanjun Li, Yucheng Wang, and Nahim B. Zahur, “Industrial Policies and Innovation: Evidence from the Global Automobile Industry,” Working Paper 33138, National Bureau of Economic Research November 2024.
- Blind, Knut, Jakob Edler, Rainer Frietsch, and Ulrich Schmoch, “Motives to patent: Empirical evidence from Germany,” Research Policy, June 2006,35(5), 655–672.
- Bloom, Nicholas, Mirko Draca, and John Van Reenen, “Trade Induced Technical Change? The Impact of Chinese Imports on Innovation, IT and Productivity,” Technical Report w16717, National Bureau of Economic Research January 2011.
- Brandão-Marques, Luis and Hasan H Toprak, “A Bitter Aftertaste: How State Aid Affects Recipient Firms and Their Competitors in Europe,” IMF Working Papers, 2024, (250).
- Branstetter, Lee, Raymond Fisman, and C. Fritz Foley, “Do Stronger Intellectual Property Rights Increase International Technology Transfer? Empirical Evidence from U.S. Firm-Level Data,” August 2005.
- Chang, Ha-Joon, “Industrial policy in East Asia: Lessons for Europe,” EIB Papers, 2006,11(2), 106–132. Publisher: Luxembourg: European Investment Bank (EIB).
- Cherif, Reda and Fuad Hassanov, “The Return of the Policy that Shall Not Be Named: Principles of Industrial Policy,” IMF Working Papers, 2019, (074).
- Choi, Jaedo and Andrei A. Levchenko, “The long-term effects of industrial policy,” Journal of Monetary Economics, June 2025,152, 103779.
- Criscuolo, Chiara, Ralf Martin, Henry Overman, and John van Reenen, “Some causal effects of an industrial policy,” The American Economic Review, 2019,109(1), 48–85.
- Evenett, Simon, Adam Jakubik, Fernando Martín, and Michele Ruta, “The return of industrial policy in data,” IMF Working Papers, 2024, (001).
- Evenett, Simon J. and Johannes Fritz, “The GTA Handbook: Data and methodology used by the Global Trade Alert initiative,” Technical Report, Global Trade Alert 2020.
- Garcia-Macia, Daniel and Alexandre Sollaci, “Industrial Policies for Innovation: A Cost-Benefit Framework,” Working Papers, IMF August 2024.
- Goldberg, Pinelopi, Amit Khandelwal, Nina Pavcnik, and Petia Topalova, “Trade Liberalization and New Imported Inputs,” American Economic Review, May 2009,99(2), 494–500.
- Goldberg, Pinelopi K., Réka Juhász, Nathan J. Lane, Giulia Lo Forte, and Jeff Thurk, “Industrial Policy in the Global Semiconductor Sector,” Working Paper 32651, National Bureau of Economic Research July 2024.
- Goldschlag, Nathan, Travis J. Lybbert, and Nikolas J. Zolas, “Tracking the technological composition of industries with algorithmic patent concordance,” Economics of Innovation and New Technology, 2020,29(6), 582–602.
- Gopinath, Gita, Şebnem Kalemli-Özcan, Loukas Karabarbounis, and Carolina Villegas-Sanchez, “Capital allocation and productivity in south europe*,” The Quarterly Journal of Economics, June 2017,132(4), 1915–1967.
- GTA, “Global Trade Alert Database,” 2022.
- Hall, Bronwyn H., Adam B. Jaffe, and Manuel Trajtenberg, “The NBER Patent Citation Data File: Lessons, Insights and Methodological Tools,” Working Paper 8498, National Bureau of Economic Research October 2001.
- Harrison, Ann, “What Makes Industrial Policy Work?,” CEPR Discussion Paper, November 2024, 19693.
- Hasna, Zeina, Florence Jaumotte, Jaden Kim, Samuel Pienknagura, and Gregor Schwerhoff, “Green Innovation and Diffusion: Policies to Accelerate Them and Expected Impact on Macroeconomic and Firm-Level Performance,” IMF Staff Discussion Notes, November 2023. Publisher: International Monetary Fund (IMF).
- Hodge, Andrew, Roberto Piazza, Fuad Hasanov, Xun Li, Maryam Vaziri, Atticus Weller, and Yu Ching Wong, “Industrial policy in europe: a single market perspective,” 2024.
- Hsieh, Chang-Tai and Peter J. Klenow, “Misallocation and manufacturing TFP in china and india*,” The Quarterly Journal of Economics, November 2009,124(4), 1403–1448.
- Hu, Albert Guangzhou and Gary H. Jefferson, “A great wall of patents: What is behind China’s recent patent explosion?,” Journal of Development Economics, September 2009,90(1), 57–68.
- Huang, Yueling, Sandra Baquié, Florence Jaumotte, Jaden Kim, Yucheng Lu, Rafael Machado Parente, and Samuel Pienknagura, “Do Industrial Policies Increase Trade Competitiveness?,” IMF Working Papers, 2025, (098).
- IMF, “Industrial policy: Managing trade-offs to promote growth and resilience?,” World Economic Outlook: Chapter 3, October 2025, Washington, DC.
- Juhász, Réka and Nathan Lane, “The Political Economy of Industrial Policy,” Journal of Economic Perspectives, November 2024,38(4), 27–54.
- and, “A Short Guide to Thinking About Industrial Policy: Takeaways from the New Economics of Industrial Policy,” SocArXiv, August 2024. Number: 4sra7v1 Publisher: Center for Open Science.
- ,, and Dani Rodrik, “The New Economics of Industrial Policy,” Annual Review of Economics, 2024,16, 213–242.
- ,, Emily Oehlsen, and Verónica C. Pérez, “Measuring Industrial Policy: A Text-Based Approach,” Working Paper, National Bureau of Economic Research June 2025.
- Kowalski, Przemyslaw and Clarisse Legendre, “Raw materials critical for the green transition: Production, international trade and export restriction,” Technical Report 269, OECD, Paris 2023.
- Krueger, Anne O., “Government Failures in Development,” Journal of Economic Perspectives, September 1990,4(3), 9–23.
- , “Trade Policy and Economic Development: How We Learn,” American Economic Review, 1997, 87(1), 1–22. Publisher: American Economic Association.
- LaBelle, Jesse, Immaculada Martinez-Zarzoso, Ana Maria Santacreu, and Yoto Yotov, “Cross-border Patenting, Globalization, and Development,” Working Papers 202507, Center for Global Policy Analysis, LeBow College of Business, Drexel University. June 2024.
- Lane, Nathan, “Manufacturing Revolutions: Industrial Policy and Industrialization in South Korea,” Working Papers 2021-10, Monash University, SoDa Laboratories.
- Liu, Ernest and Song Ma, “Innovation Networks and R&D Allocation,” Working Paper 29607, National Bureau of Economic Research December 2021.
- Machado-Parente, Rafael, Samuel Pienknagura, Sandra Baquié, Yueling Huang, Florence Jaumotte, and Jaden Kim, “Industrial Policies and Firm Performance: A Nuanced Relationship,” Working Papers, IMF 2025.
- Mazzucato, Mariana, The Entrepreneurial State 2011.
- Mealy, Penny and Alexander Teytelboym, “Economic complexity and the green economy,” Research Policy, October 2022,51(8), 103948.
- Melitz, Marc J., “When and how should infant industries be protected?,” Journal of International Economics, May 2005,66(1), 177–196.
- OECD, “An institutional framework for industrial policies,” Policy Papers 180, OECD, Paris June

*Source: wpiea2026047-source-pdf - References*

### 2025.  Edition:  180 Series:  OECD Science, Technology and Industry Policy Papers.

### Shaping Innovation: Can Industrial Policies Boost Patent Applications?

### A. Additional data description
- Cross-sectoral distribution: average number of patent applications (2009−19) displayed across numerous industries, including:
  - "Collection, purification and distribution of water"; "Construction"; "Electricity, gas, steam and hot water supply"; "Extraction of crude petroleum and natural gas"; "Manufacture of basic metals"; "Manufacture of chemicals and chemical products"; "Manufacture of coke, refined petroleum products and nuclear fuel"; "Manufacture of electrical machinery and apparatus n.e.c."; "Manufacture of fabricated metal products, except machinery and equipment"; "Manufacture of food products and beverages"; "Manufacture of furniture; manufacturing n.e.c."; "Manufacture of machinery and equipment n.e.c."; "Manufacture of medical, precision and optical instruments, watches and clocks"; "Manufacture of motor vehicles, trailers and semi−trailers"; "Manufacture of office, accounting and computing machinery"; "Manufacture of other non−metallic mineral products"; "Manufacture of other transport equipment"; "Manufacture of paper and paper products"; "Manufacture of radio, television and communication equipment and apparatus"; "Manufacture of rubber and plastics products"; "Manufacture of textiles"; "Manufacture of tobacco products"; "Manufacture of wearing apparel; dressing and dyeing of fur"; "Manufacture of wood [...]"; "Mining of coal and lignite; extraction of peat"; "Mining of metal ores"; "Mining of uranium and thorium ores"; "Other mining and quarrying"; "Publishing, printing and reproduction of recorded media"; "Recycling"; "Tanning and dressing of leather [...]".
- Cross-country distribution: average yearly number of filings (2009−19) shown for a wide country list (AGO, AIA, ALB, ARE, ARG, ..., ZWE) with both "Received" and "Submitted" series.
- Source datasets and references explicitly cited in the figures: LaBelle et al. (2024); GTA (2022); Juhász et al. (2025); authors’ calculations.

### B. Main empirical approach and notation
- Estimation method for event responses: percentage changes are estimated following Equation 1 with 100×(exp(β_h)−1) for most intensive-margin outcomes; for some extensive-margin outcomes the formula uses 100×β_h.
- Time convention: "Year 0 is the year of IP implementation."
- Fixed effects and clustering: regressions include sector-year, country-year and country-sector fixed effects; standard errors are clustered at the country level (with some robustness checks clustering at the country-sector level).
- Instrumental variables and first stage diagnostics:
  - First stage instruments include ∆ProIPStock (change in protectionist IP stock in other countries and sectors), ∆LibIPStock (change in liberalizing IP stock in other countries and sectors), "Pol Dist" (Political-distance weights), and "Trade Part" (weights by a dummy if the considered country is one of the 90th percentile largest trade partners).
  - First-stage strength reported via KP rk Wald F-statistic histograms; black line delineates F=10 as a reference.

### C. Additional results (summary of figures)
- Protectionist industrial policies and patenting:
  - Figures present the dynamic percentage change (horizons in years −3 to 4, and beyond) in patent applications submitted abroad by domestic inventors when protectionist IPs are implemented. Results are shown with 90 percent confidence intervals.
  - Separate panels analyze protectionist IPs effects on patents patented by foreign inventors and by domestic inventors for Advanced Economies (AEs) and Emerging Market and Developing Economies (EMDEs).
- Liberalizing industrial policies and patenting:
  - Figures present analogous dynamic effects for liberalizing IPs on patents patented by foreign and domestic inventors, including AE and EMDE samples.
- Climate-related versus non-climate-related protectionist IPs:
  - Effects on patent applications submitted abroad by domestic inventors are shown separately for "Green" and "Non−green" (climate-related and non-climate-related) IPs.
  - Extensive-margin (probability that at least one patent application is submitted) and intensive-margin (changes in patent applications for already-patented sectors) are reported separately for domestic and foreign inventors.
- Heterogeneity by sector centrality:
  - Effects of protectionist and liberalizing IPs on patent applications received from foreign inventors are stratified by whether targeted industries are "innovation central" or "non-central".
  - Industries are defined as innovation central if their PageRank centrality (or eigenvalue centrality) in their country’s innovation network is above the 80th percentile of the country’s innovation centrality distribution.
- OLS evidence:
  - Protectionist (purple) and liberalizing (blue) IPs and patent applications from foreign and domestic inventors are also estimated with OLS; figures display correlations with country-level clustering and 90 percent confidence intervals.

### D. Robustness checks
- Hansen J-test: p-values reported for the Hansen J-test in regressions of received patent applications on liberalizing IPs (panels present horizons −3 to 4 with the black line delineating p=0.05).
- Sample exclusion: Main results recalculated excluding the USA, China, and Germany (figures for protectionist and liberalizing policies).
- Clustering alternative: Results when clustering standard errors at the country-sector level.
- Policy universe: Robustness to using all trade policies rather than a limited subset.
- Policy-tool granularity:
  - Separate estimates for protectionist "Subsidies" and protectionist "Export incentives" and their effects on patent applications received from foreign inventors; results presented with country-sector clustering.
  - Additional IV specification where subsidies and export incentives are instrumented by the same policy tool in other countries and sectors.
- Centrality interaction: Protectionist and liberalizing IP effects shown by centrality of the targeted sector (both PageRank and eigenvalue centrality measures).

### E. Key methodological and interpretive notes appearing in figure captions
- Confidence intervals: dashed lines represent 90 percent confidence intervals.
- First-stage diagnostic display: histograms in each figure show the first stage KP rk Wald F-statistic for the considered horizon, with F=10 threshold shown.
- Controls: regressions control for non-protectionist IPs and non-IP trade policies as well as sector-year, country-year and country-sector fixed effects in many specifications.
- Sample definitions: several figures specify that the sample is limited to AEs or EMDEs when applicable.

*Source: Shaping Innovation: Can Industrial Policies Boost Patent Applications? Working Paper No. WP/2026/047 — figures, notes, and ancillary text as provided in the source PDF.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026047-source-pdf.pdf_