Innovation for Net Zero: The Technological Footprint of Corporate Climate Pledges

François Perruchas¹, Lauri Peterson², Francisco Javier Ortega-Colomer¹

¹INN4ALL, University of València & INGENIO [CSIC - Universitat Politècnica de València]
²CCEEL, Law School, University of Eastern Finland, Finland & Department of Government, Uppsala University, Sweden

Introduction

The 2015 Paris Agreement set the goal of limiting global warming to +2°C above pre-industrial levels.
The costs of inaction are high: pollution, biodiversity loss, and threats to human health and food security (Leahy, 2018)
Climate change mitigation requires accelerating the development and diffusion of low-carbon technologies (Stern, 2007) even if the solution can not be only technological (Sarewitz and Nelson, 2008)
Net-zero strategies involve all actors — states, companies, financial institutions, civil society — in an ordered pathway toward zero greenhouse gas emissions

Motivation

Global Corporate Companies represents a significant part of GhG emissions:
- half of world’s CO2 emissions come from only 36 fossil fuel firms (Carbon Major, 2023)
- the direct activities and supply chains of 157 large multinational firms jointly account for up to 60 percent of global industrial emissions (World Bank, 2023)
Role of Green technologies
- Central to strategies against climate change.
- More complex and novel than conventional technologies (Barbieri et al, 2022)
  - require targeted incentives for development.
  - specific skills to develop and implement them.

Research gap & questions

While public actors’ strategies are increasingly studied, corporate net-zero pledges remain underexamined (Hale et al, 2021)
Many corporate pledges are self-regulated, lack standardization, and may not align with the scale or urgency of required emission reductions

Research Questions:

What kind of commitments have global companies pledged to reach net zero emissions?
What types of technologies do global companies rely on to implement their net zero strategies?

Data

Corporate Climate Pledges → Net Zero Tracker (2025 snapshot, lastest)
Corporate intellectual property portfolio → Orbis + HAN OECD (v2025.2)
To identify the technologies they develop → PATSTAT 2025

Corporate Climate Pledges

Net Zero Tracker (2025 snapshot): information on targets for net zero emissions pledged by countries, states/regions/provinces (hereafter ‘regions’ for short), cities, and companies.
All of the 2,000 largest publicly-listed companies, as detailed in the Forbes Global 2000 list
Source: entities’ websites or published documentation, press releases, or news articles
Target considered: Net zero, Zero emissions, Zero carbon, Climate neutral, Climate positive, Carbon neutral(ity), GHG neutral(ity), Carbon negative, Net negative.

Technologies (green and non-green)

Identification of global corporate companies and subsidies in Orbis (companies owned with at least 50% of shares)
Patent applications of these companies (and subsidies) using HAN OECD database
Patent families in PATSTAT with CPC classes.

Depending on the technological classification:

green technologies → Y02 branches
- Y02A: technologies for adaptation to climate change
- other Y02: climate change mitigation technologies
non-green technologies → other CPC codes

Cluster analysis of corporate climate pledges

Net Zero Tracker contains 2074 companies, 66 variables about target status, target year, intermediate target, scope, etc.

First approach - focus on 4 dimensions:

Type of end target (e.g. Net zero, climate neutral, Emissions reduction…),
End target year: when they pledge it will be achieve
Interim target: existence and type of milestones
Status of the end target (e.g. proposed / in discussion, in corporate stategy, achieved…)

Cluster analysis

Hierarchical clustering

binary matrix with a column for each value of the 4 variables
euclidean distance between rows of the matrix
hierarchical clustering with complete linkage method
Define 5 clusters

Cluster validation with PCA

Principal Component Analysis to check how clusters are distributed along PCs
first 3 PCs capture 60% of data information

3D PCA plot

Dimensions associated to clusters

Mean values per cluster

variable	Cluster_1	Cluster_2	Cluster_3	Cluster_4	Cluster_5
Interim_target_Other	0.0935484	0.0111524	0.0056497	0.8095238	0.0769231
Interim_target_Reduce emissions	0.6556452	0.6282528	0.0000000	0.0000000	0.5384615
End_target_No target	0.0032258	0.0000000	0.9604520	0.0000000	0.0000000
End_target_Other	0.0225806	0.0037175	0.0131827	0.2857143	0.0000000
End_target_Zero/net/negative/emissions	0.7854839	0.8364312	0.0000000	0.6666667	0.6153846
End_target_year_2040-2059	0.6629032	0.6579926	0.0056497	0.4761905	0.0769231
End_target_year_2060+	0.0145161	0.0185874	0.0037665	0.1904762	0.0000000
End_target_year_Missing	0.0112903	0.0260223	0.9435028	0.0000000	0.3846154
Status_of_end_target_Achieved (self-declared)	0.0000000	0.0892193	0.0000000	0.0000000	0.0000000
Status_of_end_target_Declaration / pledge	0.0000000	0.8327138	0.0075330	1.0000000	0.0000000
Status_of_end_target_In corporate strategy	1.0000000	0.0000000	0.0131827	0.0000000	0.0000000
Status_of_end_target_Missing	0.0000000	0.0148699	0.9792844	0.0000000	0.0000000
Status_of_end_target_Proposed / in discussion	0.0000000	0.0483271	0.0000000	0.0000000	1.0000000

N. of firms per cluster

Cluster_1	Cluster_2	Cluster_3	Cluster_4	Cluster_5
1240	269	531	21	13

Cluster properties

cluster 1: pledge in corporate strategy, with ambitious end target and milestones
cluster 2: pledge is a declaration, but ambitious end target and milestones
cluster 3: no end target (neither objectives or year), no interim target
cluster 4: pledge is a declaration, other type of end target, no milestones
cluster 5: pledge is in discussion, interim target but no end target year

Correlations with technologies

We estimate a Multinomial logit model the correlations between belonging to a cluster and the characteristics of the Corporate IP Portfolio (green / non-green, adaptation/mitigation).

We control for:

Company annual revenues
Employees
Industry
Country

Preliminary results


	Dependent variable:

	Clusters (Ref: 3)	2	4	5
	1	2	4	5
	(1)	(2)	(3)	(4)

Green patents share	0.020^***	-0.007^***	-0.003^***	-0.001^***
	(0.000)	(0.000)	(0.000)	(0.000)
N. patents adaptation	-0.010^***	-0.029^***	-0.021^***	-0.013^***
	(0.000)	(0.000)	(0.000)	(0.000)
N. patents mitigation	0.017^***	0.017^***	0.011^***	0.005^***
	(0.000)	(0.000)	(0.000)	(0.000)
Log(Revenues)	0.000	-0.000	-0.000	-0.000^***
	(0.000)	(0.000)	(0.000)	(0.000)
Log(Employees)	0.00000^***	-0.00000^***	-0.0001^***	-0.0001^***
	(0.000)	(0.000)	(0.000)	(0.000)
N. patents	0.0004^***	0.0004^***	-0.0001^***	0.0004^***
	(0.000)	(0.000)	(0.000)	(0.000)
Constant	0.558^***	-0.259^***	-0.205^***	-0.092^***
	(0.000)	(0.000)	(0.000)	(0.000)

FE Industry	Yes	Yes	Yes	Yes
FE Country	Yes	Yes	Yes	Yes

Akaike Inf. Crit.	4,522.415	4,522.415	4,522.415	4,522.415

Note:	p<0.05; p<0.01; p<0.001

TRUE

Cluster 3 (low ambition climate pledges) is the reference

Conclusions

The biggest cluster is the most ambitious one in term of climate pledge
Different types of pledges are associated with different IP portfolios
Positive and significant correlation between mitigation technologies and ambitious climate pledges
Green patent share is only positive and significant when climate pledge is in corporate strategy, negative otherwise
Adaptation technologies are negatively associated with climate pledge

Limitations

clustering is very sensitive to parameters → requires more work on cluster definition
limits inherent to patent, alternative methods (stock, treatment effect, green R&D)
no data available on emissions yet
scopes were not included in this work

Future avenues / Questions / Comments

Any questions ? Comments ?

Any ideas about:

including public policies in the analysis
the geographical dimension of operations of global corporate companies
the characteristics of the technologies and the sector where they operate
…

Thank you!

François Perruchas - francois.perruchas@uv.es