Harnessing ESG Sustainability Uncertainty, Financial Development and Information Technology for Energy Transition

Abstract

By unraveling the electrifying nexus between ESG sustainability uncertainty, financial development, information technology, trade policy uncertainty, and economic growth, this study sheds light on how these forces collectively shape the trajectory of the United States’ energy transition. Utilizing quarterly data from 2002 Q1 to 2024 Q4, we employ the novel Quantile-on-Quantile ARDL (QQARDL) framework to capture the heterogeneous and distribution-dependent dynamics of these relationships. To the best of our knowledge, this is the first study to apply QQARDL in assessing the simultaneous effects of institutional uncertainty, financial and technological drivers, and macroeconomic growth on energy transition outcomes in the U.S. The QQARDL results confirm that ET is cointegrated with ESG uncertainty, ICT, FD, TPU, and economic growth, though the strength and direction of these relationships vary across quantiles. ICT and EG consistently promote ET, ESG, and TPU exert mixed effects, FD is generally constraining, and the negative, significant ECT confirms stable long-run convergence with faster adjustment at higher ET quantiles. Based on these findings, policies were formulated to reduce ESG uncertainty, align financial development with green priorities, expand ICT adoption, stabilize trade frameworks, and harness economic growth to accelerate the U.S. energy transition.

1. Introduction

In the United States, the accelerating impacts of climate change—including rising temperatures, intensifying extreme weather events, and shifting precipitation patterns—have reinforced the urgency of advancing a robust energy transition. The IPCC’s Special Report on Global Warming of 1.5 °C warns that breaching this threshold could trigger irreversible tipping points unless global emissions peak by 2025 and decline sharply by 2030 [1]. Against this backdrop, the 2015 COP 21 Paris Agreement represented a turning point, committing the U.S. and other signatories to limit warming to well below 2 °C and pursue efforts toward 1.5 °C through progressively ambitious Nationally Determined Contributions [2]. Although the U.S. temporarily withdrew in 2017, its re-engagement in 2021 reinvigorated domestic transition strategies, ranging from clean energy investment tax credits to new emissions regulations. Yet, the pathway to decarbonization is far from linear: ESG sustainability uncertainty, financial development, ICT adoption, and trade policy volatility all shape the pace and resilience of the U.S. transition. As the world’s largest economy and second-largest emitter, U.S. leadership is pivotal not only in meeting domestic transition goals but also in setting global standards for financing, innovation, and equitable burden-sharing in the pursuit of a sustainable low-carbon future.

Information technology (IT) has emerged as a key enabler of energy transition by optimizing energy production, distribution, and consumption across multiple sectors. Smart grids, advanced sensors, and data-driven algorithms allow utilities to forecast demand more accurately, integrate intermittent renewable sources, and minimize system losses, thereby accelerating the shift away from fossil fuels [3,4]. At the firm level, cloud-based platforms and automation tools improve operational efficiency and facilitate the adoption of low-carbon business models, while digital collaboration systems have permanently reshaped mobility patterns by reducing commuting and business travel [5]. Nevertheless, ICT expansion carries its own environmental costs—most notably rising electricity demand from data centers and embodied emissions from hardware production—highlighting the importance of green computing practices to ensure ICT contributes positively to long-run transition goals [6,7].

ESG sustainability uncertainty constitutes another critical factor influencing ET. When investors face ambiguous or shifting ESG standards, they may hesitate to channel funds into renewable energy projects, slowing down capital mobilization for clean technologies. Uncertainty over disclosure requirements or sustainability benchmarks can undermine investor confidence and delay the integration of ESG principles into financing decisions. However, at higher stages of ET, ESG commitments can also catalyze adaptation and long-term strategic planning, enabling firms to align more closely with green transition pathways [8,9]. Financial development (FD) plays a dual role in ET, acting as both an enabler and a constraint. On the one hand, deep and inclusive financial systems facilitate access to capital for renewable investments, green bonds, and sustainable infrastructure, thereby supporting ET [10,11]. On the other hand, when misaligned, FD may reinforce carbon-intensive growth by directing financing to fossil fuel projects, especially at lower quantiles of ET [12]. Thus, the impact of FD is highly context-dependent, underscoring the importance of aligning financial systems with sustainability mandates to ensure that financial deepening translates into effective transition outcomes.

Trade policy uncertainty (TPU) further complicates the ET process. Unpredictable tariffs, export restrictions, or quota adjustments on renewable energy components disrupt global supply chains, reduce economies of scale, and delay the diffusion of clean technologies [13,14]. Similarly, unstable trade frameworks create cost unpredictability for firms, discouraging investments in green equipment and sometimes prolonging reliance on legacy fossil-based systems. Finally, economic growth (EG) remains a foundational driver: while rising income levels expand resources for renewable investment and innovation, unmanaged growth risks escalating energy demand, which may offset decarbonization gains if not accompanied by structural shifts toward renewables [15,16]. Collectively, these institutional, financial, and technological forces define the pace and direction of the U.S. energy transition

The objective of this study is to examine the determinants of energy transition (ET) in the United States, with particular emphasis on institutional and technological drivers. Specifically, the analysis seeks to (i) evaluate the impact of ESG sustainability uncertainty on the pace and direction of ET, (ii) assess the role of financial development in mobilizing or constraining resources for energy transition, (iii) investigate how information technology adoption supports decarbonization through energy transition, (iv) analyze the effect of trade policy uncertainty on energy transition and (v) examining the effect of economic growth on energy transition. By integrating these dimensions, the study provides a comprehensive understanding of how uncertainty, finance, and digitalization interact to shape the trajectory of the U.S. energy transition.

This study contributes to the existing literature in several ways. First, while a growing body of research has examined the drivers of the energy transition, very few have isolated the specific role of IT adoption in shaping ET outcomes. Prior studies typically bundle IT under broader digitalization measures, which obscures the distinct mechanisms through which IT-driven efficiency gains, smart energy systems, and digital innovation accelerate renewable integration. From an energy-finance perspective, this omission is critical, as the design of digital investment incentives and green financing policies requires a precise quantification of IT’s contribution to decarbonization. By explicitly modeling ICT adoption against ET trajectories, this study fills this empirical void and provides policymakers and market actors with clear evidence of IT’s transition-enhancing potential.

Second, by explicitly incorporating ESG sustainability uncertainty and TPU into the ET framework, this study illuminates how ambiguous institutional and policy environments reshape transition dynamics. Heightened ESG uncertainty can increase the cost of capital for green technologies and delay firm-level adoption of renewable strategies, while unpredictable trade barriers disrupt clean energy supply chains and hinder technology diffusion. These effects are particularly salient in early stages of ET, when capital and technology flows are most fragile. By quantifying these uncertainty channels, the study demonstrates how policy ambiguity can slow or, under certain quantiles, catalyze adaptive transition responses, offering valuable insights for designing resilient climate and trade governance.

Third, the analysis incorporates the role of FD as both a potential constraint and enabler of ET. The existing literature [10,12,17] highlights that FD may initially finance carbon-intensive projects but, once aligned with sustainability goals, can channel resources effectively into renewable investments. By integrating FD into a quantile-based framework, this study shows how the impact of finance varies across the distribution of ET, being negative in low quantiles but supportive in advanced transition states. This heterogeneity underscores the importance of policy alignment to ensure that financial deepening reinforces, rather than undermines, energy transition pathways.

Finally, this study adopts the cutting-edge Quantile-on-Quantile ARDL (QQARDL) methodology of [8], which uniquely fuses the distributional richness of quantile regression with the long-run equilibrium structure of ARDL models. Unlike traditional mean-based or single-quantile approaches, QQARDL traces how shocks at specific quantiles of ESG uncertainty, IT adoption, FD, TPU, and economic growth propagate through different segments of ET’s distribution. This framework exposes nonlinearities, heterogeneous adjustment speeds, and persistence across market states. By doing so, the study not only addresses an important methodological gap in energy transition research but also equips policymakers with nuanced, quantile-specific insights into when and how interventions—whether through ICT investment, ESG regulation, trade policy, or financial reforms—are most effective in accelerating the transition.

The subsequent sections are as follows: Section 2 presents the summary of past studies, Section 3 presents data and method, Section 4 presents findings, and Section 5 presents the conclusion and policy recommendations.

2. Theoretical Context and Literature Review

2.1. Theoretical Context

The energy transition (ET) from fossil fuels to renewables is shaped by institutional and market uncertainties, with Environmental, Social, and Governance (ESG) sustainability uncertainty emerging as a critical factor. ESG uncertainty reflects ambiguity in sustainability disclosures, climate commitments, and governance practices, which can create barriers to green investment and slow the pace of decarbonization [18]. However, theory also suggests that under conditions of higher ESG scrutiny, firms and governments may accelerate long-term transition pathways to build legitimacy and reduce reputational risks [19]. This dual effect aligns with institutional and signaling theories, where transparent ESG practices provide confidence to investors and stakeholders, while inconsistent or uncertain signals exacerbate risks and delay transition. Financial development (FD) plays a pivotal role in mobilizing resources for clean energy and green innovation. According to finance-growth theory, well-developed financial systems can enhance ET by allocating capital to renewable energy projects, reducing financing constraints, and supporting green bonds and sustainability-linked loans [20]. Conversely, if financial development remains oriented toward fossil-fuel investments, it may hinder ET by locking in carbon-intensive infrastructure. Thus, the impact of FD on ET is theorized to be conditional on the alignment of financial flows with sustainable policies, highlighting the importance of green taxonomies and climate-risk disclosure frameworks that can rechannel resources towards clean energy [21].

Information technology (IT) contributes to ET by fostering digital innovation, improving energy efficiency, and enabling integration of renewables through smart grids and digital monitoring systems. The diffusion of innovation theory posits that ICT adoption accelerates structural change by reducing transaction costs, improving energy management, and stimulating eco-innovation [4]. Moreover, digital platforms facilitate the scaling of decentralized renewable systems, such as solar and wind microgrids, which are essential for accelerating the transition. By providing real-time data analytics, artificial intelligence, and Internet of Things (IoT) applications, IT enhances system resilience and efficiency, making ET more technologically feasible and economically attractive [22]. Trade policy uncertainty (TPU) introduces an external dimension to ET by influencing investment decisions and energy security. From the perspective of real options theory, heightened TPU increases investment hesitation in renewable projects, as firms adopt a “wait-and-see” approach in uncertain policy environments [23]. However, TPU can also stimulate adaptive strategies, where firms diversify energy portfolios and innovate to hedge against unpredictable trade conditions. At advanced levels of ET, TPU may drive resilience and global competitiveness, encouraging firms to reconfigure supply chains and adopt cleaner technologies. Thus, TPU is theorized to have nonlinear effects, initially constraining ET but fostering long-term resilience once economies adapt to uncertainty shocks [24].

2.2. Empirical Review

The literature on ESG sustainability uncertainty and ET presents mixed findings, reflecting both constraints and opportunities for transition. Refs. [19,25], using quantile-based methods in the U.S., show that ESG uncertainty exerts heterogeneous effects, sometimes hindering renewable adoption while in other contexts supporting adaptation strategies. Similarly, Ref. [9] find in China and global samples that ESG uncertainty both impedes and accelerates ET depending on institutional conditions, while Ref. [26] provide firm-level evidence that ESG uncertainty tends to weaken ESG performance, thereby slowing transition. These mixed outcomes align with institutional and signaling theories: uncertainty can deter investment when credibility is low, but may incentivize stronger long-term commitments when ESG integration is more advanced.

In contrast, financial development is generally identified as a positive enabler of ET, though with notable nonlinearities. Ref. [10] demonstrates through FMOLS and DOLS that FD supports renewable energy in Ghana, while Ref. [11] confirm similar global patterns. Doğan et al. (2025) Ref [27] further show FD’s positive contribution across RECAI countries. However, Refs. [12,17] highlight nonlinear and sometimes negative dynamics: at low levels of FD, financing may continue to favor fossil fuels, whereas beyond a threshold, FD channels capital into renewables. These results support the financial-growth literature, suggesting that the quality and alignment of financial systems with green policies determine whether FD facilitates or constrains ET.

The impact of trade policy uncertainty (TPU) is more consistently negative, though with some evidence of adaptive resilience. Refs. [14,28,29] all document that TPU reduces renewable investment and consumption in China and the United States, reinforcing the real-options perspective that firms delay long-term green investment under uncertainty. However, Gyamfi et al. (2025) ([30], p. 13) show in a U.S. multi-frequency quantile framework that TPU’s effects vary: while negative in the short term, under certain quantiles, it can push firms to diversify and innovate, thus supporting ET in the longer run. This suggests that TPU initially hampers transition but can, under adaptive conditions, stimulate resilience and technological change.

Finally, the literature on ICT and ET is broadly supportive of ICT as a driver of clean energy transformation. Ref. [31] find across the MINT economies that ICT adoption reduces emissions and fosters ET at all quantiles, while Ref. [32] highlight ICT’s mediating role in energy efficiency globally. Refs. [3,33] confirm strong long-run ICT–ET linkages in Bangladesh and China, respectively, using QARDL methods, while Ref. [34] document ICT’s heterogeneous but positive effects across OECD nations. Together, these studies reinforce the diffusion-of-innovation perspective, positioning ICT as a catalytic enabler of energy efficiency, renewable integration, and system resilience.

Table 1. Summary of past studies.

Author(s)	Period	Nation(s)	Method(s)	Result(s)
ESG Sustainability Uncertainty (ESG) and Energy Transition (ET)
[19]	2002–2023	USA	Quantile Analysis	ESG ↑↓ ET
[9]	2001–2023	China	Panel Analysis	ESG ↑↓ ET
[35]	2002–2024	Global	QQR	ESG ↓ ET
[26]	2011–2022	China (A-share firms)	Fixed Effects Panel Model	ESG ↓ ET
[25]	2002–2024	USA	Kernel Regularized Quantile Regression	ESG ↑↓ ET
[36]	2000–2024	USA	ARDL	ESG ↑↓ ET
Financial Development and Energy Transition (ET)
[10]	1990–2019	Ghana	FMOLS & DOLS	FD ↑ ET
[11]	2000–2021	Global	Panel regression	FD ↑ ET
[37]	1991–2021	RECAI countries	Panel regression	FD ↑ ET
[12]	2000–2022	OECD countries	Nonlinear panel data methods	FD ↑↓ ET
[17]	2000–2021	N-11 (Next-11) emerging nations	Regression models	FD ↑↓ ET
Trade Policy Uncertainty (TPU) and Energy Transition (ET)
[14]	2000–2023	China	Panel Asymmetric Effects	TPU ↓ ET
[28]	2000–2021	United States	Panel regression	TPU ↓ ET
[29]	1995–2021	United States	Nonlinear analysis	TPU ↓ ET
[38]	1987–2024	United States	Multi-frequency quantile regression	TPU ↓↑ ET
Information Communication Technology (IT) and Energy Transition (ET)
[31]	1990–2018	MINT	Quantile Regression	IT ↑ ET
[32]	2000–2020	Global sample	Panel regression	IT ↑ ET
[3]	1991–2023	Bangladesh	QARDL	IT ↑ ET
[34]	2000–2020	OECD countries	Panel quantile regression	IT ↑ ET
[33]	1995–2020	China	QARDL	IT ↑ ET

Note: ↑ denotes increase and ↓ denote decrease.

presents a summary of past studies.

3. Data and Method

3.1. Data

This study investigates how information technology (IT), financial development (FD), economic growth (EG), and trade policy uncertainty (TPU) affect the energy transition (ET) using quarterly data from 2002 to 2024. ET is the dependent variable, while IT, EG, CPU, and TPU are examined as key drivers influencing ET. We obtain the quarterly measured IT from [39], ET is obtained from [1], TPU from [40] and EG from EG and FD from [41]. All the data are in quarterly form. The log trend of the variables is plotted in Figure 1.

3.2. Empirical Method

We apply the Quantile-on-Quantile ARDL (QQARDL) framework of [8], which extends both the ARDL and QARDL models by simultaneously considering quantile dynamics in the dependent and independent variables. Unlike ARDL, which focuses on conditional means, and QARDL, which only incorporates quantiles of the outcome variable, QQARDL jointly models the full conditional distribution of both sides of the relationship. This feature allows the framework to uncover hidden nonlinearities and heterogeneity often overlooked by traditional approaches. To ensure comparability of short- and long-run coefficients, we follow [42] in fixing lag orders at one, and we employ [43] quantile series (QSER) technique to construct quantile-specific data before fitting ARDL(1,1) models. For inference, we implement QQARDL bounds tests, estimating long-run and short-run coefficients and error-correction terms in a two-dimensional quantile matrix that illustrates the heterogeneity of effects across τ-quantiles of the dependent variable and v-quantiles of the regressors:

$$\begin{array}{c}\mathrm{\Delta }\mathrm{l}\mathrm{n}{Y}_t^{\tau }={\varphi }_0(\tau ,\theta )+{\omega }_1(\tau ,\theta )\mathrm{\Delta }\mathrm{l}\mathrm{n}{Y}_{t-1}^{\tau }+{\omega }_2(\tau ,\theta )\mathrm{\Delta }\mathrm{l}\mathrm{n}{X}_{t-1}^{\theta }\\ +{\gamma }_1(\tau ,\theta )\mathrm{l}\mathrm{n}{Y}_{t-1}^{\tau }+{\gamma }_2(\tau ,\theta )\mathrm{l}\mathrm{n}{X}_{t-1}^{\theta }+e_t(\tau ,\theta )\end{array}$$

(1)

Here $\mathsf{\Delta }\mathrm{l}\mathrm{n}{Y}_t^{\left(\tau \right)}$ denotes the change in $\tau \text{-}\mathrm{quantile}\mathrm{of}\mathrm{l}\mathrm{n}Y\mathrm{at}\mathrm{time}t$; ${\varphi }_0(\tau ,\theta )$ is a quantile-specific intercept. The coefficients of the short-run are depicted by ${\omega }_1(\tau ,\theta )\mathrm{and}{\omega }_2(\tau ,\theta )$; and the long-run coefficients are ${\gamma }_1(\tau ,\theta )\mathrm{and}{\gamma }_2(\tau ,\theta )$. For each $\left(\tau ,\theta \right)$ pair, the QQARDL bounds test calculates an FFF-statistic to test for cointegration under:

$$H_0:{\gamma }_1(\tau ,\theta )={\gamma }_2(\tau ,\theta )=0(\mathrm{no} \text{long-run} \mathrm{link})$$

(2)

$$H_1:{\gamma }_1(\tau ,\theta )\ne 0\vee {\gamma }_2(\tau ,\theta )\ne 0(\text{long-run} \mathrm{relationship}\mathrm{exists})$$

(3)

Following the cointegration analysis, we applied the error correction form of the QQARDL model to investigate the short-run and long-run effects of the quantiles of X on the quantiles of Y, as shown below:

$$\begin{array}{c}\mathrm{\Delta }\mathrm{l}\mathrm{n}{Y}_t^{\tau }=a_0(\tau ,\theta )+{\partial }_0(\tau ,\theta )\mathrm{l}\mathrm{n}{Y}_{t-1}^{\tau }+{\delta }_1(\tau ,\theta )\mathrm{\Delta }\mathrm{l}\mathrm{n}{X}_t^{\theta }\\ +{\beta }_1(\tau ,\theta )\mathrm{l}\mathrm{n}{X}_{t-1}^{\theta }+{ϵ}_t(\tau ,\theta )\end{array}$$

(4)

Here, ∂₀(τ, θ) denotes the error-correction term (ECT), and δ₁(τ, θ) and β₁(τ, θ) are the short-run and long-run coefficients, respectively. Importantly, long-run effects of X’s quantiles on Y’s quantiles are computed only for those (τ, θ) pairs that exhibit cointegration, whereas short-run effects are estimated for all quantile combinations as specified above.

$$\begin{array}{ccccc}(0.95,0.05)& (0.95,0.1)& (0.95,0.2)& \dots & (0.95,0.95)\\ (0.9,0.05)& (0.9,0.1)& (0.9,0.2)& \dots & (0.9,0.95)\\ (0.8,0.05)& (0.8,0.1)& (0.8,0.2)& \dots & (0.8,0.95)\\ ⋮& ⋮& ⋮& \ddots & ⋮\\ (0.1,0.05)& (0.1,0.1)& (0.1,0.2)& \cdots & (0.1,0.95)\\ (0.05,0.05)& (0.05,0.1)& (0.05,0.2)& \cdots & (0.05,0.95)\end{array}$$

(5)

Here, the first set of values refers to X’s quantiles, whereas the second set corresponds to Y’s quantiles

4. Findings and Discussion

4.1. Descriptive Statistics and Correlation

Table 2. Descriptive statistics.

	EG	IT	ET	TPU	ESG	FD
Minimum	10.8250	5.3310	5.7470	2.9310	2.3530	3.8230
Maximum	11.1420	8.4140	6.6110	7.0520	4.6180	4.1010
Mean	10.9660	6.5500	6.2280	4.2190	3.2650	3.9550
Median	10.9380	6.2370	6.3130	4.0600	3.2280	3.9470
Stdev	0.0850	0.8630	0.2530	0.9860	0.3980	0.0580
Skewness	0.4620	0.6120	−0.4980	1.0960	0.4570	0.7770
Kurtosis	2.1910	2.0960	1.8860	3.6950	3.5790	3.3930
Jarque-Bera	5.7250	8.7760	8.4670	20.0370	4.4310	9.7330
Probability	0.0570 *	0.0120 **	0.0150 **	0.0000 ***	0.1090	0.0080 *

Note: *** p < 1%, ** p < 5% and * p < 10%.

presents the descriptive statistics for the variables EG, IT, ET, TPU, ESG, and FD. The results show that all variables exhibit relatively small dispersion around their means, as reflected in their low standard deviations, with FD showing the least variability (0.058) and TPU the highest (0.986). The skewness values indicate that EG, IT, TPU, ESG, and FD are positively skewed, suggesting a longer right tail, while ET is negatively skewed, implying a slight leftward distribution. Kurtosis values are generally close to three, pointing to distributions that are near normal, although TPU and ESG show slightly higher peakedness. The Jarque–Bera test statistics reveal that IT, ET, TPU, and FD significantly deviate from normality, as indicated by their probabilities being below the 5% or 10% significance levels, while EG and ESG fail to reject the null of normality. Overall, the descriptive measures suggest moderate variation, mild asymmetry, and mixed evidence of normality across the variables, which provides useful insights into their distributional properties before further econometric analysis.

Figure 2 presents the correlation result. With respect to ET, the correlation matrix shows that it is very strongly and positively associated with EG (0.87) and IT (0.86), suggesting that economic growth and investment trends move closely in line with energy transition. The correlation with TPU (0.42) is moderate, while its relationship with ESG (0.08) is weak and almost negligible, indicating that ESG scores have little direct short-run alignment with ET. Interestingly, ET is negatively correlated with FD (−0.52), implying that financial development may not always support energy transition in this dataset and could even constrain it. Overall, ET exhibits its strongest linkages with growth-related factors (EG, IT) while maintaining weaker or even adverse associations with ESG orientation and financial development.

4.2. Nonlinearity and Normality Test Result

Table 3. BDS test result.

	ET	IT	ESG	FD	TPU	EG
M2	33.286 ***	30.797 ***	7.6401 ***	17.034 ***	10.174 ***	31.947 ***
M3	35.568 ***	32.234 ***	6.6638 ***	17.397 ***	10.676 ***	33.517 ***
M4	38.269 ***	34.309 ***	6.7724 ***	18.150 ***	11.552 ***	35.621 ***
M5	42.577 ***	37.611 ***	6.8657 ***	19.689 ***	12.372 ***	39.053 ***
M6	48.341 ***	42.369 ***	7.0889 ***	21.746 ***	13.209 ***	43.951 ***

Note: *** p < 0.01.

reports the results of the BDS test for the variables ET, IT, ESG, FD, TPU, and EG across embedding dimensions M2 to M6. The null hypothesis of the BDS test is that the series is independent and identically distributed (i.i.d.), implying no evidence of nonlinearity or dependence in the data. The consistently large and highly significant statistics (all at the 1% level) across all variables and dimensions reject the null hypothesis. This indicates strong evidence of nonlinearity and dependence structures in each of the series, suggesting that linear models alone may be inadequate to capture their underlying dynamics.

The Q-Q plots (see Figure 3) for ET, EG, IT, TPU, ESG, and FD compare the distribution of each variable against the theoretical normal distribution. For EG and ESG, the points align relatively closely with the 45-degree line, suggesting approximate normality. However, ET, IT, TPU, and FD deviate considerably, especially in the tails, indicating skewness and heavy-tailed behavior. In particular, TPU and ET show strong departures from the line at both lower and upper quantiles, while FD exhibits clustering and curvature away from normality.

4.3. Unit Root Test Result

Figure 4a,b presents the quantile-based ADF and PP stationarity test results for EG, ET, IT, TPU, ESG, and FD. Across quantiles, the shaded areas represent the 1%, 5%, and 10% critical thresholds, with values below these lines rejecting the null of a unit root. The results indicate strong evidence of stationarity for TPU and ESG, as their test statistics remain well below the thresholds across most quantiles. By contrast, EG, ET, and IT approach or cross the critical boundaries at higher quantiles, suggesting weaker or partial stationarity and persistence in the upper tails. FD remains closer to the 10% threshold, reflecting mixed evidence of stationarity. Overall, we observed heterogeneous stationarity behavior across quantiles, confirming the importance of quantile-based tests to capture nonlinear distributional properties that conventional mean-based unit root tests may overlook.

4.4. QQARDL Analysis Results

Figure 5a–e displays the results of the QQARDL bounds test between ET and its key determinants (ESG, IT, FD, TPU, and EG) across different quantiles. The color intensity reflects the magnitude of the test statistics, while the significance stars indicate rejection levels at 1%, 5%, and 10%. Figure 5a shows the relationship between ET and ESG, where significant cointegration is observed consistently across most quantiles, suggesting that ESG factors play a stable long-run role in shaping the energy transition process. Figure 5b highlights ET and IT, with strong evidence of cointegration throughout the distribution, underscoring the robust contribution of innovation and technology to ET dynamics. Figure 5c,d indicate that ET also maintains significant long-run linkages with FD and TPU, though the strength of the relationship varies across quantiles, pointing to heterogeneous effects depending on the distributional state of ET. Figure 5e demonstrates a similarly strong and persistent cointegrating relationship between ET and EG across nearly all quantiles, reinforcing the fundamental role of economic growth. Collectively, the results confirm that ET is cointegrated with ESG, IT, FD, TPU, and EG, but the strength and intensity of these linkages differ across the conditional distribution of ET, highlighting the importance of using the QQARDL framework to uncover quantile-dependent and nonlinear relationships.

Figure 6a–e presents the long-run results of the Quantile-on-Quantile ARDL long-run between energy transition (ET) and its key determinants—ESG sustainability uncertainty, information and communication technology (IT), financial development (FD), trade policy uncertainty (TPU), and economic growth (EG). Figure 6a shows that ESG has a heterogeneous effect on ET, with significantly negative relationships concentrated at the upper quantiles of ESG and lower quantiles of ET, but some positive effects emerging at higher ET levels. This aligns with studies such as [44], which document that ESG-related uncertainty may initially constrain green investments but can also stimulate long-term transition strategies when ET is already advanced. Conversely, other works like [45] find that ESG integration strongly accelerates decarbonization efforts, suggesting context-dependent dynamics.

Figure 6b illustrates that IT exerts a strong and consistently positive long-run effect on ET across most quantile combinations, particularly when both ET and IT are at middle-to-upper quantiles. This provides robust evidence that advancements in digital technologies and ICT innovation play a catalytic role in accelerating the energy transition by reshaping how energy is produced, distributed, and consumed. Specifically, ICT enables the optimization of resource utilization through advanced data analytics, forecasting, and automation, allowing firms and utilities to minimize waste and improve system efficiency. In addition, the deployment of smart energy management systems—such as smart grids, grid-edge sensors, and real-time monitoring platforms—facilitates the seamless integration of renewable energy sources, improves load balancing, and reduces transmission losses. These mechanisms not only enhance the reliability of low-carbon systems but also unlock new pathways for sector-wide efficiency gains, from manufacturing and transportation to commercial buildings. In line with the findings of [46,47], this evidence underscores ICT’s role as a transformative enabler of energy efficiency and low-carbon innovation, highlighting its potential to drive both immediate decarbonization outcomes and long-term structural shifts toward sustainable energy systems.

Figure 6c, however, shows that FD has predominantly negative long-run effects on ET, especially at the lower quantiles of ET and FD, suggesting that financial development in some cases may facilitate fossil fuel financing rather than clean energy expansion. This outcome resonates with the arguments of [48], who caution that poorly aligned financial systems may reinforce carbon-intensive growth, although others like [49] provide evidence of finance-led green development in specific contexts. Figure 6d shows that TPU has mixed effects on ET, with negative relationships at lower quantiles but positive effects at higher quantiles of ET and TPU. This indicates that trade policy uncertainty can hinder early-stage transition efforts but may stimulate diversification and green innovation when ET is more advanced. Similar nonlinear effects are reported by [24,50], who argue that uncertainty shocks initially depress investment in renewables but encourage long-run resilience strategies. Finally, Figure 6e highlights that EG maintains a consistently positive and significant impact on ET across nearly all quantiles, underscoring the importance of growth as a foundation for transition. This finding supports the mainstream view in the literature that economic growth enhances the resources and capabilities needed for energy transformation [24,51,52], though some scholars like [53] emphasize the risk of growth-induced energy demand that may delay transition if not managed with green policies.

Figure 7a–e disclosed the short-run dynamics from the QQARDL framework between energy transition (ET) and its determinants: ESG uncertainty, ICT, financial development (FD), trade policy uncertainty (TPU), and economic growth (EG). Figure 7a suggests that ESG uncertainty exerts heterogeneous short-run effects on ET, with significantly negative coefficients at lower quantiles of ET and ESG but positive effects at higher levels. This highlights that ESG-related shocks can disrupt early stages of the transition but may stimulate adjustments once ET is more advanced. Similar nonlinear short-run effects of ESG performance on environmental outcomes have been observed by [19], who find that ESG activities can initially constrain firm strategies but support sustainability once integrated into decision-making. Figure 7b shows that ICT is consistently positive in the short run across most quantile combinations, with the strongest effects concentrated in the mid-quantiles of ET and IT. This underscores the role of digitalization and technological adoption in driving immediate improvements in renewable energy and efficiency, complementing long-run dynamics. Ref. [4] similarly document that ICT fosters clean energy deployment and enhances energy efficiency in the short run. Thus, the QQARDL results reinforce ICT as a robust short-run enabler of ET.

Figure 7c highlights that FD exerts predominantly negative short-run effects on ET, particularly when both FD and ET are at lower quantiles. This suggests that financial development may channel resources toward fossil-fuel financing in the short term, thereby hindering transition progress. Ref. [54] caution that unless financial systems are green-aligned, financial expansion can perpetuate carbon-intensive activities, a view consistent with these findings. However, contrasting evidence exists in Refs. [55,56], who find that well-structured financial systems can support renewable energy and sustainability goals. Figure 7d demonstrates that TPU generates mixed short-run effects on ET. At lower quantiles, TPU is associated with negative effects, reflecting policy uncertainty’s tendency to deter clean energy investments in the short run. However, at higher quantiles of TPU and ET, positive associations emerge, suggesting that firms and economies may adapt by diversifying and innovating in response to trade policy uncertainty. Ref. [57] argue that while uncertainty shocks initially suppress investment, they can promote resilience and alternative green strategies in later phases. Finally, Figure 7e shows that EG exerts strong and positive short-run effects on ET across most quantiles, with the most pronounced influence occurring at mid-quantiles. This suggests that economic growth not only supports ET in the long run but also has immediate reinforcing effects by providing resources for renewable energy projects and low-carbon technologies.

Figure 8a–e presents the Error Correction Term (ECT) derived from the QQARDL estimations, which measures the speed of adjustment of energy transition (ET) back to its long-run equilibrium following short-run shocks across different quantiles. The results reveal that the ECT coefficients are consistently negative and statistically significant at the 1% and 5% levels across all quantiles, thereby fulfilling the necessary condition for convergence. This outcome confirms the presence of stable long-run cointegrating relationships between ET and its key determinants—namely ESG sustainability uncertainty (see Figure 8a), information and communication technology (see Figure 8b), financial development (see Figure 8c), trade policy uncertainty (see Figure 8d), and economic growth (see Figure 8e). The statistical significance of the ECT underscores that any short-run deviations from equilibrium are temporary, and the system possesses a self-correcting mechanism that ensures eventual reversion to long-run equilibrium. Moreover, the magnitude of the ECT coefficients varies across quantiles, ranging approximately from −0.2 to −1.0, which demonstrates heterogeneity in the speed of adjustment depending on the distributional position of ET. At lower quantiles, where ET is relatively underdeveloped, the adjustment process tends to be slower, reflecting structural rigidities or institutional constraints that may delay the convergence to equilibrium. Conversely, at middle and upper quantiles, the coefficients become more negative in absolute terms, implying faster convergence and stronger corrective mechanisms when ET is more advanced. This pattern indicates that economies or sectors at higher levels of ET possess greater resilience and adaptability, enabling them to absorb shocks and realign with long-run trajectories more rapidly.

In summary, the QQARDL results, which reveal heterogeneous and quantile-dependent effects of ESG uncertainty, FD, ICT, TPU, and EG on ET, can be better understood through the alternating expansionary and contractionary phases of the energy transition. In expansionary phases—often marked by economic growth, favorable financial conditions, and technological optimism—ICT and EG exert strong positive influences, as seen at middle-to-upper quantiles, where digital innovation and resource mobilization accelerate renewable integration. By contrast, contractionary phases—associated with recessions, inflationary shocks, or financial tightening—dampen the transition, explaining why FD and ESG uncertainty often display negative or mixed effects at lower quantiles. Beyond macroeconomic cycles, geopolitical disruptions such as trade disputes, energy price volatility, or conflicts exacerbate these contractions, amplifying the adverse effects of TPU and delaying investment in clean technologies. Thus, the nonlinear patterns uncovered by QQARDL reflect the inherent tension between expansion and contraction in the U.S. transition process, shaped not only by institutional and technological drivers but also by broader economic and geopolitical forces.

5. Conclusions and Policy Pathways

5.1. Conclusions

Energy transition (ET) and its key drivers—ESG sustainability uncertainty, information and communication technology (IT), financial development (FD), trade policy uncertainty (TPU), and economic growth (EG)—are increasingly shaping the pathways toward low-carbon economies. Against this backdrop, our study investigates the short- and long-run effects of ESG, IT, FD, TPU, and EG on energy transition (ET) using quarterly data, thereby offering fresh insights into the nonlinear and quantile-dependent linkages that define the sustainability trajectory. In doing so, the study utilizes quarterly data spanning 2002Q1 to 2024Q4 and applies the novel Quantile-on-Quantile Autoregressive Distributed Lag (QQARDL) approach to capture the heterogeneous and nonlinear dynamics between variables across the conditional distribution. The QQARDL results confirm that ET is cointegrated with ESG uncertainty, ICT, FD, TPU, and economic growth (EG), but the strength and direction of these linkages vary across quantiles. In the long run, ESG shows mixed effects—negative at lower quantiles but supportive at higher ones—while ICT and EG consistently enhance ET, underscoring their catalytic role in technological progress and resource mobilization. FD, however, largely constrains ET at lower quantiles, reflecting its tendency to finance carbon-intensive activities, whereas TPU produces nonlinear effects, deterring ET in early stages but fostering adaptation and diversification at higher levels. The short-run dynamics mirror this heterogeneity, with ESG and TPU exerting mixed impacts, ICT remaining robustly positive, FD predominantly negative, and EG supportive across quantiles. The Error Correction Term (ECT) is negative and significant, confirming stable long-run relationships and showing that deviations are corrected over time, with faster adjustment at higher ET quantiles, indicating stronger resilience and adaptability as the transition advances.

5.2. Policy Pathways

The study proposed the following finings based on the findings. The evidence shows that ESG uncertainty constrains ET at lower quantiles but becomes supportive when the transition is more advanced. For U.S. policymakers, this underscores the importance of strengthening ESG disclosure standards through the Securities and Exchange Commission (SEC), enhancing climate-risk transparency, and ensuring regulatory consistency across states and sectors. Clearer and harmonized ESG frameworks would not only reduce investor uncertainty but also lower the cost of capital for renewable energy projects and other low-carbon investments. By providing standardized metrics and credible reporting requirements, the SEC can improve comparability across firms, preventing “greenwashing” and ensuring that investors can make informed decisions. Such clarity would encourage greater institutional investment in clean technologies and incentivize firms to align with long-term decarbonization targets. In doing so, ESG uncertainty is transformed from a barrier into a catalyst for energy transition, fostering a policy environment in which sustainable finance, technological innovation, and corporate accountability reinforce one another to accelerate U.S. progress toward net-zero commitments.

The strong and consistent role of ICT underscores the necessity of embedding digitalization into U.S. clean energy policy. Beyond traditional applications such as smart grid infrastructure, AI-enabled demand management, and digital monitoring systems, the integration of emerging ICT tools—including decision support systems, scenario planning models, and adaptive governance technologies—will be critical for managing uncertainties and supporting flexible environmental policies. Expanding investments in these areas will not only accelerate renewable integration and improve energy efficiency but also enhance the capacity of policymakers and firms to respond dynamically to evolving climate and energy challenges. Federal initiatives, such as the Department of Energy’s (DOE) digital innovation programs, can provide the necessary incentives for ICT-driven R&D, while state-level public–private partnerships can ensure that digital transformation and decarbonization proceed in tandem, building a resilient and adaptive energy transition framework.

The negative effect of FD at lower quantiles suggests that U.S. financial markets, if left unguided, may continue financing fossil fuel–intensive industries rather than green innovation. This calls for reforms in sustainable finance, such as expanding the U.S. green bond market, incentivizing climate-aligned lending, and embedding sustainability metrics into Federal Reserve supervision and stress tests. Taxonomies that classify sustainable investments, coupled with mandatory climate-risk disclosures, would align capital markets with clean energy objectives and transform FD into a driver of ET. Such measures would also reduce systemic risks associated with stranded fossil fuel assets.

The nonlinear role of TPU highlights the need for stable and predictable trade policies in the U.S. context. While uncertainty in trade rules may initially hinder clean energy investment, it can also stimulate diversification and resilience in later stages of ET. Policymakers should therefore prioritize trade agreements that incorporate renewable energy and carbon-reduction commitments, while avoiding abrupt shifts in tariffs or subsidies that could disrupt green supply chains. For example, harmonizing U.S. trade policy with global carbon border adjustment mechanisms could help stabilize investor expectations and create positive spillovers for renewable deployment. At the same time, fostering adaptive capacity within U.S. firms would ensure they can leverage TPU to drive innovation and competitiveness in clean technologies.

The robust positive effect of EG on ET underscores that economic growth remains vital for financing and sustaining the U.S. energy transition. However, this growth must be aligned with environmental priorities to avoid reinforcing carbon-intensive pathways. Federal initiatives such as the Inflation Reduction Act (IRA) already provide a framework by linking growth strategies with renewable subsidies, tax credits, and green infrastructure investments. Building on this, U.S. policymakers should expand green industrial policies, encourage circular economy practices, and promote eco-innovation across states. Such measures would ensure that economic growth strengthens the financial and institutional foundations of ET, while accelerating the decarbonization of the U.S. economy.

An additional dimension of the U.S. energy transition that emerges from these findings is the need to integrate environmental justice and equity considerations into policy design. While ICT adoption, ESG disclosure, financial development, and trade frameworks play pivotal roles in shaping the pace of transition, their benefits are unlikely to be distributed evenly without deliberate interventions. Low-income communities and marginalized groups often face barriers such as limited access to clean technologies, higher energy burdens, and restricted financing opportunities, which can exacerbate inequalities during the transition process. Policies that expand access to affordable digital infrastructure, support equitable financing mechanisms, and target subsidies toward vulnerable households are therefore essential to ensure that the shift to clean energy is not only technologically feasible but also socially inclusive. By embedding environmental justice into the transition context, U.S. policymakers can align decarbonization pathways with broader sustainable development goals, ensuring that progress in energy transition also advances equity and long-term resilience.

5.3. Managerial Implications

From a managerial perspective, the findings underscore the necessity of adopting differentiated strategies for supporting the energy transition (ET). The consistently positive role of ICT in both the short and long run suggests that managers in the energy and industrial sectors should prioritize digital transformation as a central component of corporate sustainability strategies. Investments in smart grids, artificial intelligence, and big data analytics can deliver immediate efficiency improvements while laying the groundwork for long-term competitiveness in a low-carbon economy. Furthermore, the results on ESG uncertainty highlight the importance of proactive disclosure and transparency: managers who strengthen ESG reporting practices and integrate sustainability into corporate governance structures are more likely to attract green finance and maintain stakeholder confidence, even under uncertain regulatory environments.

The evidence of adverse or mixed short-run effects of financial development (FD) and trade policy uncertainty (TPU) points to the importance of strategic financial and risk management. Managers should carefully align financing structures with green objectives, for instance, by issuing sustainability-linked bonds or leveraging climate-aligned investment instruments that reduce reliance on carbon-intensive capital flows. At the same time, diversification of supply chains and flexibility in production processes will be essential for mitigating risks associated with TPU, ensuring that firms remain resilient in the face of policy or trade volatility. Finally, given the robust positive influence of economic growth (EG) on ET, managers should frame sustainability not merely as compliance but as a growth strategy—embedding energy efficiency, renewable adoption, and green innovation into their business models to enhance resilience, secure long-term value creation, and strengthen competitiveness in the evolving global energy landscape.

5.4. Future Research Directions Toward Sustainability

Future research should extend beyond technological and institutional drivers of energy transition to incorporate broader sustainability dimensions, including environmental justice, social equity, and long-term resilience. Expanding the empirical framework to account for regional heterogeneity within the U.S. would provide deeper insight into how income inequality, energy access, and state-level policy regimes shape transition outcomes. Methodologically, integrating hybrid approaches such as machine learning with econometric models could enhance the capacity to capture nonlinear dynamics, structural breaks, and adaptive responses to shocks. At the same time, current measures of ESG sustainability uncertainty and trade policy uncertainty (TPU) remain relatively narrow, limiting their ability to capture multidimensional risks; future studies should therefore refine and expand these indicators to include firm-level, sectoral, and global perspectives. Further, research should explore cross-sectoral linkages—such as the interaction between digitalization, green finance, and trade openness—while embedding climate adaptation and biodiversity considerations into energy transition pathways. Such directions would not only strengthen the evidence base for inclusive and sustainable policy design but also align U.S. transition strategies more closely with the Sustainable Development Goals.