📈 economics

Asymmetric Energy-Price Shock Transmission and Electricity-Sector Carbon Transition in ASEAN: Panel NARDL and MIDAS Evidence

This paper utilizes panel NARDL and MIDAS models to analyze ten ASEAN economies from 2000 to 2025, finding that while international energy-price shocks exhibit short-run asymmetries and volatility-dependent effects on electricity-sector decarbonization, the long-run regional response remains largely symmetric, suggesting that higher fossil-fuel prices alone do not causally accelerate carbon transition without considering country-specific structural factors.

Original authors: Waewwan La-ongsri, Nuttaphat Sukchitt, Pasin Marupanthorn

Published 2026-06-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Waewwan La-ongsri, Nuttaphat Sukchitt, Pasin Marupanthorn

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The ASEAN Energy Balancing Act

Imagine the Association of Southeast Asian Nations (ASEAN) as a group of ten neighbors who all share a single, massive power grid. They are trying to switch their electricity from "dirty" sources (like coal and oil) to "clean" sources (like solar, wind, and hydro).

The researchers wanted to answer a simple question: When the price of oil and gas goes up or down, does it automatically force these neighbors to switch to cleaner energy faster?

Many people assume that if gas gets expensive, everyone will immediately panic-buy solar panels. This paper tests if that "automatic switch" actually happens, or if the reality is more complicated.

The Tools: How They Measured It

To figure this out, the authors built a "time machine" looking at data from the year 2000 to 2025. They used three main tools:

The "Fuel Cost Gauge": Instead of just looking at the global price of oil (like the Brent Crude), they calculated what that price actually costs each country in their own local money, adjusted for inflation. Think of it as checking the price of a gallon of gas at the pump in Bangkok versus Jakarta, rather than just looking at the global oil barrel price.
The "One-Way Mirror" (Asymmetry): They didn't just look at price changes; they looked at the direction. They asked: "Does a price hike push us toward green energy differently than a price drop pushes us away from it?" They treated price hikes and price drops as two different forces, not just opposite sides of the same coin.
The "Weather Vane" (Volatility): They checked if the chaos of the market matters. If prices are jumping up and down wildly (like a storm), does that change how countries react compared to when prices are calm?

The Findings: What the Data Actually Says

Here is what the study discovered, broken down simply:

1. The "Speed Bump" Effect (Short-Term vs. Long-Term)

When fuel prices suddenly spike, the countries do react quickly. It's like hitting a speed bump: the car (the electricity system) jolts, and for a moment, there is a strong push to look for alternatives. The study found that in the short term, when local fuel costs go up, the share of clean energy does tend to rise.

However, in the long run, the system settles down. The study found that the "push" from a price hike is almost exactly the same size as the "pull" from a price drop, just in the opposite direction.

The Analogy: Imagine a rubber band. If you stretch it (price hike), it snaps back. If you let it go slack (price drop), it relaxes. The study found the rubber band snaps back and relaxes with the same force. It doesn't mean a price hike permanently rewires the system; it just creates a temporary nudge.

2. The "One-Size-Fits-All" Trap

The researchers found that you cannot treat all ten ASEAN countries the same way.

The Hydro-Heavy Neighbor: One country (Laos) is almost entirely powered by water (hydro). Its "clean energy" score is already so high that it's hard to see the effect of oil prices. It's like trying to measure how much rain affects a swimming pool that is already full.
The Import-Dependent Neighbors: Countries that buy most of their fuel (like Singapore or Vietnam) react more strongly to price changes.
The Mixed Bag: For some countries, the data was too noisy or the history too short to draw a clear line.

3. The "Stormy Weather" Factor

The study checked if "wild" price swings (high volatility) changed the rules. They found that while wild swings might change how countries react to price drops, the data wasn't clear enough to say for sure. It's like saying, "When the wind is howling, maybe the trees bend differently," but the wind wasn't howling loud enough in the data to prove it.

4. The Prediction Game

Finally, the authors tried to use their models to predict the future.

The Surprising Winner: The simplest model (just looking at past trends) was often the best at predicting the future for most countries.
The Special Case: For Thailand, a more complex model that looked at monthly data (not just yearly) worked best. This suggests that for some places, looking at the "monthly heartbeat" of the market helps, but for others, the yearly average is enough.

The Bottom Line: What Should Policymakers Do?

The paper concludes with a very cautious message.

Don't expect a magic switch. Just because oil prices go up doesn't mean decarbonization will happen automatically or permanently. The relationship is real, but it is subtle and depends heavily on the specific country.

The Takeaway:
Think of energy prices as a thermometer, not a thermostat.

A thermometer tells you the temperature (it shows you when fuel is getting expensive and when countries are reacting).
A thermostat controls the heat (it forces the system to change).

The study says energy prices are a good thermometer to watch, but they are not a thermostat that will automatically fix the climate problem. To actually drive the transition, countries need more than just price shocks; they need specific policies, better infrastructure, and time to build new power plants.

In short: Rising fuel prices give a little push toward green energy, but they don't guarantee the journey. Every country walks that path at a different speed and in a different way.

Technical Summary: Asymmetric Energy-Price Shock Transmission and Electricity-Sector Carbon Transition in ASEAN

Problem Statement
This research investigates whether international energy-price shocks transmit asymmetrically to electricity-sector carbon-transition performance within the Association of Southeast Asian Nations (ASEAN). While regional policy ambitions for renewable energy and decarbonization are strengthening, the region remains heavily dependent on fossil fuels, with coal generating approximately half of regional electricity as of 2023. The central problem is that existing literature often treats energy-price shocks as symmetric macroeconomic disturbances or focuses on long-run policy design without empirically estimating how observed transition indicators respond to real, local-currency energy-price shocks at the country-year frequency. Specifically, the paper addresses the gap in understanding whether price increases and decreases have distinct short-run and long-run effects on the low-carbon share of electricity generation, considering the region's structural heterogeneity, exchange-rate exposures, and varying electricity-system structures.

Methodology
The study employs an unbalanced country-year panel dataset covering ten ASEAN economies (Brunei, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand, and Vietnam) from 2000 to 2025.

Variables:
- Dependent Variable: The low-carbon share of electricity generation ( $y_{i,t}$ ), defined as the percentage of generation from hydro, solar, wind, nuclear, geothermal, bioenergy, and other renewables, sourced from OWID/Ember/Energy Institute data.
- Shock Variable: Real local-currency Brent exposure ( $PE_{i,t}$ ), constructed by converting annual Brent prices into local currency using exchange rates and deflating by the Consumer Price Index (CPI). This captures both global oil price movements and exchange-rate pass-through.
- Shock Decomposition: Following the nonlinear ARDL (NARDL) framework of Shin et al., the log changes in real exposure are decomposed into cumulative positive partial sums ( $P^+_{i,t}$ ) and cumulative negative partial sums ( $P^-_{i,t}$ ) to capture asymmetric responses.
- Controls: Lagged inflation, exchange-rate changes, and real GDP growth.
- Threshold & Frequency: A volatility-threshold extension incorporates annualized energy-price volatility ( $V_t$ ) to test for regime dependence. Additionally, Mixed-Data Sampling (MIDAS) specifications utilize monthly Reﬁnitiv price data to evaluate if higher-frequency information improves annual monitoring.
Econometric Strategy:
1. Fixed-Effect Panel NARDL: Estimates separate short-run and long-run multipliers for positive and negative shocks while controlling for country-specific fixed effects and error correction dynamics.
2. Volatility-Threshold Extension: Augments the baseline model with interaction terms to test if transmission slopes differ during high-volatility years.
3. Rolling Forecast Comparison: Conducts country-specific out-of-sample forecasts comparing symmetric ARDL, asymmetric NARDL, and MIDAS-augmented models to assess practical monitoring accuracy.

Key Contributions
The paper makes three primary contributions to the literature:

Dataset Construction: It provides a transparent ASEAN country-year dataset linking electricity-sector transition performance directly to real local-currency energy-price exposure, addressing the lack of harmonized regional data on this specific relationship.
Asymmetric Estimation: It estimates asymmetric short-run and long-run transmission using a panel NARDL framework, explicitly testing whether positive and negative energy-price shocks yield statistically different long-run effects.
Monitoring Evaluation: It evaluates whether incorporating volatility states and mixed-frequency (monthly) price information improves the practical accuracy of transition monitoring at the country level, moving beyond simple annual averages.

Empirical Results

Error Correction and Short-Run Dynamics: The results indicate statistically significant error correction, with approximately half of any deviation from the country-specific transition path corrected within one year. There is a statistically significant positive short-run transmission from fuel-cost increases (positive shocks), suggesting that rising real local-currency fuel costs are associated with immediate increases in the low-carbon share, likely due to fuel-switching incentives and cost pressures on fossil generation.
Long-Run Symmetry: Contrary to the hypothesis of strong asymmetry, the estimated long-run response to fuel-cost increases and the absolute response to fuel-cost decreases are nearly identical. The Wald test fails to reject the null hypothesis of symmetry at the regional average (p-value = 0.967). The long-run multiplier for a positive shock is approximately 6.8 percentage points, while the response to a decrease is approximately -6.9 percentage points.
Volatility and Thresholds: While a high-volatility threshold (around 8% annualized volatility) was identified, the interaction effects between volatility states and shock channels are imprecisely estimated. This suggests that while high-volatility years may alter transmission channels, the evidence is not strong enough to claim a distinct structural regime.
Country Heterogeneity: Significant heterogeneity exists across nations. For instance, Lao PDR shows high sensitivity (though confounded by hydro-dominance), while Singapore and Vietnam show positive responses to price increases. Conversely, Cambodia, Malaysia, and Thailand show negative or mixed responses. Brunei was excluded from asymmetric calibration due to a flat transition share.
Forecasting Performance: No single model dominates the region. The parsimonious symmetric ARDL benchmark yields the lowest average Root Mean Squared Error (RMSE) across the region. However, NARDL improves accuracy for Indonesia, Malaysia, Myanmar, and Singapore, while MIDAS is superior for Thailand. This indicates that richer models can overfit short annual histories in some contexts but provide value in others.

Significance and Claims
The paper concludes with a cautious interpretation of its findings. It asserts that while directional fuel-cost pressure, volatility states, and country-specific electricity-system structures are informative for monitoring ASEAN transition performance, the annual panel data does not justify a causal claim that higher fossil-fuel prices directly and mechanically accelerate decarbonization.

The primary significance lies in shifting the policy narrative: rather than assuming price hikes automatically drive transition, the evidence supports a "cautious monitoring" approach. Policymakers should track real local-currency energy-price exposure, distinguish between price increases and decreases, and account for volatility states. The study emphasizes that the annual regional average masks substantial structural differences, and future research requires harmonized monthly generation-by-fuel data to move from monitoring calibration toward stronger causal identification of how imported fuel costs, policy interventions, and system constraints jointly shape decarbonization.