Avik Ghosh and Ankit Magotra�Linking fiscal strength to climate action: evidence from environmental fiscal instruments and threshold effects in Asia-Pacific

Andrew Coleman

Asia School of Business and Reserve Bank of New Zealand

July 2025

Amitav Ghosh (2016) “The Great Derangement: Climate Change and the Unthinkable”

​

Asia is at the center of the climate crisis and global responses to the crisis must center on Asian people and landscapes…..

​

Asia is “conceptually critical to every aspect of global warming.”

​

​

Avik Ghosh and Angkit Magotra

“Do weak fiscal fundamentals undermine the effectiveness of climate policies”

​

​

Baseline Model: Two-way Fixed-Effects Regression�

GDP per capita growth

Population growth (Urban or total)

Gross capital formation

Electrical power consumption (or energy?)

Renewable energy fraction

​

Please be more precise

C02 emissions

(per person or GDP intensity

​

​

Environmental taxes

(share of GDP or tax revenue)

Please be more precise

​

​

Baseline Model: Two-way Fixed-Effects Regression�

1. Are environmental taxes a good measure of the all the policies that affect C02 emissions? Are there omitted variables that affect the coefficient estimates?

​

• An ETS (not a tax) may affect emissions but raise very little revenue because permits are given away (biases coefficient towards zero)
• A tax or ETS may be introduced at a low rate but there are promises of future increases. Companies may respond to future values not current values
• Governments may introduce subsidies to encourage fewer emissions
• Government may use regulations as well as taxes (leading to an over-estimate of the effect of taxes)
• Government may use regulations instead of taxes (leading to an underestimate)

Baseline Model: Two-way Fixed-Effects Regression�

2. Reverse causality: environmental tax revenues are likely to be a function of emissions
1. When emissions are high, a government with environment taxes will make more revenue. Simultaneous equation bias will bias coefficient upwards (less negative).
2. Endogenous tax relief: taxes are sometimes temporarily lowered when oil prices increase, and oil use declines, to provide relief to consumers.
3. Endogenous taxes: Governments may increase taxes in response to unacceptably high emissions Simultaneous equation bias will bias coefficient upwards (less negative).
4. The IV strategy used does not obviously counter this

Baseline Model: Two-way Fixed-Effects Regression�

3. Energy use or green/black substitution?
1. Higher energy taxes could lead to a reduction in total energy use or a substitution of green energy for black energy, or both.
2. It is hard to tell how conditioning on energy use (electricity use?) in the Z variables affects the coefficient on environmental taxes. If environmental taxes reduce energy use, and you condition on energy use, you will mainly pick up green-black substitution and underestimate the effect of taxes on emissions.

​

Baseline Model: Two-way Fixed-Effects Regression�

Summary

• There seems to be a systematic negative relationship between environmental taxes and emissions,
• There are several conceptual and practical problems with this regression which gives us little reason to think the coefficients are accurately estimated.
• It may be better to narrow the scope of the investigation to reduce some of these conceptual problems. Eg perhaps concentrate on C02 emissions from cars or freight transport

Instrumental Variable: Instrumenting fiscal size�

Interactive term to see if fiscal size matters

​

C02 emissions

(per person or GDP intensity

​

​

​

Instrumental Variable: Instrumenting fiscal size�

The Basic problem

These instruments may not be exogenous to emissions: higher than expected GDP growth is probably correlated with higher than expected emissions

​

Much better instruments may be available eg health or education spending, correlated with fiscal size but unlikely to be correlated with emissions

​

Threshold regression

Summary

• This is an interesting research question, and useful dataset

​

• There is enough here to suggest strong fiscal fundamentals make climate policy more effective, although the magnitude does not see very large
• But….not convinced that the coefficients are unbiased and it may be helpful to discuss the nature of these biases more

​

• Thank you