Simulating the Impact of the U.S. Inflation Reduction Act on State-Level CO₂ Emissions: An Integrated Assessment Model Approach

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The draft paper explores the impacts of  some of the policy measures of the IRA programme on CO2 emissions. Using (a version of the) GCAM model the paper presents computational results for a number of policy scenario’s.

The text is mostly clearly written, but its limitation to just presenting model outputs creates a lack of clarity and transparency for readers not intimately familiar with the model(s) used. As a result, the model appears as a black box, and the paper does not provide any understanding of the reasons why the model behaves the way it does, let alone of the real-world mechanisms at play. A number of modifications and additions will be needed to achieve the goal of enhancing such understanding (as formulated in lines 96, 99) as well as enhancing trust in the results, and policy relevance.

 

First, in the current version, any attempt at explaining the mechanisms behind the model results is missing - even if these results are considered ‘somewhat’ intriguing or unexpected. Such explanations needs to be found, added and discussed. For example, how can the unexpected result of a negative effect of EV cost reductions in some states be explained? More generally, what model assumptions or relations cause the significant differences between the states? What is the explanation behind the apparent synergy between different policy elements in case of the ‘combo’ scenario’s? 

 

Second, a solid reflection on the uncertainty in the model results should be added. Results are presented in three digits, but what does this say given the multitude of uncertainties in and of the model? 

 

Third, the way in which the so-called ‘policy effectiveness density’ is used is puzzling. The numerator (as if formula (1) ) relates to the % change in total emissions per state; are these total emissions over every sector in the state, or from the transport sector only? Does the denominator (% change in the costs) reflect the % of the costs in the specific (sub)sector such as (total costs of ) EV’s – or overall costs (of what?- emission reduction costs only?). 
And then, more puzzling, in Table 2, policy effectiveness densities are expressed as %, while the definition according to equation (1) says it is a dimensionless number?

And how can, for example, one percent of EV cost reduction lead to 21% of (total??) emission reduction in California (line 260)??

Why not work with the amount of emissions reduced per dollar spent? That would also enable a clearer comparison by policy makers.

 

Fourth, a number of more specific points of attention:

 

105-106 : ‘emissions reductions may not always result in financial benefits’ Not clear how this statement relates to the argument made in the sentence before. 

 

Line 124: shouldn’t 2010 be 2050??

 

Lines 207-208 (text under figure) is confusing, PTC reduces emissions by 4% (of reference); how can this be equal to 16% (of 2005 level) at same time?

 

Line 306: “A higher value of combo-policy effectiveness density also demonstrates reducing emission reduction costs is more effective”. I am not sure what is meant with this statement; is it meant to say that the combo policy is more EFFICIENT, i.e., more result per dollar spent??

 

Table 2: The averages are apparently computed counting every state equal. So, states with a big effect will have a big influence on the averages, even if small in terms of total CO2 emissions. What is then the meaning of computing the averages?

 

Figure 2: Confusing: are ‘residential’, ‘industry’ energy sources??

Author Response

Please see the attached document for our point by point response to your comments. Thank you.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The researchers analyze the influence of macroeconomic system on the future CO2 emissions in U.S.A. The discussed topic is relevant due to the climate change and GHG emissions. It proves whether the proposed policy allows the contribution to the natural environment.  The methodology should be improved by mentioning the scenario 4 and 7 (in the methodology part: scenarios design). The conclusions clearly state the results given in the text and are consistent with the posed questions in the paper. The references listed are accurate with the topic. However, there are 10 articles cited of Shittu E. (the author of the paper) and that is a very high number. Most tables and figures are presented in a clear way. Only Table 1 and Figure 1 are hardly legible. As for Table 1 the font should be increased. After including those several comments, I recommend the paper to be published.

Author Response

Please see the attached file for our responses to your comments. Thank you.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have, in general, well understood my comments and suggestions with respect to the previous version of the paper; as a result, most of the confusing and unclear points have been resolved. 
First, the explanation in lines 240-253 of why per dollar effectiveness is not possible with the model might be technically correct, but not very clarifying for non-insiders, it doesn’t help transparency. 

Second, regarding the explanation of model outcomes, one of my major concerns remains: much of the explanations given in lines 290 – 328 is hypothetical in the sense that the authors present a number of mechanisms or processes that could explain the results. They do however not precisely explain what mechanisms actually are causing the model to behave the way it does. This requires a thorough introspection of the model – and might be difficult given the model’s foundations in econometrics, which tend to be statistically based. 
A case in point is the explanation of the negative impact of EV tax credits on CO2 emissions in some states. Basically reasoning, this is only possible if the extra CO2 emissions of electric power usage of the additional EV’s are higher than the (direct) reduction of CO2 emissions from the accompanied reduced use of petrol or diesel cars. A quick scan of internet however points out that the CO2 emissions per km of fuel cars is at least double (often more than double) than that of EV’s (accounting for the CO2 emissions of electric power generation), depending on assumptions. So it is highly unlikely that the extra E-power consumption provides the explanation for the negative impacts in some states – even if those states use non-sustainable fuels for electricity consumption. Has CO2 emission of the production of batteries been taken into account? If yes, this might provide part of the explanation (but not in the long term). 
I challenge the authors to clarify this behavior; if not, this outcoue casts doubt on the validity of the model - and its results!

 

 

 

Author Response

We wish to thank the Reviewer for the incisive comments. We have now examined the model more carefully to identify the underlying drivers of the model outcomes. We hope that this revision meets the Reviewer's expectations. Please see the attached document for more details.

Author Response File: Author Response.pdf