Air Quality Uncertainty and Earnings Management

Round 1

Reviewer 1 Report

Very interesting paper. However the research question seems quite daring. The paper seems a little bit out of scope of the journal. 

Introduction: the theory ... or how the paper enhances the literature is not very clear highlighted.

titutional background: this would be the research context. Please cite some references here!

The literature review is very scarce. It should be enhanced. 

Hypothesis：When the regional air quality uncertainty increases, firms’ managers tend to 118
improve their degree of earnings management.

where do you accept or reject the hypothesis???? Where are the calculations for that? 

Research design: it would be proper to stress out why the authors have chosen that particular research context. It also would be proper to state something about the analysis that the authors conduct and to give some proper explanations from the literature that it is a correct way to implement the analysis.

The paper lacks discussions. The authors do not compare their results with previous findings of the literature in order to highlight the novelty of their research.

 

Conclusions are quite weak. This section should be enhanced by theoretical implications - how does the paper advance the literature; managerial implications; limitations; future research perspectives. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

The research addresses the link between the changeability of air pollution and company earing management.  

Research wrote neatly in the form, with some errors:

1. The literature development does not identify the research gap, thus the hypothesis is not linked with the unidentified research gap. No test for the completeness of the literature review. Refer to methodology [1].
2. The dataset cross-sectional without control over the entities. The research makes two assumptions, ones the seat of the entity is similar to the sit of the factory plants. The second within a given province the companies react similarly. The effect of the company itself is not controlled thus the standard errors are not sufficiently clustered. Instead of the cross-sectional approach, a panel data set would be more suitable for the research, thus it constitutes the inherent limitation of the paper, which is not disclosed to the readers.
3. The model equation suffers from omitted variable bias, the number of the controls is missing e.g.: market capitalization, long-term liabilities to the asset, etc. compare [2], not explained why you use the “The shareholding ratio of the largest shareholder”, and not the largest ultimate
4. You face a similar problem for the cross-section data as Staszkiewicz and Szelągowska, they motivated the selection of the cross-sectional data as follows: ”The study relied on cross-sectional methods that embody ….”[3, p. 3807] a similar explanation or the link should be introduced in the study limitation.
5. The averaging and changing perspective to a one-year analysis of the air pollution flattens the effects, as the quality of air changes substantially over the day. Thus the transmission channel between the earing management and the air quality risk is fragile (how about the waters, wastage, temperature, etc.) thus the effect should be discussed in the study limitation section. Compare for example discussion Chen et al. [4]

Besides the above, quite an interesting study.

Literature

[1]        S. K. Tomczak and P. Staszkiewicz, “Cross-Country Application of Manufacturing Failure Models,” J. Risk Financ. Manag., vol. 13, no. 2, p. 34, Feb. 2020, doi: 10.3390/jrfm13020034.

[2]        J. B. Kim, J. W. Kim, and J. H. Lim, “Does XBRL Adoption Constrain Earnings Management? Early Evidence from Mandated U.S. Filers,” Contemp. Account. Res., vol. 36, no. 4, pp. 2610–2634, 2019, doi: 10.1111/1911-3846.12493.

[3]        P. Staszkiewicz and A. Szelągowska, “Ultimate owner and risk of company performance,” Econ. Res. Istraživanja, vol. 32, no. 1, pp. 3795–3812, Jan. 2019, doi: 10.1080/1331677X.2019.1678499.

[4]        S. Chen, P. Qin, J.-S. Tan-Soo, and C. Wei, “Recency and projection biases in air quality valuation by Chinese residents,” Sci. Total Environ., vol. 648, pp. 618–630, Jan. 2019, doi: 10.1016/j.scitotenv.2018.08.153.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments: This is a very interesting study and I like the research question which is topical in the present climate.

Minor and major comments are below:

Abstract

 

• We find that the uncertainty of air quality uncertainty affects earnings management 14 through the intermediary of government environmental investment. - This sentence is unclear

 

Introduction

• Further evidences show that the air quality uncertainty affects earnings management through 39 government environmental investment, which reveals the mechanism of the effect of air quality 40 uncertainty on earnings management.- Add a footnote explaining how this channel work, maybe through an example.
• Literature review needs to be expanded, surely there is much more prior literature that talks about earnings management and environmental factors. Some mention of literature focusing on ESG and earnings management would be better as well.
• Earnings management literature in general should be discussed to position this paper within that literature and to identify the specific channel of earnings management due to environmental factors.
• Is this the first study to analyse this for China? Contribution should be explained in more detail. And similar papers highlighted differentiating this paper with those papers.

Hypotheses Development

• Government intervention with regard to environment quality would have a lagged effect, controls from not just the firm side but the government side is also important in this context.
• It is hard to identify that the only channel is government intervention focused on improving air quality. Please provide examples here.
• There can be a lot of omitted variable issues here such as other factors that may lead a government to increase/decrease taxes such as local government elections.

Research Design

Sample selection

• More details should be provided about the air quality variables in this section as well rather than in the latter sections about the frequency of the variable. Since accounting variables are quarterly, was this air quality measure aggregated.
• Do you account for firms that shift their headquarters to different states?
• What about firm concentration and activities within and across a province.
• There may be a very small number in your sample that falls under this situation nevertheless that needs to be considered.
• There might be a sample selection problem here where firms that are more prone to conduct pollution most likely to conduct earnings management.

Empirical Analysis

• What about environmental factors other than air quality that may impact the results?
• Additional controls at government level is needed.

Robustness tests

• With regard to the IV, do you consider firm subsidiaries that maybe situated across different regions within the same province?
• Or even a case where one firm is highly concentrated in one city and another more dispersed across regions within the same province. Would be good to see the results of the IV test to see if this is a weak IV or not?

Many thanks

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The manuscript has been improved considerably. Congratulations! In my perspective it can be published. Best luck in attracting citations!

Author Response

The reviewer has no comments.

Reviewer 2 Report

No other comments. Good luck with new research.

Author Response

The reviewer has no comments.

Reviewer 3 Report

1.The study empirically examine the influence of rising political costs from the air quality, uncertainty caused by regional air quality fluctuations on firms' earnings management. However, very little is mentioned about prior research on earnings management. Especially, other factors that might cause these firms to manage earnings.

2. More controls can be added if possible given data availability to account for the following:

E.g.Prior literature state positive earnings as a common target of earnings management. Evidence of managing towards positive earnings is a larger frequency of small positive earnings (Burgstahler and Dichev, 1997; Leuz, Nanda, and Wysocki, 2003). 

E.g. If there is earnings management, most probably This if earnings are smoothed, large losses should be relatively rare.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf