Informal Employment, the Tertiary Sector, and the Gross Domestic Product: A Structural Equations Model for the Mexican Economy
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
While the authors provide a very interesting topic with quite some merit, there are some issues to be considered before publications:
- The motivation for the article is lacking, where does it provide to economic theory, in which way does in provide for practical policy making?
- The article requires significant restructuring. On the one hand there are multiple parts can could be summarized like sections 2-5 while on the other hand sections 6.2 and 6.5 would be superfluous in a scientific publication. The remainder of section 6 could focus more on the model conception and measurement of the variables and their sources.
- Very little literature on the focus of the article is cited. Is this because no other sources exist (if so, mentions this in the text as it can push the relevance of the article). If more comparable studies exist cited them and consider their methodology.
- Model conceptionalization should be rethought. inf is a percentage variable, thus the linear relations indirectly assumed in the SEM do not hold. Consequently, a conclusion like "A 1% variation in informality affects the tertiary sector by 37%." is problematic as it would translate into a 24% variation in informality affects the tertiary sector by 888%.
- At least two aspects are touched upon in the article but for some reason they are not considered in the analysis, i.e. they are considered to be constant over the considered 42 years. First, the governmental framework regarding the informal sector is missing, i.e., how is the informal sector policed, when and in how far did the government actively worked at a detection of informal workers, i.e., established penalties or task forces. Second, the situation of social security is mentioned but not considered. In general government services are incentives for working the formal sector, so changes in them should be considered in a model.
- In general the approach is a bit problematic as time series data is used in an approach that is inherently focussed on crosssectional data. Thus, path dependency effects present in all variables of the model are neither considered in the analysis nor even touched upon in the discussion.
Author Response
Comments 1: The motivation for the article is lacking, where does it provide to economic theory, in which way does in provide for practical policy making?
Response 1: Thanks for the comment, section 4 was included, where it shows the importance of public policy in the matter of informality. The main program in the fight against informality over the last 20 years are also shown. The section concludes by highlighting the importance of creating public policies that reduce informal employment.
Comments 2: The article requires significant restructuring. On the one hand there are multiple parts that could be summarized like sections 2-5 while on the other hand sections 6.2 and 6.5 would be superfluous in a scientific publication. The remainder of section 6 could focus more on the model conception and measurement of the variables and their sources
Response 2: In response to your comments, sections 2-5 were merged. The above can be seen in section 3. As for section 6.5, "Exploratory Data Analysis and Correlation Matrix," this was not included in the original manuscript. However, the academic editor made the following comment: “In this section, please describe how you conducted the exploratory data analysis and created the correlation matrix”. Therefore, it was decided to leave it in the document. Now, the section 6 is named section 5.
Comments 3. Very little literature on the focus of the article is cited. Is this because no other sources exist (if so, mentions this in the text as it can push the relevance of the article). If more comparable studies exist cited them and consider their methodology.
Response 3: In response to your comment, we have decided to expand section 5.3 by citing examples of structural equation models applied to the North American and Mexican economies, specifying the variables used in the analysis. In the same way, the review of updated literature on the topic was increased, which can be seen in section 3, 5 and 6.
Comments 4. Model conceptionalization should be rethought. inf is a percentage variable, thus the linear relations indirectly assumed in the SEM do not hold. Consequently, a conclusion like "A 1% variation in informality affects the tertiary sector by 37%." is problematic as it would translate into a 24% variation in informality affects the tertiary sector by 888%.
Response 4: Regarding the mentioned observation, the interpretation and units of the variables were reviewed. It was identified that informality is a latent variable, meaning it is not directly measured but inferred from other observed variables and is dimensionless. The manuscript analyzes the correlation between observable and latent (informality) variables based on their signs. Finally, it is indicated that informality has apositive relationship with the GDP of the tertiary sector, where a one-unit increase in informality is associated with a 0.37-unit increase in the GDP of the tertiary sector. The above can be seen in section 8 of the document.
Comments 5. At least two aspects are touched upon in the article but for some reason they are not considered in the analysis, i.e. they are considered to be constant over the considered 42 years. First, the governmental framework regarding the informal sector is missing, i.e., how is the informal sector policed, when and in how far did the government actively worked at a detection of informal workers, i.e., established penalties or task forces. Second, the situation of social security is mentioned but not considered. In general government services are incentives for working the formal sector, so changes in them should be considered in a model.
Response 5: Thanks for the observation. In response to this and with a view to strengthening the document, section 2 entitled "Theoretical frame of informality in Mexico" was integrated. This section addresses the observation made. First, the theoretical and conceptual framework of informality in Mexico is presented, showing the definition given by the ILO and experts in the field, specifying the reasons why this definition is used, since there are more than 10 definitions on the phenomenon. Second, the same section responds to the observation of social security because in Mexico the individual definition is used, that is, if an individual does not have social security for his employment, although working for formal enterprises, he is considered informal. This definition is the most widely used in the country and is used for the present research to study the phenomenon. The definition and its relation to social security is therefore set out in theory in section 2, while figures are dealt with section 3.
Comments 6. In general the approach is a bit problematic as time series data is used in an approach that is inherently focussed on crosssectional data. Thus, path dependency effects present in all variables of the model are neither considered in the analysis nor even touched upon in the discussion.
Response 6: In response to the comment, the results of the multicollinearity analysis obtained using the Variance Inflation Factor (VIF), have been added to Table 3 in the results section.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors
The paper is generally well developed and presented will need a bit more work before it can be ready.
1. The abstract is not informative and needs to be rewritten.
2. The contextualization need to justify why this research is warranted in the current context with facts and figures, business problem.
3. There were no theories to underpin the research.
4. Tables and figures should be presented more legibly.
5. There was no literature review to identify the important predictors but suddenly there is structural equation modeling.
6. Why these variables were chosen not others also have not been clearly explained.
7. Why did you choose the covariance-based SEM?
8. The assessment of univariate normality is not important as the model is a multivariate model thus multivariate normality should be assessed.
9. The authors reported several selected fit values but have left out chi square/df, TLI and p-values.
10. Discussion is very weak as there was no comparison and contrasting with the literature.
11. There are also no specific implications.
12. References are mostly dated.
13. Check reference formatting and completeness.
Comments on the Quality of English Language
Acceptable.
Author Response
Comments 1. The abstract is not informative and needs to be rewritten.
Response 1: The abstract was rewritten, however, the observations of you and the editor of the journal were considered.
Comments 2. The contextualization needs to justify why this research is warranted in the current context with facts and figures, business problem.
Response 2: The problem of informality is not a matter of industrial policy but a social issue of public policy. That is, although more than 60% of the companies in the country are informal, public policy oriented to the promotion, productivity and competitiveness of enterprises, pursues different objectives and does not apply to informality. The reason is that public policy is first sought to generate actions, instruments and mechanisms which enable micro-entrepreneurs (street vendors for example) to increase their level of well-being, for example in terms of health, and to ensure that they have decent jobs, with access to social security and the benefits provided by labor law. Once the above is met, the policy oriented in terms of industrial development, directed to financing, regulation, institutional design, rule of law, among other needs of enterprises, are industrial policy. However, the authors integrate two new sections: 2 and 4. In section 2, entitled "Theoretical frame of informality in Mexico", the definition of informality in Mexico is presented, together with the reasons why this definition is used for the Mexican case; The definition is submitted in terms of social public policy. While, in section 4 entitled "Informality and public policy in Mexico" the programs that the Mexican government has carried out in the fight against informality are explained. This section shows that all are socially related and gives the reason for this.
Comments 3. There were no theories to underpin the research.
Response 3: In response to the observation, a section entitled "Theoretical frame of informality in Mexico" is included which presents the theoretical framework of informal working conditions in Mexico under the definitions of the ILO and experts in the field, identifying how they are measured in the country, this provides a basis for econometric modelling.
Comments 4. Tables and figures should be presented more legibly.
Response 4: Thanks for the comment. Tables and figures have been in the paper.
Comments 5. There was no literature review to identify the important predictors but suddenly there is structural equation modeling.
Response 5: Lines 15-240 contain the theoretical foundations of the investigation, which allows contracting with the results obtained.
Comments 6. Why these variables were chosen not others also have not been clearly explained.
Response 6: In response to your comment, we have decided to expand section 5.3 by citing examples of structural equation models applied to the North American and Mexican economies, specifying the variables used in the analysis.
Comments 7. Why did you choose the covariance-based SEM?
Response 7: Given the importance of the topic, we have included in Section 6, dedicated to results, a discussion of the advantages and limitations of covariance-based structural equation modeling (CB-SEM) compared to other methodologies such as Partial Least Squares Structural Equation Modeling (PLS-SEM).
Comments 8. The assessment of univariate normality is not important as the model is a multivariate model thus multivariate normality should be assessed.
Response 8: Univariate normality tests were omitted from Table 3, and instead, the results of Mardia's test for multivariate normality, assessing skewness and kurtosis, were included.
Comments 9. The authors reported several selected fit values but have left out chi square/df, TLI and p-values.
Response 9: In response to the comment, the missing indices and their corresponding interpretations have been added to Table 4.
Comments 10. Discussion is very weak as there was no comparison and contrasting with the literature.
Response 10: In response to your comment, we have decided to expand section 5.3 by citing examples of structural equation models applied to the North American and Mexican economies, specifying the variables used in the analysis. The discussion of the results was adjusted. Likewise, the conclusion of the investigation was deepened.
Comments 11. There are also no specific implications.
Response 11: Thanks for the observation. In response to this and with a view to strengthening the document, section 2 entitled "Theoretical frame of informality in Mexico" was integrated, where the implications are appreciated. This section addresses the observation made. First, the theoretical and conceptual framework of informality in Mexico is presented, showing the definition given by the ILO and experts in the field, specifying the reasons why this definition is used, since there are more than 10 definitions on the phenomenon. Second, the same section responds to the observation of social security because in Mexico the individual definition is used, that is, if an individual does not have social security for his employment, although working for formal enterprises, he is considered informal. This definition is the most widely used in the country and is used for the present research to study the phenomenon. The definition and its relation to social security is therefore set out in theory in section 2, while figures are dealt with in section 3.
Comments 12. References are mostly dated.
Response 12: The service of the journal provides the Academic Editor’s Report, indicating that the document references are correct according to the parameters of the journal. Report and certificate attached.
Comments 13. Check reference formatting and completeness.
Response 13: The service of the journal provides the Academic Editor’s Report, indicating that the document references are correct according to the parameters of the journal. Report and certificate attached.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors
The manuscript titled: “Informal Employment, the Tertiary Sector, and the Gross Domestic Product: A Structural Equations Model for the Mexican Economy” is interesting and relevant, but there are several ambiguities present in the manuscript. I recommend the authors make several changes that need to be implemented before publication.
- The abstract requires substantial revision! The abstract should not include sources and needs to be reworked.
- What is the main aim of the paper? What is the primary objective of the research?
- Lines 15-240 contain the theoretical foundations, but the authors could have paid more attention to the research gap, which is not evident. I recommend shortening the processed text and focusing on the possible causes and factors of informal employment. Likewise, the authors could have formulated a research hypothesis derived from the theoretical framework.
- In terms of results, I appreciate the processing and interpretation of the results, which, in my opinion, are the best-processed part of the entire manuscript.
- However, the discussion requires revision. The discussion identifies relationships between macroeconomic factors (e.g., taxes, unemployment, wages) and the level of informality but does not provide a sufficiently deep explanation of these relationships. Why does an increase in taxes specifically lead to the shift of workers into the informal economy? The authors only briefly justify this in the discussion without detailing specific mechanisms (e.g., how exactly taxes are evaded or what types of activities occur in the informal sector). The authors mention that higher government spending is associated with an increase in informality in countries with corruption, but this argument could have been elaborated further — for instance, by providing concrete examples or countries. Similarly, I recommend adding more recent sources in the discussion... discussing findings from 2004 or 2003 is not appropriate.
- Conclusions: the authors should include the limitations of the research, which are currently absent. It is also standard practice to mention future research directions, which are also not included.
I positively assess the use of structural equations and modelling, but overall, I get the impression that the manuscript is mostly theoretical in nature and lacks practical impact on the issue being studied. It is also necessary to adjust the formal aspect of the manuscript (the readability of Figure 6 or Figure 7). Figure 1 requires reworking and atc.
Comments on the Quality of English Language
I did not find any fundamental errors and shortcomings. I recommend checking with native speakers.
Author Response
Comments 1. The abstract requires substantial revision! The abstract should not include sources and needs to be reworked.
Response 1: The abstract was rewritten, however, the observations of you and the editor of the journal were considered.
Comments 2. What is the main aim of the paper? What is the primary objective of the research?
Response 2: Thanks for the observation. The main objective of the research can be identified in the introduction.
Comments 3. Lines 15-240 contain the theoretical foundations, but the authors could have paid more attention to the research gap, which is not evident. I recommend shortening the processed text and focusing on the possible causes and factors of informal employment. Likewise, the authors could have formulated a research hypothesis derived from the theoretical framework.
Response 3: The objective of the research is not to analyze the factors that determine informal employment. Rather, the theoretical framework aims to support the behavior of the variables used in the model, since it focuses on identifying not only the correlations between observed and latent variables but also their causality as established by structural equation models. In this sense, the model aims to show the effect of informality, as a latent variable, its influence on the GDP of the tertiary sector. For this reason, the relationship between each variable, such as taxes, unemployment, public spending, among others, is delved into to explain informality. The above can be seen in sections 2 – 4. It is worth mentioning that, in response to your comments, two more sections were added, namely: section 2 and section 4, which shows the theoretical framework of informality and policies. public to combat it.
Comments 4. In terms of results, I appreciate the processing and interpretation of the results, which, in my opinion, are the best-processed part of the entire manuscript.
Response 4: The authors are grateful for the recognition of our work.
Comments 5. However, the discussion requires revision. The discussion identifies relationships between macroeconomic factors (e.g., taxes, unemployment, wages) and the level of informality but does not provide a sufficiently deep explanation of these relationships. Why does an increase in taxes specifically lead to the shift of workers into the informal economy? The authors only briefly justify this in the discussion without detailing specific mechanisms (e.g., how exactly taxes are evaded or what types of activities occur in the informal sector). The authors mention that higher government spending is associated with an increase in informality in countries with corruption, but this argument could have been elaborated further — for instance, by providing concrete examples or countries. Similarly, I recommend adding more recent sources in the discussion... discussing findings from 2004 or 2003 is not appropriate.
Response 5: In response to their comments, the relationship between the variables used in the econometric estimation and informality were included. To do this, the review of updated literature on the topic was increased, which can be seen in section 3.
Comments 6. Conclusions: the authors should include the limitations of the research, which are currently absent. It is also standard practice to mention future research directions, which are also not included.
Response 6: Thanks for the observation. Section 6 presents the limitations of the structural models, while the conclusions show the limitations of the research. Also, recommendations for future studies are shown.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for Authors
While the authors have reworked the article, most the initally mentioned issues have not yet been addressed in any decent fashion. Thus, the four/five most critical issues are repeated here again:
- Even though section 4 has been added, the main issue persists, as in the question has not been answered inhowfar the article does provide to the existing literature. The main reason for this is, that the state of the literature is not, in any way, being presented in the article. While two new sources are introduced with regard to the used methodology, a decent literature review on studies working on the informal sector in Mexico is missing.
- The comment with regard to restructuring the article did not imply that the respected titles should be deleted, but that the previous sections 2-5 can and should be significantly shortened to focus on the main ideas.
- With regard to comment 5, the two relevant aspects have been introduced theoretically, but the question remains, why they are not reflected in the empirical analysis.
- The solution to comment 6 does not solve the respective issue. If at all path-dependency is detected my auto-correlation tests not multicollinearity tests. Relevant changes to reflext the time-series nature of data, however, have not yet been considered in the empirical analysis itself and neither are any reasons discussed why this has not been done and what limitations result from omitting this type of analysis.
Author Response
Comment 1: Even though section 4 has been added, the main issue persists, as in the question has not been answered inhowfar the article does provide to the existing literature. The main reason for this is, that the state of the literature is not, in any way, being presented in the article. While two new sources are introduced with regard to the used methodology, a decent literature review on studies working on the informal sector in Mexico is missing.
Response 1: In response to your observation, the following change was made. In addition to the section on the definition of informality and public policy carried out in Mexico in the fight against informality, an exclusive section was integrated into the document that reviews the in-depth literature about informality, under the escape and exclusion approach, which strengthens the theoretical framework of informality and supports the choice of variables and their relationship to the study problem. The new section is supplemented by 31 references, of which 15 are new in the document. Overall, the document was composed of 72 references.
Comment 2: The comment with regard to restructuring the article did not imply that the respected titles should be deleted, but that the previous sections 2-5 can and should be significantly shortened to focus on the main ideas.
Response 2: In response to your observation, the merging of sections 2 to 5 was left. However, references to theoretical issues were removed from the reports. These references were placed in the new section 2, which fully explains the theoretical framework of informality. This new section is integrated by 31 references, of which more than 10 are new, while the remaining ones were taken from sections 2 to 5. In this sense, section 3, which includes the union of sections 2 to 5, only addresses the current state of the Mexican GDP, unemployment, public expending, taxes, incomes and informality at present. It is important to note that the section was not further reduced, as another reviewer requested that this section be expanded.
Comment 3: With regard to comment 5, the two relevant aspects have been introduced theoretically, but the question remains, why they are not reflected in the empirical analysis.
Response 3: In response to your observation, although informality is measured in terms of access to social security, the unemployment variable included in the model captures those people who are part of the labor force seeking employment and who are highly likely to work informally. Therefore, it is not considered a specific variable of social security. In the case of the Mexican economy, the behavior of informality in relation to social security is hampered by social programs that provide health services to people who do not have formal jobs. That is, people can move from one market to another without any restriction. Therefore, the theoretical framework of informality was deepened and expanded, from the study of the escape and exclusion approach, providing support and response to the issue of social security. This also allows for better linking the different types of public policy programs carried out by the government, in which it is seen that there are actions from both approaches, strengthening the analysis.
Comment 4: The solution to comment 6 does not solve the respective issue. If at all path-dependency is detected my auto-correlation tests not multicollinearity tests. Relevant changes to reflext the time-series nature of data, however, have not yet been considered in the empirical analysis itself and neither are any reasons discussed why this has not been done and what limitations result from omitting this type of analysis.
Response 4: In response to your comment, the Durbin Watson test for first-order autocorrelation was included in the model results to examine the independence of the residuals from the unique error terms associated with each observed variable (WA, TAX, GOB, UNE). These error terms represent the portion of each indicator that is not explained by the latent informality variable, besides that the model is not recursive and therefore decreases the effects of autocorrelation.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors
The authors of the manuscript, Informal Employment, the Tertiary Sector, and the Gross Domestic Product: A Structural Equations Model for the Mexican Economy, made a number of changes that helped revise the manuscript. I appreciate the editing of the Conclusions, which the authors have edited to the required level. Unfortunately I still remain of the opinion that the discussion should be edited and additional and more recent sources should be added.
Author Response
Comment 1: The authors of the manuscript, Informal employment, the tertiary sector and the gross domestic product: a structural equation model for the Mexican economy, made a series of changes that helped revise the manuscript. I appreciate the editing of the conclusions, which the authors have edited to the required level. Unfortunately, I still think the discussion needs to be edited and additional, more recent sources added.
Response 1: In response to your observation, the following changes were made:
- An exclusive section was integrated into the document, which reviews in-depth literature on the proposed topic and supports the choice of variables and their relationship with informality. In this, 15 new investigations were integrated into the document. Overall, the manuscript was 72 references, of which 50 investigations are no more than 10 years old, while 22 references are prior to 2014. The above reinforces the theoretical framework.
- The discussion section was modified, in which the results were reinforced with 16 more references, which confirm the results obtained by econometric analysis; 8 references are from 2019 onwards.
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
Comments and Suggestions for Authors
All existing issues have sufficiently been addressed by the authors.
