The Influence of Sentiments of Economic Agents on Pedestrians and Vehicle Crossings along the US–Mexico Border
Round 1
Reviewer 1 Report
This paper carries out time series regression analyses of monthly border crossing totals from Mexico to the US, explaining them as a function of macroeconomic and sentiment variables. It first-differences the data to address non-stationarity and carries out several robustness checks including a set of rolling regressions.
On the whole, the paper is well-structured, uses suitable data and applies a reasonable methodology. I have some suggestions for ways in which it might be improved, but it is worth considering for publication.
In section 3 The selection of search terms for the BCS and Imm proxy variables is not very well described and comes across as sounding rather arbitrary. Also, you could consider including a simple measure of consistency for the terms included in BCS, such as Cronbach’s alpha, given that this proxy variable is based on several search terms. Finally, it would be better to start each of these paragraphs (i.e. at lines 224 and 234) by explaining what the proxies are trying to do and then describe how they were constructed, rather than the other way around.
Later in Section 3 you describe the other explanatory variables, and in Section 4 there is a very useful sub-section setting out briefly prior expectations for most of the coefficients. The passages in Section 3 are rather confusing, because you don’t really explain why variables such as exchange rates and gasoline prices are included. That information only comes in the following section. Maybe it would be worth moving up some of the explanatory material justifying inclusion of each of the right-hand-side variables to the first time they are mentioned.
In lines 425-428, 474-477 and 527-528 you compare coefficients across models, drawing inferences from their relative sizes. Are you sure these differences are statistically significant?
Lines 507-512 note that adding Imm removes the significance from the real exchange rate variable for Texas, and it suggests that this implies “that an anti-immigrant environment may affect crossings”. Actually, the implications subtler than that. It might imply that an anti-immigrant environment weakens the economic motive for crossing; in itself the moderation effect described here doesn’t imply anything about direct effects on the level of crossings.
Detailed comments
- In the abstract, “trends” should be capitalized in “Google Trends”
- In fig 1, it would be better to add Y-axis titles to make it easier to see which axis relates to which line.
- In the introduction, at line 51, and also later in the paper at line 249, you refer to the downturn in crossings as being “influenced by the sub-prime mortgage crisis”. First, you don’t justify that assertion, so maybe “was likely influenced” would be better. Second the “subprime mortgage crisis” was only one component of the downturn and probably not the one that deterred most of those crossing. A broader term like “global financial crisis” or “Great Recession” might be better for making this point.
- On line 70, the reference to investor sentiment seems out of place.
- On line 79, an apostrophe is needed in “decision-makers[‘] sentiments”
- In line 182, you should drop “the” in “the internet search volume data”
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper is well structured.
I suggest improving the introduction, highlighting the aim of the paper and the conclusions, maybe adding a discussion section.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors examine the impact of human sentiment on Mexico-US border traffic with consideration of the pedestrian and personal vehicle crossings. Overall, the selection of dependent and independent variables seems (with the caveats given in Section 4) adequate to the issue under study, and the data sources are reliable. However, the article has several serious shortcomings, which I list below:
- The title of the article is too broad in comparison with the presented research and should be changed, which may increase the number of potential readers. After such a change, the paper would indicate a very topical and debatable issue. I explain why this is so. Based on terms such as "migrants," "deportations," "border wall," and "border patrol," the authors have constructed a special sentiment index they call the Border Crossing Sentiment (BCS). However, it is actually a much narrower index that should be called Border Economic Migrant Sentiment (BEMS), since it is based only on terms related to population migration. For example, it omits other typically commercial terms mentioned in the article, such as hedonic shopping motivation related to the fun and pleasure experienced during cross-border shopping (lines 64-65). Those interested in legitimate cross-border trade, which can contribute to the development of cities on both sides of the border, are mainly interested in the economic conditions of trade, not in deportations, border walls, or border patrols. These latter concepts are central to illegal migrants. Fortunately, this deficiency is offset to some extent by economic independent variables. Accordingly, the title of the article should read as follows: "Influence of Sentiments of Economic Migrants Crossing the Mexico-US Border on Traffic on Pedestrian and Personal Vehicle Crossings." Now everything is in place and the basic elements of the analysis are properly related to each other.
- The line of reasoning of the authors is quite interesting, as they claim that since financial markets are under the influence of the sentiment of economic agents, perhaps the same can be said about the economic development of border regions. This is the basic economic hypothesis that is tested in the paper. However, the economic effects of this similarity are omitted. As it is known, waves of optimism and pessimism in financial markets can be predicted and practically used, as they can be used to develop appropriate investment strategies. On the other hand, in the case of the sentiments of people (economic migrants) crossing the Mexico-US border, some new regularities were identified, but their practical applications were not given.
- Accordingly, the article should be supplemented with a concise section on the practical applications of research results. It should be placed before the Conclusions section. I see the following three possible directions for the use of research results.
- Migration policy. Research shows that the United States, and partially Mexico, can pursue a more or less balanced migration policy by exerting a specific influence on some independent variables, such as decision-maker sentiments (BEMS and immigration variables) and the real exchange rate (to a lesser extent).
- Development of infrastructure at border crossings. As the data show a significant variability of the traffic intensity at crossings depending on the sentiment and the exchange rate, the infrastructure at the crossings should be flexible, and thus prepared for alternately increased and relatively small traffic. This thread can be further analyzed in the context of the effectiveness of the involvement of human and capital resources.
- One might be tempted to discuss the applicability of the model to trade balancing policies in both countries. The influence of economic authorities on the variables listed in point A could be used again.
- I was also thinking about the issue of information redundancy in the model, namely the lack of necessity (or possible necessity) of distinguishing the "immigration" variable, which is associated with an increase in anti-immigration sentiment in the USA. It may seem unnecessary, since we already have a BCS index (although it should be a BEMS index) based on such terms related to immigration as "migrants", "deportations", "border wall" and "border patrol". Probably all these terms have a strong anti-immigration overtone. Therefore, you should consider removing the variable "immigration". Perhaps the authors will come to the conclusion that the separate variable "immigration" should remain, but then they should prove that it does not cause information redundancy in the model.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
Thank you for the opportunity of reading and reviewing your interesting manuscript. It is a well structures, well written and adequate for the journal's scope. To further improving the quality and readability of the paper I have the following comments and suggestions:
1.please clarify the objectives of your work
2.please formulate the research hypotheses or at least define the research questions
3.please consider including a section called Discussion or at least include this part together with the Results section
4.please discuss the applications/managerial, policy and practical implications of your findings. This is particularly important in a journal called "Applied Sciences".
Good luck!
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 5 Report
In the assessment of the paper submitted for the review, I specifically focussed on the discussed issues, applied research methods and the scope of analysis of research results, as well as substantive content of the article and its structure.
The considerations conducted in the paper are focused on such categories as: border crossings, sentiments, personal vehicles, Mexico, U.S; Google trends.
The subject area discussed in the paper is important .
The research procedure has complex character.
The article, however, needs to be refined to be published.
Therefore, it is specifically recommended to:
- provide in-depth and expanded discussion of results,
- deepen the conclusions,
- develop the description concerning research limitations and trends for future research.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The amended text represents a much better scientific level. The authors introduced a more precise definition of sentiment, following the guidelines in the previous review. The research hypothesis was also clarified and a new subchapter was added, which suggested some practical applications of the discoveries made. It was also possible to clarify the controversy related to the “immigration” variable, which is associated with an increase in anti-immigration sentiment in the US. However, there are still opportunities to improve the article.
-
The first issue concerns the title of the work, which does not quite accurately reflect what is in the text. The first version talked about people's sentiment, while in the new version only sentiment was left. This is an imperfection that could affect the readers' perception of your article. The key and decisive issue should be the source of the data used in the article, i.e., the answer to the question of whose sentiment is being tested. The content of the article shows that the authors mean the sentiment of people crossing the border, and therefore it is about economic migrants. This matter should be clarified, as the correct wording of the title can have a large impact on the interest in this paper, which will certainly be reflected in the number of citations. I would not avoid including the phrase "Sentiments of Economic Migrants" in the title of the article. The issue of migrants is a pressing problem all over the world today, both in the US and in Europe. Moreover, this issue is highly controversial, some are in favor of policies beneficial to immigrants, others are not, and still others do not have a crystallized opinion on it. However, the article is balanced, the authors do not choose any of the sides and interpret only the research results, so there is nothing to be afraid of. For this reason, an article with migrants in the title may attract a lot of attention. Accordingly, it would be advisable to go back to the title I suggested in a previous review. Without it, the new title is inadequate to the research presented in the paper.
- The second question is how the authors distinguish between legal and illegal border crossings. This problem did not exist in the first draft of the article and unexpectedly appeared in the second draft. In the model used, there is no reference to any independent variable related to a legal border crossing, so we do not know whether the tested sentiment concerns people legally crossing the border or illegally. This is confirmed by the authors themselves, who write that the search for phrases using the Google Trends tool contains "terms associated with both legal and illegal migration" (lines 240, new version). Therefore, all references to legality or illegality should be removed from the text (as well as from the title of the work). The authors deal with economic migration in general, both legal and illegal.
After these defects are removed, the article may be published.
Author Response
Comments and Suggestions for Authors
COMMENTS FROM REVIEWER
The amended text represents a much better scientific level. The authors introduced a more precise definition of sentiment, following the guidelines in the previous review. The research hypothesis was also clarified and a new subchapter was added, which suggested some practical applications of the discoveries made. It was also possible to clarify the controversy related to the “immigration” variable, which is associated with an increase in anti-immigration sentiment in the US. However, there are still opportunities to improve the article.
- The first issue concerns the title of the work, which does not quite accurately reflect what is in the text. The first version talked about people's sentiment, while in the new version only sentiment was left. This is an imperfection that could affect the readers' perception of your article. The key and decisive issue should be the source of the data used in the article, i.e., the answer to the question of whose sentiment is being tested. The content of the article shows that the authors mean the sentiment of people crossing the border, and therefore it is about economic migrants. This matter should be clarified, as the correct wording of the title can have a large impact on the interest in this paper, which will certainly be reflected in the number of citations. I would not avoid including the phrase "Sentiments of Economic Migrants" in the title of the article. The issue of migrants is a pressing problem all over the world today, both in the US and in Europe. Moreover, this issue is highly controversial, some are in favor of policies beneficial to immigrants, others are not, and still others do not have a crystallized opinion on it. However, the article is balanced, the authors do not choose any of the sides and interpret only the research results, so there is nothing to be afraid of. For this reason, an article with migrants in the title may attract a lot of attention. Accordingly, it would be advisable to go back to the title I suggested in a previous review. Without it, the new title is inadequate to the research presented in the paper.
Following your suggestions: we propose to change the title to: “The Influence of Sentiments on Economic Agents on Pedestrians and Vehicle Crossings along the U.S.-Mexico Border”.
Initially, the paper’s title included “people’s sentiments”, since we can only restrict Google Trend searches by geography (i.e., Mexico) and timing criteria (i.e., our sample). And following the literature, it can be argued that Google Trends’ information can be related to people online behavior at the aggregate level. We understand the relevance and popularity of migration issues, however we believe that using the term “Economic Migrants” could be misleading to the reader of the paper. This considering that we are dealing with legal crossings of economic agents (e.g., shopping-trips travelers) and not those of permanent or temporal migrants. To be clear, there is no formal accounting of the migration of people crossing the U.S. Mexican border by land: pedestrians and vehicles.
2. The second question is how the authors distinguish between legal and illegal border crossings. This problem did not exist in the first draft of the article and unexpectedly appeared in the second draft. In the model used, there is no reference to any independent variable related to a legal border crossing, so we do not know whether the tested sentiment concerns people legally crossing the border or illegally. This is confirmed by the authors themselves, who write that the search for phrases using the Google Trends tool contains "terms associated with both legal and illegal migration" (lines 240, new version). Therefore, all references to legality or illegality should be removed from the text (as well as from the title of the work). The authors deal with economic migration in general, both legal and illegal.
After these defects are removed, the article may be published.
Following our previous observation, our Google Trend searches neither distinguish between the different motives that individuals possess to cross the border nor the type of individual (legal or illegal crossings) that perform such searches. Therefore, we have adjusted the manuscript to prevent a misunderstanding regarding the analysis of illegal crossing, which is outside our analysis scope.
Submission Date
26 October 2021
Date of this review
21 Dec 2021 11:55:54
Author Response File: Author Response.pdf
