The Impact of Port Performance on Trade: The Case of Selected African States

Round 1

Reviewer 1 Report

Dear authors,

The first model of the article - Panel ARDL: The Impact of African Countries' Ports on International Trade is well grounded in the paper. The statistical sample size  is named: ten African countries, time series for the period from 2005 to 2018. The use of the ARDL model is justified. The endogenous variable and exogenous variables are well explained in the context of economic meaning.

The second model of the article is the Gravity Model: a model that describes the economic interaction between countries. The statistical sample size characterization needs a better description. The economic meaning of interaction between countries, resulting from the Gravity Model, has not been described.

The meaning of deriving the equation (line 305) from the equation (line 298) is clear, but why  was replaced by , and C by K is not clear.

Author Response

We acknowledge the reviewer’s comments and rectifications were made. A description of the economic meaning of interaction between countries, resulting from the Gravity Model, was added (line 319-323).

The equations in line 305 and line 298 were revisited. They second equation was addressed accordingly. An error had been made.  Tij was not supposed to be replaced and the  c (the constant) was also not supposed to be replaced by K.

Author Response File: Author Response.pdf

Reviewer 2 Report

Review of the paper titled „The Impact of Port Performance on Trade: The Case Of Selected African States” submitted to the Economies journal:

Major remarks:

• Perhaps the problem with GDP and its relation with trade (in results section, table 7), stems from potential endogeneity relationship between these variables? Authors could dwell on this issue.

Minor remarks:

• Author(s) could work on graphical side of the tables. Please consider using the font size comparable or smaller then the rest of the text. One could also consider adding a few cases to reach top-10 and bottom-10 countries in Table 1.
• Consider sorting values on Figure 1.
• Add missing variables and their descriptive statistics, including dependent vars to table 2.
• Line 375: “Results from the Pedroni test show that seven of the eleven statistics reject the reject the null hypothesis of no cointegration at the conventional size of 0.05” – consider rewriting the sentence.
• Line 388: “The study performed tow dynamic models” –> two dynamic models
• Line 444: two spaces between sentences?
• Line 346-7: “The result is insignificant as reflected by the p value 436 (0.1280) which is less than 0.05.” Shouldn’t it be here higher than 0.05?
• Line 516: consider breaking the paragraph.
• Correct the missing DOIs and improper links in the references section, e.g. https://doi.org/10.1007/s00168-019-00947-6 instead of https://doi.org/10.1007/s00168-019-00947-

Author Response

The study was cognizant of the endogeneity challenge that is faced when examining the relationship between trade on economic growth. The problems of endogeneity are related to both (1) the proposition of potential reverse causality between economic growth and trade and (2) common unobserved factors of influence that may give rise to positive-biased estimations. However, as shown by Rodriguez and Rodrik, the relationship between trade and growth may be difficult to determine due to methodological deficiencies and factors of influence that researchers do not consider (Rodriguez and Rodrik, 1999). Having this in mind, the study had to complement the ARDL with the Gravity model (GMM). This is in line with Bond et al (2001) who argue that among the various econometric techniques proposed to overcome these problems is the Arellano and Bond (1991) approach (GMM estimator) which has been applied to growth regressions by many analysts (Dollar and Kraay, 2003; Amable, 2000; Bond et al., 2001, among others). The main advantage of this estimator is that it does not require any external instrument to deal with endogeneity (Huchet et al. 2018). The GMM model showed a significant relationship between trade and GDP.

The tables and Figures were also reworked.

All typographical and grammatical errors that were highlighted by the reviewer were corrected.

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The review of the paper titled „The Impact Of Port Performance on Trade: The Case Of Selected African States” submitted to the Economies journal:

Major concerns:

• In line 37 after “Ports are a primary conduit of international trade and are central to growth of the global economy.” I would add: Seaports are also an important locational determinant of exporters’ location https://link.springer.com/content/pdf/10.1007/s00168-019-00947-6.pdf

Minor concerns:

• Please add keywords to the paper
• Line 83: “The paper is divided into six sections…” - please the text starting from this place to the next paragraph
• In table 4, there is a lack of one variable name
• Line 399: “results” should start with a capital letter. The same at line 417 with “furthermore,”
• Please tidy up the references section: text formatting and remove unnecessary spaces.

I recommend publishing the paper in the Economies journal after introducing these few amendments.

Author Response

Responses

We acknowledge the comments raised by the reviewer and we incorporated all these comments.

• We added the line: Seaports are also an important locational determinant of exporters’ location. We also cited the source of that information.
• We added keywords as suggested by the reviewer.
• We paragraph outlining the structure of the paper was put on its own.
• In table 4, we realised that we had made an error and we deleted the figures that did not have a variable
• All other grammatical errors were addressed accordingly.

Author Response File: Author Response.pdf

Reviewer 2 Report

• In the abstract and in the introduction, the added value / novelty of this study is not mentioned.
• The hypotheses to be studied are not presented.
• In the PWC reference, lines 68 and 200 are missing the reference to the year.
• PWC citation (2018) does not exist in the references.
• On line 272, the acronym PMG is missing.
• On line 290 there are a few more parentheses (..).
• In tables 5 and 6, the level of significance must be identified.
• In tables 4, 7, 8 and 10 there is no result for a significance level of 10%.
• The indication of the levels of significance in table 8 need to be extensively revised (Example: GDP => 5% **).
• It remains to introduce a section with discussion of results.
• Review the references and place the DOI.

Conclusion: the article needs to be thoroughly revised, especially the part of the methodology (more organized and well-founded) and the results (there are several errors in the classification of the levels of significance as mentioned above).

Author Response

Responses

• An explanation of where the study made an original contribution was added both to the abstract and also in the second last paragraph of section 1. This was done in line 89.
• The study put a research question instead of a hypothesis. This was done in line 82.
• The missing year was added.
• The PWC (2018) citation was added to the reference list
• The acronym PMG was defined. Line 264.
• The parentheses were deleted.
• The austreki for significance at 10% was deleted since no variable was significant at 10% significance level.
• The indication of the levels of significance in table 8 was revised. It was seen that an error had been made on the GDP variable.
• All other grammatical errors were also addressed.

Reviewer 3 Report

This study tries to address a big problem in economic development in Africa and addresses an interesting research question “does port performance affect trade?”. The author(s) apply an Autoregressive Distributed Lag, (ARDL) model in a panel dataset that covers 10 African countries for years 2010 to 2018. The first paragraphs were written rather in proper English language, but the starting from the literature review section onward the English language becomes poor. In general, I consider this paper as a work in progress as there are many issues apart from English language which need serious consideration. Therefore, it is too early to consider this work as finished scientific work. My main comments are as following:

1. The authors indicate in the literature review that one of the problems in port delays refers to the corruption and weak institution quality, but unfortunately this effect has not been considered in their analysis.
2. There are many abbreviations mentioned in the paper without clear definition for example; ARDL model, Table 1 column 4 “no”
3. It is confusing that in the abstract the authors indicate that their panel dataset includes 2010 to 2018 and then in the data section they mention a different time period (2005 to 2018). I could not find in the paper which one is the correct time period.
4. There are ambiguous claims for example, “An analysis by the PwC in 2018 revealed that a 25% upgrade in port efficiency may increase growth by 2%.” I could not understand to which growth it refers to. And another one “the study performed tow dynamic models”
5. Sorry to say, but funny results; “this may suggest that when inflation is increasing trade tend to increase also” and “there is no relationship between trade and economic growth”

Reviewer 4 Report

The paper presents an econometric estimation for Africa of the effects of several variables on trade, with special focus to port performance. The aim of the article is really interesting. The structure of the paper is correct, the methodology applied is also correct. Nevertheless, the paper presents important flaws which succinctly are going to be explained below.

2. The dependent variable (Trade) is incorrectly specified:
   1. It is defined as “the contribution of trade to GDP” (see Table 1). What does it mean? Exports/GDP? (Exports+Imports)/GDP? (Exports-Imports)/GDP? Imports/GDP? I assume it is Exports/GPD.
   2. Whatever the definition of the ratio, none can provide the numerical values described at column TR at Table 2.
   3. There is no theoretical nor empirical support to justify the relationship between this ratio (whatever the definition) and the Port Performance. E.g., USA has very low ratios and PP is high; Singapore has very high ratios and PP is high.
   4. The dependent variable should be real exports. The use of the ratio is an error that invalidates the analysis.
3. There is no sense including the variable D due to the presence of the richer variable PP.
4. Some variables can make sense in a gravity model (exchange rate, inflation), but the empirical results explained for them in section 5.5 are nonsense. Probably this is due to the flaw in the definition of the dependent variable.
5. The author/s explained several variables which it is said that are related to PP (e.g., country’s link to supply chains (line 51), country’s corruption (line 70), public investment (line 77), the transportation sector (line 419),…). They could be used to explain trade instead of less frequent variables such as inflation.
6. The presentation of the references is extremely poor:
   1. It is really difficult to check the quality and suitability of the references in the present format.
   2. The authors refer most of the papers through a weblink, not with the journal/working paper series/… information. This is not admissible.
   3. Additionally, many of the references seem superfluous and unnecessary all across the paper, and especially in the Introduction.
   4. Many references in the main body of the paper are not present in the References section (see references in lines 67, 81, 87, 102, 108, 129, 173, 197, 251, 252, 265, 269, 272, 279, 444,…). Other are in the References section but are not present in the main body of the paper (513, 522, 558, 569, 595, 598, 604, 607, 694). Other mistakes: Benamara et al. (2019) is included twice.; Humphreys et al. (2019) is wrongly located; UNCTAD (2018) must be defined as UNCTAD (2018a) and UNCTAD (2018b) or similar.
   5. A very deep revision of the presentation of the references must be conducted if the author/s is/are going to submit this article to a scientific journal.
7. Comments at lines 333-337 are absurd. What about the scale? What about the economic sense of each variable? The key is the regression coefficient, not the size of the value of the variable.
8. Line 431: There is a lot of empirical evidence about the link between imports and inflation.

8. Minor points:

1. 1. Table 1: Some countries are misspelled.
   2. It is more scientific to quote the original source. Avoid sentences such as lines 34, 135 or 265.

Round 2

Reviewer 2 Report

The recommendations were resolved by the authors.