Examining the Shifting Dynamics of the Beveridge Curve in the Turkish Labor Market during Crises

Round 1

Reviewer 1 Report

Dear Authors,

The article presents an interesting research problem. The research assumptions were properly formulated and then verified. I do not comment on the article's assumptions and methodology.  The Figures are prepared appropriately. However, the authors should expand the content in terms of:

- the literature review should be in-depth, also including international literature,

- a discussion that should be more in-depth in the context of the presented research and a look also in a global context,

- conclusions that are too general and much too short should be deepened, also in the context of the research problem of the article,

- point out research limitations and further research challenges. This can be included in the ending.

Moreover, in editorial matters, a dot should not be placed at the end of the title, this also applies to subheadings, where it should be unified.

Author Response

We are grateful to the reviewer for sparing time and giving comments and suggestions which have really improved the manuscript. Following are the responses to the reviewer's comments and in the revised manuscript the changes are made in a different (red) font color.

1. The literature review has been extensively adjusted to capture global studies accordingly.
2. The discussion section is adjusted to have in-depth analysis of the problem. In addition, the discussion in the overview of the Turkish labour market adds insights to the discussion.
3. The review regarding editorial issues, conclusion and direction of further research has been revised.
4. The conclusion is adjusted accordingly and limitations particularly on data issues and direction for further research are discussed accordingly based on recommendation of reviewer. 

Reviewer 2 Report

Very few typos, e.g. in line 8, please "exhibits". Language of article is fine.

In my opinion the article should be reorganised and explain better the theoretical labor economics background which is hard to understand in the article without prior knowledge of labor market models. Issues such as (comparative) labour market regulation and product market regulation or e.g. insider-outsider- or hysteresis-problems are not really included.

Only on p. 8 (line 280) the concept of the Beveridge curve is described in some detail, but still too brief. This central concept should be described much earlier and also with the support of the basic diagrams to explain the theoretical concept. Also potential measurement problems especially of vacancy-rates are hardly discussed.

A sentence such as "The BC, which indicates labour market efficiency by illustrating the relationship between unemployment and vacancies" (line 50 f.) is in my opinion not really clear to readers that do not know the BC concept already before in detail - explaining a bit more on this only several pages later is no good idea.

In a comparative perspective, it is also unclear what "high employment levels" (line 139 f.) really means, the discussion of unemployment rates on the same page is vague (a graph could help). Saying that "the government implemented measures to reduce unemployment..." (line 153) does not tell a lot, because it remains unclear by which measures, how lastingly and whether the measures adopted addressed the true causes of unemployment (and perhaps also a generally comparatively low OFFICIAL employment level compared to many other nations - line 172). In line 159 what is meant most likely both "demand for AND SUPPLY OF Labour". 

In figure 1 about "Outcomes" one wonders why in panel f "Minimum wages" are included, as this is a policy measure and not an outcome.

Instead of discussing the central concept of the article early in some detail, the article repeatedly from the beginning makes very general quotations of international institutions' reports without offering page numbers. Without a clear theoretical idea of the labour market concept in the article (e.g. natural rate of unemployment NRU, NAIRU-models etc.) it remains unclear, what "functioning of the labor market" really means. Is e.g. an increased worker transition (line 32) always a problem in the longer term or only in the shorter term. A functioning labour market could be e.g. understood as one where NRU is not lastingly rising (or the causes of this are theoretically clarified, e.g. in the comparatively easy models created by noble prize winner Edmund Phelps - other models are also possible; it appears to me, however, that such a theoretical basis is missing in the article because the policy implications by the authors appear to be missing such theoretical foundations and remain vague).

Lines 401 ff. demonstrate again that only a very basic understanding of labor markets is guiding the authors since it is usually not the question whether higher growth creates jobs but more whether the job creation is persistent or only temporary etc.

The mechanisms why young persons have a particular jobs problem remain also very unclear (even though this problem is well known in many other labour markets e.g. Spain). Causes of this are however not discussed in the article. This is also one key explanation why the policy implications are very vage in lines 480 ff.

In other words, though perhaps the use of statistical methods/econometrics appears mathematically fine in the article (disregarding perhaps huge measurement problems of some variables), overall the theoretical basis used by the authors should really be improved to avoid that the article may attract other articles discussing the problems mentioned above. 

Author Response

We are thankful to the reviewer for sparing time and giving comments and suggestions which have really improved the manuscript. Following are the responses to the reviewer's comments and in the revised manuscript the changes are made in a different (red) font color.

1. The concept of BC is discussed in the introduction section and in other part of the study according to reviewer’s suggestion.
2. The potential measurement issues, particularly with vacancy rate is discussed in the methodology section.
3. The discussion of measures has been addressed with specific examples.
4. We have also added specific policy measures to guide readers comprehension in the context of Turkish labour market.
5. Some editorial issues are also addressed to convey clarity as identified by reviewer.

Reviewer 3 Report

The objective of the paper is to analyse potential shifts in the Beveridge curve for Turkey during the global financial crisis and the Covid-19 pandemics. With this aim, the authors specify and estimate an econometric model relating unemployment rates with vacancies and other control factors using data covering the period comprised between 2005 and 2021. The authors conclude that the two considered crisis didn't have any significant effect on the unemployment rate-vacancies dynamics.

While the paper is well structured, well written and the topic is well motivated, there are several caveats related to the econometric analysis that prevent me to recommend publication at the current stage.

1. The paper lacks a proper description of data sources and variable definitions that would allow the reproducibility of the research. For instance, are data adjusted for seasonality? Have you checked the robustness of the results to the use of quarterly vs monthly data?

2. Results of cointegration tests that would allow expressing the Beveridge curve in the form of an error correction model are not adequately presented in the paper.

3. Figure 2 can be clearly improved. For instance, separate periods into two panels. The same applies to tables 3 and 4 where additional information should be provided (i.e., number of observations, Adj R2, etc)

4. In order to identify potential shifts in Beveridge curve and its drivers, recent papers such as Bonthuis et al (2016) -cited by the authors- apply a two step procedure that should be also appropriate in the context of this paper. Please consider this possibility or explain why you deviate from the literature and what are the advantages of the approach in your paper.

5. Literature review is clearly incomplete. Please include Bova et al (2018) and Destefanis et al (2020) and the literature cited there.

6.  Minor issues: There are two "table 1". The paper by Blanchardr et al (2013) is not in the proper position in the reference list.

References

Bonthuis, B., Jarvis, V. & Vanhala, J. (2016) Shifts in euro area Beveridge curves and their determinants. IZA J Labor Policy 5, 20. https://doi.org/10.1186/s40173-016-0076-7

Bova, E., Tovar Jalles, J. & Kolerus, C. (2018), Shifting the Beveridge curve: What affects labour market matching?. International Labour Review, 157: 267-306. https://doi.org/10.1111/ilr.12046

Destefanis, S., Fragetta, M., Mastromatteo, G. et al. (2020) The Beveridge curve in the OECD before and after the great recession. Eurasian Econ Rev 10, 411–436. https://doi.org/10.1007/s40822-019-00140-2

Author Response

We are grateful to the reviewer for sparing time and giving comments and suggestions which have really improved the manuscript. Following are the responses to the reviewer's comments and in the revised manuscript the changes are made in a different (red) font color.

1. The description of data issue has been addressed. All data are seasonally adjusted according to their sources. The study is limited by the availability of extensive data for the Turkish labour market for it to allow robustness checks based on frequency. This is not possible.
2. The bound test performs the function of the cointegration test. We have included it in the study.
3. Separating the figure based on period is not feasible. It will erode the basis for comparison. Having the curve in a single figure makes it easy for comparison and gives a clear picture of the shift and at what point rather than disaggregated.
4. Despite the advantage of using 2 approaches by Bonthius et al (2016) in their paper, the second approach has its weakness in assuming that shocks are orthogonal and response to a shock is constant overtime. This is not the reality in Turkey and may limit the accuracy of our result.
5. Literature review has been updated as suggested.

Round 2

Reviewer 1 Report

The authors took into account the comments formulated in the review. The article has been supplemented with the required content.

Reviewer 2 Report

I looked through the corrected manuscript of  Examining the Shifting Dynamics of the Beveridge Curve in the Turkish
Labour Market during Crises.

In my opinion, the revisions are sufficient in order to allow the publication of the manuscript now.

Reviewer 3 Report

Thank you very much for your additional efforts in response to my suggestions and comments to the previous version. I am satisfied by the way you have revised the document and I am now happy to recommend publication.