Corporate Failure Risk Assessment for Knowledge-Intensive Services Using the Evidential Reasoning Approach

Round 1

Reviewer 1 Report

I thank you for the opportunity to review this paper, which presents the author's approach to assessing the corporate failure risk of knowledge-intensive industry using an evidence reasoning (ER) approach. The analysis of this area in the context of a specific industry (in this case, knowledge-intensive industry) is a very interesting idea. The issue of company failure prediction is very timely, and the literature in this area has grown tremendously in recent years. We observe a growing interest in scientific studies that propose new approaches (both theoretical and empirical) to the exploration of knowledge in this area.  Using the theory of enterprise risk management (ERM) is certainly a new look at this area of research.

However, in my opinion, the manuscript could be even stronger if the authors considered a few observations. Please, see my detailed comments below. Good luck with your research!

I have provided my comments as follows:

First of all (after reading the title and abstract), it is clear to me what is the purpose of this article.  The title is very informative. The abstract clearly indicates the purpose and key assumptions and findings. Both the title and the abstract, in my opinion, encourage the reader to read the article further.   

I also found an interesting and informative introduction.

We can read that "This paper focuses on corporate risk assessment by using various consequences or causal factors of failure to classify companies into different levels of failure risk."(v.57-58) However, I did not find information on how the author defines "Corporate failure risk". (see title). In my opinion, it is not clearly explained to the reader. I encourage you to clearly define how the author understands it. In my opinion, clarifying this concept will make it easier to understand the model presented.

I encourage you to specify your research questions at the end of this section to help the reader follow the logic of further considerations.

I consider Section 2 "Literature review" to be well presented. The considerations presented are interesting, logical, and supported by evidence from the literature. The author has included a critical assessment of the contribution of the cited studies, which indicates to the reader the theoretical basis for building the model presented in the following section.  My reservation, however, is the actuality of the cited research.  Throughout the article, the author cites 51 references (most of which are in this section). However, only 12 references are from the last 5 years with as many as 24 references older than 10 years. The problem of corporate failure is developing quite dynamically; therefore, I encourage the authors to complement their considerations with the latest scientific reports in this area. I recommend updating this section with works from the last 5 years.   I also encourage you not to use the symbols in title 2.2.

Section 3 Reserach methodology

Subsection 3.1

Please provide a rationale for why the years 2006-2019 were selected for analysis and why the Fame database was used. What is this database? Why was it considered reliable? It is not clear to the international reader.

The author writes v. 219-220 "In this paper, we link the concept of risk with bankruptcy." Where is this contribution of ERM theory ?

I encourage you to describe in more detail how the sample for the study was selected. It is not clear to me? The current description makes it impossible to replicate the study.

In Section 4.2, the author indicated (v. 580-581), "...in this study, 100 active companies and 50 in-liquidation companies are randomly selected to test the new model.” How were these companies selected? Why these proportions? This is not clear and, in my opinion, may have affected the reliability of the model. Explain this at the beginning of the section.

Sections 3.3. and 3.4

The rationale for the selection of variables and the description of how the inference was carried out in these sections were well presented and credible.

Section 4.2

I encourage you to change the title of Section 4.2. In this section, the author does not present a discussion of the results obtained but only the results. The title of the section may be misleading to the reader.

I encourage the creation of a separate "discussion" section, where the author would confront the obtained results with previous scientific reports in a similar field.

Finally, I encourage the authors to express their opinion: Can the model presented be adapted to assess the risk of failure in other industries?   To what extent do the authors consider the data sets used in the analysis to be specific to the knowledge-intensive industry ? I believe that such an opinion could enrich the "discussion" section.

I believe that this study adds new value to the already available literature on the subject.

In my opinion, after minor changes, the research note is suitable for publication in the journal ‘Journal of Risk and Financial Management’.

Once again - congratulations. I encourage the author to do further research in this area. I wish the author all the best.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,
Your article is written on a relevant topic.
However, it contains an unacceptable number of self-citation. Therefore, the review of methods and literature cannot be considered sufficient.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Authors,

Your paper regarding corporate failure (bankruptcy) prediction based on ER approach is an extremely interesting work. It wasn't an easy Sunday afternoon reading' It requires knowledge from multiple fields. It uses a calculus based technique to predict failure, based on some hierarchical attributes.  I'll not debate the usefulness of such models, - it is a very common research topic-, but still without general validity.  I hope at least in the case of knowledge-intensive organizations your model will have better results than the similar approaches. This, of course is yet to be tested.

I have the following proposals.

1. You should more emphasize why you decided to analyze knowledge-intensive organizations, and the model is expected to deliver the same results for other type of organizations.
2. For me it was not clear how this model can be applied in practice. It seems you intend to use it as a tool to alert investors about companies with high risks, and less as a tool for the top management of organization or authorities.  I think you assume, that a company with high risks should be avoided for investment purposes, (and your main goal to detect in time these organizations). Well, I think this is not that simple.  I'd like to recommend to explain how this model will become an useful tool, and for whom.

3. In the study, your definition company failure means a bankrupted firm, and this is equivalent to the status ‘in liquidation’. We can agree on, that the day of bankruptcy is the last day of the firm, and the last year before this date is interesting from your model's point of view. I don't know the legal procedures in UK, but according to my international experience  -under normal circumstances- the financial figures of a company that in the next 365 days will be bankrupt are extremely bad much earlier. Sometimes there are several legal steps -in order to protect the creditors- to avoid bankruptcy, and it is a many year agony till the company goes out of the business. On the other hand, -and this might be the case of the knowledge-intensive industry- the bankruptcy is almost sudden, when these mostly ICT services not fulfill the market expectations. I recommend to explain how the model can prove it's utility under these circumstances. In the first case, if you take the data 365 days before bankruptcy it is obvious that the company represents a high risk, in the second case, there is no sign at all about the problems.

   4. For me it was not clear how the hierarchy of the attributes are  decided, and how the financial variables are selected.  If I'm understanding well,  the macroeconomic and industry indicators are more important than the internal risks... You mention that industry indicators  are half important than macroeconomic indicators.  How you decide this? In my assumption, exactly the opposite should be set, when analyzing the risk level of an individual (!) organization (the economy might flourish, but at the same time a certain industry within it might be in fall, thus the organizations in this industry might have a higher risk score than organizations form other industries). I'd like to recommend to explain a little bit more how you set the grades and weights in IDS and in your model.

As final remark l think the manuscript sounds scientific, tries to use provide calculus based arguments for bankruptcy, it is challenging , rises several questions, in certain statements can be considered even provocative, but definitively is not... boring.

Success in upgrading it. Hopefully it will be published soon!

Best regards!

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

With reference to knowledge-intensive corporates in the UK, in this paper you propose a risk assessment model and imply evidence reasoning to assess the risk of failure.

Some suggestions for you:

1. The abstract section is cumbersome and unclear;
2. The contribution of your paper to the existing literature must be clearly stated in the introducition;
3. In my opinion, you should better explain to the reader what knowledge-intensive corporates are and what kind of risks they handle;
4. A more specific definition of ERM is needed, by adding references too;
5. In section 3.1 you have to deeply describe the dataset and add references;
6. You wrote " unlike in some papers 222
   (e.g. Bhattacharjee and Han, 2014; Tinoco and Wilson, 2013) that use some indicators to 223
   define financial distress, which may lead to an untrustworthy model" (lines 222-224). Justify your assertion.
7.  An extensive editing of English language an style required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Dear Authors,

Thank you for taking in account my suggestions, and also to answer to my remarks. The paper is much improved, some of the weaknesses are no longer present. Especially the scientific level and the methodology is better. Still, most of the remarks I made are valid even now. If you can please improve it. Consider the level of understanding of the average reader, and also, emphasize the utility of you work. Congrats!

Reviewer 4 Report

The authors revised the paper taking into account all the suggestions.

I think that know it is ready for publication, after revising english again.