Social Security and Fighting Poverty in Tunisia
Round 1
Reviewer 1 Report
The description of the social security is by far too scarce. There is no information on criteria to be included in the system or not; is the system voluntary (thus exposed to adverse selection)? How is it financed? Do those included, who have higher income than those not included, pay premiums? Income related or a fixed rate?
I also notice that Nicholas Barr, the grand old man when it comes to social insurance, it not in the reference list. I think you need to study his textbook in order to become familiar with insurance and effects of different ways of organizing it: utility of insurance, adverse selection, moral hazard et c. Now there is a (too) long literature review of why people would like constant consumption even if income changes. I suggest that you shorten this part and instead discuss insurance. The wish to keep consumption constant, whatever hits you, follows from utility maximization. Theory of insurance shows why this is not possible or very expensive (see Kruse, A and A-C Ståhlberg: Welfare Economics. Theory, empirical evidence and the Swedish experience. Studentlittertur 2013, for a fuller explanation).
Complete versus partial risk sharing: I think you have to explain the meaning of these two concepts, not only by referring to (rather!) old articles. What about the Tunisian social security system? Is it organized so that you can choose degree of co-finance, i.e. partial insurance?
I take it that you explain/discuss the concepts by referring to the empirical results, i.e. β being closer to 0 or 1. In my view, a rather odd way; but anyhow you have to explain it.
Line 45: “ … the impact of social security … of household investment in human capital”. Is that really what you do? I do not find any variable showing human capital.
Section 3.1.2. Good presentation of the questionnaire. However, nothing on family size. Do you not have information on this? It most certainly influences consumption.
Lines 183-184: “ … GDP fluctuation … not an important role … consumption of a country”. Discuss whether this also goes for poor households.
When discussing the results of eq. 3: Hugh effect for those not being covered … and none for those being covered. Again: please explain the kind of cover.
(lines 309-310 are repeated once again in lines 313-314)
Apart from not clearly telling what the design of the insurance looks like, I find your discussion of your results clear and thorough. However, the last lines, 553-505, seem rather funny (sorry!). “The role of social security … important, but it remains insufficient especially for uncovered … “. Of course, if they are uncovered!!
Author Response
Answers to Reviewer 1:
1- Comment : The description of the social security is by far too scarce. There is no information on criteria to be included in the system or not; is the system voluntary (thus exposed to adverse selection)? How is it financed? Do those included, who have higher income than those not included, pay premiums? Income related or a fixed rate?
Answer: As you suggested, we include now a paragraph in the introduction just after line 35 describing social security in Tunisia.
2- Comment : I also notice that Nicholas Barr, the grand old man when it comes to social insurance, it not in the reference list. I think you need to study his textbook in order to become familiar with insurance and effects of different ways of organizing it: utility of insurance, adverse selection, moral hazard et c. Now there is a (too) long literature review of why people would like constant consumption even if income changes. I suggest that you shorten this part and instead discuss insurance. The wish to keep consumption constant, whatever hits you, follows from utility maximization. Theory of insurance shows why this is not possible or very expensive (see Kruse, A and A-C Ståhlberg: Welfare Economics. Theory, empirical evidence and the Swedish experience. Studentlittertur 2013, for a fuller explanation).
Answer: As you suggested, the reference is now included in the text. Kruse, A. and Ståhlberg, A-C. 2013. "Welfare Economics. Theory, Empirical Results and the Swedish Experience", Studentlitteratur, Lund. The following paragraph, "Kruse and Ståhlberg (2013) present various welfare programs that alleviate risks, which households can encounter such as not being able to support themselves due to sickness, unemployment or old age" , is include in literature review.
For a short time, it is difficult to study the book in details, but it will be done for future studies in the subject of welfare economics.
Literature review is reorganized by putting the accent on social insurance and risks.
3- Comment : Complete versus partial risk sharing: I think you have to explain the meaning of these two concepts, not only by referring to (rather!) old articles. What about the Tunisian social security system? Is it organized so that you can choose degree of co-finance, i.e. partial insurance?
I take it that you explain/discuss the concepts by referring to the empirical results, i.e. β being closer to 0 or 1. In my view, a rather odd way; but anyhow you have to explain it.
Answer: As you suggested, both concepts of complete risk sharing and partial risk sharing are explained in subsection 3.2.
4- Comment : Line 45: “ the impact of social security … of household investment in human capital”. Is that really, what you do? I do not find any variable showing human capital.
Answer: No, there is an error. The phrase is reformulated ; After a brief literature review, the empirical analysis uses data from households surveyed for the purposes of evaluating the impact of social security schemes on the strategies adopted by households to support themselves due to sickness, unemployment or old age shocks in Tunisia.
5- Comment : Section 3.1.2. Good presentation of the questionnaire. However, nothing on family size. Do you not have information on this? It most certainly influences consumption.
Answer: Some variables (such as family size and the head of family) are not included in the analysis due to their strong collinearity with the other variables. This phrase is included as a footnote and in the conclusion as a limit of the study.
6- Comment : Lines 183-184: “ … GDP fluctuation … not an important role … consumption of a country”. Discuss whether this also goes for poor households.
Answer: The following phrase is included after the lines 183-184: "In poor countries, the social security is weak or inexistent and the risks are present so frequently, especially in the case of unemployment, sickness, death, and weather shocks. For these reasons, social assistance between households is crucial for poor households."
7- Comment : When discussing the results of eq. 3: Huge effect for those not being covered … and none for those being covered. Again: please explain the kind of cover. (lines 309-310 are repeated once again in lines 313-314).
Answer: The huge effect for those not being covered is explained by a lower degree of consumption insurance and thus a higher vulnerability to income risk. However, for those being covered, there is a higher degree of consumption insurance and a lower degree of vulnerability to income risk. Social security is associated with a decreased vulnerability to risk for covered households. The households, who are covered by social security, are insured against sickness, death and old age.
8- Comment : (lines 309-310 are repeated once again in lines 313-314)
Answer: The repletion is deleted.
9- Comment: Apart from not clearly telling what the design of the insurance looks like, I find your discussion of your results clear and thorough. However, the last lines, 553-505, seem rather funny (sorry!). “The role of social security … important, but it remains insufficient especially for uncovered … “. Of course, if they are uncovered!!
Answer: The phrase is deleted from the text.
Reviewer 2 Report
This paper proposes an interesting research question that is framed within a unique dataset. Potential for a valuable contribution to the field of social welfare literature or inequality literature is present, but major changes are needed first.
This paper fails to connect relevant theory throughout the paper. I’m lost wondering if this is a policy paper, an inequality paper, a paper about economic risk and shocks. Additionally, no attempt is made to connect ideas issued in this paper with theories of the welfare state. Social security seems to be used as an umbrella term, but not one that is well defined.
The authors have not included enough background information for the reader to clearly see why assumptions about social security income is or is not helping households. Reorganizing the literature review into sections synthesizing research on poverty in Tunisia, the use of welfare policy to alleviate poverty or redistribute wealth across the population, and the effects of economic shocks to poor populations, needs to occur. I’m not sure if those are exactly the subsections that are needed, but they may serve as an example.
Tables have not been formatted consistently.
The conclusion lacks critical analysis of the data. Sure, public programs help sometimes. What does that matter? Again, I feel like this is where a theoretical argument needs to be addressed.
Author Response
Answers to Reviewer 2:
This paper proposes an interesting research question that is framed within a unique dataset. Potential for a valuable contribution to the field of social welfare literature or inequality literature is present, but major changes are needed first.
1- Comment: This paper fails to connect relevant theory throughout the paper. I’m lost wondering if this is a policy paper, an inequality paper, a paper about economic risk and shocks. Additionally, no attempt is made to connect ideas issued in this paper with theories of the welfare state. Social security seems to be used as an umbrella term, but not one that is well defined.
Answer: The paper is dealing with social security and insurance risks. The introduction is rewritten where social security in Tunisia is well described.
2- Comment: The authors have not included enough background information for the reader to clearly see why assumptions about social security income is or is not helping households. Reorganizing the literature review into sections synthesizing research on poverty in Tunisia, the use of welfare policy to alleviate poverty or redistribute wealth across the population, and the effects of economic shocks to poor populations, needs to occur. I’m not sure if those are exactly the subsections that are needed, but they may serve as an example.
Answer: Literature review is reorganized by putting the accent on social insurance and risks.
3- Comment: Tables have not been formatted consistently.
Answer: If the paper is accepted, the associate editor of the journal will format the tables.
4- Comment: The conclusion lacks critical analysis of the data. Sure, public programs help sometimes. What does that matter? Again, I feel like this is where a theoretical argument needs to be addressed.
Answer: We add at the end of the conclusion the following paragraph: "It is also clear that coverage against poverty cannot be clearly elucidated in this small sample because the questionnaire may have some limits since the households surveyed may hide some information that present themselves in a unfavorable situation. In addition, some variables such as family size are not included in the analysis due to the problem of multicollinearity.
Round 2
Reviewer 1 Report
Revised version
Lines 122-131 are added into the revised version. I do not find that it does any good to your paper.
Lines 139 and the following: I think you have to think a bit more on how you formulate the problem: You are discussing social insurance, i.e. insurance organised by the public sector/the state. With social insurance it is certainly possible to cover also aggregate shocks. Actually, this is one of social insurance’s raison d’être!!
Author Response
Answers to Reviewer 1:
1- Comment : Lines 122-131 are added into the revised version. I do not find that it does any good to your paper.
Answer: As you suggested, we delete lines 122-131.
2- Comment : Lines 139 and the following: I think you have to think a bit more on how you formulate the problem: You are discussing social insurance, i.e. insurance organized by the public sector/the state. With social insurance, it is certainly possible to cover also aggregate shocks. Actually, this is one of social insurance’s raison d’être!!
Answer: As you suggested, this paragraph is revised and modified. All modifications are done on the text.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
I feel that researchers have addressed my previous concerns. I think some additional clarity could be provided to the methods and results sections.
Author Response
Answers to Reviewer 2:
1- Comment: I feel that researchers have addressed my previous concerns.
Answer: Many thanks.
2- Comment: I think some additional clarity could be provided to the methods and results sections.
Answer: We tried to better clarify the methods and results. All modifications are done on the text.
Author Response File: Author Response.pdf
