Household Economics, Information Sources and Annuity Choices: Annuitisation Preferences of Members of the Slovak Private Pension Pillar

Round 1

Reviewer 1 Report

The article deals with an important and current topic. In many countries, citizens face a decision on how to properly, not to say optimally, distribute the accumulated private pension capital. In the face of the longer life expectancy and the associated risk of longevity, the literature states that annuities are the optimal way of disposing of them. However, I would argue whether this is the case at every level of accumulated capital. In my opinion, the small amount of capital allocated to the annuity is not essential for the pensioner's finances.

I highly appreciate the reliability of the study conducted by the authors. It is a very good and valuable contribution to the theory and practice of finance - especially regarding the elderly.

For my part, I suggest you consider investigating the following:

1. does the declared religion influence the decision about the annuity? I think that the family model and what we consider "family property" is connected with religion. This may affect the willingness or unwillingness to allocate capital to annuities, as opposed to inheritance.

2. And inheritance. Maybe it would be worth checking whether there is any level of inheritance after which the retiree feels that he has already secured his descendants (e.g. by handing over the house to his only son) and can distribute the accumulated capital to his retirement security. At the same time, he did not feel that he had not taken care of his offspring in terms of inheritance.

These are my two reflections. And I wish you success in your further work, because it is a very important topic, and you already have a lot of knowledge about it, and it is worth exploring.

Author Response

The article deals with an important and current topic. In many countries, citizens face a decision on how to properly, not to say optimally, distribute the accumulated private pension capital. In the face of the longer life expectancy and the associated risk of longevity, the literature states that annuities are the optimal way of disposing of them. However, I would argue whether this is the case at every level of accumulated capital. In my opinion, the small amount of capital allocated to the annuity is not essential for the pensioner's finances.

I highly appreciate the reliability of the study conducted by the authors. It is a very good and valuable contribution to the theory and practice of finance - especially regarding the elderly.

For my part, I suggest you consider investigating the following:

1. does the declared religion influence the decision about the annuity? I think that the family model and what we consider "family property" is connected with religion. This may affect the willingness or unwillingness to allocate capital to annuities, as opposed to inheritance.
2. And inheritance. Maybe it would be worth checking whether there is any level of inheritance after which the retiree feels that he has already secured his descendants (e.g. by handing over the house to his only son) and can distribute the accumulated capital to his retirement security. At the same time, he did not feel that he had not taken care of his offspring in terms of inheritance.

These are my two reflections. And I wish you success in your further work, because it is a very important topic, and you already have a lot of knowledge about it, and it is worth exploring.

Thank you for kind review and suggestions.

As for the declared religion influence, I believe, this is a misunderstanding. The paper considers impact of region rather than religion.

This research did not consider impact of inheritance on perceived retirement security, but I think this is an interesting idea. I added this idea to suggested directions for further research: ‘Qualitative research may also shed light on subtle details regarding relation between the bequest and survivor benefit motives on one hand and perceived security of retirement income’.

Reviewer 2 Report

This paper investigates the impact of socioeconomic and sociodemographic variables on annuity choices. The annuity puzzle is an interesting problem in financial studies.

1. The contributions of this paper should be argued by comparison with existing literature and will become persuasive.
2. What is the basis for the research hypothesis? Is it completely the same as the existing literature?
3. The data source and its processing methods should be described in detail.
4. The table format can be improved to be more beautiful.
5. Whether the empirical analysis results are robust？ The robustness test needs to be added.

Author Response

This paper investigates the impact of socioeconomic and sociodemographic variables on annuity choices. The annuity puzzle is an interesting problem in financial studies.

The contributions of this paper should be argued by comparison with existing literature and will become persuasive. What is the basis for the research hypothesis? Is it completely the same as the existing literature?

Thank you for the comments. The extensive literature survey is provided in section 1.2

The data source and its processing methods should be described in detail.

You are right to point that data source and its processing methods should be described in detail. The data section was expanded and following text added:

The poll agency has been active in Slovakia over 30 years. It implements both regular (omnibus) and ad hoc surveys. Unlike other poll agencies, this agency collects data on the face-to-face basis, rather than online forms. The online-based collection of data tends to filter out the elderly and/or less educated population groups. The agency maintains a network of pollsters in Slovakia. The pollsters, in turn, uphold contacts with selected members of public, as to provide for a stratified sample of the Slovak population. The stratified sample mirrors structure of the Slovak adult population in terms of gender, age, educational attainment, income, territorial distribution (NUTS 3 level) and settlement size. Sample of 600 individuals is representative of the working population aged 40-62. Network of pollsters is major advantage of the data collection process in case of demanding topics (such as annuity choices). The agency firstly discusses contents of the questionnaire draft with pollsters. It is important to ensure that pollsters understand all questions in the survey. The final questionnaire is produced once all issues are clarified. The pollster then approaches his/her respondents in persons and records their answers. The pollster explains any questions a respondent may have. Results of the survey are recorded and provided to clients in the Excel and SPSS files.

The table format can be improved to be more beautiful.

We assume that the journal provides standard editing for the text and tables in the final version of the paper.

Whether the empirical analysis results are robust？ The robustness test needs to be added.

Thank you for the comment. The robustness tests for overall results are provided in section 4.2. As for the individual variables, standard Wald statistics are provided.

Reviewer 3 Report

The paper discusses up-to-date and interesting topics related to voluntary retirement savings and annuity choices. The paper is well structured, the reasoning is clear. However, before publication, I would recommend the following:

1.       Introduction: The justification for the importance of the topic should be deepened. Furthermore, the main goal of the study is not clear. It should be clearly formulated. Also, the introduction should be a separate part of the study. There is no need to specify sections as: 1.1. Annuity preferences.

21.       Literature review: The authors present a broad overview of previous research on the determinants of annuity choices. They show who has dealt with similar research topics before and what the results of previous studies are.  From my point of view, this part lacks a theoretical foundation for the study. It would reinforce empirical analyses and policy implications.

Moreover:

a.       the literature review should constitute a separate part of the part (it should not be a part of the introduction).

b.      page 2, lines 80-82: the authors could briefly explain the content of Table 1. All symbols in the table require explanation (0, +, n/a).

32.       Methodological part:

-          The formulation of the hypotheses requires reconsideration. From my point of view they are too wide (too general) and should be formulated more precisely. For example, in the case of H1, marital status seems to influence the annuity choices; however, this factor seems to have an effect on the bequest motives but not on accumulation motives. Therefore, can we say that hypothesis H1 has been fully verified?

Furthermore, singles were more likely to opt for an annuity than married (with partner). These aspects were not fully captured by H1.

-          Information about the survey is very narrow. Sample selection should be explained in more detail. When was the study conducted? How was the research sample selected? When and how were the responses collected? How was the representativeness of the sample verified? The declaration, one statement that the sample is representative, is not sufficient.

-          The explanation of multinomial logistic regression should be part of the methodological section and should be expanded. In particular, the regression equation should be presented. Furthermore, the authors should indicate how the reliability of the results obtained was verified.

43.       Results: The clarity of presentation is sufficient, but could be improved:

-          This section presents results and their discussion, which was not shown in the title of this part. Another solution would be to distinguish two separate parts: Results; Discussion. It is also reasonable as in the case of accumulation motives, the results are only presented (there is no discussion on it).

-          overall results: to increase the clarity of presentation, please indicate which conclusions are related to Table 3, and which to Table 4.

-          p. 11, lines 377-379: this paragraph fits better to the section “overall results”

-          p. 11, Bequest motive….: It would be easier for the reader if the order of the factors discussed corresponded to their order in Table 4

-          Table 2: It is not clear why Cramer’s V is presented for Education; The title “main information resource” instead of “information resource” would be more informative.

Other comments:

4. Some conclusions are not fully consistent with the results (or their presentation is misleading). For example:

-          “Milder and medium-sized declines were generally associated with higher preferences for accumulation motives, while the severe ones linked to bequest motives (Hypothesis 5 confirmed)” (p.12). In Table 4: “people expecting medium-sized fall in their public pension were more likely to consider accumulation motive” (p. 11). In case of bequest motive, the variables “decrease in the Social Security pension 10%/ 25%” are negative and significant (these respondents are less likely to consider the bequest motive?)

-          “Furthermore, the research also confirmed that not all types of information resource are of equal importance for informed pension decisions. Professional advice provided by financial institutions and government bodies was associated with higher probabilities of annuity uptake” (p. 15) – In Table 4, this variable is statically insignificant.

-          “The use of informal channels, on the other hand, resulted in disproportional preference for accumulation and bequest motives” – This variable is statistically significant only for the bequest motive.

-          Some final conclusions require further explanation and links to research results, e.g. “Low-income earners, benefitting from solidarity transfers, should be discouraged from diverting their Social Security contributions to private pension funds”

5. I consider that the conclusions part of the work should be expanded. In particular, the conclusions at the end of the paper should be expanded showing the economic policy implications of the research results.

Author Response

The paper discusses up-to-date and interesting topics related to voluntary retirement savings and annuity choices. The paper is well structured, the reasoning is clear. However, before publication, I would recommend the following:

Introduction: The justification for the importance of the topic should be deepened. Furthermore, the main goal of the study is not clear. It should be clearly formulated. Also, the introduction should be a separate part of the study. There is no need to specify sections as: 1.1. Annuity preferences. Literature review: The authors present a broad overview of previous research on the determinants of annuity choices. They show who has dealt with similar research topics before and what the results of previous studies are.  From my point of view, this part lacks a theoretical foundation for the study. It would reinforce empirical analyses and policy implications.

The literature review should constitute a separate part of the part (it should not be a part of the introduction).

You are right to point to structure of the paper. The paper is empirical by its nature, but surely needs some theoretical introduction. We followed your advice and added some more discussion on the theory behind the annuity choices in chapter 1. The literature review was made a separate Chapter 2.

We also stated the main aim of the research: This research aims at shedding some light on rational and semi-rational motives for choices alternative to an annuity purchase.

Please see respective changes in the paper structure.

Page 2, lines 80-82: the authors could briefly explain the content of Table 1. All symbols in the table require explanation (0, +, n/a).

Thank you for spotting this, We added explanatory text to the Table footnote.

Methodological part:

The formulation of the hypotheses requires reconsideration. From my point of view they are too wide (too general) and should be formulated more precisely. For example, in the case of H1, marital status seems to influence the annuity choices; however, this factor seems to have an effect on the bequest motives but not on accumulation motives. Therefore, can we say that hypothesis H1 has been fully verified? Furthermore, singles were more likely to opt for an annuity than married (with partner). These aspects were not fully captured by H1.

This is really a good point. H1 and H2 were changed accordingly

Information about the survey is very narrow. Sample selection should be explained in more detail.

You are right to point that data source and its processing methods should be described in detail. The data section was expanded and following text added:

The poll agency has been active in Slovakia over 30 years. It implements both regular (omnibus) and ad hoc surveys. Unlike other poll agencies, this agency collects data on the face-to-face basis, rather than online forms. The online-based collection of data tends to filter out the elderly and/or less educated population groups. The agency maintains a network of pollsters in Slovakia. The pollsters, in turn, uphold contacts with selected members of public, as to provide for a stratified sample of the Slovak population. The stratified sample mirrors structure of the Slovak adult population in terms of gender, age, educational attainment, income, territorial distribution (NUTS 3 level) and settlement size. Sample of 600 individuals is representative of the working population aged 40-62. Network of pollsters is major advantage of the data collection process in case of demanding topics (such as annuity choices). The agency firstly discusses contents of the questionnaire draft with pollsters. It is important to ensure that pollsters understand all questions in the survey. The final questionnaire is produced once all issues are clarified. The pollster then approaches his/her respondents in persons and records their answers. The pollster explains any questions a respondent may have. Results of the survey are recorded and provided to clients in the Excel and SPSS files.

the explanation of multinomial logistic regression should be part of the methodological section and should be expanded. Furthermore, the authors should indicate how the reliability of the results obtained was verified.

We followed your advice and moved the explanation of multinomial logistic regression to the methodological section. The SPSS software produces standard tests for the MLR. We report these tests in the Table 3. We also discuss some test statistics in the Chapter 4.

Results: The clarity of presentation is sufficient, but could be improved:

This section presents results and their discussion, which was not shown in the title of this part. Another solution would be to distinguish two separate parts: Results; Discussion. It is also reasonable as in the case of accumulation motives, the results are only presented (there is no discussion on it).

Thank you for suggestion. We created a separate chapter ‘4.3 Discussion of accumulation and bequest motives’

Overall results: to increase the clarity of presentation, please indicate which conclusions are related to Table 3, and which to Table 4.

Thank you. We added respective references.

1. 11, lines 377-379: this paragraph fits better to the section “overall results”

Good point. We transferred the paragraph to previous section

1. 11, Bequest motive….: It would be easier for the reader if the order of the factors discussed corresponded to their order in Table 4

We have rearranged some factors accordingly.

 Table 2: It is not clear why Cramer’s V is presented for Education; The title “main information resource” instead of “information resource” would be more informative.

Thank you for spotting this. We corrected the text.

4. Some conclusions are not fully consistent with the results (or their presentation is misleading). For example:

Milder and medium-sized declines were generally associated with higher preferences for accumulation motives, while the severe ones linked to bequest motives (Hypothesis 5 confirmed)” (p.12). In Table 4: “people expecting medium-sized fall in their public pension were more likely to consider accumulation motive” (p. 11). In case of bequest motive, the variables “decrease in the Social Security pension 10%/ 25%” are negative and significant (these respondents are less likely to consider the beques motive?)

The text was corrected

Furthermore, the research also confirmed that not all types of information resource are of equal importance for informed pension decisions. Professional advice provided by financial institutions and government bodies was associated with higher probabilities of annuity uptake” (p. 15) – In Table 4, this variable is statically insignificant.

Thank you. The text was corrected.

The use of informal channels, on the other hand, resulted in disproportional preference for accumulation and bequest motives” – This variable is statistically significant only for the bequest motive.

The text was corrected.

 I consider that the conclusions part of the work should be expanded. In particular, the conclusions at the end of the paper should be expanded showing the economic policy implications of the research results.

Some final conclusions require further explanation and links to research results, e.g. “Low-income earners, benefitting from solidarity transfers, should be discouraged from diverting their Social Security contributions to private pension funds”

We refer to policy implications in the concluding part of the paper. The text was amended in following way:

Low-income earners, for example, benefit from solidarity transfers in the Social Security. They should be discouraged from diverting their Social Security contributions to private pension funds, as they lose some part of their life-long public pension, but save little with the private pension company.