Kindergarten Proximity and the Housing Market Price in Italy

Round 1

Reviewer 1 Report

The article "Kindergarten area and the price in the housing market in Italy" deals with an important topic: the real estate market. The authors have attempted to link the real estate market with the distance from the kindergarten. They also took into account the type of kindergarten (state and private), which shows that the latter neighbourhood generates a more significant capitalization effect. I think the topic is interesting and important. However, I have doubts about the selection of the sample. Work is biased because the presence of a kindergarten is essential for people who have children. However, the proximity to school may not be an asset for people who do not have children. So in both cases, the dependence should occur, but it should have a completely different direction. The authors did not specify how the sample was selected. The whole work is not very detailed and therefore raises my reservations.

Detailed comments:

[lines 79-84]: I would expect a more detailed description of the research carried out by other authors or the removal of the part that does not affect the quality of the work

[lines 161-165] I did not understand whether it was about municipalities or cities. This part of the work is unclear to me. Besides, I wonder how these cities were chosen at random? Have entire cities been surveyed? What is a micro-zona?

[verse 199-211]: Why do the authors consider that these indicators indicate the kindergarten's quality?

[lines 319-328] Do the column numbers in the table match the formula numbers? Placing this information in the text would greatly improve the work's readability.
I did not understand what it determines and what the measure contained in Table 2 for individual models, e.g. Min - 0.352. It would have to be better explained.
What test or hypothesis does the p-value in Table 2 refer to?
The quality of the models, measured by the R-squared coefficient, is no good idea because the different number of explanatory variables get a higher value. A better indicator would be an adjusted R-squared

[line 334-342] How was the impact of the proximity of the kindergarten on the price of the apartment found?

[lines 343-355] This part is incomprehensible to me. The authors write that all variables affect prices, which is not reflected in the results. This part is too generalized.

[lines 358-362] I was not convinced of it. This needs to be explained in more detail concerning the results.

[lines 426 -437] I do not know how the authors concluded that non-state kindergartens significantly impacted the price of flats. This part requires a more precise interpretation of the results obtained.

Author Response

The article "Kindergarten area and the price in the housing market in Italy" deals with an important topic: the real estate market. The authors have attempted to link the real estate market with the distance from the kindergarten. They also took into account the type of kindergarten (state and private), which shows that the latter neighbourhood generates a more significant capitalization effect. I think the topic is interesting and important. However, I have doubts about the selection of the sample. Work is biased because the presence of a kindergarten is essential for people who have children. However, the proximity to school may not be an asset for people who do not have children. So in both cases, the dependence should occur, but it should have a completely different direction. The authors did not specify how the sample was selected. The whole work is not very detailed and therefore raises my reservations.

 

We thank the referee for the detailed comments. Although we recognize the limitations of our work, we have tried to better clarify in the new version of the text how we mitigate the criticism that the Referee correctly raised. Regarding the selection of the sample, we recognize in the final remarks that this is a limitation of our analysis. However, we have also clarified that only the largest cities provide us with enough heterogeneity to identify the impact of kindergartens’ location on the housing market for two main reasons: first the presence of many facilities distributed across the cities; and second the possibility to measure housing prices at the level of the neighbourhood. Regarding the potential bias in our estimates, we recognize the impossibility to identify families with and without children in our dataset, since at the moment this kind of information is not available at the neighbourhood level and to collect it we should implement a specific survey that goes beyond the scope of this paper. However, we would remark that with the two stages procedure we have tried to mitigate this potential bias. We think that further research in the future may address these points.

Detailed comments:

 

[lines 79-84]: I would expect a more detailed description of the research carried out by other authors or the removal of the part that does not affect the quality of the work

 

Thanks for suggestion. We have removed the sentence and we begin with: “Up to now, only few studies have explored the relationship between housing market prices and school proximity despite this factor may affect house values as the attractiveness of a house increases with the proximity to a school, especially with school-aged children due to commuting and safety worries in the district (Huang and Hess, 2018).”

 

 [lines 161-165] I did not understand whether it was about municipalities or cities. This part of the work is unclear to me. Besides, I wonder how these cities were chosen at random? Have entire cities been surveyed? What is a micro-zona?

We thank the referee for the request of clarification. Now at the beginning of §3 we have included the following clarification. “Moreover, we restricted the analysis only to the largest cities because they provide enough heterogeneity to identify the impact of kindergartens’ location on the housing market for two main reasons: first the presence of many facilities distributed across the city; and second the possibility to measure housing prices at the level of the neighbourhood.”

[verse 199-211]: Why do the authors consider that these indicators indicate the kindergarten's quality?
We thank the referee for this request for clarification. In line with this request, we denominate all the variables regarding the kindergarten as indicators that allow us to identify the “structure of service” rather than the “quality”. However, we decided to retain also the name “quality of service” especially for those indicators such as the average class size, the average size in square meters of playgrounds etc., since they measure the “intensity” of the service. Instead, the level of the service is typically identified only by the presence of the kindergarten and by the number of pupils.

 

[lines 319-328] Do the column numbers in the table match the formula numbers? Placing this information in the text would greatly improve the work's readability.
I did not understand what it determines and what the measure contained in Table 2 for individual models, e.g. Min - 0.352. It would have to be better explained.
What test or hypothesis does the p-value in Table 2 refer to?
The quality of the models, measured by the R-squared coefficient, is no good idea because the different number of explanatory variables get a higher value. A better indicator would be an adjusted R-squared.

We thank the referee for this request of clarification. First of all we have described in more detail the structure of the results’ tables linking the tables to the equations. Moreover, in the footnote of all tables reporting the results of the estimates, we have clarified that: “Each line of the table refers to three different levels of housing prices per sq. meter: minimum, average and maximum. Each column refers to a different specification of the model in terms of control variables included among the regressors. Columns from 1 to 4 consider as dependent variables the raw price of housing, instead, columns from 5 to 8 consider as dependent variables the price of housing depurated from the neighbourhood effect. Different blocks of the table consider different time spans of the housing prices, from 2011-2012 average up to 2016-2017 average, finally, the last block refers to the average price over the entire period 2011-2017.” As requested we included Adjusted R-sq in all tables.

 

[line 334-342] How was the impact of the proximity of the kindergarten on the price of the apartment found?

 

Thanks. We have written: “The main results confirm the impact of the proximity of the kindergarten on the price of the house.”

 

[lines 343-355] This part is incomprehensible to me. The authors write that all variables affect prices, which is not reflected in the results. This part is too generalized.

 

 We have changed. Thanks.

“More in detail, the school proximity coefficient estimates suggest that, overall, close location to kindergarten has a significant and positive effect on housing price. As we expected, the capitalization effect becomes smaller after we depurate the housing price from the neighborhood effect but remains in most of the specification positive and statistically significant. In addition, comparing the results over time (2011-2017), we can observe a persistent capitalization effect. Specifically, the proximity of kindergarten to the house generates stronger capitalization considering the maximum housing value, a weaker impact is observed on minimum values, instead the impact on the average value is in between.”

 

[lines 358-362] I was not convinced of it. This needs to be explained in more detail concerning the results.

 

We have changed and we have added a footnote 13. Thanks.

 

“Adding the variables that capture the quality of schools (see Table A.3 in Appendix)[1], we find that several quality variables such as the presence of foreign pupils, people who take care of the disabled, canteen service, number of schools opened on Saturdays can positively impact housing market prices. All the considered variables impact on housing market prices even if they present a different magnitude. Foreign pupils' presence has the highest value and is equal to 1.225, while the lowest value, equivalent to 0.240, is for people who take care of the disabled. These results are signals that parent not only cares about the location but also about the quality of kindergarten. This result is in line with the work of Turnbull et al. (2017) which shows how parents search and then choose schools that offer specific and additional services to solve organizational and working problems.”

 

 “Tables A.3 in the Appendix contains the complete empirical results obtained when we consider as dependent variable the mean value of housing price during the period 2012-2013. We have chosen to focus on this period since it better explains the degree of capitalisation of kindergartens on housing prices with respect to other years.”

 

 

 

[lines 426 -437] I do not know how the authors concluded that non-state kindergartens significantly impacted the price of flats. This part requires a more precise interpretation of the results obtained.

 

Thanks. We have explained better this point. “These original results can be interpreted as evidence of the higher utility that non-state kindergartens provide to households with respect to state institutions. Our results, in fact, show that public schools have a more homogeneous distribution on the territory; on the contrary, private schools can have an asymmetrical dislocation and they are present in any location where it is deemed necessary to provide this facility.”

 

 

 

Reviewer 2 Report

The author(s) have conducted a detailed study on assessing the impact of kindergarten proximity on housing market prices in Italy. The study is well structured, and the findings are interesting.  That said, I would like to make minor comments for the consideration of the authors:

1. The theoretical and practical implications of the study's findings should be included in the abstract. These are missing in the current state of the manuscript. 

2. The author(s) are entreated to thoroughly read the manuscript again in order to correct some misspellings such as 'wat' instead of 'want' on page 5 line 194; 'ad' instead of 'at' etc. 

Author Response

The author(s) have conducted a detailed study on assessing the impact of kindergarten proximity on housing market prices in Italy. The study is well structured, and the findings are interesting.  That said, I would like to make minor comments for the consideration of the authors:

1.The theoretical and practical implications of the study's findings should be included in the abstract. These are missing in the current state of the manuscript. 

 

Thanks. In the abstract we have added the following sentence:” Finally, the empirical evidence could be useful to several actors involved in urban planning when developing plans for the construction of new kindergartens in order to create a more homogeneous city.”

 

2. The author(s) are entreated to thoroughly read the manuscript again in order to correct some misspellings such as 'wat' instead of 'want' on page 5 line 194; 'ad' instead of 'at' etc. 

 

Done. Thanks

 

Author Response File: Author Response.docx

Reviewer 3 Report

Тhe article deals with a very interesting topic, moreover it is well written and easy to read.   below are just a few suggestions in order to improve the article:  1) in the introduction section more emphasis should be given to the research questions; 2) also in the introduction section we need to mention what are ( always if any) the papers on kindergarten that have been performed with reference to the Italian context. 3) in the conclusion section I think it is appropriate to expand the part on the limitations of the work by better explaining how future research can fill the gap in the present work. 

Author Response

Тhe article deals with a very interesting topic, moreover it is well written and easy to read.   below are just a few suggestions in order to improve the article:  1) in the introduction section more emphasis should be given to the research questions; 2) also in the introduction section we need to mention what are (always if any) the papers on kindergarten that have been performed with reference to the Italian context.

 

 Thanks for suggestions. In the introduction we have explained the key objectives of the paper and we have specified that there are not research with reference to the Italian context:

 

“While the existing literature on schooling and house market prices has investigated the impact of school quality on housing price in numerous countries (Black, 1999; Downes and Zabel, 2002; Kane et al., 2003, 2006; Figlio and Lucas, 2004; Brasington and Donald, 2006;  Clapp et al., 2008; Gibbons and Machin, 2008; Nguyen-Hoang and Yinger, 2011; Machin, 2011; Gibbons et al., 2013; Livy, 2017; Yi et al., 2017; Towe and Tra, 2019; Turnbull et al., 2017; Turnbull and Zheng, 2019; Bonilla-Mejìa et al., 2019), there is no research focusing on the relationship between accessibility to kindergartens and housing price in Italy.”

 

 

3) in the conclusion section I think it is appropriate to expand the part on the limitations of the work by better explaining how future research can fill the gap in the present work.

 

Thanks for suggestion. We have added another limitation and re-written this part.

 

“Finally, the crucial caveat to be highlighted descends from: (i) the limited number of Italian Municipalities even if they present homogeneous characteristics and (ii) the nature of data employed that do not allow us to understand the different effects of kindergarten proximity between households with children and those without children. Thus, further research on this topic could be based on larger and more detailed datasets to go beyond the limitations of this current work.”

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thank you for the included corrections. Congratulations on your good work.