-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Charitable Giving, Tax Reform and Political Trust #45
Comments
A list of feedbacks1/10に佐々木周作先生と1/12に大竹文雄先生・佐々木勝先生のゼミで現状の結果を報告し、いくつかのフィードバックを得ました。ここにその一覧をまとめておきます。 この研究の売り方
租税改革について
分析について
|
@daraterkim の感想(?)この研究の売り方 1)に対して、以下に自分の考えを共有します。 1.「寄付の租税政策が有効に機能するのは政治家への信頼があるときか?」というQuestionで売れるのでは Table. NasTabの政治家に対する信頼 (2015からスタート)
注目する年度は、2018年度ですが、政治家に対しての信頼が非常に下がりました。 2014年 4月16日: セウォル号船の転覆事件-朴槿恵(パク・クンヘ)大統領の指導力に疑い (寄付金の税額控除を開始した年度) 個人的には、上に書いてる政治的な重要な点を用いて、加藤さんの分析結果を説明できたら面白いかなと思ってます。 租税改革について
分析について 3.個人的に面白いコメントでした。まだ別の論文になる話しですが、寄付行動だけではなく、税に対する意識などどう変化したかな...! |
後藤です。加藤くん、金さんありがとうございます。 加藤くんから話は伺っていて、加藤くんのコメントには異論なく、個人的には加藤くんの方向性で行くのがいいかと思います。
をみるかぎり、価格弾力性の話を避けなければならないのかと心配しています。 |
「tax deductionからtax benefitへの租税改革の目的をクリアにするべき」というコメントは佐々木周作先生からのものです。 あと、個人的に「税金が最上の寄付」という言葉は今回の研究のResearch Questionに深く関わっていると思います。 個人的には、価格弾力性という指標を用いて、「税金が最上の寄付」と考えている人は租税改革にどのように反応したのかというQuestionが一番いいかなと思っています。そのために、租税改革を用いて価格弾力性をpreciseに推定したことを強調して、その上でこのQuestionに答える分析をしたという流れになるかと思います。 乱雑な文章で申し訳ないです。 |
@gt510244 、@KatoPachi さん、いつも素晴らしいです。
そのところが、自分のメイン研究と完全に同じですが、全体的には「「税金が最上の寄付」と考えている人は租税改革にどのように反応したのか」が、メインのquestionなら、ぜんぜん別の話なので大丈夫です! 自分の前に書いた論文 |
Dear Mr. Kim: I have received a report from my advisors on your manuscript, "The price elasticity of charitable contributions in a tax-benefit system". With regret, I must inform you that, based on the advice received, your manuscript cannot be accepted for publication in International Tax and Public Finance. The main reason for this is that there has been no attempt to address the endogeneity of the tax price. While the estimates appear within reasonable ranges across quantiles I do not have any confidence in method. The tax price is a function of after-tax income and thus donations under the tax deduction regime. Under the tax credit regime, the tax price is not likely to be endogenous. However, with both regimes considered in a single equation, the bias from the deduction period affects all estimates. It would be interesting to study the tax price elasticity of donations under the tax credit system in isolation. However, this would be a different paper. But this is not what you have done. Attached, please find the reviewer comments for your perusal. I would like to thank you very much for forwarding your manuscript to us for consideration and wish you every success in finding an alternative place of publication. Sincerely yours,
Comments for the Author: Reviewer #1: Review of MS D1900071: This is an interesting paper asking an interesting question and applying a novel approach. The author estimates the tax price elasticity of charitable contributions through two different tax-price settings, within the same country (South Korea), providing an answer to the question: is there a difference between the tax price elasticity of donations under a tax deduction and a tax credit? The author analyzes panel data from 2007 to 2016, with the switch from a tax deduction to a tax credit taking place in 2014. Estimates of the tax price elasticity range from -1 to -4. Comments:
|
Dear Dr. Kim, Thank you for submitting your manuscript to the Journal of the Japanese and International Economies. The comments of the referee are included below. The referee points out several serious problems in the current version of the manuscript, which has convinced me that we are unable to accept your manuscript for publication in the Journal of the Japanese and International Economies. While you may be disappointed by this decision, we hope that the thoughtful comments will help you in your future studies. The quality of submissions for the Journal of the Japanese and International Economies has improved substantially in the last several years. As a result, we often have to reject pretty good papers. I would like to urge you to continue to consider the Journal of the Japanese and International Economies for publication of future manuscripts. Sincerely, Shin-ichi Fukuda Editor in Chief Journal of The Japanese and International Economies Reviewers' comments: This paper calculated the effects of different tax incentives for charitable contributions using South Korean household level panel data over the period from 2007 to 2016, applying the censored quantile regression method. The topic of this paper is important. However, I have four major concerns about the current version of the paper. First I have a concern about the log-log regression model in page 7. It is a model in which the marginal tax rate determines the total amount of charitable contributions. It looks odd to me because the marginal tax rate should affect the marginal behavior of the taxpayer. For example, consider the case where the taxpayer's marginal tax rate was 35% before giving charitable contributions but 25% after giving the contributions. In this case, both 35% and 25% should affect the amount of charitable contributions for the taxpayer. However, the model seems to assume that only the marginal tax rate after giving charitable contributions affects the amount of charitable contributions. In addition to the above conceptual problem, the author should have explained the log-log regression model more in details. I could find no explicit definition of Di[tax deduction] and Di[tax credit] in the paper. What is the expected sign of 1 and 2? Since i lies between zero and one, log[1-i] should be negative. On the other hand, logYi is positive if Yi is greater than one but negative if Yi is less than one. This means that the sign of 1 and 2 depends on the unit of Yi. If the unit is KRW, 1 and 2 should be negative. However, if the unit is trillion KRW, 1 and 2 should be positive. The author should have given extensive discussions on the expected sign of 1 and 2 before estimating the model. Second I could not understand how the author calculated i. In page 7, the author stated that "i has been calculated in accordance with the nonlinear marginal income tax rate, which is summarized in Fig. 1". However, the author should have explained the calculation method more precisely. I believe that it was based on the NasTaB panel data. If this is the case, the author should have explained which variables were downloaded from the data set and how these variables were transformed to derive i. We need to see the exact formula to derive i. Otherwise, we cannot evaluate whether the methodology is appropriate or not. Third the author should report the coefficient estimates of the censored quantile regression of all quantiles. In the current version, Table 4 only summarizes the coefficient estimates of the upper 0.7 quantile for the whole sample. To the extent that the censored model is correctly specified, not only the estimates of upper quantiles but also those of lower quantiles would have useful information on the taxpayers' behavior. Discussions based only on the estimates of upper quantiles might not be sufficient to derive important policy lessons. Finally I could not see what new implications this paper derived. In the conclusion, the author stated that "(t)he estimated elasticities, between -4 and -1, are in the higher range of elasticities compared to results from previous studies" and that the estimated elasticities of large donors are "consistent with results from the US." Does this mean that the paper just simply confirmed the previous estimates in the other countries and had no additional implication? If this is the case, the contribution of this paper would be limited. In addition, the author never discussed implications for Japan. This was embarrassing because JJIE is a journal that specializes in the Japanese economy. |
加藤くん このコメント非常に興味深いと思いました。
これをRQに据えるのはいいかもしれないですね。面白いと思います。 金さん 素早い返信ありがとうございます。自分の研究とバッティングする内容の研究になりそうであれば上手く回避するように具体的な部分を伝えていただくのがいいかと思います。 |
ここまでの議論で必要な追加分析をまとめました。
他に必要なことはありますか? |
とりあえずはこれでいいのではないでしょうか。追加で何かあればまたお伝えします。 |
後藤です。気になったので、今まで出ている結果を既存の理論研究からみてどう解釈できるかを考えてみました。 寄付の価格弾力性についてみた、Saez (2004)では寄付の価格弾力性が
となっています。(注意:原論文では価格が下がったときにいくら寄付が増えるかを考えているので符号は逆。) この結果を分析結果に当てはめると、
ということになるかと思います。 また、Saez (2004)の(22)式では、政府が公共財を供給する価格と寄付で直接公共財を供給するときの価格を比べて、政府が非効率に公共財を供給する場合に弾力性がどうなるかということも調べています。ここでは
というが示されています。これは、政府が公共財を効率的に供給するのであれば、政府が供給する場合と比べて1単位の寄付の費用が高くなるので、寄付価格が安くならない限り個人が寄付をするのは合理的でないので、個人がより寄付税制に敏感になるためです。 この結果は「政府が効率的であると思っているひとの寄付の価格弾力性が低い」という結果とは全く異なる結果だと思われるので、Saez (2004)の考えているような価格弾力性とクラウドアウトの枠組みでは、なかなか我々の分析結果は解釈しにくそうです。 なので、いま出ている実証の分析結果から「政府への信頼度が低い人は、(寄付による公共財供給に比べて)政府(による公共財供給)の効率性が低いと思っている」という結論を導くのは理論的なサポートが得られないのではないかと思います。 ですので、実証の分析結果についてあまり理論分析の解釈と結びつけた解釈はやめておいたほうがいいのかなと思います。 これらの点について、またdiscussionできればと思います。 |
この話ですが、決していまの研究の方向性に冷や水を浴びせるというわけではなくて、こうした議論を踏まえて方向性を検討できればいいという前向きな話だと思っているので、是非意見がほしいです。 また、いまの分析ではIntensive marginとExtensive marginが区別されていないので、区別すると、違う結果が出てくる可能性があります。Almunia et al.(2020)では、低所得者層のほうがExtensive marginでの弾力性が弾力的、Totalでみると低所得者層のほうが弾力性が弾力的という結果が出ています。 個人的にはSaez (2004)のような枠組みを考えずに「政府への信頼度が低い人は寄付価格がどうであろうが寄付行動を変えないので(寄付をし続けるので)非弾力的」「政府への信頼度が高い人は寄付価格が変化すると寄付行動を変える(政府の政策誘導にそって行動する)ので弾力的」と解釈していいと思います。 |
@gt510244 さん、コメントありがとうございます。
これは「絶対値が小さい」という意味でしょうか?それとも、「マイナスの符号を含めて小さい」という意味でしょうか? |
ほんとこれややこしいよね。 |
わかりました。ありがとうございます。 |
すみません、寄付の価格弾力性と政府の効率性の議論について事実誤認があったと思うので訂正しておきます。 これはこのIssueの
から始まる一連の議論に事実誤認があったということを示すものなので、非常に申し訳ないです。すみません。つまり正しく言えばSaez(2004)では
ではなく、
ということが言えていると考えられます。(これも事実誤認だったらきついですが) |
ただし、Saez(2004)の世界は政府が最適化を解けている世界なので、個々人がGivenの寄付価格などにどう反応するかという話とは違います。なので、実証分析では、政府が非効率だと思う人の寄付の価格弾力性が弾力的ということもありえると思います。 1.人々は寄付のほうが相対的に効率的に思うならば、政府のつける寄付価格の1単位の低下によって限界的により増やそうとする(弾力性大) 2.最適な税制のもとでは政府は寄付のほうが相対的に効率的ならば、寄付を優遇したほうがいいので、寄付価格を低下させ、寄付を増やそうとする。 1は個人を主に考えるとこうなると思われます。私の夏頃に作っていたモデルがこれですね。このモデルでは政府の最適化問題はでてきません。 |
このテーマはお蔵入りにします。さようなら。close |
ベースライン分析を固めました
皆さん、ふらふらして申し訳ありませんでした。
1/10に私の大先輩の佐々木周作先生(東北学院大学)に相談し、ベースラインの分析を固めました。
Issueのタイトルはtentativeですが、この研究をよく表していると思います。
内容としましては、
今後はこのメカニズムとロバスト分析を検討します。
今後の分析結果の共有方法について
今まで分析結果はHTMLファイル(report.html)で共有していました。しかし、皆さんが逐一ローカルにダウンロードして開くという面倒くさいところがありました。今後はreportブランチのslides.pdfに逐一結果を反映していきます。pdfはgithubのページでも開けるので確認の容易性が高まると思います。また、学会発表もこのpdfを使いまわせばいいので、いろいろ便利かと思います。report.htmlは近いうちに削除します。ただ、せっかくですので、ローカルに最新状況を反映しておくことをお勧めします。
The text was updated successfully, but these errors were encountered: