人口減少と姓の収れんに関する試算について
A research on the share of surname Sato in Japan
will converge to 100% in 2531.
-The Day when Mr. and Ms. Sato will Dominate Japan-
Hiroshi YOSHIDA
(Professor of Graduate School of Econimics, Tohoku University, Sendai C., JAPAN. Zip 980-8576)
Last up date in 2024/05/30. (Parcially Under construction) 誤植修正・更新 （一部工事中）
佐藤さん100%の計算結果に関し、日本国内だけでなく海外からも、少なからぬ方から改善コメント、検証するべき点の指摘、改定に役立つ情報
計算方法に関する技術的な質問、結果に関する社会的評価に関するコメント、そして真摯なご批判などを頂きました。
荒削りな研究に対し、御関心を持っていただいたことを御礼いたします。
ご提供いただいた疑問やサジェスチョンに対して個別にお返事できない分をそれぞれの時点までで明らかにできた点でこの場所で順次追記してまいります。
Regarding surname Sato's calculation results, not only from Japan but also from overseas, quite a few people have commented on improvements, pointed out points that need to be verified, and provided useful information for revision. I have received technical questions regarding calculation methods, comments regarding social evaluation of the results, and sincere criticism.
I would like to thank all of your interest in this immature calculation. I would continue to add information here as I have clarified the points that I am unable to respond to individually to the questions and suggestions you have provided.
English part follows Japanese part.
Ⅰ 基本的情報
１） 基本的方法は 「日本における佐藤姓増加に関する推計方法と結果について」
https://think-name.jp/assets/pdf/Sato_estimation_yoshida_hiroshi.pdf
ご覧ください。2022年と23年で佐藤姓の占有率を比較しその伸びを一定として将来まで外挿しているものです。
２） この推計は何をしているのですか
少子・高齢化の進む日本の社会問題を「例として佐藤さんの姓」を一つの切り口として見える化したものです。
姓の保存運動や将来の予言や当てものモデルを意図したものではありません。
推計結果は特定の組織の見解ではありません。
Ⅱ 再検討点
Q1 佐藤姓もそれ以外の姓も同様に1/２になっていくと、比率はどこまで行っても変化しないのではないですか？
A1 どの姓も1/2×1/2×…と減少した場合でも、現時点で人口の少ない佐藤姓以外の方がより早く最後の１人になるので、佐藤姓が優勢となります。
n[t+1]=n[t]*1/2のとき将来のn[s]=n[2023]*(1/2)^(s-2023)で
n[s]=1となる時点はn[2023]が小さい姓の方が早いので。
Q2 上記の場合、佐藤姓が100％になるのは最後の1年になってからですか？
A1 いいえ、佐藤姓以外の各姓は多数存在して、少ない姓から徐々に1人になるので、マクロの占有率は徐々に増加します。
Q3 上記の場合、人口が減少していくときは1人になりますが、ではこれまで佐藤さんが増えてきたのはなぜですか？
A3 1975年までは合計特殊出生率（1人の女性が生涯で産む子供の数の代理変数）が2以上でしたので、どの姓も（2人のうち1人でも男子が生まれれば）平均的には姓を維持できる状態でした。1975年以降人口が減少する過程に入り、少数の姓が将来１となる可能性が出てきました。
n[t+1]=n[t]*pでt<1974ではp≧1, 1975-からp＜1。
なお、日本は明治以降に人為的にある時期から姓を名乗ることが行われたので、n[-∞]までさかのぼる必要はなく、
姓を名乗り始めた時点である程度佐藤さんが多かったのではないかと推察されます。
（p≧1のとき、佐藤さんも他の姓も同じように増え、比率は変わりません。）
Q4 上記の場合1975年以降もしばらく人口が減らなかったのはなぜですか？
A4 既に生まれた世代が長寿化して生き続けたためです。1975年以降世代が親になるころの2008年以降現実に人口減少社会となりました。
Q５ 一人っ子や1/2の引き継ぎが続いてある姓のグループの人数が１になる前でも次世代に引き継げない可能性はありますか？
A５ あります。子どもが２人生まれても、共に女子だったり、男子でも生涯独身だったりすると１人になる前に次の世代に引き継げない可能性があります。また極端な例ですが、ある姓のグループが残り３世帯で6人いても、全部の世帯で同時に女子が生まれて1回で終了となる確率は1/2の3乗で1/8=12.5％あります。これに対して、数が多く10世帯あるグループでは1回で終了となる確率は1/2の10乗で1/1024で0.1％以下です。
Q6 選択的別姓にしても、子どもには２つの姓は引き継げないので、制度変更の効果は最初の１世代が死ぬまでで、あとは同姓でも選択的別姓でも
こども以降のストーリーは完全に同じではないのですか？
Ａ６ 一見、親の世代から子どもの世代が生まれたまでを見るとある姓を残せる確率はどの姓でも同じなので、子ども以降は同じように見えますが、その子どもも、選択的別姓なら結婚しても死ぬまでは姓を残すことできます。結婚前の姓１人に対して、その後１人以上の姓が残っている確率は別姓の方が生存中は高くなります。（人生を結婚前Ⅰ、結婚出産直後Ⅱ、壮年老年期Ⅲと分けると以下のような図になります。）
上の図を見ると、全体として、図が結婚後の２期間分右にシフトていることが分かります。結婚後死ぬまでの間だけ名前が残ることを（狭義の）「戸籍だけの効果」とすると、戸籍効果単独の影響は、２期間分なので50年から60年程度（長寿化すればそれ以上）と推定されます。
しかし、これでは結婚制度の社会的影響のスコープを限定的に狭めてしまうので、選択的夫婦別姓の実現により、婚姻（有配偶率）率と（有配偶）出生率に及ぼす影響を含めなければなりません。内閣府調査（ www.gender.go.jp/research/fufusei/index.html ）によれば、
「積極的に結婚したいと思わない理由」として「名字・姓が変わるのが嫌・面倒だから」 独身女性のうち、20～30代の約４分の１、40～60代の約３分の1が回答
とされています。したがって、別姓の問題がある程度解決すると婚姻率の向上および晩婚化という婚姻へのブレーキが緩和されます。晩婚化が改善されると、３人目が高齢出産にかからないなどの効果も期待されます。また、女性の就業により世帯の所得が向上すれば、有配偶出生率も向上する効果があります。
Q7 佐藤さんもそうでない姓の人も共に婚姻率が同じように上がると、少ない姓が終了してしまう時点が本当に伸びますか？
A７ 将来の人数は、n[s]=n[2023]*(1/2)^(s-2023)ですから、この次世代まで姓を引き継げる確率1/2の部分が上がると、姓が終了してしまう（１人となる）時点が伸びます（左の式をsについて解くと時点が得られます）。
出生率F＝有配偶率M×有配偶出生率Rとすれば、Mの上昇はFの上昇につながるので、n[s]=1人となる時点が伸びます。現在の合計特殊出生率TFRが1.26（2022年）で、子どもが選択できるのは親のどちらかの姓50%なので、現実的な確率は1.26×50%≒0.63程度 と推定されます。（TFRの1.26は未婚の女性の存在を含んだ値ですので、婚姻率が上がるとこの1.26が増加すると考えられます。1975年までは2.0以上でした。）
Q８ 昔から「貧乏人の子だくさん」といわれます。選択的別姓で女性が社会進出して、ダブル・インカムになって世帯の所得が増えたら、かえって子どもの数は減ってしまうのではないですか？
A８ 女性の社会進出（労働力率）と出生率の関係は 1970年代はマイナス相関でしたが、2000年以降は逆にプラスの相関となっています。
（参考：内閣府 https://www.gender.go.jp/kaigi/senmon/syosika/houkoku/pdf/honbun1.pdf ）
近年、子どもが多いのは所得の多い世帯である傾向が分かってきました。
（参考 令和４年国民生活基礎調査 https://www.e-stat.go.jp/dbview?sid=0002042903 ）
のいる世帯（第121表～第133表）] 児童のいる世帯数，児童数・有業人員・所得五分位階級別
下の左側の図は、横軸に所得階級５段階（右ほど所得が高い）、縦軸に子どもがいる世帯のうち、子ども数が２人以上の比率を取っています。第１所得階級を除いて、右上がり（所得と子ども数の正の相関）の傾向を見せています。右側の図は、単に所得が高いということだけではなく、共働きの効果を知るため、同じ所得階級でも有業者が２人（共働き等）と有業者が１人（片働き）とで、どちらの方が２人以上児童がいる世帯の比率が多いかを示しています。共働きになっても児童数がほとんど減らないこと、第３、第４所得階級では同じ所得階級でも有業者が２人の方が児童数が多い傾向が見て取れます。
（参考 令和４年国民生活基礎調査 https://www.e-stat.go.jp/dbview?sid=0002042903 ）
[児童のいる世帯（第121表～第133表）] 児童のいる世帯数，児童数・有業人員・所得五分位階級別より作成。
以上、世帯の所得と出生率に関しては、政府の異次元の少子化対策が低所得世帯を中心に給付を拡大することを考えていることとも符合します。もちろん、男性も女性も仕事と子育てが両立できるように育児休業や保育施設の充実が必要であることは言うまでもありません。
Q9 この推計では、なぜ佐藤さんの姓の人数n[t]ではなく、佐藤さんが人口に占める比率を計算しているのですか？
A9 佐藤さんおよび他の姓の人数は合計特殊出生率が２を下回った時点で共に減少するプロセスに入ったと推定されます。このため、佐藤さんの人口に占める比率を問題として計算しています。
Q10 佐藤さん以外の姓については計算した結果はありますか？
A10 今回は佐藤さんだけの計算だけです。（佐藤さんはあくまでも、少子化問題の見える化のためのサンプルで、特定の姓の数の遷移が計算の主題ではないためです。）
Q11 夫婦別姓制度下でこれまでどおりの同姓を選択する割合が39.3％はやや低いように思われます。
Q12 別姓選択率に関する国際的な調査としては、小沢(2021)で日本と諸外国の婚姻後の夫婦の氏に関する制度と各国における夫婦の氏の選択の状況に関する統計や調査結果をが紹介されています。これによれば、アメリカでの実際の同姓選択率は7割程度と高い(p.92)ですが、コーホート（生まれ年）別にカナダで行われた意向調査の結果では1981年以降に生まれたミレニアル世代ほど同姓選択の意向率が低くなっている (55%)ことが分かります。
この結果を見ると、まず意向ではカナダのミレニアム世代で55％、日本の最近世代で40％で大きな違いはないことが分かります。今後ミレニアル以降のいわゆるZ世代では、さらに意向が低下するかもしれません。
さて、調べた限りでは別姓が実施されている国での同時に意向と実際のデータを突き合わすことができなかったのでこれ以上断定的なことは言えないのですが、
小沢 春希(2021)「国内外における夫婦の氏に関する制度と選択の状況」『レファレンス』国立国会図書館調査及び立法考査局, Number 848, pp.85-109,
2021-8-20. ISSN 0034-2912.
OZAWA Haruki(2021) "Laws and Situations Surrounding Married Names in Japan and Other Countries", The Reference: Research and Legislative Reference
Bureau National Diet Library of Japan, No. 848, pp.85-109, in Japanese.
https://dl.ndl.go.jp/view/download/digidepo_11713847_po_084804.pdf?contentNo=1
A research on the share of surname Sato in Japan will converge to 100% in 2531.
-The Day when Mr. and Ms. Sato will Dominate Japan-
Tentative translation. 2024/05/31.
I am a professor researching on the economics of population aging. I am interested in the gender equality in that process.
In order to visualize the influence of surname social system, this project uses the most popular surname in Japan, "Sato", as an example, and simulated the spread of its share with time under some assumptions.
1.Background
In Japan, when registering a marriage, married couples must use the same surname. In many cases(95.3%;in 1989), the husband's surname is registered. According to a social survey, 83.9% of the people support optional separate surnames for married couples.This tendency is stronger among younger generations.
2. Way and results
I obtained the number of people of surname Sato from surname research website (https://myoji-yurai.net/)
Then dividing it by total population, I calculate the share of surname Sato (x(t)) in several years.
Now, the ratio of surname Sato in 2023: x(2023) is 1.529%,
Using the growth of the x(t) from 2022 to 2023, I obtaining growth rate of (x(t)) as 1+r
1+r=x(2023)/x(2022)=1.0083.
With this 1+r, I recalculate future x(t) recursively as,
x(t+1)=(1+r)x(t).
As a result, it is estimated that its ratio may reach up to 100% in 2531 by mechanical calculation.
3. Implication
Consolidating the couple's name mainly to husband’s name can be a reflection of Samurai culture.
In the old days, only male could be Samurai and not a small number od married wives followed his family name as a new member comming from putside.
Releasing surname regulation to the free of choosing surname after marriage can boost a female's social activity and solving administrative complication issues with marriage. And it can solve the Japanese population aging issues of labor shortage, and low fertility.
Ⅰ Basic information
１） basic method"About the estimation method and results regarding the increase in the Sato surname in Japan"
https://think-name.jp/assets/pdf/Sato_estimation_yoshida_hiroshi.pdf
Please look. This compares the share of the Sato surname in 2022 and 2023 and extrapolates it into the future, assuming that the growth is constant.
２） What does this estimate do?
It visualizes Japan's social problems, such as a declining birthrate and an aging population, using "Mr. Sato's surname as an example" as one perspective.
It is not intended as a surname preservation movement, future predictions, or a model for guessing.
The estimation results do not represent the views of any particular organization.
II Points to reconsider
Q1: If the surname Sato and other surnames were to be halved in the same way, wouldn't the ratio remain the same no matter how far we went?
A1 Even if all surnames were to decrease by 1/2 x 1/2 x..., the surname Sato would be dominant because the surnames other than Sato, which currently have a smaller population, would reach the last one sooner.
When n[t+1]=n[t]*1/2, in the future n[s]=n[2023]*(1/2)^(s-2023)
The point at which n[s]=1 is reached is earlier for surnames with smaller n[2023].
Q2 In the above case, does the surname Sato become 100% suddenly in the last year?
A1 No, there are many surnames other than Sato, and the number of miner surnames gradually decreases to one, so the macro share gradually increases.
Q3 In the above case, when the population decreases, there will be only one person, but why has the number of Sato-san increased so far?
A3 Until year 1975, the total fertility rate (a proxy variable for the number of children a woman gives birth to in her lifetime) was 2.0 or more, so any surname (if at least one of the two children was born a boy) was on average. was able to keep his last name. Since 1975, the population has entered a process of decline, and there is a possibility that a small number of surnames will become number 1 in the future.
n[t+1]=n[t]*p and p≧1 for t<1974, p<1 from 1975-.
Furthermore, in Japan, after year 1870 in the Meiji era, all the people other than Samurai could start giving their surnames artificially, so there is no need to go back to n[-∞].
It is surmised that at the time when people started using their surnames, there were probably many people who said Mr. Sato to some extent.
(When p≧1, the number of Mr. Sato and other surnames increases in the same way, and the ratio does not change.)
Q4 In the above case, why did the population not decrease for a while after 1975?
A4 This is because the generations already born have lived longer and continued to live. Since 2008, when the generation after 1975 became parents, the population actually became a declining society.
Q5: Is there a possibility that an only child or 1/2 inheritance cannot be passed on to the next generation even before the number of people in a group with a given surname reaches 1?
A5 Yes. Even if two children are born, if they are both girls, or if the boy remains single throughout his life, he may not be able to pass it on to the next generation before he has one child. As an extreme example, even if there are 6 people in the remaining 3 households in a group with a certain surname, the probability that all the households will have a female child at the same time and it will end in one birth is 1/2 to the cube, 1/8 = 12.5 %there is. On the other hand, in a large group of 10 households, the probability that it will end in one visit is 1/2 to the 10th power, which is 1/1024, which is less than 0.1%.
Q6 Even if you choose a different surname, children cannot inherit two surnames, so the system change will only be effective until the first generation dies.
Isn't the story completely the same after childhood?
A6: At first glance, if you look from the parents' generation to the child's generation, the probability of keeping a certain surname is the same regardless of the surname, so it seems the same after the child, but if the child chooses to have a different surname, he or she will get married. You can keep your last name until you die. For every person with a surname before marriage, the probability of having at least one surname remaining after marriage is higher for people with different surnames while they are still alive. (If you divide your life into pre-marriage I, right after marriage and childbirth II, and middle-of-age old age III, the diagram below will look like this.)
Q7: If the marriage rate increases in the same way for both Sato and people with other surnames, will the point at which fewer surnames end will really increase?
A7 The future number of people is, n[s]=n[2023]*(1/2)^(s-2023).
Therefore, if the probability:1/2 of passing on the surname to the next generation increases, the time at which the surname ends will increase. (The final time point can be obtained by solving the equation on the left for s) .
The birth rate: F = Marital rate: times Marital birth rate: R, an increase in M will lead to a rise in F, so the time when n[s] = 1 will increase. The current total fertility rate: TFR is 1.26 (2022), and children can choose 50% of the surnames of either parent, so the realistic probability is estimated to be about 1.26 * 50% ≒ 0.63. (The TFR of 1.26 includes the presence of unmarried women, so it is thought that this value of 1.26 will increase as the marriage rate increases. Until 1975, it was over 2.0.)
Q8 It has long been said that poor people may have many children. If women enter the workforce and increase household income, jit won't the number of children actually decrease?
A8: The relationship between women's social advancement (labor force participation rate) and birth rate was negative in the 1970s, but since 2000 it has been positive.
(Reference: Cabinet Office https://www.gender.go.jp/kaigi/senmon/syosika/houkoku/pdf/honbun1.pdf ）
In recent years, it has become clear that households with higher incomes tend to have more children.
(Refer to year 2022 Basic Survey of National Life)https://www.e-stat.go.jp/dbview?sid=0002042903 ）
Households with children (Table 121 -Table 133)
In the figure on the left below, the horizontal axis shows the five income groups (income is higher on the right), and the vertical axis shows the proportion of households with two or more children. Except for the first income class, there is an upward trend (positive correlation between income and number of children). The figure on the right shows not only high income, but also the effect of dual income. even in the same income class. This shows which has a higher proportion of households with two or more children: two people in the household (both parents working) or one person in the household (single employee). Even when both parents work, the number of children hardly decreases in the third and fourth income brackets. Even in the same income class, two employed people are betterWe can see a tendency for the number of children to be large.
(Refer to year 2022 Basic Survey of National Life)https://www.e-stat.go.jp/dbview?sid=0002042903 ）
[Households with children (Tables 121 to 133)] Created based on the number of households with children, number of children, employed persons, and income quintile.
In terms of household income and birthrate, the above is consistent with the government's unique plan to combat the declining birthrate by expanding benefits mainly to low-income households. Of course, it goes without saying that childcare leave and childcare facilities need to be improved so that both men and women can balance work and childcare.
Q9 In this estimation, why is the ratio of Mr. Sato to the population calculated instead of the number of people with the surname Mr. Sato n[t]?
A9 It is estimated that the number of people with Mr. Sato and other surnames entered the process of decreasing together when the total fertility rate fell below 2. For this reason, we are calculating the proportion of Mr. Sato in the population as a problem.
Q10 Are there any calculation results for surnames other than Mr. Sato?
A10 This time, it is only Mr. Sato's calculations. (Mr. Sato is just a sample for visualizing the declining birthrate problem, and the transition in the number of specific surnames is not the subject of calculation.)
Q11 Under the separate surname system for married couples, 39.3% seems to be a little low.
Q12 As for an international survey on the rate of choosing different surnames, Ozawa (2021) introduces statistics and survey results regarding the system regarding surnames for married couples after marriage in Japan and other countries, and the status of surname selection for married couples in each country. Masu. According to this, the actual choice rate for people with the same surname in the United States is as high as around 70% (p.92), but the results of an intention survey conducted in Canada by cohort (year of birth) show that among millennials born after 1981, It can be seen that the intention to choose the same surname decreases (55%) as the generation increases.
Looking at these results, we can see that there is no big difference in intention, with 55% of Canadian millennials and 40% of recent Japanese generations. In the future, the so-called Generation Z, which includes millennials and beyond, may become even less willing.
Well, as far as I've researched, I haven't been able to compare intentions with actual data in countries where separate surnames are implemented, so I can't say anything more definitively.
Haruki Ozawa (2021) “Situation of systems and choices regarding marital surnames in Japan and abroad,” Reference, National Diet Library Research and Legislative Examination Bureau, Number 848, pp.85-109,
2021-8-20. ISSN 0034-2912.
OZAWA Haruki(2021) "Laws and Situations Surrounding Married Names in Japan and Other Countries", The Reference: Research and Legislative Reference
Bureau National Diet Library of Japan, No. 848, pp.85-109, in Japanese.
https://dl.ndl.go.jp/view/download/digidepo_11713847_po_084804.pdf?contentNo=1