www.debito.org
来日外国人犯罪の現状
国内総合犯罪との比較
FOREIGN CRIME STATISTICS FOR JAPAN
COMPARED TO JAPANESE CRIME RATES
1993-2003年度の統計 (page
down)
警察庁来日外国人犯罪等対策室 出版
及び
刑法犯犯罪の特徴的傾向 (page down)
警察庁 出版
1994-2003年度の統計
及び
外国人不法滞在の1993-2003年度統計 (page down)
法務省入国管理局2003出版
及び
警察庁の外国人犯罪統計に対する疑問（英字テキスト）
FOREIGN CRIME IN TOTAL 1993-2003 (page
down)
and
CRIME IN JAPAN IN TOTAL (1994-2003) (page
down)
(Both Courtesy National Police Agency)
Also
ILLEGAL OVERSTAYERS IN JAPAN 1990-2003 (Courtesy Ministry of
Justice, Immigration Bureau, 2003) (page down)
Finally
CAVEATS IN AND QUESTIONS ABOUT THE SAMPLING AND ANALYSIS OF
THE NPA'S FOREIGN CRIME STATISTICS (page
down)
来日外国人犯罪の現状
1993-2003年度の統計
FOREIGN CRIME STATISTICS FOR JAPAN
FOREIGN CRIME IN TOTAL 1993-2003
（http://www.npa.go.jp より)
注意１：「特別法犯」は外国人のみが犯せる犯罪の統計（外国人登録法犯・ビザ関係・不法滞在等なら日本人は逮捕不可）
注意２：「総検挙 人員」の線で、外国人登録者数の増加のなか、「外国人犯罪急増中」（例えば、産経新聞２０００年５月１日front
page）とは言い切れない。
注意３：警察庁は「外国人の犯罪が多く、外国人なら犯人だと疑う相当な理由」の大前提で外国人を捕まり（例えば、東京で自転車に乗っている場合、交番に訪ねると警察官は「外人登録証を見せなさい」の場合）、いわゆる「検挙人員インフレ」になる可能性がある（もし、日本人をもっと捕まると日本人検挙人員も増える）。ひいては、2000年２月、静岡県警察本部は「外国人犯罪の特徴」を商店に配付し、外国人はグループで入店すると、「グループが乗ってきた車を確認、ナンバーをメモして警察に通報」のアドバイスを店員にあげた。それなら「外人ハンティング」を促進していると感じるのはやむを得ないであろう。しかも、上記のレポートは「警察庁来日外国人犯罪等対策室」で発行されたので、このタイトルはまるで警察庁は「外国人に対し対策が必要」である。いつ犯罪率の急増している日本の中・高校生の対策室を設けるのか。
刑法犯犯罪の特徴的傾向
警察庁 出版
1994-2003年度の統計
FOREIGN CRIME STATISTICS FOR JAPAN
ILLEGAL OVERSTAYERS IN JAPAN 1990-2003 (Courtesy Ministry of
Justice, Immigration Bureau, 2003)
外国人不法滞在の統計（法務省入国管理局２００３出版）
Look at the Orange Chart. According to IMMIGRATION 2003, Shutunyuukoku Kanri,
"Ru-ru o Mamotte Kokusaika", the number of illegal overstayers ("Gaikokujin
Fuhou Zanryuusha") in Japan been dropping EVERY year since 1993 (peaking at
299,000, now 220,000 in 2003). Available at any Immigration Office.
上記のオレンジのチャートをご覧下さい。外国人不法滞在者は「増加」していますか。いや、1993年から2003年の間、毎年もれなく低下しています。
CAVEATS IN AND QUESTIONS
ABOUT THE SAMPLING AND ANALYSIS OF
THE NPA'S FOREIGN CRIME STATISTICS
By Arudou Debito
Email To: The National Archive of Criminal Justice Data (NACJD)
nacjd@icpsr.umich.edu
http://www.icpsr.umich.edu/NACJD/
http://www.icpsr.umich.edu:8080/NACJD-STUDY/02565.xml
January 5, 2003
My name is ARUDOU Debito, speaking for internet human-rights group The Community
(http://www.debito.org/TheCommunity)
and United for a Multicultural Japan (http://www.tabunka.org),
and I have a question of importance about the reportage and analysis of crime statistics
in Japan.
My question is statistical in nature. Japan's National Police Agency (NPA http://www.npa.go.jp/english/index.htm)
in recent years has been announcing large rises in the foreign crime rate, and using
this to justify increased budgetary outlay for specific crackdowns on foreign crime,
often through racial profiling. (http://www.debito.org/foreigncrimeputsch.html)
We believe that the statistical analysis offered by the police is flawed, but as
amateurs in the field of statistics, we would like your opinion (or an agency you
could recommend) as an impartial, informed third party.
The data set is as follows:
STATISTICS FOR CRIME IN JAPAN, JAN-JUNE 2003
(Courtesy http://www.npa.go.jp/toukei/index.htm
in Japanese only)
Total Crimes Reported: 1,338,983 cases
Total Cleared Cases of Crime: 290,480 cases (clearance rate of 21.7%)
Total Number of Cleared Perpetrators: 169,992 people
These statistics include foreign crime cases and criminals.
STATISTICS FOR FOREIGN CRIME IN JAPAN, JAN-JUNE 2003
(Courtesy http://www.npa.go.jp/kokusaihanzai/index.htm
in Japanese only)
Total Cleared Cases of Foreign Crime: 12,507 cases
(Total Cleared Cases including visa violations: 18,579 cases**
**NB: We prefer to leave out visa violations because Japanese by definition cannot
commit these crimes, and we want comparable numbers for comparable crimes.)
Total Number of Cleared Foreign Perpetrators (leaving out visa
violations): 4016 people
The NPA provides far more details and breakdown of changes in foreign crime than
for Japanese crime, and with no real comparison between the two. When releasing crime
statistics to the press, the NPA compares the foreign crime numbers for last year
(or for the same six-month period for the previous year) with the current foreign
crime numbers, then indicates whether foreign crime has increased. For example:
Total Cleared Cases of Foreign Crime Jan-June 2003: 12,507 cases
Total Cleared Cases of Foreign Crime Jan-June 2002: 10,628 cases
Change in numbers: +1879 cases
Change in crime rate during Jan-June 2003: +17.7%
Total Cleared Foreign Perpetrators Jan-June 2003: 4016 people
Total Cleared Foreign Perpetrators Jan-June 2002: 3645 people
Change in numbers: +371 people
Change in crime rate during Jan-June 2003: +10.2%
These numbers, when announced to and by the mass media, are not compared with the
rise in Japanese crime (see below), and are not grounded
in the context of a rising foreign population and a static Japanese population.
THE QUESTION
How do we calculate a more accurate crime rate, using the above data set, comparing
foreign crime to Japanese crime? In sum, when talking about crimes cases, what goes
in the denominator?
We have a bounded set 1,338,983 total cases. We have a sampling of 21.7% (290,480
total cleared cases). We have 12,507 total cleared foreign cases.
So is the crime rate:
12,507 / 1,338,983 = 0.9%
(i.e. a function of the total cases in the bounded set, using the only
hard number we have available in the denominator)
or
12,507 / 290,480 = 4.3%
(i.e. a function of the cases within the sample size, but using a softer
number in the denominator, potentially skewed due to biases in the sampling I note
below)
This matters, because the number of "registered foreigners" in Japan is
around 1.7% of the total population, and depending on the denominator, the crime
rate is either remarkably higher or remarkably lower than the per capita proportion
of foreigners in Japan.
Same with the perpetrators. We have 169,992 total perps. From there, we have 4016
foreign perps. That comes out to 2.4% of the total number of sampled perps. However,
we believe that neither number is hard or reliable due to the following caveats:
CAVEATS TO THE DATA include:
1) The definition of "foreigner".
Only those who are here on visas for more one year or more are "registered foreigners",
so those here temporarily, including tourists, illegals, and temporary workers have
the same potential to commit domestic crime yet are not included as a measurable
percentage of the population. Meaning we don't exactly know how many "foreigners"
are in Japan making up the number. On the other hand, Japanese law requires blood
for citizenship, so up to four generations of ethnic Asians born and raised in Japan
are still registered as "foreign", again skewing the numbers registered
significantly (by a factor of nearly two).
2) The demographics of "foreigners".
Foreign workers, immigrants, and illegals have a higher number of young adults than
the average Japanese population. Which means they are in an age group statistically
more likely to commit crime.
3) The economics of "foreigners". Many are also in trades (the licenced
quarters, temporary workers) which can be more closely linked to crime, or have below-average
economic stability compared to Japanese.
4) Racial profiling.
The Japanese police in recent years have been "cracking down" on crime,
singling out visibly-foreign people for street checkpoints and carrying out raids
on the licenced quarters. This means that foreigners are more likely to be stopped
and caught red-handed than Japanese, skewing the numbers upwards.
5) No "population growth deflator".
Japan, the fastest-aging nation in the world, has practically no population growth.
Except for the number of foreigners--perennially on the increase. More people means
more potential crime, and a rise in the Japanese crime rate is not the same as the
rise in the foreign. That is not taken into account.
In sum, how does one account for the low sample size due to the low clearance rate,
and the potential biases in the sampling due to biased samplers? Although there is
no accounting for the lack of comparative analysis either by the police or by the
mass media, we need to know a more accurate way of calculating crime rates in accordance
with international standards.
With the beginning of the new year, we will soon have another report from the NPA
on foreign crime statistics, and have another periodic media blitz and political
outcry on how foreign crime must be brought under control. Your input on more sensible
statistical analysis would be most helpful, in fact essential for helping us as minorities
live less targeted lives. Thank you very much indeed for your time and assistance.
Sincerely yours,
ARUDOU Debito, The Community, Sapporo, Japan
debito@debito.org, http://www.debito.org
http://www.debito.org/TheCommunity
Jens WILKINSON and Imtiaz CHAUDHRY, United for a Multicultural Japan
http://www.tabunka.org
jowilkinson4@yahoo.com
chaudhry@thechaudhry.com
SUPPLEMENTARY DATA FOLLOWS
SUPPLEMENTARY DATA
What follows is the comparable data and rates for Japanese cases and perpetrators
(my calculations, using the same method of calculation the NPA uses on foreign crime):
Total Cleared Cases of Crime Jan-June 2003: 290,480 cases
Total Cleared Cases of Japanese Crime only Jan-June 2003:
(290,480 - 12,507 foreign cases) = 277,973 Japanese cases
Total Cleared Cases of Crime for Jan-June 2002: 271,404 cases
Total Cleared Cases of Japanese Crime only Jan-June 2003:
(271,404 - 10,628 foreign cases) = 260,776 Japanese cases
Change in numbers (277,973 - 260,776) = +17,197 cases
Change in Japanese crime rate during Jan-June 2003: (17,197/260,776) +6.7%
Total Cleared Crime Perpetrators Jan-June 2003: 169,992 people
Total Cleared Japanese Perps only Jan-June 2003:
(169,992- 4016 foreign cases) = 165,976 Japanese people
Total Cleared Japanese Perps for Jan-June 2002: 162,282
Total Cleared Japanese Perps only Jan-June 2003:
(162,282 - 3645 foreign cases) = 158,637 Japanese people
Change in numbers (165,976 - 158,637) = +7339 people
Change in Japanese crime rate during Jan-June 2003: (7339/158,637) +4.6%
ENDS