Herfindahl - Hirschman Index input the right data structure

[Lonneke de Nooijer]

Dear all,

I am currently doing my thesis. As part of this, I would like to calculate the Herfindahl-Hirschman Index. I want to calculate the concentration of employment in a specific sector in a specific year. However I am wondering if somebody can give me some more information how to work with the HHI command function in Stata. My version of Stata is 14 on Mac.

I have 530 regions (Nuts 2 regions of Europe), 14 Industry sectors and the information from the years 1995 till 2017. See below a small example how my data set is organised. I already look at ''search hhi'' in Stata and read all the information in this Stata forum about this topic, however I am confused if my data structure is correct and how to use the HHI command correctly.
So far I have tried different commands, such as ''hhi Employment, by (Industry Year Region)'' however the results are only value 0 or 1...

Sorry I am new here and new with working with Stata, so I am a bit in panic how to get started with this function. Furthermore sorry if I forget to include certain important things, just let me know!

Year	Region	Industry	Employment
2008	Groningen	Mining and quarrying	674
2009	Groningen	Mining and quarrying	633
2010	Groningen	Mining and quarrying	626
2008	Groningen	Manufacture of food products	3,727
2009	Groningen	Manufacture of food products	3,713
2010	Groningen	Manufacture of food products	3,643
2008	West-Nederland	Mining and quarrying	3,668
2009	West-Nederland	Mining and quarrying	4,065
2010	West-Nederland	Mining and quarrying	3,870
2008	West-Nederland	Manufacture of food products	44,232
2009	West-Nederland	Manufacture of food products	44,646
2010	West-Nederland	Manufacture of food products	45,947

I hope you get my problem and idea and if somebody can help me it would be great!
Thank you in advance.

Kind regards,
Lonneke

[Nick Cox]

I doubt that anyone has read everything about this measure on Statalist, as over the years there have been hundreds of posts about it, sometimes under quite different names. it's typically economists who talk about Herfindahl and/or Hirschman either knowing or not knowing that the measure goes back at least 30 years before either.

I don't know anything much about the command (not function) hhi (which is from SSC, as you are asked to explain), but it seems to me that specifying industry, year and region just specifies individual observations in your dataset. To be blunt, I didn't think the help of hhi is especially clear, but then others might say the same about my files.

I am more familiar with entropyetc, also from SSC.. I get this result (note the use please of dataex to show data).

* Example generated by -dataex-. To install: ssc install dataex
clear
input int year str14 region str28 industry long employment
2008 "Groningen"      "Mining and quarrying"           674
2009 "Groningen"      "Mining and quarrying"           633
2010 "Groningen"      "Mining and quarrying"           626
2008 "Groningen"      "Manufacture of food products"  3727
2009 "Groningen"      "Manufacture of food products"  3713
2010 "Groningen"      "Manufacture of food products"  3643
2008 "West-Nederland" "Mining and quarrying"          3668
2009 "West-Nederland" "Mining and quarrying"          4065
2010 "West-Nederland" "Mining and quarrying"          3870
2008 "West-Nederland" "Manufacture of food products" 44232
2009 "West-Nederland" "Manufacture of food products" 44646
2010 "West-Nederland" "Manufacture of food products" 45947
end

entropyetc industry [w=employment], by(region year)
(analytic weights assumed)

--------------------------------------------------------------------------------
              Group |  Shannon H      exp(H)     Simpson   1/Simpson     dissim.
--------------------+-----------------------------------------------------------
     Groningen 2008 |      0.428       1.534       0.741       1.350       0.347
     Groningen 2009 |      0.415       1.515       0.751       1.331       0.354
     Groningen 2010 |      0.417       1.517       0.750       1.334       0.353
West-Nederland 2008 |      0.270       1.310       0.859       1.165       0.423
West-Nederland 2009 |      0.287       1.333       0.847       1.181       0.417
West-Nederland 2010 |      0.273       1.314       0.857       1.167       0.422
--------------------------------------------------------------------------------

As a check for Groningen 2008 I see two sectors in the example data, so the sum of squared probabilities is

. di (674 / (674 + 3727))^2  + (3727 / (674 + 3727))^2
.74061399

which is where .741 comes from the output.

As this measure is just based on calculating fractions and summing their squares, it could also be calculated directly without community-contributed commands.

I trust that helps. Please note especially https://www.statalist.org/forums/help#stata for your future posts.

[Lonneke de Nooijer]

Hi Nick,

Thank u for the quick response! This information helps me a lot!

However I am wondering if you also can make a concentration of Employment of a specific Industry sector of a specific Region by an given year, with your 'entropyetc' function?
At the end I want to have for example: the % of Employment of Mining and quarrying in Groningen in 2008.
So I have 530 regions and they are al a part of different country's. I wanted to use the the HHI index at first to get a concentration of employment per industry sector per region. See below the formula of the HHI index:

Where
E = Employment
i = specific sector
r = region
n = national average

So if we say that the Netherlands have 2 Regions (Groningen and West-Nederland) I want to calculate for example
In 2008:
(Employment 'Mining and quarrying' Groningen / Employment 'Mining and quarrying' of the Netherlands ) – ( Employment total Groningen/ Employment total Netherlands) ^2

Your command 'entropyetc' seems to work as well, but I don't get a value by Industry, Region and Year. (See below):

Code:
entropyetc Employment, by(Industry Year Region)

-----------------------------------------------------------------------------------------------------
Group | Shannon H exp(H) Simpson 1/Simpson dissim.
-----------------------------------------+-----------------------------------------------------------
Manufacture of food products 2008 Gronin | 0.000 1.000 1.000 1.000 0.917
Manufacture of food products 2008 West-N | 0.000 1.000 1.000 1.000 0.917
Manufacture of food products 2009 Gronin | 0.000 1.000 1.000 1.000 0.917
Manufacture of food products 2009 West-N | 0.000 1.000 1.000 1.000 0.917
Manufacture of food products 2010 Gronin | 0.000 1.000 1.000 1.000 0.917
Manufacture of food products 2010 West-N | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2008 Groningen | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2008 West-Nederland | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2009 Groningen | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2009 West-Nederland | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2010 Groningen | 0.000 1.000 1.000 1.000 0.917
Mining and quarrying 2010 West-Nederland | 0.000 1.000 1.000 1.000 0.917
-----------------------------------------------------------------------------------------------------

I hope you have an answer for my question and thank you in advance for your time and effort! It means a lot to me.

Kind regards,
Lonneke

[Nick Cox]

Sorry, but no. The help for the entropyetc command (again, not a function) explains what it calculates and there are no hidden options to produce other measures. This one looks highly programmable and the copy for entropyetc may help in writing code for your purpose.

[David Benson]

So far I have tried different commands, such as ''hhi Employment, by (Industry Year Region)'' however the results are only value 0 or 1...

I think that's because your data are already aggregated at the industry-year-region level. Per post#3 it looks like what you want is "Is an industry concentrated in a single region or is it spread across many regions?"

If that is the case, it would look like:

ssc install hhi
hhi emp, by(industry year)
sort year industry
rename hhi_employment hhi_emp2 // Called hhi_emp2 because I created hhi_emp to match Nick's answer in post #2

** These 2 lines are just to make hhi easy to check
egen total_emp = total(employment), by(year industry) 
gen percent = employment / total_emp

** Just formatting some variables
format total_emp %9.0gc
format percent  %10.3fc
list year industry region employment total_emp percent hhi_emp2, sepby(year industry) noobs abbrev(12)

  +----------------------------------------------------------------------------------------------------+
  | year                       industry           region   employment   total_emp   percent   hhi_emp2 |
  |----------------------------------------------------------------------------------------------------|
  | 2008   Manufacture of food products        Groningen        3,727      47,959     0.078   .8566539 |
  | 2008   Manufacture of food products   West-Nederland       44,232      47,959     0.922   .8566539 |
  |----------------------------------------------------------------------------------------------------|
  | 2008           Mining and quarrying        Groningen          674       4,342     0.155   .7377355 |
  | 2008           Mining and quarrying   West-Nederland        3,668       4,342     0.845   .7377355 |
  |----------------------------------------------------------------------------------------------------|
  | 2009   Manufacture of food products   West-Nederland       44,646      48,359     0.923   .8582305 |
  | 2009   Manufacture of food products        Groningen        3,713      48,359     0.077   .8582305 |
  |----------------------------------------------------------------------------------------------------|
  | 2009           Mining and quarrying        Groningen          633       4,698     0.135   .7668324 |
  | 2009           Mining and quarrying   West-Nederland        4,065       4,698     0.865   .7668324 |
  |----------------------------------------------------------------------------------------------------|
  | 2010   Manufacture of food products   West-Nederland       45,947      49,590     0.927   .8638687 |
  | 2010   Manufacture of food products        Groningen        3,643      49,590     0.073   .8638687 |
  |----------------------------------------------------------------------------------------------------|
  | 2010           Mining and quarrying   West-Nederland        3,870       4,496     0.861   .7603029 |
  | 2010           Mining and quarrying        Groningen          626       4,496     0.139   .7603029 |
  +----------------------------------------------------------------------------------------------------+

* I created some firm-level toy data to demonstrate as well

dataex year firm industry employment region  // Example shared using dataex command. To install: ssc install dataex
clear
input int year str12(firm industry) long employment str5 region
2005 "Boeing"      "Aircraft Mfg" 25000 "North"
2005 "Airbus"      "Aircraft Mfg" 25000 "North"
2005 "McDonald's"  "Fast Food"    10000 "North"
2005 "Burger King" "Fast Food"    10000 "North"
2005 "Wendy's"     "Fast Food"    10000 "North"
2005 "Tim Horton"  "Fast Food"    10000 "North"
2005 "Boeing"      "Aircraft Mfg" 75000 "South"
2005 "Airbus"      "Aircraft Mfg" 25000 "South"
2005 "McDonald's"  "Fast Food"    40000 "South"
2005 "Burger King" "Fast Food"    30000 "South"
2005 "Wendy's"     "Fast Food"    20000 "South"
2005 "Tim Horton"  "Fast Food"    10000 "South"
2006 "Boeing"      "Aircraft Mfg"  6000 "North"
2006 "Airbus"      "Aircraft Mfg"  4000 "North"
2006 "McDonald's"  "Fast Food"     1000 "North"
2006 "Burger King" "Fast Food"     1500 "North"
2006 "Wendy's"     "Fast Food"      500 "North"
2006 "Tim Horton"  "Fast Food"     1000 "North"
2006 "Boeing"      "Aircraft Mfg"  7000 "South"
2006 "Airbus"      "Aircraft Mfg"  3000 "South"
2006 "McDonald's"  "Fast Food"      200 "South"
2006 "Burger King" "Fast Food"      300 "South"
2006 "Wendy's"     "Fast Food"      400 "South"
2006 "Tim Horton"  "Fast Food"      500 "South"
end

------------------ copy up to and including the previous line ------------------

 list year firm industry employment region, sepby(year region) noobs

  +-------------------------------------------------------+
  | year          firm       industry   employ~t   region |
  |-------------------------------------------------------|
  | 2005        Boeing   Aircraft Mfg     25,000    North |
  | 2005        Airbus   Aircraft Mfg     25,000    North |
  | 2005    McDonald's      Fast Food     10,000    North |
  | 2005   Burger King      Fast Food     10,000    North |
  | 2005       Wendy's      Fast Food     10,000    North |
  | 2005    Tim Horton      Fast Food     10,000    North |
  |-------------------------------------------------------|
  | 2005        Boeing   Aircraft Mfg     75,000    South |
  | 2005        Airbus   Aircraft Mfg     25,000    South |
  | 2005    McDonald's      Fast Food     40,000    South |
  | 2005   Burger King      Fast Food     30,000    South |
  | 2005       Wendy's      Fast Food     20,000    South |
  | 2005    Tim Horton      Fast Food     10,000    South |
  |-------------------------------------------------------|
  | 2006        Boeing   Aircraft Mfg      6,000    North |
  | 2006        Airbus   Aircraft Mfg      4,000    North |
  | 2006    McDonald's      Fast Food      1,000    North |
  | 2006   Burger King      Fast Food      1,500    North |
  | 2006       Wendy's      Fast Food        500    North |
  | 2006    Tim Horton      Fast Food      1,000    North |
  |-------------------------------------------------------|
  | 2006        Boeing   Aircraft Mfg      7,000    South |
  | 2006        Airbus   Aircraft Mfg      3,000    South |
  | 2006    McDonald's      Fast Food        200    South |
  | 2006   Burger King      Fast Food        300    South |
  | 2006       Wendy's      Fast Food        400    South |
  | 2006    Tim Horton      Fast Food        500    South |
  +-------------------------------------------------------+


*** Create the measure using the HHI command (from SSC)
ssc install hhi
hhi emp, by(year industry region)
gen hhi = hhi_employment * 10000  // for various reasons, in econ HHI is scaled to be from 0 to 10,000
format hhi %9.0gc

*** Create the measure manually
egen total_emp = total(emp), by(year industry region)
gen percent = employment / total_emp
gen p2 = (percent)^2
egen hhi_manual = total(p2), by(year industry region)
format total_emp %9.0gc

. list, sepby(ind year region) noobs abbrev(12)

  +---------------------------------------------------------------------------------------------------------------------------------+
  | year          firm       industry   employment   region   hhi_employ~t       hhi   total_emp    percent         p2   hhi_manual |
  |---------------------------------------------------------------------------------------------------------------------------------|
  | 2005        Boeing   Aircraft Mfg       25,000    North             .5     5,000      50,000         .5        .25           .5 |
  | 2005        Airbus   Aircraft Mfg       25,000    North             .5     5,000      50,000         .5        .25           .5 |
  | 2005    McDonald's      Fast Food       10,000    North            .25     2,500      40,000        .25      .0625          .25 |
  | 2005   Burger King      Fast Food       10,000    North            .25     2,500      40,000        .25      .0625          .25 |
  | 2005       Wendy's      Fast Food       10,000    North            .25     2,500      40,000        .25      .0625          .25 |
  | 2005    Tim Horton      Fast Food       10,000    North            .25     2,500      40,000        .25      .0625          .25 |
  |---------------------------------------------------------------------------------------------------------------------------------|
  | 2005        Boeing   Aircraft Mfg       75,000    South           .625     6,250     100,000        .75      .5625         .625 |
  | 2005        Airbus   Aircraft Mfg       25,000    South           .625     6,250     100,000        .25      .0625         .625 |
  | 2005    McDonald's      Fast Food       40,000    South             .3     3,000     100,000         .4        .16           .3 |
  | 2005   Burger King      Fast Food       30,000    South             .3     3,000     100,000         .3        .09           .3 |
  | 2005       Wendy's      Fast Food       20,000    South             .3     3,000     100,000         .2        .04           .3 |
  | 2005    Tim Horton      Fast Food       10,000    South             .3     3,000     100,000         .1        .01           .3 |
  |---------------------------------------------------------------------------------------------------------------------------------|
  | 2006        Boeing   Aircraft Mfg        6,000    North            .52     5,200      10,000         .6        .36          .52 |
  | 2006        Airbus   Aircraft Mfg        4,000    North            .52     5,200      10,000         .4        .16          .52 |
  | 2006    McDonald's      Fast Food        1,000    North         .28125   2,812.5       4,000        .25      .0625       .28125 |
  | 2006   Burger King      Fast Food        1,500    North         .28125   2,812.5       4,000       .375    .140625       .28125 |
  | 2006       Wendy's      Fast Food          500    North         .28125   2,812.5       4,000       .125    .015625       .28125 |
  | 2006    Tim Horton      Fast Food        1,000    North         .28125   2,812.5       4,000        .25      .0625       .28125 |
  |---------------------------------------------------------------------------------------------------------------------------------|
  | 2006        Boeing   Aircraft Mfg        7,000    South            .58     5,800      10,000         .7        .49          .58 |
  | 2006        Airbus   Aircraft Mfg        3,000    South            .58     5,800      10,000         .3        .09          .58 |
  | 2006    McDonald's      Fast Food          200    South       .2755102   2,755.1       1,400   .1428571   .0204082     .2755102 |
  | 2006   Burger King      Fast Food          300    South       .2755102   2,755.1       1,400   .2142857   .0459184     .2755102 |
  | 2006       Wendy's      Fast Food          400    South       .2755102   2,755.1       1,400   .2857143   .0816327     .2755102 |
  | 2006    Tim Horton      Fast Food          500    South       .2755102   2,755.1       1,400   .3571429    .127551     .2755102 |
  +---------------------------------------------------------------------------------------------------------------------------------+

Hope that helps!

[Lonneke de Nooijer]

Dear all,

I am sorry for the late response David, but thank you so much for the helping and you too Nick!

I am now further in my progress of my thesis and just organise my data of employment over 400 regio's (of Europe) and 13 sectors (Industry, Services and Households). Due to the big data set I organise my data at a different way than mention before. See below my example data file. I hope somebody can help me with some questions:

1. I want to merge some data/columns so I will have 13 sectors at the end. My data is from 5 datafiles and of the year period of 1999-2016. But from 2008 there is a different way of naming the sectors, this is why I have to SUM some columns and merge them together
For example: Iron and steel is 1 sector, but I have to merge Emp_DJ27 and C24 together.
And for example: For the sector services I have to sum 'G-Q' , 'G-I' , J-K' , 'L-Q' and sum 'G-I' , 'J' , "K', "L' , 'M-N' , ' O-Q' , 'R-U' and if I have sum them to 2 columns I have to merge them together.
My question: can you do this easily in Stata or is it better to do this in Excel?

2.I have a lot of 'No-Data' in my datafile and it has now the figure ':' or '###' how do you deal with this in Stata, so you do not get errors?

3. My most important question is that I want to calculate the concentration of employment in a specific sector in a specific year.
For example you have the country Belgium with 4 areas with 14 sectors. Take for example the sector ' iron and steel'. I want to know the share of the sector Iron and steel for the years 1999 til 2016 in every area. So for example
2007. Iron and steel. Belgium.Brussel 0,2
2007. Iron and steel. Belgium.Brussel 0,1
2007. Iron and steel. Belgium. Vlaams gewest 0,4
2007. Iron and steel. Belgium.Prov. Antwerpen. 0,3


The command 'hhi; was not able to do this, so I found some formules where I want to write a code for:
- Herfindahl-Hirschman-index (other version than mention in my last post): HHI = Zigmaⁿ_i=1 M²_i
n = amount of jobs of a sector
M = Market share
This calculate the market share of a sector in a country.
And
- The Krugman index: Ks = Zigma_i (v_i^k - v_gem^-k)
v = total activity of a country
k = industry
i = location
This calculate the difference between a share of a sector in a specific area and the national average of the sector in a country.

At the end I want a concentration level of different sectors in the specific area's comparing at country level.
I was wondering if this is possible, because I have 'Country' in a column, 'Area' in a column and 'Year' in a 'Column', but i have the 14 sectors in a different column, so those are in a row.
I never wrote a code before with so many variables. I am wondering if somebody can tell me if it's possible to write a code with variables in columns and rows and give me a small example of how to do this?

Thank you all so much in advance! Your help means a lot to me, because I am sometimes not able to figure it out by myself and I feel a lot of pressure of time for my thesis!
I hope you can help me out!

My version of Stata is 14 on Mac.
Kind regards,
Lonneke

				iron and steel		non-ferrous metals		chemical and petrochemical		non-metallic minerals		transport equipment			machinery		mining and quarrying		food, beverages and tobacco				paper, pulp and printing			wood and wood products		construction		textile and leather					Services											Households
Country	Area	Year	Yearcode	Empl_DJ27	C24	Empl_DG24	C20	Empl_DF23	C19	Empl_DI26	C23	Empl_DM	C29	C30	Empl_DK	C28	Empl_C	B	Empl_DA	C10	C11	C12	Empl_DE	C17	C18	Empl_DD20	C16	Empl_F	F	Empl_DB	Empl_DC	C13	C14	C15	G-Q		G-I		J_K		L-Q	G-I	J	K	L	M_N	O-Q	R-U	Households
Belgium	Belgium	2006	1	38882		66653		5139		35497		64908			45222		3682		101986				9834			13936		243108		55874	2749				2838.9	1011.7	446.8	1380.5								10213752
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2006	1	:		3379		616		1452		7735			3470		102		6877				:			438		19708		2109	377				288.2	93	66.9	128.3								954460
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2006	1	:		:		:		:		:			:		:		:				33253			:		:		:	:				288.2	93	66.9	128.3								954460
	Vlaams Gewest	2006	1	21282		47054		4077		18108		51472			28546		935		72993				10693			10137		152640		48937	2183				1691.9	641.2	268.3	782.4								5926838
	Prov. Antwerpen	2006	1	7608		28454		3552		4313		19904			6813		365		18667				3383			1654		41542		1966	216				452.4	176.9	77.2	198.3								1640966
Germany (until 1990 former territory of the FRG)	Germany (until 1990 former territory of the FRG)	2006	1	265216		478087		20891		:		903184			981359		128720		562548				80323			114488		1115081		197311	26578				22618.1	8194.6	3968.4	10455.2								82037011
	Baden-Württemberg	2006	1	24262		58936		:		:		220374			264414		4385		64998				35314			21083		115617		43369	:				2655.9	936.3	488.2	1231.4								10426040
	Stuttgart	2006	1	:		12559		:		:		145237			116016		1684		24735				20385			6933		44973		11804	:				990.9	358.7	194.7	437.5								3898171
Belgium	Belgium	2007	2	40844		69350		5190		:		64390			44236		3635		101645				53974			14963		251329		54255	2638				2957	1022.5	503.9	1430.7								10239085
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2007	2	357		3340		783		:		8118			3359		102		7075				9443			450		20696		2190	310				293.7	91.1	74.1	128.5								959318
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2007	2	:		:		:		:		:			:		:		:				:			:		:		:	:				293.7	91.1	74.1	128.5								959318
	Vlaams Gewest	2007	2	23667		50264		3999		:		50861			27556		951		71837				33692			10929		156596		47496	2134				1759.7	633.9	309.2	816.5								5940251
	Prov. Antwerpen	2007	2	7672		30805		3569		:		17772			6659		358		18314				11150			1752		43377		1854	191				464	180.7	78.6	204.7								1643972
Germany (until 1990 former territory of the FRG)	Germany (until 1990 former territory of the FRG)	2007	2	261149		470312		21560		:		913572			981136		118055		566409				412641			113549		1033653		187733	25839				22973.5	8283.5	4178	10512.1								82163475
	Baden-Württemberg	2007	2	24317		59783		:		:		225693			267473		4392		64277				81177			20838		112346		40978	:				2719.1	924.5	555.3	1239.3								10475932
	Stuttgart	2007	2	:		12571		:		:		149034			117085		1752		24952				35587			6961		44237		11099	:				973.4	345	220.2	408.2								3917305
Belgium	Belgium	2008	10		39111		45612		:		33582		42525	7323		42695		:		88905	10554	:		15228	21490		14001		293967			26593	6899	:					983	129.4	175.6	22.7	364.4	1369.2	212.2	10584534
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2008	10		:		:		635		1338		:	:		:		212		4348	34	:		:	:		:		24267			412	:	:					83.3	21	21.1	3	50	108	44.4	1031215
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2008	10		:		:		635		1338		:	:		:		212		4348	34	:		:	:		:		24267			412	:	:					83.3	21	21.1	3	50	108	44.4	1031215
	Vlaams Gewest	2008	10		:		:		:		:		:	:		:		:		64005	:	:		10284	:		9501		184336			:	:	:					617.5	75.5	106.6	12.4	225.7	795.2	109.4	6117440
	Prov. Antwerpen	2008	10		9855		:		:		4485		11865	:		:		:		17954	:	771		2760	4676		:		51368			:	:	:					184.1	18.6	27.7	3.7	60.3	198.8	30.7	1700570
Germany (until 1990 former territory of the FRG)	Germany (until 1990 former territory of the FRG)	2008	10		267115		311098		20478		192103		770377	113549		987402		80018		469142	62554	9565		137416	120184		84404		642398			72568	37882	17649					63	7.9	6.7	2.7	16.7	89.6	12.3	82314906
	Baden-Württemberg	2008	10		26008		32199		:		20882		211434	6097		297221		4052		58693	8208	:		29322	20476		15123		83777			13379	11088	1592					8354.7	1209.3	1304.2	209.2	3595.4	9377.3	2033.3	10738753
	Stuttgart	2008	10		6522		10925		:		6006		135146	989		135542		1448		23697	2933	0		13438	7575		5364		33197			3414	2742	:					1017.9	194.1	179.1	18.1	437.8	1156.8	224.5	4005380
Belgium	Belgium	2009	11		31769		43510		4429		31551		38750	7116		38551		2700		85815	9949	1572		13776	21363		12889		291381			24168	6477	1482					377.6	78.7	77.9	5.6	157.6	391	82.8	########
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2009	11		441		2665		:		1023		4351	1138		3132		:		4381	36	0		354	2525		578		24069			365	789	:					971.9	139.8	155.9	20.3	381.8	1395.8	226.8	1048491
	Région de Bruxelles-Capitale / Brussels Hoofdstedelijk Gewest	2009	11		441		2665		:		1023		4351	1138		3132		:		4381	36	0		354	2525		578		24069			365	789	:					85.1	18.5	21.9	3.3	53.3	111.7	50.4	1048491
	Vlaams Gewest	2009	11		18261		31818		3655		18181		31353	1737		24804		689		62285	5541	:		9384	14704		8778		181108			21444	4934	:					85.1	18.5	21.9	3.3	53.3	111.7	50.4	6161600
	Prov. Antwerpen	2009	11		6057		18382		:		4174		:	:		6542		303		:	1029	750		2452	4734		1484		50329			718	840	85					616.3	84.8	89.8	11.9	236.8	806.4	116.3	1715707
Germany (until 1990 former territory of the FRG)	Germany (until 1990 former territory of the FRG)	2009	11		248693		308020		20205		181845		730167	110309		937119		76119		475044	61180	9743		132615	113830		80391		1590079			65028	35256	16732					184.5	25.2	20.7	2.9	62.9	203.7	33.9	########
	Baden-Württemberg	2009	11		22350		31027		:		19981		196417	7721		284957		4034		59091	7943	:		28072	18850		14148		199280			11974	10513	1337					8338.4	1205.8	1313.2	256.4	3737.6	9571.8	1841.2	########
	Stuttgart	2009	11		5861		10730		:		5617		129101	1174		131179		1455		24101	2796	0		13007	6750		4905		75786			3123	2627	:					998.4	190.1	181.8	23.2	439.9	1201.2	223.6	4007095