By Dr. Ole Jørgen Anfindsen
Editor, HonestThinking.org
The article Cheating with immigration numbers (”Juks med innvandringstall”), published 15 may 2005 in the Norwegian daily Verdens Gang (VG), is accompanied by additional information below, as well as a further discussion of several aspects. This document consists of two main parts, each of which is in turn split up in a part A and a part B.
· Part 1: Hard facts:
o A offers documentation of just what SSB and Lars Østby have done to deserve my criticism of them.
o B discusses some ‘what-if’ analyses.
· Part 2: Discussions of principal as well as practical issues.
o A discussion of political, ethical, and philosophical problems.
o B some suggestions for the future.
Here I focus on the so-called hard facts. Part 1A contains documentation concerning my contentions about SSB and its demographic researcher Lars Østby. Part 1B presents numbers that hopefully will make it more convenient for the reader to assess the current situation.
To provide documentation that SSB, and not least researcher Lars Østby, has attempted to beautify reality, let’s look at the book “Felleskap til besvær – om nyere innvandring til Norge (Inconvenient community – about recent immigration to Norway), issued by “the leading Norwegian academic publisher” Universitetsforlaget in 1992, with economic support from Utlendingsdirektoratet (Norwegian Directorate of Immigration).
One chapter in said book is written by demographic researcher Lars Østby of SSB. There he writes, among other things, on pages 153-154:
“We do notice from time to time that opponents of immigration make the assumption that present time immigration will produce a coloured majority, or even a Muslim majority ‘ in the lifetime of our children’, ‘within a generation’ or similar. […]
By means of commonly accepted demographic models we have made calculations to find what can be expected in the future, considering various assumptions for future immigration and the observed fertility rates for the group in question. Even when we consider the most extreme alternate will we in 2050 see no more than 20% of the total population having origins in the third world, and then most often one or more generations back in time. The more probable alternatives give significantly lower numbers; see [references to SSB reports].
As of today registered Muslims constitute a small portion of immigrants from the third world. By early 1990 we had approximately 20 000 Muslims in this country. This amounts to roughly 1/3 of all domiciled immigrants from the third world. Even if this portion conceivably may increase, an anticipation of a Muslim majority in Norway must assume that so many ethnic Norwegians convert to Islam that they will constitute a large majority among Muslims.”
As can be seen, Mr. Østby’s contention is that only in an extreme situation will we see 20% (i.e., about 1 million – Norway’s population is expected to grow from the current 4.5 to about 5 million by then) in Norway having their origins in the third world, while more likely this segment will be “considerably lower”. But in 2004, according to SSB, we already had 250 000 immigrants from non-western countries (and this does not include third and later generation immigrants, see below).
For some time now it has been clear that the prognoses for the period 1990 – 2050 have been too low (freely admitted by Mr. Østby during a telephone conversation with me in the autumn of 2004). Norway needs no more than a good 3% annual growth to see more than a million non-western immigrants by 2050. What SSB claimed only a little more than 10 years ago was an extreme maximum is no longer extreme at all. And if the growth rates for our immigrant population continue to be sustained at the levels we have consistently seen since the 1970’s, Norway will have a non-western majority by 2050.
How could SSB be so mistaken? Well, by reading the research reports Mr. Østby referred to at the time he wrote the above, it can quickly be established why. These reports employ a very specific but quite unrealistic assumption as basis for their prognoses. One assumption used in all the reports is that net immigration will be more or less constant throughout the entire prognosis period.
To show that this assumption is unrealistic, it can conveniently be pointed to the fact that the more immigrants are accepted, the more applications for family reunion or entry applications for spouses being brought from abroad, will be submitted. This gives a self-enforcing effect and typically results in so-called exponential growth.
That this is an actual fact is being verified by SSB’s own numbers. These numbers show that immigrant population was exponentially growing in the years before the publication of the reports around 1990, and that this pattern has continued since then (see e.g. the nice graph at the bottom of this web page: www.ssb.no/innvstat/).
Mr. Østby does not offer any reason for his extremely modest prognoses in the above mentioned book but limits himself to referencing some research reports from SSB. In this context, the more interesting report is the following:
Report from Statistics Norway no. 91/10
“The number of immigrants and their descendants up towards the year 2050”
And, sure enough, here we find some very modest prognoses for the development in number of immigrants up towards 2050, but the reason for the low numbers is that they are based on the peculiar and totally unrealistic assumption that net immigration would be constant throughout the period from 1990 to 2050. Obviously impossible with the policies we had at that time, and even more so today
The author of the research report no. 91/10 apparently wanted to prevent any unwarranted interpretations of his calculations, and so emphasized on page 67 of said report:
“Selection of assumptions for the annual net immigration of the various groups is a central point in these calculations. Only if the choices of policies implicit here are interesting as practical or hypothetical possibilities are the resulting prognoses of any interest. We have made our choices based on deliberations about this. But we have no guaranties that our choices are good. Others could have made other choices. Whoever makes use of the results should therefore make up their own mind about whether our alternatives are interesting.”
One possible interpretation of this statement is that the researcher in question has been coerced to build his prognoses on assumptions that he himself is not quite comfortable with. To further guard himself against being misunderstood he adds the following understatement on page 68:
”We assume that in the future, too, immigration will be regulated by the authorities. […] Many circumstances, like e.g. family reunion, could make it difficult to implement [such regulations].
Here it is implicitly stated that should our politicians continue to implement the policies that were in force before 1990, then the results would be totally different from what SSB’s prognoses predicted.
As could be expected, the above predictions do not correspond with reality. To give the reader an idea of how misleading Mr. Østby’s contentions are, we will look at some simple numbers from SSB. These numbers are derived from, among others, www.ssb.no/innvstat/. Note: Most numbers below are put in round figures.
In the above mentioned book Mr. Østby wrote: “At the beginning of 1990 we had approximately 20 000 Muslims in this country. This is about 1/3 all domiciled immigrants from the third world”. It follows that we had approximately 60 000 domiciled immigrants from the third world at that time. Diagrams found on the just mentioned SSB web page, however, indicate that numbers were somewhat higher than the 60 000 stated. If we include immigrants from Eastern-Europe, it looks like we had about 80 000 non-western immigrants here in 1990. (Note: on 14 June 2005 Mr. Østby, together with his colleague Mr. Brunborg, in the context of an article in the Norwegian daily Klassekampen, published figures that indicate that 80 000 non-western immigrants was too low for 1990 – I have asked SSB for clarification on this.)
According to SSB the corresponding number as of 2004 had increased to 249 000. An increase in number of non-western immigrants from ca. 80 000 in 1990 to ca. 249 000 in 2004 constitutes an average annual increase for this group in excess of 8% during that period. (Note: SSB now claims that the average annual growth has been 6.5% - se my note in the previous paragraph).
In contrast, Mr. Østby has contended that it would be absolutely extreme if as much as 20% of the population here should have non-western background in 2050. Mr. Østby must have envisioned an increase in non-western immigration from 80 000 in 1990 to maximum 1 000 000 in 2050. Some simple calculations show that Mr. Østby expected an annual increase in non-western immigration in this period to be a little over 4% at most, but more likely considerably lower, i.e. in the order of 3% annually (some readers may wonder about this but 3% is actually considerably lower than 4%, just as 5, 6, or 7% is considerably higher – see the below tables and you will understand why).
So far evidence indicates that Mr. Østby was wrong, and the situation is now turned upside-down. During the last 15 years the annual increase has been slightly over 8% - in average, but need to come down to about 3.5% annual average the next 45 years to see us end up with “only” 1 million non-western immigrants by 2050. What Mr. Østby presented as an extreme scenario just over a decade ago, does not appear to be extreme at all. But Mr. Østby did not miss out because he is a second rate professional. He missed – for reasons we can only speculate about – because he gave in to the temptation of beautifying reality.
If we look at only Islam’s growth in Norway, the numbers are even more dramatic. Mr. Østby submitted in the reference above that Norway had about 20 000 Muslims in 1990, but that number has since increased to about 80 000 in 2004. This means a quadrupling of number of Muslims in the course of 14 years – or an annual increase in numbers exceeding 10%.
For the record: The statistics from SSB simply show the number of people who are members of various Islamic congregations. Muslims who are not members of such congregations are not included.
At the time SSB published the research reports referred to by Mr. Østby, one tried to make future estimates about “immigrants and their descendants”. That method has been discontinued and in addition there has been made a new definition of the concept of ‘immigrant‘(ref.: www.ssb.no/innvstat/):
This means that if for example a person born in Norway by immigrant parents marries a person from the country of origin (which indeed the majority of the second and third generation from our largest immigrant countries do), then their offspring will not be counted as an immigrants in the statistics of SSB.
There are conceivably good reasons for adopting such a definition, but I find it inadequate in as much as it does not include “immigrants and their descendants”. By not having such a term one withholds important demographic information.
An often heard argument in support of the revised definition presented above is a typical counter question of this kind: “For how long should people really be considered immigrants?” That is, admittedly, an interesting question, and there are hardly any clear and simple answers. But it should be considered fair to look at this in relation to how quickly various immigrant groups become integrated in our society. I cannot see it substantiated that integration of non-western immigrants takes place sufficiently quickly to warrant “disappearance from the statistics” after a mere two generations.
Development over the past 20-30 years indicates that Norway will become a non-western and/or Muslim country in the course of 50 – 150 years from now. It is striking to see how this is under-communicated by the SSB. Further, SSB’s present definition of ‘immigrant’ tends to cover up reality.
Based on the above documentation we can conclude as follows:
Both of the above is of serious concern.
SSB’s action on this issue is serious because political rather than professional considerations have been allowed to dominate. This reflects negatively on SSB.
On 26 December 2004 scientist in Thailand could observe on their instruments that a possible tsunami had been triggered on the ocean floor. For fear of causing undue panic, they elected not to warn about the danger. The result was both a tragedy and a scandal.
I know from experience that anyone who dears criticise SSB will quickly be required to put forward alternative numbers for future immigration. Such a requirement is based on at least one misunderstanding, viz. that one attempts to compete with SSB and produce better prognoses than they do. However, it is quite possible to find faults and shortcomings in SSB’s material without presenting alternate prognoses. Such prognoses would require information about fertility rate, age distribution, expected longevity, migration patterns, etc., information only professional demographists tend to have access to.
Even though I do not present prognoses in competition with SSB, I do present some what-if-analyses. But first a quick look back.
The number of Muslims in Norway has increased from 20 000 in 1990 to a good 80 000 in 2004. By use of mathematical calculations or a little trial and failure with a spreadsheet one will find that this means we have had an average annual growth of 10.5%. This can be illustrated by the following figures (produced by means of a spreadsheet).
|
Annual growth in percent: |
10,5 % |
|
This is the percentage figure x |
|
|
Annual growth factor: |
1,105 |
|
Growth factor, 1+x/100, used from row to row below. |
|
|
|
|
|
|
|
|
|
Year |
Number |
|
|
|
|
1990 |
20 000 |
|
This figure is given by SSB |
|
|
1991 |
22 100 |
|
This figure is 10,5% larger than the one for 1990 |
|
|
1992 |
24 421 |
|
This figure is 10,5% larger than the one for 1991 |
|
|
1993 |
26 985 |
|
This figure is 10,5% larger than the one for 1992 |
|
|
1994 |
29 818 |
|
And so forth... |
|
|
1995 |
32 949 |
|
|
|
|
1996 |
36 409 |
|
|
|
|
1997 |
40 231 |
|
Carefully note that the numbers from 1991 to 2003 are “artificial”, |
|
|
1998 |
44 456 |
|
i.e., they are just intermediate calculation results. |
|
|
1999 |
49 124 |
|
The real number for each year could be above or |
|
|
2000 |
54 282 |
|
below the ones shown here. |
|
|
2001 |
59 981 |
|
|
|
|
2002 |
66 279 |
|
|
|
|
2003 |
73 239 |
|
|
|
|
2004 |
80 929 |
|
This is very close to the SSB figure for 2004 |
An annual growth of 10.5% may not appear all that much, but as can be seen from the table above, this leads to a quadrupling within 14 years. If this growth continues, it will produce a new quadrupling within the next 14 years, and so on. The growth may very well shrink in the future, but this is nonetheless dramatic.
Now, if we look at the number of non-western immigrants in Norway, we see that these have increased in numbers from 80 000 in 1990 to 249 000 in 2004. It is possible to produce a similar table as seen above and find that average annual growth for this group has been roughly 8.4% during that period (but see note above indicating that corrected numbers from SSB might result in average annual growth of 6.5% - still a dramatic figure).
Having shown the actually growth since 1990, we will go on to ask what the future consequences will be with various growth rates. Note: What is presented below is called a what-if-analysis, which is not to be confused with a prognosis. The latter is complicated and requires information about birth rate, expected longevity, migration patterns, etc. A what-if-analysis is much simpler. This is both its advantage and its disadvantage.
Special note for journalists, bloggers, letter-to-the-editor writers etc.: Unless you fully understand the difference between a prognosis and a what-if-analysis, please leave this topic to others.
One may of course ask: Has a what-if-analysis any value at all? Yes, they are very interesting provided sensible assumptions are applied. My assumptions are as follows:
1. The current growth pattern is not linear. It is exponential by nature (the growth curve is getting increasingly steep). It is therefore natural to look at the average annual growth rate over the period, given as a percentage.
2. Future growth will not significantly change from the pattern already established in Norway during the last 15 - 20 years; i.e. growth will most likely be in the area of 3% - 9% annually.
3. Moderate growth rates are the more interesting to analyse. Higher ones are too extreme to be considered here.
Objection 1: One cannot extrapolate into the future like this; that’s not the way the things work. Answer: This objection is very common but is based on a fundamental misunderstanding. When e.g. 10, 25 or 50 years from now we look back to see what has happened since 2005, we will conclude that increase in the number of Muslims or non-Westerners is presented as a certain factor (which could hypothetical be negative – which will mean that number of Muslims has decreased). When we know that factor we can calculate the average annual growth factor. That’s the way things work. And the growth that we currently experience clearly has an exponential component.
Objection 2: Growth will gradually come to a stop all by itself. Sure. And that’s why each column stops when a certain number is reached (which should not be interpreted as a claim as to exactly when, or at what level, growth is going to stop).
Objection 3: Growth will most likely decrease with time. Yes, quite possibly; due to e.g. negative economic trends, lack of housing, etc. But it is also possible that the growth will increase. Both are possible. But for that very reason we include calculations showing just that. And if growth is about 8 – 10% for a few years and thereafter decreases to for example 4%, then the average for the period could end up being 5 to 7% for the whole period. This is exactly one of the possibilities included in my what-if-analysis.
Possible future growth in number of Muslims in Norway
This table shows how the number of Muslims in Norway might increase in the future, given alternative assumptions about average annual growth. Keep in mind that the current population of Norway is around 4.5 million.
|
Alternative: |
|
A |
B |
C |
D |
E |
|
Annual growth: |
9 % |
7 % |
5 % |
4 % |
3 % |
|
|
Growth factor 1 year: |
1,090 |
1,070 |
1,050 |
1,040 |
1,030 |
|
|
Growth factor 5 years: |
1,539 |
1,403 |
1,276 |
1,217 |
1,159 |
|
|
|
|
|
|
|
|
|
|
|
Year |
Number |
Number |
Number |
Number |
Number |
|
|
2005 |
78 000 |
78 000 |
78 000 |
78 000 |
78 000 |
|
|
2010 |
120 013 |
109 399 |
99 550 |
94 899 |
90 423 |
|
|
2015 |
184 654 |
153 438 |
127 054 |
115 459 |
104 825 |
|
|
2020 |
284 114 |
215 204 |
162 156 |
140 474 |
121 521 |
|
|
2025 |
437 144 |
301 835 |
206 957 |
170 908 |
140 877 |
|
|
2030 |
672 600 |
423 340 |
264 136 |
207 935 |
163 315 |
|
|
2035 |
1 034 879 |
593 756 |
337 112 |
252 985 |
189 326 |
|
|
2040 |
1 592 289 |
832 773 |
430 249 |
307 795 |
219 481 |
|
|
2045 |
2 449 935 |
1 168 008 |
549 119 |
374 480 |
254 439 |
|
|
2050 |
3 769 528 |
1 638 191 |
700 831 |
455 612 |
294 964 |
|
|
2055 |
5 799 887 |
2 297 648 |
894 457 |
554 321 |
341 945 |
|
|
2060 |
|
3 222 570 |
1 141 579 |
674 417 |
396 408 |
|
|
2065 |
|
4 519 821 |
1 456 976 |
820 531 |
459 545 |
|
|
2070 |
|
6 339 283 |
1 859 512 |
998 301 |
532 739 |
|
|
2075 |
|
|
2 373 261 |
1 214 586 |
617 590 |
|
|
2080 |
|
|
3 028 950 |
1 477 730 |
715 956 |
|
|
2085 |
|
|
3 865 792 |
1 797 884 |
829 989 |
|
|
2090 |
|
|
4 933 840 |
2 187 401 |
962 185 |
|
|
2095 |
|
|
6 296 968 |
2 661 308 |
1 115 436 |
|
|
2100 |
|
|
|
3 237 888 |
1 293 097 |
Possible future growth in number of people in Norway with non-western background
This table shows how the number of non-Westerners in Norway might increase in the future, given alternative assumptions about average annual growth. Keep in mind that the current population of Norway is around 4.5 million.
|
Alternative: |
|
A |
B |
C |
D |
E |
|
Annual growth: |
9 % |
7 % |
5 % |
4 % |
3 % |
|
|
Growth factor 1 year: |
1,090 |
1,070 |
1,050 |
1,040 |
1,030 |
|
|
Growth factor 5 years: |
1,539 |
1,403 |
1,276 |
1,217 |
1,159 |
|
|
|
|
|
|
|
|
|
|
|
Year |
Number |
Number |
Number |
Number |
Number |
|
|
2005 |
250 000 |
250 000 |
250 000 |
250 000 |
250 000 |
|
|
2010 |
384 656 |
350 638 |
319 070 |
304 163 |
289 819 |
|
|
2015 |
591 841 |
491 788 |
407 224 |
370 061 |
335 979 |
|
|
2020 |
910 621 |
689 758 |
519 732 |
450 236 |
389 492 |
|
|
||||||