Tuesday, March 28, 2006

OECD factbook 2006

I have already posted on this over at my own blog, but I still think it merits an entry here at Demography.Matters. The topic is the new issue (2006) of the OECD factbook and specifically the part about migration and population which presents a good entry to some updated basic demographic data and charts.

The most basic yet also telling charts and conclusions can be found in this PDF.

"In 2003, OECD countries accounted for just over
18% of the world’s population of 6.3 billion. China
accounted for 21% and India for just over 17%. The
next two largest countries were Indonesia (3%)
and the Russian Federation (2%). Within OECD, the
United States accounted for nearly 25% of the OECD
total, followed by Japan (11%), Mexico (9%), Germany
(7%) and Turkey (6%).

Between 1991 and 2004, population growth rates

for all OECD countries averaged 0.8% per annum.
Growth rates much higher than this were recorded
for Mexico and Turkey (high birth rate countries) and
for Australia, Canada, Luxembourg and New Zealand
(high net immigration). In the Czech Republic,
Hungary and Poland, populations declined from a
combination of low birth rates and net emigration.
Growth rates were very low, although still positive, in
Italy and the Slovak Republic.

Total fertility rates have declined dramatically over

the past few decades, falling on average from 2.7 in
1970 to 1.6 children per woman of childbearing age
in 2002. By 2002, the total fertility rate was below its
replacement level of 2.1 in all OECD countries except
Mexico and Turkey. In all OECD countries, fertility
rates have declined for young women and increased
at older ages, because women are postponing the
age at which they start their families."

For many of our visitors this might seem very basic, but I still think some clear and worth while conclusions can be drawn.

- Fertility trends; an overall declining rate of fertility well below replacement level coupled with the postponing of birth.

- Population Growth rates; a high divergence within the OECD.

A more broader question here is that if demographics show us the real emerging economies can we use the same measure for OECD?


S.M. Stirling said...

Odd; the CIA estimates show Turkey to be at a TFR of 1.92, well below replacement level.

Edward Hugh said...

I think this must be some sort of error in the CIA documentation you were looking at. The Population Reference Bureau gave Turkey a TFR of 2.4 in 2005 (you can find the PRB link under Population Statistics in the sidebar). Basically Turkey has two fertility regimes (in the same way the US has three). One is in the poorer and less educated Kurdish Eastern part - and there TFRs are well above replacement, and the other is in the Western (Turkish??) part where TFRs are, as the CIA suggest, below replacement.

I have a page on my website about Turkey, and there is a paper linked to there which explains what is happening in Turkey in greater depth:


S.M. Stirling said...

The CIA World Factbook lists Turkey as having a TFR of 1.94 as of 2005, and a crude birth rate of 16.83/1000. According to them the TFR dropped below 2.1 several years ago, and they're usually pretty good at these things.

Iran is listed as having a TFR of 1.82 as of 2005 and as also having a crude birth rate of 16.83

S.M. Stirling said...

Actually, as an aside, the US has rather more than 3 fertility regimes.

Eg., the usual "Hispanic" category includes Mexican-Americans (TFR around 2.7), Puerto Ricans (TFR around 1.75) and Cuban-Americans (TFR around 1.6).

Hispanic fertility levels are higher than the national average but converging with it rapidly, due both to assimilation and of course to the continuing decline of levels in the countries of origin, particularly in Mexico, which is probably at or below replacement level now.

Of course, immigrants aren't a random sample -- they're usually biased towards rural and lower-income individuals, though Cuban-Americans are something of an exception.

American blacks have a dual regime, with lower marital fertility than whites but higher extra-marital levels and much lower marriage levels. Or to put it another way, they differ much more steeply than whites by class and amount of education.

The steep drop in the teen pregnancy rate over the past two decades, especially steep for blacks in the 90's, accounts for part of the convergence of black and white fertility levels, which are now roughly equal.

Likewise, the "non-Hispanic white" group is usually lumped togther in discussions of fertility.

But that's 220,000,000-odd people and they differ very widely by region, religion, and other benchmarks.

The overal TFR for that category is around 1.9-2.00 but within that there are two main groups.

One, which you might call Group A, the "costal-urban-secularist" is around the upper range of European levels, about 1.7 or so, and has been since the 1970's. A typical state domianted by that group would be Vermont, with a current TFR of 1.6. It also has the lowest rate of church attendance in the nation.

The other, Group B, would be the "continental rural/suburban/exurban religious-conservative", with Utah as the archetypical state, and a TFR of around 2.5-2.7 and rising.

All states contain both groups, of course, in differing proportion, and they have rather different responses to the same stimuli; for example, in Group B fertility doesn't drop with the amount of female education the way it does in Group A.

Group B is the basic reason why US TFR's have been slowly increasing since the late 70's/early 80's.

(Apart from immigration, of course, but immigration accounts for only 30% of the increase.)

Group B is steadily increasing its relative weight in the group and among the overall population.

Plus the most recent waves of immigration are assimilating, and in particular intermarrying, much more rapidly than the previous big surge 100 years ago.

American-born Hispanics and Asian-Americans have rates of intermarriage that weren't reached by Polish-Americans or Italian-Americans until after WWII. They're making the linguistic transition to being monoglot English-speakers much more rapidly too.

This is visible in other fields as well -- somewhere between a quarter and a third of US Hispanics are evangelical Protestants now, whereas a generation ago they were almost all Catholic, albeit often nominally.

S.M. Stirling said...

Then there are Asian-Americans, another 'group' that shoehorns together sub-groups of very divergent demographic characteristics.

Eg., people from the Philippines are an entirely different story (higher TFR than the national average) vs. a vs. people from China or Korea (lower than the national average).

Edward Hugh said...

"Then there are Asian-Americans, another 'group' that shoehorns together sub-groups of very divergent demographic characteristics."

Again I agree with this. This is really why instead of looking at this 'ethnically' I am trying to think about it in terms of reproductive cultures (see most recent posts). I'm not saying I have fully understood this yet, merely that I keep trying, and I am learning as I go along (which is why your comments are most welcome). I see understanding what is happening to US fertility as an important key for helping unlock the whole economic growth mystery and also an indispensible platform from which to be able to seriously address the issue of economic global imbalances (like the US-China one).

Edward Hugh said...

"and they're usually pretty good at these things."

I'm sure you're right, indeed I often use world factbook data myself. But I think we have to accept that in this case the CIA have got it wrong somehow. Does this prove that they are, after all, human :).

The data Yahuz (2005) quotes comes from the latest Turkey Demographic and Health Survey, so I don't know where the CIA would be getting their data from (satelite surveilance? :))

As Yahuz says:

"The latest nationwide Turkey Demographic and Health Survey (TDHS) reveals that the current TFR is close to reproduction level, with a wide range of west-east regional disparity."

This is presumeably the data the PRB have drawn on.

"Iran is listed as having a TFR of 1.82 as of 2005"

Well the prb again have Iran slightly higher (2.1) but there is no doubt that Iran's fertility is falling below replacement.

Basically cases like Iran (and Cuba which you also mention, or North Korea) raise questions about whether these countries are not missing out on the economic demographic dividend altogether. If this is the case (Belorusia would be another example), the consequences may well be grave. This would be an example of what is often called 'path dependency'. We need to understand all this much better.