According to the Economist last week “Reports of Europe’s death are somewhat exaggerated“. I can only whole-heartedly agree. I think though, it only fair to add, that reports of Europe’s impending old age are almost certainly not, indeed generally it might be felt that the significance of this phenomenon were rather underestimated, than overstated.
Let me explain.
As the Economist article itself points out, here in Europe a good deal more attention has been being focused on the potential impact of climatic change (which is in and of itself undoubtedly an important topic), whilst, and in contrast, comparatively little coverage is being given to our need to develop a population policy:
though every rich country has a climate-change policy, few have a population one (there are historical reasons for that). And just as everyone whinges about the weather, but does nothing about it, so everyone in Europe complains, but does nothing, about population.
Again I tend to agree. Part of the difficulty comes, I think, from our undoubted tendency to try – as the Economist also notes – to simplify what are undoubtedly complex topics. This simplification processes can in itself produce rather sudden and noticeable shifts in opinion, as we have recently seen in some quarters in the case of climate change. What was previously thought by some to be benign, now is thought to be not quite so benign, and in the process a new global consensus emerges, even if comparatively little seems to have changed in the way of available evidence.
And so it will probably be with demography. In part, if this does turn out to be the case the Economist itself may turn out to be one of the guilty parties, since interesting and useful as this article is, it does most definitely fall into the complacent – things aren’t so bad as was feared – camp.
The article makes 6 main points:
i) “This article will argue that pessimism is no longer justified. It would be too much to say Europe’s population is bouncing back. But its long-term decline is starting to bottom out, and is even rising in a few places.”
ii) A long list of US observers – ranging from American observers from Walter Laqueur, an academic, to Mark Steyn, a conservative polemicist – who have been arguing that “Europe is fast becoming a barren, ageing, enfeebled place” are wrong.
iii) That changes in population are not – in and of themselves – either a good or a bad thing in economic terms, since “there is no short-term correlation between population change and wealth” and “Japan and South Korea have even lower fertility than Europe”.
iv) Europe is simply not in decline. “Rather…. it no longer makes sense to talk about Europe as a single demographic unit at all” since “There are two Europes.”
v) Some “very-low-fertility countries can fall into a trap”. (This is a reference to a hypothesis which has been advanced by the Austrian demographer Wolfgang Lutz and his collaborators at the Vienna Institute of demography, although strangely, even while the Economist author uses adjusted data from the VID for the article, Lutz himself doesn’t appear to warrant a mention. I have posted on this hypothesis extensively both on Afoe and elsewhere, and a list of posts can be found here)
vi) “16 European countries, with a total population of 234m, now have fertility rates of 1.8 or more…..They are rare examples of bucking the trend that, as countries get richer, their birth rates fall. Why? There are no obvious answers.”
Of these (iv) (with qualifications see below) and (v) seem to be arguably very much to the point, (vi) is undoubtedly true, (iii) is highly questionable (in substance, though not in the rather constrained form in which the argument is presented, again see below), (ii) is undoubtedly the case, due to the simplistic way in which the argument is often put, and (i) is really not only deeply questionable, but fall foul of exactly the same kind of oversimplification process which the article’s author would want us to reject from Europe’s US critics. A case of double standards?
Well, let’s take a look at what is actually happening.
In the first place, as the Economist argues (and this is undoubtedly one of the strong points of the article) it is simply not satisfactory to talk about Europe as one single demographic whole. There are several Europe’s, and perhaps not two, but four. The general situation can be rapidly grasped by a quick glance at this map which I have put online here.
In the first place we have those countries – essentially France, the UK, Ireland, the Netherlands and Scandinavia – where fertility is at, or near, population replacement rate. The population path here, if you add in a certain quantity of immigration which the comparatively strong economic dynamic of these countries naturally attracts, would certainly seem to be pretty sustainable, and at least a lot more sustainable than in many other countries. As noted above these countries vary considerably in their welfare and tax systems, so it is hard to identify any specific feature which has contributed to their relative stability. This being said, that isn’t the end of the problem, unfortunately, since demographic processes are not only about fertility, they are also about life expectancy, and increases in the latter, which seem to form part of what Federal Reserve Chairman Ben Bernanke recently referred to as an ongoing demographic transition, a transition which is associated with rising population median ages and which is destined, with or without fertility-related problems, to place growing pressure on the health and pensions systems of all OECD countries.
In the second place, and at, as it were, the opposite extreme, we have the former member States of the Eastern Bloc. I single this group out as a special category since they are arguably still operating under the weight of what could well be termed an “asymmetric demographic shock” since their fertility generally plummeted following the coming down of the Berlin Wall. In addition, prior to the coming down of the wall, the mean age at first birth of mothers was significantly below that which could be found in Western Europe (see this map here for an at a glance appreciation) and below ages which are now considered to be the norm for developed societies with services-oriented economies. As a result these countries face what could be called a continuing “birth dearth” as mean first-birth ages move steadily upwards over – and probably over a good number of years to come – as women systematically put off having children to ever-higher ages.
This postponement process can lead many astray into thinking that the impact the process has on Total Fertility Rates (TFRs) is benign, since eventually TFRs may well recover somewhat (if there is not a trap, again see below), and although this debate gets incredibly technical involving comparisons of Completed Cohort Fertility Rates and TFRs, and the study of an issue which has become known as Quantum vs Tempo, one of the obvious impacts is easy enough to understand: with each passing generation the size of the cohort base from which children can be born is reduced, and substantially so – as a result of the missing births. The structural damage which this does to the shape of the population pyramid is known as the negative momentum effect, and this is one of the mechanisms which has been identified as a factor in any possible low-fertility trap.
In the third place we have the ‘Latin’ cultures of Southern Europe – Spain, Italy, Portugal and Greece – where, by and large, significant birth postponement has already taken place (Portugal is something of an outlier here), but where fertility still stubbornly sticks near to the lowest-low TFR 1.3 zone. I think entering the specifics of these countries is going to have to remain beyond the scope of the present post, but my feeling is that Portugal and Italy are much more stuck in the fly-trap than Greece and Spain are (this remains outside my present scope since the explanation of why I think this is the case rests on a development of the economic dynamics of the trap which Claus Vistesen and I are currently working on, which I briefly outline here, and which I sort of spell out in the case of Italy here. In a nutshell, it depends on whether – as a population – you are still young enough to get a housing boom or not).
Fourthly and lastly we have the case of the German speaking countries, namely Germany and Austria (and a part of Switzerland). The German case is by now reasonably well known. Aggregate fertility was, of course, negatively affected by the fertility “crash” in the former DDR, but as the graph appearing in the middle of this post – and which compares the two constituents independently – reveals, fertility in the West is low in its own right, and has been so for a very long time now.
As the Economist notes:
Germany not only has low fertility now, but has had for more than a generation. This suggests that “exceptionally†low rates can persist for decades. Admittedly, points out Michael Teitelbaum of the Sloan School in New York, Germany may simply be odd demographically.
Now while the German fertility pattern is decidedly odd, perhaps one of the oddest of odd features in the recent childbirth patterns there is omitted from mention in the article, namely the relatively higher numbers of women in German-speaking cultures who remain childless (see this chart where you can see the very rapid and significant rise in childlessness – up towards the 25% mark – among German women since the 1950 cohort) and indeed the proportions of women in these cultures who have considered it normal not to have a child. As can be seen in this chart, in answer to the question asked of women in the 2002 Eurobarometer survey about what their “ideal” number of children would be some 16.6% (in the 18-34 age group) declared “none†to be their ideal number of children in Germany and 12.6% in Austria.
These results do tend to give credence to the idea that some part of the low fertility in Germany is structurally different from low fertility in other members of the “lowest-low” group, in that a more significant part of the childlessness may be due to a free and voluntary decision rather than a result of biological infertility produced by excessive postponement.
But high levels of childlessness are not the only significant characteristic of low fertility in Germany, as can be seen from a glance at this chart, which compares the parity composition of childbirth (ie numbers of children) in six EU countries – Italy, Federal Republic of Germany, the UK, the Netherlands, Finland and France – for the 1935 cohort. If we make a direct comparison between Germany and France we can see that not only does Germany have more women who remain childless, of those who have children, a far lower percentage were having third and fourth children.
If we then take a look at the time-series chart for the percentages of children born out of wedlock to mothers in a number of EU countries which I have at the bottom of this post, we can see that in the case of Germany it is noticeable that the percentage of children born out of wedlock remained low in comparison with the UK, Sweden and France right though the second half of the last century, and that the level had stabilized by the 1990s (at around one-sixth of the birth total): this is an interesting result since marriage and the family are specifically protected by the German Constitution and since we have seen how since unification the number of such births has been halved in the east, where “illegitimacy†was previously massive.
So we may well have a rather perverse situation here, whereby “family” (as opposed to child oriented) policy specifically targeted married couples, and – at least in terms of tax concessions – favoured the father rather than the mum, with the result that – given the significant social transformations which were taking place in family types during the period in question – less children where born. Such at any rate is the opinion of the Max Planck Institute for Demographic Research demographer Jan M. Hoem, as argued in this paper (PDF).
So now lets go to point (iii) in my list from the Economist, namely the idea that population change is economic growth neutral. I would say that this was perhaps the most controversial idea in the whole article. The key point to note here I think, is that it is not population SIZE that matters, but population age structure. Changes in age structure effectively produce – as was mentioned in the context of Ben Bernanke and the Demographic Transition earlier – shifts in median ages, and these shifts in median ages do seem to have significant economic consequences. Basically, if we look – yes, actually look – at those societies whose median age has reached the highest level – around 43 – so far – Germany, Japan, and Italy – we can note straight off that each of these has been experiencing economic problems in recent years which to some extent break away from the traditional pattern. I do not wish to go into this in any great detail here (that will be, I think, another post), but basically it could be argued that these three countries all tend to be suffering from congenitally weak domestic consumer demand, and as a result tend to depend on export lead growth for increases in GDP (increases which in the case of Italy remain exceedingly small, due to the inability to meet the export-lead growth challenge).
I have recently gone into all this in some considerable depth in the German case (and here) so I will simply refer the interested reader to this line of argument. But this kind of economic problem will undoubtedly feed-back into the fertility trap problem (if one exists), and in particular by maintaining downward pressure on the disposable income available to young people, both via the tax squeeze that ageing and the associated higher elderly dependency ratios produces (viz, the 3% VAT rise in Germany) and the downward pressure on wages which is being systematic and relentless in both Germany (see this remarkable Q1 2007 wage data from Eurostat, just 0.1% growth in wage costs y-o-y after the boom year of 2006) and Japan (where again wages continue to fall, and here).
So, in summing up, what can we now make of the Economist’s claims that “pessimism is no longer justified” and that “Europe’s population is bouncing back”? Well, I would say that pessimism is rarely justified, since it tends to produce fatalism. On the other hand realism leads me to want to qualify the Economist’s claims in the following way:
* Europe is only bouncing back in parts, so it is hard to draw any real conclusions, in particular a very large part of Europe still has – as can be seen here – around 70% of its population with TFRs below 1.7, and 1.7 is already significantly below replacement level.
* Demographic changes are not processes which only go to work in the very long term, the short term consequences of changing median ages are already real and present.
* The economic consequences of changing population age structures are not growth neutral, but are real and significant.
* As a consequence of all of this we simply cannot afford to continue to give demographic changes the back seat. Europe needs above all policy – rather than complacency – in the face of these changes, and such policies ought to be just as evident in the minds of our citizens as the recent declarations of good intent about the need to act on climate change.
