Changing Perspectives On Immigration.

Views of immigration are changing. Back in the mists of time, when I first came to the conclusion that ongoing demographic changes were going to be important, the voices in favour of a reconsideration of immigration policy were few and far between. Perhaps the first and most notable of these voices was the UN population division. Now things are different, and a series of recent international conferences and reports highlighting the positive advantages of immigration as an economic motor only serve to underline the fact that discussion of this important topic is very much back on the agenda.

Earlier this week the Japanese Ministry of Intrenal Affairs announced that on April 1 2005 there were 150,000 fewer children under 15 in Japan than there were on April 1 2004 . This was the 24th consecutive annual drop.

As a fraction of the total population, Japan’s 13.8 percent for the 0-14 age bracket is among the lowest in the world. Whatsmore this percentage has been on the decline now unabated for the last 31 years. To get a measure of how far the Japanese have gone we could note that US the equivalent figure is currently some 20.7 percent, whilst in Germany the age group has dropped to an all time low of 14.7 percent of the population.

Now before going any further I want to make clear that this post is not primarily about fertility. It is important to take on board the simple fact that any improved fertility in our societies would – if we assume a labour market entry age of twenty as a rule-of-thumb working measure – not have any impact till 2025, and between now and then increased fertility would only constitute a net fiscal liability. The key question is how some of our economies are going to sustain growth during the intervening years (even assuming fertility could be increased, which is far from clear).

Why does growth depend on population? Well there is no big mystery here, the gross national product of any country is simply the result of combining the two principal inputs – labour and capital – with the relevant technological mix.

Now according to another sub-section of economics – growth accounting – you can break-down those annual economic growth numbers into three components: more labour, more capital, and improved efficiency in the combination of the two (or productivity if you prefer). Despite a lot of obfuscation there really is no big mystery at all here.

Now lets apply this simple understanding to the case of Italy. Italy’s working age population is already declining. A recent OECD report on ageing and employment in Italy suggests that if the participation rates remain constant (ie the proportion of the working age population actually employed doesn’t change) then the 2000 workforce of 23.6 million could be reduced to 16.8 million by 2050.

As the OECD points out, the Italian labour force grew at an average rate of 0.4% per year between 1950 and 2000. Under the constant participation assumption it would decline by 0.67% per year. This change is estimated to knock 0.7% of Italy’s growth potential. This may not seem like a lot, but remember that in recent years Italy has been struggling to achieve 1% growth, so even this small reduction may imply a flatline for Italian output.

It is also important to remember two more things here: firstly that the reduction in labour force is only one of the negative consequences, other factors need to be added (or should that be subtracted), and secondly most financial stability projections for things like pensions and health services assume something like a growth potential of some 2%. Zero growth, if it arrived, would have important consequences.

Now all is not lost here. Clearly the entire Lisbon agenda exists precisely to help to find ways to increase employment levels across EU economies. A lot could be said about this process, but this is not the place. I would merely note that to date we have not been startlingly successful in achieving specified objectives. Additionally there is no good reason why we need to put all our eggs in one basket, and lean on one exclusive policy pillar.

If we need to wait 20 years to increase our workforces via improving fertility, importing workers via immigration is obviously a far more rapid (and indeed cheaper) means to the same result. Immigration also has the added advantage that it attracts people over a range of ages, and thus can help with another aspect of our demographic change: the inversion of the age pyramid. Obviously immigration cannot stop this process, but it can ease it.

So it is that, as I mentioned at the start, the positive desireability of immigration is gradually working its way as a theme into our economic literature.

Perhaps the most recent and possibly startling example of this trend can be found in Japan. The latest economic strategy paper of the Japanese government ? the Takenaka Report – advocates that Japan greatly expand the skill sectors where immigrants are permitted, and that the government take steps to guarantee those same immigrants against discrimination in the workplace. Peter has more info and the original Morgan Stanley link over at EuroPolyphony. Now this may well be a case of ‘far too little, far too late’ in the case of Japan, but it is revealing as an indication of the way things are moving.

At EU level the Brussels Economic Forum at the end of April had a whole session dedicated to The Economic Impact of Labour Migration (a pdf version of one of the papers can be found here).

Again the EU in itsFirst Annual Report on Immigration and Integration published last July had this to say:

Economic theory is relatively optimistic with respect to the economic impact of immigration, suggesting overall welfare gains. Immigrant workers can improve the allocation of workers to firms and may ease labour shortages in areas in which natives do not want to work. As they tend to be more responsive than local workers to labour market conditions, immigrant workers may help to smooth the adjustment of labour markets to regional differences or shocks. Moreover, the increase in human capital from immigration contributes to long-term growth, in addition to the purely quantitative impact of increases in the labour force.

The IMF in last September’s World Economic Outlook had a whole chapter on demographic issues: How Will Demographic Change Affect the Global Economy. The IMF illustrate how migration could help with the aid of a study:

The role that labor mobility could play in helping economies respond to demographic change was investigated in a simple two-country, two-period model…………. The model features an advanced and a developing country, each with a population composed of two age groups (workers and the retired). The population characteristics of each country are set to represent actual UN projections for advanced and developing countries, so that the advanced country ages more quickly than the developing country………..

The implication of allowing full capital mobility (with no labor mobility) was compared with the situation where there is full labor mobility (with no capital mobility). The results?in terms of per capita consumption, a proxy for household?s welfare?of having full capital mobility are broadly similar to those discussed earlier.

Specifically, demographic change will reduce per capita consumption growth in advanced countries and will raise it in developing countries over the next half-century. If there is labor mobility but no capital mobility, the results change. Specifically, the advanced country benefits from labor mobility, and its per
capita consumption loss is considerably less than when there is no labor mobility. This is because the advanced country already has in place the capital stock, and without inward migration the declining size of the domestic workforce?as its population ages?means this capital stock becomes less productive. Consequently, migration cushions the impact of the declining work-force, although growth is still slightly negative over the period.

The IMF goes on to note that migration is, of course, only a ?temporary? remedy to the aging of populations. In the long term, population ageing is a global event, and one that migration alone cannot solve, given that immigrants themselves get old and over time also tend to embrace the fertility standards of the host country. But what this temporary remedy can do is ease an impending crisis and buy time, time to look for better solutions to handling all of this on a global basis.

In conclusion I would simply like to state the obvious: with unemployment running at the 10% level in some major EU countries it will be an uphill struggle to convince some people that systematic immigration is part of the solution to our problems (but then someone once had difficulty getting people to accept that the earth orbited the sun), even with the economic evidence running in favour it will only really be possible to move forward if we can also make progress in improving the labour participation rates amongst the existing population, and that means the Lisbon agenda, structural and labour market reforms etc etc.

Update (8 May): I’ve just notice that New Economist blog has a handy link to a couple pieces on migrant labour in the UK which have some relevance to this post.

28 thoughts on “Changing Perspectives On Immigration.

  1. Edward,

    This indeed an interesting subject but it is also one that is very emotive in Europe. The economics may make sense but there has, over the past few years, been a distictly anti-immigration backlash which has entered into politics – people need to be convinced, and, prejudices overcome ….

    There is also the matter of “integration”. France takes a very different approach to the UK on this. Then there is the Danish approach …

    All that said, for some reason whenever I read all the predictions about the aging population, and the costs involved, I always wonder whether a flu virus, say a mutation of Avian flu, or one of those nasty diseases that for some reason seem to always emerge in Africa, might not come along and solve the problem of the aging population. I hope not and that our governments are properly prepared.

    Avian Flu – What we need to know is a weblog that is quite informative, although it can seem somewhat alarming, that is dedicated to the issue of a possible flu pandemic – but don’t let it spoil your weekend!

  2. Of course no mention of the obvious negatives of immigration. It’s not prejudice or backwards thinking to have a negative view of immigration, in particular in Europe, where the some of the most absurd immigration laws exist and very little debate is involved. I mean really, your comparing 16th century thinking to people who dislike immigration?!? How disingenuous is that? Great way to stifle debate.

    Lisbon isn’t going to happen. Bringing in more immigrants from primarily poor, impoverished, and fanatical Muslim countries won’t solve your problems either. Having more kids will. God forbid you should have more children at the expense of ‘fiscal liability’. The problem isn’t that people need to be convinced of anything, but rather Western Europe’s absurd immigration policies need to be rectified.

    Good luck on that one.

  3. Fear you not! There is a solution: more robots and machines.

    Robotic Nation Evidence
    http://roboticnation.blogspot.com
    http://marshallbrain.com/robots-in-2015.htm
    http://marshallbrain.com/manna1.htm

    Where is the Robot-Valley?
    “How many robots are now working out there in industry?
    Worldwide at least 800,000 units (possibly the real stock could be well over one million units), of which 350,000 in Japan, close to 250,000 in the European Union and about 112,000 in North America. In Europe, Germany is in the lead with 112,700 units, followed by Italy with 50,000, France with 26,000, Spain with 20,000 and the United Kingdom with 14,000.”
    http://divedi.blogspot.com/2004/10/where-is-robot-valley.html

    See also:
    The Pentagon’s Robotic Army
    http://divedi.blogspot.com/2005/02/pentagons-robotic-army.html

  4. “I always wonder whether a flu virus, say a mutation of Avian flu, or one of those nasty diseases..”

    I presume Peter you wouldn’t want to be amongst those affected.

    Seriously though I keep following the avian flu issue and it seems entrenched in Vietnam, possibly mutating to spread on a person to person basis. Quite preoccupying. In general I prefer not to speculate about this type of thing. Curiously though, those economists who have tried to find parallels in history for what is now happening have talked about the consequneces of the ‘black death’, following which population was seriously reduced, and where you had a century or so of economic dislocation and even falling living standards as an aftermath.

    Any pandemic wouldn’t just involve old people, young children (eg) would also be disproportionately affected. I don’t think anyone should wish for this kind of thing even in the craziest of dreams.

    “Where is the Robot-Valley?”

    Now robots is an interesting question. I am convinced that of course as technology they can help enormously in making life easier for older dependent people, but the economics of robots is another question entirely.

    Lets do a ‘thought experiment’ for a moment, and imagine a world populated entirely by robots. Would what we term ‘economic science’ operate. I doubt it. What you would have is one giant machine , whose functioning you could understand by studying the technical science of ‘robototics’, not the human science of economics. (I guess of course, this depends what views you hold on Azimov’s laws of robotics and things like that which are generally beyond me :).

    The whole problem is fairly complicated, and you need to keep in mind the facts that robots could probably have almost ‘perfect information’ about the workings and plans of other robots, perfect markets could operate, there would be complete clearing of markets, and no “multi factor” productivity. Tricky things like that.

    In our world, having extra robots working in production would only be like having a lot of very capital intensive industry: say like a modern petro-chemical factory, or an automated steel rolling mill.

    Now if the machinery for these plants were also designed and manufactured by robots, and if the robots doing this activity were a universally available and widely distributed technology, then costs of doing all this (apart from the raw material costs) would be relatively low.

    If, as is more likely, know-how was centered in one country (in the real world Japan seems to be a clear case in the offing) and strategic raw materials to be found in a limited numbers of others, then these two groups would be clear beneficiaries, and the other ‘technology’ and raw material importing countries would in fact be worse off.

    How far Japan can address its population problems by developing a limited number of ‘super companies’ which generate wealth by keeping the rest of the world supplied with robot technology, and then redistribute the wealth to support its ageing domestic population is an ‘open’, empirical question, where only time can give us the answer.

    OK, this is all a bit muddled and ill-organised, but I guess there is some food for thought here. Interesting problem robots :).

  5. “it is also one that is very emotive in Europe.”

    “There is also the matter of “integration””

    Yes I understand this, and this is why I tried to stick in this post to the economic arguments. If people were better informed on this level I perhaps idealistically hope that the debate could be less emotional. The ’emotive’ dimension here has even notably extended recently to the movement of fellow European citizens from the accession countries of the east.

    “I mean really, your comparing 16th century thinking to people who dislike immigration?!?”

    No. I think you missed the point. I was comparing the ‘common sense’ and the reflective views. Looking up at the sky, the sun seems to be moving. Looking at the unemployment numbers in a country like Germany or France you could come to the conclusion that introducing more people would be madness. But the common sense response may be misleading. Digging a bit deeper you may discover that – counter intuitively – more people is precisely what you need.

    “Lisbon isn’t going to happen”

    Do you have some information on this? As I understood it this agenda was moving forward, and that some progress was being made.

    The Brussels Economic Forum (linked to in the text) had a session dedicated to the Lisbon agenda. The following seems to be a reasonable summary of the state of play so far:

    http://europa.eu.int/comm/economy_finance/events/2005/bxlforum0405/doc8en.pdf

    “Having more kids will. God forbid you should have more children at the expense of ‘fiscal liability’.”

    Again I don’t think you’ve quite followed my argument. I am not arguing *against* having children – nor am I arguing in favour either, since that is an individual decision for each person and couple (another thing would be that we could have more child-friendly and mother-friendly tax and social policy).

    My point is that in the context of the current EU problem of reducing populations of working age and growing government debt, children born tomorrow won’t help in a time scale of the coming twenty years. Indeed insofar as fiscal sustainability could become a problem new children (welcome as they may be for other reasons) only add to costs.

    Generally I am sceptical that declining birth rates – which is a global phenomenon – can be easily reversed. Even the US, whose demographic position is notably better than the EU, only enjoys this respite thanks to massive immigration from the mid-eightees on.

  6. “even with the economic evidence running in favour it will only really be possible to move forward if we can also make progress in improving the labour participation rates amongst the existing population, and that means the Lisbon agenda, structural and labour market reforms etc etc.”
    But reforming the labour market is the last thing European politicians – left or right – really want to do. Even the Lisbon agenda could not fundamentally move things forward in this regard.

  7. “But reforming the labour market is the last thing European politicians – left or right – really want to do.”

    They may have little appetite for this, but continuing low growth and mounting fiscal problems are really leaving them with little alternative.

    “Even the Lisbon agenda could not fundamentally move things forward in this regard”.

    Are you sure about this? This is a strong statement. It’s one thing to say – as one commentator does – that Lisbon won’t happen, and something else entirely to say that even with the Lisbon goals achieved nothing would have really changed.

    Incidentally, when I say ‘Lisbon’ I’m obviously not thinking of the silly rhetoric, I’m talking about the concrete proposals which are emerging in the light Lisbon (again see papers in the linked conference).

  8. It’s an important issue. Immigration will be increasingly important and an enormous challenge particularly for nations that define themselves along ethnic lines. That continues along a spectrum with some nations (like Japan) very much ethnically defined and others (like the United States) less so.

    I think you might want to re-examine your assumptions. First, what is the degree of social mobility in the society? If it’s low (as in much of Latin America, for example), immigrants just don’t replace lost workers in the upper strata of the native population. Second, are the fertility rates for the various strata in the society the same? If not and the degree of social mobility is low, immigration may merely create intense competition for low-level jobs. Third, what sorts of workers are being replaced? 10,000 doctors of medicine are not replaced by 10,000 (or even 20,000) day laborers. Fourth, recall that immigration is disruptive for both the destination nation and the source nation. Fifth, where will all of these immigrants come from? In the case of the United States and Mexico, for example, if trends continue the dependency ratio in Mexico will actually be higher than in the United States. That makes it unlikely that a continuous stream of Mexican immigration is the long-term solution to U. S. dependency ratio issues. China is already below replacement level (as is much of India). The only really good candidates for sources of new immigration are Africa and the Middle East.

  9. Edward

    Nice to have you writing again for all. Your ideas were missed.

    Anne

  10. [in] a world populated entirely by robots [would] ‘economic science’ operate. I doubt it.

    Why not? Economic science is little more than an analysis of complex systems and ‘organizations’ attempting to optimize some output on the basis of incomplete information. This would not change fundamentaly if the world was all people, all robots or some mixture of the two. Of course, the results of the analysis may differ for machines 🙂

    Economic theory and closely related game theory already provide the basis for management and organization of ‘extremely complex systems’ such as gangs of autonomous robots. Perhaps, the biggest effect on economists will be to turn their science from a mainly descriptive discipline to more of a proscriptive one. Like experiments in communism but without the human cost when it all goes wrong.

  11. Even the US, whose demographic position is notably better than the EU, only enjoys this respite thanks to massive immigration from the mid-eightees on.

    Well, no. The “native” US population is growing, though slowly — TFR around 2.15.

    Adding immigrants boosts growth considerably, sure. But the US’ population would be growing even without them.

    The /why/ of this is a fascinating question. One interesting data point: families are smallest, and TFRs lowest, in New England. Meanwhile the biggest families and highest TFRs are found in a handful of western states, including Arizona, Texas, and (surprise) Utah.

    — It’s a bit tricky because TFRs are a composite statistic. During a period when women are shifting towards having children later, TFRs will fall — sometimes to below replacement rate — temporarily. Yet long-term population growth will still be positive.

    This isn’t a hypothetical situation, BTW. This seems to be exactly what’s been happening in France over the last generation. France’s official, sub-replacement TFR is partly the result of Frenchwomen having children later and later. So it disguises the fact that the French population is probably going to grow by 10 million over the next 25 years or so. So the low TFRs in the Northeastern US may conceal the same phenomenon. (Though I suspect not.)

    Doug M.

  12. Doug: you raise two questions, migrant fertility, and the composite TFR.

    On the first issue it seems to me that there is some evidence that those societies with relatively higher immigration (apart from the US the UK and France) have TFRs which are much nearer to replacement rates. There are also what appear to be sound intuitive reasons why this should be.

    Immigrants on arrival tend to be clustered in the 20 – 40 age range, which means you give an automatic boost to women in the childbearing age. This then obviously has an impact on TFR.

    Secondly migrants tend to come from cultures where age at first child is *lower*, and where the number of anticipated children per family tends to be slightly higher.

    The studies suggest that after the second generation this gap closes: ie the number of children considered desireable drops.

    (Current data in Spain make quite explicit that the steady recent rise in fertility is due to the recent massive immigration – in fact since 2000 proportionately much greater than anything seen even in the US: 1.5% of the population per year – and a quick glance around the nursery schools reinforces this impression).

    A further piece of evidence might be that in the 70’s US TFR’s were dropping below replacement rates just like all other OECD countries, and that it is only after 1988 that levels have climbed back up again (some of this is of course consonant with increasing fertility among older women, of which more later).

    Now in the US you also keep data on ethnic identities, which enables you to break the numbers down a bit. For example I have seen numbers for ‘Anglo’ TFRs which are around 1.7 and in this sense broadly consistent with some European readings post the ‘age-displacement’ TFR rebound.

    A significant part of the post 1988 migration has come from Latin America, and the Hispanic TFR’s seem to be significantly higher.

    Obviously this issue suffers from getting ‘overcharged’ emotionally. Patterns of Hispanic TFR’s are perfectly explicable culturally, as I am indicating.

    In fact of course all of Latin America is now moving through the first part of the demographic transition (where TFRs fall to replacement level) and some regions are now entering the second stage (where TFRs fall below replacement): this is later process may well now be present in some parts of urban Mexico. I’m not up to date on country of origin data for recent US migration, but I suspect Mexico is slowing down (and will probably eventually ‘dry up’ as a source), whilst new points of origin – like Ecuador and Columbia – may be on the rise.

    So as fertility drops in Lat Am the ‘boost’ from cultural differences on arrival will also gradually fade out.

    Incidentally, it would be interesting to see a study on distribution of TFRs and immigration: as an outsider I have the impression that New England may have received relatively less immigration from the south, whilst in the case of Arizona and (certainly) Texas this could be important. Utah is surprising, but since I know relatively little about Utah (not to say next to nothing) I cannot say anything useful. California also – unsurprisingly – seems to be experiencing a ‘relative boom’ in fertility.

    Now on the statistical issues of the composite TFR. You are undoubtedly right: falling fertility has two components, less absolute numbers of children per women, and later age at first child.

    The point is that while this ‘postponement’ of having children is misleading statistically since some of the children are born later, it is important to note that this process is ongoing, and that the collective average age at first childbirth is still rising in all OECD societies, and may continue to rise for some decades yet.

    What does this mean: that the initial rapid rise in age at first child sends the numbers plummeting: to the 1.2 TFRs seen in Italy and Spain. These numbers are artificially low, and there is then a ‘rebound’ but only up to around 1.7. Without immigration (and repeat, for cultural reasons migrants may have the first child earlier) it is extraordinarily difficult to get much above this level.

    Apart from the statistical component, it is important to bear in mind that the rise in childbearing age has a biological component: the older the woman the less fertile (basically). And also there is a strong social component of what we could call ‘couple instability’ as you move up in the thirties age group: ie people separate more. More people are living alone. At the aggregate level such factors affect overall TFR’s, and there is no getting away from this.

    Well, a lot of issues here. And it is impossible to do then all justice in one comment. Still I hope the thinking lying behind that paragraph you picked up on is now a bit clearer.
    You do in the US keep

  13. “[in] a world populated entirely by robots [would] ‘economic science’ operate. I doubt it.

    Why not?”

    As I say I think this is an extraordinarily interesting and stimulating question. Doing the ‘thought experiments’ here, can clarify a lot of issues.

    In the first place there is a basic strategic decision whether or not the robots have “Kurzweil type” properties. (ie where they spiritual machines). I am assuming that they don’t. If they did, then the distinction between an economy which was all people and one which was all robots would disappear as an ‘in principle’ question, and all you would be left with would be the probable greater efficiency of the latter: in which case ‘watch out human economy’.

    So I assume that the robots are machines of some ‘non-spiritual’ variety, running on some kind of algorithm or another.

    Now we could separate three cases:

    Case a: an all robot universe.

    Case b: a universe populated by two people who compete with each other having access to the same robot technology and natural resources.

    Case c: A universe composed of a number of small open economies where one of these economies has exclusive patent rights over the whole range of robot technology. For ease of exposition lets call this economy ‘Japan’.

    Now in case (a) it is my intuition that you would not have an economic system as we know it.

    “Economic science is little more than an analysis of complex systems and ‘organizations’ attempting to optimize some output on the basis of incomplete information.”

    I agree more or less with this description. The point is in the fully ‘robotised’ world why would there be ‘incomplete information’. Essentially modern economic science is founded – following Marshall – on the so called ‘scissors’ of demand and supply. The robot world would effectively remove one blade from the scissors: the demand related side. What you would have would of course still be a complex system, but one why would simply be run on the basis of an ‘input/output’ programme. Incidentally many existentialist philosophers have already suggested that human life is essentially ‘meaningless’. Somehow, and against all odds, we have managed to stand our ground, and keep going. But for the life of me, I can’t manage to see what the ‘point’ of the robot world would be :).

    “Like experiments in communism but without the human cost when it all goes wrong”.

    Obviously your reference here is pretty much to the point, social science in the robot world would be something like the historical materialism once famously advocated by Bukharin.

    Now case (b) is even more interesting. Would these two people become richer and richer? Against all odds my intuition is a resounding *no*, and seeing why is where things get fascinating.

    Imagine for a moment that we live in a world of zero inflation, zero population change, and a population column rather than a pyramid. Now imagine technology goes on changing. Would GDP grow? I don’t think it would. Economically speaking we would be no better off, and no worse off with the passage of time.

    However in terms of quantities and types of goods at our disposal, our world, and of course what people often misleadingly call our ‘living standard’ would change over time beyond recognition. That is the physical quantity of goods and services you could consume would rise and rise, but your ‘economic’ wealth, measured in terms of some standard currency, would be unchanged.

    So lets get back to our two people. Imagine each of them (lets have a nice egalitarian world) started out with 100 units of some agreed common currency. Then at the end of time period ‘x’ (however big you want to let ‘x’ be) each of them would still only have the same 100 units. Meantime off course the planet would be littered with stockpiles of essentially ‘useless’ products, but that’s another story.

    Now case (c). Japan supplies robot technology to the rest of the world to help with the declining labour force.

    Well initially Japan would earn some sort of Schumpeterian monopoly rent for supplying the technology. This would enable the wealth from super-rich Japanese corporations to be redistributed and maintain the rest of the Japanese economy afloat. Patent protection would be important, since otherwise more would enter the market and these gains would disappear.

    But wait a minute. Japan would be getting richer at the expense of all the remaining economies due to a drastic improvement in its terms of trade. But this would simply divert resources to Japan and lead to the gradual relative impoverishment of the rest. Initially Japan would probably lend money to the others to keep buying, but as the realisation spread that this money could never be paid back, interest rates would rise, eventually prohibitively, demand would drop off, and finally Japan would have a tremendous surplus of robot products which it would have difficulty selling. Consequently the price would drop, and the initial advantage would gradually disappear.

    The only real lesson from this last case is that any global economy with ‘massive imbalances’ is unstable and ultimately unsustainable. This lesson, however, may be directly applicable more to matters rather nearer home than towards the actual robot case.

  14. “where will all of these immigrants come from? In the case of the United States and Mexico, for example, if trends continue the dependency ratio in Mexico will actually be higher than in the United States. That makes it unlikely that a continuous stream of Mexican immigration is the long-term solution to U. S. dependency ratio issues.”

    Dave, you raise a number of ‘to the point’ questions.

    I think the issue of the impact of migration on the ‘sending’ society is an important one.

    There has been a good deal of study of the pattern of migratory processes, and one of the best examples of current work is perhaps the Harvard economic historian Jeffrey Williamson:

    http://post.economics.harvard.edu/faculty/jwilliam/papers.html

    At the above link you will find a variety of relevant papers.

    Now migratory processes in modern societies (I would argue that in fact migratory processes form an important part of our collective history as a species, but let’s stay with the modern world for now) have certain identifiable characteristics. In particular they need to be located in terms of what is known as the ‘demographic transition’ (which, as I am indicating I would be inclined to divide into two phases).

    Thinking aloud as I am writing, the demographic transition ‘kicks off’ at the end of the 18th century in the UK with a small rise in fertility and then a huge population explosion which has, over the last couple of centuries, spread right round the globe.

    After the initial ‘burst’ the pace of expansion starts to slow, and fertility rates fall, following – as we are noting – an overshoot path which takes us initially way below replacement rates.

    Now for thousands of years before the UK industrial revolution our global population as a species was in some kind of ‘homeostatic equilibrium’ with our environment: the rate of population growth was generally pretty small.

    So – and this is the thinking aloud bit – it is quite probable that at some stage this ‘homeostatic equilibrium’ will be recovered. The problem is that the time scales involved here are very large indeed, and – looking at current global momentum – we are going to remain seriously out of equilibrium for a century or more at least.

    The point I would stress here is that the rapid acceleration of population which accompanies the ‘explosion’ characteristic of the first phase of the transition produces a whole array of difficult problems while are, alas, already only too well documented. So……… it could be anticipated that the very rapid ‘break’ and reversal of this process which is characterising the second phase of the transition will have consequences which are difficult to foresee and difficult to handle too.

    Now one of the points about the first phase of the transition is that during this stage most countries pass through a period of being ‘population exporters’. This happened first in Europe, with North Europe being first, and then Southern Europe following.

    The poorest of poor countries are not ‘exporters’. They are just stuck in a poverty trap. (See Williamson various on this). Then, with a certain improvement in economic performance and decline in fertility, the migration process gets going. It could even be that exporting migrants could be an early indicator of coming progress in development.

    So it is that as you suggest, Mexico’s situation may be changing. In general in Latin America the Spanish experience suggests Ecuador and Columbia (followed by Peru and Bolivia) as important source points. But if things go well in development terms Lat Am may no longer be a source of migrants in a generation or so.

    As Williamson suggests the next major wave may well come from Africa (although AIDS adds an important new dimension in this case). And then soon mass migration may become a thing of the past.

    So I am clear: migration is not a permanent solution. In the medium term however it can buy us time, and this could be tremendously important if we are to find ways of dealing with the unwinding of the second phase without sending all our economies into crisis.

    Remember it has only been during the last few years that we have seriously started to talk about all this. We have no ‘ready made’ answers, and we still are not clear about economic and social consequences. One thing we may well need is time, immigration policy, wisely used, can help buy us that time.

  15. Firstly, to which extent is “working age” a statistical fiction? And what will medical science do?
    Secondly, this depends on having a comparable level of economic participitation in the labor market among the immigrants. Thinking about, eg French banlieues, how realistic is that?
    Thirdly, to which extent will using a quick fix hurt in the long run lessing pressure to solve the underlying problem?

  16. OK Oliver, I’ll take your third point first.

    “to which extent will using a quick fix hurt in the long run lessing pressure to solve the underlying problem?”

    The question here is: what is the underlying problem? This whole demographic transition thing is a trajectory which runs over centuries. You don’t have to be especially Malthusian to imagine there are built-in stabilisers, and that eventually some new homeostatic state will set in, but with what population level, what combination of robotics, what level of genetic engineering? I have no idea, and I don’t want to get into this since it is a problem for future generations.

    Our problem is now, and let’s say over a time horizon of a decade or so. Here we can do things, and I don’t think immigration is a ‘quick fix’ in quite the same way as running up unsustainable fiscal deficits is.

    I think it is an intelligent policy option, along with others. I’m not sure what the ‘hurt in the long run’ on-cost is.

    As the IMF indicate, we have globalised capital markets without globalising the labour ones to the same extent. There can be gains for everyone by allowing increasing global labour movements.

    Of course all of this is liable to the same sort of objections as ‘globalisation’ in general is (eg cultural impacts) but that is another, and much more explored debate.

    “And what will medical science do?”

    The answer to this is that we don’t really know yet. Of course there are lots of ‘tremendous’ speculations, but to date there is little to show. What we do know is that medical science is able to increase general life expectancy more rapidly than it seems able to increase the desired working life. Germany and the US are moving towards retirement ages of 67, but this seems to be being greeted with a notable lack of enthusiasm.

    Meantime medical science is increasing life expectancy for the over 80’s. But the costs of doing this are increasing rapidly. So – leaving other things equal – medical science may well serve to make the problem worse, at least in the short term which is what I’m looking at.

    Certainly, and against all cynicism to the contrary, I keep mentioning the Lisbon agenda, and this envisages a longer working life and higher participation rates among older people. I am sure that this, if achieved, will help.

    “to which extent is “working age” a statistical fiction”

    Of course, I completely agree. This is why I keep harping on about 20 as a viable estimate of the number of years involved before someone enters the labour force, and why I keep trying top avoid applying the 0 – 15 category which invariably shows up in the population tables.

    Having said this continuing to use these age groups has some justification as it enables comparative studies of societies across time.

    That being said, it’s about time we changed this, as it is time we changed the way we change the way we measure ‘unemployment’. In Japan unemployment is dropping, but so too is the labour force, so from an economic point of view the statistic in itself is less than useless.

  17. No, working age is not merely statistical it is both a cultural construct and, in some cases, a physical reality. If you don’t believe it’s a physical reality, try sending your 70 or 80 year old grandfather into a coal mine.

    One of the problems we’re running into with our systems of social services is the apparent conflict between the goals of equality and justice. It doesn’t appear unjust to ask an office worker to work for a few more years while they’re still physically able. But physical requirements for a farmer, miner, fisherman, or factory worker differ from those of an office worker (and from each other). Establishing a retirement age may be equal but it has disparate impact.

    Don’t downplay cultural constructs, either. They’re as real as the ground. The employment rate for those over 60 is much lower in China, for example, significantly lower than in the West. Health issues don’t appear to completely explain the disparity: the veneration of elders even in post-Mao China may be a factor. How this translates into the modern youth-oriented Chinese economy is a mystery.

  18. The question here is: what is the underlying problem?

    In the short run, probably health care.
    And I wouldn’t limit this to a decade. It’ll be there for at least 30 years. But what makes you think that there is a steady state of population in an industrialized society? For all we have hard evidence, the natural state of such a population may be fluctuating.

  19. Dave:

    “Establishing a retirement age may be equal but it has disparate impact”.

    I accept the force of this point, but this just complicates things further.

    “Don’t downplay cultural constructs, either.”

    No, I don’t and I hope you haven’t given that impression. In the OECD world since 68 we have probably had an increasingly ‘young’ oriented society: this may well be about to change. Certainly there will be plenty of money to be made out of catering for the needs of old people. These are demand-related items, which here I have been deliberately leaving top one side.

    One thing does worry me though: in China, and in earlier times in Europe older people were respected. In part this was because they were relatively few in number, and because their experience was seen as having value. My worry is that now there will be many more, and that maintaining them (us) will be very costly, and (and this is a separate argument for another post), with an increased rate of technological and social change meaning that the accumulated experience has less apparent value, well my worry is that social attitudes towards the elderly could become very negative indeed in some sections of the population.

    “In the short run, probably health care”

    But then I don’t see how fleshing out the pyramid will be a downside now. Obviously 30 years out going forward more population now will mean increased health costs later. What I am worried about is relatively short term buckling of the financial system under the weight of inverting the pyramid too rapidly.

    “But what makes you think that there is a steady state of population in an industrialized society.”

    This is just a guess. There is no way we can know this, I’m suggesting that things may achieve a new steady state, not that they have to, and I am suggesting that this would appear to be at least a century away due to inbuilt momentum globally going forward.

    I know it is probably only a case of you using a convenient shorthand, but obviously we won’t be talking about an ‘industrial society’. Industry is already responsible for less than 30% of GDP in most of our economies. What I’d like to do is strip out reference to any particular set of social arrangements and look at this on a very abstract – species/environment – level. In this case either we could move on a continuing ‘boom-bust’ cycle, like we’ve had over the last 300 years, or things could start to steady themselves up. My guess is that the latter could happen, but it is only that, a guess. And in any event we most likely won’t be around to see :).

  20. Returning to Doug Muir’s points, I do think it is important to distinguish between the US case and all the other OECD societies. The US faces a much easier scenario looking forward than any of the rest. (Of course I would want to ask that if the better TFR in the US doesn’t reflect immigration, what else is there that would make the US relatively unique?).

    At the same time the US pyramid still looks decidedly wobbly:

    http://www.un.org/esa/population/publications/migration/usa.pdf

    And it is, of course, important to remember that the changing dependency ratios are a result of two processes, declining fertility and increasing life expectancy. The US is going to be affected by this latter component just like all the rest.

  21. But then I don’t see how fleshing out the pyramid will be a downside now.
    Fleshing out the pyramid is a very good idea, anything else being equal.
    However, it isn’t. This will take resources. Initially monetary resources, after that, political capital. What worries me is that governments that fight a hard battle to allow increased immigration will not raise retirement age or curtail benefits.
    Secondly, should the initial monetary cost be spend that way now? Or should it better be spent on eg. Alzheimer research?
    Thirdly, practically we must recognise that there’s a limit to the amount of immigration that societies can stand. Is it per annum or a grand total? If the latter, how do we optimally time immigration?
    Fourthly, accepting the point that part of the current economic woes is caused by demographics, is the damage reversible and how long will it take?

    what else is there that would make the US relatively unique?
    Population density, a relatively unbroken history of prosperity, frequent foreign wars, size

  22. One thing does worry me though: in China, and in earlier times in Europe older people were respected. In part this was because they were relatively few in number, and because their experience was seen as having value
    There’s a fundamental difference between a literate and an oral culture. In a literate culture when you want to know something you look it up in a book. In an oral culture when you want to know something you find somebody who knows. The old have been around longer and, presumably, know more. Literacy can erode the value of the old as transmitters of knowledge.

  23. Hi Edward,

    Couple of points here.

    Apart from the statistical component, it is important to bear in mind that the rise in childbearing age has a biological component: the older the woman the less fertile (basically).

    In rich, post-demographic transition countries, this effect is pretty secondary. If the average woman only wants 2 kids anyway, it’s probably no big deal if she starts at 31 instead of 21.

    Frex, in the US — and especially in the Northeast — teenage births have been dropping for the last couple of decades. Meanwhile births to women over 30 have been rising at a rate of about 3% per year, doubling since the early ’80s:

    And also there is a strong social component of what we could call ‘couple instability’ as you move up in the thirties age group: ie people separate more.

    It’s a lot more complicated than that, especially if you’re looking at people with kids. Short version: in the US, at least, divorce rates drop smoothly with age of first marriage.

    Immigrants on arrival tend to be clustered in the 20 – 40 age range, which means you give an automatic boost to women in the childbearing age. This then obviously has an impact on TFR.

    Secondly migrants tend to come from cultures where age at first child is *lower*, and where the number of anticipated children per family tends to be slightly higher.

    Yeah, but this is not the main reason for higher USAn TFRs and birthrates. It helps, yes, but it’s secondary. Mexicans make up the biggest immigrant group in the US, and Mexican-American family patterns converge rapidly on USAn norms, becoming almost indistinguishable by the second native-born generation. The fertility boost thus comes only from the immigrants themselves and their children.

    Since the US Census tracks data by race and state, it’s also possible to tease out some interesting data bits. So, for instance, white non-Hispanic Texans have a TFR of 2.13: they’d be well over replacement rate even without immigrants or their children.

    A further piece of evidence might be that in the 70’s US TFR’s were dropping below replacement rates just like all other OECD countries, and that it is only after 1988 that levels have climbed back up again

    Yeah, but US rates never dropped as low as European or Japanese.

    Now in the US you also keep data on ethnic identities, which enables you to break the numbers down a bit. For example I have seen numbers for ‘Anglo’ TFRs which are around 1.7

    No, that’s too low. You want to look at National Vital Statistics Reports, Vol. 52, No. 19 (5/10/2004)

    http://www.cdc.gov/nchs/data/nvsr/nvsr52/nvsr52_19.pdf

    Averaged data over three years, from 2000 to 2003.

    For a detailed look at one state, with projections:

    http://txsdc.utsa.edu/tpepp/2004projections/2004_txpopprj_txtotnum.php

    — scroll down and stare at the “0.0” scenario, which assumes the state sealing its borders to all immigrants. White and black (native) populations continue to grow slowly; Hispanics grow quickly at first, then slow down to native growth rates. The population as a whole grows by about 30% by 2040.

    I’m not up to date on country of origin data for recent US migration, but I suspect Mexico is slowing down

    Actually, not. But that’s another story.

    Incidentally, it would be interesting to see a study on distribution of TFRs and immigration: as an outsider I have the impression that New England may have received relatively less immigration from the south, whilst in the case of Arizona and (certainly) Texas this could be important.

    It gets complicated. Short version:

    New England — very “advanced” native demographics, with late marriages, late first births, and small families. Not much immigration. Result: declining TFRs, low birth rates, low or zero population growth. (Although note that New England overall has a TFR around 1.70, which is still higher than most countries in Europe.)

    Mid-Atlantic (New York, New Jersey) — advanced demographics, though not as much as New England. Lots of immigration. Native population growing slowly or not at all, but decent growth from immigration.

    The South — less advanced demographics… earlier marriages, earlier first births (and more divorces). Very little immigration outside of Florida. Most Southern states have modest growth rates, almost entirely driven by native growth and TFRs around 2.1 – 2.2.

    The Southwest — even less advanced demographics (partly a product of _internal_ immigration, which is yet another story) plus lots of immigration. Result: growth at a rate more typical of the Third World than the First. Arizona, for instance, went from 750,000 people in 1950 to 5.1 million in 2000, and is projected to have 8.2 million by 2020. Texas’ population has nearly doubled since 1970, and is expected to double again by 2040.

    Utah is surprising, but since I know relatively little about Utah (not to say next to nothing) I cannot say anything useful.

    Mormons.

    California also – unsurprisingly – seems to be experiencing a ‘relative boom’ in fertility.

    No, actually California looks more like New York — advanced native demographics, lots of immigrants. The greater growth rate comes from internal immigration: people moving to CA from Kansas, more or less.

    You are undoubtedly right: falling fertility has two components, less absolute numbers of children per women, and later age at first child.

    Later age for all children, to nitpick.

    What does this mean: that the initial rapid rise in age at first child sends the numbers plummeting: to the 1.2 TFRs seen in Italy and Spain. These numbers are artificially low, and there is then a ‘rebound’ but only up to around 1.7.

    Actually, in Italy and Spain the rebound hasn’t gone nearly that far. They really aren’t having kids.

    In France, OTOH, it looks like they’re going to stabilize at a TFR around 2.2 for a while.

    Without immigration (and repeat, for cultural reasons migrants may have the first child earlier) it is extraordinarily difficult to get much above this level.

    Which loops us back to my original point: the US has managed it, even taking out immigrants /and/ the first generation of immigrants’ kids. Native born USAns are replacing themselves.

    What makes the US different from all other advanced countries is a separate, fascinating question.

    Doug M.

  24. I’d just like to say thanks to everyone who has
    contributed to this thread. The debate has been
    civilised and I personally am slowly getting some
    things clearer in my mind.

    In particular the discussion with Doug (Muir) has
    brought home forcefully to me just how differently all
    this will be perceived in the US, and how this
    difference will only *add to the list of global
    imbalances*.

    Dave:

    “There’s a fundamental difference between a literate
    and an oral culture.”

    Yes, this is a topic that has long interested me. I am
    pretty much an ‘unrepentant McLuhanite’ on this
    front. This interest also lead me at one stage to a
    more or less obscurantist interest in the ‘authorship’
    of the Odyssey and the Illiad.

    But I’m making a bigger distinction here. In literate,
    but pre ‘post-modern’ societies, grandparents have
    played an important role in transmiting cultural
    knowledge and experience to their grandchildren (again
    depending on family structure, but this was clearly
    true in Southern Europe eg). This link is now largely
    broken, and the interesting question may be why, and
    with what consequence: post coming on this in the near
    future.

    Your orginal point that immigration represents an
    “enormous challenge particularly for nations that
    define themselves along ethnic lines” is also
    important though it hasn’t been taken up here. Clearly
    the US, Canada, Australia and New Zealand are special
    cases on a very large continuum.

    Oliver:

    “current economic woes is caused by demographics, is
    the damage reversible and how long will it take?”

    Difficult to say. My guess is that it is irreversible,
    but that the impact can be reduced and that is what we
    should be working on.

    “political capital. What worries me is that
    governments that fight a hard battle to allow
    increased immigration will not raise retirement age or
    curtail benefits.”

    OK this is surely a real worry. Non of these measures
    is popular. No easy answer here.

    “Or should it better be spent on eg. Alzheimer
    research?”

    Two points here. In the first place advances in the
    treatment of Alzheimer (desireable as they may be) may
    not keep many people working longer, they will
    probably help many people live longer with an
    increased quality of life. ie in the near term it
    seems unlikely that we will eliminate Alzheimer, we
    may however slow down the deterioration. Secondly, in
    general in this area it may be better to encourage the
    private pharmaceutical industry to shoulder the cost.

  25. Doug Muir.

    I don’t want to risk being too pedantic, but…..

    “If the average woman only wants 2 kids anyway, it’s probably no big deal if she starts at 31 instead of 21”.

    I’m not ready to buy this. Virtually all the EU
    studies show that women generally ‘want’ 2 children. Over the reproductive cycle wish and reality don’t coincide. There are many reasons for this, and one of them is fertility which declines with age.

    One book which has influenced my thinking on this a lot is: On Fertile Ground – A Natural History of
    Reproduction by Harvard Anthropologist Peter Ellison.

    Just one extract:

    “Understanding the natural arc of female fecundity has come slowly and is as yet incomplete. But what we do know suggests that the relationship of age and female fecundity is a basic part of human biology, shaped by forces of natural selection that are both very old and possibly, on an evolutionary timescale, quite new……..

    The most convincing evidence of age-specific changes in female fecundity comes from the realm of ‘assisted reproductive technologies’ or ART…. Age related declines in pregnancy rates have now been documented in virtually every ART procedure, including ovulation induction and in-vitro fertilization. Indeed, so common and well documented is the effect of female age
    that it’s absence in any substantial database would be a significant, publishable finding” (pages 218-220).

    Equally the measure of first child age is useful since it gives a rule of thumb measure of ‘family formation’ age, and this again has some significance for fertility.

    One UN Demographer – John Bongaarts – has tried hard to use differences in this statistic (age at forming first family) to differentiate between North European and South European fertility: since North Europeans tend to leave home earlier, and form an ‘independent’ couple sooner.

    “Actually, in Italy and Spain the rebound hasn’t gone nearly that far”.

    No you are right. Bad wording on my part here. They are rebounding, but the impact is relatively small to date.

    The important ongoing problem is how representative of one possible future the Italian and Spanish (and Greek and Portuguese) cases will turn out to be. A lot hangs on the rapidity of the decline from high fertility to below replacement.

    Some years ago a demographer named Keyfitz introduced an idea which is actually quite important: population momentum.

    Momentum gives an inherent ‘snapshot’ of a country’s demography. The key distinction would be between stationary and stable population. It is possible to have a stationary (in numbers) ageing population where the pyramid is not stable (the US towards mid century may become like this). Or conversly a population where the fertility rate is stabilising but the momentum is still moving towards large population increases (the current case of many third world countries).

    Now momentum equations have been used to show that a real problem can come if the drop in fertility happens too rapidly: in less than a reproductive generation.

    This can produce a structural break and a very low
    momentum figure which it may be difficult to climb
    back from.

    The old EU countries have had fertility declining over a relatively long time period, but the situation is accelerating with each new arrival below replacement.

    “No, that’s too low.”

    Yes, you’re right. But not much too low. According to the document (CDC National Vital Statistics Reports) you linked (for which thank you) the TFR for non-hispanic whites has been dropping from 1.866 in 2000 to 1.828 in 2002, whilst during the same period the Hispanic TFR has gone from 2.780 in 2000 to 2.718 in 2002.

    The aggregate non-Hispanic black TFR’s hover around the national average 2.178 in 200, 2.047 in 2002.

    Now part of the difficulty we may be having is about the definition of native/non native, and immigrant/non immigrant. Clearly second generation hispanic citizens are *non-immigrant*, but they are still under the impact of the migration process. Since it seems the vast majority of the hispanic population are post 1970 migrants or their descendents I maintain my view, whilst accepting the modification that immigration is
    far from the only pertinent factor here.

  26. Two points here. In the first place advances in the
    treatment of Alzheimer (desireable as they may be) may
    not keep many people working longer, they will
    probably help many people live longer with an
    increased quality of life.

    It seems to me that the largest cost of an ageing population is medical. Not only are the old unable to work, but caring for them takes labor unavailable elsewhere. It is enough if we can keep the old healthy and we still get enormous benefits.

  27. If you want to know how would it look a society built on immigrant labor, go to the Gulf countries. Very clean, full of smiling black faces. And State enforced discrimination. Nice if you are of ruling class, Gulf Arab inthis case.

Comments are closed.