Ten Theses on Longevity, Technology, Fertility and Economic Growth
Life is an energy harvesting process
Life expectancy and fertility are intimately interconnected. The principal interconnection derives from the simple relation that the greater the length of the average human lifespan the higher the rate of return on the initial investment in each individual child. (Kaplan et al 2003, Kaplan and Robson, 2002). The underlying decision procedure which faces each individual in making such investment decisions has two components, investment in self and investment in offspring. Up to a certain point investment in self is also investment in offspring as there are externalities in the learning and cultural transmission process, and children obviously benefit from their parents own educational investment. But beyond this investment in children is a net charge on the individual consuming significant quantities of economic, emotional and time resources. The net outcome is that longer lives mean less newborn children, as parents invest more time and energy in both themselves and in each individual child, while the length of initial parental dependency increases steadily as both lifespan and the technological level of a society increase. In many ways Robert Fogel's idea of techno-physio evolution is in considerable harmony with this thesis...| continue reading #
Each society has a technological frontier. Life expectancy, the technological frontier , and years of pre-productive preparation are interconnected. The further out the frontier in any given society, the longer the life expectancy, the more 'formation' the individual needs to operate on the frontier with optimum effectivenes (ie the later the age at which the individual becomess fully self-sufficient resourcewise), and thus the higher the mean age of mothers at first birth (MAFB) in the society.
A society's standard of living is a derivitive of its technological frontier. Per capita income economic growth is a function of the rate at which the technological frontier moves forwards. There is a global technology frontier, and some societies are either on or near that frontier. Others are some distance away from it but closing, while yet others are caught in a virtual 'sink'.
Once a society starts to reduce the distance between its frontier and the global one, an increasing returns feedback process is set in motion. Life expectancy increases, fertility falls, and the standard of living rises. Ceteris paribus the rate of economic growth of a society, the rate of its fertility decline and the rate of increase in its life expectancy are all a function of the distance between the local and the global technology frontier at the time of initiating the 'convergence process', and the speed with which convergence occurs.*
Globalisation and openness are important catalysts** which facilitate the development process once 'catching up' starts. This is important point to note is that 'technology transfer' may also be considered to include 'mindset' and 'institutional' transfer (competitive product markets, gender equality etc etc). Put another way, the diffusion of the mobile phone is not simply the distribution of yet another product, it is also the initiation of a whole new set of communicational and behavioural patterns.
The life expectancy - technological frontier - years of pre-productive formation triad contains an important increasing returns dimension in terms of the evolution of human cerebral capacity and creativity and innovation. This is what gives rise to the phenomenon of 'acceleration' in the process of technological change. ***
Post transition societies tend technologically towards dual labour markets, with a corresponding 'rectangularisation' of fertilty around two 'attractor' ages for first births, a lower and a higher one. In both cases the MAFB will steadily rise, but possibly at different rates, reflecting movements in the differing cognitive and skill requirements of the two poles.
At the core of the fertility and dual-labour-market transition lies a gender one. Hunter gatherer societies already had a dual labour market, although since this was gender-driven, there was only one single fertility regime. Essentially there was a division between 'gathering' (which was principally female or young child oriented) and 'hunting', which needed training (or energy/resource investment) and was male.
As Kaplan points out, hunting, as opposed to gathering of animal protein in small packets, is largely incompatible with the evolved commitment among primate females toward intensive mothering, carrying of infants, and lactation on demand in service of high infant survival rates. Hunting is often risky, involving rapid travel and encounters with dangerous prey. Also, it is often most efficiently practiced over relatively long periods of time rather than in short stretches, due to search and travel costs. Finally, it is extremely skill intensive, with improvements in return rate occurring over two decades of daily hunting. The first two qualities make hunting a high-cost activity for pregnant and lactating females. The third quality, in interaction with the first and second, generates life course effects such that gathering is a better option for females, even when they are not lactating, and hunting is a better option for males. (Kaplan et al., 2001).
With the move to horticulture this balance changes, as child labor starts to become more valuable to families since horticulture provides a number of relatively low-skilled tasks that older children can perform. In fact fertility rises slighly in the horticultural context (Bentley et al., 2001). Kramer (2002) offered some explanation for this when he demonstrated that among Maya horticulturalists older children contribute at the level of “helpers-at-the-nest,” significantly increasing their parents’ fertility and without whose help their parents could not add further offspring to the family.
The transition from this earlier homeostatic regime to the modern fertility/longevity trade-off one has been marked by a steady evolution towards gender equality. Modern labour markets are essentially based on formal equality of the sexes in the presence of biological asymmetry. In the long run this is not sustainable and will surely have consequences which are going to be important, but which are, at this stage, difficult to foresee
Life cycle productivity is variable. In the dual labour market context, the less skilled pole exibits less fluctation in performance across the cycle. At the more skilled pole however the situation is markedly different. There is a cross-cycle trade-off between speed and experience, and this produces the well known hump-shape curve across the age span. Put bluntly it is the evolution of the peak of this hump which determines whether a society ages gracefully, or with difficulty.
The societal ageing process inevitably produces a constant increase in calendar median ages, the real issue is whether this increase also occurs in what Sanderson and Scherbov term life-cycle-rescaling calibrated terms.
As stated in (3), per-capita-income growth in a society is a function of the rate at which the technological frontier of that society moves forwards. This is in fact an oversimplification. It is important to also specify that an integral component of that technological frontier is the level of the human capital which is 'embedded' in the frontier. Now following Altig et al (2001), it is plausible to assume that technical progress causes the time endowment of each successive generation to grow at a certain determinate rate. The proposition Altig et al advance is not that time, per se, expands for successive generations, but rather that each successive generation is more eﬀective in using time to either perform work or enjoy leisure. What Altig et al do not seem to do is to map this idea onto the hump-shaped curve. The intuition here is that across generations age-specific time-effectiveness increases at some constant rate, but that within any given generation time effectivess reaches a peak and then enters decline.
The transition from the earliest homestatically-regulated hunter-gatherer societies to the modern long-run steady state-growth ones (which still lie in some distant and probably never realised 'homeostatic' future - never realised due to the ongoing impact of the technophysio evolution feedback process) is one from relative stasis in fertility and longevity (which only fluctuate as a result of changes in climate and the disease environment) to one where there is a systematic trade-off between lifespan and fertility. At present there is little noticeable movement in the upper reaches of our lifespans, and thus a kind of 'rectangularisation' process is occuring with mortality being clustered around the extremes, but there is no in-principle reason to imagine that increased knowledge and technical change will not mean that today's apparent limits are not breached at some stage in the future, in which case there is no reason to assume any end to this co-evolution of lifespan and fertility, exept the one rhetorically referred to by Wolfgang Lutz recently when he stated that 'of course if one day we do 'live forever' then maybe there will be no children. This, of course, is simply a theoretical limit case.
In any event during the evolution from the first to the second state much of economic growth is driven by transitional dynamics, so analytically it is, in fact (if not in principle), impossible to distinguish which components of growth come from such dynamics and which from movement in the technology frontier. In the hypothetical long-run post-transition steady state of course, global populations should converge in their fertility behaviour, lifespans and economic growth rates, and at such a level of abstraction all of this seems perfectly sustainable. But then, as Keynes said, in the long run we are all dead (or not, depending on what happens to life expectancy), and what is more interesting for us in the here and now are the short run, or transitional dynamics.
Whether continued population ageing is a socially problematic process depends in large measure on the degree of life cycle rescaling and intitutional flexibilisation which is achieved. As is well known as societies move through the demographic transition there is a positive growth impact, or demographic dividend. Rather less well explored is the possibility that, as the large generations which characterised the earlier parts of the transition (and which were laregly the product of the first-stage mortality decline) age, there may not also be a demographic penalty, to be charged above all in terms of increasing dependency rates and declining productivity for the most numerous generations.
The key to whether or not this eventuality is realised will be the extent to which the peak in the hump shaped performance curve is able to move to the right when compared with the increasing distribution of the age structure towards the higher ages. Put another way, the key question would be what happens to the mean aggregate time effectiveness of a society. If this continues to rise, or at least remain constant, then handling rapid population ageing of the kind associated with the unwinding of the demographic dividend should be achievable without major systemic shocks. Should the mean aggregate time effectiveness of any society enter decline, ageing will become problematic and the consequences hard to foresee.
I think the above ten 'theses' are, in-and-of themselves, sufficient, since adding more would only needlessly complicate an already complex picture. Everything else can in-principle be incorporated in the form of feedback-loop-type 'plug-ins' (cohort factors, fertility traps, intergenerational transfers, welfare systems etc etc).
* A-priori it is possible to postulate that the later the process starts the greater the distance of the initial position from the frontier, and the more rapid the convergence. This should manifest itself in the form of ever higher 'full steam ahead' growth rates, and ever lower birth-postponement fertility 'floors'. The Asian Tigers would be expected to peak out at a higher GDP growth rate and lower TFR floor than Japan (which they did), China than the Asian Tigers (which it will), and India (say 10 to 15 years from now) ditto vis-a-vis China (which is, of course, currently the economic growth record holder, and may yet become the next champion of the lowest-low fertility-floor league).
** Something they may not be before that time, at least in such an unequivocal fashion. Hence much of the confusion in the globalisation debate.
*** The increasing returns aspect actually comes from the Michael Kremer generated literature. Essentially, if increasing returns were a product of the number of people then fertility would never have any good reason to fall, and we could just go on happily to infinity (across the universe I suppose). But this isn't what happens. So it is important to distinguish between extensive and intensive brain growth. The former, as we know, stopped many thousands of years ago, but the latter (if the scanner reports are anything to go by) continues, not only unabated, but possibly at an ever increasing rate.