If Demographics is the Problem, What's the Solution

In the comments column of Brad's blog Amit Dubey asks the following:

So if demographics is the problem, what's the solution? And how should other countries handle it e.g. a big tax hike now in order to fund government consumption when demographic inflection hits? I don't think "immigration" is a good answer, because then what do India and China do a couple decades down the road? (and, perhaps more optimisitcally, what will Nigeria and Congo do a couple decades after that?)
Source: Semi Daily Journal
LINK

My attempts to answer this quite important and legitimate question have given rise to the following thoughts.

Firstly 'What's the solution'? Apart from stating there is no insta-pundit solution on this one, there are obviously things we can do. Reduce our deficits in the good times for one. Carry out pension reforms and encourage private saving for two. Increase the retirement age for three. Individually have more children, for four. And most important of all change our attitudes to immigration.

Of course, none of these 'solve' the problem. But they may make things better rather than worse. If Britain scores rather favourably on the ageing vulnerability index, for example, it is because it has made important progress on pension reform, immigrant acceptance and multicultural identity when compared with its European neighbours. Is this unequivocally good? No. The pension reform comes at a price of increased poverty among old people. Is there much alternative? I don't think so.

The point about India and China is a good one: one day the planet will run out of immigrants. But what may be only a stop-gap solution does buy time, and that could be important as this is what we need to feel our way forward.

Secondly: "if demographics is the problem"

Interestingly more and more people seem to be waking up to the fact that it might be at least part of the situation. Last month it was Alan Greenspan, now it is the deficit 'hawks' Volcker and company:

"Looming at the end of the decade is a demographic transformation that threatens to swamp the budget and the economy with unfunded benefit promises, like Social Security and Medicare, of roughly $25 trillion in present value. Our children and grandchildren already face unthinkable payroll tax burdens that could go as high as 33 percent to pay for these promised benefits. It is neither fiscally nor morally responsible to give ourselves tax cuts and leave future generations with an even higher tax burden."

Remember: the 'pension reform' which is most the likely outcome, even in the US, is tantamount to a massive Enron-style default on previously contracted obligations, only this time the defaulting party is the national exchequer. This is bound to have an impact on saving, consumption and the output gap.

What's the solution. This is a fair question, and the honest answer is that I don't know.

The best answer I can come up with right now is that since an economy is a form of complex adaptive system, then there will be some automatic stabilisers. Maybe our expectations form part of the picture. Maybe we are assuming as inbuilt growth patterns which have only characterised the last two hundred years.

Taking three parameters - technology, population and per-capita incomes - as key variables, we can see that for most of human history prior to the industrial revolution (outliers like the Greek and Roman empires being exceptions, and for interesting reasons) better technology meant more people and near stagnant living standards. This was Malthus' argument. Then came the industrial revolution, and both population and living standards increased rapidly with technological change. Now we have the information age, and technological change accelerates (possibly following a power law distribution?) population begins to decline, and we don't know what will happen to per capita incomes. Period: we don't know.

The point about Nigeria and Congo is also interesting. A demographer called Keyfitz once developed something called the population momentum equation. This is much more imporatnt than the name sounds, as it enables forward looking calculations on demography. Have you ever asked why the demographic problem has hit Japan first? Well Keyfitz goes a long way towards the answer. It's all about how quickly you make the first part of the demographic transition (the part which goes from 3-4 children per woman to below 2.1). Well if this occurs in less than one generation (simplifying greatly), then the population effectively becomes rapidly de-structured and enters a downward spiral. That is why Japan is first, and other countries like Italy, Spain , North Korea, Taiwan etc (who begin and complete their transitions later, and more rapidly than EU 'core' countries like the UK and Germany, and who also historically have less immigration) follow in quick succession. This could lead us to expect that in India, China and Brazil the situation could be still more dramatic (since latest word from the UN is that the transition is accelerating in these countries), and so on till we reach Nigeria and the Congo. That is if HIV-AIDS doesn't get there first. Remember, things get faster, faster.

Lastly, in terms of analysis, and especially in terms of more conventional economic analysis, this process can be approached from both a macro and a micro angle. With the proviso that it's only when you come to lay the macro-horizon on its micro-foundations that, I feel, the extent of the problem becomes clear.

On the macro side standard growth theory can serve as well as any other starting point.

Solow suggested that Y= A f(K,L)

Well I would suggest(following Mokyr) that this characterises fairly well industrial society with high proportions of fixed capital , but that in the information age (the age of human capital, remember the adage: a firms most important asset are its workers), a better formulation might be:

Y = A f(K, L1, L2,........Ln).

If the function looks something like this, then obviously if the sum of the growth in value of each of the various components of human capital exceeds the reduction in quantity, you can still have growing GDP. This problem is extremely tricky to model, as Solow notices in his critique of endogenous growth theory. It is easy to build a model with exploding, knowledge driven growth, it is far more difficult to envisage this occuring in practice. In part this is for reasons associated with our understanding (or lack of understanding) of Total Factor Productivity (or TFP). At present my understanding of this is still mirky, and it is better described as work in progress.

For the other part the actual values of the various human capital elements depend on the demand side equations, and it is here that consumption enters. I find Angus Deaton's notions of liquidity constraints enormously suggestive. Basically the collective constraint will be related to perceived rates of growth going forward, likelihoods of government debt (or pension) default, and the respective age structures and dependency ratios of the population. The nub of the matter seems to me to be how rapidly the younger age groups in society are willing to advance consumption, and, given their relative numberic weakness, at what rate of indebtedness.

This is as far as I have gone to date. Anyone willing and able to take this type of problem forward to a succesful conclusion deserves a Nobel in my book. Any good suggestions or lines of investigation, please contact me.