Not so. The very part you quote contains the phrase "As the population declines".
Yes, the population will keep declining. But the
ratio of workers to retirees will stop declining once it reaches a new equilibrium point. This is because eventually with some time lag until the people born when birth rates were still much higher have died off, the number of retirees will shrink at the same rate as the number of workers.
Which is kind of irrelevant because no-one denied or failed to understand that a below-replacement fertility means a declining population (once the generation born when fertility rates were still higher has died off).
But hey, if you're allowed to calculate with a 5% pa decline for effect, do I get to calculate with a 5% growth?
After a century of declining at an annual rate of 5%, the US population will be 1.9 million (starting with 325 million). After a century of growing at 5%, the US population will be 42.7
billion. Neither is ideal, but I know which I'd prefer (and no, a more realistic rate won't make this go away, it'll just delay it by a few short centuries).
I don't think anyone denied that
all else equal, the burden on the younger generation will be higher. What people are denying is that all else will be equal, or that immigration is the only parameter that can reasonably be tweaked.
And the ratio of 1.12 workers supporting a retiree is more than doubling the burden over today when there are about 2.9 workers for each retiree.
A lot of that increase is due to increased life expectancies alone, though. If you run the calculations with my parameters and simplifications and a fertility rate exactly at the replacement level, that is population growth rate 0.0%, you still get a ratio of 1.6 exactly. In that model, each one-year-cohort is exactly as strong as the previous one, so all we need to do is compare the 40 cohorts between ages 25-65 to the 25 cohorts 65-90. A more realistic age-dependent mortality, instead of pretending everyone drops dead on their 90th birthday, only makes things worse, though not by much as long as few people die before 65.
Indeed, even if we plug in the US's current population growth rate of 0.7% annually, the ratio is still 2.01 workers to every retiree at a 90 year life expectancy. In order to maintain a ratio of 2.9 with a life expectancy of 90 years, a growth rate of 1.8% annually is required. Conversely, if you believe that life expectancies will plateau at 80 or 85, even the declining population corresponding to a fertility rate of 1.5 will still have a workers/retirees ratio of 1.98 or 1.44 respectively.
What these calculations also ignore is that not only are the retiree cohorts numerically stronger in a declining population than the working-age cohorts, but the cohorts of children and students are weaker than the working age cohorts. These effects almost cancel each other out: If we assume a stable fertility rate high enough to guarantee an annual population growth of 1% and life expectancy of 90 years, the ratio of workers/non-workers (where non-workers include both retirees and children and full-time students, modeled as anyone below age 25 or above 65) will be 0.7629. For population declining at -1.1% annually (which is what I get for TFR of 1.5) that same figure will be 0.7547. In fact those ratios are symmetric: A positive growth rate of +1% or 101/100 produces the exact same ratio as a negative growth rate of -0.99% (=100/101), and the optimal ratio of 0.8 (40 cohorts of working age people to 25 cohorts each of children/students and retirees) is achieved only at 0.0 growth, where all cohorts are the same size. This is an artifact of the fact that our pre- and post-productive life phases are the same length (25 years before 25, 25 years after 65 till death at 90). With a lower life expectancy, children and students will be more of a burden than retirees and thus the ratio worse under a growing than a declining population.
And we could improve this ratio by increasing immigration of young workers. This is my argument.
We could. Or we could reap the fruits of productivity increases and realise that in the 21st century there's no longer any good reason why the ratio has to be high to provide everyone with a good life.
The birth rate is the running average number of children each woman has in their lifetime. The replacement rate is the birth rate required to keep the population constant without any migration or immigration, people leaving or entering the country. The replacement has to be above two to account for the factors you mentioned but it must always be above two.
No, the
birth rate is the number of children born per year per 1000 inhabitants. What you're talking about is the
total fertility rate.
Terminology aside, the replacement fertility rate
can be below 2.00 under one condition: If more girls than boys are born, and the bias is strong enough to offset girls' and women's death rate before the end of fertility.
I don't follow your math, you seem to be going down the well worn path of convincing at least yourself that you are correct. The math is overburdened with your assumptions, another way of saying that you are right if you accept my assumptions.
You can just ask
Which of the assumptions do you find dubious? Do you expect life expectancies to grow indefinitely, of fertility rates to decline indefinitely?
I'll explain my formula though.
At a given rate of population growth or shrinkage, the number of people born each year will be proportionally smaller than last year's number. Say the rate is -1.1% (approximately what we get with TFR 1.5 and generation span 30). If the births in year 0 are 1 unit (say a million people), the births in year 1 will be 0.989 units, and the births in year 2 0.989 x
that, or 0.989^2, and so on and so forth. For every year y, the births will be 0.989^y. To find the total number of retirees, we sum over the births from 90 years ago to the births from 65 years ago. For ease of calculation, the births 90 years ago will be our unit. We thus sum over the numbers {0.989^0 ... 0.989^(90-66)}, and likewise for the workers, though here the range will start at 0.989^25 or as I expressed it 0.989^(90-65): the oldest, and most numerous, cohort of workers will already be smaller than each retiree cohort. In fact, it will be 25 years worth of a 1.1% decline smaller than our unit cohort of 90 year olds.
The problem is simple. If the birth rate is below the replacement rate, which can never go below 2.0, the population will decline without immigration. It is inherent in the definitions of the two terms. The replacement rate is the birth rate required to keep the population constant.
Thanks for telling me nothing new. But we weren't talking about keeping the
population constant, we were talking about keeping the ratio of workers/non-workers constant constant.
But it doesn't matter, your conclusion is the same as mine, the ratio of workers to retirees will decline. This is sufficient to say what I am saying, the burden on each worker will increase and we can relieve this burden by increasing immigration.
Yes, there are things that we could do to mitigate the problem without immigration. We could do what civilizations have done before. We could institute polygamy. Only the wealthy can afford to have many wives and they would be less restrained by the costs of raising a child. Of course, we could suffer then from the problem of too many sexually frustrated young males willing to blow themselves up to achieve relief from the 78 virgins in heaven. And I think that polygamy isn't so much a reaction to declining birth rates as it is reflecting the desire of rich and powerful men to have sex with many different women.
But the conservative approach is to assume that everything else remains as it is. That we don't turn to polygamy, that we don't as bomb suggested increase the retirement age to lower the mortality age at retirement by working people to death.
This is wrong. What automation does first of foremost is increase the amount of products produced without a proportional increase in the size of the workforce, or no increase at all. I don't want to go into that same old derail with Bomb#20 about the labor theory of value, and fortunately I don't have to: Simple suppy-and-demand reasoning will tell you that producers will have to make their products cheaper relative to wages in order to find buyers in that scenario. The only situation in which this isn't so is if supply is upper-bound due to relying on a scarce natural ressource. If that's what you're basing your reasoning upon, you are the Malthusian, not Bomb#20. Also, in that case, population growth would surely make the situation worse.
I am the all time champion of economic theory derails.
I will try to minimize that to answer this, but it is hard for me.
My statement that automation isn't the answer for this problem is based on the conditions that exist today continuing. Here are those conditions,
<snipped long-winding discussion which might be relevant directed at a defender of neoliberalism, but you're barking up the wrong tree>
The reason that a worker can't pay 90% of his income in payroll taxes is because the workers are not sharing in the rewards from automation. It is now the aim of our current economic policy to pass all of the gains from all of the growth in the economy to the already wealthy. Both you and bomb seem reluctant to acknowledge this point but seem to ignore it rather than to offer your reasons to oppose it.
If I understand you correctly, you're admitting that automation has the potential to allow everyone to have a high living standard despite a declining ratio of workforce/adult population but contending that this potential isn't being actualised because of how neoliberal economies prioritise production and organise distribution of the produce. Is that a fair paraphrase?
If so, I don't see how what you're proposing - to keep the ratio artificially high in order to not have to address those priorities - is any different from a man who keeps throwing more and more fuel into the fireplace to fight the chill while ignoring the open windows.
And to stick with this analogy, he's already low on fuel and will soon be starting to burn the furniture: Continued exponential growth is not a feasible long-term solution due to rather simple math. You can call me a Malthusian till the cows come home, but the fact remains that, at a continued exponential growth of just 1% per year, the population multiplies by a factor of 2.7 per century, 20959 per millennium, and over 9 trillion in 3000 years, still well within the timescales of human civilisation. Starting with todays population of 7.6 billion and assuming an average human to have 50kg, half of the earth's mass, liquid iron core and all, will be human flesh by the year 5018.
I'm very much
not a Malthusian in that I accept the empirical evidence that this is not in fact happening, that population growth is already slowing down at an encouraging speed globally and the population likely will peak within the century, early next century at the latest. But his math was right, it's his assumptions that were wrong. Continued exponential growth is a mathematical impossibility in a universe with a finite maximum speed and a finite number of dimensions, much more so within any one country.
