The thought hadn’t occurred to me until I read this.
The Impact of Income Distribution on the Length of Retirement
By: Dean Baker
David RosnickSocial Security has made it possible for the vast majority of workers to enjoy a period of retirement in at least modest comfort without relying on their children for support. The average length of retirement has increased consistently since the program was started in 1937. However, the increase in the normal retirement age from 65 to 67 that is being phased in over the years 2003 to 2022 largely offsets the increase in life expectancy. As a result, workers who work long enough to collect their full benefits will see little gain in the expected length of their retirement over this period. These gains have gone overwhelmingly to workers in the top half of the income distribution. Consequently, the increase in retirement age will offset the gains in retirement lengths for the bottom half — even if there is no further inequality in improvements in life expectancy. If such inequality in improvements persist, then the bottom half of workers born in 1973 will have retirements no longer than those born in 1937.
Well I’m still in favour of increasing the retirement age. But I don’t know what to do about this. What could be done without messing with basic norms of ‘equality’ before the law, which is to say how can we improve substantive equality without mucking up procedural equity – without mucking up the idea of treating people of like age similarly. If we made the pension dependent on health, that creates moral hazard – there would be clinics to get you in really bad shape for your medical to qualify for the pension.
I find this very hard to believe. Life expectancy has increased by only TWO years from 1937 to 1973?
I wonder if this amounts to nothing more than a particularised restatement of the fact that, generally, the rich really do have it better. Which is why we have progressive taxation and progressive pensions, isn’t it?
Yet we need to consider all forms of moral hazard, I think there is a moral hazard in forcing people to work on in physically demanding jobs (pretty much the entirety of which are in the lower half of the income range) when their bodies are starting to give out.
Personally I’m fine, in a interesting, white-collar, physically undemanding job which I’m going to be happy to work in until at least 70, but I have two brothers in-law who are in their late 40s and early 50s and have spent 30 plus years already working in physically demanding blue-collar work. They both have significant health issues as a result of their work and to insist on a longer retirement age for them, because us professional/managerial types are happening to live longer verges upon the cruel.
It seems incredible to me, given the massive increase in wealth since 1937, that we would be looking at equivalent benefits for a current 37 year old upon retirement.
As I canvassed in a recent post, I don’t favour extending the mandatory retirement age beyond the planned 67 for blokes, but I do favour implementing a system of modest tax and other incentives for people who may choose voluntarily to work on until (say) their early 70s.
It would be inappropriate for the incentives to be so generous that they create significant inequality among retirees, with tradies and others in physically demanding jobs which give them little practical alternative but to retire in their 60s being condemned to a relatively impoverished retirement by comparison with those in less physically draining white collar jobs.
However, I suspect that for a lot of people the incentives would not need to be very great at all. At least as I currently see it, I expect I’ll want to work on until 70 or so almost irrespective of extra incentives. On balance, and despite the problems of the higher education sector, I find university teaching, administration, writing etc sufficiently inherently rewarding and worthwhile that I would like to keep doing it on some level (possibly part-time and casual/flexible) for quite some time yet.
I strongly suspect that there would be enough people like me, who find their career more a joy than an imposition that they’re anxious to escape, so that very moderate incentives (perhaps even just facilitating more flexible part-time working arrangements) would be more than enough to cancel out the adverse inter-generational economic effects of babyboomer aging.
One possibility, which I imagine should be pretty easy to do now given superannuation is compulsory and hence how long you have worked can be tracked easily, would be to allow people to retire after a certain number of years worked, rather than a particular age. Given that people in higher income brackets tend to have higher education levels and hence have spent more time studying, at least this group would have to retire a few years later, and people that had slogged away all their lives could retire earlier. It also has some sense of fairness built in, because it would mean that you would have to pay tax for at least a certain number of years before being able to retire. There are obviously groups where corrections would need to be made (e.g., people out of the workforce due to children), but I don’t think these would be impossible to work out.
That’s a great idea Conrad. Shows the value of ‘crowdsourcing’. Only problem is the richies would be onto it and would all be working for their Dad’s business all through school! I guess one can catch that by removing people who’s main activity is education.
That is a great idea, particularly since only ‘richies’ tend to take extended ‘sabbaticals’ or career breaks in foreign beach/ski resorts. Also, apprenticeships could be included as work.
Conrad’s idea is implemented in France. My understanding is that the strikes there were also against extending the length of time in work before you received the full pension, not just extending the age at which you could retire. So for example French people could retire at 60 but to get the full benefit of the pension most people needed to work to 65. I think Sarkozy has added 2 years to both the retirement age and the time required to be in the workforce before you receive the full pension. The detail might be incorrect but I think the thrustof what I have said is correct.
In effect you are right, JP. A full pension is conditional on:
1 having passed 62 (to be eligible for anything);
2 having passed 67 (to be eligible for the full rate); and
3 having contributed for 41 years (increasing under current reforms).
The obvious weak link is that current retirements are largely paid out of current taxes. There are myriad other weak links such as the ‘regimes speciaux’ which mean that millions of public servants and partly-privatised public servants qualify for a full retirement at barely 55.
Amongst the absurdities, train-driver remains classified as a ‘penible’ occupation which means that train-drivers retire early. This has its roots in the pre-1910 pension regime and has not adapted to the change in job description over the intervening 100 years (speaking as someone who has driven a French train, death by boredom and fatness appears to be the biggest risks of the job at present).
“speaking as someone who has driven a French train”
I’m intrigued. Tell us the story.
You haven’t driven a French train Ken? I’m surprised and frankly appalled. I thought pretty much everyone on Troppo had driven a French train. I certainly have.
It’s very easy, a friend is a French train-driver, so we just took a quiet midday service along a sleepy outer-urban/country line, strolled into the cabin, wandered around, had a drive, and then got off. The whole place is the ultimate closed shop, so no-one blinked, not even the conductor. We didn’t even have to pay for the trip.
Even my then 4-yo son had a drive! After all it isn’t like you can steer it off the rails :)
The main thing is to keep your grip on the dead-man (bizarrely like a truck’s steering wheel), and intermittently flick the switch to indicate that you have seen a signal. It is a sophisticated dead-man, though, if you don’t periodically release it that also triggers it.
Oh, and you toot before going under bridges. My friend was at a loss as to why, since this presumably makes any would-be-jumpers’ job much easier. But you do.
Braking is a little more of an art, if you flick the air on too abruptly you can really slam the thing to stop. In fact that was the hardest part, after staying vaguely focused.
Pleasant day out, really, spent a few hours in a picturesque little town having coffee and then took the train back.
Oh, and my friend? He swears by the ‘pénibilité’ of the work :) They have to get up at all hours depending on their shifts. The night/dawn shifts really throw his body clock, apparently. He really believes this, too. He was a bit surprised to hear that professionals sometimes work close to a train-driver’s week in a day :( He didn’t think my work (for example) was nearly so ‘pénible’.
The best bit is that his dad is a retired train driver. Well that’s not the best bit. The best bit is that his dad recently completed a couple of lucrative (post-retirement) years driving private trains in Taiwan for an affiliate of the SNCF. The remuneration was generous enough but of course it was wholly additional to his pension. For those hard-workers in the SNCF the conditions were (at least in his father’s time) something like 30 years service and at least 50 years old!
How that stacks up against ‘pénibilité’, let alone ‘sustainability’, I leave to you.
As I keep saying, in setting a retirement age mortality is not the main issue. Morbidity is.
The fact, as people keep saying, that Otto Bismark set pension age at 65 assuming most workers would not survive to that age, and that most now do, is not relevant. What is relevant is whether today’s 65 year old is as capable of earning a living as an 1890 German 65 year old. And that is not a gimme.
It’s not a gimme for two reasons:
– selection effects: only the toughest workers survived to 65 in 1890.
– modern medicine is far better at keeping people alive than keeping them fit. That’s why the rate of disability in the population is consistently rising (many disabled people in earlier decades would have been dead and hence not counted).
The over 55 non-working basically fall into two camps:
(1) The poor bloody workers whose health has gone, living in near-penury.
2) The better off who can afford to retire early (NTTAWTT at all if no-one’s subsidising them, and arguably it’s not even a problem if there is a modest subsidy).
In the long run the latter group will become bigger as society gets richer. But in a society with increasingly unequal earnings, the voluntary retiree group will grow much faster. Yet raising the retirement age basically aims at poor involuntary retirees.
Fiscal sustainability may mean some countries are forced to do this (though this doesn’t remotely apply to Australia – our system is eminently sustainable). But its not ipso facto good policy unless you are forced to do it.
A few observations on the issues raised:
– Patrick is right; overall life expectancy at 65 has risen enormously in recent decades (in Australia, they rose about 10 years in the 30 years since 1970). So as a statement about the recent history of the general US population, Baker & Rosnick’s third sentence makes no sense. It is better read as a statement about outcomes for the bottom half of the income distribution. (Thankfully, the paper gets better from there.)
– The Baker & Rosnick paper argues that statistics show most of the US life expectancy gain has gone into the top half of the income distribution. That is new information to me. If it’s true in the US, I would expect it to be true in Australia.
– The Baker & Rosnick paper refers to expected gains in life expectancy for male workers born in 1960 i.e. today’s 50-year-olds. This is not the only way to do the analysis. Most people see rises in the pension age as a process of catch-up for the gains already made in recent decades – that is, in a perfect world you would have raised the pension age some time ago. Also, projections of future life expectancy are pretty unreliable, and probably more so when you try to break them up by income. Projecting the future relationship between life expectancy and income, as Baker & Rosnick implicitly do, might be a stretch.
– The Baker & Rosnick paper’s worst results come when you build in a projection of continued rises in income inequality, which is by no means assured (though you wouldn’t want to bet your house against it).
– Patrick’s right again on the potential regressivity of age pensions. One repercussion of the Baker and Rosnick paper is that to the extent that it delivered benefits to the top half of the income distribution, the age pension system itself would be in some sense regressive. To the extent that you take the pension away from the upper half of the income distribution, you remove this regressivity.
– Reformers should therefore concentrate on making the age pension more targeted. It’s a two-for-one, increasing fairness while lowering costs.
– DD is right in pointing to morbidty (the incidence of ill-health) as a crucial issue in the ageing debate. I differ from DD in being less sure that “modern medicine is far better at keeping people alive than keeping them fit”, i.e. that old-age morbidty is rising. There, surprisingly is a lot less data than you’d want on this point, but the weight of evidence seems to be that people stay relatively healthy until about three years before they die, no matter how long they live. (I am very open to persuasion that this entire statement is wrong.)
– So while we talk about this issue as “ageing”, in fact “years until death” is a crucial concept in this debate. Baker & Rosnick are arguing that years until death at the pension age has soared for the top half of the income distribution, but has grown only slightly for the bottom half and could even fall in the years ahead.
– As Martin’s comments suggest, any decent pension age reform should ensure that workers who are genuinely disabled by years of hard physical work can be taken care of. (You could, for instance, introduce some new and more generous form of disability pension for older workers with long working histories. Nick is rightly worried about moral hazard, but the problem which vexes him arrived long ago in the form of the disability pension. We have slowly and partially solved it.) The most recent thorough Australian proposal on the pension age, by David Knox for CEDA, specifically said we must deal better with those whose working history has left them disabled. The Rudd government’s failure to address this issue in 2009 tends to underline its tendency towards careless policy implementation. It was really sloppy work.
– As Ken suggests, participation incentives for aged workers really do matter in an ageing society. They should receive more attention, and almost certainly will be an issue over the next decade.
Note: Be aware of the distinction between pension age and retirement age, often glossed over but particularly important in a system with large amounts of private retirement funding, like Australia’s.
I’m pleased DW has acknowledged that my contribution isn’t completely misconceived. Everyone else completely ignored it for some reason.
However it seems to me that, if the major reason that this issue is being discussed is the inter-generational effects on the Federal Budget bottom line (and I presume that’s why we’re discussing it), then surely we should be exploring whether the problem can be solved or radically reduced in scale by incentives towards remaining in the paid workforce longer voluntarily instead of the blunt instrument of a higher mandatory retirement and/or pension age.
I would be most interested to know, for instance, how many Troppo readers for whom retirement is something proximate enough to be something they’ve actually thought about would seriously contemplate working for more years (at least part-time) if there were either tax incentives, a slower deduction of earnings from pension payments than at present, or award and agreement restructuring to facilitate much more flexible working arrangements. A straw poll of troppo readers would of course have no statistical validity whatever because almost certainly our readership is skewed towards the upper end of the while collar workforce, but it would be interesting just the same. Even if one only gets a significant uptake among white collar workers I suspect that would make a big difference to the Budget effects.
It should be clarified that ‘retirement age’ is the point at which State social security payments can be accessed, not the point at which anyone can choose to leave the workforce. Anything else will have folks thinking you’re advocating slavery as Conrad “allow people to retire” is doing above.
Any increase in the age of accessing social security will impact ONLY those in the lower income groups who cannot afford to become self-funded.
Any attempt to control when people are permitted to leave the workforce is slavery. So long as those individuals can fund their lifestyle from their own resources, there is zero rationale for the State to interfere.
Re Ken’s comments (14), I retired (left the workforce) some 20 years before the supposed ‘retirement age’. We are totally self-funded, having a great time, and
there is nothing short of a complete economic meltdown that could force either of us back into any form of active economic activity. Much better to be an investor.
Ken, I’m getting towards retirement age and have thought quite a bit about just that issue. In my case I can tell you that the income effect – retaining more money from working means I can afford to retire earlier or at least move to part time work – would dominate the substitution effect.
This was a criticism made, for example, when the extraordinarily generous tax treatment of superannaution withdrawals for over 60s was introduced in 2006. The stated purpose was to encourage people to have more super to draw on, the actual effect may have been to make some people retire earlier because they can now afford to. But the truth is that, except perhaps for low income groups, the empiric data is pretty lacking on a lot of this.