The PBO Report on sustainability of benefits to the elderly

The Parliamentary Budget Office has released a Report (pdf) on the fiscal sustainability of benefits to the elderly.

This is not my area. I don't have a lot to say on the subject. I want to say a couple of things.

1. Do the numbers look right?

Stephen's old post linked to Statistics Canada demographic projections. This is from the Statistics Canada highlights:

"In 2009, for every 100 people in the working-age population, there were 24 children aged 14 years or under and 20 people aged 65 years or over. According to the medium-growth scenario, there would be 26 children and 39 seniors per 100 working-age people in 2036."

That increase from 20 to 39 old people, per 100 working-age people, is nearly doubling.

Let me do a very rough back of the envelope calculation. Assume that all of GDP is produced by working-age people. Assume that the number of old people per 100 working-age people exactly doubles. Assume that benefits per old person stay exactly the same percentage of GDP. Then benefits will double as a percentage of GDP.

Let's compare my BOE calculation with the PBO Report:

"PBO's 2011 FSR baseline projection of federal elderly benefits shows an increase in the cost of the program of 1 percentage point of gross domestic product (GDP) from 2.2 per cent of GDP ($36 billion) in 2010-­‐11 to a peak of 3.2 per cent of GDP ($142 billion) in 2036-­‐37 (see below), reflecting the entry of the baby-­‐boom cohorts into the 65-­‐and-­‐over segment of the population. This projection is somewhat higher than projected by the Office of the Chief Actuary as it includes some 'enrichment' to benefit payments over and above the statutory indexation to inflation."

An increase from 2.2% of GDP to 3.2% is GDP is an increase of 45%. That is much less than the 100% increase (doubling) given by my BOE calculation.

OK, my BOE calculation is probably an overestimate. Not all of GDP comes from labour, some comes from capital income, and the old have more capital than the young. Plus some old people will continue working. So let me revise my estimate. Assume that each old person contributes half as much to GDP as each working-age person.

So in 2009 we have 100 working-age people plus 20 old people, producing a GDP of 110. And in 2036 there are 100 working-age people plus 40 (should be 39, but lets keep the arithmetic simple) old people producing a GDP of 120. So we go from old people per GDP of (20/110)=0.182 in 2009, to (40/120)=0.333 in 2036. The increase from 0.182 to 0.333 is an increase of 83%. That's less than doubling, but still a lot bigger than the PBO's estimate of 45%.

Even if I make the extreme assumption that old people contribute exactly as much to GDP as working-age people, I still get an increase of 71% in old people per unit of GDP.

I don't get it. My guess is that the PBO is assuming that benefits per old person fall as a percentage of income per person, even though they are assuming full inflation-indexation plus some "enrichment". But the fall would have to be a big fall, to make our numbers compatible.

Maybe I got something wrong. I was never great at math.

2. Is "sustainability" enough?

Here's the basic message of the Report. Assume Canada recovers from the recession and the budget goes back to normal. Assume taxes and spending (excluding interest on debt) stay the same as a percentage of GDP. Then the Federal Debt/GDP ratio will decline over time and eventually fall to zero. Now add in the demographics. The increase in benefits to the elderly will slow down the rate at which the debt/GDP ratio falls to zero, but it will still fall to zero. Therefore it's sustainable.

And, from my quick reading of the papers, the reaction has been: "OK then, nothing to worry about here; we don't need to do anything, so let's do nothing".

I don't think that conclusion follows. If your plans for the future are not sustainable, you've got a problem, and need to change your plans. But just because your plans are sustainable doesn't mean you don't have a problem and don't need to change your plans. And what were our plans anyway? Were we planning to keep on doing exactly what we are doing now? Were we planning to keep taxes and spending exactly the same percentage of GDP as they are now? It is very obvious we weren't, or shouldn't have been. Because we presumably weren't planning to have the debt/GDP ratio fall to zero, then keep on falling into negative territory, and only stop falling when the government owns everything and runs out of things to buy. That's not sustainable either!

Presumably we weren't planning to make the Federal Debt/GDP negative. We were planning to reduce it to some lower level, and then cut taxes and/or increase spending to keep it at that level. So that when an emergency came up (like the recent recession) we would have plenty of room to allow a temporary deficit without worrying too much about it.

It would be nice to think that we had already factored demographics into those plans. After all, we could have seen this one coming since 1966, when the baby boom ended. But did we? David Dodge says that previous governments wanted to take action in the 1990's, but were scared to. Maybe the rest of us just didn't want to think about it, until we had to. OK, now let's think about it.

3. Federal benefits and provincial health.

Demographics raises two big fiscal problems. The first is Federal benefits, which is what the PBO Report is about. The second is health spending. And health is a provincial question. And from Bill Robson's estimates (pdf), health spending for the elderly is a much bigger deal.

If we had a unitary system of government, we would roll the two in together, and talk about demographics and the fiscal sustainability of the pair. But we don't. But we can't ignore the relationship between Federal and provincial finances. Either the Federal government will have to increase future transfers to the provinces to help them cope, or the Federal government will have to cut future tax rates to allow provinces to increase theirs. Or, we just ignore the question and let the younger generations carry the load with higher taxes to pay for boomers' benefits plus healthcare.

Comments

[Frances Woolley]

Nick: "Assume that the number of old people per 100 working-age people exactly doubles. Assume that benefits per old person stay exactly the same percentage of GDP. Then benefits will double as a percentage of GDP."

I haven't read the PBO report, but my guess is that they're assuming that productivity grows at a faster rate than OAS/GIS/CPP.

I completely agree with you that "sustainable" doesn't necessarily mean optimal. It's driving me crazy that hardly any pundits seem to get this point.

If the price of carrots increased tomorrow to $20 per pound, my current carrot consumption would be "sustainable" in that my weekly carrot habit isn't going to bankrupt me. But would I *choose* to buy carrots if they cost $20 per pound? No.

[Frances Woolley]

Nick, reading the report, I'm pretty sure that all of the action is coming from the underlying assumptions about economic growth.

They assume that elderly benefits will increase by less than the rate of GDP growth thus "In  the  absence  of  population   ageing,  projected  federal  spending  on  elderly   benefits  would  decline  to  1.4  per  cent  of  GDP  over   the  long  term." Population aging doubles that to 2.8 percent of GDP, and since this is "over the long term" rather than "at the peak", that's not totally out of line with your back of the envelope calculations.

But I can't figure out what growth rate they assume.

[Nick Rowe]

Frances: Yep. And if we had had lower taxes or higher spending now, the benefits would be unsustainable. Would that be an argument to cut benefits? Or an argument to raise taxes or cut other spending? There's just something very strange about the logic of the argument being made.

In defence of the PBO though, they don't really make that argument. They are just answering a question: "will we go bust if we keep on doing what we are doing?". But it's a strange question to tie to the OAS/GIS question. And the pundits do seem to be taking an answer to the debt question to be an answer to the OAS/GIS question.

[Nick Rowe]

Frances: I think that must be it. But is that a reasonable assumption? Is that an optimal policy? And is that what people think will happen? If you told people: "Don't worry, we will just make sure that OAS/GIS benefits fall relative to the average income" would they say "OK, nothing to worry about then"? Plus, it doesn't really line up with the assumption that all other taxes and spending stays the same as a percentage of GDP.

What do we want: delay till 67, or lower benefits relative to average incomes? Personally, I think I would want to have at least a bit of delay in the mix. Especially since people live longer and are (supposedly) healthier nowadays, and fewer of us do heavy manual work.

[Bob Smith]

Frances, the growth rate assumption is contained on page 34 of the 2011 fiscal sustainability report (there's a link in this report). They're assuming a 1.8 growth rate (1.1% on a per capita basis) over the next 75 years.

Nick, agreed, not clear why the mere fact that we can pay people to sit on their duff for the last quarter of their life, means that we should. Personally, I'd take this report to confirm that a polity which cuts all OAS spending and gives it to me is "sustainable" over the next 75 years. Sweet!

There's a few questionable assumptions built into this report.

First, I note that the PBO seems to be assuming that all "operating" spending (undefined) will grow at the rate of population and inflation over the next 75 years (see page 6). From the context (it refers to Ottawa's ongoing review of its operating spending), it suggest that that's Ottawa's expenses (military, RCMP, federal civil service) rather than transfer payments. If true, then yes, it's easy to see why the PBO concludes that OAS is "sustainable".

That's also a preposterous assumption. Can we really assume that federal civil servants will accept a real wage freeze for the next 70 years (while the rest of us, presumably, get much richer?). Certainly that hasn't been the case over the past half century. A more realistic assumption would be that federal program spending increases at the rate of nominal GDP (which would catch population growth, inflation and increases in real GDP per capita), which was the assumption made by the PBO in the the 2011 fiscal sustainability report for all program spending (other than health and seniors). Given that that's a dramatically different assumption from what was in the 2011 FSR, I hope that's a typo on the part of the PBO.

The PBO's conclusion is also premised on the assumption that the federal governments "de-escalation" of health transfers will stick. But for that assumption, the PBO would have concluded, as it did last fall, that the federal government had a 1.2% fiscal gap over the next 75 years. Maybe that change will stick (although I can't help but notice that the same people who are opposed to changes to OAS also though that the change to health care transfers "gutted" medicare and should be reversed, so go figure). On the other hand, and its a point the PBO acknowleges in footnote 8, that's a change that doesn't make Canada's fiscal structure as a whole more sustainable, we still have a 2.7% fiscal gap over the next 75 years, it just shifts the burden to the provinces (who, at least with respect to health, probably should bear it, since they're the level that might actually be able to reign in health care spending).

Canadian governments collectively, according to the PBO, aren't sustainable, so we'll have to decide where we come up with that gap. In today's terms, call it $50 billion a year. I suppose we could triple the GST or the provinces could cut all welfare spending, but I suspect that modest changes to OAS benefits would be preferable to either of those options.

[Ryan]

I agree with Frances Woolley; typically these kinds of "predictions" make large assumptions about capital growth. The Wall Street Journal put out a graph a couple of years ago showing actual GDP growth graphed against Congressional GDP annual growth estimates. The result was pretty laughable.

I think everyone knows that socialism doesn't work, that it's mathematically impossible and economically destructive. I think these "studies" only serve to cast a veneer of credence over an untenable position.

What bothers me (as I indicated on my blog a few days ago) is the fact that no one in Canada seems to mind that old-age pensions to Baby Boomers are a transfer payment from the comparatively poor to the comparatively rich. It is hideously unethical considering the fact that younger generations are going to face much dimmer economic prospects than the Baby Boomers. For that reason alone, these benefits should be eliminated. Any parent who would knowingly leave his/her children worse-off than they themselves were just isn't being a good parent.

[Jacques René Giguère]

Ryan: boomers made massive investment in human capital, among other things, that made our children more productive and able to sustain us. So did our parents. As my father is fond of saying : "We paid your university. I'll enjoy the rest of my life."
I have a far greater standard of living than my stevedore grandfather and my nephews seem weel-started in their life.

[Nick Rowe]

Bob: good comment (substantive, as usual).

"They're assuming a 1.8 growth rate (1.1% on a per capita basis) over the next 75 years."

I would have thought that a declining percentage of working-age people would mean a slower growth rate of GDP per capita. Increasing capital per worker (with more old capitalists per young worker) might increase the growth rate per worker a bit, but otherwise I would go for constant growth rate of GDP per worker, not per capita.

[whitfit]

Nick says:

"What do we want: delay till 67, or lower benefits relative to average incomes? Personally, I think I would want to have at least a bit of delay in the mix. Especially since people live longer and are (supposedly) healthier nowadays, and fewer of us do heavy manual work."

Ok as far as you and I are concerned, but I see this position coming from all sorts of intellectuals with good jobs who are likely the people that are getting the good part of the expansion in life expectancy, and where working past 65 isn't a hardship. People who are poor aren't likely to live as long, and are likely to be in worse shape (and working unpleasant jobs) at 65. I would agree with a means tested delay in these programs, but for people who's life expectancy hasn't increased, it seems like a real significant reduction in their benefits to extend the eligibility age. Bob Smith argued in another thread that because the poorer people are dying young from their own poor choices, we shouldn't be concerned with that discrepancy, because they aren't, to borrow another threads terminology, deserving poor - not a position that I find convincing.

Ryan says:

"I think everyone knows that socialism doesn't work, that it's mathematically impossible and economically destructive. I think these "studies" only serve to cast a veneer of credence over an untenable position."

I'm not sure what you mean by socialism, but if you mean that the state relieves the poverty of some of its citizens (which I think is an idiosyncratic definition), I think that it is far from economically destructive, and that it is not true the everyone knows that it doesn't work.

[Nick Rowe]

whitfit: I think your point is generally correct. Some jobs are harder to do at 65 than others. So it's always a compromise where we draw the line. But my point is that I think there are fewer jobs that are harder to do at 65 than there were in the past. So, other things equal, we ought to be drawing the line at a later age than we did in the past. Of course, other things are not equal. And maybe we didn't draw it in the right place in the past anyway.

[Patrick]

Ryan: Not so sure about that claim of transfer from poor to rich. For that to be true, the rich would have to receive more in CPP, OAS, GIS during retirement than they payed in taxes into those programs during their working lives. I don't know enough about the programs to say for sure, and the structure of them might make it hard to figure out, but off the cuff it seems unlikely that it's an issue.

Really, this is a demographic thing. There's no grand allegorical tale to be spun, no good heroes and villains. Just a bunch of people getting old, sick and ultimately dying. I won't dispute that it's possible that the oldsters will vote themselves a giant kegger on the youngins tab but that hasn't happened yet.

If anything, I'd like them to party a bit more! Where's my Moon base!? And it's been 50 years and nuclear fusion still doesn't work... I'm closing in on forty FAST and all the great stuff I dreamed about as a kid has been supplanted with hip replacements and pensions. It's gawd awful depressing. I didn't sign-up for this. Where's Zephram Cochrane when you need him?

[Ryan]

Jacques, I think the idea that the Boomers' investment in human capital was genuinely of productive value is not entirely supported by reliable evidence. We know a great deal of the supposed value has been over-stated by changing benchmarks of inflation rates, by tech bubbles, by housing bubbles, higher education bubbles, etc. It's clear that the Boomers were able to generate some cash for themselves, but a lot of it came at the cost of other generations. (i.e. Anything that came as the result of a market bubble came at the cost of future generations and was a net *decrease* of capital. And an over-credentialized job market is *hardly* a blessing for younger generations and should never be misconceived as real human capital.)

Whitfit, there is only one definition of socialism: state ownership of the means of production. In the case of old-age pensions, that means state ownership of the means of producing pensions, i.e. investment capital. Of course, this is merely one example, and things like Canada Post, provincial health care plans, Petro Canada, Export Development Canada, VIA Rail, and whatever else you like also certainly qualifies.

[Determinant]

OAS goes to low and middle income seniors and starts being clawed back at $66K. Further most pension income, OAS and CPP is taxable. The increase in OAS is not directly proportional to the increase in population, it is less due to the preceding factors.

Further the federal government only spends 1/3 of its revenue on its own programs, including the Public Service. 20% goes to the National Debt and the rest goes as transfers.

[Nick Rowe]

Determinant: "OAS goes to low and middle income seniors and starts being clawed back at $66K. Further most pension income, OAS and CPP is taxable. The increase in OAS is not directly proportional to the increase in population, it is less due to the preceding factors."

Your third sentence doesn't seem to follow from your first two. For a simple thought-experiment, suppose we duplicated every old person. So we double the size, leaving the distribution of income the same. Then every measure of expenditure would double. Before tax would double, net of tax would double, etc.

[Frances Woolley]

whitfit: "Ok as far as you and I are concerned, but I see this position coming from all sorts of intellectuals with good jobs"

Why, at 65, are people suddenly entitled to dignity and a decent level of income support? Why not at 60? 55? 50? Heck, why not at 20?

[Determinant]

If every measure of expenditure doubles due to population doubling, you get doubling of supply and income too. Those new seniors have taxable incomes. It's taxes in the form of money that pay for OAS, not GDP directly. Rich retirees pay for OAS as much as a young worker does.

[Bob Smith]

Whitfit: "I would agree with a means tested delay in these programs, but for people who's life expectancy hasn't increased, it seems like a real significant reduction in their benefits to extend the eligibility age."

But a "means-based" delay raises other problems, doesn't it? Why should someone who is poor, but healthy, get to retire at 65, while someone whose middle-class, but sick, has to work until 67? That's an odd result, no? It's one thing to say that "poor" people have shorter life expectancy than richer people, on average, but individuals aren't average. Also, what is the relevant time for measuring "means"? If you started your life working in a mine, but rose to affluence in your older age, should you be able to retire at 65 or 67, in your model. Conversely, if you're a trust fund kid who blew it all on the ponies, should you be able to collect OAS at 65 or 67? The former may be richer than the latter, but we both know whose life was harder.

Also why do we privilege (no pun intended) those whose life expectancy is limited by income/wealth? Life expectancy has a significant genetic component. Shouldn't we allow people who "lost" the genetic lottery to retire early too? What about people who have life limiting diseases or conditions (obesity, cancer, MS, etc.) shouldn't they get to retire early? If my doctor tells me that, statistically, I have five years to live, wouldn't it be fair to let me hang up my spurs at 34? Men live shorter lives than woman, British Columbians live longer than Newfies, aboriginal Canadians have lower life expectancies than white Canadians, how far down this road do we go? And yet, there's no rational distinction between the income/wealth based argument that you're making and some of these other distinctions (other than the fact that people have absolutely no control over their genes, gender or race).

To answer my own question, I think the reason we don't go down that road is because the purpose of CPP/OAS/GIS is not to provide people with a given number of years of retirement. It's to provide for them when they can't work. So the fact that people who are poor, male, aboriginal, obese, Newfoundlanders, what-have-you (and you can try to guess which two of those catagories I fall into) live shorter lives is irrelevant to the policy aim of the CPP/OAS/GIS system. What is intended is to provide benefits for people who can't work any longer (albeit, on a very crude, rough-and-ready basis). If, as Nick suggests, we're now living longer, healthier lives, it makes sense to push that threshold level back.

You might want to revisit my earlier comments to Frances on exactly this point. My suggestion was to continue to permit (or at least fund) retirement at 65 on the basis of health (as we do now with disability benefits under CPP for people under 65). Clearly, if you can't work, asking you to work two more years is harsh. But that's a proposal that links early retirement directly to medical need, rather than indirectly (and rather vicariously) through income.

[Ryan]

Patrick, The villain is socialism. It robs us of our capital, our economic freedom, and our sole means of relating to each other as equals. If you want Baby Boomers to party, then that's great. I don't want any generation to do anything as a generation at the expense of any other generation. What I want is for all wealth to be invested freely in the prospect that yields the highest profit for the investor. What I want is freedom.

The fact that this "demographic shift" has been obvious since the 1950s and no one has bothered to do anything about it until now demonstrates both the impotence of central planners and destructiveness of socialism.

The fact that Boomers made a lot of cash is great for them. The fact that governments forced society to consume its investment capital, creating a situation in which past pension schemes are no longer financially viable and future economic growth will forever be lower than it otherwise would have been is reason enough to see that socialism is the villain.

AFAIC, there's no other way to look at it. A person's income at retirement shouldn't be a matter of the Social Contract. Public pensions should have been eradicated the minute someone realized that they were not indefinitely sustainable, i.e. before they were even created.

[Nick Rowe]

Bob: "...British Columbians live longer than Newfies..."

Sample selection bias. Some people die before they get to retire to BC. ;)

[Nick Rowe]

Determinant: OK, I think I see what you are saying.

1. The PBO did benefits as a percentage of GDP, so I did the same.

2. Ultimately, the elderly want consumption of goods, not money, and that consumption has to come from production of goods. So I think benefits/GDP is (roughly) the right way to look at it.

[Frances Woolley]

Ryan:

Sorry, you got me in a nitpicking frame of mind...

"The fact that this "demographic shift" has been obvious since the 1950s"

Since the 1970s, actually. Birth control pill was introduced in 1962.

"The fact that Boomers made a lot of cash is great for them."

Huge difference in economic outcomes between people born right at the front of the boom, who are just riding that sweet wave all the way through, and people born right at the end of the boom, who always arrive just after the party's ended.

[Ryan]

Frances, maybe we're using competing definitions. My understanding is that the Baby Boomers are the ones born after the end of WWII, when the birth rate temporarily exploded before eventually subsiding again. But perhaps you're right that the generational bulge didn't become obvious until later.

As for your other point, we can put those Boomers-at-the-end-of-the-Boom in the same category as "the younger generations" and the point still holds.

I'm not trying to put blame on anyone, I'm simply making the point that it is generally true of transfer payments that they go from haves to have-nots. Baby Boomers' pensions are the opposite.

[Nick Rowe]

Ryan: have a look at Stephen's graphs here

The Canadian baby boom began in 1947, peaked in 1961, and ended in 1966, IIRC.

[Nick Rowe]

Frances: "Huge difference in economic outcomes between people born right at the front of the boom, who are just riding that sweet wave all the way through, and people born right at the end of the boom, who always arrive just after the party's ended."

And my guess is that something similar may happen again, if we don't do something now. We will let it slide, until eventually the working age people revolt, and raise the age for eligibility, or cut benefits, just as the late boomers hit 65, which is when the problem peaks.

[Bob Smith]

Nick: "Sample selection bias. Some people die before they get to retire to BC. ;)"

I was thinking measurement error, people move to Victoria and it takes two years before any one realizes that they're dead.

[Determinant]

Ryan:

Canada's baby boom began a year later than the American one; it is a documented fact that Canada was slower to demobilize than the US was.

The current definition of Baby Boom is 1947-66 when the fertility ratio, the number of children per mother, reversed the previous trend of decline and rose. The fertility ratio had been declining in all industrialized countries since the late 19th Century. At it's peak Canada's was also the biggest, the average Canadian mother was having 4.0 children in the late 1950's. In the US it was 3.8.

[tyronen]

You people are making the same mistake as the gullible Canadian pundit class.

OAS and GIS are not indexed to GDP growth. They are indexed to inflation, nothing more. Whenever the economy experiences real GDP growth, this *reduces* OAS/GIS as a percentage of GDP, if the senior population was constant.

So your assumption at the beginning "assume benefits per old person stay exactly the same percentage of GDP" is flat-out wrong.

This effect has been seen before. OAS/GIS peaked at 2.7% of GDP in 1993, fell to 1.8 earlier this decade, and are at 2.3 today. They rise to 3.1% by 2030, since the senior population is not constant, but this hardly seems unmanageable. If we could afford 2.7% in 1993, why can't we afford 3.1% in 2030?

[Nick Rowe]

tyronen: There is formal de-jure indexation, and informal de-facto indexation. Administered prices, and benefits, may be adjusted over time even if it's not written into the law. Do you think it's plausible that real (i.e. inflation-adjusted) OAS/GIS would stay constant if real incomes were rising over time? Do you think there would be political pressure to increase them, based on relative incomes, or relative conceptions of poverty? What do you think most people assume?

My guess, or benchmark assumption, is that, if it weren't for demographics, we would want to keep roughly the same relationship between pre-retirement incomes and OAS/GIS benefits. Otherwise people would face a bigger and bigger drop in income at retirement over time.

"OAS/GIS peaked at 2.7% of GDP in 1993, fell to 1.8 earlier this decade, and are at 2.3 today."

Interesting. Do you know why it fell and rose again? That can't have all been demographics, can it?

"If we could afford 2.7% in 1993, why can't we afford 3.1% in 2030?"

1. There are lots of things we could afford to do, that we choose not to do.

2. I don't think it will be 3.1%.

[Bob Smith]

Tyronen,

Actually we're not making that assumption, the PBO is. Hardly the "gullible pundit class". And it's making that assumption on the pretty plausible basis that people will expect support for the elderly to broadly track increases in the general living standards, and that the relevant laws will be amended to reflect that.

"If we could afford 2.7% in 1993, why can't we afford 3.1% in 2030?"

Do you recall the $40 odd billion dollar deficit that Canada was running in 1993 (or our characterization as an honourary member of the third world by the WSJ a few years later)? We couldn't afford 2.7% in 1993.

[Kathleen]

A comment on the impact of having a reduced number of workers compared to the number of retirees. There are three factors that I would like to see addressed:

1. When Canada imports labour intensive goods are we not importing labour? To the extent that these goods are intermediate goods will this not contribute to GDP growth? I agree we will not receive the income tax that we would if there were actual immigration, but won't we benefit from the "foreign" labour productivity.

2. There is room for improvement in Canadian labour productivity. There are many reasons that Canada lags in labour productivity growth but one of them is not that Canadian businesses do not seek to maximize profits. With a shrinking labour force and higher Canadian dollar, are we not seeing the conditions for improved labour productivity?

3. I am of the opinion that Canada's unemployment rate is high, and has been for decades. Why are other countries operating at a consistently lower unemployment rate? If we reduce our unemployment rate will this not contribute? What is needed here: training or higher wages (Canadian workers are choosing unemployment over minimum wage jobs).

Thanks for this excellent post thread. It is really great to have access to this type of discussion.

[Ryan]

All:

I'm not sure why the dates matter with regard to my core point. Draw the line wherever it is appropriate. My original point has little to do with dates. If the boom wasn't obvious by the 1950s, then it was obvious by the 1970s, and we are simply talking about 40 years of inaction as opposed to 60.

[Bob Smith]

Kathleen,

I'm not sure about point (3). Admitedly, comparing unemployment rates between countries is a tricky business, but my impression has been that, at least since the 1990's, Canada's unemployment rate, or perhaps more importantly, Canada's employment rate (which reflects the participation rate) has been higher than that of most other developed countries (with the notable exception, until recently, of the US). I'm pretty sure there is OECD data on this point.

I'm also not sure about point (2), all the businesses I work with are pretty keen on maximizing profit. There are a lot of reasons for Canada's lagging labour productivity (I'm partial to the argument that for many years we didn't need to improve productivity, because we could (and did) devalue our currency, but lack of greed probably isn't one of them.

[Bob Smith]

Tyronen,

Just to follow-up on my earlier point, we've been focused on the demographic and economic implications of an aging population, but we should also think about the political implications of an aging population. Amongst other things it means that the percentage of the voting population that will be collecting OAS/GIS will be significantly larger than it presently is. The over-65 set are already a politically powerful group, but they'll be that much more powerful by 2031.

That has two implications. First, as Nick and I mentioned above, that makes it very likely that OAS/GIS payments will increase to reflect general living standards. Second, it means that the (political) ability of the federal government to introduce reforms to constrain or reduce the OAS/GIS burden (as with the ability of provincial governments to contrain health care spending) is likely to decline over time. Given those pressures, if we're going to be making changes, now's the time to do it, because we won't be able to do it in 15 years time.

Hey, if the economy booms over the next decade and the demographic pressures get better (higher immigration, people start dying sooner), it'll be easy to increase OAS/GIS benefits, or loosen eligibility. But, if economic growth is slower than expected over the next decade (a real possibility given the organized anarchy going on in Europe and the political dysfunction of our US cousins), or demographic preassures get worse (someone discovers a cure for cancer), we aren't going to be in a position to make those reforms down the road. I'm inclined to err on the side of caution.

[Kathleen]

Bob Smith,

On point 3: yes, my opinion is just that, an opinion. Nevertheless, is there not room for improvement on our unemployment rate? I really have to wonder why workers and businesses can't seem to agree on a wage. It can't be true that we have a labour shortage and high unemployment at the same time. Where is the disconnect? Is it training or wages?

On point 2: my wording was awkward. My point is that Canadian businesses do seek to maximize profits therefore if they are not pursuing productivity enhancements there is some cost or revenue barrier that they are facing. I agree that with a low dollar perhaps they didn't "need" to invest in productivity enhancements. If they begin to do so now, won't that lift our GDP growth rate?

It is really point 1 that I am most interested in. I think of importing labour intensive goods as a form or offshore outsourcing that does not have the disadvantage of displacing Canadian workers.

[Nick Rowe]

Bob: "Actually we're not making that assumption, the PBO is."

I've been in email contact with Chris Matier, from the PBO, who wrote the report. He's been very helpful. It turns out to be more complex and a lot weirder than that. There's a very interesting point here, that's totally non-obvious. I will be doing an update post soon.

[Patrick]

Kathleen: "It can't be true that we have a labour shortage and high unemployment at the same time. Where is the disconnect? Is it training or wages?"

FWIW, I think the problem is in thinking about *a* labour market. There's no such animal. There's one market for each job opening, and each is dysfunctional in its own special way. If each job is essentially unique, but firms want to find experienced labour (which obviously doesn't exists since each job is unique), then you I can easily imagine a world where workers end-up unemployed and firms think there is a shortage of skilled labour.

As the boomers filling all the one-of jobs retire en masse this problem is only going to get worse.

[Jacques René Giguère]

Bob Smith: "or our characterization as an honourary member of the third world by the WSJ a few years later)?"
It was not the WSJ, at the time a very respectable and professionnal publication. It was the WSJ editorial page, two pages of absolute lunacy under Jude Wanisky. A big , big difference ( less so today).

[Jim Sentance]

Ryan, I think the problem is less the boom, which would have been apparent by the 50's as you say, but the bust, which would have been well established by the later 60's. If we'd kept it up there would be plenty of people to cover the cost.

[Determinant]

Stephen Gordon had a great graph where he showed we are "short" 11 million people because of the decline in birth rates.

[Yancey Ward]

Well, what has happened historically since 1950, for example? Has the average senior's total entitlements per capita grown relative to the income of the working age group? If so, then the argument that growth can explain the discrepancy seems a bit odd to me.

[Ryan]

Jim, we experienced an anomalistic population growth spurt. The idea that "old age pensions would be fully affordable, if we'd keep to Baby Boom birthing rates" is a little silly, IMO. Sort of like saying "We'd never experience an aggregate demand shock if we'd all just earn more money."

The question becomes: What datum is going to be the one that finally convinces us that these pension schemes are destructive and unaffordable? How badly do we have to fail before we're willing to confront our mistakes?

[Nick Rowe]

Yancey: I don't know the historical experience. But one thing that has changed, I've heard, is that OAS used to be payable only at age 70? Presumably there have been other changes too, to the level of benefits (relative to incomes).

Ryan: I thought Jim's point was sensible. People in 1965 couldn't predict future birth rates. Yes, maybe we should have made provision for the possibility that birth rates would fall. If these sort of future costs were budgeted in a fully-funded plan, as opposed to Pay As You Go, that might help.

[Frances Woolley]

Yancey: "Has the average senior's total entitlements per capita grown relative to the income of the working age group? "

The most informative stats on the relative size of pension benefits v. social assistance and other benefits for "working age" people with no other source of income v. low wage work is the poverty rate.

Back in the 1960s, seniors, especially single senior women, had a very high poverty rate - my image of poverty as a child was a little old lady eating cat food.

Now seniors have a fairly low poverty rate - the demographic least likely to live in poverty in Canada is married couples over 65, and a disproportionate share of the poor in that group are immigrants who are not entitled to full OAS benefits.

Today the face of poverty - the person scavenging food from a dumpster - is a single person unable to find work.

[Ryan]

Nick,

Fair enough, but 1965 was almost 50 years ago. And the Baby Boom was being called "the Baby Boom" since before I was born. We're talking decades. We can perhaps make an excuse for the central planners of 1965 (although not a very good one), but we can't make that same excuse for every central planner of the last 50 years.

My point is that the excuses themselves are a little irrelevant because the pension plan itself is based on socialism: a factual impossibility. The transversality conditions of public debt prevent any country from being able to deficit-spend into infinity. The market-shrinking impact of public revenue generation prevents any country from experiencing a permanent level of economic growth adequate enough to sustain a socialist country into infinity.

We should be honest with ourselves. These government pension schemes aren't sustainable; the transfer payments are going into the hands of the "haves" rather than the "have-nots;" the younger generations cannot afford to pay for them.

Am I the only one not willing to make any more excuses for it? It's bad policy.

[Mandos]

Ryan has proven socialism wrong with SCIENCE! Therefore we must make grandma eat cat food.

[Bob Smith]

"Today the face of poverty - the person scavenging food from a dumpster - is a single person unable to find work."

A good point that has been missed by all sides in the debate. I would have thought the principled left-wing approach to reforming OAS (or playing hardball with public sector unions - but I digress) is that it would allow for the better use of public funds for those in greatest need. Not that I expect politicians (of any political stripe) to be particularly principled when it comes to getting votes. Still, if the Tories were clever they'd devote some portion of whatever savings they think they can generate through OAS reform to providing, say, long-term sustainable funding for on-reserve schools (young aboriginal Canadians being a group that's more genuinely in need of government assistance than your average 65 year old).

[Mandos]

One thing that's astonished me since I started reading this blog is the number of people who think that the principled left is similar in goal to a charity operation...

[Bob Smith]

One thing that's astonished me since I started reading this blog is the number of people who think that the principled left is similar in goal to a charity operation...

And one of the things that's astonished me ever since I got interested in politics is the number of people on the "principled" left who, notwithstanding their rhetoric about assisting the most vulnerable, sticking it to the man, yada, yada, yada, believe that the role of government is to act as a charity operation for particularly privileged groups in society (seniors, public servants, etc.). Then again, it didn't take me long to figure out that much of the "principled" left weren't so principled.

[Nick Rowe]

I have to confess, Mandos: I am one of those people. Until I saw your name at the bottom, I thought that your comment had come from a cynical righty.

In other news:

I'm still emailing back and forth with Chris Matier at the PBO, trying to reconcile our estimates.

OAS did originally only start at age 70, and was gradually lowered to age 65 from 1965 to 1969.

The graphs of Marginal Effective Tax Rates for seniors are even weirder and spikier than for the rest of the population. (We really need to start a "Bulldozer Club", similar to Greg Mankiw's "Pigou Club", that would advocate putting a bulldozer to flatten out the spikes and fill in the ravines in METRs.)

CPP was (slowly) transformed from a PAYGO to a funded system in the 1990's, in large part because of demographic issues, but OAS is still seen as PAYGO. I do not understand why they would be treated differently. (You have to be a bit aspy to ask this question, and see through the framing that totally snows all non-economists and other normal people.)

[Frances Woolley]

Nick: "CPP was (slowly) transformed from a PAYGO to a funded system in the 1990's," I don't think it'll ever be fully funded, but is much much closer to a funded system than it was.

Why they were treated differently? We learn from our mistakes. In the 1950s, would you think investing in stocks was a good way of ensuring a safe retirement? No, probably you knew people who had had all of their savings wiped out by the great depression. You would be looking back over the past 20 or 30 years and seeing equities as pretty crummy all told. PAYG would look pretty good.

In the 1990s, however, people looking back over the past 20 or 30 years saw an equity premium! Equities were great! Equities would give 5 % real returns! Hence the PAYG.

[Mandos]

I have to confess, Mandos: I am one of those people. Until I saw your name at the bottom, I thought that your comment had come from a cynical righty.

And that's the problem, quite possibly the main split between mainstream economics and the left. Getting money and goods to poor people is a blessing. It is not a political program, and it's not a program for a principled moral and social order. Because economists have settled on a particular dichotomy between means and ends, the overall policy discussion, the choice of what to measure, and so on is inherently narrowed.

For the left, the end is a particular distribution scheme---or more correctly, to restrict or abolish a particular distribution scheme...