Leviathan, Exit, Voice and Ontario Teachers

The last few weeks has seen the death of two economists – James Buchanan and Albert O. Hirschman - whose work has influenced my intellectual development and thinking over the years.    Their thoughts combined with tomorrow’s “political action” by Ontario teachers against the soon to expire McGuinty government has caused me to think about what forces are at work here.   To be human is to see patterns (I can’t remember who came up with that line but I’m sure it was not me) and the dispute with Ontario teachers can certainly be framed in my mind by the concepts of Leviathan, exit and voice.

Nobel Laureate James Buchanan (1919-2013) is associated with public choice theory but he also wrote the classic American Economic Review Article in 1950 – “Federalism and Fiscal Equity” which made the theoretical case for interregional fiscal transfers  - equalization payments – which I think makes him along with Charles Tiebout one of the founders of the field of fiscal federalism. Another contribution Buchanan made was in work with Geoffrey Brennan on government as a tax revenue maximizing “Leviathan”.  The Leviathan Hypothesis that was put forward by Brennan and Buchanan in The Power to Tax: Analytical Foundations of a Fiscal Constitution argued government was a revenue-maximizing Leviathan operating subject to constitutional fiscal constraints.  The size of government and the centralization of power were positively related and therefore as they wrote: “Total government intrusion into the economy should be smaller, ceteris paribus, the greater the extent to which taxes and expenditures are decentralized.”

Albert O. Hirschman (1915-2012) was an economist who wrote on a wide variety of topics including economic development but what I was chiefly influenced by was his work on how people deal with events and environments that are unpleasant.  His views were presented in a book titled Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations and States.  Briefly, when faced with a situation whose outcome you disagree with, what do you do?  Do you persist out of loyalty? Voice criticism in the hope of changing the outcome? Or simply decide to exit the situation?  What option you select I suppose is ultimately a function of benefits and costs.

So, this brings me to Ontario and the acrimonious situation with its public school teachers.  The government of Ontario has been a tax revenue-maximizing Leviathan over the last decade and has used those revenues to pursue its agenda – which of course has included a lot of spending on education.  Indeed, based on the most recent edition of the Federal Fiscal Reference Tables, between 2002-03 and 2011-12, Ontario’s total provincial government spending grew by 64 percent while its revenues grew only 46 percent suggesting that when it comes to pursuing its objectives, the Ontario government has not been all that fiscally constrained.

Fiscal reality has finally entered the picture and the government is also reining in its teachers.  So how can the teachers respond to what they see as a fairly unpleasant situation?  I suspect the government anticipated the response would be one of “loyalty” after years of generous funding to the education system and that teachers would accept the restraint.  This has not been the case.  For the teachers as it is for most members of the human species, its not what you’ve done for me but what you can do for me.  How about exit?  Well, those teachers that can either retire or feel strongly enough to find another job no doubt have done so but for the most part – they are not exiting.   Indeed, for most teachers, exit is not an option given that they actually do like their jobs, which are meaningful and stimulating, along with good pay and benefits.   This leads to the final strategy left – voice.  The teachers are voicing their dissatisfaction and hoping to change the outcome.  Will they be successful?  Not likely.

Leviathan is not just a revenue and expenditure maximizer, it is also a vote maximizer and the teachers have delivered Leviathan an excellent gift.  In this case, the public school teachers might have been better off in the long run if they had simply accepted the restraint package out of “loyalty” – as the Catholic teachers have done – and embraced the old adage, of “he who fights and runs away, lives to fight another day”.  For the Ontario government, the fight with the teachers is a political gift.  It makes the public school teachers a lightening rod for public dissatisfaction and diverts the general public’s attention away yet again from the proroguing of the legislature, the province’s debt, the very expensive wind and solar energy policy, rising hydro rates, the eHealth scandal, the Ontario Lottery and Gaming scandal, the situation at ORNGE, the power plant scandal during the most recent election, etc. Leviathan is quite happy to see the teachers subjected to vociferous public criticism for the next few weeks until the cleansing ceremony of selecting a new leader is concluded.  Having the public focus their anger on teachers followed by a new leader, new options and the expectation of a fresh start is not a bad strategy for the incumbent government in the run up to the next election.  It might even work. Leviathan has been cynical but brilliant.  I wonder who is advising the teachers?

Comments

[Mark]

Looking at PEA data up to 2009 (CANSIM 384-0004), Ontario revenues as a share of GDP in 2003, 2007 and 2009 are 15.4%, 16.3%, 16.2% in both years. Excluding transfer payments they are 13.2% 13.7% and 13.2%.

These revenue ratios are lower than any other province except Alberta.

Current spending was 16.1% in 2003, 17.0% in 2007 and 19.4% in 2009, lower than any other province except Alberta and Sask in 2009.

If a Leviathan is a 'revenue maximizer' based on these numbers, I wouldn't classify the Ontario government as a Leviathan.

[Livio Di Matteo]

Hi Mark:
Well I suppose it is all about how we want to operationalize the definition of "Leviathan". In the Brennan and Buchanan world all governments are "Leviathans". I don't think there is a continuum with one province being less "Leviathan-like" than another. All governments are revenue maximizers - it is the operating assumption of the Leviathan hypothesis.

[Jacques René Giguère]

Livio: "fights another day"? Which fight will they be able to wage tomorrow that is so unlike today that it will be "victorious"?
The worst with fighting teachers is that it corroborates the popular idea that "fiscal probkems" are due to "overspending" on useless governement workers, as you point out. But teachers are always a good target. For people on this blog, school was a mostly pleasant and rewarding experience,so much so that we went as long as we could and continue working there. For most voters, after kindergarten school is barely above a P.O.W. camp.
A cartoon I saw last week show a man sentenced to hell repeating 5th grade with dreary Ms. Smith. And Satan telling the guy: "But it is Ms. Smith heaven!"

[Determinant]

Along with the old trope that teachers get paid too much. With the sub-trope that their pension plan is way too cushy.

Teachers are the front-line in Canada of the cleavage between the public sector, the vast majority of which has defined-benefit pensions and the private sector, which has either DC pensions or nothing at all, in the main. Group RRSP's count closer to nothing as they often feature the worst of all worlds, high MER's and little choice.

[Livio Di Matteo]

Jacques:
I think the teacher position in Ontario has been weakened by the fact that there is not a united front amongst all the unions as some have accepted the government's position.
Determinant:
Interesting point - given their numbers, the teachers are a front line and if the government gets them in line it will represent a major setback for public sector unions in general.

[Andrew F]

Determinant, defined benefit pensions are either horribly expensive or prone to blowing up. To the extent that workers should have a (nearly) risk-free pension, it must eventually be backed by government with its power to tax. CPP seems reasonable to me. I don't see why only public sector workers should be able to enjoy that guarantee. It seems to me that the pension promises made by government to public sector unions are underpriced, for the usual reasons of agency issues and moral hazard on the part of politicians and unions.

[Determinant]

To the extent that workers should have a (nearly) risk-free pension, it must eventually be backed by government with its power to tax.

False. A Life Annuity provided by a life insurer such as Manulife or Sun Life does the same thing, with the full backing of a deposit-insurance type plan (Assuris). The difference in cost between a life annuity premium and contributions to a DB plan represent the risk discount for the pension.

Further, CPP is not guaranteed by government, it is funded out of contributions and investment returns, eligibility for which is determined solely by a formula (age and years of work). CPP contributions are identical to DB pension premiums, "payroll tax" is an American term most related to Social Security that is entirely inappropriate to Canadian institutional arrangements.

Or in other words, a DB pension premium is a voluntary tax.

The problems with DB pensions may be summarized by the fact that a DB pension is a life annuity a company that runs one runs an insurance subsidiary. In insurance the bigger and more diversified the better, hence the problem with DB pensions in the private sector.

If you think your employer could not competently run an insurance subsidiary, that explains why the private sector had such trouble with this product. DB resorts to an employer guarantee if the saved funds aren't enough; fear of this drove the private sector away. Governments have to pay those costs anyway in other areas so they are more comfortable with them.

All pensions can "blow up", its just DC pensions or RRSP's blow up differently. Poor market returns and high MER's can leave you permanently poor on retirement. One bad investment at 63 and your retirement is toast. Like most good economic problems, it boils down to "which price do you want to pay, and why?"

More along my point, to Livio, the private sector since 1980 has had to adjust to the demise of the idea of an employer as an institution and wages as a form of welfare provision. DB pensions, drug, dental and disability insurance are all forms of private-sector provided welfare. But welfare is best provided by a source that is itself stable and intends to have a stable relationship. But those are also costs and costs that private sector management resists paying. But government has continued with the institutional employment model and its disappearance in the private sector leads to much political strife.

[Andrew F]

CPP is guaranteed by its ability to extract contributions from current workers. Any shortfall is made up by a combination of benefit changes and contribution rates. The only way to flee CPP is the leave the country or start your own business and not pay yourself a salary.

The Assuris guarantee is only as good as the insurance industry itself. If our insurance industry were to fail, or even just one large insurer, either those annuities would be defaulted upon or (more likely) the government would recapitalize the industry and make whole those policies.

My preference would be to have DB pension income sufficient to ensure a reasonable standard of living relative to working year income. I don't think final salary plans fall into that category. I would support an expansion in CPP income replacement rates, and all other retirement income moving to a DC model. Some risk should be borne by the saver. I don't disagree that the private sector has done an appalling job of serving their clients. A pooled DC plan could be administered by CPPIB or another public manager, with private options being allowed to compete under a capped fee regime. The important thing is that pensions should be portable to facilitate labour market flexibility.

Even if DB plans were the norm in the private sector, tenures there are shorter and pension portability becomes a big problem.

[Bob Smith]

"The worst with fighting teachers is that it corroborates the popular idea that "fiscal probkems" are due to "overspending" on useless governement workers, as you point out. But teachers are always a good target."

Not always. People may not have liked school when they were kids, but as adults they generally have a fair regard for their own kids' teachers. In the past the teachers have had a fair bit of political success using that goodwill in their fights with governments. Certainly, that gave them a lot of weight in their fights with the former Conservative government (and meant that their support for the McGuinty government went a long way to help him win in 2003, 2007 and 2011).

I think the difficulty is that the goodwill only holds when the teachers can present their argument as fighting for quality education (which they suceeded in doing, accurately or not, in their fight with the Harris/Eves Tories). When they're seen as fighting for their paychecks, and holding their students hostage to achieve that, not only does it blow through that goodwill, but they hits voters in the place that matters the most to them - their kids. People who might otherwise be sympathetic to the teachers' beef with the government are going to lose that sympathy when the teachers decide to hold their students hostage to try to win it.

All of which makes the teachers' current strategy completely moronic (unless one believes that they're in the pockets of the Conservatives and eager to get Tim Hudak elected). Having built McGuinty up as the education premier over the last decade (with hefty dollops of union support each election), the teachers aren't going to have any success portraying him as a threat to quality education (and McGuinty has been smart in framing this as a choice between wage freezes and cutting all day kindergarden, which remains wildly popular with Joe Q Voter), and not one is buying their "its not about the money" line of argument, because it's plainly about the money (more to the point, if it's not about the money, are you telling me that you're sticking it to my kids on a point of principal - that's almost worse).

Talk about not having a learning curve - the teachers played the same game with Bob Rae vis-a-vis Rae Days (which, in retrospect, werea decidedly labour friendly form of austerity), and look how that turned out for them.

[Bob Smith]

"To the extent that workers should have a (nearly) risk-free pension, it must eventually be backed by government with its power to tax."

Mind you, on one level, government pensions are more risky than, say, a proper life annuity. Yes, governments have the power to tax, and so can back up their promises further than, say, an insurance company (although even then, as the Europeans are learning, the power to tax is not unlimited). But they also have the power to rewrite laws and impose contracts (as Ontario's teachers are learning). If the CPP were ever to run into trouble, one response could be for the government to raise taxes, but the other would be to cut CPP benefits or push back the age for eligibility (as European pensioners are learning). In contrast, your insurer (or Assuris, as a backstop - and if the entire insurance industry goes under, well, civilization as we know has likely ended, and forget pensions, I'm stocking up on canned goods and small arms) can't raise taxes (so in theory, can go belly-up), but it also can't rewrite the terms of your annuity. A government backed pension may have less solvency risk, but it has greater legal/policy risk.

I think the issue with an expanded CPP is the concern that it may provide "excess" pensions, i.e., force people to save more than they want to (or more than that they should). One of Stephen or Mike (I think) had a twitter link recently to an article that made the case that, at least for low-income Canadians, the current regime works pretty well. A reform that resulted in shifting more current income to retirement (i.e., higher CPP premiums, and higher CPP benefits) might well make them worse off. I suppose an optional CPP expansion might address that issue.

[Determinant]

One good reform would be to put the T1213 on a firmer statutory basis. This gem of a form allows you to claim deductions on your paycheque instead of getting them as a refund at the end of the year. Right now it is actually thinly supported, the underlying provision allows "relief to taxpayers who face an excessive burden" for paying their tax liability.

The CRA is slipshod in approving them and employers shy away from implementing them and give you grief if you ask for it. As one of the key ways to implement the retirement saving "nudge", this is ridiculous. The statutory basis should be clear in that you can use it as long as you submit proof of the tax-favoured activity, i.e. a receipt or savings agreement. If you wind up with a tax liability at the end of the year, your employer will have no withholding liability.

In the UK when you save in a personal pension the government gives the provider (usually a life insurer) a cheque directly for the tax relief for your account.

Combine this with increased annuitization requirements and government policy to make pensions the "fifth pillar" of the financial system. Ensure full portability (not really a problem now). There are some decent attempts at local pooled pensions, Moncton has a local DC plan run by Assumption Life for all employers in Moncton, small businesses mainly. The most important reform would be to end DB pensions run totally by employers, the annuity portion should be provided by a life insurer who has a strong incentive to stay in business. Once you are not working for a company, you should no longer be their problem, not even for your pension.

What Bob Smith said about the insurance industry, plus I will note that Assuris took care of Confederation Life when it collapsed and that was just after Assuris was formed. Assuris works by transferring policies to solvent companies. The industry as a whole is based on trust and has a strong incentive to root out problems and make sure they are isolated. It works.

[Bob Smith]

Determinant,

There is a provision which allows you get around the CRA process in respect of RRSP or pension plan contribution where the employer duducts from your paycheck and makes a contribution directly to an RRSP or registered pension (since, in that case, obviously, there's no risk that the employee won't make the contribution). That amount isn't included in computing the tax base for withholding purposes.

Your point, however, is well taken with respect to the rest of us who aren't part of employee group RRSPs or pension plans. It's also worth noting that your proposed arrangement would put people who don't (can't) participate in such plans on a level playing field vis-a-vis those who do, which is presumably a sensible arrangement. If nothing else, you might say that provided you provide your employer with a pre-authorized contribution agreement to an RRSP for a particular year, they can take that into account in computing your withholding without needing to seek relief from the CRA.

[Determinant]

That provision, which I have *tried* to use before is one of those things that works in theory, but doesn't in reality and therefore which needs tinkering. The T1213's underlying statutory provision is weak.

There are good 1% or less MER funds out there that are pension-like (ING is one, for how long after the Scotia acquisition is anyone's guess) but there should be more of them. That's sort-of a government problem but not really one that requires government money to be spent.

[Andrew F]

Indexed annuities are absurdly expensive, given that real return bonds currently yield about zero. That's why I said that DB pensions are either liable to collapse or are absurdly expensive.

Risk pooling on the scale of CPP (75 year sustainability horizon) means that small course corrections can keep the plan sustainable. Smaller, less well diversified plans are far likelier to fail or require big adjustments (in contributions or benefits) to remain solvent.

TD actually offers better index MF solutions than ING, called e-series, for about half the MER of ING's streetwise funds. And of course, the lowest cost alternative is to use ETFs. A pooled pension manager with a few billion in AUM should be able to operate for <50 bps. CPPIB currently costs about 20 bps of AUM.

I worry less about political interference in CPP, given the thresholds for amendments, than I do insurance industry solvency. Assuris isn't even a full guarantee. Imagine if we increased the annuity market by orders of magnitude....

[Andrew F]

Stephen Gordon wrote a piece in Macleans about retirement income by income quintile. It's true that low-income individuals have high replacement rates, but that is virtue of the welfare system in place to increase their incomes. GIS and OAS should be much smaller programs, and intended primarily to help those who had low lifetime incomes, not to help those that had reasonable lifetime incomes but didn't save.

I'd be all in favour of increasing CPP replacement rates and using the foregone GIS and OAS expense to raise incomes of the working poor, particularly to offset the additional CPP contributions they make during their working years.

[Determinant]

OAS is not that big, nor that burdensome. It also works. Very few people game the system to have low retirement incomes to get OAS and GIS. Your post also goes into the "deserving" and "undeserving" poor, which I reject as a concept.

ING offers a good, off-the-shelf balanced (pension plan in a box) for 1% MER. The comparable e-series product runs at 1.2%. ING offers direct sales to individuals without additional costs, TD requires a brokerage account. Again, depends how sophisticated an investor you are.

Indexed life annuities add a load of 17% for the indexing, significant but not absurd. People are comfortable decreasing their income in retirement anyway and if you know it in advance and plan for it, it is manageable.

See here: http://retirehappy.ca/indexed-life-annuity/

Imagine if we increased the annuity market by orders of magnitude....

To match the size of the life insurance market then? The two products do form a longevity hedge pair so an increase is no bad thing for life insurers.

[u]

A mutual fund is not a pension in the sense of an indexed life annuity (what people commonly mean when they say pension). If you are concerned about fees (and if you aren't, you should be), then you should avoid balanced funds. ING is better than some of its peers, but its MF products are still expensive.

Annuities are already fairly expensive when compared to more traditional savings, and indexed life annuities are, as you say, 20% more expensive yet (for someone at age 65).

A substantial part of OAS is paid to middle class seniors, since the clawback doesn't even begin until $70k per year. I question the appropriateness of using so much revenue to supplement the incomes of seniors who are well above the median income in the country. OAS is a PAYGO pension--why not roll some of that into a partially pre-paid pension like CPP?

Annuities and life insurance are not perfect hedged, due to information assymetry. An individual who bought both whole life insurance and a life annuity might as well just burn their money.

[Bob Smith]

Its a pain in the butt to set up an account to hold the TD e-series funds (i.e., you have to go into the bank and talk with someone), but once the account is set up they're painless - i use them for my kids RESPs. TD should do a better job of marketing them, since it would allow them to compete with BMOs etf products (most of the e-series funds have MERs in the 0.3-0.6 range - higher than etfs, but without the commission).

Andrew, no one worries about political interference... Until the politicians interfere. I doubt a decade ago many Greeks were worried about the security of their public pensions. They should have been.

[Determinant]

Annuities and life insurance are a good hedge for a life insurance company; I wasn't speaking of the individual.

CPP is a defined-benefit pension, so contributions are pooled and benefits reflect the benefit formula (which itself reflects years of contributions and income level, not level of contributions directly) so there is income redistribution within CPP, just not as much as with OAS. CPP and OAS were designed to work together as OAS preceded CPP.

The price of annuities reflects the guarantee of annuities; it is a true price for a given level of service. Cheaper products have less security. That's price-as-information, I haven't been that classical in a while on this blog.

[Andrew F]

So maybe a reasonable compromise to provide sufficient CPP benefits to replace a reasonable amount of income. Perhaps an increase from 25% to 50-60% current pensionable income.

For any supplemental employer retirement benefits, the employee has their choice of some combination of group RRSP, pooled retirement investment (like say, the Saskatchewan Pension Plan), or deferred annuity purchased from a private insurer. The employer is indifferent the which is chosen, and undertakes no risk. Essentially ban in-house DB plans or benefits.

I think this should apply equally to public and private sector employers. My concern is with the moral hazard involved with politicians and unions colluding to create large off-balance sheet liabilities. If all retirement benefits were reflected in current year cash costs, this would not be possible.

[Sean O]

It's not just the teachers in Ontario. Other public service unions in Ontario have done the same. Back in the good old Days of Rae, the unions worked actively to ditch Rae and the NDP. It never occurred to them that would get a government much less sympathetic to their views. Now they are doing the same with the Liberals. Maybe they could change their slogan to: "We're getting what we wish for."

[Bob Smith]

"So maybe a reasonable compromise to provide sufficient CPP benefits to replace a reasonable amount of income. Perhaps an increase from 25% to 50-60% current pensionable income"

Is that a reasonable compromise? It would entail a doubling (or more) of the current CPP contribution rates. And since CPP contributions are regressive in nature (i.e., the CEO makes the same contribution as the autoworker making $60K, and a proportionately smaller contribution than the recent grad making 35K), I suspect both voters and business (if you think they bear the incidence of the employer contribution - they probably don't, but they also probably don't realize that) might not be so keen on that compromise.

The problem with a more generous minimunm CPP arrangement is that it implies a measure of forced savings that, for many people, will be sub-optimal during large swaths of their lifetime. So, for example, the young family quite reasonably might save little (no) money for retirement because they're taking maternity leave, paying for child care or paying their mortgage. But they might make up for that lack of savings in their late 40's or 50's (when their incomes are, generally, higher, and their kids start coming off the books). Granted, that's a problem with the current CPP, but given the relatively low contributions (and, yes, benefits) it's less likely that the degree of forced savings are going to be sub-optimal. The problem with a one-size-fits-all pension proposal, is that one-size doesn't fit all.

[Bob Smith]

"Other public service unions in Ontario have done the same. Back in the good old Days of Rae, the unions worked actively to ditch Rae and the NDP. It never occurred to them that would get a government much less sympathetic to their views. Now they are doing the same with the Liberals."

Or more recently, think Toronto civil servants and David Miller. David Miller was probably the most pro-union mayor that CUPE could ever expect to have and they destroyed him (Miller, mind you, didn't help his cause by not standing up for his voters during the strike). For their trouble, Rob Ford was elected and promptly contracted out half of the city's garbage collection, reduced the public payroll and forced them to agree to the most favourable contract (for the city) in decades (Ford may be a bumbling idiot on most files, but his (or his handlers') dealings with CUPE last year was probably the most successful labour negotiation Toronto has seen in decades. That may be damning with faint praise, though).

I was listening to Buzz Hargrove on the radio this morning, and while he blames the government for the current fight with the teachers (hey, he's a union man, I get that), he made it very clear that he thought the teachers were badly misplaying their hand by cancelling extra-curriculars and threatening walkouts. Then again, he probably learned his lesson from the labour movement's treatment of Bob Rae when he was in charge. They should listen to him.

[Determinant]

Is that a reasonable compromise? It would entail a doubling (or more) of the current CPP contribution rates. And since CPP contributions are regressive in nature (i.e., the CEO makes the same contribution as the autoworker making $60K, and a proportionately smaller contribution than the recent grad making 35K), I suspect both voters and business (if you think they bear the incidence of the employer contribution - they probably don't, but they also probably don't realize that) might not be so keen on that compromise.

It depends on how you define regressive. The CEO is only deducted the maximum on his YMPE (51,500 next year). People making under 51,500 pay 4.5% of their income (9.9% in total with the employer contribution). The CEO only gets 25% of $51,500, $12,875. The rest is his own responsibility.

CPP was never intended to be an pension by itself. It was introduced when DB pensions were expanding in the private sector. Now those DB plans have turned out to have their own serious problems.

I actually endorse Andrew F's proposal. Defined benefits should not be offered unless guaranteed by an insurer.

[Andrew F]

"Is that a reasonable compromise? It would entail a doubling (or more) of the current CPP contribution rates."

Because CPP rates were increased to make up past underfunding, you would actually only need to increase contributions by ~60% to double benefits (over a phase in period). Thus employee+employer contributions would rise from 9.9% to 16% of pensionable income.

The purpose of CPP, to me, is to force a certain minimum amount of savings on the middle class to prevent them from being a welfare burden in retirement. It addresses the moral hazard problem of having generous income supports for low-income seniors. Those with low working-years income will still receive GIS, which is defensible from an equity perspective.

CPP, being mandatory, can also offer better implied yields because it does not have the problem of adverse selection (those with lower than average expected life expectancy not purchasing the annuity). With annuities, insurance companies must assume that there is some adverse selection and offer commensurately lower implied yields. And having a higher stream of 'strongly' guaranteed and inflation protected income allows savers to invest the remainder of their retirement savings more aggressively.

[Bob Smith]

"Defined benefits should not be offered unless guaranteed by an insurer."

I wouldn't go that far, people should be able to contract for whatever sort of relationships they want. But they have to understand what they're (implicitly) paying for and what the risks are. People can make risky investments if they want - I could pull all my RRSP savings into penny stocks if I were so inclined (I'm not) - but they have to understand that a employer sponsored DB plan is not riskless, and unfortunately, in the past they have been presented that way.

[Andrew F]

The problem is that it is essentially riskless when your employer is the government. This creates a large asymmetry between the public and private sectors.

Running a DB plan means running a life insurance business. We don't let just anyone do that--why should we allow private employers to?

[Determinant]

"Defined benefits should not be offered unless guaranteed by an insurer."

I wouldn't go that far, people should be able to contract for whatever sort of relationships they want. But they have to understand what they're (implicitly) paying for and what the risks are. People can make risky investments if they want - I could pull all my RRSP savings into penny stocks if I were so inclined (I'm not) - but they have to understand that a employer sponsored DB plan is not riskless, and unfortunately, in the past they have been presented that way.

Most people can't understand that DB carries its own risks and trying to fight against three generations of belief is nigh impossible. Better to clarify the guarantee or not make it at all. There is too much information asymmetry. Tying the guarantee to an insurance guarantee makes the cost of the DB explicit. If it is too much for an employer to pay, at least it starts the conversation.

We spent 30 years after 1945 treating private-sector firms like institutions that were immortal. That turned out to be a fallacy and we are living with the consequences.

Andrew F:

The fact that the government doesn't go bankrupt like normal businesses is simply a fact of life. But it does lay off its staff, so the risk of not accumulating enough service to get a substantial pension is real for the individual. Different form of risk, but still a substantial risk. The fact that government plans are final-salary formulas increases this risk for the individual.

Pensions essentially started as benefits paid from current revenues, and only moved into a secured-by-assets model in the 1960's. GM initiated the change in the US in the 1950's. It was supposed to increase the security of the pension. Pension Benefits Acts in Canada mandated the move to that model plus evaluation of the plan every three years by an actuary, PBA's date from the 1970's.

Pension contributions were deductible from corporate income tax and exempt from personal income tax and payroll levies on the date of contribution. In the era of 50% CIT rates in the early 1970's made pensions an economical form of compensation in that the pension contribution was paid through avoided taxes, in the same way that RRSP contributions are paid through marginally avoided taxes. The decline in corporate income tax rates has shifted the burden of the pension off the avoided taxes and onto corporate income and logically most companies have protected their income by ending their DB pension arrangements.

[Bob Smith]

I'm not sure why the fact that government pensions are relatively riskless is a reason for getting rid of them. It is, however, a good argument for pricing them (implicitly in wage negotiations) in a way that reflects that low risk - i.e. If your compensation scheme includes a (relatively) riskless DB plan, all else being equal, your other compensation should be lower (compared to a person offered a similar private sector DB plan).

Mind you, the problem may be, a determinant suggests, that if people think all DB plans are riskless, there's no reason for the government to, implicitly, charge a premium for a public DB plan.

I'm not sure that taxes are necessarily the driving factor behind the decline of the DB (after all, that doesn't explain why firm shift a DB plan to a DC plan). I think the bigger factors are life expectancy (pensioners are living longer than expected) and increased competition as a result of globalization. Coupled with fairly steady decline in interest rates over the past 20 years, you can't blame companies for wanting out.

[u]

Government DB plans being relatively riskless is a great way for the public sector unions to extract rents, especially given that politicians are all too willing to underprice them in exchange for labour peace. Allowing DB plans in the public sector makes the temptation too great.

It seems to me that all citizens should benefit from government-backed DB plans equally. And our government should not undertake such vast risks as it currently is with underpriced final salary DBs.

DB plans should be limited because they cause financial instability in both public and private sector employers who offer them.

[Bob Smith]

"Government DB plans being relatively riskless is a great way for the public sector unions to extract rents, especially given that politicians are all too willing to underprice them in exchange for labour peace."

I agree, although that strikes me as an argument for better government, and for expecting our governments to take a hard line in negotiating with their employees, rather than doing away with government DB plans (which, after all, if priced properly may be a way to reduce overall public sector compensaion, since governments can offer a benefit that one else can). Taking government DB plans out of the compensation mix doesn't eliminate the incentive for governments to overpay (i.e.,underpricing the DB plan) in exchange for labour peace, it just eliminates one means of doing so.

I'm not sure DB plans neccesarily cause financial instability. If poorly managed or improperly hedged they can cause a company to go bankrupt, but the same is true of lots of things that businesses do. The solution is to ensure that they're properly managed and hedged, not to ditch them entirely as a tool. Of course, a lot of companies have decided that they can't do that, or at least not at a cost that's feasible, and that's fine, but DB plans should be an option, for parties who want to go down that road.

[u]

DB plans are an option: the employer can buy an annuity for its employees from an insurance company.

[Matawan Tax Return preparer]

This issue came up off-line, so I figured I should post about it. Some people in the oil industry claim that the deduction for intangible drilling costs is not a subsidy but an allowance for real expenses. And they are correct! Which is why nobody proposes to eliminate it.