« Instrumentalism | Main | Plastic and glass »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Interesting post, Stephen. But I hope you're not using the Low Income Cutoff Rate as your "Statistics Canada's poverty threshold for Quebec." That's a real pet peeve of mine — LICO doesn't measure poverty.

"Statistics Canada has clearly and consistently emphasized … that the LICOs are quite different from measures of poverty. They reflect a consistent and well-defined methodology that identifies those who are substantially worse off than the average."

One effect that hasn't been mentioned here, but I think is important is the REGIONAL effect of introducing a Citizen's Income. Citizen's income amounts to both a wage subsidy and an injection of purchasing power into poor areas. I reckon that makes its regional effects likely to be startling. Does anybody know of any evidence on this?

doesn't the conclusion high taxes are necessary place more importance on the labour supply issue? even if the incentives for getting work are more favorable, how many more people would choose to stay at home under CBI? Trying to raise more money to fund payments to a large non-working population has potentially vicious circle dynamics, if raising taxes causes the number choosing not to work to rise.

I'm pretty sure that if I could live reasonably comfortably without working, I'd choose to do that (I have lots of hobbies). My feeling is that very many other people would too. I don't choose that now, partly because UK unemployment benefits (job seeker allowances) are low, partly because they try to make you accept work/training or they'll withdraw payments, but mostly because of housing and savings. I have some savings, I have a mortgage, if I quit work I'd be ineligible for housing benefits. Talking of housing, when I see CBI debated in the UK, it's usually housing that comes up as the main problem. If CBI pays for housing costs in London, it would be insanely generous. If it doesn't pay for housing, then the state would still have to house the unemployed and they'd still face marginal benefits withdrawal when finding work.

if you're interested in UK writers on this, google stumbling mumbling and citizens basic income, he's written a fair bit on it.

Luis Enrique - that is why he thought raising VAT would be the tax change chosen.


These amounts are far from lavish, but the costs are surprisingly high.

The costs may not be as high. Over at the Volunteer.ca, Mike Brock brought up an interesting point about how a GAI administered by Revenue Canada would lead to a reduction in government bureaucracy (ie. welfare, unemployment, old age security, etc. would no longer be needed). Those savings would reduce the costs considerably.

One concept a basic income would tie in very nicly is an offset flat tax. The idea there is that you are charged the same marginal rate on all income (from the 1st dollar on), but you are given a basic income as an offset, based on family size and possibly cost of living where you live. All of our current walfare, OAS, and measn tested benifits would be rolled into the basic income.
The key point of this is that at any income level there is always a consistant and releativly low marginal imapct of eraning your next dollar, so there never is a crushing dis-insetive to improve your income, while also significantly simplifiing the tax system.
The average tax rate would still be higly progressive, and with the correct values of rate and basic income woudl closly mimic the current tax structure.

As an example say the rate is 25%, and the basic income is $10,000. If you had 0 market income, you would have a net income of $10,000. If you had a marget income of $10,000, you would pay $2500 in tax, and recieve $10,000 back, resulting in $17,500. At $40,000 you would break even, at $100,000 you would pay $15,000 net in taxes, etc.
These numbers are just examples, i suspect the actual rate would be near 40% if you included provincial income tax, but the average tax rate would be much lower than that for most people.

I wonder what effect GAI might have on things like crime and health. Presumably it would improve outcomes and thus save the government money. Is there any work estimating what the effects might be?

If we are going to institute fair/efficient program like a like a basic income we might as well start off by funding it from fair/efficient taxes rather than raising the economically harmful ones we've already got.

A carbon tax of $200/tonne of carbon (about $56/tonne CO2 or $0.13/l gas) would raise about $31Bn/year on 550 million tonnes of annual CO2 emissions. That's about $1000/Canadian, including the 25% of the population that's children, so $1333/Canadian if you pay them nothing.  It's a start. We don't have to do it all at once.  There are other highly efficient possible sources of revenues.  Other ecotaxes like a land value tax come to mind.  To the extent that these taxes are welfare improving or, in the case of the land value tax, in principle 100% efficient, we'll get a big benefit from the overall scheme.

Finally, and I'll try to stop repeating this (sorry), if this is ever going to happen, labeling it as a citizen's dividend with clear connotations of rents from public resources, would be far more politically palatable (in my opinion) than guaranteed or basic income which have welfare/charity connotations.

Jesse: There are serious political economy issues with incorporating a flat tax. Numerous studies have shown that the scheme that you are proposing is good for the poor and good for the rich, but not good for the middle class (the largest block of voters).

Stephen, all this focus on finding the revenue through higher taxes. Couldn't the GMI scheme be funded through spending reductions due to the elimination of less efficient income redistribution schemes and social programs? (I know, I know, I mentioned political economy concerns about a flat tax and that probably applies to program cuts too).

There would be some savings, certainly, but it seems to me as though they would be small compared to the extra revenues that would be required.

Going to how you pay for it, that is where the flat marginal tax helps.

For the levels beign contimplated (12k per family) would be about 5k per person, or 150B for the country. That would work out to be something like 10% income tax rate, just to cover the basic income. A good chunk would come from eliminating a lot of programs that this would replace, but for the middle and upper class the extra rasied by the flat marginal rate, over the progressive rates now, would cover the rebate for the same people.
(the government would have more money going through it's hands, but a good chunk would go straight back to the same people it came from). Part of the advantage there is simplicity in that no means testing is required, and tax can be deducted at each source of income, without having to know what other income someone has to determine the rate. Theoretically everyone with normal employment earnings, and investments through full service bank/bbrokerage would have an exactly 0 tax balance for their income tax return, possibly only adjsuted for chartiable recipts.

JDUB: based on the relativly crude decile and 5% income stats available form stats canada I found that you can pick a combination of rebate and rate that closly matches the current distribution of earned income vs. after tax income. If I remeber correctly (don't have it in front of me), it was basically $10-15k per family, 33% rate). Obviosly this would have to be worked through closely, and we would probably need to remove the lower capital gains rates to make sure we are not lowering the burden on the wealthy too much, at the expense of the middle class. Obsiosly this system would be good for the very low income, as that was the whole point at the start of the discussion

Stephen: I really don't get it. Why does it cost anything?

Suppose you assume zero behavioural effects. And zero administrative savings.

Set the basic income at the same level as welfare. Then take Andree-Anne's graph above, and smooth it out. Shave off the hills, and use the excess earth to fill in the valleys. Set the same (flat) marginal tax rate everywhere. My eyeballs say 60% marginal tax rate ought to roughly do it. The budget should still balance before and after. (Actually, you would really need to redraw the graph so the x axis measure income *percentiles*, rather than income, for this to work exactly.)

Any behavioural effects and administrative savings are then gravy, that we can use to either increase the basic income above welfare and/or reduce the marginal tax rate below 60%.

The Basic Income is just welfare with a smooth(er) structure of marginal tax rates and a simplified administrative structure.

Thanks for doing some of the legwork on this Stephen.

I'd second some of the earlier comments in wondering what the impact of reduced spending on administration would be from a guaranteed income scheme (I'd guess pretty small, but enough to be worth taking into consideration.)

Personally, I think the most practical approach is an incremental one. Gradual increases to HST credits, with offsetting reductions in welfare payments will move us in the right direction. Maybe getting the marginal effective tax rate down to 60% for the first $15,000 of income would be a good intermediate goal (as Nick points out, this shouldn't involve too much extra cost over the current approach. Long term, I think somewhere around 40-50% is probably the ideal range, based on the tradeoffs you identify in your post between cost of program and incentive to work.

Personally, I am less concerned about the efficiency loss from the high marginal rates, as I am about the dehumanizing aspect of a system where people must jump through a hundred hoops to prove they are unable to find a job and the inefficiency of maintaining a bureaucracy dedicated to this cat and mouse game.

With respect to Luis' comment above about living comfortably without working, I (and I think I'd be in the majority on this one) don't support a guaranteed income (or welfare payments) high enough to make people able to live comfortably. With dignity, yes, but comfortably, no.

I think a basic income would also boost innovation, by giving someone who wanted to start a business or become an artist something to live on while getting started. Plus, I would like to emphasize the point made earlier about reducing bureaucracy.

I'm rather amazed at the lack of negative comments. Any time I have suggested a basic income, I always get smacked down pretty hard. Taxes Taxes Taxes.

Nick: Your suggestion only works if the distribution of incomes across the graph were uniform. They are not. You are on the right track when you note that you would need percentiles, rather than dollars to do your 'smoothing out' trick.

Here is why it would be more expensive, even without behavioural effects.

If you had a big enough transfer to leave those currently on SA with incomes unchanged, you would not save any money. Then, if you are going to have a lowish clawback rate, you are going to have to extend partial benefits well up the income distribution--at parts of the income distribution that are particularly thick. Many people not currently getting cheques would start getting them. That's where the cost comes from. Perhaps worth doing, but expensive.

If the GAI comes with a 25% clawback rate, it seems like we'd be getting something like flat tax anyways, since 25% is somewhere around the difference between the top and bottom marginal tax rates.  Depending on where the clawback ends, the rate might be slightly lower in the middle...

But I think that conception as an income "guarantee" puts too much emphasis on income tax as a way to recuperate the costs.  Like nick says, the marginal rate is likely around 60% if it's going to be revenue neutral.  Much better to just pay a smaller but non-taxable income to everyone, and then make a separate decision about how to fund it.  For example, create a flat tax around 50-55%: whatever level insures that nobody will be worse off in terms of GAI+income tax alone. That way the incentive to participate in the workforce will be significantly increased (rather than decreased) for a lot of people.  If the social benefits are supposed to include increased independence for all citizens, then let's put more emphasis on consumption and other taxes (see e.g. my comment above) which will actually encourage productivity, savings and investment.

A potential great benefit of a fixed universal annual payment might be if it helps to compartmentalize the social justice decision so that we can make tax decisions based principally on efficiency considerations.

Patrick: "I... don't support a guaranteed income... high enough to make people able to live comfortably"

That's because you think of it as charity.  What if the oil spouted out the ground like it was Saudi Arabia?  Personally I say we split the spoils and live like princes.  In addition to our hard labour, the land yields a certain amount of natural wealth.  I don't see why you or I should get more of it than anyone else.

Patrick: "I... don't support a guaranteed income... high enough to make people able to live comfortably"

That's because you think of it as charity. What if the oil spouted out the ground like it was Saudi Arabia? Personally I say we split the spoils and live like princes. In addition to our hard labour, the land yields a certain amount of natural wealth. I don't see why you or I should get more of it than anyone else.

Tragedy of the comments.

The tragedy is that I blamed Patrick when it was Declan who said it.  Sorry, Patrick.  Apart from that I am not sure I understand what you mean, JVFM.

K: Sorry, it was an irreverent drive by. Something about producing oil like there's no tomorrow.

The graph makes a rather good case for tax simplification. Why not abolish all these dedicated levies and fund them out of the Consolidated Revenue Fund?

I just added an update in order to quote Kevin Milligan's neat twitter post from earlier today.

Kevin: we agree on the details. It's just I find it hard to think of smoothing out the marginal tax rates as "expensive". *High* marginal tax rates are expensive, because they cost efficiency. *Smooth* marginal tax rates don't cost anything, if you leave the average height the same. Plus, if you add in the behavioural response, under the assumption that the deadweight cost function is quadratic in marginal tax rates, there's a negative cost to smoothing.

Kevin: " Many people not currently getting cheques would start getting them."

But Stephen's graph is the effective marginal tax rate.  I.e. you should think of it as everyone gets welfare and every other benefit and then have it taxed away again depending on their marginal income.  So from that perspective there's no additional cheques; everybody already gets everything.  If that graph was the real marginal tax structure then, all other things being equal, we could fund the payment of all current benefits to all of us.  (At least, that's what I understood Nick to be saying.) So the important things are 1) to figure out what the value of those benefits are (that's the value of basic income that could be paid) and 2) If desired create a saner marginal tax curve.  This could be done, all other things being equal, by just smoothing out the effective marginal curve over the income percentiles, as suggested by Nick.

The concept generally appeals to my prejudices, but my concern: Guaranteed Basic Income can only be business cased as cash-neutral if you eliminate all other forms of social subsidy and the administrative overhead that comes with them...


There will always be in society a tiny number of people (substance or behaviour dependents, mentally ill, intellectually marginal, etc) that simply cannot be relied upon to look after themselves with any amount of money. Indeed, resources of $20K a year could conceivably make them considerably worse off. Do we completely wash our hands of these people and permit nature to take its course?

No, of course not. But this means that you cannot eliminate the bureaucratic overhead associated with existing social programs, or even the programs themselves.

So how do you make a cash-neutral business case for the Guaranteed Basic Income?

I meant:"everyone gets welfare and every other *currently income dependent* benefit". And these are no longer means tested. And the new marginal tax rates are no higher than the old effective ones. So as Nick said, how is it more expensive than before?

I often think about this exact issue. Let's say we set a BI. Maybe $1000 per month for every citizen, including children. What government expenditures could we CUT (completely or partially)? I don't seen any sensible excuse to maintain agricultural subsidies. I think we should expect pretty much every part of government to take a major hit in the reshuffle. I think it would be fair to cut back many social services dramatically. There would still be many individuals after whom the government would need to watch. The most important would be children. Each child's BI should go to someone who can be counted on to use that money in the child's interest. That would require bureaucracy, but I think those departments would be just as valuable without BI as they would be with BI. Similarly, cases in which an adult can not be counted on to spend the money wisely should be considered, although I think social services could still be trimmed massively. I think it might also be possible that a BI could be structured in ways that would work especially well for people who have problems handling money. BI funds could go directly toward their services, which could put government funds into the places where they would do the most good. Direct support for the poor might also increase the political feasibility of other methods for raising revenue. Pigouvian taxes are one such method. I'm sure there are others.


Re Children - you could also allow start charging for public schools with the money you give families for BI. No more tax deductions - effectively a voucher system.

The change really would be quite dramatic (and I think positive).

As for people with special needs (handicapped, psychologically ill etc) - I agree you need a burocracy for them - but that is positive - instead of employing clerks you employ social workers - to do proper social work. And in such case they should be empowered to control some of the money and replace it with in kind support.

P.S. Dropping the minimum wage could also be part of the deal.
P.P.S. By the way, I'm a bit disappointed that nobody has picked up on the regional implications. You are not only redistributing from rich to the poor people, but from rich to poor regions. Low rent areas would receive a face lift in my view. It would work a bit like the industrial/military/prison complex works in the US in supporting remote areas (i.e the GOP welfare system).

Luis Enrique,
re hobbies etc,

I think you will find that eventually you get pulled back into the work world - at least on a part time basis. In fact, I suspect there would be a boom in part-time employment - and that is not a bad thing.

What do ex-hippies do these day? I think many run there own businesses. (They make their hobbies into businesses in many cases).

Luis Enrique,
re housing support - I think the UK system here is unusual. Other countries of course have mortgage and rent systems that allow some relief in cases of unemployment - but often at the expense of the landlord or bank.

And of course the correct answer is to phase in replacement of the housing support system with private (or private with a public option) insurance.

I'd like to propose something a bit more radical: institute a guaranteed minimum income (something on the order of $5,000 basic rate per household + an additional $12,000 per person), abolish all government transfers & subsidies, as well as personnal and corporate income taxes + pension & UI payments, and just have a VAT that would be universal (no exceptions). Unfortunately a quick back of the envelope calculation shows you would need a 100% tax rate (or 50% if you consider prices after the VAT is applied). Politically, this would never fly. However this system would be much simpler, more just, respect privacy, encourage savings and eliminate a lot of perverse misincentives.

Alex: sounds OK to me (i.e. your BOE calculation sounds about right, and the policy proposal sounds good)!

the problem with any change from our current mess of programs to something new is the fact that there will be som ewinners and losers, assuming we do not increase expend the total amount of money spent and that administrative savings will not be large enough to prevent some people from getting less.

and then how do you transition with regard to programs that people have paid into - such as CPP, EI, workmen's comp, etc., or would these continue to be on top of a minium income?

my own solution to poverty is simpler - raise the minium wage to $15 across the country, and cut immigration in half. low unemployment and high wages are the best way to get people into the workforce. forget programs like public housing that are only avaiable to a few lucky beneficiaries, and lack "horizontal equity". put more money into apprenticeships and training.

and for example, i don't see how a AGI will deal with aboriginal poverty at all - the idea of an AGI is a bit simplistic, and it also fails to account for the higher cost of living in cities.

reason - yes I see all sorts of positives from this idea. More happy people working part time being one of them. Just wish I believed it was feasible.

so housing isn't generally seen as the deal breaker, in other countries, as it is in the UK. I know, reason, you like the regional distribution arguments, but you may also see 'social cleansing' of rich neighborhoods, like everybody is (rightly?) complaining about with current plans to cap housing benefits. i.e if you have to rely on CBI, you have to move out of South East UK.

ah - but you see I think we need more worthwhile places to live, and lower housing costs overall. I think there is a vicious circle making some places very expensive, and some places very cheap and this will break out of it. But that is another story.

Alex Plante,
add environmental (and other taxes on negative externalities) and inheritance taxes and I would sign up.

"*High* marginal tax rates are expensive, because they cost efficiency."

Hi Nick. In the optimal income tax rate formula I have in front of me, we multiply the cost you mention above by the density of the income distribution at that point.

reason: "we need more worthwhile places to live, and lower housing costs overall."

Exactly.  Housing subsidies are real estate market subsidies.  As usual, just give people the money and let them decide where they want to live and how they want to spend it.  What we need is a land value tax (the opposite of a subsidy).  Apart from being efficient, it would also counteract real estate bubbles.

And good idea on the inheritance tax.  Big inheritances must be one of the greatest wastes of resources known to our economy.

Personally, I would start with land value tax, then inheritance and then VAT for whatever is left.  VAT is demonstrably less efficient than the other two, especially land.  And a big VAT will drive large sectors of the economy underground.

Good grief. The Flat Taxers and VAT-in-lieu of Income Tax people are out in force. It's Amateur Hour here at WCI. :(

First, there are lots of worthwhile expenses that are not geared to income, like health care. Therefore, we will always need more revenue than a flax tax will provide. We plain out need income proportional redistribution.

Second, a VAT suffers the same flaw as an excise tax or import duty. It's a voluntary tax. In fact, its promoters sell this as a feature. Every argument they make implies that a person can be thrifty enough to minimize their rate of personal tax. They don't want to pay. A VAT will allow them not to. That's why they like it.

The problem is on a macro level is that the government needs a certain level of revenue to pay for our desired programs. Consumption is more variable than income. That's why in a monetary economy we need, not want, need an income tax. An income tax, an involuntary tax, is ultimately what is needed to pay for programs and services.

Thirdly, as have said in the past and will say again, abolishing OAS will not save one penny. OAS is paid for out of the Consolidated Revenue Fund, that is current tax revenue. Canada is not the United States where Social Security does both preretirement income-proportional replacement and bare-minimum social welfare. In Canada we explicitly split those jobs between OAS and CPP.

Consumption is more variable than income

This is simply wrong; look at the data. Most of the rest is as well, come to that.

I find it delightful that even in a 'room' full of economists, we can't agree fully on our pros and cons...how on earth do we expect the rest of the world to? The last two 'basic income' proposals I've seen are actually means-tested: last week's federal HUMA Committee report (p. 141-143) focusing on disabled people and the general one proposed by Sen. Hugh Segal, so even the basic concept of lowering METRs hasn't cleared the academic-political hurdle. Political hurdles aside, I suspect there are efficiency-equity wins out there, even after taking into account the need for additional revenue (I'm on board with you here, Kevin). It would be nice to see a half decent comparison of a real BI program with labour supply effects included, as suggested by Stephen.

A quick note is that the CIRPÉE paper is from 2006, so doesn't include the WITB. When you do that, the marginal rate drops 25% (25 points) between $3,600 and $7,250 and then rises 20% between about $16,000 and $21,000 (the graphic is four a family of four, two adults, two kids).

Further responses, since I'm a day late getting here:

Geoff, bingo. There still need to be social supports, though they can be for fewer people and, as reason points out, qualitatively different (hopefully more effective). Still, best not to overestimate the savings on this front.

Declan, I love your comment: "With dignity, yes, but comfortably, no." Each of us has our own definition of dignity and comfort and agreement on what is which is one of the hardest parts of setting transfer payment levels (effectively, you're defining the social welfare function). A basic income is high on my list because I have a great utility for other people's dignity and, qualitatively, I agree that this is a massive improvement on that front. I also think it holds its own efficiency-wise, but we could certainly use some more data.

David, the LICO for a single person is around $20,000 - I think he's using the Market Basket Measure here (which is regionally based and comes in around $12,500-$15,000).

The links got rejected from my previous post...let's see if I can put them up here:

HUMA report - http://www2.parl.gc.ca/Content/HOC/Committee/403/HUMA/Reports/RP4770921/403_HUMA_Rpt07_PDF/403_HUMA_Rpt07-e.pdf

Sen. Segal describing his GAI plan - http://www.hughsegal.ca/index.php?option=com_content&view=article&id=33%3Athe-real-recovery-challenge-basic-income-security&catid=20%3A2009&Itemid=28&lang=en

Yeah, we need redistribution, and that means succession tax, property tax, and progressive income tax. VAT and Carbon just won't do the job.

Regional impact would be generally good for the consumption levels of those outside major centres.

Unfortunately a lot of Canada's poverty happens in our main urban centres--partly because of high housing costs.

A program which tries to account for regional differential housing costs would have very perverse effects-- what could be more perverse than subsidizing urban real estate?

But programs which distribute income to smaller centres, and to the hinterlands, would tend to move workers out of the places which have been designated the main foci of our "new economy," which has always been explicitly, indeed exultantly, metropolitan.

Would major capital investments necessarily follow a BI-subsidized outflow of low-end workers to the hinterland cities and towns?

As a long-suffering migrant from BC's hinterland, I would love to stick it to urban taxpayers and boost incomes in the Interior. But if we're pitting a BI policy against our investments in higher education, transport, R & D, etc. then we'd be raising taxes only to work at cross-purposes.

GAI is a broad stroke. Many unemployed people would derive mental health (socializing like forced in school) and fitness benefits from PT work. Might be significant enough to clawback from an optimal GAI, a concurrent negative income tax if this health benefit can be captured. But coal miners get dead, truckers sleep deprived under present arrangements (don't earn danger pay from cargo consumers)...is broad.
I just can't see why the Right is so hostile to the idea of broadening the number of people have the opportunity to volunteer, career-build, etc; why do the GOP and our PCs, Cons, Quebec Right; why so hostile to a Northern European level of quality-of-living minus the pissed immigrants?
I wish I lived in Ukraine. Housing is free there. Instead some idiot in a Cgy suburb gets an extra few hours with a hooker. Sickening the lack of philosophical duality.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad