Time to Think about a UBI?
The CERB program has its flaws: There are good argument that it isn't enough and it's underinclusive, and I've written before about how it strips out some of the best features of the well-established EI program (which works, even though it has its own problems).
Problems are understandable. This is a complex policy initiative rolled out in a rush, to deal with a crisis. Perfection isn't the standard here, and everyone's largely agreed that something like CERB was necessary to address the pandemic.
Andrew Scheer has raised another concern about the implementation of CERB: The $1000 income cutoff creates a problem for people phasing back into the workplace.
CERB eligibility has some pretty rigid criteria, and among them is a requirement that the individual not make more than $1000 in the 4 week eligibility period (there's a different treatment for the initial claim period, though). For the sake of simplicity, let's call this $250/week. If you're a hair under the cutoff and otherwise eligible, you get the full $500 per week from CERB. If you're a dollar over the cutoff, you get zero.
This is a harsh result, presumably for the sake of streamlining a system facing millions of claims. Scheer is right to say that it creates a difficult situation for individuals looking to return to work, where an individual might get significantly less money because she took all the shifts she could get.
Scheer is taking a lot of heat for his articulation of a 'gradual type of reduction' of benefits, as if he wants to reduce CERB incrementally for everyone across the board to force people back to work. My read of his comments is that this is an unfair characterization, and what he's actually proposing is a progressive clawback, similar to what we do with programs like OAS: If you make over a certain threshold, then you don't immediately lose your entire benefit, but instead there's a formula. For every dollar you make over the threshold, you lose a portion of that dollar in a benefit clawback.
So, conceptually, Scheer seems to be proposing that, if you make $1200 or $1300 in the 4 week eligibility period, you only lose a portion of the CERB benefit. He doesn't propose a whole lot of detail, but he does provide an overarching paradigm that "it's always better off for Canadians to work". (Let's call this the "Scheer Paradigm".)
These clawbacks are good policy, but in this particular context, even aside from dramatically increasing the complexity of the CERB system, it creates other challenges.
The biggest challenge surrounds a particular group of people who are not eligible for CERB: Low-income essential services workers who did not lose their jobs. So, for instance, grocery store workers. Here in Alberta, somebody earning minimum wage (which is higher than much of the country) for a 37.5 hour work week grosses $562.50 per week. Not much more than CERB. And that's without considering that many such workers are part-time only.
It's hard to envision an iteration of Scheer's clawback where we're not giving residual CERB benefits to people making significantly more than ineligible, still-employed, front-line grocery store workers. It's especially hard to imagine such a system without dramatically reducing existing CERB benefits (recall what I said earlier that it should be one of the governing paradigms of crisis programs that "once a benefit is rolled out, it's guaranteed"). And Scheer has expressly clarified by Twitter that he's calling on the government to make the program "more generous"...so I don't think I'm taking him out of context if I suggest that his plan does not include a reduction of existing benefits.
If we use the threshold of $250 per week before the clawback begins (based on the current threshold of $1000 per four weeks for CERB), and we were to consider a dollar-for-dollar clawback, then your entire CERB benefit gets clawed back at $750/week in income. That's the floor. Remember, the Scheer Paradigm is that Canadians should be better off for working, which means a clawback of something less than dollar-for-dollar.
If we used a 50% clawback on income from $250-$500/week, and than a 75% clawback after that, then the full benefit isn't clawed back until you reach $1000/week in income.
Can we tell a still-employed minimum wage grocery store clerk earning less than $600 per week that he doesn't get any benefit at all, but turn around and give free cash to the person making $950 per week in employment income? Even aside from the poor optics in light of the pandemic, it still violates the Scheer Paradigm: Those grocery store employees are worse off for having kept their jobs.
The only rational solution here is to extend the CERB benefit beyond those who 'lost their jobs due to COVID-19', and to all individuals in the applicable income ranges.
In other words, to fix CERB in any way that doesn't violated the Scheer Paradigm, we need to turn it into a Universal Basic Income (or UBI).
A UBI is a mostly-theoretical social program - there have been a few pilots in different areas - that essentially means ensuring that absolutely everyone is getting enough money to survive, even if they aren't working.
There are different iterations. Some models are means-tested, where you only cut the cheque to people earning less than a fixed amount up front. Other models involve cutting the cheque to everyone, and then clawing or taxing back the money from those with sufficient other income. It's really a 'six of one, half a dozen of the other' kind of thing, where there are limited practical differences, and the main differences between the systems are administrative and political.
It's currently the progressive Holy Grail, the goal of which is essentially the eradication of extreme poverty.
Yes, yes it is. But not as much as you might think. The Parliamentary Budget Officer modelled a UBI a couple years ago based on an Ontario pilot. At a UBI level of about $17k per year for an individual and about $24k for a couple (which, to be fair, is less than CERB would work out to annually), the PBO estimated a cost of about $76 billion per year Federally. However, this would presumably subsume existing Federal benefits such as CPP and EI, and so the PBO concluded that the net cost would be about $43 billion.
Moreover, the PBO only netted the cost against existing Federal benefits, without considering the cost of Provincial benefit programs which would be rendered largely obsolete by a UBI - Ontario, for instance, spends about $10 billion per year on social assistance.
In addition, poverty amelioration costs bleed into a lot of other areas of public spending, such as education and criminal justice.
So while the logistics of implementing a permanent UBI might have their challenges, the net cost to the taxpayer is actually a lot less than you'd expect. In a country where the Federal budget is in the ballpark of a third of a trillion dollars, a new $43 billion outlay is big, but not obviously implausible.
How Would a UBI Impact the Economy?
This is the tougher question. An immediate answer you'll hear from most boomers and conservatives is "A lot of people will decide they don't have to work, and just take the handouts". There's certainly some truth to that in existing social assistance programs - though those of us who have worked with those communities know that it's not necessarily about laziness, but about other underlying disadvantages and disabilities.
The key here is the one highlighted by the Scheer Paradigm: A good UBI is one that preserves the incentive to work.
But there's a flip side to the UBI that the conservatives tend to ignore: Stimulus. When you give money to low-income people, what do they usually do with that money? Why, they spend it, obviously. And with a permanent adequate safety net, the impact on consumer confidence would be very positive. A team at the Berkeley Institute ran models that showed substantial economic growth driven by UBIs.
Still, an honest appraisal of a UBI has to acknowledge that there's still much that we don't know about the macroeconomic impacts of such a program - and small-scale pilot projects are unlikely to shed much light on such issues.
The National Bureau of Economic Research released an excellent (and well-balanced) analysis of the outstanding questions and concerns about a UBI. The authors, Hoynes and Rothstein, suggest that there may be a reduction in labour supply in the short term, but that this isn't necessarily a bad thing. Perhaps more importantly, in the long term, such a support encourages higher levels of education and skill development, and improves outcomes in early childhood development, and would therefore promote a more skilled and productive workforce in the long run.
I've historically been sceptical about the concept of a UBI - I regard it as interesting, with the potential to do a lot of good, and yet the cost and overall impact seem to have their risks.
But the thing that makes me increasingly more attracted to it is the trend toward automation of labour. Traditional economic drivers like manufacturing and resource extraction have significantly moved to be less labour-intensive.
Labour obsolescence is nothing new, and there are those who believe that what we are seeing is nothing new: Historically, technological development has wiped out entire fields, and yet simultaneously created new fields and freed up the labour to pursue them. (Consider switchboard operators, for example. Automation replaced them a long time ago...but automated switchboards created efficiencies that promoted economic growth, and created new jobs in other areas.) Or in other scenarios it has improved efficiency to the point where it actually caused businesses to hire more employees - like Henry Ford's moving assembly line, which made the automobile producible at prices available to the mass market.
But I would posit that there is a threshold at which automation will eliminate more jobs than the corresponding efficiencies can possibly create via economic growth - primarily because the corresponding growth opportunities are also highly automated. It is not difficult to imagine a futuristic world where essentially all tasks associated with agriculture, manufacturing, construction, natural resources, transportation, retail, telecommunications, and other industries are fully automated. It is a bit harder to imagine full automation of certain professional services, yet even in fields like accounting and law, there are already moves toward automation with AI and machine-learning.
In such a future world, it is very difficult to imagine productive employment for all members of that society: In essence, we'll be at the point where are productive capacity is no longer capped by the cost and/or availability of labour, but by the availability of raw materials and/or our collective capacity to consume. So where do we cross the threshold from a world where efficiency-driven growth translates into a net increase in employment and wages, to a world where efficiency-driven growth translates into a net decrease in employment and wages?
I'm not sure, but that threshold certainly exists, and it's not totally unreasonable to think that we're approaching it, if we haven't already reached it.
The bottom line is that automation creates an oversupply of labour, cheapening it. That is, after all, the point. And yet, at the same time, businesses rely on a prosperous middle class: If our economy, on the whole, successfully reduces its labour cost, then that generates a corresponding reduction in the ability of the economy at large to consume.
This is what I call the macroeconomic prisoner's dilemma: If I own a business, then my best conceivable outcome occurs if I incur little-to-no labour cost, but if everyone else pays their labour enough money to be able to purchase my product. But if other businesses cut back on their labour in the same way as me, and don't pay workers enough money to purchase my product, then my sales will struggle despite the fact that I can produce more inventory while paying virtually nothing for labour.
A UBI addresses this in multiple ways: Firstly, as Hoynes and Rothstein note, it likely reduces the labour pool, offsetting some of the broader economic implications of automation. Secondly, it provides relief and a safety net to those whose jobs have been eliminated, or those who fear elimination of their jobs. Thirdly, it facilitates workers seeking retraining when faced with obsolescence. And fourthly, it provides unemployed and underemployed individuals with cash in hand to be able to purchase from the private sector.
I've thought for a while that a UBI is likely something that will become necessary at some point in the future. Andrew Scheer, however, has brought me around to thinking that it's probably the right way out of our current mess.
The CERB itself is a program on a clock; it's only available for a fixed time. As I've said, I think think we can pare back any benefits already offered, but once the current program has run its course, all reforms are on the table.
For the moment, however, there are certain constraints to be observed: Everyone already eligible to receive the full benefit...gets the full benefit. Clawbacks should be based on income over $1000 per 4-week eligibility period. No special treatments or reductions based on spousal income.
So the first question is "who gets it"? The obvious answer is 'everyone', but that's going to come with some question-marks: Are we just talking about people who are working or were working in the recent past? Or are we including people who aren't labour force participants? Stay-at-home parents? Retirees? Persons with disabilities?
I don't see a compelling reason for any such exclusion classes. How the benefit interacts with other types of income and benefits is a complex question, but we've developed a similar program for students (which should be rolled into this one), and I'm not too concerned about other categories. When dealing with crisis programming, and particularly when dealing with changing programs that have already been rolled out, it's better to be overinclusive than underinclusive.
The second question is what the clawbacks should be. I think the numbers canvassed about - no clawback up to $1000 of income (over the four week period), claw back 50% of income from $1000 to $2000, and 75% of income from $2000 to $4000.
As Scheer says, we want people to want to work. Too high of a clawback removes the incentive.
Once the CERB runs its course, converting it into a permanent system requires a lot of structural changes to other Federal and Provincial programs, but first let's start with the ways that the payout itself would differ.
The numbers would, of course, have to be revisited. I'm not sure that $500 per week is a terrible starting point, generally speaking (on the one hand, it's substantially higher than the program previously costed by the PBO; on the other hand, it's still a poverty-level income), but there is a certain logic that a couple residing together should get less than two individuals living on their own, and to means-test based on family income. (Also, parents should get something toward the kids...but there's already a framework in place for this.)
More importantly, I think the clawback threshold would need to be revisited. I'm reluctant to suggest reducing the threshold under $1000 under the already-implemented emergency program, because that pulls back money that people are legitimately counting on. But as far as an expansion of the project goes...use a more gradual clawback, and start it at a lower level. One of the problems with too steep of a clawback is that it does create a disincentive to productive labour.
I think it would be perfectly reasonable to have a 20%-33% clawback starting from dollar one - so, playing with $500/week benefit and a 33% clawback for the sake of argument - somebody earning $100 per week in employment income would only get $467 of their benefit. To the extent that we want people earning toward the higher end of the resulting range (up to $1500 per week) to be getting less of a benefit, we do that by playing with marginal tax rates.
Let's look at it this way: In Alberta, somebody making $1000 per week is in a ~30% marginal tax bracket - meaning that for every additional dollar of income, they pay about 30 cents in tax. Under the iteration of UBI we're discussing, that additional dollar in income would mean 33 cents less in the UBI, and 20 cents less in tax. As you can see, scaling up the UBI clawback would really mess with incentives, if even at $52000 of income you're still paying more than half of every dollar you earn back to the government. The solve has to be through marginal tax rates adjustments - and given that the entire system of CPP and EI premiums would have to go to the wind anyways, there has to be an adjustment made to taxation regardless.
As noted, CPP and EI would be done. This would completely and permanently supplant EI (another reason why I think the benefit can't be much lower than $500, which is already lower than the max EI benefit), and given that it's nearly double the max CPP benefit, it would (in the long run) supplant that as well.
There are challenges for CPP, though: There's no clawback on a CPP pension, regardless of what you earn elsewhere, so you can't necessarily do a straight substitution for people with vested CPP entitlements. (Most likely, I'd say that earned CPP entitlements still get paid out, but the program would otherwise be suspended.)
Still, elimination of CPP and EI is a big change. For someone making $60,000 per year, CPP and EI premiums combine to a cost of about $3750 per year just for the employee contributions, with the employer contributing another $4000. (In Alberta, you're paying about $12,000 in Federal and Provincial tax, but yes, the government gets another ~$8000 on your account toward EI and CPP, if you're an employee.)
So you'll see marginal tax rates go up. A lot. Including at lower income levels. If you're an employee under the EI max insurable earnings, then a tax hike of under 7% isn't even really a tax hike. Really, the break-even point for somebody earning $52,000 per year would be around a nine percent tax increase on everything over the Federal basic personal amount. Combined with the UBI as discussed, a ~12% tax increase would leave that $52k earner with the same take-home they're getting now. It's a stunningly huge number. The folks who will pay more are those who aren't 'employees' - i.e. successful entrepreneurs, contractors, and gig workers - and those making over $60,000 to begin with. (But let's face it: They should. As I noted before, if there's one thing we've realized during this crisis, it's that it's not just conventional employees who should have an income security safety net. Even if we don't turn CERB into a UBI, we'll definitely have to extend the coverage (and burden!) of EI-type coverage to other groups. This restructuring of our payroll taxes simply has to happen, regardless of whether we adopt a UBI.)
Provincial welfare programs would be completely subsumed by this benefit. However, to the extent that Provincial disability programs provide assistance with healthcare costs not necessarily covered otherwise, you couldn't just abandon those supports altogether. Nonetheless, Provinces would save a mint on the cost of administering and policing these programs. (Administering a UBI is far cheaper than administering conditional benefits like welfare.)
There are a range of other programs and benefits that would likely get rolled into this - carbon tax rebates, GST rebates, etc.
There may be those who would argue that a UBI should be paired with scrapping the minimum wage. After all, if the government is providing a basic standard of living, doesn't that mean we're okay with no longer putting that burden on employers?
I'm not so sure. As I argued a while back, social assistance programs that support underpaid workers turn out to be subsidies for worthless business models. And remember: Wages take part of the burden off the taxpayer. If there's a 33% clawback, then for every hour that person works at a $15/hour minimum wage, that's $5 saved by the government. It certainly changes the policy considerations underlying the minimum wage, but I still don't conclude that the minimum wage is unnecessary.
There's a larger question here, and I think it significantly turns on the adequacy of the UBI: Will unskilled workers, in receipt of a UBI, still be willing to work for whatever they can get, because they still need more money to subsist? Or will the UBI give them the bargaining power to reject exploitative working conditions and wage rates? It's supposed to be the latter, but there will, no doubt, still be cases where the former remains true. If the UBI does effectively limit the supply of unskilled labour to a sufficient degree to drive up wages in a modest way, that's not necessarily a terrible thing either - for the same reasons that I argued that a $15/hour minimum wage wasn't the economy-destroying disaster that was projected in some circles.
To that extent, we'd need to pay attention to wages, prices, etc.: A UBI is a major economic policy that may need to be tweaked over time if the specific implementation were impacting wages and prices in a way likely to affect production sufficiently to offset its stimulus effect.
But the concept appears to be sound; Canada is likely a good candidate to make it work; and now may well be the right time to try it.
Problems are understandable. This is a complex policy initiative rolled out in a rush, to deal with a crisis. Perfection isn't the standard here, and everyone's largely agreed that something like CERB was necessary to address the pandemic.
Andrew Scheer has raised another concern about the implementation of CERB: The $1000 income cutoff creates a problem for people phasing back into the workplace.
CERB eligibility has some pretty rigid criteria, and among them is a requirement that the individual not make more than $1000 in the 4 week eligibility period (there's a different treatment for the initial claim period, though). For the sake of simplicity, let's call this $250/week. If you're a hair under the cutoff and otherwise eligible, you get the full $500 per week from CERB. If you're a dollar over the cutoff, you get zero.
This is a harsh result, presumably for the sake of streamlining a system facing millions of claims. Scheer is right to say that it creates a difficult situation for individuals looking to return to work, where an individual might get significantly less money because she took all the shifts she could get.
Scheer is taking a lot of heat for his articulation of a 'gradual type of reduction' of benefits, as if he wants to reduce CERB incrementally for everyone across the board to force people back to work. My read of his comments is that this is an unfair characterization, and what he's actually proposing is a progressive clawback, similar to what we do with programs like OAS: If you make over a certain threshold, then you don't immediately lose your entire benefit, but instead there's a formula. For every dollar you make over the threshold, you lose a portion of that dollar in a benefit clawback.
So, conceptually, Scheer seems to be proposing that, if you make $1200 or $1300 in the 4 week eligibility period, you only lose a portion of the CERB benefit. He doesn't propose a whole lot of detail, but he does provide an overarching paradigm that "it's always better off for Canadians to work". (Let's call this the "Scheer Paradigm".)
These clawbacks are good policy, but in this particular context, even aside from dramatically increasing the complexity of the CERB system, it creates other challenges.
The biggest challenge surrounds a particular group of people who are not eligible for CERB: Low-income essential services workers who did not lose their jobs. So, for instance, grocery store workers. Here in Alberta, somebody earning minimum wage (which is higher than much of the country) for a 37.5 hour work week grosses $562.50 per week. Not much more than CERB. And that's without considering that many such workers are part-time only.
How would such a clawback work?
It's hard to envision an iteration of Scheer's clawback where we're not giving residual CERB benefits to people making significantly more than ineligible, still-employed, front-line grocery store workers. It's especially hard to imagine such a system without dramatically reducing existing CERB benefits (recall what I said earlier that it should be one of the governing paradigms of crisis programs that "once a benefit is rolled out, it's guaranteed"). And Scheer has expressly clarified by Twitter that he's calling on the government to make the program "more generous"...so I don't think I'm taking him out of context if I suggest that his plan does not include a reduction of existing benefits.
If we use the threshold of $250 per week before the clawback begins (based on the current threshold of $1000 per four weeks for CERB), and we were to consider a dollar-for-dollar clawback, then your entire CERB benefit gets clawed back at $750/week in income. That's the floor. Remember, the Scheer Paradigm is that Canadians should be better off for working, which means a clawback of something less than dollar-for-dollar.
If we used a 50% clawback on income from $250-$500/week, and than a 75% clawback after that, then the full benefit isn't clawed back until you reach $1000/week in income.
Can we tell a still-employed minimum wage grocery store clerk earning less than $600 per week that he doesn't get any benefit at all, but turn around and give free cash to the person making $950 per week in employment income? Even aside from the poor optics in light of the pandemic, it still violates the Scheer Paradigm: Those grocery store employees are worse off for having kept their jobs.
The only rational solution here is to extend the CERB benefit beyond those who 'lost their jobs due to COVID-19', and to all individuals in the applicable income ranges.
In other words, to fix CERB in any way that doesn't violated the Scheer Paradigm, we need to turn it into a Universal Basic Income (or UBI).
Universal Basic Income
A UBI is a mostly-theoretical social program - there have been a few pilots in different areas - that essentially means ensuring that absolutely everyone is getting enough money to survive, even if they aren't working.
There are different iterations. Some models are means-tested, where you only cut the cheque to people earning less than a fixed amount up front. Other models involve cutting the cheque to everyone, and then clawing or taxing back the money from those with sufficient other income. It's really a 'six of one, half a dozen of the other' kind of thing, where there are limited practical differences, and the main differences between the systems are administrative and political.
It's currently the progressive Holy Grail, the goal of which is essentially the eradication of extreme poverty.
Wait, Isn't That Expensive?
Yes, yes it is. But not as much as you might think. The Parliamentary Budget Officer modelled a UBI a couple years ago based on an Ontario pilot. At a UBI level of about $17k per year for an individual and about $24k for a couple (which, to be fair, is less than CERB would work out to annually), the PBO estimated a cost of about $76 billion per year Federally. However, this would presumably subsume existing Federal benefits such as CPP and EI, and so the PBO concluded that the net cost would be about $43 billion.
Moreover, the PBO only netted the cost against existing Federal benefits, without considering the cost of Provincial benefit programs which would be rendered largely obsolete by a UBI - Ontario, for instance, spends about $10 billion per year on social assistance.
In addition, poverty amelioration costs bleed into a lot of other areas of public spending, such as education and criminal justice.
So while the logistics of implementing a permanent UBI might have their challenges, the net cost to the taxpayer is actually a lot less than you'd expect. In a country where the Federal budget is in the ballpark of a third of a trillion dollars, a new $43 billion outlay is big, but not obviously implausible.
How Would a UBI Impact the Economy?
This is the tougher question. An immediate answer you'll hear from most boomers and conservatives is "A lot of people will decide they don't have to work, and just take the handouts". There's certainly some truth to that in existing social assistance programs - though those of us who have worked with those communities know that it's not necessarily about laziness, but about other underlying disadvantages and disabilities.
The key here is the one highlighted by the Scheer Paradigm: A good UBI is one that preserves the incentive to work.
But there's a flip side to the UBI that the conservatives tend to ignore: Stimulus. When you give money to low-income people, what do they usually do with that money? Why, they spend it, obviously. And with a permanent adequate safety net, the impact on consumer confidence would be very positive. A team at the Berkeley Institute ran models that showed substantial economic growth driven by UBIs.
Still, an honest appraisal of a UBI has to acknowledge that there's still much that we don't know about the macroeconomic impacts of such a program - and small-scale pilot projects are unlikely to shed much light on such issues.
The National Bureau of Economic Research released an excellent (and well-balanced) analysis of the outstanding questions and concerns about a UBI. The authors, Hoynes and Rothstein, suggest that there may be a reduction in labour supply in the short term, but that this isn't necessarily a bad thing. Perhaps more importantly, in the long term, such a support encourages higher levels of education and skill development, and improves outcomes in early childhood development, and would therefore promote a more skilled and productive workforce in the long run.
The Robot Revolution and the Macroeconomic Prisoner's Dilemma
I've historically been sceptical about the concept of a UBI - I regard it as interesting, with the potential to do a lot of good, and yet the cost and overall impact seem to have their risks.
But the thing that makes me increasingly more attracted to it is the trend toward automation of labour. Traditional economic drivers like manufacturing and resource extraction have significantly moved to be less labour-intensive.
Labour obsolescence is nothing new, and there are those who believe that what we are seeing is nothing new: Historically, technological development has wiped out entire fields, and yet simultaneously created new fields and freed up the labour to pursue them. (Consider switchboard operators, for example. Automation replaced them a long time ago...but automated switchboards created efficiencies that promoted economic growth, and created new jobs in other areas.) Or in other scenarios it has improved efficiency to the point where it actually caused businesses to hire more employees - like Henry Ford's moving assembly line, which made the automobile producible at prices available to the mass market.
But I would posit that there is a threshold at which automation will eliminate more jobs than the corresponding efficiencies can possibly create via economic growth - primarily because the corresponding growth opportunities are also highly automated. It is not difficult to imagine a futuristic world where essentially all tasks associated with agriculture, manufacturing, construction, natural resources, transportation, retail, telecommunications, and other industries are fully automated. It is a bit harder to imagine full automation of certain professional services, yet even in fields like accounting and law, there are already moves toward automation with AI and machine-learning.
In such a future world, it is very difficult to imagine productive employment for all members of that society: In essence, we'll be at the point where are productive capacity is no longer capped by the cost and/or availability of labour, but by the availability of raw materials and/or our collective capacity to consume. So where do we cross the threshold from a world where efficiency-driven growth translates into a net increase in employment and wages, to a world where efficiency-driven growth translates into a net decrease in employment and wages?
I'm not sure, but that threshold certainly exists, and it's not totally unreasonable to think that we're approaching it, if we haven't already reached it.
The bottom line is that automation creates an oversupply of labour, cheapening it. That is, after all, the point. And yet, at the same time, businesses rely on a prosperous middle class: If our economy, on the whole, successfully reduces its labour cost, then that generates a corresponding reduction in the ability of the economy at large to consume.
This is what I call the macroeconomic prisoner's dilemma: If I own a business, then my best conceivable outcome occurs if I incur little-to-no labour cost, but if everyone else pays their labour enough money to be able to purchase my product. But if other businesses cut back on their labour in the same way as me, and don't pay workers enough money to purchase my product, then my sales will struggle despite the fact that I can produce more inventory while paying virtually nothing for labour.
A UBI addresses this in multiple ways: Firstly, as Hoynes and Rothstein note, it likely reduces the labour pool, offsetting some of the broader economic implications of automation. Secondly, it provides relief and a safety net to those whose jobs have been eliminated, or those who fear elimination of their jobs. Thirdly, it facilitates workers seeking retraining when faced with obsolescence. And fourthly, it provides unemployed and underemployed individuals with cash in hand to be able to purchase from the private sector.
I've thought for a while that a UBI is likely something that will become necessary at some point in the future. Andrew Scheer, however, has brought me around to thinking that it's probably the right way out of our current mess.
What would a COVID UBI look like?
The CERB itself is a program on a clock; it's only available for a fixed time. As I've said, I think think we can pare back any benefits already offered, but once the current program has run its course, all reforms are on the table.
For the moment, however, there are certain constraints to be observed: Everyone already eligible to receive the full benefit...gets the full benefit. Clawbacks should be based on income over $1000 per 4-week eligibility period. No special treatments or reductions based on spousal income.
So the first question is "who gets it"? The obvious answer is 'everyone', but that's going to come with some question-marks: Are we just talking about people who are working or were working in the recent past? Or are we including people who aren't labour force participants? Stay-at-home parents? Retirees? Persons with disabilities?
I don't see a compelling reason for any such exclusion classes. How the benefit interacts with other types of income and benefits is a complex question, but we've developed a similar program for students (which should be rolled into this one), and I'm not too concerned about other categories. When dealing with crisis programming, and particularly when dealing with changing programs that have already been rolled out, it's better to be overinclusive than underinclusive.
The second question is what the clawbacks should be. I think the numbers canvassed about - no clawback up to $1000 of income (over the four week period), claw back 50% of income from $1000 to $2000, and 75% of income from $2000 to $4000.
As Scheer says, we want people to want to work. Too high of a clawback removes the incentive.
How about later?
Once the CERB runs its course, converting it into a permanent system requires a lot of structural changes to other Federal and Provincial programs, but first let's start with the ways that the payout itself would differ.
The numbers would, of course, have to be revisited. I'm not sure that $500 per week is a terrible starting point, generally speaking (on the one hand, it's substantially higher than the program previously costed by the PBO; on the other hand, it's still a poverty-level income), but there is a certain logic that a couple residing together should get less than two individuals living on their own, and to means-test based on family income. (Also, parents should get something toward the kids...but there's already a framework in place for this.)
More importantly, I think the clawback threshold would need to be revisited. I'm reluctant to suggest reducing the threshold under $1000 under the already-implemented emergency program, because that pulls back money that people are legitimately counting on. But as far as an expansion of the project goes...use a more gradual clawback, and start it at a lower level. One of the problems with too steep of a clawback is that it does create a disincentive to productive labour.
I think it would be perfectly reasonable to have a 20%-33% clawback starting from dollar one - so, playing with $500/week benefit and a 33% clawback for the sake of argument - somebody earning $100 per week in employment income would only get $467 of their benefit. To the extent that we want people earning toward the higher end of the resulting range (up to $1500 per week) to be getting less of a benefit, we do that by playing with marginal tax rates.
Let's look at it this way: In Alberta, somebody making $1000 per week is in a ~30% marginal tax bracket - meaning that for every additional dollar of income, they pay about 30 cents in tax. Under the iteration of UBI we're discussing, that additional dollar in income would mean 33 cents less in the UBI, and 20 cents less in tax. As you can see, scaling up the UBI clawback would really mess with incentives, if even at $52000 of income you're still paying more than half of every dollar you earn back to the government. The solve has to be through marginal tax rates adjustments - and given that the entire system of CPP and EI premiums would have to go to the wind anyways, there has to be an adjustment made to taxation regardless.
What Else Changes?
As noted, CPP and EI would be done. This would completely and permanently supplant EI (another reason why I think the benefit can't be much lower than $500, which is already lower than the max EI benefit), and given that it's nearly double the max CPP benefit, it would (in the long run) supplant that as well.
There are challenges for CPP, though: There's no clawback on a CPP pension, regardless of what you earn elsewhere, so you can't necessarily do a straight substitution for people with vested CPP entitlements. (Most likely, I'd say that earned CPP entitlements still get paid out, but the program would otherwise be suspended.)
Still, elimination of CPP and EI is a big change. For someone making $60,000 per year, CPP and EI premiums combine to a cost of about $3750 per year just for the employee contributions, with the employer contributing another $4000. (In Alberta, you're paying about $12,000 in Federal and Provincial tax, but yes, the government gets another ~$8000 on your account toward EI and CPP, if you're an employee.)
So you'll see marginal tax rates go up. A lot. Including at lower income levels. If you're an employee under the EI max insurable earnings, then a tax hike of under 7% isn't even really a tax hike. Really, the break-even point for somebody earning $52,000 per year would be around a nine percent tax increase on everything over the Federal basic personal amount. Combined with the UBI as discussed, a ~12% tax increase would leave that $52k earner with the same take-home they're getting now. It's a stunningly huge number. The folks who will pay more are those who aren't 'employees' - i.e. successful entrepreneurs, contractors, and gig workers - and those making over $60,000 to begin with. (But let's face it: They should. As I noted before, if there's one thing we've realized during this crisis, it's that it's not just conventional employees who should have an income security safety net. Even if we don't turn CERB into a UBI, we'll definitely have to extend the coverage (and burden!) of EI-type coverage to other groups. This restructuring of our payroll taxes simply has to happen, regardless of whether we adopt a UBI.)
Provincial welfare programs would be completely subsumed by this benefit. However, to the extent that Provincial disability programs provide assistance with healthcare costs not necessarily covered otherwise, you couldn't just abandon those supports altogether. Nonetheless, Provinces would save a mint on the cost of administering and policing these programs. (Administering a UBI is far cheaper than administering conditional benefits like welfare.)
There are a range of other programs and benefits that would likely get rolled into this - carbon tax rebates, GST rebates, etc.
There may be those who would argue that a UBI should be paired with scrapping the minimum wage. After all, if the government is providing a basic standard of living, doesn't that mean we're okay with no longer putting that burden on employers?
I'm not so sure. As I argued a while back, social assistance programs that support underpaid workers turn out to be subsidies for worthless business models. And remember: Wages take part of the burden off the taxpayer. If there's a 33% clawback, then for every hour that person works at a $15/hour minimum wage, that's $5 saved by the government. It certainly changes the policy considerations underlying the minimum wage, but I still don't conclude that the minimum wage is unnecessary.
There's a larger question here, and I think it significantly turns on the adequacy of the UBI: Will unskilled workers, in receipt of a UBI, still be willing to work for whatever they can get, because they still need more money to subsist? Or will the UBI give them the bargaining power to reject exploitative working conditions and wage rates? It's supposed to be the latter, but there will, no doubt, still be cases where the former remains true. If the UBI does effectively limit the supply of unskilled labour to a sufficient degree to drive up wages in a modest way, that's not necessarily a terrible thing either - for the same reasons that I argued that a $15/hour minimum wage wasn't the economy-destroying disaster that was projected in some circles.
To that extent, we'd need to pay attention to wages, prices, etc.: A UBI is a major economic policy that may need to be tweaked over time if the specific implementation were impacting wages and prices in a way likely to affect production sufficiently to offset its stimulus effect.
But the concept appears to be sound; Canada is likely a good candidate to make it work; and now may well be the right time to try it.
Comments
Post a Comment