What is Employment Insurance For?

 

If Canada's Employment Insurance program was designed solely to insure workers against the loss of employment, it would look very different.

For one thing, the premiums that employees and employers contribute would go towards paying benefits to people who have lost their jobs. But in 2013/14 - the most recent year for which I can find data - less than half the monies collected in EI premiums went to regular benefits for unemployed workers. The rest went to special benefits such as parental and sickness leave, to fishing benefits, to fund "Part II" benefits  (mostly labour market development agreements with the provinces), to cover administrative costs and to run a surplus.

 

If Employment Insurance worked like insurance, people who were at greater risk of being laid off would pay higher premiums or receive lower benefits. There would be experience rating, so workers or employers who used the EI system repeatedly would see their premiums increased or benefits decreased.

But that's not how it works. As the next picture shows, workers in a handful of industries receive, on average, substantially more than they contribute in benefits from Employment Insurance. But because EI has administrative costs, pays for training, and in some years runs a surplus, workers in most industries have a benefit-to-contribution ratio below one.

I could have drawn this graph twenty years ago, and the overall pattern would have looked much the same.  The only major change in recent years is that the expansion of maternity and parental benefits has bumped up the total income benefits/contribution ratios in industries with female-dominated workforces, such as health care and social assistance.

 Not only does Canada's Employment Insurance system not experience rate workers, firms or industries, it actually employs a system that is close to reverse experience rating. The number of hours required to qualify for benefits, the duration of benefits, and even the amount of benefits received can depend upon the local employment rate - what the Mowat Centre EI Task Force dubbed the postal code lottery. The impact of these rules - working alongside the presence of special EI benefits for fisheries workers and in combination with the varying industrial mix and economic conditions of different provinces - is an EI system that benefits some provinces, on average, more than others. 

 Any policy maker whose sole aim was to create a plan to insure workers against job loss would never have come up with our present system.

Which raises the question: why do we have the system we do?

 There are some cynical, and not entirely inaccurate answers:  Employment Insurance is for winning votes in Atlantic Canada. Employment Insurance is for making depressed regions of the country economically viable; for keeping small communities alive. Employment Insurance is for subsidizing industries that are, for one reason or another, considered to be important to the Canadian economy, like the fisheries. Yet these answers raise the question, Why EI? Why use Employment Insurance for regional subsidies, rather than another scheme, such as equalization?

 One way of understanding the structure of our current EI system is through James Buchanan's analysis of the adage "an old tax is a good tax". It is politically easier to maintain a tax that is already in place than introduce a new tax. The figure below shows the evolution of Employment Insurance premiums over time. Canada's Unemployment Insurance program (as it was then known) was radically overhauled in 1971. The new system was immediately hit by the OPEC oil price shock, and costs began to rise. Premiums rose to cover expenses until the early 1990s, when the system was again overhauled and made substantially less generous.

That 1990s reform changed EI from a system that chronically run deficits to one that often ran surpluses. Those surpluses did, eventually, lead to reduced premiums. But they also created fiscal room which could be used to finance new programs like compassionate care and expanded parental leaves, or the labour market development agreements. These programs were not only desirable in and of themselves, they also, by delivering benefits to groups of workers traditionally ill-served by EI, such as employees in the health care industry, bolstered the political base of the EI program.

Yet in the long run this evolution has led the Employment Insurance system away from its core focus of insuring against loss of employment and towards a greater focus on social development-type goals, and leaves us asking - again - what is Employment Insurance for?

Looking at the existing range of programs financed through EI, and how they are structured, it seems that there are four things that Employment Insurance is for at Canada at the present time.

• Income redistribution: EI is for redistributing income to those who are in need
• Social policy: EI is for making it easier for families to cope with sickness and caregiving responsibilities
• Work supports: EI is for making it easier for people to re-enter the labour market after job loss
• Regional and industrial policy: EI is for supporting certain Canadian regions and industries.

Let's take each of these in turn.

Income redistribution: 

Here, EI could do better. The number one item on my wish list is elimination of the regional differences in eligibility for benefits, and for receipt of benefits. 

People who are at greater risk of becoming and staying unemployed already benefit more from the existence of EI than other workers because they are more likely to claim EI and - to the extent that it's harder for them to find work - they claim EI for longer. There is no sense in extending their benefits further.

Now it could be argued that people who remain on EI in low unemployment regions are lazy people who are simply not trying, whereas people who remain on EI in high unemployment regions are hard working individuals for whom there is no job available. Yet there are all sorts of factors that make it hard for people to find work after they have lost a job - being older, being a recent immigrant, having low levels of education, having a disability, and so on. What is the sense behind privileging region of residence, and not the many other factors that contribute to the ease by which people can find employment?

Moreover, Canada implements regional differences in eligibility badly. When people are laid off, their eligibility for benefits is based upon the seasonally adjusted unemployment rate over the past three months for which data is available in their region. Someone who is laid off right at the beginning of a recession will be eligible for fewer weeks of benefits, even though she may struggle to be re-employed (although there is some scope for adjustments). By way of contrast, someone who is laid off towards the end of a recession will receive relatively more generous benefits, even though the economy at this point is starting to recover, and job prospects will be better. The illogic of this system becomes starkly apparent when a part of the economy experiences sudden fluctuations in the employment rate.

The proposal in the 2016 Budget to extend the period for which people can claim benefits in 12 regions that have experienced sudden rises in unemployment rates is a flawed attempt to introduce greater rationality into a fundamentally irrational system. It would be better just to get rid of regional differentiation of benefits.

The second item on my wish list is a more creative and flexible approach to determining eligibility for benefits. Part-time workers and students can pay into EI for years, working summers and weekends, yet never accumulate enough hours to be eligible for benefits. It wouldn't be hard to say to potential claimants who lack sufficient hours, "Sorry, you don't qualify for benefits, here's a cheque that reflects all the amounts that you have contributed to Employment Insurance over the past five years." It should be noted, however, that the announcement in the most recent budget to eliminate the tougher eligibility requirements placed on new labour market entrants will go some way towards easing the situation for part-time workers and students.

The third item on my wish list would be some kind of brake on the subsidies EI delivers to the fishing and construction industries. As the table below shows, the hourly wage rate in construction is above the national average; why subsidize the industry further through the EI system?

Social Policy

EI is not jut about benefits for unemployed workers: the parental, compassionate care, sickness and other provisions are key social policy tools. 

Yet it is not obvious that EI is the best vehicle for financing programs with broader social goals such as training or compassionate care leave. First, EI premiums are generally thought to be a regressive form of taxation (except to the extent that the employer share falls on the owners of capital), and one might have doubts about the desirability of using regressive taxation to fund general social policy programs. Second, use of EI to fund benefits excludes people who are not in the labour market, and often the self-employed as well.

The self-employed can, in fact, chose to participate in Employment Insurance and claim maternity and parental benefits. But according to the 2015 Employment Insurance Monitoring and Assessment Report, there were just 558 maternity claims among self-employed workers in 2013/14, compared to 169,640 maternity claims among employed women. To put that number into perspective, in 2015 fifteen percent of Canadian workers were self-employed (CANSIM 282-0012). Now it may be that self-employed women are choosing to finance their own parental leaves, rather than opting into EI. However to the extent that low take-up for EI special benefits reflects fewer maternity and parental leaves, and to the extent that maternity and parental leave benefits have positive impacts on children's and mother's well-being, the low participation rates of the self-employed in EI special benefits is a matter of concern.

On my wish list: better access to maternity and parental leaves among the self-employed. A re-examination of the desirability of funding social programs through EI premiums.

Work Supports.

There are some telling numbers buried in the Employment Insurance Monitoring and Assessment Report. In 2013/14, over a million people - 1,325,840 to be precise - claimed regular EI benefits. About one quarter of that number, 348,909, received "Part II" benefits - skills development, and so on - while on claim.

These numbers reveal just how disconnected many EI recipients are from the skills development part of the EI program - or indeed human interaction with Service Canada workers. The entire process of claiming EI is now completely electronic. The form to apply must be completed on-line. The bi-weekly reports must be completed on-line. I have someone in my household right now who is claiming EI benefits, and I am truly stunned by the absence of any kind of behavioural nudges, anything that points a claimant in the direction of seeking work. 

On my wish list: behavioural nudges integrated into the EI on-line system, for example, links to job sites, productivity software, etc.

 That's all for now. It should be pretty obvious that I have serious concerns about using EI as a tool for industrial and regional policy.

I'm talking about EI on March 31st here - https://www.cdhowe.org/essential-public-policy-events/employment-insurance-wish-list-reform. 

 

 

 

Comments

[Kevin Milligan]

Hi Frances, wish I could be there to see you and DavidG present on this!

About the EI/UI contribution rates, before 1975 it was a 'tabular' system. I was just looking at this yesterday!

Here's 1962, from the National Finances:
Weekly earnings:
<$9 10cents
9-14.99$ 20 cents
15-20.99 30 cents
21-26.99 38 cents
....
63 - 68.99$ 86 cents
69$+ 94 cents

Employer and employee both paid the amounts above. Note that marginal tax rates at the threhsolds become infinite.

This system was in place until 1972, when they phased in the new (current) system. They mixed the two systems 40% in 1972, 60% in 1973, 80%in 1974 then 100% in 1975.

I put in a 'fudge' number in CTaCS that was an approximate average of that system in the 1960s. But maybe I shouldn't. Anyway, because of that, the part up to 1974 in CTaCS likely doesn't give a correct impression.

I am trying to get the 2016-1 version of CTaCS out the door, and so I don't have the time to do the schedular 1960s UI coding--do you think I should just zero that out in CTaCS for now so that people don't get confused? Thanks, Kevin

[Frances Woolley]

Kevin, thanks for this. May not have a chance to correct this for a little while - see how the morning goes tomorrow. It is clearly differentiated by being in red on the spreadsheet, but I didn't read the footnotes.

Sorry about that.

[Kevin Milligan]

No problem--not trying to force last minute revisions on you. Just wanted to let you know for future reference :)

[Frances Woolley]

Just embarrassed by my carelessness....

[Vladimir]

Unless I'm mistaken there is a clawback of EI benefits if the recipient reports more than about 50k in earnings. Something like 30%. At least it's what I remember from preparing a tax return. That would mitigate dome of your concerns re: the construction industry.

[Kevin Milligan]

Not trying to embarrass you! Just coincidental this was in my mind because I was looking at my 1960s National Finances yesterday looking up some CTaCS stuff.

[Angella MacEwen]

Regarding behavioural nudges, claimants can opt-in to the Job Alert system which will send twice daily emails with ads from the Job Bank: http://www.jobbank.gc.ca/job_alert.do

[Frances Woolley]

Vladimir - was just looking that up today for the EI claimant in my household. Here's the clawback info: http://www.esdc.gc.ca/en/reports/ei/repayment.page. It starts at net income around $63,000, and people who haven't claimed in the past 10 years are exempt from the taxback.

Remember the regular benefit/contribution ratio for the construction industry is more than twice the Canadian average. The clawback will make a bit of a difference, but it's still way out of whack.

[Frances Woolley]

Angella - interesting. That doesn't look like it's specifically for EI claimants, though, and I don't know if there's anything that encourages people to sign up.

I'm thinking of things like this reporting calendar - http://www.esdc.gc.ca/en/reports/ei/calendar.page - wouldn't it be possible to make that a little bit sexier and also encourage people to use is to do thinks like keep track of networking, reaching out for contacts, etc? People must have turned their mind to this....

[Avon Barksdale]

I find the maternity and parental leave benefits in the EI program to be particularly odd. Since most people on the whole have about two children, lumping this program in with insurance makes little sense. To the extent that these benefits are self-financed by the individual's EI premiums, parental leave benefits act more like an implicit loan paid back over the individual's working life.

Suppose that we eliminated the parental leave benefit component from EI. If we agree that maternity and parental leave benefits have positive impacts on well-being, and if we agree that borrowing constraints create some sort of market failure which prevents private self-funding of that benefit, then why not have the government create an explicit loan program? Lower the EI premiums, refocus on actual insurance (which is still a bit of an oxymoron given that unemployment is ultimately voluntary), and let people take advantage of the loan program for parental leave. Those who don't have children don't take the loan.

To me the word insurance should be reserved for providing protection against those events that don't happen with probability one. EI has little to do with actual insurance (and neither does our health care system). Once the government decides to be an insurance company with an army, we quickly find that insurance ends up being the last thought on the government's mind.

[Jacques René Giguère]

To make a comparison, when QC instituted the Régime d'Assurance Parentale (Parental leave) they used the Régie des Rentes machinery but the financing was different and there was no commingling of the two programs.
To mix the two, as Oscar would say , is more than a misfortune, it's mistake.

[Frances Woolley]

Jacques Rene - it would be really interesting to compare maternity/parental leave taken by self-employed women in Quebec/ROC. I wonder if anyone has done that.

Sick leave is also an issue - I have a friend who is self employed and having some health issues and it's totally brutal because no work=no income,

Frances

[anon]

> Lower the EI premiums, refocus on actual insurance (which is still a bit of an oxymoron given that unemployment is ultimately voluntary), and let people take advantage of the loan program for parental leave. Those who don't have children don't take the loan.

This looks like a near-wash at best to me. The main effect is removing a minor subsidy to having kids while working that's arguably a legitimate policy goal anyway - and possibly re-introducing credit-failures, depending on the details of the newly-established loan program. Parental benefits are broadly sensible - IMHO, that's hardly the worst distortion in this Employment Insurance programme.

[Avon Barksdale]

anon,

But it's not a near wash. It confuses what insurance is suppose to be. Let's use insurance for insurance – that is to protect against financially crippling random events. Parental leave benefits fails to qualify as insurance problem. As for the loan system, it just appears as a line item on your tax return every time you use the program until the loan is discharged. We do this for RRSP home-ownership loans, so it's not hard to do. Of course the size of the loan is a function of your earnings when you make use of the program, so delinquency need not be a serious problem.

If we decide that we would like parental leave benefits, fine, whatever, but then let's do that through a separate program – loans or otherwise. Write cheques to the people you would like to write cheques to. But for crying out loud don't call it insurance and certainly don't convolve it with an existing insurance program.

[Determinant]

If Employment Insurance worked like insurance, people who were at greater risk of being laid off would pay higher premiums or receive lower benefits. There would be experience rating, so workers or employers who used the EI system repeatedly would see their premiums increased or benefits decreased.

Not all insurance is rated on an individual basis. Health insurance, for example, is done on a community-rated basis, which is why it is usually sold a group. Disability insurance is rated on a community basis after issue, and is sold as "guaranteed renewable", which means just that. Insurance always contains an element of redistribution from most policyholders to the claimants. Individual rating, which is the standard in property or auto insurance is on one end of the spectrum, community rating is in the middle and pure redistribution through taxation is the other end of the spectrum.

[Avon Barksdale]

"Insurance always contains an element of redistribution from most policyholders to the claimants." ????

Not in expectation, otherwise it's charity.

[whitfit]

I also find the whole system mystifying. I am about to go on parental leave benefits (for the second time) and another thing that is interesting about the parental leave benefits is that many workplaces tie their benefits to the EI benefits. I couldn't get the parental leave benefits through work if I didn't go on EI, but if my wife wanted to use all of the EI for her benefits, I couldn't take advantage of my workplace benefits. My understanding is that this is pretty common.

My wife doesn't get EI parental leave benefits because she doesn't pay into EI. She could, but we did the math and for self-employed people, it works out to what "Avon Barksdale" is suggesting above - a loan that is paid back over time. For us, that loan wouldn't make sense because then I wouldn't be able to access my parental leave benefits, so it would actually cost us either me not taking time off, or money to have her take the EI.

I do think it makes some sense to offer parental benefits, but not sure why it should come out of the EI system (ie taxing EI paying workers), and not the broader tax base.

I also don't think that community rating would be that different for EI than individual rating. Certain groups of workers (university professors, public servants etc...) have stable jobs. Community rating only helps for health insurance because, although correlated with job types, health is not as closely correlated with job types as unemployment.

[Erik Lund]

http://www.dippam.ac.uk/eppi/documents/11292/page/152137(Appendices 6 and 7 of the Irish Fish Inquiry of, I think, 1833; digitised by the University of Southampton. The rest of the commission's report doesn't seem to be digitised, but I found a summary by an American member of this or another commission of inquiry, here. https://archive.org/stream/deepseafishingfi00hold/deepseafishingfi00hold_djvu.txt. The Irish section is pretty deeply buried, but it's there, and seems to cover the issues well enough.)

That was a bit abrupt. I'm sorry to be a bit late to the party, but I'll compensate by being extra narcissistic and telling everyboy that I was a bit busy this week, being a night manager Wednesday night, and a category manager in the morning. Ten hours between shifts, no time to do either job properly. It looks like a labour shortage, but obviously it can't be, because we're not raising wages.

It being a bit late in the day for being arch and ironic, I will stress that this labour market dysfunction is my point, that it is already affecting the reader (unless the reader doesn't eat), and will have its effects on more vital sectors soon enough --if if hasn't already. Canada has an urgent need for new workers which cannot be met from immigration. An effective neo-natal policy is not going to address that problem in less than sixteen year plus nine-and-a-half months at a minimum, but if we move our policy horizon out to the day after "sixteen years plus nine and a half months," the labour market situation doesn't magically get any brighter!

If this doesn't motivate discussion of more effective maternal/parental leave funding schemes, it damn well should. (Although I tend to think of this as tinkering around the edges of the problem. The solution is a social guarantee that a potential parent is saved from all harm. And that goes way beyond a year, or even two years, of partially-paid leave.)

Now I'll return to the subject and make try to make some kind of historian's point. Although I won't be able to resist a bit more Chicken Littling. The case for subsidising fishing is that the business collapses without subsidies. Fisheries fail from year to year. This would be enough risk to gradually shut fishing communities down even if there were not other unpredictable downside risks such as major disasters. Faced with these risks, communities tend to take the windfalls of a good catch and use them to abandon the trade. If you want to have fish, you have to be prepeared to intervene. I am sure that we do not want to go down the road recommended by the Lords Commissioners. Bounties for landed fish are a subsidy for overfishing. The alternative is some kind of support during the off-season. The fact that we do it through UI may or may not be ideal, but it is the way we do it.

One can take a hardline libertarian position and decide that if the outports cannot make it by themselves, they should be abandoned. The problem with that, however, has been demonstrated off Somalia. And do not for a minute imagine that the rest of the world's abandoned, hardscrabble, seaside, boat-owning communities have never been tempted by the prospects of piracy before, or that our own Maritimes are somehow immune from ever feeling that temptation. Though I'm looking for modern piracy to appear in the First World in Greece, first. http://sputniknews.com/europe/20150814/1025748642.html

People always have a resort: crime and flight into substance abuse being the most common in our modern times. They both have victims, and once we get to the point where stealing stuff from other people is the most constructive resort, what kind of society do we have?

[Frances Woolley]

whitfit, thanks for those observations, that's a really interesting point.

Erik - you've made an argument for supporting the fisheries, but what I'm not hearing is the argument for supporting the fisheries through EI, especially given that there's a good body of opinion out there that EI premiums are somewhat recessive (though it really does depend upon assumptions about the incidence of employer premiums). Why EI and not general tax revenues?

[Erik Lund]

Frances, I suppose my answer is the same one you've come up with. EI is what we have. Though it does protect the paid labour force. And it replaced subsidies, which tended to promote over-fishing. On the other hand, slacks in seasonal employment are in part predictable, but in part highly unpredictable. EI already takes that into account: you're only qualified when you're not working! A seasonal employment tax credit --doesn't something like that already exist?-- might be designed as the perfect wonkish alternative. Except, well...

I guess I'm being an old fogey here. I have some sense of how the argument has gone over the last six centuries or so, and the consequences of policy errors has often been quite serious. Ruined towns, lost fisheries, outbreaks of smuggling, wrecking and piracy. And while I'm not being entirely serious in suggesting that Newfie fishers are about to turn to piracy, I do have it in mind that as long as we've got illegal immigration and drugs, they don't need to do anything so crazy. And I'm not sure that we want to encourage drug dealing and people smuggling, either.

[Determinant]

"Insurance always contains an element of redistribution from most policyholders to the claimants." ????

Not in expectation, otherwise it's charity.",

Redistribution is paying $150,000 to a claimant from a pool of $50 premiums from policyholders.

Furthermore, as I demonstrated, there is a large swath of insurance products that are not rated on an individual basis. Any employer-based group health plan which accepts all employees engages in redistribution to employees with pre-existing conditions.

I also don't think that community rating would be that different for EI than individual rating. Certain groups of workers (university professors, public servants etc...) have stable jobs. Community rating only helps for health insurance because, although correlated with job types, health is not as closely correlated with job types as unemployment.

Depends on how you define the community. You want a community with diverse risks. Occupations may be uniform, but for instance municipalities, provinces or countries are much more diverse.

In insurance, not everything is about the individual, and often too much individualism is bad business, both for the insured and the insurer.

[Frances Woolley]

Erik - "and it replaced subsidies, which tended to promote over-fishing." o.k., I missed that point, which is an important one. But wouldn't you figure EI promotes over-fishing too by encouraging entry into the fishery, also makes the entire enterprise more lucrative therefore facilitates capital investment etc?

[Erik Lund]

Frances:

Any scheme that is intended to keep people in the fishery carries a risk of promoting over-fishing. We've seen two solutions for that: the oldest is licenses, but these trust too much in the licensee's interest in preserving, and breaks down entirely when poaching becomes a problem. (Plus, fish swim around.) More recently, and requiring more social control, we have quotas, which are pretty hard to reconcile with bounties. Everyone wants to be the admiral/highliner! On their face, quotas are ideal for combination with employment insurance, because the boat crews won't mutiny when told that they have to go in. Their incomes will be made good by society, the same society that says that, in the interest of the common good, they must stop fishing when they hit their quotas.

The issue here is that it is "intended to keep people in the fishery." This tends to be what sticks in the craw here in the economic core regions. Why are we paying these people subsidies? They should just move to the oilfields! (Intentional irony is intentional.) In highlighting the Nineteenth, and even Eighteenth Century experience, I want to make it clear that we've had a pretty extended run at exactly this kind of sink-or-swim approach to the fishery, and the result has been perverse. There is a resource (fish), and willing buyers for the fish, and somehow the market cannot coordinate the two.

I guess that's leading with the conclusion. What turned my opinion was a very simple argument from one of those many, forgotten parliamentary inquiries, framed in Napoleonic-era logic. The fisher goes out, and does well in a good year, but badly in a bad year. In a good year, the fishing family eats from the catch. What does it eat in a bad year? Well, the Commissioner says, that's a problem, and the solution inherited by the community we looked at was cottager plots. The family has a garden plot, where it grows potatoes, muesli, truck vegetables. . Whatever.

Unfortunately, this placed the family in a bind. On the one hand, Dad can go to sea with the fleet. He does not know if this will be a good year or a bad year. Eventually, he knows that there is a chance that it will be a very bad year, and that he will die at sea. In a bad year, the cottage plot will provide for the family. In a good year, it quite possibly will not, because he won't be there to dig it. In a very bad year, he will impoverish the family by depriving it of his labour for ever thereafter. So the only reason that he will go out to the fleet is if the plot's income is meagre to the point of subsistence.

But! In a good year, he makes a profit. Now, on the one hand, his boat needs a refit. On the other, his neighbour's plot is up for sale. Should the profit be invested in the fishing boat, or the plot? The question, I think, answers itself. There are fish, and there are willing buyers, but the fishing family declines to be the intermediate. I'm not sure I see a non-social, old-timey "liberal economics" solution to this problem.

[Nathan W]

Also, some people who cannot qualify for benefits are still required to pay EI. For example, students must pay EI but if your status is as a student you don't quality for EI. Also, single people are required to pay into a system that subsidizes parental leave, regardless of whether they will ever be in a position to take advantage of it.

Even worse, people who work in temporary jobs and occasional work who desperately need additional benefits but cannot get sufficient hours to qualify, still have to pay into a system that will never benefit them.

The matching of who pays and who stands to benefit could be done much better. I don't know how many thousands I've paid into EI over the years, but between various alternating situations, I've been ineligible for benefits any time I've been unemployed. Now I'm self employed, so my social contributions are more in line of the basic tax and redistribution system, without the additional EI tax that I will never benefit from.