Self-insurance for workers doesn’t work

This is a guest post from Rod Hill, a Professor of Economics at the University of New Brunswick, Saint John campus. A previous version of this post first appeared in the New Brunswick Telegraph Journal.

In a report this month for the Halifax-based Atlantic Institute for Market Studies (AIMS), entitled “An Alternative to Employment Insurance”, Justin Hatherly proposes replacing the Employment Insurance (EI) system. A look at the proposal quickly reveals how unsatisfactory it is.

Instead of EI, Mr. Hatherly wants individuals and employers to contribute to Personal Security Accounts (PSAs). These accounts would be the property of the individuals, which they could draw upon in certain circumstances in the event of unemployment. The funds would be invested in the stock market by an independent board.

In effect, he is proposing to eliminate EI while expanding the current Registered Retirement Savings Plan (RRSP) system with some compulsory contributions, while restricting the withdrawal of those additional funds.

He writes “Persons who lose work through no fault of their own can draw 55 percent of their wages [up to the insured maximum] for 24 weeks, provided they had contributed for 960 hours” (about 24 weeks of full-time work). “Those who left their prior employment voluntarily would be ineligible” – but why deny them access to their own savings? Quitting a job get a better one is something to be encouraged.

Crucially, “those with insufficient savings receive benefits from a common fund financed by general revenue. However, they incur a negative balance and must pay back the government before contributing to their PSA” to be eligible for further withdrawals or loans.

AIMS is proposing that individuals should rely entirely on their own savings or borrowed money to survive during a period of unemployment.

Every insurance system, public or private, has the feature that those who experience a bad outcome (a house fire, a car accident, a health crisis, layoff, and so on) have benefits that are paid by those who have not (yet) experienced a bad outcome. That is the whole point of insurance. Risk for everyone is reduced as risk is pooled across the whole population.

Instead, Mr. Hatherly is inviting people to ‘self insure’ like people do if they fail to buy house insurance. We all know how that turns out if your house burns down.

A few lengthy periods of unemployment would be no more pleasant. When people self-insure, they bear the entire risk themselves. Those with high and steady incomes may be able to shoulder that risk, but most people, particularly those with lower incomes, would not.

I did a calculation to see how this system would work. Someone earning $50,000 a year and making contributions of 4 percent could take 6 years to accumulate enough resources to cover the proposed maximum withdrawal from their Personal Security Accounts. (Under the current EI system, such a person would be guaranteed a minimum of 36 weeks of benefits, not the 24 in the AIMS scheme.)

This assumes that the invested funds would grow steadily. When the last recession began in 2008-2009, the national unemployment rate rose from 6.1 percent to 8.3 percent, while the Toronto Stock market index fell by more than 40 percent. If unemployed workers had been relying on Personal Security Accounts, their funds would have been decimated at the time they needed them the most.

In his report, Mr. Hatherly notes that even unemployment might not diminish the Personal Security Accounts very much because of “restrictive conditions on benefit withdrawal and duration” – a point which underscores the inadequacy of his proposal for maintaining income and spending after job loss.

An important feature of EI is that benefits and the spending they support kick in quickly where and when layoffs occur. This helps shorten recessions by maintaining total spending.

Mr. Hatherly is right about one thing. With workers left to support themselves during periods of unemployment, they will have an incentive to find employment quickly – assuming, as he seems to, that jobs are available. (Particularly in recessions, the number of people looking for work far exceeds the number of job openings.)

However it’s better for both workers and employers if people to take time to find a job well suited to their skills. As well, a lack of income support during unemployment would increase the bargaining power of employers and push down wages.

No one would argue that the existing EI system is perfect. A much criticized feature is its division of the country into regions where eligibility criteria and benefit duration vary greatly.

In those with the lowest unemployment rates, typically urban areas, a minimum of 700 hours of work are required to be eligible for only 14 weeks of benefits. A minimum of 1820 hours (about 46 weeks of full-time work) are needed for 36 weeks of benefits.
In regions with the highest unemployment rates, 420 hours of work gives eligibility for 32 weeks of benefits. The result is a permanent subsidy to regions of high unemployment and inadequate access to EI benefits for many in urban areas. Just because the unemployment rate is low does not mean that it is easy to get a job. Many people are increasingly stuck in ‘precarious work’, temporary or part-time with no job security.

Any change to this system towards one with greater national uniformity would have to be done gradually to avoid undue hardship in high unemployment regions. It would be best done in conjunction with other changes to income supports, such as guaranteed minimum incomes, an idea governments are now seriously considering.

However, but AIMS’ radical proposal to scrap Employment Insurance completely and to leave individual workers on their own to bear all the risk of unemployment is not an improvement.