Generous Benefits Raise Long-Term Unemployment

10/01/2006
Featured in print Digest

If you provide very generous unemployment insurance, you may end up with more long-term unemployment. That's what economists Peter Kuhn and Chris Riddell find when they compare the long-term impact of a highly generous unemployment insurance (UI) program in the Canadian province of New Brunswick with the more modest UI program in the neighboring state of Maine. In Maine's northernmost countries, about 6.1 percent of employed men worked fewer than 26 weeks (half a year) in 1990. Across the Saint Croix River in New Brunswick, the comparative figure was more than three times as high, 20.8 percent. The more-generous UI program in New Brunswick accounts for about two-thirds of this difference, the authors estimate.

A 10 percent UI-induced increase in the income associated with working for less than half a year raises the number of persons working less than a half year by about 10 percent.

In The Long-Term Effects of a Generous Income Support Program: Unemployment Insurance in New Brunswick and Maine, 1940-1991 (NBER Working Paper No. 11932), Kuhn and Riddell use what they term a "dramatic natural income-support experiment" spanning fifty years. Both areas are similar, known for their coastal scenery, cold climate, rural character, and relatively low incomes. Both have relatively sparse populations - 1.2 million in Maine, 740,000 in New Brunswick - that have grown less rapidly than their national averages over the fifty years.

In Canada, though, UI is financed and administered by the federal government. This allows for generous payments that, the authors note, "would likely be unsustainable if UI was self-financing within the region." Personal per capita annual income in New Brunswick is 27 percent below the Canadian average. However, as a statement by the Canadian Construction Labor Relations Association cited by the authors notes, one consequence of the Canadian UI system is "�that many persons voluntarily make what amounts to a way of life out of working only long enough to establish benefits, then drawing them for the maximum period, and then repeating the cycle."

In 1950, Maine and New Brunswick had quite similar UI systems. Since then, New Brunswick's system, with two major expansions, has become much more generous. By 1980, ten weeks worked per year in New Brunswick on average provided an annual income, if UI benefits are included, equivalent to 33 weeks of earnings. This part-time work could be repeated year after year without penalty. In Maine, ten weeks of work plus UI benefits added up to only 13 weeks of pay. Maine's UI payments have remained roughly constant since the 1950s.

Over time, both workers and firms adjusted to the generous UI program in New Brunswick. Although the statistical analysis in their paper focuses only on the effects of UI on weeks worked, the authors suggest that UI might also have had an impact on "education decisions, occupational choices, fertility decisions, migration (workers with high tastes for leisure may be induced to remain in New Brunswick by UI policy), learning effects (it takes time and/or experience to understand the workings of the UI system), and the development of informal institutions." The latter point refers to the likelihood that some firms may re-label a worker who quits a job as a layoff to allow that person to be eligible for UI, or that firms may permit sequential job-sharing in a single job so that two employees can take advantage of the UI system. Indeed, because of its prevalence, the stigma of living off UI may have shrunk in New Brunswick, making it even more attractive, the authors note.

The authors also attribute even larger increases in the part-year work of women, and of less-educated men, in New Brunswick to the UI system. This took place especially in service jobs, which account for half of female employment, as well as in inherently seasonal activities such as primary industries. For example, forestry, fishing, and farming jobs decline considerably in the winter. "Thus, for women, it appears that UI has not just preserved an existing lifestyle, but actually created a new one based on chronic seasonal unemployment," the authors write. For women, 13.8 percent worked less than half the year in Northern Maine compared to 26.2 percent in New Brunswick.

Overall, the authors find that a 10 percent UI-induced increase in the income associated with working for less than half a year raises the number of persons working less than a half year by about 10 percent. In New Brunswick, UI payments account for 6 percent of the province's gross domestic product. That's six times the proportion of GDP representing UI in Maine

-- David R. Francis