A new report by the Conference Board of Canada says the income gap between richest and poorest Canadians has been widening over the past 20 years, with the richest increasing their share of national income. The poorest and middle-income Canadians have lost share since 1993.

The report, posted on the Conference Board website, says Canada's Gini index -- a measure of income equality -- reached a low of 0.281 in 1989, but began rising in the 1990s and has remained at about 0.32 for the last decade.

The Conference Board explains how median income has risen by just 5.5 per cent between 1976 and 2009, while the richest Canadians have seen their incomes rise far more than that:

Median income increased from $45,800 in 1976 to $48,300 in 2009. This increase has been very modest—just 5.5 per cent over 33 years. Moreover, the gap between average and median income has been growing, which is not good news. The growing gap signals that income growth is distributed unequally.

Paul Krugman, winner of the 2008 Nobel Prize in Economics, gives a simple but effective example of the difference between average and median income. Suppose there are 10 people in a bar and they each earn between $34,000 and $36,000. The average income in that room is around $35,000. The median income is in the $35,000 range as well.

Now let's assume that Bill Gates walks into the bar and that his annual income is $1 billion. All of a sudden, the average income of the people in the bar soars to $91 million, but the 10 people who were already there before Bill Gates walked in are no richer than they were before. The median income remains around $35,000. Only one person is better off, and so inequality rises.

The report says the improved incomes of the top 20 per cent of Canadians are due mostly to very high salaries:

Most gains have gone to a very small group of "super-rich." To understand this, we have to look in more detail at the top quintile group—the top 20 per cent of Canadians. For this we need a different data set, one that provides more information on this group of super-rich. In a study for the Canadian Centre for Policy Alternatives, researcher Armine Yalnizyan uses tax file data to track the richest 1 per cent of Canadians; that is, those whose income was higher than 99 per cent of Canadian tax filers.

She found that this group—the 246,000 people whose average income was $405,000—took home almost a third of all growth in incomes from 1998 to 2007, a decade that saw the fastest economic growth in this generation. She notes that: "The last time the economy grew so fast was in the 1950 and '60s, when the richest 1 per cent of Canadian took only 8 per cent of all income growth."

The phenomenal growth in incomes of the super-rich is not due to the assets they own. Yalnizyan points out that, while it is true that historically the super-rich relied mostly on unearned income from assets, "the income of the richest 1 per cent is due mostly to the lavish sums they are paid for the work they do."

The report finds that among Canadian cities, Vancouver has the highest share of its population (over 16 per cent) living on low incomes, compared to 8 per cent in Victoria and under 4 per cent in Quebec City.

Meanwhile, "Canada's elderly poverty rate has been rising since the mid-1990s, a worrisome trend."

Taxes and transfers, the report says, are helping to reduce income inequality, but not as they once did:

In 2009, the richest group had 24.6 times as much market income as those in the poorest group. After taking into account taxes and transfers, the richest group had 9.1 times as much income as the poorest group. This means that government taxes and transfers are reducing income inequality.

As we saw using the Gini index, however, the tax and transfer system is not reducing the income gap as much as it once did. In 1989, the richest group had 7.2 times more income after taxes and transfers than the poorest group. This increased to 9.1 times in 2009.

One explanation for the weaker impact of the tax and transfer system on income inequality in Canada is the restructuring of many programs that help to reduce inequality. For example, there has been erosion in the employment insurance (EI) program. One indicator of EI generosity is the beneficiaries-to-unemployed ratio. Ken Battle, Michael Mendelson, and Sherri Torjman estimate that this ratio fell from 82.9 per cent in 1990 to 47.8 per cent in 2009. This means that a smaller share of unemployed people are getting unemployment benefits.

Crawford Kilian is a contributing editor of The Tyee.