This chapter highlights broad trends in income inequality, both on a market income basis and an after-tax income basis. The difference between these two concepts of income is government transfers and income taxes, so the chapter also provides information on the redistributive impact of the tax-transfer system.

Quintile analysis is frequently used in this publication. For more information about quintiles, see Chapter 1  and the paragraph titled "Percentiles" in Data quality, concepts and methodology - Notes and definitions .

Has income inequality changed in recent years?

This question can be answered in a few ways. One way is to look at income in absolute terms: did the differences in income across the population become larger or smaller in dollar terms? It can also be looked at in relative terms: how did the distribution of income change in terms of the share held by the families with the lowest and the highest income. Depending on how the income of all families changed, each approach could lead to a different conclusion. The following analysis illustrates changes in inequality among Canadian families using both points of view.

In absolute terms, i.e., in dollar terms, the disparities in after-tax income became wider over the years from 1996 to 2001. This happened mainly because of a greater improvement in the average income of the one-fifth of families with highest income (top quintile). In 2002, the distribution of after-tax income remained about the same as in 2001.

In relative terms, the distribution of after-tax income for the whole period from 1996 to 2002 changed very little.

Incomes improved over several years, across the distribution

Since changes in aggregate income can have an impact on the conclusions to be made about inequality, it is important to recognize that average after-tax income for families had been on the rise since 1996. Most of the increase occurred over the five years from 1997 to 2001. Average after-tax income for families changed minimally between 2001 and 2002 (up 0.3%).

When the population of families is broken down each year into five equal-sized groups or "quintiles", from lowest after-tax income to highest after-tax income, it can be seen that all five quintiles shared to some extent in the increases in after-tax income since 1996. The same is true for market income.

Slight narrowing of the gap in 2002 (in dollar terms), after it widened from 1996 to 2001

The dollar difference between the average after-tax income of the highest and lowest quintiles increased yearly, from $78,000 in 1996 to $94,900 in 2001. In 2002, this gap narrowed slightly to $94,100 (-0.8%). Over the period from 1996 to 2002, the average after-tax income of the highest 20% of families of two or more persons rose by an estimated $19,300 or 20%. Although the lowest quintile had a $3,200 increase in average income, this was a 17% improvement. The middle three quintiles had increases of 15% to 16% in their average after-tax income. In short, the absolute gains of the highest quintile were the largest, in dollar terms and as a percentage of the income they started with. The lowest quintile had the smallest gains in dollar terms, but had the second highest gains as a percentage of the income they started with.

Looking at market income over the same period, the bottom two quintiles (still defined on an after-tax basis) had much larger percentage improvements in income than they did on an after-tax basis, but their market income is so low that even a small variation in dollar terms appears as a large percentage change. For example, the lowest quintile had an increase in average market income of 35% between 1996 and 2002, while the second quintile had an increase of 23%. But in dollar terms, the lowest quintile gained $3,200 in average market income and the second quintile gained $6,400, contrasted with a gain by the highest quintile of $21,100. For the highest quintile, this gain represented a 17% increase in their average market income.

Chart 7.1 shows the average after-tax income for families in each quintile, at three points in time covering the period 1996 to 2002. A line drawn between the points of each adjacent pair of quintiles shows the degree of inequality in the distribution between the five groups of the population, in absolute terms. An increase in the steepness of the line between two years suggests that the distribution has become less equal, while a flatter line suggests it has become more equal.

Between 1996 and 2000, the line in chart 7.1  shifted upwards and became steeper. This is true for all segments of the line, but especially between the fourth and fifth quintiles. This suggests that the inequality of the distribution of income grew across all five quintiles, but especially between the highest quintile and all other families. Between 2000 and 2002, the line shifted upwards slightly, but did not become much steeper. This suggests that the inequality of the distribution did not change much between 2000 and 2002.

Chart 7.1
Average after-tax income of families by quintiles, for years 1996, 2000 and 2002

Income inequality in relative terms

By expressing the income of each quintile as a share of the income of all families, we concentrate on relative changes among quintiles. Any increase to a particular quintile is necessarily a decrease for some other quintiles. How did the shares of income received by lower-, middle-, and higher-income families evolve in the last several years?

There was a very small and gradual shift in favour of the highest quintile families from 1996 to 1998, as their share of after-tax income rose from 38% to 39%. Their share did not fluctuate between 1998 and 2002, at an average of 39%. Any changes in the shares of market income were even less evident over the period from 1996 to 2002.

Chart 7.2
Shares of aggregate after-tax income by quintiles, 1980 to 2002

Increasing ratio of the top to the bottom

Another relative measure of income inequality is the ratio of average income of the highest income families to that of the lowest income families. This measure focuses on the two ends of the income distribution. Again, after-tax income quintiles are used here to identify these two groups of families.

This ratio shows quite starkly the impact that taxes and transfers have in moderating differences between the outer ends of the distribution. In terms of market income, the ratio of average income received by the families in the highest quintile versus those in the lowest quintile was 11.7 to 1 in 2002, i.e., $11.70 held by the highest quintile for every one dollar held by the lowest quintile. However, after taxes and transfers, the ratio was much lower, 5.2 to 1.

As for the movement in the after-tax income ratio of top to bottom, it remained stable at about 4.8 to 1 for several years up to 1995. It then rose in 1996 and 1997 to 5.3 and fluctuated very little in the four years leading up to 2002, when it was estimated to be 5.2 to 1.

While this measure would suggest that income inequality increased during 1996 and 1997, the interpretation of changes in this ratio has to be done carefully. A given dollar increase or decrease for the lowest quintile will always be larger in percentage terms than for the highest quintile. With the ratio measure in particular, when the income of the lower quintile is rising, the value of the ratio can decrease, even while the gap in dollar terms between the top and the bottom is widening.

This apparent contradiction occurred, in fact, when the ratio for market income fell between 1998 and 2000. Over that period, the gap in average market income between the lowest 20% of families and the highest 20% increased by over $4,600, as a result of a $2,300 increase for the lowest quintile and a $6,900 increase for the highest quintile - clearly a larger dollar gain for the higher quintile. But because it represented a 23% increase for the lowest quintile and only a 5.0% increase for the higher quintile, the ratio of average market income of the top to the bottom declined; it fell from 13.8 to 1 in 1998, to 11.8 to 1 in 2000. The ratio of average market income of the top to the bottom then rose to 12.1 to 1 in 2001, but fell again in 2002 to 11.7 to 1.

Chart 7.3
Ratio of average income of the highest quintile families to the lowest, showing market income and after-tax income, 1980 to 2002

The Gini coefficients declined for market income and remained stable for after-tax income in 2002

The Gini coefficient is a number between zero and one that measures the degree of inequality in the distribution of income. Again, this is a relative measure of inequality. The coefficient would register zero (perfect equality) for a population in which each member received exactly the same income and it would register a coefficient of one if one member received all the income and the rest received none. Even though a single Gini coefficient value has no simple interpretation, comparisons of the level over time or between populations are very straightforward: the higher the coefficient, the higher the inequality of the distribution, and vice versa.

The Gini coefficients were fairly stable from 1991 to 1995, at about 0.43 for market income and 0.30 for after-tax income. They then rose for the next two years. After this point, Gini coefficients for market income declined slowly to 0.42 in 2002 and Gini coefficients for after-tax income remained stable at 0.31.

Chart 7.4
Gini Coefficients for families, 1980 to 2002