Thank you very much for the invitation to speak to the committee today. I very much appreciate this opportunity to discuss some really critical issues with you.
The federal government spends billions of dollars every year on policies and programs directed toward poverty, and I include here transfers to the provinces, some of which is used for social assistance and housing and other programs. If someone, let's say a reporter or a citizen, were to ask a member of the government, let's say the Prime Minister, about accountability for the funds spent, what could the government say?
Let's be more specific. Suppose someone asked the Prime Minister how many poor people we have in Canada at this time. What would his answer be? And let's suppose he were asked whether the government policies have been effective in reducing poverty over the years. What would his response be? In both cases, he would have to honestly say he doesn't know. And those of us in the research community, if we were being honest, would have to say we don't know either.
There are two main reasons for this unfortunate situation. There's no official poverty line, or even generally accepted measure in Canada. That's number one. Number two, the indicator that we use to tell us how well off people are, which is income, is so badly flawed that we couldn't estimate poverty accurately even if we all could agree on a poverty line.
Let me look at the second reason first. Income is used in almost all studies that estimate the number of poor people. However, the income data that we all use for estimates comes from Statistics Canada and is actually drawn from surveys in which households are asked to reveal their income for the previous year. So the data we all use is reported income, which may not be the same as actual income. For a variety of reasons, people do not always accurately report their income.
One of the reasons for this may be linked to the deliberate hiding of income as part of tax avoidance. The 1999 Auditor General's report focused some attention on this issue, and estimated that the legal part of the hidden economy was about 4.5% of GDP. The illegal part, things like gambling, prostitution, drug dealing, theft and so on, would be very significant as well. The report suggested that the hidden income problem would grow over time due to the growth in self-employment and electronic commerce.
Permit me to give you a little peek at the 2004 distribution of reported income drawn from the Survey of Household Spending micro-data file. We'll start with slide one.
At the top end of the distribution, we had more than 1,000 households with a reported income of $3.7 million. Because the survey is a random sample, every record in the file represents a certain number of households in the country. At the bottom of the distribution, we have almost 2,000 households with negative reported incomes. That's due largely to small-business losses. The average reported income in 2004 was $63,400.
Now, we're interested in poverty, so let's take a closer look at the very bottom of the income distribution. That is all households with incomes of $5,000 or less. If you'll turn to slide two, that will give you that bottom part of the distribution.
I used $5,000 as an arbitrary point, simply for this illustration, because a household or even a single person cannot survive in Canada on less than $5,000. They simply can't purchase all the basic necessities. Yet in 2004, we had 185,000 households that apparently did so.
What is puzzling is that our last-resort program, social assistance, provides in every province more than $5,000 per year for households in need. So there are some real questions about the reliability of the data that tell us that 185,000 households had incomes this low. And I should mention that 41,000 of these household had zero reported income. Overall, this group at the bottom of the distribution, which is to say $5,000 or below, had average reported income of just $1,951, below $2,000. By anyone's standards, this group would be the poorest of the poor in Canada.
However, the same Statistics Canada database provides additional information about these households. The average reported spending by this low-income group of households was over $20,000, which is more than 10 times their average reported income. The significant discrepancy between reported income and reported spending at the lower end of the distribution calls into question the basic data that we all use to measure poverty. We will never be able to estimate the extent of poverty with any accuracy until we deal with the issue of reported incomes. In all my writings on poverty, I have pointed out this problem and I've urged Statistics Canada and government policy-makers to address this issue of poor data.
The other reason we cannot answer the question of how many poor we have in Canada is that there is no accepted poverty definition that we can use as a basis for a measure. There are two broad approaches, as Mr. Finnie mentioned, to defining and measuring poverty. The relative approach identifies someone as poor if they have substantially less than most others in the society or the community. A commonly used relative measure is half the average income. This approach is popular within the social welfare community. It views poverty as a problem of inequality. The absolute approach views poverty as a problem of insufficiency. According to this approach, you are poor if you lack or cannot acquire any of the basic necessities. Clearly, even this has a relative component. The basic necessities have to be those that are accepted within your own society or community.
An example of a relative line, which is frequently used in Canada, is Statistics Canada's low-income cut-off, known as LICO. There are some concerns with this measure besides the fact that it has a very complex construction and it's very difficult to explain to Canadians. One concern expressed by the provinces, which has prompted the development of the market basket measure, is the fact that the lines are too high to represent “poverty” as most people understand that term. The provinces were quite clear that they were tired of having their social assistance rates compared to the LICO lines as if LICO represented poverty.
Another concern is that Statistics Canada does not support the use of LICO as a measure of poverty. Chief Statistician Ivan Fellegi has made the following statement. This was in 1997:
For many years, Statistics Canada has published a set of measures called the low income cut-offs. We regularly and consistently emphasize that these are quite different from measures of poverty. They reflect a well-defined methodology which identifies those who are substantially worse off than the average. Of course, being significantly worse off than the average does not necessarily mean that one is poor.
This is a direct quote from the Chief Statistician of Canada.
If we were to use the LICO measure, which is used by the National Council of Welfare, the poverty rate for Canada is now exactly where it was in 1981, 15.9%. There has been no improvement over that period.
An example of an absolute measure of poverty is the one that I've developed. It's called the “basic needs poverty line”. One often-heard criticism of this measure is that it is too low, that it's even mean-spirited. I've always been puzzled by this critique. Why wouldn't we want, as part of our need to understand poverty, to have a measure that could tell us how many of our fellow citizens simply cannot afford the basic necessities of life? The lines are low, of course, because they represent the cost of basic needs in the various parts of Canada. I should tell you, the basic needs poverty rate has declined by about 31% between 1981 and 2004, from 7.1% to 4.9%. This estimate, of course, uses reported incomes, and we know that's flawed.
Just to conclude, I have been consistent in my recommendation that Canada adopt a small number of measures, perhaps two or three, that would give us a clear picture of the prevalence of poverty in Canada. One single measure may not tell us everything we need to know. This was, in fact, a recommendation of the Copenhagen agreement in 1995, to which Canada was a signatory. However, we simply cannot make any progress in determining the extent of poverty until we get the data question settled. I cannot stress enough the importance of this step.
Thank you.