Poor Poverty Statistics

Poverty is an awful social condition and many studies show that it produces a wide-array of negative externalities, including crime and substance abuse. Although eradicating poverty is a noble cause, elected officials frequently politicize the issue and use “poverty reduction” as a means of saving face and many interest groups stretch the definition of it to absurd lengths, which distracts us from discussing real, abject poverty in Canada and abroad.

The Cape Breton Post and the Chronicle Herald have given significant coverage to the issue following a report released by the Canadian Centre for Policy Alternatives (CCPA). The most shocking statistic in the report is that 33 per cent of children in Cape Breton are living in poverty. In 2014, one out of every three children in Canada live in poverty, which, from the outset, appears to be a terrible social blight that the government needs to rectify. Those calling for action would certainly be justified in doing so if the situation were as bad as they describe.

Statistics Canada does not have an official measure of poverty and the statistical agency said it would not institute one without Parliamentary consensus. It does, however, measure low-income individuals and families using multiple methods: the Low-income Cut Off (LICO), the Low-income Measure (LIM), and the Market Basket Measure (MBM). LICO and LIM measurements are relative measures of poverty. In other words, these measures determine poverty relative to another statistic, such as the median income. If income gains in the top 50 per cent of the income distribution were faster than in the bottom 50 per cent, for instance, these measures would make it seem as though more individuals have entered poverty. In the case of children, for instance, Statistics Canada noted that there was a statistically insignificant decrease in LICO rates between 1979 and 2009, whereas the LIM and MBM rates increased between 2008 and 2009.

The CCPA study relies on the After-tax Low Income Measure (AT-LIM) to gauge poverty rates in Nova Scotia, which defines poverty as 50 per cent of median income adjusted for family size. This measure and the CCPA study are problematic because relative measurements of poverty are not indicators of absolute wellbeing and the AT-LIM does not reflect income growth of low-income individuals: an increase of income for all groups would indicate that poverty has not improved. The nature of this measurement also allows for groups and politicians to define poverty according to their own ideological beliefs and gives them justification for implementing large social programs in the interest of eliminating poverty.


Measuring absolute poverty is a more accurate method to assess how many people are living in actual poverty. Chris Sarlo, a Senior Fellow with the Fraser Institute and economist at Nipissing University, used the “basic needs” approach to measure poverty. This method attempts to define the basic needs of individuals to sustain long-term physical wellbeing. Essentially, there is a basket of goods and the poverty line represents the amount of income necessary to obtain those goods. According to Sarlo’s calculations, this measure fell from 12 per cent of the population in 1973 to 5 per cent in 2004. Statistics Canada uses the MBM, which is a much broader basket than the Fraser Institute, and ideology, too, can shape the “basket of goods.” For instance, some baskets include university education, but one could reasonably argue that a lack of university education does not constitute abject poverty.

In its truest form, abject poverty is extremely problematic and requires a solution, however, its prevalence in Nova Scotia and Canada is largely exaggerated. Relative poverty, on the other hand, is typically rooted in a lack of economic opportunity in a given area and the best response is pro-growth policy–something the region has been sorely lacking for quite some time.

Corey Schruder is an AIMS on Campus Student Fellow who is pursuing an undergraduate degree in history at Cape Breton University. The views expressed are the opinion of the author and not necessarily that of the Atlantic Institute for Market Studies

Reconsidering Income Statistics

The measure most often used to track prosperity is some form of income statistic. Although this intuitively makes sense, the devil can be in the details. Income statistics, in general, can be misleading in many ways. Furthermore, some measures of income seem to be better than other measures. Nevertheless, measures of prosperity are likely to continue comprising of income measures and, therefore, it is important to understand their flaws and limitations.

The limitations of income statistics as a measure of prosperity fall into two categories. First, income is only one part of the picture. Second, it is not always clear that changes in income statistics accurately reflect changes that occur in actual per-capita income.

To begin with, income is not an all-encompassing measure. Other contributors to prosperity include job satisfaction, health, and leisureliness, which income statistics do not typically capture. In addition, certain aspects of the economy remain unaccounted for using income statistics, such as volunteer work, the black market, the household division of tasks, and other activities that do not appear on an individual’s tax return. As a result, using income on its own to measure prosperity is clearly misrepresents an individual’s level of affluence.

There is yet a deeper flaw in the measurement of income statistics, which lies in the imperfect science of trying to draw conclusions using statistical abstractions.

For instance, a common way to measure, and think about, income statistics is to consider the household unit. After all, most people are familiar with their household’s monthly budget and use this to help their understanding of the economy.

Household income, however, is, perhaps, the worst way to measure income. This is fundamentally a result of household variation, not only over time, but also between different demographics and income brackets. Thus, changes in household composition often cause variations in household income. Economist Thomas Sowell has studied this phenomenon extensively, offering three general observations:

1) Household size has decreased over time. Between 1967 and 2005, real per-capita income in the United States grew 122% (whereas household income grew 31%). In this case, shrinking household size has “weighed down” apparent income growth
2) Household composition, including the number of working household members, varies between groups. For instance, the highest earning 20% of American households consists of 64 million people, while the lowest earning 20% of households consists of 39 million people, including single-parent families, single-earner families, and so on. It is therefore easy to speculate that differences in household income can partially be attributed to differences in households
3) Most household income statistics do not follow flesh-and-blood individuals over time. This is important because a given income bracket does not perpetually consist of the same individuals, despite some narratives. While many people are surely living in conditions of poverty, individuals gradually climb the income ladder throughout their career. Sowell’s research suggests that individuals in the bottom 20% of households are more likely to end up in the top income bracket than to remain in the bottom bracket after ten years.

In any case, ‘household income’ is an abstract measure. Consider a fictional economy, for instance, consisting of a married couple living in a single dwelling. If said couple separates, however, “household” income decreases by 50%. Although per-capita income does not change, using household income as a measure would not lend that impression. This is precisely the rationale for distinguishing per-capita income from other statistical categories, such as household income.

Overall, it is important to recognize the limitations of income statistics when attempting to assess the wellbeing of specific groups, entire populations, or whole nations. While household income is particularly flawed, it is unlikely that any single income measure wholly measures prosperity or happiness.

Michael Craig is a 2013-2014 Atlantic Institute for Market Studies’ Student Fellow. The views expressed are the opinion of the author and not necessarily the Institute