New findings on asset poverty in Canada and developments on asset poverty measurement

by David W. Rothwell and Yunju Nam

Asset poverty

The most common method of economic poverty measurement requires defining of a minimal level of basic needs and economic resources. Economic poverty can be measured using income, consumption or wealth; subjective or objective criteria; multiple or single dimensions; and, from relative and absolute perspectives. In advanced economies, the predominate method for understanding economic poverty relies on annual household income as an indicator of command over resources. There are disagreements over methods to understand income poverty (e.g., see the Institute for Research on Poverty’s overview here). The U.S. poverty line is an absolute measure. Relative measures are used for international comparison.

In the 1980s and early 1990s scholars began challenging the reliance on income to understand household well-being. They argued that saving and asset accumulation function as more than stored income to be used for future consumption (seminal works by Oliver and Shapiro; Sherraden). According to Sherraden, when people accumulate assets, they think and behave differently; and the world responds to them differently. Assets function not as a flow but as a stock and are more permanent. Following these ideas, economic deprivation, i.e., poverty, can be measured using assets instead of income. Studies of asset poverty complement and contrast our understanding of the poverty condition.

Economists Robert Haveman and Edward Wolff considered a household or person asset poor if their access to wealth-type resources is insufficient to meet basic needs for some predefined period of time. Wealth-type resources usually involves financial assets or net worth; basic needs can be approximated with the income poverty threshold; and, period of time has usually been set at three months. Therefore, a person or household is considered asset poor if their asset resources fall below one-fourth of the official income poverty line. Concretely, assuming the annual income poverty threshold for a family of four is $32,000, the household would be considered asset poor if owned assets were less than $8,000.

Asset poverty in Canada

Recently, we estimated the first known asset poverty measures in Canada using the 1999 and 2005 Survey of Financial Security. We produced asset poverty rates based on (1) both financial assets and net worth, (2) the Canadian Low Income Cutoff as the threshold of basic needs and (3) three months as the period time. In the paper we reported the national asset poverty rates to be 53% based on financial assets and 34% based on net worth. Below we highlight key findings and contextualize them with US asset poverty findings.

Asset poverty is closely related to age. Figure 1 shows how the rate of financial asset poverty decreases with age. The line represents the rate of asset poverty across the life course. The shaded area reveals the uncertainty around the given asset poverty estimate. It is well known that younger people have struggled in the recovery from the Recession. This finding suggests a need for social policy targetted at younger households.


Joint income and asset poverty

When the joint distribution of poverty based on income and assets is considered we were able to identify three sub-populations: (1) 14% of households were joint income and asset poor, (2) 3% of households were income poor but asset non-poor, and (3) 40% of households were income non-poor but asset poor. The third group reveals that a large segment of the Canadian population has sufficient income to be non-poor but lacks access to assets to survive for three months at the low income threshold. Future policy efforts will play a role in reducing or reinforcing this economic vulnerability.

Comparison with U.S. asset poverty

Differences in measurements and survey design make cross-country comparisons difficult. (For example, using 2001 data, Haveman and Wolff reported that 25% of U.S. households were poor based on net worth compared to the Canadian estimate of 34%. Using financial assets, the disparity was greater: 53% in Canada compared to 38% in the U.S.) It is better to examine systematic cross-country data. Brandolini and colleagues used data from the Luxembourg Wealth Study (years 1999-2002) to compare the asset poverty rates in several OECD countries. Using 50 percent of the median income as the income threshold, they reported that Canada had the highest financial asset poverty rates at 56.5 followed by the US (52.6).

It is perhaps more useful to examine over representation among the asset poor. The most recent CFED scorecard, showed the 2011 liquid asset poverty in the US was 43.5% and these rates were analyzed separately based on race and family structure. The rate of white households compared to households of color was 34.7 to 60.6. The rate of asset poverty among single parent households is nearly double that of 2 parent households (1.94). The analysis showed that 17% of Americans are in extreme asset poverty, i.e., they have negative or zero net worth. In the Canadian study, we decomposed asset poverty rates to understand over-representation. We created a disproportionate index where 1 is perfectly representative of the population rate.

Variable Population Asset poor Disproportionate index
Female lone parents 3.91 6.01 1.54
Age under 25 6.26 9.65 1.54
Renters 36.37 51.11 1.41

Although not perfectly comparable with the CFED and Haveman results (e.g., race and ethnicity is not measured in the Canadian survey), there are some parallels between Canada and the U.S. Asset poverty is disproportionately experienced by female single parents, younger people, and renters.

Our recent study reinforces the importance of asset poverty measurement to understand dimensions of poverty and economic vulnerability that go unnoticed when using an income based measure of poverty.

Importantly, the method of asset poverty measurement used in the U.S. and Canadian studies assumes that households need a certain amount of asset to meet consumption needs at the poverty threshold. In addition to asset poverty measures described above, new and more comprehensive asset-based economic well-being measures have been recently developed in the United States. These measures incorporate ‘asset for development’ perspective in that they recognize assets’ roles in promoting long-term economic development as well as assets’ roles in protecting families from unexpected economic emergencies (Nam, Huang, & Sherraden, 2008). For example, the Asset Security and Opportunity Index produced by the Institute on Assets and Social Policy at Brandies University includes asset opportunity as well asset security. The asset security index measures economic stability and a type of precautionary savings. While asset security is similar to asset poverty (i.e., assets needed for the period of unemployment), asset opportunity is a fundamentally different concept: it is based on the amount of economic resources needed for a family’s investment for the future (i.e. assets for college education, homeownership, and business start-up). Using this concept, over 50% of U.S. households have insufficient assets to promote social development (Shapiro, Oliver, and Meschede 2009).

In addition to precautionary savings for the time of unemployment, the US Department of Commerce (2010) highlighted the importance of two asset measures: savings for college education and retirement. College savings amounts are estimated from zero to $6,800 per year depending on family type and income while retirement savings is assessed as 1.2% to 3.3% of annual family income (U.S. Department of Commerce, 2010). The most comprehensive asset measure is included in a new framework called the Basic Economic Security Table (BEST): precautionary, retirement, homeownership and college education (McMahon, Nam, & Lee, 2011). Using the BEST, we estimated that monthly savings required to meet all four saving needs ranged from $155 to $572 depending on family size and conditions.


Brandolini, A., Magri, S., & Smeeding, T. M. (2010). Asset-based measurement of poverty. Journal of Policy Analysis and Management, 29(2), 267-284. doi:10.1002/pam.20491

Radio pieces on poverty and social development

I’m not sure if it’s my confirmation bias, but it certainly seems that US media are talking more and more about poverty, inequality, and social stratification. Over the last few weeks, I’ve come across the following pieces (they’re great to listen to while doing the dishes!).

On Point from WBUR Boston, interviewed Richard Florida on September 26, 2013. He spoke about a recent article he wrote, at the Atlantic. It’s a nuanced argument. He finds that the post-crash US growth model is working, but that high inequality threatens the sustainability. In the interview at minute 32:00, he recognizes that his previous arguments about clustering (advanced in his work on the creative class) were misguided. Richard Florida is the director of the Martin Prosperity Institute at the University of Toronto’s School of Management.

This American Life from PRI did a piece on unconditional cash transfers in developing African countries over the summer. The piece, explains what UCTs are, why they might work, and why charities (in this case, they interviewed someone from Heifer International) by and large are against them. It’s a great, thought-provoking, and complicated discussion. A few weeks ago, an economist from the World Bank came to McGill to discuss findings that show that conditional cash transfers are related to better outcomes for families than unconditional cash transfers.

Planet Money from NPR also has been delving into the issue of poverty with episodes (they’re short!) on the trouble with the US poverty line and welfare. The poverty line episode is a great introduction to why poverty has been measured the way it has in the US and exactly why this methodology is often divorced from current phenomena that better describes deprivation and hardship.

Dazinger – The Mismeasure of Poverty

In the following article, Sheldon Dazinger explains how a limited measure of poverty skews our perception of the degree of poverty in the United States.  He argues for the use of a measure that incorporates anti-poverty policies in order to provide evidence of the success of the “safety net” in reducing poverty overtime.  In addition, Dazinger explains that economic growth has done little to alleviate poverty in the last few decades.  Growth has not reached the poor because of high levels of inequality and the concentration of growth at the highest end of the income distribution.  He argues for an increase of the minimum wage to begin correcting this.

How do, and should we, measure poverty?

I recently came across this story from the Columbia School of Social Work. Three notable academics address the topic of how to best measure poverty. The analysis is US-centric but includes some valuable content on comparing measures and how municipalities can use poverty measures to inform policy. The story with audio files can be found here.


Blog authors are solely responsible for the content of the blogs listed in the directory. Neither the content of these blogs, nor the links to other web sites, are screened, approved, reviewed or endorsed by McGill University. The text and other material on these blogs are the opinion of the specific author and are not statements of advice, opinion, or information of McGill.