November 11, 2003
Mr. Charles Louis Kincannon
Director, U.S. Census Bureau
4700 Silver Hill Road
Washington, D.C. 20233-0100
Dear Mr. Kincannon:
In September, the Census Bureau released the official poverty rate for 2002, using a method of calculation that was created by a government statistician some 40 years ago. By officially adopting this measure, the United States has committed to annually documenting and publicizing the extent to which the country is progressing towards a more just society. However, we are concerned that today, the official poverty rate as calculated by the Bureau misses important elements of well being and does not adequately inform policy makers who utilize it to craft effective policy interventions.
The Census Bureau uses a set of money income thresholds that vary by family size and composition to determine who is in poverty. The official poverty definition uses money income before taxes and does not include capital gains accumulation and non-cash benefits (such as public housing, Medicaid, and food stamps). The threshold for poverty for a family of five in 2002 was an income of $22,007.
But an even more meaningful indicator of well being than cash income - the amount of assets (or wealth) a family has accumulated - is currently not being considered in our poverty determinations at all. This is a crucial omission. While income refers to the flow of dollars during a set period, typically a year, wealth is the total extent, at any given moment, of an individual's accumulated assets over time and access to resources.
Sufficient levels of income can generate wealth, and most types of assets can quickly be turned into income, making it a convenient source of potential consumption. Both are important to well being.
Yet, when judging the well being of families and individuals, especially over the long term, assets are an even more pertinent indicator: while income is used for day-to-day necessities, assets, by contrast, can be turned into income, relieve individuals from being dependent on others for income, and provide a crucial safety net during times, such as unemployment or illness, when income is disrupted. Assets are also directly transferable from generation to generation and are key to planning for the future, including for retirement and education. Assets are a basic statement of what one owns and a sure indicator of one's capability to be truly self-reliant.
We believe it is important that assets be counted when determining the well being of those citizens who are defined as poor. As long as economic well being for the poor is perceived as only a question of income, or the lack thereof, then social policy options that are designed primarily to help the poor accumulate financial or other assets will likely not be fully understood, appreciated, or utilized. For example, legislation is under consideration in Congress to support Individual Development Accounts (IDAs), a kind of Individual Retirement Account (IRA) for the poor that is designed to help low-income families accumulate assets that can be used for housing, to start a small business, or for education. IDA demonstration projects around the country are proving that citizens who are poor, with the proper incentives, can save and accumulate the assets necessary to step by step improve their lives and break cycles of poverty. But since the federal government's official measure of poverty only focuses on cash income, strategies like IDAs designed to help the poor accumulate assets are almost by definition outside of the "mainstream" of policy prescriptions, regardless of their impact on economic well being. When assets are counted by federal agencies, it is primarily for the purpose of means testing anti-poverty programs, thus becoming a disincentive to the poor to save.
Assets should also be considered in determining poverty because, as some social scientists have now documented, they are important to changing attitudes among low-income citizens themselves. Some researchers contend that assets improve household stability and create an orientation towards the future, but also stimulate development of other assets, increase civic participation, and enhance the well being of offspring. It should be obvious that understanding the extent to which poor families are, or are not, accumulating assets, would be vital in shaping social policies that provide long term benefits to the poor, and to society.
Finally, assets should be measured for the poor because it is essential that policy makers understand the impact of asset-based social policies on the overall distribution of opportunities in our society. The reality is that, though there are several strategies to help citizens in our society develop assets, the poorest citizens rarely benefit from these programs. These include mortgage tax deductions, as well as 401(k)s, IRAs, Roth IRAs, the Federal Thrift Savings Plan and other tax-based incentives to promote savings. At present, most of the benefits of these programs go to households with over $100,000 in income. Over 90% goes to households with over $50,000 in income. The official poverty statistics should include asset estimations in part to specifically assist policy makers in understanding the extent to which these or other asset development strategies actually benefit all citizens, including those who are poor.
The Census Bureau already provides several alternative calculations for the poverty rate, but none addresses the issue of assets. We believe that it is time for the Census Bureau to officially update its methodology to provide policy makers, and the public, a more accurate and comprehensive understanding of the economic well being of our most vulnerable citizens. The official poverty measure as is simply does not provide a truthful representation, nor is it complete.
We would appreciate your response to the following questions:
1. Does the Census Bureau agree with the general thesis that assets (along with income) are also a key determinant of the well being of families and individuals and that overall welfare can best be understood by considering how they control all of the resources under their disposal? If not, please explain why the thesis is incorrect. If so, why has the Census Bureau not included assets as an element in poverty calculations, or developed an additional definition or calculation to determine who is "asset poor"?
2. Does the Bureau believe that the assets a family or individual own are also related to the amount of consumption they can afford? If so, isn't it inherently inaccurate to purport to measure a family's well being without also considering assets?
3. What specific value do various assets, including home ownership, savings, stock ownership, etc., have as social indicators? Does the Bureau believe it is in the interest of all citizens, including those who are poor, to accumulate assets? Does the Bureau agree with the proposition that asset ownership provide citizens a stake in society that income alone cannot provide?
4. Does the Bureau believe that it would be useful to understand how "resource deficient" families are; meaning the extent to which their available financial resources will allow them to survive at least three months should their normal income be disrupted?
5. Does the Bureau agree with the proposition that the way a problem is defined has a critical impact on the development of proposed solutions? If so, what are the implications for the way the Bureau now defines poverty solely in terms of income?
6. What factors would the Bureau recommend be considered in defining and measuring "asset poverty"? How difficult would it be for the Bureau to develop data necessary to make this determination?
Joseph I. Lieberman
United States Senator
Committee on Governmental Affairs