[Congressional Record Volume 141, Number 28 (Monday, February 13, 1995)]
[Extensions of Remarks]
[Page E324]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]


                      POVERTY DATA IMPROVEMENT ACT

                         HON. THOMAS C. SAWYER

                                of ohio

                    in the house of representatives

                        Moday, February 13, 1995
  Mr. SAWYER. Mr. Speaker, today I am reintroducing legislation that 
will help Congress target billions of dollars annually in Federal 
program funds to populations most in need, as well as measure the 
effectiveness of public assistance programs in a timely way.
  The Poverty Data Improvement Act passed the House unanimously in 
November, 1993. The bill requires the Census Bureau to produce and 
publish poverty estimates for States, counties, cities and school 
districts every 2 years. Currently, the only source of reliable poverty 
data below the national level is the decennial census. According to the 
General Accounting Office, $22 billion is allocated through 19 Federal 
grant programs each year to State and local governments based on those 
poverty figures.
  Clearly, the infrequent production of small area poverty data has 
undermined the ability of many critical Federal programs to reach their 
target populations effectively. As Federal dollars become more scarce 
in the effort to balance the budget, it will be even more important to 
ensure that these programs are serving communities that are most in 
need. Concentrations of poverty are not stagnant over the course of a 
decade. The movement of lower-income populations into rapidly growing 
areas, as well as the abandonment of older cities by the middle class, 
causes a shift in demographic patterns that must be measured more often 
than once every 10 years.
  A notable case in point is the title 1 grant program for elementary 
and secondary schools, which Congress reauthorized as part of the 
Improving America's Schools Act in the 103d Congress. The 1990 census 
income data, which reflects 1989 economic conditions, was used for the 
first time to allocate title 1 funds in the 1993-94 school year. At 
their best, the figures were 4 to 5 years old. And the year before 
that, 1980 census data--reflecting 1979 income--was still being used to 
allocate title 1 funds. Imagine using figures that are nearly 14 years 
old to allocate nearly $7 billion to counties and school districts 
across the country. How can we have any confidence that those funds are 
reaching children and schools that need the most help?
  Unfortunately, the Senate did not act on the Poverty Data Improvement 
Act in the last Congress. But Congress saw the folly in relying on 
outdated poverty numbers to develop and administer important programs 
such as chapter 1, the Job Training Partnership Act, Community 
Development block Grants, and rural housing programs, to name a few. In 
its reauthorization of the title 1 program, Congress called for the use 
of updated county poverty estimates by 1996 and updated school district 
poverty estimates by 1998, in allocating program funds. We also asked 
the National Academy of Sciences to undertake a multi-year study of the 
Census Bureau's effort to produce poverty estimates for States, 
counties, cities and school districts every 2 years. Timely data are an 
important factor in policy development, but it's also important for 
policymakers to have confidence in the numbers on which they rely.
  To its credit, the Census Bureau has recognized the critical policy 
need for more frequent poverty numbers below the national level. The 
Bureau has started the research and development phase of its small area 
poverty estimates program, and reports that it is on schedule to 
release poverty figures for States and counties in the fall of 1996.
  Given the significant amount of taxpayer dollars that are distributed 
according to poverty data, the Census Bureau's effort is a bargain. In 
fiscal years 1994 and 1995, the Bureau will spend $600,000 per year to 
develop its first intercensal poverty numbers. In subsequent years, the 
annual cost will rise to approximately $800,000, with additional costs 
likely to produce poverty estimates for school districts. Nevertheless, 
that's far less than a hundredth of a percent of the funds allocated 
each year on the basis of that data.
  The Poverty Data Improvement Act addresses one important element of a 
growing debate about the accuracy of data we use for Federal program 
purposes. That element is the question of timeliness. Data that are old 
may look precise, but they simply aren't accurate.
  The bill does not address broader--and very legitimate--concerns 
about the way we define poverty. In fact, today we are using 
definitions that were developed nearly 30 years ago. Fortunately, the 
Committee on National Statistics of the National Academy of Sciences is 
completing a comprehensive study of the definition of poverty. That 
study includes a review of consumption patterns, differences in cost of 
living across geography, and the effect of noncash benefits on living 
standards. The academy expects to release its findings and 
recommendations in May.
  Mr. Speaker, we need the capacity to identify demographic and 
economic forces that are changing more rapidly than our ability to 
measure them using traditional data collection methods. Accurate, 
useful, and timely data can serve as a solid foundation on which to 
build sound and cost-effective programs. The Poverty Data Improvement 
Act represents an important start toward achieving that goal. I urge my 
colleagues to support this worthwhile legislation.


                          ____________________