This report is looking at these specific provisions of AAR:
QUOTE
The Center’s analysis, which covers 36 states and the District of Columbia, examines the effect on poverty of seven ARRA provisions: the expansion of three tax credits for working families, two provisions that strengthen unemployment insurance assistance, a provision that boosts food stamp benefits, and a one-time payment for retirees, veterans, and people with disabilities.
1. Expansion of Tax Credits for working families
2. Improvements to Unemployment Insurance
3. Improvements to Food Stamp benefits
4. One time payment to retirees, veterans and persons with disabilities
All things which have happened. It then speaks of "estimates."
Quote:
Nationally, these provisions are keeping more than 6 million Americans out of poverty and reducing the severity of poverty for 33 million more. (These figures include both people whom ARRA has lifted out of poverty and people whom ARRA has kept from falling into poverty.)
These estimates are conservative. The seven provisions examined cover only about one-fourth of the recovery act’s total spending.
The estimates are about the literal number of people presented lifted from poverty or prevented from falling into poverty . It's not a "projection" of the future. But it is an estimate of the past/current effect because no one can say to the number exactly how many people have been helped. Statistical estimates are required. But in this particular case, this is an estimate based on data that exists now about actions already taken. It is
not a projection about the future.
The other 3/4 of AAR spending - well perhaps that's where there would be projections, about programs and spending that have not been fully implemented yet and need more data to fully evaluate. But that's not what this report is about.
The report even goes on to literally explain its methodology to you:
QUOTE
In brief, the analysis uses Census data to examine how these policy changes will affect family income and poverty status by state. The estimates start with data collected in March 2004, March 2005, and March 2006 through the Census Bureau's Current Population Survey. Data for three years are combined to increase the reliability of the state-by-state estimates.
We make three adjustments to these Census data. First, we correct the tendency of Census and other surveys to undercount receipt of certain public benefits, using the data and methods for making such adjustments that are reflected in the U.S. Department of Health and Human Services’ TRIM model. <4> Next, we adjust the data to approximate recent economic and demographic conditions in each state, including labor-market conditions and state population levels in April through June of 2009. Finally, we adjust the food stamp participation data to approximate actual food stamp participation levels by state in May 2009. For each family in the resulting data, we estimate the family’s 2009 income with and without the seven recovery act provisions.
The analysis considers a family to be kept out of poverty if its estimated income is below the poverty line without the recovery act provisions but above the poverty line with the provisions. We use a measure of poverty that adheres to National Academy of Sciences poverty measurement recommendations by including after-tax cash and non-cash income, while subtracting child care and work expenses and out-of-pocket medical expenditures.
We provide estimates for 36 states and the District of Columbia. For the remaining 14 states, our data are insufficient to show reliable results. <5> Given the uncertainty associated with using a sample of the population, we show a range of estimates for each state. This range can be substantial, particularly for states with the smallest survey samples. For example, estimates for Iowa range from 25,000 to 55,000 residents kept out of poverty, with our best estimate falling in the middle (40,000).
The CBPP is estimating how many people have been lifted out of poverty of prevented from slipping into poverty. It is not, however, "projecting" about parts of ARR that have not been implemented yet. It's discussing seven provisions already implemented and their effects.