Sullivan and Meyer Report on Poverty and the effects of inaccuracies in CPI.   Report on the Def of poverty is flawed:  PDF file.  Some exerpts:

We show that bias in the CPI-U has a very sizable effect on changes in poverty.  Between the early 1960s and 2009, an income poverty measure that corrects for bias in this price index declines by 13.5 percentage points more than a comparable measure based on the CPI-U.  The patterns are very different across family types, with consumption poverty falling much faster than income poverty since 1980 for the elderly, but more slowly for married couples with children.  Income based poverty gaps have been rising over the last two decades while consumption based gaps have fallen. We show that how poverty is measured affects the composition of the poor, and that the consumption poor appear to be worse off than the income poor.  We also examine potential explanations for the patterns we estimate. Demographic changes, with the exception of rising educational attainment, do not account for the declines in poverty. Changes in tax policy explain a substantial part of the decline in income poverty particularly for families with children. Other than social security, cash and noncash government transfer programs have only a small impact on changes in poverty.  Measurement error in income is likely to explain some of the most noticeable differences between changes in income and consumption poverty, but saving and dissaving do not appear to play a large role.

Official poverty estimates suggest deprivation has become more widespread over the past four decades—the rate in 2009 is nearly 2 percentage points higher than the rate in 1970—despite a doubling of real GDP per capita and trillions of dollars spent on antipoverty programs. These official estimates, however, suffer from well-known flaws, which include a narrow definition of income, an odd adjustment for family size, and a biased adjustment for price changes (Citro and Michael 1995; Besharov and Germanis 2004, Jencks, Mayer and Swingle 2004a).

First, we show that estimates of changes in poverty over the past five decades are very sensitive to how resources are measured. Broader measures of income poverty have very different patterns than official poverty or other measures that are based on pre-tax money income. Although some previous studies have also documented differences,3 several other studies have argued that the trends are quite similar across these measures.4   We show that an income poverty measure that incorporates taxes declines by about 2 percentage points more during the 1960s, and by another 1.2 percentage points more during the 1990s, than a pre-tax money income measure. In addition, we show that the patterns for consumption based poverty are quite different from those for even a broad measure of income poverty.5 Income based poverty falls more than consumption based poverty during the 1960s. The reverse is true for the 2000s, although in 2009 consumption poverty rises more than income poverty. The patterns are very different across family types, with consumption poverty falling much faster than income poverty since 1980 for the elderly, but more slowly for married couples with children.   Income and consumption measures of the poverty gap have generally moved sharply in opposite directions in the last two decades with income based poverty gaps rising, but consumption based poverty gaps falling.

Second, we show that upward bias in the Consumer Price Index (CPI-U), the index used to adjust official poverty thresholds for inflation, has a very substantial effect on changes in poverty over long periods. Between the early 1960s and 2009, an income poverty measure that corrects for this bias declines by 13.5 percentage points more than a comparable measure based on the CPI-U.

Third, we show that how poverty is measured affects not only changes over time, but also who is designated as poor.

Compared to the income poor, the consumption poor are less educated, less likely to own a home, more likely to live in married parent families, and much less likely to be single individuals or elderly. The fraction of the consumption poor living in married parent families is 80 percent higher than the fraction of the income poor living in such families in recent years.

Fourth, we examine potential explanations for changes in poverty over time and investigate why the trends for income and consumption poverty differ sharply for some family types.  Demographic changes over the past five decades do a poor job of explaining poverty changes, although rising educational attainment does account for some of the decline in poverty.  Changes in tax policy explain a substantial part of the decline in income poverty particularly for families with children.  Rising social security benefits account for a decline in income poverty, particularly in the late 1960s and early 1970s, but other cash and noncash government transfer programs have only a small impact on changes in poverty.  We suspect that measurement error explains much of the large differences between income and consumption measures that focus on the distribution below the poverty line such as poverty gaps. Given the evidence on low asset holdings, particularly for groups such as single parents, saving and dissaving are likely to explain only a small portion of the differences between income and consumption measures of poverty.

Finally, this paper provides improved methods for measuring the material well-being of the poor.  We improve upon consumption measures used in previous studies by calculating better measures of housing consumption for those living in public or subsidized housing and by imputing the flow value of vehicle ownership using detailed information on the cars a family owns.  We also provide estimates of the value of public and private health insurance coverage that allow us to incorporate insurance coverage into a measure of poverty. And, we address concerns about increased under-reporting of consumption in survey data by constructing a measure of core consumption that relies on the components of consumption that are reported consistently well over time compared to the national income accounts.

Our consumption data come from the Consumer Expenditure Survey (CE), which is the most comprehensive source of consumption data in the U.S. We use the CE Interview Survey component for the years 1960-1961, 1972-1973, 1980-1981 and 1984-2009 (see Data Appendix for details).  To convert reported expenditures into a measure of consumption, we make a number of adjustments. While previous studies have made similar adjustments, our approach involves several important methodological improvements. First, we convert vehicle spending to a service flow equivalent. Instead of including the full purchase price of a vehicle, we calculate a flow that reflects the value that a consumer receives from owning a car during the period that is a function of a depreciation rate and the current market value of the vehicle. To determine the current market value of each car owned, we use detailed information on vehicles (including make, model, year, age, and other characteristics). This approach accounts for features and quality improvements through what purchasers are willing to pay.

Second, to convert housing expenditures to housing consumption for homeowners, we substitute the reported rental equivalent of the home for the sum of mortgage interest payments, property tax payments, spending on insurance, and maintenance and repairs. Third, for respondents living in government or subsidized housing, we impute a rental value using detailed housing characteristics available in the survey including the number of rooms, bedrooms and bathrooms, and the presence of appliances such as a microwave, disposal, refrigerator, washer, and dryer.

Finally, we exclude spending that is better interpreted as an investment such as spending on education and health care, and outlays for retirement including pensions and social security.9  We exclude out of pocket medical expenses because high out of pocket expenses are arguably more likely to reflect substantial need or lack of good insurance rather than greater well-being.  However, given the importance of health coverage and changes over time in public and private insurance, we report alternative consumption measures that include a value for public and private health insurance (more details of our use of the CE to measure consumption are in the Data Appendix).  Rather than using the official poverty thresholds, for these alternative measures we specify thresholds that equate poverty in the baseline year (1980). This anchoring of poverty rates in 1980 facilitates comparisons of trends across different measures of poverty. Specifically, for each alternative poverty measure we find thresholds such that the poverty rate for that equivalence scale-adjusted measure is equal to that of the official poverty rate in 1980 (13.0 percent).11

Anchoring our alternative measures to the official measure allows us to examine the same point of the distribution in 1980 so that different measures do not diverge simply because of differential changes at different points in the distribution.12  To obtain thresholds for other years, the thresholds are adjusted for inflation using a price index. Haider and McGarry (2006) show that the share of household income coming from household members outside the nuclear family increased noticeably during the 1990s. For the income poverty results from the CPS, we follow the official measure, designating the family as the resource sharing units. There is also substantial evidence that consumption is under-reported in the CE and that the under-reporting has increased over time.

This core consumption measure consists of food at home, rent plus utilities, transportation, gasoline, the value of owner-occupied housing, rental assistance, and the value of owned vehicles.

Because the official poverty thresholds are adjusted over time using the CPI-U, bias in this price index will lead to bias in poverty trends. Although this bias can be very substantial for changes over long time periods, this criticism has received little attention in the poverty literature.17 The BLS has implemented several methodological improvements in calculating the CPI-U over the past 25 years (Johnson, Reed, and Stewart 2006). Although the BLS does not update the CPI-U retroactively, it does provide a consistent research series (CPI-U-RS) that incorporates many of the changes.18 As we will show, these two price indices yield very different patterns for poverty changes over longer periods.  Between 1960 and 2009 the CPI-U grew on average by about 0.4 percentage points per year faster than the CPI-U-RS, with nearly all of this difference occurring between 1978 and 1998.  However, a consensus view among economists is that the CPI-U-RS does not make sufficient adjustment for the biases in the CPI-U. Unfortunately, there is no clear consensus on the exact amount of the bias over time.

Substitution bias refers to the bias in the use of a fixed market basket when people substitute away from high relative price items. Outlet bias refers to the inadequate accounting for the movement of purchases toward low price discount or big box stores. Quality bias refers to inadequate adjustments for the quality improvements in products over time, while new product bias refers to the omission or long delay in the incorporation of new products into the CPI.   The Boskin Commission (Boskin et al. 1996), a group of distinguished economists appointed by the Senate Finance Committee, provides an authoritative source on the extent of these biases. They concluded that the annual bias in the CPI-U was 1.1 percentage points per year at the time of the report, but 1.3 percentage points prior to 1996 (the extra 0.2 percentage points was due to an inadvertent bias added by a 1978 change that was later corrected).

In the analyses that follow, we use the CPI-U-RS as our base price adjustment.19  However, given the estimated bias in the CPI-U of greater than one percentage point per year, the CPI-U-RS will not fully correct the problem. Thus, we report results using an adjusted CPIU-RS that subtracts 0.8 percentage points from the growth in the CPI-U-RS index each year. We also base this adjustment on Gordon (2006) who argues that even with recent alterations to the CPI-U methodology that make it and the CPI-U-RS essentially the same for recent years, a bias of 0.8 percentage points per year remains. Berndt (2006) reports that the bias remaining in 2000 as estimated by each of the individual Boskin Committee members ranged from 0.73 to 0.9 percentage points per year.