Modified Adjusted Gross Income (MAGI) Income Conversion Methodologies. Section I: Methods Considered by the Department for Income Conversion


The Department evaluated a number of potential conversion methodologies.  To inform this evaluation, it consulted with states, solicited public comments through a request for information (“Solicitation”) in June 2012, and worked with 10 pilot states to test the feasibility of potential conversion methodologies, including analyzing the impacts of proposed methodologies using state data. The Department used the following criteria to assess the proposed methods for income conversion:

  • Unbiased: Across all eligibility categories, the method does not systematically increase or decrease the number of eligible individuals within a given eligibility group or systematically increase or decrease the costs to states.
  • Accuracy: To the extent possible, the method minimizes changes in eligibility status by minimizing losses and gains in eligibility for a given category of coverage.
  • Precision: The converted standard must be stable and repeatable.  In other words, if the methodology to arrive at the converted standard were repeated, it would arrive at the same result.  For example, if a sampling methodology is used, the sample size must be large enough to ensure that the conversion method, if calculated on another sample, would in general yield the same converted standard.
  • Data Quality: The data used to conduct the conversion method are representative of the income and disregards of the population so as not to bias the converted standard due to poor data quality.
  • Administrative Burden: The method minimizes demands on state administrative resources and capacity.

The Department evaluated two broad categories of methodologies: “Disregard Methods” and “Disregard combined with Household Composition/Income Counting Methods” (“Disregard/HCIC Methods”).  There are three potential components of a MAGI-based conversion: disregards; household composition; and income counting.  Disregard Methods establish converted standards incorporating the effects of disregards only.  Disregard/HCIC Methods attempt to account for differences in household composition and income counting rules as well as disregards. Each category has three methods, listed below:

Disregard Methods:

  • Same Number Net and Gross (SNNG)
  • Average Disregard Method (ADM)
  • Marginal Disregard Method (MDM/25)

Disregard/HCIC Methods:

  • Same Number Net and MAGI (SNNM)
  • Average Difference
  • Marginal Difference

As discussed below, the Department’s analysis led to a recommendation of the Marginal Disregard Methodology as the Standardized Conversion Methodology with the margin defined as 25 percentage points below the existing net income Medicaid standard for each eligibility category in a state. 

The Department used both survey data and state administrative data in developing and analyzing the methods. 

The survey data came from the Survey of Income and Program Participation (SIPP), conducted by the Census Bureau.9  The SIPP was selected because it contains data on monthly (rather than annual) income.  It also has the level of detail needed to distinguish income sources that may be treated in different ways and to model certain disregards currently available for a given state and eligibility group.  The SIPP provides detailed data on characteristics such as age and family relationships that are needed to place respondents in the appropriate eligibility categories.   SIPP data for development of the MAGI methodologies came primarily from the April 2010 Cross Section of the 2008 panel.  As in any longitudinal survey, individuals move in and out of the sample depending on whether they complete the questionnaire in a particular month.

A limitation of the SIPP is that sample sizes are roughly proportional to state populations, and therefore can be fairly small for smaller states.  Considering that many SIPP respondents are above the Medicaid or CHIP income standards, or age 65 and older, the effective number of cases available for the conversion analyses in a particular state from that state’s respondents will be even smaller.  Moreover, although state identifiers are available in the public-use data, the survey is not designed to be representative of the low-income population at the state level.  

In order to improve the accuracy of the state-specific analyses, the Department reweighted the SIPP data.  In essence, the full national sample is made to resemble any given state by placing more or less weight on each individual in the sample in proportion to the extent that the state differs from the nation.  For example, in a relatively low-income state, low-income individuals in the national sample will be given more weight. 

For more information on the preparation and use of SIPP data for income conversion see the HHS document “Data Sources for Modified Adjusted Gross Income (MAGI) Conversions” available at

State administrative data on Medicaid and CHIP enrollees were provided by pilot states that agreed to serve as partners with the Department in the development of MAGI methods.10  The Department tested the methods using the most complete data available at the time from the pilot states.

Finally, whether based on SIPP data or state data, all tables, numbers, and thresholds contained in this paper are illustrative.  They should not be interpreted to represent the actual threshold that will apply for any state or eligibility category.  All states, including the 10 pilot states, have the option of choosing either the Standardized MAGI Conversion Methodology or an alternative method that is approved by CMS.  In addition, the tables and numbers in this paper do not reflect the final weighting of the SIPP that will be used to calculate converted thresholds.  While the final converted numbers and thresholds may therefore differ from those in this paper, the methodology used for the calculations will not change.  It is the Standardized MAGI Conversion Methodology as defined below.

View full report


"rb.pdf" (pdf, 747.87Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®