FMAPs and the Impact of Decennial Census Data

FY 2013 Federal Medical Assistance Percentages; Decennial Census Data Affect the Flow of Medicaid Funds

by Vic Miller*

 *Vic Miller is a Medicaid finance expert who was the founder and first director of Federal Funds Information for States (FFIS). Mr. Miller, now an independent contractor, prepared this paper for the National Association of Medicaid Directors. It is one in a set of occasional papers by Mr. Miller on Medicaid and fiscal federalism.

Summary

The Federal Medical Assistance Percentage (FMAP), the share of a state’s Medicaid costs borne by the federal government, is recalculated each year. The FY 2013 FMAPs will again show substantially more reductions than increases, with 24 states experiencing a decline of 0.50 percentage points or more and only 12 receiving that level of increase. This partly reflects personal income shifts among the states, and partly the impact of a first-time use of the 2010 decennial census data.

Background

Medicaid is a joint federal-state program of health insurance, primarily for those of low income.  Costs of health benefits are shared between the federal government and each state depending on a state’s federal medical assistance percentage (FMAP). [1] [2]  The FMAP, which is recalculated each year, is based on each state’s per capita personal income over the most recent three calendar years compared to the national average for those years.[3]

FMAP = 1 – .45 x [(State PCI)2 / (U.S. PCI) 2]

A state with average per capita income receives an FMAP of 55 percent, and itself pays 45 percent of the cost. No state may receive an FMAP less than 50.00 (where the federal government provides one dollar for each state dollar) or higher than 83.00 (where the federal government provides $4.88 for each state dollar). On average, this formula has resulted in the federal government paying for about 57 percent of spending on Medicaid benefits nationally and states 43 percent.

The personal income data used to develop the FMAPs are based on a three-year average of data published by the Department of Commerce’s Bureau of Economic Analysis (BEA).  FMAPs are recalculated each year and published annually between October 1 and November 30 in the Federal Register for the federal fiscal year that begins the following October.  For example, the FMAPs that apply in FY 2012, which began October 1, 2011, were published in November 2010, and were calculated using the latest per capita personal income available at that time, for calendar years 2007, 2008, and 2009.

The FY 2012 FMAPs are thus the first to incorporate fully state income data from the recent economic downturn.  While the recession officially ended in June 2009, states experienced much of the worst of the impact from the recession into 2010, both in terms of increased service demands and lost revenues.  There have been discussions over the history of the program to change the formula to incorporate more timely data, but changes would result in substantially different results across states, and these efforts have not borne fruit.

FY 2013 FMAP Projections 

FY 2013 FMAPs. Table 1 displays FY 2013 FMAPs based on just-released per capita personal income data. Mississippi will again have the highest FMAP (73.43), as it has since the Medicaid program began. However, this FY 2013 FMAP will be its historical lowest, almost 10 percentage points below the 83.00 it began with in 1969.  Most states with high FMAPs have seen them decline substantially over time. By comparison, 14 states again will be at the 50.00 minimum, a total first reached in FY 2012. This compression of FMAPs reflects the comparatively slow economic growth of the Nation’s wealthier states over time.[4]

Twenty-four states will experience a decline of 0.50 or more in FY 2013 and only twelve will receive a similar increase. This continues the experience of FY 2012, when 17 states lost 0.50 or more and only eight had that level of increase. These changes partly reflect ongoing personal income shifts among the states. However, FY 2013 FMAPs will be the first to use the decennial census population counts to develop per capita estimates, and some state FMAPs will be as affected by changes in their population estimates as by changes in their personal incomes.

The most substantial FMAP increase in FY 2013 will be for Nevada (+3.54), following an increase of +4.59 the previous year. This means that Nevada will receive $1.48 from the federal government for each state Medicaid dollar it expends in FY 2013, up from $1.07 per state dollar in FY 2011, a 38 percent increase. Other substantial increases in FY 2013 are regionally diverse, including Florida (+2.04), Delaware (+1.50), Hawaii (+1.38), Texas (+1.08) and Idaho (+.77). The many substantial losses are spread through all regions, with the greatest concentration in the Plains states. The most substantial state losses in FY 2013 will be experienced by North Dakota (-3.13), South Dakota (-2.94), Missouri (-2.08), Arizona (-1.62), Vermont (-1.54), Utah (-1.38), Iowa (-1.12) and Nebraska (-0.88).

Enhanced FMAP. The enhanced FMAP used for the Children’s Health Insurance Program (CHIP) and small parts of the Medicaid program itself is calculated by reducing the state FMAP share by 30 percent, with the resulting 65.00 for 50.00 FMAP states the CHIP minimum. Table 1 displays the FY 2013 enhanced FMAPs, and compares their levels to those of FY 2011.  Nevada receives the largest two-year increase of 2.48 percentage points, followed by Delaware (+1.05). North Dakota (-2.19) and South Dakota (-2.06) experience the largest two-year reductions.

Data Analysis

Federal income and population data. Per capita personal income data used in the FMAP calculation are in fact two separable data streams, income and population, which are calculated respectively by two separate Department of Commerce units, the BEA and the Bureau of the Census. BEA combines the two streams to produce per capita estimates.

Both the income and population data are periodically adjusted to reflect new information, definitions or data structures. One of the major changes occurs once a decade, after the decennial population census population data counts are released for April 1 of the decennial year. Between the decennial censuses, population data used for the state per capita estimates are those generated by annual estimates produced each year from the July Current Population Survey (CPS). These CPS estimates are based on samples rather than counts, and are center points of confidence intervals whose size may be relatively large. Following the decennial counts, two changes are made to produce population data for the BEA September publication of per capita income estimates:

  • The decennial April data are “aged” to produce July estimates for the decennial year that will be consistent with other years’ July data.
  • Previous year state population estimates are readjusted to produce a smooth trend of intercensal population with the new decennial data.

As a result, the new per capita income data published in the September following each decennial census year include not only the new population data for the decennial year but also adjusted data for the previous intercensal years. These changes often are larger for smaller states, since the annual CPS data are produced for them with smaller samples and generally higher standard error intervals.

Historical Perspective. Substantial changes in states’ FMAPs occur for a variety of reasons. In some cases, these changes reflect long-term secular shifts in a state’s economy. Michigan’s economic experience over the past decade illustrates this effect, with its FMAP rising steadily from 52.72 in FY 1999 to highs of 66.14 in FY 2012 and 66.39 in FY 2013. This means that Michigan in FY 2013 will get $1.98 for each state Medicaid dollar extended, as opposed to $1.12 in FY 1999, a 77 percent increase. Wyoming’s energy-based economy illustrates the opposite effect. The state, which had experienced FMAPs near or above 60.00 for most years, rose to a high of 64.60 in FY 2001—reflecting the decline in its energy industry at the end of the previous decade—but then fell rapidly, reaching the minimum 50.00 in FY 2008 and staying there through FY 2013.

Other reasons for major changes include cyclical shifts in states’ economies, including the federal government’s involvement. Shifts in farm income and federal farm payments caused major swings in Plains state FMAPs in the 1970s, as shown in Table 2. North Dakota’s experience is the extreme.  While all states were affected, North Dakota’s shifts are extraordinary—falling almost 20.00 percentage points between FY 1974-75 and FY 1978-79, and then swinging back more than 10.00 percentage points in FY 1980-1981. This experience was repeated in the 1980s, and was a primary impetus to the change beginning in fiscal year 1987 to an annual adjustment in FMAPs from a biennial adjustment.

More recently, Louisiana has also experienced a major shift in its base FMAP because of shifts in federal disaster relief spending. Louisiana’s FMAP, at or near 70.00 in most years, fell to 63.61 in FY 2011 and 61.09 in FY 2012. This means that for every state dollar expended the federal share declines from $2.35 at 70.00 to $1.57 at 61.09, a 32 percent decrease.

A third major reason for major FMAP changes for a state is the introduction of decennial census population counts and the subsequent effect on all three years of per capita income estimates. As an example, 23 states experienced a shift of 1.00 or more, a very substantial shift, for the FY 1984-1985 FMAPs following the introduction of 1980 census count data. The Census Bureau had previously underestimated both the overall size of the U.S. population and its migration out of the northeast. Massachusetts and New York lost substantially while many southern and western states gained.[5]

This effect has continued for the following decennial censuses. FY 1993 saw 19 states with FMAP changes of 1.00 or greater following the introduction of 1990 census data. California, whose population had been accurately estimated, nonetheless saw a decline in its FY 1993 FMAP because shifts in other states changed the national average to which it was compared. [6]

FY 2003 saw 11 states with changes of 1.00 or greater compared to an average of just over 6 for the other years between FY 1999-2007. All states with the largest population adjustments–Nevada, Arizona, Rhode Island, Florida, North Carolina, Georgia and Utah—saw substantial increases in their FMAPs.[7]

The impact in FY 2013 is more nuanced. States that appear to receive the largest population adjustment upward after the 2010 census are Hawaii, North Dakota, Wyoming, New Mexico and Nevada. Hawaii and Nevada receive substantial FMAP increases as a result, and Wyoming continues as a 50.00 state. However, the differential economic growth of energy states such as North Dakota and New Mexico is so large that the influence of the population adjustment is not to increase their FMAPs but rather to reduce their loss.

Concluding Observations

Reductions in FMAPs in FY 2013 continue the ongoing compression of FMAPs since Medicaid program began. As a result, the extent of redistribution of wealth intended in the initial FMAP formulation has declined. In addition, the impact of using new decennial population data can result in substantial multi-year underfunding of Medicaid programs in given states, which is never remedied until the new data are used. Nevada, which has experienced such underfunding before two consecutive decennial censuses, is only the worst example.



[1] Administrative cost-sharing is done under a different structure.

[2] The District of Columbia has a statutory 70.00 FMAP. Five federal insular areas also have Medicaid programs, but they are reimbursed under a different structure.

[3] All FMAP references in this paper are to the base FMAPs established in the Social Security Act, not the increased FMAPs provided for in the American Recovery and Reinvestment Act of 2009.

[4] For further explanation of the historical compression of FMAPs see Miller, Vic; An Overview of Changes in the Federal Medical Assistance Percentages (FMAPs) for Medicaid; Kaiser Commission on Medicaid and the Uninsured; July 2011.

[5] See Miller, Victor J.; “Recent Changes in Federal Grants and State Budgets,” in Lewin, Marion Ein, The Health Policy Agenda, Some Critical Questions; American Enterprise Institute; 1985.

[6] See Federal Funds Information for States (FFIS) Issue Brief 01-04 for a discussion of this shift.

[7] Colorado, which also had a substantial increase in its population in the 2000 census had an FMAP of 50.00 at that time, and received no adjustment.

Read the full report here.

Comments are closed.